^i. m ^.^ i UNITED STATES EPARTMENT OF X)MMERCE •UBLICATION iv^/ \/ \ I %^ I ij~i—\ / Fishery Bulletin U.S. DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration National Marine Fisheries Service Volume 71 \ Number 1 Vol. 71, No. 1 January 1973 WIGLEY, ROLAND L., and FREDERICK CHARLES STINTON. Distribution of macroscopic remains of recent animals from marine sediments off Massachusetts 1 VENRICK, E. L., J. A. McGOWAN, and A. W. MANTYLA. Deep maxima of photo- synthetic chlorophyll in the Pacific Ocean 41 KNIGHT, MARGARET D. The nauplius II, metanauplius, and calyptopis stages of Thysanopoda tricuspidata Milne-Edwards (Euphausiacea) 53 PERKINS, HERBERT C. The larval stages of the deep sea red crab, Geryon qidnquedens Smith, reared under laboratory conditions (Decapoda: Branchyrhyncha) 69 KARNELLA, CHARLES. The systematic status of Merluccius in the tropical western Atlantic Ocean including the Gulf of Mexico 83 JACKSON, RODNEY G., and MARTIN SAGE. Regional distribution of thyroid stim- ulating hormone activity in the pituitary gland of the Atlantic stingray, Dasyatis sabina 93 DUBROW, DAVID L. Effect of drying and desolventizing on the functional properties of fish protein concentrate (FPC) 99 CHENOWETH, STANLEY B. Fish larvae of the estuaries and coast of central Maine . . 105 SANDIFER, PAUL A. Effects of temperature and salinity on larval development of grass shrimp, Palaemonetes vulgaris (Decapoda, Caridea) 115 SHERBURNE, STUART W. Erythrocyte degeneration in the Atlantic herring, Clupea harengus harengus L 125 HIDA, THOMAS S. Food of tunas and dolphins (Pisces: Scombridae and Coryphae- nidae) with emphasis on the distribution and biology of their prey Stolephorus bucca- neeri (Engraulidae) 135 TAYLOR, JOHN L., CARL H. SALOMAN, and KENNETH W. PREST, JR. Harvest and regrowth of turtle grass (Thalassia testudinutn) in Tampa Bay, Florida 145 O'HARA, JAMES. The influence of temperature and salinity on the toxicity of cadmium to the fiddler crab, Uca pugilator 149 LINDALL, WILLIAM N., JR., JOHN R. HALL, and CARL H. SALOMAN. Fishes, macroinvertebrates, and hydrological conditions of upland canals in Tampa Bay, Florida . . 155 KROUSE, JAY S. Maturity, sex ratio, and size composition of the natural population of American lobster, Homarus americanus, along the Maine coast 165 Le GUEN, J. C, and GARY T. SAKAGAWA. Apparent growth of yellowfin tuna from the eastern Atlantic Ocean 175 CONOR, S. L., and J. J. CONOR. Descriptions of the larvae of four North Pacific Porcellanidae (Crustacea: Anomura) 189 CONOR, S. L., and J. J. CONOR. Feeding, cleaning, and swimming behavior in larval stages of porcellanid crabs (Crustacea: Anomura) 225 HASTINGS, ROBERT W. Biology of the pygrmy sea bass, Serraniculus pumilio (Pisces: Serranidae) 235 eattle. Wash. NUARY 1973 (Continued on back cover) U.S. DEPARTMENT OF COMMERCE Peter G. Peterson, Secretary NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION Robert M. White, Adminisirator NATIONAL MARINE FISHERIES SERVICE Philip M. Roedel, Direcfor Fishery Bulletin The Fishery Bulletin carries original research reports and technical notes on investigations in fishery science, engineering, and economics. The Bulletin of the United States Fish Commission was begun in 1881 ; it became the Bulletin of the Bureau of Fisheries in 1904 and the Fishery Bulletin of the Fish and Wildlife Service in 1941. Sep- arates were issued as documents through volume 46; the last document was No. 1103. Beginning with volume 47 in 1931 and continuing through volume 62 in 1963, each separate appeared as a numbered bulletin. A new system began in 1963 with volume 63 in which papers are bound together in a single issue of the bulletin instead of being issued individually. Beginning with volume 70, number 1, January 1972, the Fishery Bulletin became a periodical, issued quarterly. In this form, it is available by subscription from the Superintendent of Documents, U.S. Gov- ernment Printing Office, Washington, D.C. 20402. It is also available free in limited numbers to libraries, research institutions, State and Federal agencies, and in exchange for other scientific publications. EDITOR Dr. Reuben Lasker Scientific Editor, Fishery Bulletin National Marine Fisheries Service Southwest Fisheries Center La JoUa, California 92037 Editorial Committee Dr. Elbert H. Ahlstrom National Marine Fisheries Service Dr. William H. Bayliff Inter-American Tropical Tuna Commission Dr. Daniel M. Cohen National Marine Fisheries Service Dr. Howard M. Feder University of Alaska Mr. John E. Fitch California Department of Fish and Game Dr. Marvin D. Grosslein National Marine Fisheries Service Dr. J. Frank Ilebard National Marine Fisheries Service Dr. John R. Hunter National Marine Fisheries Service Dr. Arthur S. Merrill National Marine Fisheries Service Dr. Virgil J. Norton University of Rhode Island Mr. Alonzo T. Pruter National Marine Fisheries Service Dr. Theodore R. Rice National Marine Fisheries Service Di-. Brian J. Rothschild National Marine Fisheries Service Mr. Maurice E. Stansby National Marine Fisheries Service Dr. Maynard A. Steinberg National Marine Fisheries Service Dr. Roland L. Wigley National Marine Fisheries Service The Secretary of Commerce has determined that the publication of this periodical is necessary in the transaction of the public business required by law of this Department. Use of funds for printing of this periodical has been approved by the Director of the Office of Management and Budget through May 31, 1974. DISTRIBUTION OF MACROSCOPIC REMAINS OF RECENT ANIMALS FROM MARINE SEDIMENTS OFF MASSACHUSETTS Roland L. Wigley' and Frederick Charles Stinton" ABSTRACT Macroscopic animal remains are common constituents of bottom sediments on the conti- nental shelf and upper continental slope south of Cape Cod, Mass. The largest quantities are in sandy deposits in the vicinity of Nantucket Shoals, where they form nearly 30% by volume of the total substrate. The smallest quantities are along the outer continental shelf and upper slope, where animal remains generally make up less than 1% of the substrate. Representatives of all three major realms of aquatic animals contribute to the prefossil skeleton assemblages; benthic forms are the principal components, nek- tonic forms are common, and planktonic forms are rare. The quantitatively dominant taxonomic groups present in the sediments are: echinoderms, mollusks, and teleosts. Typical specimens of all groups represented in the samples are illustrated. Charts and graphs show the geographic and bathymetric distributions of the common species. Durable remains of recently (up to several thousand years) deceased animals and plants constitute an important, but frequently over- looked, link between living organisms and their fossils. Reconstruction of the marine environ- ment that existed in past geological ages can be better approximated when present-day marine populations and processes are well understood. A conventional approach used in paleobiological investigations is to equate the habits, ecological requirements, and functional morphology of fos- sil species with their living relatives (Ladd, 1957; and others). Consequently, a thorough knowledge of existing life is valuable to geologi- cal advancement. Events during the transitional phase between death and fossilization may strongly influence the dispersal, shape, and as- sociated species of fossil remains. Frequently these events must be clearly understood to in- terpret fossil findings correctly and completely. It is in this context that the prefossil stage is considered to be significant in determining the history of life. A series of samples collected from the ocean bottom off southeastern Massachusetts provide ^ Northeast Fisheries Center. National Marine Fish- eries Service, NOAA, Woods Hole, MA 02543. - Bournemouth, Hants, England. an insight into the composition and the geo- graphic distribution of macrobenthic, nektonic, and planktonic animal skeletons — or portions thereof — that occur in continental shelf bottom sediments and that are available for fossiliza- tion. Thus the purpose of this report is to de- scribe qualitatively and quantitatively the mac- roscopic animal remains (durable portions of recently dead animals) in the bottom sediments of this representative portion of the continental shelf in New England. To avoid undue repetition of the words "dead," "deceased," "remains," and similar descriptive terms throughout this report, it must be empha- sized at the outset that all samples of animal materials dealt with in this report are the re- mains of dead animals. Accounts of the living organisms obtained in these collections will be dealt with in other reports. Previous studies of paleontological interest pertaining to prefossil marine animal remains are very diverse in subject. A few examples of these studies include such dissimilar topics as: the composition and distribution of mollusk shell rem.ains (Habe, 1956; and others), bio- logical alteration of bottom sediments (Schafer, 1956; Rhoads, 1966; and others), comparison Manuscript accepted May 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. FISHERY BULLETIN: VOL. 71, NO. 1 of the feeding habits and sediment types inhab- ited by epifaunal and infaunal benthic animals (Craig and Jones, 1966), catastrophism in the sea (Gunter, 1947; and others), position of pelecypod shells in different environments (Em- ery, 1968), burial of mollusk shells (Johnson, 1957), radiocarbon dating of relict oyster shells (Merrill, Emery, and Rubin, 1965), and other related subjects. Most of these studies are re- stricted to one specific topic. The present study, likewise, has a limited objective: to describe the species composition and distribution of mac- roscopic prefossil animal remains. Literature pertaining to present-day mollusk remains in marine bottom deposits is relatively common; see references in Habe (1956), Schafer (1956), Johnson (1957), Belyaev (1970) , and others. In contrast, however, a pau- city of reports dealing with prefossil fish re- mains became strikingly evident during our lit- erature search. Research on this subject tends to be regionally oriented. For example, the study by Jensen (1905) deals with otoliths from an Arctic basin. David (1947) and Soutar (1967) described fish remains from ofl^ southern California, and Belyaev and Glikman (1970) describe selachian teeth from a broad expanse of the Pacific Ocean. A major exception to this regional basis is the report by Brongersma- Sanders (1949), which summarizes the earlier literature pertaining to fish remains (albeit mostly fossil) from many parts of the world. Prefossil remains of marine organisms are more easily obtained than are those of most ter- restrial or aerial forms. Macrobenthic and nek- tonic organisms are usually abundant on conti- nental and insular shelves, and their skeletal components are massive compared with those of microplanktonic pelagic forms. As a result, the "fossil assemblages" (Craig, 1953) of the conti- nental shelf are dominated by macroscopic or- ganisms, as opposed to planktonic forms that make up the bulk of deep-sea fossils. Likewise, the prefossil material of organic origin on con- tinental and insular shelves is generally of a larger size, and the macrofaunal components are considerably more abundant than they are in the deep sea. MATERIALS AND METHODS Samples were collected 11-20 June 1962, from the Bureau of Commercial Fisheries (now the National Marine Fisheries Service) RV Dela- ware at 62 stations south of Martha's Vineyard, Mass. (Table 1; Figure 1). Stations were spaced at intervals of 16 km on a grid pattern having eight north-south transects at right angles to the depth contours. Quantitative bot- tom samples, including sediments and the con- stituent benthic fauna, were collected with a Smith-Mclntyre grab sampler (Smith and Mc- Intyre, 1954), This instrument effectively sam- pled a 0.1-m- area of bottom to a depth of about 10 to 17 cm. The volume of bottom material analyzed from individual samples averaged 8.9 liters. At sea, contents from the grab were washed on a 1-mm mesh sieving screen. Ma- terial remaining on the screen after washing was removed and preserved in a solution of neutral Formalin.' In the laboratory ashore. * Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. . . 7,1' 7,0« , , , , Wf; '-'[T^-'-'^m^^ NANTUCKET '20' -^^^^- '._X 41*- BLOCK '' •'AETNA'S VINEYARD _ ^^jl^ '' ( \ \ ^-7>63' .46 :45,.3o »2V:^; / ^i .62 .47 .44 .31-. .28 ",'>./ A „. ^^^ • 60' \ (- ' <* / / /^'--'^y .48 .43 .32 .27 --'/.|5 -""Z /.' .60' .49 .42'^^ .33 .26 .|7 .|4 -3 ^N ---' ~'^\. .59 _J'50__.4I .34 .25 .|8 °I3 ".4"'' ^ ^ ^ "^ - ... ^ - .58 ^l5J---»40 .35 .24 .|9-.°I2 °5/~- ''"' ""^^---^ ^^^---'' 40*- /.57 .52 .39 .36 °23 '"'.20-^°ll "6 ,200 -^SB -- J\ ^j" ^~ '~-~~ -o'^'rx-'^-:: 500 ,1 ^^J 1 ,-^^-''~i-'^V^'^^~"' . , IPOO METERS 1 1 1 f\^» 1 1 1 1 ' 7'0* ' ' ' 4r -40* Figure 1. — Location of stations at which bottom samples were collected for determining the distribution of the remains of marine animals. Isobaths are indicated by dashed lines. WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 1. — Station location, water depth, sediment type, ime of bottom samples collected south of Martha's d, Mass., 11-20 June 1962. Lot Long Water Sediment Sample N W depth type volume m titers I 40''58 69°30' 46 Sand & gravel 2 2 A-d°6]' 69°31' 46 Sand 41/2 3 40°40 69°31' 51 Sand 23/4 4 40° 30 69° 29' 62 Sand 63/4 5 40°21 69° 30' 76 Sand 33/4 6 40° 10' 69°31' 91 Sand 53/4 8 39°57' 69°30' 183 Sand 43/4 9 39°56 69° 45' 201 Sand 33/4 10 40°O0' 69°45' 139 Silty sand 4'/4 11 40° 10' 69°45' 95 Sllty sand 4'/4 12 40°2O' 69°46' 79 Sand 61/4 13 40°30' 69° 45' 73 Sand 93/4 14 40° 40' 69° 45' 59 Sand 41/2 15 40°50 69°45' 37 Sand 6 16 40°46 70°OO' 38 Sand 23/4 17 40°39' 69°59' 49 Sand 4% 18 40°30' 70°O0' 73 Sand 113/4 19 40° 20' 69° 59' 91 Sand 7 20 40° 10' 70° 00' 117 Sand 3/4 21 40° 00' 70°00' 165 Sand 101/4 22 40°03' 70° 15' 183 Silty sand 93/4 23 40° 10' 70° 15' 113 Silty sand 123/4 24 40° 20' 70° 15' 90 Silty sand 143/4 25 40° 30' 70° 15' 70 Sand 14 26 40° 40' 70° 15' 51 Sand 10 27 40°50' 70° 15' 44 Sand 91/4 28 41°00' 70° 15' 33 Sand 73/4 29 41°11' 70° 16' 27 Sand 43/4 30 41° 10' 70° 30' 38 Sand 10 31 4rOO' 70° 30' 48 Sand 93/4 32 4O°50' 70° 30' 59 Sand 141/4 33 40°40' 70°30' 62 Silty sand 143A 34 40°30' 70° 30' 73 Sandy silt 133/4 35 40° 20' 70° 30' 97 Sandy silt 103^ 36 40° 10' 70°30' 128 Silty sand 143/4 37 40°04' 70° 29' 220 Sand 11 1/2 38 40°02' 70° 44' 194 Silty sand 53/4 39 40° 10' 70°45' 132 Silty sand 143/4 40 40°20' 70° 46' 106 Sand-silt-clay 143/4 41 40°30' 70° 45' 79 Sandy silt 141/4 42 40° 40' 70°45' 66 Silty sand 71/2 43 40°50' 70°45' 55 Sand 934 44 41°0O' 70° 45' 51 Sand 73/4 45 4ri0' 70° 45' 38 Sand and gravel 5 46 41°I0' 7rO0' 40 Sand 6I/2 47 4 TOO' 71°00' 51 Sand and gravel 121/4 48 40°50' 7r00' 59 Sand 43/4 49 40° 40' 7 TOO' 70 Sandy silt 143^ 50 40°3O' 71°O0' 84 Clayey silt 14 51 40°21' 7roo' 99 Sandy silt 1434 52 40° 10' 7roo' 146 Silty sand 43/4 53 40°06' 7 TOO' 179 Silty sand 113/4 54 39°59' 71°00' 366 Silt 1I3^ 55 39°56' 71°00' 567 Silt 10 56 40°03 7ri6' 183 Sand 103/4 57 40° 10 7ri5' no Silty sand 73/4 58 40° 20' 7ri5' 91 Silty sand 141/2 59 40°30' 7ri5' 77 Silty sand 141/4 60 40°40 71° 15' 62 Sand 141/2 61 40°50' 71° 15' 62 Sand 123/4 62 41°01' 71° 16' 48 Sand 3/4 63 41°10 71°15' 38 Sand 43A mineral matter and associated debris were re- moved by hand sorting, and the animals and an- imal remains were separated by species, identi- fied, and counted. Only animal remains are considered in the present report. Water depths at which samples were collected ranged from 27 to 567 m. Sediment samples were collected at each sta- tion and at two localities equally spaced between stations along the cruise track. Of the 186 sam- ples collected, 60 were analyzed in detail for par- ticle size, and the remaining 126 were examined in the laboratory by field techniques. Names of the various sediment types are in accordance with the classification reported by Shepard (1954) and Emery (1960). DESCRIPTION OF THE AREA Three major physical features that have an important impact on the occurrence, distribu- tion, and condition of the prefossil animal re- mains in this area are: physiography, bottom sediment composition, and hydrography. These features are briefly discussed below. PHYSIOGRAPHY The area studied is about 130 km square and extends across the continental shelf and the up- per portion of the continental slope. Bottom configuration is moderately smooth; water depths increase gradually and rather uniformly from shore outward to the shelf break, which is at a depth of about 120 m. Beyond the shelf break, on the continental slope, the depth gradi- ent is relatively steep, averaging 4°. Detailed bathymetric charts of this area having contour intervals of 1 fathom were published in 1967 by the U.S. Department of Commerce and U.S. Department of the Interior (Coast and Geodetic Survey, Bathymetric Maps numbers: 0708N-52 and 53; 0808N-51 and 52; and 0807N-51). BOTTOM SEDIMENT COMPOSITION Bottom sediments in the area are composed of relict glacial material — principally nonbio- genic sands and silts plus a few gravel patches FISHERY BULLETIN: VOL. 71, NO. 1 AO" NANTUCKET 41" Mo; ^J^llMi:: 40° .BLOCK IS ,- MARTHA'S VINEYARD _ '.•:S:-:-^^iv: ■Iv/iir.v!!-! ,-^ . ^..--x-xvx-x':-:-./ A .•.■.•.•.■.v.v.vf.-.v.v' l_ ;.;.-.;.. ^V .;.;.;.;.-. w. ;.;.;.;. v.v Q. . . . .\, . . .g . p: v.".vav.'.v,-.v.-.'.w.v 80. jfSv/.v.v.;. :::':^ii - 500 ■.;.;.;.;.;*.;.;.;.; • vi/XylvX ipoo METERS BOTTOM SEDIMENTS ^aORAVEL-SANO ^3 SANDY SILT [JS3SAND [IZ]SANO-SILT-CLAY CZHSILTY SAND JMlSILT 40° 1 1 tHS 1 1 1 1 1 Tlni — I 1 1 1 7'l 7'0» Figure 2. — Distribution of the various types of bottom sediments in the study area. Terminology is based on the classification reported by Shepard (1954) and Emery (1960). of glacial erratics. Six major sediment types occur in the area (Figure 2). The terminology used is based on the standard Wentworth par- ticle size classification (Twenhofel and Tyler, 1941; and others) and the nomenclature is that of Shepard (1954) and Emery (1960). Three types — sand, silty sand, and sandy silt — are dis- tributed over a rather large area; the other three — gravel-sand, sand-silt-clay, and silt — have limited areal distributions. Sands cover more than half of the area. They occur mainly in shallow water (less than 60 to 80 m), except in the eastern sector and in a narrow (6 km) band parallel to the isobaths just below the outer periphery of the continental shelf. Sands and silts in the vicinity of the shelf break are primar- ily glauconitic. In shallow waters near Nan- tucket and Martha's Vineyard and in the vicinity of Nantucket Shoals, the sands are silt free and occasionally mixed with large quantities of shell. Mixtures of sand and gravel also occur in scat- tered patches in the shallower waters of the northwest sector and in Nantucket Shoals. Li- monitic pellets and sand particles heavily stained with iron oxide are common in the northwest sector. Admixtures of silt occur with the sand over most of the remaining area. A large (80 by 100 km) area of fine-grained sediments is situated in the southwestern sector. A relatively small circular area of sand-silt-clay near its center is surrounded by an inner band of sandy silt and an outer band of silty sand. Characteristically, the relatively large sand grains throughout the area of fine-textured sed- iments are frosted rounded quartz particles. Pyrite-filled foraminiferal tests occur in the east- ern portion. On the continental slope below the sand zone, the dominant sediment component is silt. Additional information concerning sediments of this area and references to the geological lit- erature were given by Uchupi (1963), Wigley and Mclntyre (1964), Emery, Merrill, and Trumbull (1965), Emery (1966), Garrison and McMaster (1966), McMaster and Garrison (1966), and Wigley and Emery (1967). HYDROGRAPHY Within the area the water temperature regime is typically warm-temperate, although the bo- real influence is seasonally significant. Surface temperatures are substantially higher than bot- tom temperatures; off'shore surface waters are somewhat warmer than inshore waters through- out most of the year; temperatures of the entire water column change seasonally and to some ex- tent from year to year. Most pertinent to the subject of this report are bottom water temper- atures and nontidal currents. A cell of cold (6.6°C in June 1962) bottom water extends in an east-west band from the New York region eastward to long 70°W (east- ern Nantucket Island). This cell occurs at depths of about 40 to 80 m, which is roughly the midshelf region. At 300 to 600 m the bottom water temperatures are low and nearly constant throughout the year; they generally range be- tween 4° and 7°C. Near the shelf break and upper continental slope the bottom temperatures are substantially higher, but also nearly con- stant; values range near 10° to 12°C through- out the year. Offshore shelf waters, especially in shallow sectors, may range from 3°C in Feb- VVIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS ruary-March to 14°C in September-November. Temperatures of inshore surface waters sub- stantially exceed these values, Nontidal movements of water masses on the continental shelf within the area are generally westward. Water in the Gulf of Maine and Nantucket Sound tends to flow southwesterly across Nantucket Shoals and into the area. Con- versely, surface waters offshore beyond the continental shelf flow easterly. Authors who have published further information on the hy- drography of the area include Bigelow (1927, 1933), Bumpus and Day (1957), Bumpus et al (1957), Day (1958), and Colton (1964, 1968, 1969). ORDER OF DISCUSSION The most common animal remains in the sam- ples studied were echinoderms, mollusks, and fish. Considerably less common than the fore- going were remains of crustaceans and coelen- terates. The order in which these groups are discussed below is according to the abundance of remains in each major group, namely, echi- noderms, mollusks, fish, and crustaceans and coe- lenterates. REMAINS OF ECHINODERMS Echinoderms were the most numerous and quantitatively dominant group of animal re- mains occurring in the area. The sole contrib- utors in this group were the echinoids. Spines and test fragments were rare to very abundant and were widely distributed. Presumably the skeletal fragments of asteroids and ophiuroids, of which living members of both groups are common in this region, were generally too small to be recovered using the 1-mm mesh screen. Of all macroscopic animals in the samples, the common sand dollar, Echinarachnius parma (discussed in the following subsection), was by far the leading component. Spines were the principal structures recovered from heart ur- chins and sea urchins. Some examples of typical echinoderm remains are illustrated in Figure 3. The size of fragments of most organisms dis- cussed in this section ranged from 1 mm (sand size) to 1 cm or more. The largest remains were tests of whole or nearly whole E. pai-ma. Adult size of living members of this species (about 7 cm) is less than some of the other non- molluskan species, but the comparatively strong, compact test is much more resistant to fracture. This durability, plus the enormous supply in the form of living individuals, contributed to the abundance of fragments of this species in the sediments. Counting the E. parma and other echinoids was impractical owing to the enormous numbers of small fragments, plus a gradation in size that precluded the separation of major fractions from minor ones. Occurrence of Brisaster fragilis, Echinarachnius parma, and Strongylocentrotus drobachiensis are listed by stations in Table 2. ECHINARACHNIUS PARMA Remains of E. parma were widespread (Table 2) and numerous. This species ranked first in volume and number of fragments of all organic remains in the study area; it occurred at 73 Cr of the stations. It was most abundant in the vicinity of Nantucket Shoals (stations 2, 3, and 16). E. parma fragments made up nearly 30^ (by volume) of the substrate near station 3. In deepwater areas in the vicinity of the middle and outer shelf, the density of fragments was low — occasionally less than 50/m= or about 0.01 '"r by volume. (All animal remains combined gener- ally formed less than 1% by volume of the sub- strates of the outer shelf and slope.) The distribution of E. parma extended from the shallow inshore areas across the entire shelf to the upper portion of the continental slope (Figure 4). Surprisingly, it was rather sparse near the middle of the shelf. Fragments from inshore areas were diff'erent in size, color, and sphericity from those collected oflfshore. Test fragments from the inshore zone, which extends out to 50 or 70 m, were whitish, usually larger than 5 mm in greatest dimension, and had sharp and angular edges and apexes (Figure 3A). Fragments from depths of about 80 m or more were greenish-brown, commonly less than 5 mm long, and had rounded edges. In contrast to the fresh, new appearance of the test fragments FISHERY BULLETIN: VOL. 71, NO. 1 A -H Figure 3. — Skeletal remains of echinoderms. A - E chinarachnius parma, test remains from shallow water; B - £". parma, test remains from deep water; C - Brisaster fragilis, test fragments and spines; D - Stron- gylocentrotus drobachiensis, test fragments and spines. Each scale bar is 5 mm. 6 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 2. — Occurrence, by station, of the remains of three species of echinoderms. Present ( + ), absent ( — ). Station number E china rachntus parma Brisaster jragilis^ Strongylo- ctntrotus drobachensis I + — + 2 + — — 3 + — — 4 + — — 5 + — — 6 + — — 8 — + — 9 + — + 10 + — — 11 + + — 13 + — — 14 + — — 15 + — — 16 + — + 17 + — — 18 + — — 20 — + — 21 + + — 22 + + — 23 + + — 24 + + — 26 + — — 27 + — — 28 + — — 29 + — — 30 + — — 31 + — — 33 + — — 35 + + — 36 + + — 37 — + — 38 + + + 39 — + — 40 + + — 41 + — — 43 + — — 44 + — — 45 + — — 46 + — — 47 + — — 49 + — — 51 + + + 52 + + + 53 + + — 54 — + — 56 + + — 57 + — — 61 + — — 62 + — — 63 + — — 1 May include some frogments of Echinocardium or Schizasttr. from the inshore zone, shells from deep water appeared old and worn (Figure 3B). The char- acteristics of specimens from the area between the two depth zones were intermediate. OTHER SPECIES Brisaster fragilis (Figure 3C) was distrib- uted in a broad east-west band along the outer 7.I' I 41* .BLOCK ISLAND -, '■ ~40 - '~. . 60 40'-' , 7,0' J^— ' L_r4 , - UABTHA'S UIWEVABD NANTUCKET U II / ' \ » '.-> * '"-V / ' ' -41* IPOO METERS ECHINARACHNIUS PARMA k::::] light color, larger size, angular edges f^ intermediate fz/i dark color, smaller size, rounded edges T jV^i I I T- 7'0' trS I I" 40* Figure 4. — Geographic distribution of remains of tests of Echinarachnius parma. Three categories of size, color, and sphericity are shown separately. continental shelf and upper slope (Figure 5). The fragments were generally small and scat- tered. Even when all species are considered as a group, only a few spines or test fragments oc- curred in any one sample. Range in water depth for this echinoid was 90 to 366 m. Water depth range is compared with that of other species of echinoderms in Table 3 and illustrated in Fig- ure 6. The green sea urchin, Strongylocentrotus drobachiensis (Figure 3D), was represented chiefly by spines and less commonly by test frag- ments. The remains were rather widely scat- tered among six stations; two were in the Table 3. — BathjTnetric distribution of three species of echinoderms and the number of stations at which each occurred. Species Water deptti Number of M nimum Maximum Mean stotions m m m Brisaster jragilis' 90 366 155 18 Echinarachnius parma 27 201 34 45 Strongylocentrotus drobachiensis 38 201 121 6 1 May include some spines of Echinocardium and Schizaitfr. FISHERY BULLETIN: VOL. 71, NO. 1 7 1* 7n® 1 I I 'i* 1 1 II 1 'I*-* 1 1 1 1 Wm; ''-'; .'^"-V^^^i NANTUCKET Cff^^^. , k 4I«- .BLOCK .'' •'*«^"*'S ^"^EYARO -.«4«iP,; I ISLAND ' n,'^' „ "-,-"// ' , ' -60,' ,, \ --, / o _o_ O O O O O ' 0ooooosA$S5x5$^yx^ ^^99888888888888888^^ 40'- ^°° " ^ rjl'\^-^ A'---'^~— --' ~ ^ IpOO METERS BRISASTER FRAGILIS 1 1 1 7l|o 1 1 1 1 ' J'O' ' ' ' ' r^ 7,1 Jj I - I L. 7,0° 4I°- 40' 20 .BLOCK ISLAND MARTHA'S VINEYARD NANTUCKET 40 60 <;--^'^ '''-'''' $^f 80 100 200 500 o -- ^ o 'M 1.000 METERS S r/?0/VG Y LOG EN TRO TUS DROBACHIENSIS T 1 TTTi 1 1 1 1 T T'rii 1 1 1 ' 41* 40" 7'l 7'0° Figure 5. — Geographic distribution of skeletal remains of the echinoderms Brisaster fragilis and Strongylocen- trotus drobachiensis. WATER DEPTH (METERS) SPECIES 50 100 150 200 250 Echinarachnius parma Sfrongylocentrotus drobachiensis Bnsasfer fragilis i_ ^TO 366 • Figure 6. — Bathymetric range and mean depth of oc- currence of echinoderm remains. (Mean values are listed in Table 3.) shallow waters of Nantucket Shoals and the other four were in moderate to deep water near the shelf break (Figure 5). Bathymetric range at the locations where this species was found was from 38 to 201 m (Table 3, Figure 6). REMAINS OF MOLLUSKS Remains of mollusks were among the most common organic seabed constituents. In total abundance they ranked second ; only the echi- noderms were more plentiful. Four major groups of mollusks were represented in the material analyzed. Pelecypods were the most abundant molluscan group, gastropods ranked second, and the cephalopods and scaphopods were present in relatively small quantities. These groups are discussed below in the order of their abundance. PELECYPODS Pelecypod shells were abundant and conspic- uous components of the prefossil animal re- mains. In addition to their relatively large size, often the color and texture of the shell surface contrasted sharply with the sediments in which they occurred. Size of the shells ranged from such large, robust species as Spisula solidissima (12 cm), Arctica islandica (10 cm), and Placo- pecten magellaniciis (10 cm) , to such small, frag- ile forms as Thyasira gouldi, Nucula proxima, Bathyarca pectuvadoides, and others, all of which were 5 mm or less. A total of 57 species representing 40 genera were collected. Typical species are illustrated in Figure 7. Pelecypod remains were very widespread; they were col- lected at all stations except three (3, 47, and 62) . 8 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Figure 7. — Representative pelecypods from off southeastern Massachusetts. A - Arctica islandica (X0.7); B - Astarte subequilatera (XI); C - Astarte 2(ndata (X2); D - Cerastodervia pinnulatum (X2.6) ; E - Crenella glandida (X4) ; F - Modiolus modiolus (X0.7) ; G - Nucula proxima (X5) ; H - Nucula proxima, interior (X5) ; I - Nucula tenuis (X3.3) ; J - Nucula tenuis, interior (X3.3); K - Nuculana acuta (X4); L - Phacoides filosus (X2.6) ; M - Periploma papyracea (X2.6) ; N - Periploma papyracea, interior (X2.6); O - Placopecten magellanicus, left valve (X0.7); P - Placopecten magellanicus, right valve (XI); Q - Thyasira trviimiata (X3.3); R - Venericardia borealis (X1.3); S - Yoldia sapotilla (X2) ; T - Yoldia sapotilla, interior (X2). FISHERY BULLETIN: VOL. 71, NO. I Table 4. — Species and density (number per square meter), by station, of the comiaon pelecypods. s in = It V E u 01 1 s ■s a •s 3 3 C ij P « s c (n o c CO o o x> ■o « < u a < < < < s, o o o c3 "1 B a id 3 0) z: £ :^ z ^ z (0 a a i £ 2 o. (O i 1 e g ^ o B 3 I 20 2 4 5 6 30 8 30 9 70 10 10 11 12 13 - 14 - 15 16 10 17 18 19 20 - 21 22 40 23 24 25 26 27 28 - 29 30 - 31 32 33 34 35 - 36 37 38 - 39 - 40 - 41 42 43 - 44 45 - 46 48 49 50 - 51 - 52 20 53 20 54 - 55 56 80 57 58 - 59 - 60 - 61 63 10 80 160 130 40 40 20 40 1,560 30 10 110 660 10 40 130 110 50 10 20 90 90 10 100 10 120 60 10 50 250 20 170 - 150 30 10 20 110 140 110 10 30 40 10 - 60 150 2,120 160 60 20 10 90 160 200 30 70 80 10 20 20 30 30 10 30 10 760 20 70 2,080 40 50 50 290 10 10 50 280 40 170 10 30 10 150 50 80 10 70 30 40 10 30 10 120 10 10 20 550 760 80 20 60 10 50 250 40 - 1,360 230 - 1,360 250 10 80 - 190 - 240 40 10 30 20 30 40 - - 90 - 10 10 10 10 230 50 60 10 40 10 80 1,160 70 20 50 50 160 150 160 70 90 120 20 20 - 350 40 - 360 30 10 10 870 220 430 20 190 20 30 20 30 90 60 - 20 20 20 140 140 10 30 10 20 10 40 70 160 150 10 390 20 170 90 30 30 10 10 30 90 20 10 10 - 120 10 10 360 - 150 - 250 20 50 10 10 30 30 20 70 240 90 10 130 330 40 220 10 80 110 20 30 50 90 90 70 60 30 60 490 10 90 20 30 10 10 130 60 30 920 1,340 210 430 - 210 20 50 30 20 60 20 100 40 10 20 20 10 10 220 1,050 1,000 10 20 30 10 10 20 20 - 70 20 40 20 10 70 440 300 300 100 20 40 30 20 100 20 10 20 20 300 980 90 110 20 110 30 - 800 20 20 280 70 120 200 430 30 70 10 20 - 60 90 10 90 310 3,780 290 460 10 350 140 10 20 30 880 800 1,440 150 50 390 10 70 10 240 - 80 - 10 80 - 130 20 - 10 30 - 170 10 50 80 - 180 160 20 10 160 500 10 10 - 20 490 30 30 - 50 - 120 80 10 10 30 50 - 70 80 240 4,600 40 630 200 40 240 30 30 80 10 10 80 120 30 140 80 70 60 Members of this group were by far the most abundant mollusks. Densities were as high as 8,230/m- (all species combined). The occur- rence records of pelecypods are presented in two tables: Table 4 gives the number of shells per square meter, by stations, for the 35 more com- mon species and Table 5 lists the number of shells per square meter for each of the 22 species that occurred at only one station. Each pelecy- pod shell was counted separately. No attempt was made to enumerate the left and right valves separately, because of the fragmentary nature of many specimens. Distribution and Density Pelecypods, all species considered, were gen- erally most abundant in a band extending north- east-southwest across the study area and in another narrower band parallel to the depth con- tours near the shelf break (Figure 8) . Densities were frequently less in the northwest, northeast, and south-central sections of the continental shelf and along the continental slope. As ex- pected, the density of the group was strongly influenced by a relatively few species that were both abundant and widely distributed. 10 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 5. — Species and density, by stations, of pelecypods that occurred at only one station each. Species Station Specimens Abra tongicallis Aequipectfn gtyptus Anadara ovalis Axinopsis orbiculata Bathyarca anomala Crenetla ptctinula Cuspidaria striata Cyrtodaria siligua Liocyma jluctuosa Macoma balthica Modiolus dftnissus Myonrra limatula Mytilus edulis Nucula dflphinodonta Nuculana tenuisulcata Panomya arctica Ptriploma afjinis Siliqua costata Solimya velum Tellina agilis Thracia conradi Thracia myopsis 2) 10 16 8 56 33 8 30 30 6 46 3 1 11 38 25 4 17 4 29 13 5 Nolrrfi 10 10 10 50 80 20 20 20 10 10 10 10 10 10 10 10 20 20 10 10 40 20 41' 40'- -'0° T 200 ipq^^ii: 500 ipOO METERS t -. _n ^ '■:>.»-.-- '^ EZI3 0-1,000 PELECYPODA NUMBER PER M* F=^ lOOO-SpOO T771 OVER jpoo T 1 r T\' ro* Figure 8. — Density distribution of pelecypod shells, all species combined. Pelecypods were sparse (less than 50/m^) or absent at 8 stations, common (50 to 1,000/m-) at 32 stations, abundant (1,000 to 3,000/m-) at 16 stations, and very abundant (more than 3,000/m-) at 6 stations. Pelecypod shells were present at all depths sampled (Table 6). Densities were lowest (30 to 40/m-) at both the shallowest and deepest stations; moderate (100 to 1,100 'm-) between 30 and 89 m and between 200 and 249 m; high (greater than l,100/m2) between 125 and 199 m; and highest (more than 2,000/m-) from 90 to 124 m. Table 6. — Density distribution of pelecypod shells, all species combined, in relation to water depth. Water depth class Samples collected Samples containing pelecypod shells Mean number of shells Mettrs 20-29 30-39 40-49 50-59 60-69 70-79 &0-89 90-99 100-124 125-149 150-174 175-199 200-249 250-567 Number 1 6 7 8 5 9 1 7 4 4 I 5 2 2 Percent 100 100 86 63 100 100 1O0 100 100 100 100 100 100 100 No/m^ 30 140 440 320 670 1,030 620 2,250 3,080 1,110 1,460 1,970 690 40 Relations of Density to Sediments Pelecypods were generally more abundant in moderately fine-grained sediments than in either coarse or very fine types. Silty sand, sandy silt, and sand-silt-clay were most commonly associ- ated with high density. Average shell density in these three substrate types ranged from 1,800 to 3,300/m-. Shells were absent or sparse in gravel-sand substrates (average density 80 /m=) and silts (average density 40/m-), and moder- ately low (average 650/m-) in sand. Distribution and Density by Species Geographic distributions of the 35 more com- mon pelecypod species are illustrated in Figure 9. These charts are based on information listed in Table 4. No two species had identical distri- butions, but the distribution of a number of spe- cies in east-west bands across the study area sug- gests correlations with hydrographic features or bottom sediments, or both. 11 FISHERY BULLETIN: VOL. 71, NO. 1 MARTHA'S VINETARD &STARTE CASTANEA , I ISL4MD '_ ^;'' ^/ _ ipOO ME Tens ASTARTE SUBEQUILATERA rrir*- rr>^ ,JLOC« " ""'"A-S VlNSr.BO BATHYARCA PECTUNCULOIDES wir*- IpOO METEftS CERASTODERMA PINNULATUM rrirV" ..LOW ,•- »•"'"'■= "«•""" llSLftND'^-,' .40*."W /"' ipOO METERS CRENELLA DECUSSATA 20 MARTHA'S vmEVARO . ^M^ 1 ISLAND '__^;^ • ,., • r--.'.''' . ; ^ \ ■40 -' ~-._ .-- .--' ~~. ,-:' ■' 1 / }:■ • '-'. X'-' ';-' ^ '■■' 60 ^ ,, • _: • \ '-'- i^i^ • — • B --V , '•'-■i.-' BO .-' :-■--'["'•'"■' ■-.;_ •,/- ,200 500 ^^^^ -.-:-^ . ipOO METERS CUSPIDARIA PERROSTRATA , ■^rr; — p ^ 1 1 \ — n;^ ,,,,'- rr«fT^ -'■ '.^'^ - MARTHA'S VINErARD 1.000 METERS /' ;. --'/ CYCLOPECTEN THALLASSINUS m^^^ .'-- >^' ' ' ' ' ,' :, '.' W -r ^^^j NANTUCKET 20 .BLOCK [island -^.; MARTHA'S VINErARO ~,*^B|^^,'< 40 - ' ' 60' :#' V 1>5^ '"-~"-»>0Ok #5 ' oJ'" 90 , w^' ^^ •'"■••- V-:v-L:~^!'r^ _ipoo MET ens E/VS/S DIRECTUS Figure 9. — Geographic distribution of the common pelecypods. 12 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS r*->- .BLOCK _ I ISLAND '^^,* lATHA'S VINEVARO /•,-■• , IPOO METERS LIMATULA SUBAURICULATA t"' "^L* /"T' ■' -BLOCK MMTHA-S ViNETftRD ^ .^^/j^,' 60 :.---. • ''\M'--''I^ . >^.^ • . . • «C J ^. -.-.---. . .. '00 / J"- ' '/-■--■■-=■■ • ^"■'"'■ ioo , ._, , 400 " --------•■"--''.''-■- -:-\*?^.' --.'---:-• . IPOO METERS LIMOPSIS SULCATA ..LOO. ,•■ ""'"*•» "«•"" ;«• •0 ' --_ . eo; ^_,, 1 •' • • ' • ^ [ ■BO.-'' ^ - , i .wo' /-'"~ "'--•- .--'*"'*--*\ .!00'„ -»,.-., ,'' .^'".^ soo "' ' _ ipOO METE"S LYONSIA HYALINA ' 7'l' T-O* ' ' m^rir^ '-' >^ — ' ' 1 l-r— ' Wf '- .^Li NANTuc-ET 20 MARTHA'S V^NE'ARO .*^B^ < [iSLftND '^,; --'._' ' '''=-■:■■: : /i. eo," ^_,, • • • \%f t- , ,,-'"■ '■' • -- • • -w • • . ' ,-- AflL ""■--- ' * ' ' i l:l^'--_v • v • . #- # , 'oo' ^/-'' - -'~ ,. • - .?00 .^ -'»■—--. .-' ■- -'' '-"-Vc-- ';i:l:::. _ IpOO METEOS MESODESMA ARCTATUM -4I»- „ „, .- MARTHA'S V'NeraRD . ^BM - ^ >-•< «.-;' . .V"' -- — , . , 40*- . t.OOO METERS MODIOLUS MODIOLUS ' ' 7l|- Tf ' ' ' wrtr'^ A'S ViSE'ARO MUSCULUS NIGER rrif^*~ ,- MARTHA'S vtNE^aflO NUCULA PROXIMA ^^W^W . .^.^ > ','. 4 NANTUCXET 20 -BLOCK MARTHA'S VINETARO -TMB^.-t :' \ " ■40-' '-_ -'"._ '-, ' '''':-^' Vi, • • ' '-\ ' '■:■■''■'' "*■' -.-- i^ k ^-^ •# V-- Z: :■# ,200 - SOO"' ' ixrVx--'-'' , ipOO METERS /VUCUL4 TENUIS T'l* ' ' T'O- ' • rrtr*- MARTHA'S ViKETiRO NUCULANA ACUTA w 20 ■^;-^-^ , «A«Tuc«tT ; 'v,;, MMTHA'S VINE' "■» '■•«•,• [isLSNo ■;-; 40 ' ' - - -----:XX -. ' 'V^}> / /I :* "^'^'""•"^^i-^ ■0 .'[ -•^Wv W 1 ^^'5 ': ^ ? •" - -\^? .500 ' ' --——.-■ •-:V;^^cl:5s^iC^ , IPOO METERS PANDORA G0ULD//3W/1 ' T'C ' T-O* • MJ.it-' '--••^^i Hp '-■;-'"--^^ NAUTUCET ■20 .BLOCK ,'' MARTHA'S VINE'ARD --U \ ISLAND ', -./ . • '->■ . i? ■40-' '-_. ----'' i\^ '-:■•.■- / . Bo; _^_, • :# X ^5^;v J,-' --■''' "' .'- « ^-. "■ ^ --' ^^^ _-•---.' • • • " <'>4 kJc ~-- '"^--''' /• . -,-.,. . ,IO0' fJ''' -"--..-_•- --"s "' ,200 ' -^,— --''''- ~r"~- -•-''V-*''' _ 1,000 METERS PERIPLOMA LEANUM Figure 9. — Geographic distribution of the common pelecypods. — Continued. 13 FISHERY BULLETIN: VOL. 71, NO. I p>*^ _ ( ISL'MD \ 40-' '■ ,' MMTHA'S ViNETkWD .,« X , ipOO METERS PHACOIDES BLAKEANSIS rF>n*- llSLAMO PHACOIDES FILOSUS . . . T,l' 7.0" , . . , BLO» ,-' ''**'''***'S VINEY4B0 ~.«9[9,< 1 #-— -: "■-'■•""• ■■' j 40* , IPOO METERS P/rAff MORRHUANA 1 1 1 7I,, 1 1 1 1 1 ylQ. , 1 . . rF>'*^ ,' MARTHA'S VINEYI NANTUCHET , , •\ : ^\~-^' ■■:■ •M-- "» r- 5^ THYASIRA GOULDI J.lf^t" '-r ', >«R0 _ «|^ C«£T Ik. ; /V 60 ^OO InnrSc ' 1 1 ^K ^ • '' 80 OOG ! «OC ^ Sfevl ,J00 ' , __ 500'' ' _ ipOO METERS YOLDIA SAPOTILLA Figure 9. — Geographic distribution of the common pelecypods. — Continued. 14 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS The 10 most widely distributed species, in de- creasing order, were: Yoldia sapotilla, Ceras- toderma pinmdatum, Astarte undata, Thyasira trisinnata, Venericardia borealis, Arctica islan- dica, Placopecten magellaniciis, Phacoides filos- us, Crenella glandula, and Nucula proxima. All of these species inhabited rather broad east-west zones across the study area, except for Nucula proxima, which was absent in shallow water in the western sector. Its distribution was alig-ned in the north-south direction, and to some extent east-west. Bathymetric distributions differed greatly among various pelecypod species. Depth range in which each species was found is listed in Table 7, and data for the most common species are plotted in Figure 10. Species dispersed over the widest depth range (38 to 567 m) were Astarte imdata and Placopecten magellanicus. Depth ranges of 21 species were very narrow, but nearly all of these were based on few col- lections. Most of the species that occurred in two or more collections were taken over rather broad depth ranges. Species found in shallow water (less than 50 m) were: Anadara ovalis, Cyrtodaria si- liqua, Liocyma fhictuosa, Lyonsia hyalina, Modi- olus demissus, Mytilus edulis, Siliqua costata, and Tellina agilis. Species found in deep water (taken at depths greater than 200 m) were: Anomia aculeata, Astarte subequilatera, A. undata, Cerastoderyna pinnulatum, Limatula subauriculata, Nucula proxima, N. temtis, Nuculana acuta, Phacoides blakeansis, P. filosus, Placopecten magellanicus, Thyasira plana, T. trisinuata, Vene7'ica7^dia bo- realis, and Yoldia sapotilla. Density of shells of individual pelecypod spe- cies ranged from 10/m- to 4,600 /m- (Table 4). Densities tended to be high for the more widely distributed species and low for species with a re- stricted geographic distribution. Species found in greatest density were: Venericardia borealis — 4,600/m-, Arctica islandica — 2,080/m-, As- tarte subequilatera — 1,560/m-, Nucula proxima — 1,360/m-, Asta^'te undata — 1,440/m-, and Thyasira trisinuata — 1,160/m^ All of these, except Astarte subequilatera, were among the 10 species with the widest geographic distribu- Table 7. — Bathymetric distributions of 57 species of pelecypods and the number of stations at which each occurred. Species Water depth Number Minimum Maximum Mean ot stations m HI m Jhra longicallis 165 165 165 1 Aequipecten glyptus 139 139 139 1 Anadara ovalis 38 38 38 1 Anomia aculeata 38 201 139 10 Arctica islandica 44 110 75 27 Astarte castanea 38 97 64 6 Astarte subequilatera 62 366 152 15 Astarte undata 38 567 123 36 Axinopsis orbiculata 183 183 183 1 Bathyarca anomala 133 183 183 1 Bathyarca pectunculoides 128 194 164 7 Cerastoderma pinnulatum 38 220 96 38 Crenella decussata 73 84 77 3 Crenella glandula 46 194 99 20 Crenella pectinula 62 62 62 1 Cuspidaria perrostrata 146 194 177 5 Cuspidaria striata 183 183 183 1 Cyclopecten thallassinus 91 183 156 5 Cyrtodaria siliqua 38 38 38 1 Ensis directus 51 62 56 3 Limatula subauriculata 165 201 183 3 Limopsis sulcata 99 146 124 4 Liocyma jluctuosa 38 38 38 1 Lyonsia hyalina 27 49 38 2 Macoma balthica 91 91 91 1 Mesodesma arctatum 91 113 101 3 Modiolus demissus 40 40 40 1 Modiolus modiolus 38 146 68 5 Musculus niger 33 62 52 3 Myonera limatula 183 183 183 1 Mytilus edulis 46 46 46 1 Nucula delphinodonta 95 95 95 1 Nucula proxima 33 220 S3 19 Nucula tenuis 44 220 68 10 Nuculana acuta 91 366 161 17 Nuculana tenuisulcata 194 194 194 1 Pandora gouldiana 51 110 88 7 Panomya arctica 70 70 70 1 Periploma ajjinis 62 62 62 1 Periploma leanum 48 91 70 2 Periploma papyracea 44 106 77 19 Phacoides blakeansis 62 220 126 7 Phacoides filosus 44 201 122 22 Pitar morrhuana 38 139 88 4 Placopecten magellanicus 38 567 116 25 Siliqua costata 49 49 49 1 Solemya velum 62 62 62 1 Spisula solidissima 37 9'1 62 7 Tellina agilis 27 27 27 1 Thracia conradi 73 73 73 1 Thracia myopsis 76 76 76 1 Thyasira gouldi 84 179 113 4 Thyasira ovata 62 99 77 5 Thyasira plana 106 201 171 4 Thyasira trisinuata 48 220 100 36 Venericardia borealis 38 366 116 33 Yoldia sapotilla 33 220 96 38 tion. Among the species collected at 10 stations or less, few occurred in densities greater than 15 FISHERY BULLETIN: VOL. 71, NO. I WATER DEPTH (METERS) SPECIES 50 100 150 200 250 Lyonsia hyalina L-,- Musculus niger 1 H Nucula proximo Yoldia sopotillo Cerostoderma pinnulatum Venericardia borealis , _j .TO 366 TO 567 Placopecten magellanicus TO 567 1 1 Spisula solidissima Astarte castanea l_ 1 L_ 1 Arctica islandica L J Periploma popyraceo Crenella glandule Nucula tenuis L J L 1 1 Thyasira trisinuata Periploma leanum Ens is di rectus L I rJ Pandora gouldiana Phacoides blakeansis 1 _i 1 Astarte subequilatera Thyasira ovata Crenella decussota 1 .TO 366 ^ Mesodesma arctatum L_ ^ Ttiyasira gouldi Cyclopecten tt^allassinus J 1 .TO 366 Limopsis sulcata Thyasira plana Bathyarca pectunculoides Cuspidaria perrostrata Limafula subauriculata 1 1 L 1 1 Figure 10. — Bathymetric range and mean depth of oc- currence of the common pelecypods. (Observed values are listed in Table 7.) 500/m^; the maximum density for most of these species was less than 100/m-. Exceptions to the direct relation between high density and wide distribution were two types: (1) species widely distributed, but present in low density, such as Cerastodetma pinnulatum, Crenella glandula, Placopecten magellanicus, and Yoldia sapotilla; and (2) geographically restricted species of relatively high local den- sities, such as Bathyarca pectunculoides and Thyasira ovata (shells of these two species oc- curred at only seven and five stations each, but densities were as high as 300 and 430/m-). Four patterns of geographic distribution re- vealed by these samples are: (1) Narrow band extending east-west across the area, such as: Bathyarca pectunculoides, Crenella decussata, Cuspidaria perrostrata, Cyclopecten thallassin- us, Limatula subauriculata, Mesodesma arctat- um, and Nuculana acuta. (2) Broad east-west band exemplified by: Arctica islandica, Peri- ploma papyracea, Placopecten magellanicus, and Thyasira trisinuata. (3) Encircling distribu- tion surrounding the center of the area, illus- trated by Astarte undata. (4) Wide inshore-off- shore distribution, as typified by: Anomia aculeata, Cerastoderma pinnulatum, Crenella glandula, Nucula proxima, Venericardia boreal- is, and Yoldia sapotilla. Hydrographic conditions and the type of bot- tom sediments appear to have a substantial in- fluence on the suitability of a habitat for some species of bivalves in this region. Unfortunately the common co-occurrence of fine-grained sedi- ments in areas of low energy and relatively stable water temperature, as opposed to coarse sediments in high-energy and changeable water temperature does not lend itself to an evaluation of the specific conditions that limit the occur- rence of the various species. Additionally, the presence of fossil shells invalidates a detailed evaluation of inferred habitat based on the pres- ence of shell remains. For example, the shells of Mesodesma arctatum from depths of 91 to 113 m probably are remains of populations that inhabited nearshore areas during the rapid rise in sea level of the post-Pleistocene period. Ra- diocarbon age determinations for shells collected in this region at depths between 86 to 130 m, studied by Emery and Garrison (1967), range from 10,850 ± 150 to 14,850 ± 250 years be- fore present. Species that occurred in moderately deep wa- ters and appeared to require a stenothermic ha- bitat were: Arctica islandica, Nuculana acuta, Thyasira plana, and Thyasira trisinuata (Fig- ure 9). Species that inhabited stenothermic 16 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS waters, but also showed special sediment require- ments, were: Bathyarca pectunculoides, Cuspi- daria perrostrata, Limatnla suhauriculata, and Thyasira gouldi. Conversely, those bivalves that occurred in eurythermic habitats (and coarse sediments) were: Ensis directus, Lyon- sia hyalina, and Musculiis niger. Relations of pelecypod distributions with bot- tom sediments are discussed below. Species-Sediment Relations Shells from 89 9^ of the 35 more common pele- cypod species represented in the area were found in several different sediment types; only 11% were associated exclusively with one sed- iment type. All of the common species were primarily in sediments in which sand or silt was the chief constituent. Species frequently taken in sand sediments were: Ensis directus, Limat- ula suhauriculata, Lyonsia hyalina, and Muscu- lus niger. Species most commonly found in silty sand and sand were: Bathyarca pectuncjiloides, Cuspidaria perrostrata, Cyclopecten thallasimis, Limopsis sulcata, Nucula proxima, N. tenuis, Nuciilana acuta. Pandora gouldiana, Phacoides filosus, Pitar morrhuana, Spisula solidissima. Thyasira plana, and Venericardia borealis. The only two species found mainly in sandy silt or silty sand were Limopsis sulcata and Thyasira gouldi. The absence of Astarte undata in the center of the area is probably due to the presence of fine-grained sediments there. All the remain- ing common species were collected from several different sediment types. The presence of a narrow band of sand ex- tending parallel to the depth contours near the outer margin of the shelf (Figure 2) appears to be a major feature affecting the distribution of many species having a narrow-band distribution (Figure 9). GASTROPODS Remains of gastropods formed an important component of organic origin, but compared with other mollusks they were far less common than pelecypods, but considerably more abundant than scaphopods and cephalopods. Gastropods were widely distributed throughout the area and ranged in density (all species combined) from to slightly over 1,000/m-. All remains were shells, except for one operculum of Polinices dupUcata. Forty-four species of gastropods were found in the samples. Some typical examples are shown in Figure 11. A large majority of spe- cimens were small, less than 1 cm in shell height. Some of the smallest specimens, averaging be- tween 2 and 5 mm, were Alvania carinata, Cyl- ichna alba, Retusa obtusa, and larval forms, pre- sumably of Thais. The larger species, averaging between 1 and 5 cm in greatest dimension, were: Buccinum undatum, Colus pygmaeu^, Crucib- ulum striatum, Crejridjila fornicata, and Nassar- ius trivittatus. Gastropod occurrence records are listed in Tables 8 and 9. Table 8 gives the species-station record for the 24 more common species. Table 9 lists the occurrence record for species taken at only one station. Distribution and Density Gastropod shells (all species combined) were rather widely distributed throughout the area, occurring at 80% of the stations. Highest con- centrations (250 to 1,050/m-) were in the cen- tral part of the area in a lens-shaped patch at depths between 40 and 80 m (Figure 12). Although gastropod shells were collected at all depths, the average concentration increased in each 10-m depth class from 20 m down to about 80 m (Table 10). Concentrations were lower in deeper water, except for a zone of greater density between 175 and 250 m. Alvania cari- nata and Cylichna goiddi made up the bulk of the gastropod remains in the shallowwater zone: Mitrella zonalis, a gastropod larva, and a variety of species accounted for the high abundance in the deepwater zone. Relations of Density to Sediments The density of gastropod shells as a group was related to sediment type only in a rather general way. No gastropod shells were found in the coarse sand or gravel-sand substrates. Concen- 17 FISHERY BULLETIN: VOL. 71, NO. 1 / *f f H Q 18 Figure 11. — Representative gastropods from oflf southeastern Massachusetts. A - Alvania carinata (X14) ; B - Coins pygmaeus (X1.5) ; C - Cylichna alba (X4) ; D - Cylichna gouldi (Xll-7) ; E - Drillia lissotropis (X5.5) ; F - Drillia sp. (X4.7) ; G - Epitonium dallianum (X3) ; H - Epitonium groenlandicum (X0.8) ; I - Mitrella zonalis (X7); J - Nassarius trivittatus (X2.3); K - Odostomia canaliculata (X?) ; L - Turbonilla interrupta (X4.7); M - Buccinum undatum (XI); N - Nep- tunea decemcostata (X0.8) ; - Eidimella smithi (X2.3) ; P - Polinices duplicata (XI) ; Q - Crepi- dula fomicata (XO-8); R - Crepidula fornicata (X0.8). WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 8. --Species and density (number per square meter), by station, of the common gastropods. « e ■U 3 n3 (0 to 4J j-i •H E u CO (15 -o D •H ■H ■u w c ^ 3 E cfl 3 M 4J 4J u •o QJ O (A U 0) C UH o •u D E E u c U CO 3 •H e >. r-( •—t w 3 a 3 3 •H u c TJ XI s: ■H •H m ■H o CJ CJ 3 a to .— 1 U r— 1 (U 3 c m 3 O M U < pq CP U U o O CO 3 Vi O o •H c u ■o c > o 0( -H (U N -1 s E t-( n 3 3 CO c « •H •H to r-4 x; •H c c 'H 1 — 1 o .—1 o o 4J at •H 1 — r 4J ■u CO i-i T— 1 ■H •H -H c 4J >, >J a a 3 •H u Q w1 [51 iJ a CO j-i CO t— ) to 3 ■u O CO -f-( t— ( CO •— 1 C a CO 3 o 13 (0 td CO a. 3 -:j (U CO -H e u H o -H CO CO 4J c o (A m •H Ul (0 O .—1 CO CO ■o o ^ Z o ^ CtS CO •o to to 1— 1 r-4 T-i r-4 'M •H c c n o Xi XI h h ■n 3 H H c 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 30 31 32 33 34 35 36 37 38 40 41 42 43 44 46 49 51 52 53 54 55 56 57 58 60 61 62 63 30 20 - . - . . 10 - - - 20 -20 ----20 ---10-50- -10 30 - 10 40 - - - - ----- 50 - 10 - 10 - - 10 - - 20 - - - - - 10 40 10 - - 60 10--- 20 ----20- -20 60 ---10 10-30 10 40-10 40 --10 ------- 10- 200 - - - 10 - - - 100 - - - - 40 - - 30 - - - 10 - 10 20 20 20 10- 10 10 30 10- 10 - - 10 - - 10 - - 10 - - - 10 - 20 - 20 -20 ---10- 50-10- 10 630 - - - 10 - - - 410 - - - - 530 20 10 - - 10 - - 70 130 20 ---10 ------ 50 -10 20 10 --10 90 --10 ----40 ------ -. 10 ------ 600 - - - 10 - - - 150 - - - - 20 - 10 - - - 40 40 10--- 80 10 20- 50 20 10 -----so- lo 10 10- -10 so- lo - 10 - 10 30 - - 20 10 - - - 10 - 10 - - - 20 ------- --20---- --------- 50- - 10--- - 10 10 - - 30 10 ------ - 10 10 - - 10 10 10 10 --10- --30---- -20---10--- 10 10 --20 --30 10 10 10 - - - - so- lo 10- ..--10 --40- 10 ---10- .-...20--- 10 50---- -10 10 20- 10 ...--10-- ---10 10 ---10 --30 10 10 20 --- 20-- 20-- trations were high in silty sand in some areas but were intermediate or low in other locations at similar depths. Densities of gastropod shells were high, intermediate, and low in sand, with no apparent correlation. Distribution and Density by Species The geographic distributions of the 24 most common forms of gastropods are shown in Fig- ure 13. These charts are based on the data in 19 FISHERY BULLETIN: VOL. 71, NO. 1 Table 9. — Species and density, by station, of gastropods that occurred at only one station each. Table 10. — Density distribution of gastropod shells, all species combined, in relation to water depth. Species Station Specimens Alvania janmayeni Calliostoma occidentalis Calliostoma sp. Cavolina longirostris Cavolina tridentata Epitonium groenlandicum Epitonium multistriatum Eidimella smithi Eulimella sp. Eupleura caudata Fossarus elegans Lunatia heros Mitrella lunata Neptunea decemcostata Odostomia gibboia Pieudorotetla solida Pyramidella sp. Retuia obtusa Solariella iris Taranis cirrata Turrtellopsis acicula 55 23 53 51 53 52 23 37 6 29 52 1 27 31 1 38 21 5 9 9 37 Nolnfi 20 20 10 10 20 10 30 10 20 10 10 10 10 10 40 40 10 40 30 10 10 7,1' , 20 .BLOCK ISLAND J I I Il2! — L.^ \ 1 pf. .' MARTHA'S VINEYARD NANTUCKET ipOO METERS CrZ3 0-50 GASTROPODA NUMBER PER m' [°1 60-250 ^3250- IPOO T 1 1 — yTji 1 r 41' ■40» T — ^TQi — I 1 1 r Figure 12. — Density distribution of gastropod shells, all species combined. Table 8. Species with a wide geographic dis- tribution were, in decreasing order: Alvania carinata, Mitrella zonalis, Cylichna gouldi, Colus pygmaeus, Odostomia canaliculata, Epitonium, Samples Mean Water Samples containing number depth collected gastropod of shells shells Miters 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99 100.124 125-149 150-174 175-199 200-249 250-567 Number 1 6 7 8 5 9 I 7 4 4 1 5 •2 1 Percent 100 83 71 63 100 89 100 100 75 100 100 100 100 No/m^ 10 35 84 96 190 221 51 72 80 60 102 140 65 dallianum, Cylichna alba, Nassarius trivittatus, and Turbonilla interrupta. Four patterns of geographical distribution are revealed: (1) The most common is a rela- tively narrow east-west band across the area in either shallow or deep water, typified by Balcis intermedia, Crepidula fornicata, Drillia lissotro- pis, Epitonium dallianum, and others. (2) Com- paratively broad east-west bands across the area are illustrated by Mitrella zonalis and Turbonilla interrupta. (3) Peripherial occurrence around the fine-grained bottom sediments located in the center of the area is illustrated by Cylichna alba, Odostomia canaliculata, and to some extent by Rissoa sp. (4) Distribution in the central part of the area, a pattern nearly opposite that of peripheral distribution (pattern number 3), is illustrated by Alvania carinata. Bathymetric range differed markedly among species. The minimum, maximum, and mean depth of occurrence for each species is listed in Table 11 and illustrated for the more common species in Figure 14. Mitrella zonalis was the only species taken over a wide range of water depths (62 to 567 m). Species that had a mod- erately wide depth range were: Buccinum undatum, Cylichna alba, Epitonium novangliae, Odostomia canaliculata, and Rissoa sp. Species found in shallow water — those re- stricted to depths of 50 m or less — were: Crepi- dula fornicata, Eupleura caudata, Lunatia her- 20 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS rrtPT^ /;. ^ ^.'-^ ANACHIS COSTULATA r>^ HJMTRA'S VINCOO ^(,...100 ,,- BALCIS INTERMEDIA ,SL.»o-..,' . .-^,-.v, ; . J«>'\ '''•',■'-/ >''/ ■•-■;-'■ • .-• <:9 •'■, ;/•,--- .;■■ " _^_»__,^» • • • • • '-' . ■ao_--' ^ "^^ , ^ • -:— - >V' 40" 100 ,.- ^ --•--.' ---i\ . . „ 200 „ ..---^-.--.^ ,'' '>,-^'' '-V r"c~ -—'';':_'■-'-' ,IP00 METEBS BUCCINUM UNDATUM -I III rrJr*- COLUS PyGM4fUS Tir— ' • ' ' ' — Tc'-r WTlT^ a'S . netlBD _ lOOO METEBS CREPIOULA FORNICATA rtrr*" ,- UMTMA'S VIME'ftBO ^' .zoo' .^ -»- ^_ _,' i^^j' ^-.~^~^~— •-''iJC--''^ _,500"'' '" i"^ '^^J*' ,-C-'''"V»-'-'' *''"■'' IPOO METERS CRUCIBULUM STRIATUM rr^r*~ ,' MMTHA'S VINET4B0 i?^^ 1000 ItfTERS CYLICHNA ALBA ' ' ' 7'l' 7'0" ' ' ' ' , , . 7A' 7,0- . . , , .0-' '- ..-•' "- 'c?'J ^ k •••-:'■'•• • ..: • •■. ■."/•.-''• .;'.' ''.J ,^''' \ "■---- 'j • - . . v_,^ . , . • ~> ,-- ■^v. • ■■••.. V"^'' ao.--' ~'~- , ,''' x>OOOo "^--'' 0* . lOOO METERS DRILLIA LISSOTROPIS If If.*, y- >' CYLICHNA COULDI pipr*~ ,' MWTHA'S VIMEVARO ZPITONIUM DALLIANUM BLOCK ' ' *'*'•''''*'* VINE*»RO 40 - ' .'"'<. • ».'-.'.''_ KW ■>,- A . '."' \ "■' J ''/ .--■■ - \ * ""'•v..-.'. /-:---" ./-' • *-t^^ — ■->^" , , ^ , , • V'--' ----, •0 , "■» -, • • #^-' 800 ^ -'» 5.J#: IpOO METCftS EPITONIUM NOVANGLIAE \ -r-- -1 . — . — . — Ty,. h Figure 18. — Geographic distribution of the common gastropods. 21 FISHERY BULLETIN: VOL. 71, NO. I 7,1' T,0* , , , , 40-- !oo ' -»- /\ ^y'tS'^z — -_— 'c.-*- ' MO -" _ ;- - ^^„^.-:>-^^- v*-'- 1,000 METERS LUNATIA LEVICULA ' ' ' 7'l- Tto iP>-* ,' './-L _ I ISLAND ^,, ' MARTHA'S VIMt'l ->--■-. "Sw/i MITRELLA ZONALIS pir^ MARTHA'S VINE'ARO MITRELLA SP rrir*- tpoo wcTEns NASSARIU5 TRIVITTATUS TJrr>r - MARTHA'S VINETARD , 1,000 METERS ODOSTOMIA CANALICULATA ■'-T,^-" • ;..'^ -m. //\ '■'li>>^]^<^ 40-^ .oo' ,.•- ' /■'— --'*"^--:\ . . „ '"'' 1,000 P«T£B5 POLINICES DUPLICATA ■1 ' ' ' T' Tyj- ' ' ' 1 7, • , 7,0 1 . 0'^- ._- 1M ^ NANTUC-E r 20 i^ • '._A -SLOW ,'' [ ISLAND '^^;^ *o -' '-._ •o; ^_„ j5& , -• 'S VINE >ARD . <^Bip •0,'-' -•- - •^ '\ %Vv.-' „_^ ,wo' .,^ , 900"' ' - -i ^ ^:5'S -^^ ^i^- . / ./■ ipOO METERS TURBONILLA INTERUPTA rirn^ . ' MARTHA'S VtNEVARQ ^- ^-^X--:/ TURBONILLA SP rrir^ MARTHA'S VINETARD / ,-. GASTROPODA LARVA rf«f;,*- ,'-; >J. lisLANo '^-;^' , . , MARTHA'S VINEYARD GASTROPODA SPR Figure 13. — Geographic distribution of the common gastropods. — Continued. 22 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 11. — Bathymetric distributions of 44 gastropods and the number of stations at which each occurred. Water depth Number .jpc;i.ic;a Minimum Maximum Mean stations m m m Alvania carinata 59 220 112 19 Alvania jatvmayeni 567 567 567 1 Anachis costulata 201 567 384 2 Balds intermedia 110 183 144 4 Buccinum undatum 59 146 102 2 Calliostoma occidentalis 113 113 113 1 Calliostoma sp. 179 179 179 1 Cavolina longirostris 99 99 99 1 Carolina tridentata 179 179 179 1 Colui pygmaeus 37 90 58 13 Crepidula jornicata 38 48 43 3 Crucibulum striatum 91 146 118 2 Cylichna alba 49 220 125 10 Cylichna gouldi 38 79 63 15 Drillia lissotropis 113 201 162 5 Epitonium dallianum 91 201 143 11 Epitonium groenlandicum 146 146 146 1 Epitonium multistriatum 113 113 113 1 Epitonium novangliae 95 366 215 3 Eulimella smithi 220 220 220 1 Eulimella sp. 91 91 91 1 Eupleura caudata 27 27 27 1 Fossarus elegans 146 146 146 1 Lunatia heros 46 46 46 1 Lunatia levicula 38 40 39 2 Mitrella lunata 44 44 44 1 Mitrella zonalis 62 567 157 17 Mitrella sp. 179 194 186 2 Nassarius trivittatus 33 62 45 10 Neptunea decemcostata 48 48 48 1 Odostomia canaliculata 40 220 113 12 Odostomia gibbosa 46 46 46 1 Polinices duplicata 38 110 61 4 Pseudorotella solida Dall 194 194 194 1 Pyramidella sp. 165 165 165 1 Retusa obtusa 76 76 76 1 Rissoa sp. 62 194 101 4 Solariella iris 201 201 201 1 Solariella sp. 139 139 139 1 Taranis cirrata 201 201 201 1 Turbonilla interrupta 73 201 140 10 Turbonilla sp. 51 139 70 3 Turrtellopsis acicula 220 220 220 1 Gostropod larval 73 567 200 11 1 Only one species appeared to be represented— possibly a Thais. OS, L. levicula, Mitrella lunata, Neptunea decemcostata, and Odostomia sp. Species taken only in deep water — those re- stricted to depths greater than 200 m — were: Alvania janmayeni, Anachis costulata, Eulimella smithi, Taranis cirrata, sindi Turrtellopsis acic- ula. All these species were taken at only one station, except Anachis costulata, which oc- curred at two stations. Shells of individual species of gastropods gen- erally occurred at low or moderately low densi- ties. Only three were found in high or moder- WATER DEPTH (METERS) SPECIES 50 100 150 200 250 Nassarius trivittatus Lunatia levicula Crepidula fornicata Cylichna gouldi Colus pygmaeus Polinices duplicata Odostomia canaliculata Cylichna alba Buccinum undatum 1 1 J l_ L_ L, i 1 Alvania carinata Rissoa sp. Mitrella zonalis Turbonilla interrupta Crucibulum striatum L_j Epitonium dallianum Epitonium novangliae Balcis intermedia L_ 1 1 Drillia lissotropis Anachis costulata 1 TO 54rT Figure 14. — Bathymetric range and mean depth of occurrence of the more common gastropod species. (Observed values are listed in Table 11). ately high concentrations: Alvania carinata (630/m'-), Cylichna gouldi (530/m-), and Nas- sarius trivittatus (130/m-). The density of other gastropods (41 species) was 60/m- or less. Species-Sediment Relations The majority of gastropod species occurred in sand and silty sand. None was in coarse sand or gravel-sand substrates, and none appeared to be restricted to silt. Widely distributed species generally occurred in a variety of sediment types ranging from medium sand to silt. A few species were associated with specific sediment types. Gastropods found chiefly in sand sub- strates were: Colus pygmaeus, Crepiduki for- nicata, Cylichna alba, Lunatia levicula, Nassar- ius trivittatus, and Odostomia canaliculata. The only species found principally in fine-grained sediments was Epitonium dallianum; it was mainly in silty sand. 23 FISHERY BULLETIN: VOL. 71, NO. 1 CEPHALOPODS AND SCAPHOPODS Remains of cephalopods and scaphopods were found in moderate to low densities, were small, and occurred in a relatively limited area. Only a few species of each group were represented in the samples. Illustrations of typical examples are shown in Figure 15. Cephalopod remains consisted solely of beaks (jaws or mandibles) of Decapoda (squid). All were black and 4 to 6 mm long. The animals from which the beaks came were adults and probably rather small (less than about 10 cm in mantle length). Their uniformity in configu- ration and size suggests that only one or a few species are represented. Scaphopod remains consisted only of shells or fragments of shells of a few species of the genus Dentalium (15 to 35 mm long) and one species of the genus Cadulus (mostly 10 to 13 mm ^•^ B Figure 15. — Cephalopod mandibles and scaphopod shells from off southeastern Massachusetts. A - cephalopod beaks; B - shells of Cadulus pandionis; C - shells of Dentalium spp. Each scale bar is 5 mm. Table 12. — Density of cephalopod beaks and scaphopod shells, by stations. Cephalopods^ Scaphopods Station Cadulus Dentalium pandionis spp. No/m2 No/mi No/m^ 5 6 8 10 — — 40 20 9 __ 90 10 10 10 __ 11 10 __ 21 30 — . 20 22 20 20 23 20 __ 110 3^^^^V ^ 200 "^JQcJ )\ - ^ ^"^"^^^'VVXJ sXy 500 'xj , IpOO METERS 5A ^— "-'^ - ^ '^/ ~ V,-- ^ l^ v_ — - CEPHALOPODS 7l|, 1 ^IQO 1 1 -41° 40° Cephalopod beaks were present at 12 stations, at a depth range of 76 to 567 m. Their average density was 38/m-, and maximum density 130/m-. Highest densities were at the deepest stations sampled, stations 54 and 55, where water depths were 366 and 567 m. Scaphopod shells were collected at 11 stations. Caduhis pandionis was present only on the con- tinental slope at depths between 139 and 366 m. Average density at the six stations where it oc- curred was 41/m^ and maximum density was 110/m-. Dentalium spp. occurred along the con- tinental shelf margin at depths of 91 to 183 m. Table 13. — Bathymetric distribution of cephalopod beaks and scaphopod shells and the number of stations at which each occurred. Figure 16. — Geographic distribution of cephalopod beaks. Species Water deptti Number M nimum Maximum Mean stations m m m Cephalapods 76 567 201 12 Scaphopods Cadulus pandionis 139 366 213 6 Dentalium spp. 91 183 134 7 , 7iO° 41*- . 6o; 40* 20 [BLOCK ISLAND 40 y ' . - MARTHAS VINEYARD NANTUCKET -41''- ..-o \ 80. 100 200 500 o ^ o "S --^ o ^ ^ o ^ Si. - -- ' ipOO METERS CADULUS PANDIONIS 7'l I 7,0'' , - 60? 1 I I TTTi I I I 1 1 ^t;^; — I 1 1 r' BLOCK [ ISLAND ' -./^ 40 ^- MARTHA'S VINEYARD J NANTUCKET ■a. ^-l 1 r-^ I kill \ / IV I / 1 V \ ( .f r -41* o \ o 80 . ' •404 '°° r 200 ' 500 '^ 1000 METERS ,1 \\ - .^•'-^^"V^^ DENTALIUM SPP. To' 1 1 1 — ^TT^ — I 1 1 1 1 — ^x^s — I r -40* 7'l 7'0* Figure 17. — Geographic distribution of shells of the scaphopods, Cadulus pandionis and Dentaluim spp. 25 FISHERY BULLETIN: VOL. 71, NO. 1 Average density at seven stations was 33/m-, and maximum density was 110/m-. Relations with Sediments The kinds of cephalopods that are abundant in this region are pelagic and their occurrence would not ordinarily be expected to be directly related to substrate composition. The fate of the remains of these animals that drop to the ocean floor may depend indirectly on sediment type because these species generally occur in deep or moderately deep water. Cephalopod re- mains were absent in coarse sand or gravel. Densities were moderate (10 to 40/m^) exclu- sively in silt. Caduhis and Dentalium remains also were found only in fine-grained sediments; fine sand, silty sand, sandy silt, and silt. No areas of coarse sand, gravel, or mixtures of the two yield- ed scaphopod shells. A large majority were in areas where the sediments are fine sand and silty sand. Cadulus was densest in fine sand, and Dentalium in silty sand. REMAINS OF FISH Vertebrate remains in the bottom sediments were represented exclusively by fish otoliths and small numbers of bones, teeth, and scales (Table 14). Some examples of typical otoliths are il- lustrated in Figure 18. Otoliths were rather broadly distributed over much of the area but were particularly common in the deepwater sec- tion. The otolith density was strikingly high, 3,020/m-, near the shelf break south of Nan- tucket Shoals. All samples combined included 18 genera and at least 26 species of fish ; all but one were identified from otoliths. A record of the otoliths of each species recovered at diff^erent stations is given in Table 15. Eleven of the spe- cies are bottom-dwelling types, and 11 are epi- pelagic or mesopelagic (Table 16). Three spe- cies, Merluccius albidus, M. bilinearis, and Pe- prilus triacanthus, represented by otoliths range widely from the sea bottom to upper water lev- els; they remain unclassified for the purposes of this discussion. Clupeoids, scombroids, and other common pelagic groups were lacking. The collections included many more otoliths from pe- lagic species (1,288), however, than from groundfish (141); the average otolith density (based only on samples containing one or more otoliths), of pelagic species was 379/m^ com- pared with 41/m2 for groundfish. All fish re- mains were less than 2 cm in greatest dimension, and most were less than 3 mm. The sizes of fish from which these remains came ranged from lanternfish only a few centimeters long to sharks estimated to be 2 to 3 m long. Table 14.— Density of fish remains' by stations. all species combined. Station number Otoliths Bones No/mi No/m2 4 10 __ 5 10 __ 6 30 __ 8 3,020 _^ 9 1,250 __ 10 360 10 11 10 20 12 10 --. 13 __ 10 16 ^_ 10 17 10 __ 18 _^ 10 20 20 __ 21 760 10 22 2,200 10 23 170 40 24 __ 10 25 __ 30 27 30 10 31 10 __ 33 10 34 10 10 35 40 10 36 30 10 37 1,460 80 38 1,270 39 150 10 40 20 10 41 20 20 50 ^^ 60 51 50 10 52 50 53 830 54 870 40 55 580 56 1,380 57 60 10 58 10 59 10 __ 61 10 — ' Scales occurred only at stations 11, 17, and 40 (10 to 20/m2) and teeth only at stations 53 and 54 {20 to 30/m2). 26 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS ^mg ^B^^^^;^ *«^ id 4» B ^^^y H .-''''!!f^'»- '4lE> AJt**' M r ^* u «j5S<5?- w I^IGURE 18. — Representative teleost otoliths from off southeastern Massachusetts. Inner face of otolith above, outer face below. A - Acanthuroidei sp.-l (XIO) ; B - Benthosema glaciale (X6.5); C - Centropristis ocyurus (X8); D - Ceratoscopelus maderensis (X4.5); E - Citharichthys larctifrotis (X9) ; F - Dtap/iws sp.-l (X^); G - Diaph- us sp.-2 (X4.5) ; H - Diaphus sp.-3 (X4) ; I - Diaphus sp.-4 (X6-5); J - Lepo- phidium cervinum (X5.2); K - Lobianchia dofleini (X7) ; L - Lopholatilus chamae- leonticeps (X2.5); M - Merluccius albidus (X4.5); N - Merhiccius bilinearis (X2); - Myctophum punctatum (X4.5); P - Myctophum sp. (X6.5); Q - INotoscopelus (X3.2); R - Peprilus triacanthus (X4.5); S - Phycis chesteri (X5.8); T - Poma- canthus arciiatus (XIO) ; U - "Stromatejis" (X5.8) ; V - Urophycis chuss (X2) ; W - Urophycis 1 floridanus (X2); X - Urophycis tenuis (Xl-3). 27 FISHERY BULLETIN: VOL. 71, NO. 1 Table 15. --Species and density (number per square meter) of fish otoliths, by station. e 3 c c o (U ■o -) 3 x: 4J u n B 3 nj -H U C) V^ l-l CO a> en •r4 4-) M C/1 CO u 0) 3 JIJ XI en o i-i 3 c (n CO •r-l .H CJ 1-1 o a ?^ 0) x; o (ij Pj IX 3 C (3 0) lO e •O M •H 01 CO h •H U O 3 u-t c . CH CJ o. •o C" 4J CO cu ca CO CO UH ■H ■H ■H •H o 4J >, >, >-, C x: x: x: O) a a a •o o o o .r4 ^ n M c D 3 3 3 4 5 6 8 9 10 11 12 17 20 21 22 23 27 31 33 34 35 36 37 38 39 40 41 51 52 53 54 55 56 57 58 59 61 10 10 20 40 20 2,160 650 220 10 130 20 180 - 270 40 10 30 - 440 10 10 - 30 50 - - 40 20 - 120 10 80 10 10 10 10 10 10 10 10 20 520 1,480 40 40 110 50 20 - 110 40 20 - 330 80 - - 30 - 10 20 - 60 - 10 70 10 10 - - 20 - - - 20 - - - 10 10 10 10 10 10 10 1,100 970 70 10 40 30 20 10 10 20 10 200 20 110 40 20 20 10 10 60 50 20 20 10 30 10 10 10 20 10 10 10 10 20 10 40 10 50 460 820 500 1,010 10 50 20 30 10 100 20 50 10 160 20 80 10 20 20 10 10 110 20 10 - 10 20 20 40 10 + + 10 Table 16. — Comparison of density and abundance of otoliths of pelagic fish and groundfish. Item Pelagic Groundfish U nclassified Number of otoliths Average otolith density (per m^) Including all samples (62) Only samples with otoliths (34) Number of species 1,288 (87%) 208 379 11 (44%) 141 (10%) 23 41 11 (44%) 41 (3%) 7 12 3 (12%) IDENTIFICATION OF OTOLITHS Otoliths of relatively deepwater teleosts form the major portion of all species dealt with here; littoral species rarely occurred in the samples. Myctophids (lanternfishes) contributed most of the otoliths. Many of these have been referred tentatively to various genera in this group, but specific determinations are not possible in the absence of suitable identified material for com- parison. It is very likely that the species are already in ichthyological collections, but most 28 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS preserving fluids soon render the otoliths use- less, if they do not destroy them completely. Nearly all the otoliths had suflTered some ero- sion that may have resulted from abrasion on the sea bottom, possibly preceded by partial dis- solution in the digestive system of predatory ani- mals and later by the reworking of bottom sedi- ments by deposit-feeding benthic invertebrates such as polychaete worms, holothurians, starfish, and many others. One or several of these agents resulted in the destruction of the rostral area on all percoid otoliths. The outer rims of some mer- luccid otoliths were damaged sufficiently to make identification difficult. DISTRIBUTION AND DENSITY Fish remains, all species combined, occurred at 65% of the stations. The remains were not uniformly distributed over the area but occurred mainly in the southern, offshore sector (Figures 19 and 20). More than 90 Cf of all fish remains were taken at depths greater than 150 m, where- as less than 1 9r came from depths less than 50 m. Only 369^ of the samples collected at depths less than 100 m contained one or more otoliths, whereas all samples taken at depths greater than 100 m contained otoliths (Table 17). Highest densities, 500 to 3,030/m-, were in a band par- allel to the isobaths along the outer portion of the continental shelf and upper part of the con- tinental slope. Density of fish otoliths was correlated closely Table 17. — Density distribution of fish otoliths in re- lation to water depth. Water depth Samples collected Samples containing otoliths Mean number of otoliths Meters Number Percent No/m2 20-^9 1 30-39 6 40-49 7 43 7 50-59 8 60-69 5 60 6 70-79 9 56 7 80-89 1 90-99 7 71 20 100-124 4 lOO 70 125-149 4 100 142 150-174 1 100 740 175-199 5 100 1,724 200-249 2 100 1,365 250-567 2 100 735 7,1 ll I _ I I I , 7i0' ■^-^— ' '. ^ 41*- - 60' 40* .BLOCK [island ' ,-> 1 MARTHA'S VINEYARD J NANTUCKET ( ( > '•.'■.V'i' -80 20-500 ^ZZl 500 - 3.000 T 1 1 — ■^m — I 1 1 1 1 ^t;^; — I 1 1 r 41* -40* 7'! 7'0« Figure 19. — Geographic distribution and density of fish otoliths, all species combined. - 20 I ^1** I I I I I Zi2_ '-\-;' •' - I , I L. MARTHA'S VINEYARD J NANTUCKET T 1 f Figure 20. — Geographic distribution of fish bones. 29 FISHERY BULLETIN: VOL. 71, NO. with water depth (Table 17). Average densities ranged from to 20/m^ between 20 and 100 m and from 735 to l,724/m2 between 150 and 567 m. Environmental features that contributed sub- stantially to the observed correlations are the low energy environment combined with the rel- atively mild abrasive characteristics of the bot- tom sediments. Densities of fish teeth, bones, and scales were low (80/m- or less). Teeth were recovered at only two stations (53 and 54), where water depths were 179 and 366 m; densities were 20 to 30/m-. The teeth at station 53 were from the blue shark, Prionace glauca, a cosmopolitan spe- cies that commonly attains lengths of 2 to 3 m. Fish scales were found at three stations, at water depths of 49 to 106 m, and at densities of 10 to 20/m-. Fish bones were detected at 22 sta- tions (Figure 20). Vertebrae and rib bones were encountered most frequently but occasion- ally skeletal sections from the oral and branchial regions were taken. The small thin bones gen- erally had a fresh appearance, whereas the lar- ger thicker bones were often badly eroded and stained brown. Fish bones were collected at water depths from 38 to 366 m; densities ranged from 10 to 80/m-. RELATIONS OF DENSITY TO SEDIMENTS A broad comparison of the geographic distri- bution and density of fish remains (Figures 19 and 20) with bottom sediment types (Figure 2) disclosed a moderately close correlation. The most obvious aspects were the absence of fish remains in gravel-sand mixtures, and an exceed- ingly low density in coarse and medium sand sediments. Conversely, fish remains were com- paratively common in silt, sandy silt, and fine- grained sand. Otoliths had highest densities in the fine sand, whereas bones were common in sediments composed chiefly of silt and clay with admixtures of fine sand. DISTRIBUTION AND DENSITY BY SPECIES Of the 26 fish species whose remains were re- covered from the bottom sediments, only six were abundant or moderately abundant; Cera- toscopelus maderensis, Citharichthys larctifrons, Diaphus sp.-4, Lepophidium cervinum, Merluc- cius bilinearis, and Myctophum sp. (Fish spe- cies represented by otoliths are listed by station in Table 15.) Each of these species occurred at eight or more stations, and maximum densities ranged from 60 to 2,160/m-. Four of the six abundant species are pelagic forms (exceptions are L. cervinum and M. bilinearis, although M. bilinearis frequently is mesopelagic) . The most common species was Ceratoscopelus maderensis. Otoliths of this species occurred at 19 stations and average density was 530/m-. Nearly all fish remains were collected in the southern half of the area. The geographic dis- tribution of otoliths of diff"erent species is illus- trated in Figure 21. With few exceptions, oto- liths of individual species were geographically distributed in an east-west band across the area, roughly parallel to the depth contours. A major exception to this distribution was that for M. bilinearis, the most widely distributed species. It was found at 15 stations, most of which were located on the outer continental shelf, but a few otoliths occurred on the central and inner por- tions of the shelf. This species is one of the few whose remains were found in the inner-shelf region. Water depths at which remains of individual fish species occurred ranged from 44 to 567 m. Considerable differences in depth range were evident among species, probably in part because of the sparse representation of some. Depth-of- occurrence data, by species, are summarized in Table 18 and Figure 22. Only two species, "Stromateus" and Merluccius bilinearis, were found at depths shallower than 50 m, and only six occurred at less than 100 m. On the other hand, 15 species were recovered from depths greater than 200 m, and 6 from depths greater than 360 m. The species that were distributed over the widest depth range are Ceratoscopelus maderensis and Citharichthys "larctifrons; their remains were taken at depths from 95 to 567 m. Other species whose remains were spread over a wide depth range are: Benthosema glaciale, Diaphus sp.-2, Diaphus sp.-4, and Lepophidium cervinum. About 61 Vf of the species were found 30 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS _^,. r"^" 41»- 40 CO : : : /:n .,_- 60 ao • . " • . • -40* m^ 40^ WO lOOO ME TEW 100 ^ '000 WTEBS ACANTHUROIDEI SP 1 a SP. 2 BENTHOSEMA 6 L AC 1 ALE T'O- ' ' T'C" T'O- ' ' ' ' ^BLO« -"^-'^ (island ,. «0 ~ . . / •**' .--• BO. -'' 5^: 100 f*'' ~--*- JOO ' -»-.-„. 500 '' ' ^ 1000 MTEBS CENTROPRISTIS OCYURUS ' ' 7V ' ' ' ' 41* . 40*- ^fi^P ' >^' ' ' ' ' ' "-^ 1^ '' ^^Lj *(AMTL.C«ET ..LOO. , - ••»'"»'5 .i»E>..0 ^^j^ • • • "--.•' ^ -:.-... . . . ■. / : 'to _ K» . -,,J^^V* , iOO ' "" .- . tOOO WTEW DIAPHUS SP - 3 r> - * ' >^ ' LOBIANCHIA DOFLEINI ^rlr-* ' — > ' ' — ' — ' — >- — 1 — ' — ~*- i* ' ^mL» N«t.Tuc«ET ^^^^*- -. k 41* .BLOC- "*'""■* -«'"«' (iSLftNO •0 60 80 : : : ; 40* aoo ^'VvyO^^^ ^^m^ 500 ^pOOO >i"ET£<»S DIAPHUS SP - 4 ' ito- ' ' ' ' r>T^ IRTmA'S VINE-i 1000 ii«eTt«s L EPOPHIDIUM CER VINUM LOPHOLAT/LUS CHAMAEL EON TICEPS P>"T^ 7'0- 7.0* -BlOCh I'SlAND MERLUCCIUS ALBIOUS Figure 21. — Geographic distribution of fish otoliths, by species. 31 FISHERY BULLETIN: VOL. 71, NO. 1 .^ "f MARTHA'S V.NETAnO tj^/^ - MARTHA'S VINEtARD NANTUCKET _ [slano ^ /» I'Slano '■ - ^^ - _ [island , ■;' , i? 40 , .41' .. . : : : \ .i:/: -4I'- 40 - _ . - 60 „ , •0 yi'>^ MARTHA'S V(NE»ARO 1000 METERS PHYCIS CHESTER/ rf>T'>~ .BLOCK , [island ■;^;' MARTHA'S VINEtARO (000 METERS POMA CA N THUS ARCUA TUS 7,1 ',0' 1 , W" >^ ' NTUCXE 20 .BLOW _ 1 ISLAND '^ 40 MARTHA'S VINEYARD ."^ 4 / -^' - • -' . .^.^ • D0_ ' - _.-, ------ '"■~'^,' aoo ' ,_ (000 METERS ..^- ""'■"--'.',""7-" ,• - 'STROMATEUS" ' ' . 7to- ' wrir'^ ..toe- "'""" •'""■ ^.^ . - / '° . . . . 00 " ^ w 200 . "OCXV OO" 500 _i000 METERS UROPHYCIS CHUSS Figure 21. — Geographic distribution of fish otoliths, by species. — Continued. 32 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS Table 18. — Bathymetric distribution of teleost and selachian remains by species or higher taxon, and the number of stations at which each occurred. All entries are based on otoliths except Prionace glauca, which is represented by a tooth. [P — pelagic, G — groundfish, and Un — unclassified.] Species or Water depth Number Environmental higher taxon Minimum Maximum Mean OT stations classification m m m Acanthuroidei sp.-l 179 179 179 1 G Aconthuroidei sp.-2 179 179 179 1 G Benthosema glaciate 183 567 296 5 P Centropristis ocyurus 110 183 146 2 G Ceratoscoptlus maderemis 95 567 185 19 P Citharichthys farctifrons 97 567 196 10 G Diaphus sp.-l 132 220 176 3 P Diapkus sp.-2 110 567 227 7 P Diaphus sp.-3 183 194 188 2 P Diaphus sp.-4 132 366 196 10 P Lfpophidium cervinum 97 366 169 17 G Lobianchia dojUini 183 201 193 3 P Lopholatilus chamaeleonticeps 113 113 113 1 G Merluccius albidus 79 79 79 ] Un Merluccius bitinearis 44 220 126 15 Un Merluccius sp. 91 113 102 2 Un Myctophum punctatum 139 201 174 5 P Myctaphutn sp. ?39 201 178 8 P tNotoscopelus 183 220 195 4 P Peprilus triacanthus 183 183 183 1 Un Phvcis chesteri 91 220 171 3 G Pomacanthus arcuatus 220 220 220 1 G Prionace glauca 179 179 179 1 Un ^^Stromateus'' 44 44 44 1 P Urophycis chuss 113 194 163 3 G Urophycis ^. jloridanus 113 113 113 1 G Urophycis tenuis 183 201 189 3 G Urophycis sp. 73 366 220 2 G SPECIES WATER DEPTH (METERS) 50 100 150 200 250 'Slromateus' Merluccius bi linearis Urophycis sp. Merluccius albidus Merluccius sp. Phycis chesleri Ceroloscopelus moderensis Cilharichlhys 'arclifrons Lepophidium cervinum Centropristis ocyurus Diapltus sp.-2 Lophololdus cbomoelaonticeps Urophycis .'floridanus Urophycis chuss Diaphus sp- I Diaphus sp - 4 Myc top hum sp. Myctophum punctatum Acanthuroidei sp - 1 Acanthuroidei sp-2 Peprilus triacanthus Diaphus sp- 3 Lobianchia dofieini Urophycis tenuis fNotoscopelus Benthosema glac/ale Pomacanthus arcuatus TO 567 ■TO 567 -I- T0 56T only in the general vicinity of the shelf break (100 to 220 m) . None was restricted to a depth below 220 m. REMAINS OF CRUSTACEANS AND COELENTERATES Crustaceans and coelenterates were the least numerous of all taxonomic groups represented in the samples and formed only a small portion of the total macroscopic animal remains. These two groups differed markedly in geographic dis- tribution, bathymetric distribution, and abun- dance. Thus, each is treated in a separate sec- tion below. Figure 22. — Bathymetric range and mean depth of oc- currence of fish species represented in the samples by otoliths. (Observed values are listed in Table 18.) 33 FISHERY BULLETIN: VOL. 71, NO. 1 CRUSTACEANS Remains of two groups of crustaceans — cir- ripedes and decapods-^were present in the sam- ples. Cirriped (barnacle) remains consisted of calcareous plates, primarily compartments (wall plates) plus a moderate proportion of opercular valves. Only balanomorph types were present, and generally the thicker, more durable portions were most numerous. Examples are illustrated in Figure 23. None of the chitinous parts of the skeleton, such as the covering of the appendages, was present. Decapod crustaceans were repre- sented by anomuran (hermit) crabs and brachy- B LirTrt-iiirrSi ii'rrniiiiriirrwnwi ^ tmi f HI II mill D Figure 23. — Skeletal remains of crustaceans and coelenterates. A - cirripedes, scutum and compartments; B - anomuran and brachyuran, chelipod remains; C - Flabellum, corallite fragments; D - Acanella (?), axial skel- eton remains. Each scale bar is 5 inm. 34 WIGLEY and STINTON; REMAINS FROM MARINE SEDIMENTS uran (true) crabs. Remains of the latter group consisted of the larger more massive and durable parts of the skeleton (mainly the carapace and chelipeds) , and the anomuran remains consisted only of chelipeds. Occurrence records for both groups of crustaceans are included in Table 19 ; bathymetric data are given in Table 20. The geographic and bathymetric distributions are il- lustrated in Figures 24 and 25. Remains of cirripedes (Figure 23 A) were widely scattered over the area (Figure 24) . The density of major fragments ranged from 10 to 90/m-; densities were substantially higher in shallow water than in deep water. The depth range was 27 to 567 m with the average depth at 123 m. Remains of crustaceans carapaces and che- lipeds were from anomuran and brachyuran crabs ( Figure 23B ) . They were sparse to mod- erately dense and had a somewhat limited geo- graphic distribution near the central part of the shelf (Figure 24) at depths from 51 to 113 m. Their distribution was much more restricted than that of cirripedes. Also, this part of the shelf is a low-energy region, as compared with the Nantucket Shoals and the shallow inshore areas where cirriped remains were prevalent. Table 19. — Density of crustaceans and coelenterates, by station. Station number Crustaceans Coelenterates Cirripedes Anomuran- brachyuron Flabellum Acanellai'i) Nolvfi No/r> I 90 __ 2 50 __ 12 — 20 16 50 _^ 18 —^ 10 21 10 __ 23 10 10 24 __ 10 25 20 26 __ 10 29 10 __ 32 __ 70 34 __ 10 35 __ __ 36 10 _^ 38 .— __ 41 __ 20 49 10 __ 51 10 ._ 52 __ __ 53 10 __ 54 __ 55 30 __ 59 10 10 63 80 __ No/nfi Nolrrfi 20 50 30 80 10 50 Table 20. — Bathymetric distribution of crustaceans and coelenterates, and the number of stations at which they occurred. Group Water de pth Number Minimum Maximum Mean stations m m m Crustaceans Cirripedes 27 567 123 13 Anomuran-brachyurans 51 113 76 10 Coelenterates Flabellum 146 366 221 4 Acanrlla (?) 90 97 94 2 COELENTERATES Coelenterate remains were the rarest group of animals in the prefossil assemblage. They consisted solely of corals: Flabellum alahastrum i=goodei Verrill), a cup coral, and Acanella ( ?) , a bush coral. Some examples of each kind are illustrated in Figure 23. Flabellum, a solitary coral of the madrepo- rian group, has a rather large (4 by 6 cm) polyp and a typical calcareous skeleton (corallite) with well-developed septae. Corallite remains con- tained a large proportion of septae and were commonly 4 to 8 mm long. This species occurred only in a limited area on the continental slope south of Martha's Vineyard (Figure 24) at depths of 146 to 366 m (Figure 25). Densities of fragments were as high as 80/m'-, but the average density at the locations where they oc- curred was about 40/m-. White calcareous rodlike structures about 0.5 mm in diameter and 0.5 to 1 cm in length (Figure 23D) were provisionally classified as Acanella, a colonial alcyonarian coral. The fragments appeared to be internodal portions of the axial skeletons. Acanella normani Verrill is not uncommon in the region. The multi- branched colony of this species is composed of numerous slender, jointed segments. Total height of a full-grown colony is usually less than 30 cm. Remains of this coral were found at two stations near the center of the area (Figure 24) at depths of 90 to 97 m and in densities of 20 to 50/m^ 35 FISHERY BULLETIN: VOL. 71, NO. 1 - ^ ^- ~ -o^rX-^'-: ipOO METERS CIRRIPEDES 7^1 tt; — I 1 1 1 1 — Ttpii — I 1 1 ' I ^''* ' ^ ' ' I I ^^ U L. -J r+ 20 .BLOCK I ISLAND - 40 41'- -40'- ' ,' MARTHA'S VINEYARD / NANTUCKET / ' \ ( I 60,' 80 _ 100 200 500 ^ ^ ' — ^ "V ipOO METERS ANOMURAN-BRACHYURAN 7,1* J __i L. 7'0' J Zi2I L^ I ' , i 1 1 1 — tTT? — I 1 1 1 1 Ttp^ — I 1 r 41* -40* 7'l . I ISLAND \-^l^ -fZO .BLOCK 41'- - MARTHA'S VINEYARD NANTUCKET 1. 1 1 I - 20 40 ' 60' 1 / e \ o 80. 404 '°° r- .200 ' ,. 500"^ ° -- -TI -, o ^ ^ o \ /. , \_'^ ipOO METERS 1 — ^ij; — I 1 1 r ^ ra 3 I I I f 41' . 60' -40*-' 7'0* J I Zi2 L^ I I .BLOCK I ISLAND 40 MARTHA'S VINEYARD ^ NANTUCKET TT- t I I / l\ / ' ' ( I \ •■ I I I ''-'f 41* -o S 80 _ 100 200 ^ 500 ' ipOO METERS o r J r- ? ACANELLA 40* 1 I I ^Tj^ r T I I 7^o» ' ' I I Figure 24. — Geographic distribution of skeletal remains of crustaceans and coelenterates. 36 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS WATER DEPTH (METERS) SPECIES 50 100 150 200 250 CIrnpedes Anomurons and Brachyurans Acanella (?) .TO 567 y Figure 25. — Bathymetric distribution of skeletal re- mains of crustaceans and coelenterates. COMPARATIVE DISTRIBUTION OF ALL TAXONOMIC GROUPS groups whose centers of concentration were in- shore in relatively shallow waters were cirri- pedes and Echinarachnius parma. The dominant midshelf and outer-shelf components were gas- tropods, pelecypods, and decapods. Pelecypods, scaphopods, Brisaster, and Flabelhim were com- mon along the outer portion of the continental shelf and upper portion of the continental slope. The chief components in deeper sections of the continental slope were fish otoliths and cephalo- pod mandibles. The distribution of the principal animal re- mains in relation to each other, sediment type, and water depth, and, to a limited extent, their north-south geographical position on the conti- nental shelf and slope are illustrated in Figure 26. This chart is a generalized profile of the study area with the inshore (north) section on the lefthand side and the oflFshore (south) sec- tion on the righthand side. Broad, diagonally striped bands indicate relatively high density, and narrow lines indicate low density. Animal SEDIMENT TYPE in 0. lij o a: i m " / Sri i -4 IHIGH DENSITY I LOW DENSITY _ SILTY SAND SAND SANDY SILT SILTY SAND SAND Figure 26. — Schematic diagram of the density distribu- tion of macroscopic remains of the major animal groups represented in bottom sediments arranged according to water depth and sediment type (see text for details). SUMMARY Skeletal remains of deceased animals were common seabed components on the southern New England continental shelf and the upper part of the continental slope. In some sections, par- ticularly in shallow water, skeletal remains con- stituted a substantial portion of the substrate volume — up to nearly 30 9f in the vicinity of Nantucket Shoals. Offshore, near the margin of the continental shelf and in the upper portion of the continental slope, macroscopic animal re- mains generally constituted less than 1 9f of the substrate. Remains of benthic, pelagic, and nektonic or- ganisms were present; benthic forms were dom- inant. Planktonic animals (represented only by pteropods) were sparse. Fish and cephalopods were the principal nektonic forms. They were rather abundant in the deeper waters, particu- larly on the outer portion of the continental shelf and on the continental slope. The two animal groups that contributed the largest quantities of material to the substrate were echinoid echinoderms and pelecypod mol- lusks. Although remains of a wide variety of fish species were present, the quantity was mod- erate and the sizes small; consequently the vol- ume of fish remains was rather small. ECHINODERMS The exceedingly abundant remains of echino- derms consisted exclusively of echinoids. Onlj'- one species — Echinarachnius panna — occurred in high densities and was the most abundant and widely distributed component of organic 37 FISHERY BULLETIN: VOL. 71, NO. I origin in the sediments. Geographically it had a wide distribution, occurring at 72% of the stations. Depth range was 27 to 201 m. Size, shape, and color of Echinarachnius fragments differed markedly with water depth and sedi- ment type. The E. parma fragments in the inshore localities were whitish, relatively large, and had angular edges and corners; in offshore localities, the fragments were light greenish- brown, smaller, and had rounded edges. Den- sities of echinoids other than Echinarachnius were low, and except for Brisaster, remains were found at only a few localities. Density of Bri- saster remains were low but the remains were rather widely distributed along the outer portion of the continental shelf. Remains of Strongylo- centrotus drohachiensis were sparse and widely scattered in both shallow and deep water. MOLLUSKS Pelecypods ranked first in diversity of forms (57 species) and second in volume of remains in the bottom sediments. They were present at all depths sampled, from 27 to 567 m, and were widely distributed geographically. Densities were high in a wide band extending from Nan- tucket Shoals southwestward across the area, and in a narrow band parallel to the isobaths near the shelf break. Pelecypods were very abundant (more than 3,000/m-) at 6 stations, most of which were along the outer margin of the continental shelf; common to abundant (50 to 3,000/m-) at 48 stations; and sparse (less than 50/m2) or absent at 8 stations. In general, the species with the broadest geographic distri- butions occurred in highest densities. The six most abundant and widely distributed pelecy- pods were: Venericardia borealis, Arctica is- landica, Astarte subequilatera, A. undata, Nu- cula proxima, and Thyasira trisinuata. Pelecy- pod shells were more abundant in moderately fine-textured sediments than in either the coarse or very fine sediments. Silty sand, sandy silt, and sand-silt-clay yielded the highest densities of pelecypod shells. Size of shells ranged from 10 to 12 cm (Spisula, Arctica, Placopecten) to less than 5 mm {Thyasira, Nucula, Bathyarca) . Gastropods ranked third in volume of skeletal material in the substrates. Shells of gastropods were distributed widely throughout the area, but highest densities were near the center. A total of 44 species were present, but only 2 were gen- erally abundant — Alvania carinata and Cylichna gouldi. Shells were taken at all depths, and were particularly common between 60 and 80 m and moderately common between 175 and 250 m. Density was correlated in a general way with bottom sediments. High densities were in silty sand and sand sediments, whereas shells were absent in coarse sand and mixtures of sand and gravel. A large majority of gastropod shells was less than 1 cm in height. Cephalopod remains, consisting entirely of beaks, were present at only 12 stations, all of which were from the outer portion of the conti- nental shelf and upper part of the continental slope at depths between 76 and 567 m. Densities were generally less than 40/m- at the shallower depths, but ranged to ISO/m^ at a depth of 366 m and ll/m^ at 567 m. Remains of this group ranged in size from 4 to 6 mm and were rela- tively fragile. They were recovered only from fine-textured sediments. Distributions of scaphopods were rather lim- ited geographically and densities were low. The two genera collected, Cadulus and Dentalum, were present at 11 stations, geographically lim- ited to the deepwater areas on the outer portion of the continental shelf and the upper continental slope. The bathymetric range was 139 to 366 m for Cadulus, and 91 to 183 m for Dentalium. Sed- iments at the scaphopod localities were gener- ally fine-grained, but Cadulus occurred in slightly coarser sediments than Dentalium. Densities at the stations where they occurred averaged about 30 to 40/m-; maximum density for both genera was 11/m-. Cadulus shells were 10 to 13 mm long, and Dentalium shells were 15 to 35 mm. FISH Fish were the only vertebrates in the samples. Otoliths were the main component and bones were moderately common, but teeth and scales were rare. Remains of 26 species were collected, nearly half of which were from epipelagic or mesopelagic forms. Myctophids were the most numerous and widely distributed, and they con- 38 WIGLEY and STINTON: REMAINS FROM MARINE SEDIMENTS tributed the greatest number of species. Thirty- six percent of the species and 87 ?f of all otoliths were Myctophiformes. The six most abundant fish, based on otolith identifications, were: Cer- atoscopelus maderensis, Citharichthys tarcti- frons, Diaphus sp.-4, Lepdphidium cervinum, Merluccius bilinearis, and Myctophum sp. The estimated length of the fish whose remains were encountered ranged from a few centimeters to several meters. Remains of fish were at depths between 38 and 567 m, and an overwhelming ma- jority was found at depths greater than 150 m. More than 909^ of the otoliths were at depths below the 150-m isobath ; bones were less com- mon and more uniformly distributed, from 38 to 366 m. The remarkably high otolith density of 3,030/m2 was found near the edge of the conti- nental shelf south of Nantucket Shoals. Remains of most individual species were geographically distributed in east-west bands across the area, generally oriented parallel to the isobaths. Fish remains were absent in coarse-grained sedi- ments, and most abundant in fine sands and silt-clay. CRUSTACEANS AND COELENTERATES Crustaceans and coelenterates were the only other nonmolluscan invertebrates, in addition to those previously described, that were present in the samples. The quantity of their remains was very small. Crustaceans were generally sparse and rather widely distributed. Cirripedes consisted exclu- sively of shells of sessile forms; they were geo- graphically scattered and were taken at all depths sampled. Cirripedes were only slightly more common in shallow water than in deep water. They were one of the few animal groups whose remains occurred in coarse-grained sedi- ments. Fragments of skeletons of anomurans and brachyurans were encountered only in the midcontinental shelf in sediments primarily of silts and fine sands. They were collected between 51 and 113 m. Densities were low, from 10 to 70/m^ Remains of coelenterates occurred in low den- sities (10 to 80/m^) and were geographically restricted to small areas in the south-central and southwestern sectors. Two genera — both corals —were represented, Acanella (?) (at 90 to 97 m) and Flabellum (between 146 and 366 m). Both kinds were restricted to fine-textured sed- iments. ACKNOWLEDGMENTS The staflF members of the Northeast Fisheries Center, National Marine Fisheries Service, Woods Hole, Mass., who aided in collecting and processing the samples are: Harriett E. Mur- ray, Samuel R. Nickerson, Ruth R. Stoddard, and Roger B. Theroux. Officers and crew of RV Delaware assisted in collecting the samples, Arthur S. Merrill, National Marine Fisheries Service ; Earl Reed, Museum of Science, Spring- field, Mass.; and Roger B. Theroux, National Marine Fisheries Service, identified mollusks; Malcolm R. Clarke, National Institute of Ocean- ography, Wormley, England, contributed infor- mation regarding cephalopods; and Richard H. Backus and James E. Craddock, Woods Hole Oceanographic Institution, Woods Hole, Mass., identified fish and provided reference specimens for otolith identification. K. 0. Emery, Woods Hole Oceanographic Institution, Woods Hole, Mass., provided information on bottom sedi- ments and suggestions for improving the man- uscript. LITERATURE CITED Belyaev, G. M. 1970. The rostra of squids in the bottom sediments of the Pacific Ocean. [In Russian, English summ.] Tr. Inst. Okeanol. Akad. Nauk SSSR 88: 236-251. Belyaev, G. M., and L. S. Glikman. 1970. The teeth of sharks on the floor of the Pa- cific Ocean. [In Russian, English summ.] Tr. Inst. Okeanol. Akad. Nauk SSSR 88:252-276. BiGELOW, H. B. 1927. Physical oceanography of the Gulf of Maine. U.S. Bur. Fish., Bull. 40(2) :511-1027. 1933. Studies of the waters on the continental shelf, Cape Cod to Chesapeake Bay, I. The cycle of temperature. Pap. Phys. Oceanogr. Meteorol. 2:1-135. Brongersma-Sanders, M. 1949. On the occurrence of fish remains in fossil and recent marine deposits. Bijdr. Dierkd. 28: 65-76. 39 FISHERY BULLETIN: VOL. 71, NO. 1 BuMPUS, D. F., J. Chase, C. G. Day, D. H. Frantz, Jr., D. D. Ketchum, and R. G. Walden. 1957. A new technique for studying non-tidal drift with results of experiments off Gay Head, Mass., and in the Bay of Fundy. J. Fish. Res. Board Can. 14:931-944. BuMPUS, D. F., AND G. G. Day. 1957. Drift bottle records for Gulf of Maine and Georges Bank, 1931-56. U.S. Fish. Wildl. Serv., Spec. Sci. Rep. Fish. 242, 61 p. CoLTON, J. B., Jr. 1964. History of oceanography in the offshore wa- ters of the Gulf of Maine. U.S. Fish. Wildl. Serv., Spec. Sci. Rep. Fish. 496, 15 p. 1968. Recent trends in subsurface temperatures in the Gulf of Maine and contiguous waters. J. Fish. Res. Board Can. 25:2427-2437. 1969. Temperature conditions in the Gulf of Maine and adjacent waters during 1968. J. Fish. Res. Board Can. 26:2746-2751. Craig, G. 1953. Discussion: Fossil communities and assem- blages. Am. J. Sci. 251:547-548. Craig, G. Y., and N. S. Jones. 1966. Marine benthos, substrate and palaeoecology. Palaeontology 9:30-38. David, L. R. 1947. Significance of fish remains in recent deposits off coast of southern California. Bull. Am. As- soc. Pet. Geol. 31:367-370. Day, C. G. 1958. Surface circulation in the Gulf of Maine as deduced from drift bottles. U.S. Fish. Wildl. Serv., Fish. Bull. 58:443-472. Emery, K. 0. 1960. The sea off southern California, a modern habitat of petroleum. Wiley, Lond., 366 p. 1966. Atlantic continental shelf and slope of the United States, geologic background. U.S. Geol. Surv. Prof. Pap. 529-A:Al-A23. 1968. Positions of empty pelecypod valves on the continental shelf. J. Sediment. Pet. 38:1264-1269. Emery, K. 0., and L. E. Garrison. 1967. Sea levels 7,000 to 20,000 years ago. Science (Wash., D.C.) 157:684-687. Emery, K. 0., A. S. Merrill, and J. V. A. Trumbull. 1965. Geology and biology of the sea floor as de- duced from simultaneous photographs and sam- ples. Limnol. Oceanogr. 10:1-21, Garrison, L. E., and R. L. McMaster. 1966. Sediments and geomorphology of the con- tinental shelf off southern New England. Mar. Geol. 4:273-289. Gunter, G. 1947. Catastrophism in the sea and its paleonto- logical significance, with special reference to the Gulf of Mexico. Am. J. Sci. 245:669-676. Habe, T. 1956. Studies on the shell remains in l?ays. [In Japanese, English summ.] Contrib. Physiol. Ecol. (Kyoto Univ.) 77:1-31. Jensen, A. S. 1905. On fish-otoliths in the bottom-deposits of the sea. I. Otoliths of the Gadus-species deposited in the Polar Deep. Medd. Komm. Havunders., Ser.: Fisk. 1(7), 14 p. Johnson, R. G. 1957. Experiments on the burial of shells. J. Geol. 65:527-535. Ladd, H. S. 1957. Introduction. In H. S. Ladd (editor). Trea- tise in marine ecology and paleoecology. Vol. 2, p. 1-29. Geol. Soc. Am., Mem. 67. McMaster, R. L., and L. E. Garrison. 1966. Numerology and origin of southern New England shelf sediments. J. Sediment. Petrol. 36 : 1131-1142. Merrill, A. S., K. O. Emery, M. Rubin. 1965. Ancient oyster shells on the Atlantic Conti- nental Shelf. Science (Wash., D.C.) 147:398-400. Rhoads, D. C. 1966. Missing fossils and paleoecology. Discovery (New Haven) 27l) : 19-22. Schafer, W. 1956. Wirkungen der Benthos-Organismen auf den jungen Schichtverband. Senckenb. Lethaea 37: 183-263. Shepard, F. p. 1954. Nomenclature based on sand-silt-clay ratios. J. Sediment. Petrol. 24:151-158. Smith, W., and A. D. McIntyre. 1954. A spring-loaded bottom sampler. J. Mar. Biol. Assoc. U.K. 33:257-264. SOUTAR, A. 1967. The accumulation of fish debris in certain California coastal sediments. Calif. Coop. Oceanic Fish. Invest., Rep. 11:136-139. Twenhofel, W. H., and S. A. Tyler. 1941. Methods of study of sediments. McGraw- Hill, N.Y., 183 p. Uchupi, E. 1963. Sediments on the continental margin off eastern United States. U.S. Geol. Surv. Prof. Pap. 475-C:Cl32-C137. WiGLEY, R. L., and K. O. Emery. 1967. Benthic animals, particularly Hyalinoecia (Annelida) and Ophiomusium (Echinodermata), in sea-bottom photographs from the continental slope. In J. B. Hersey (editor). Deep-sea pho- tography, p. 235-249. John Hopkins Oceanogr. Stud. 3. Wigley, R. L., and a. D. McIntyre. 1964. Some quantitative comparisons of offshore meiobenthos and macrobenthos south of Martha's Vineyard. Limnol. Oceanogr. 9:485-493. 40 DEEP MAXIMA OF PHOTOSYNTHETIC CHLOROPHYLL IN THE PACIFIC OCEAN E. L. Venrick, J. A. McGowAN, and A. W. Mantyla^ ABSTRACT Data collected on several expeditions through the temperate and tropical Pacific Ocean show that during most of the year the maximum concentrations of chlorophyll occur below the surface, typically in a narrow layer near or below the depth of penetration of 1% of the surface light. The layer appears to be continuous across most of the Pacific although the depth and chlorophyll concentration vary regionally. The depth of the layer is more closely related to the depth of the nitrite maximum and to the position of the nutricline than to either light or density regimes. Productivity within the layer is low but positive, and contributes substantially to the total production of the water column. The maximum layer may be a seasonal phenomenon developing in the summer after the stabilization of the water column and mixing to the surface during the winter. Year to year fluctuations of depth and concentration of chlorophyll within the maximum layer may be related to large-scale meteorological fluctuations. Doty and Capurro (1961) have tabulated the position, date, depth, and values of chlorophyll and productivity in the world's oceans. There are several thousands of these measurements in the Pacific. Most are in the Northern Hemi- sphere, and most are near land masses or is- lands (e.g., Hawaii, Luzon, Hokkaido, New Cal- edonia, New Zealand), along the equator, or north of lat 40°N. Of the values from the oceanic Pacific, between lat 50 °N and 50 °S, less than 10% of the chlorophyll values represent depths greater than 25 m; in the same region, over half of the productivity measurements were obtained at the sea surface. Koblenz-Mishke, Volkovinsky, and Kabanova (1970) have used these data and additional data available to them to estimate the plankton primary production of the Pacific, to construct tables and charts of its geographical variability, and to compare production in the Pacific with their estimates from other oceans. Their estimates of primary production, ex- pressed in milligrams carbon per square meter of sea surface per day, represent production in- ^ Scripps Institution of Oceanography, University of California at San Diego, La Jolla, CA 92037. Manuscript accepted August 1972. FISHERY BULLETIN: VOL. 71, NO. I, 1973. tegrated through the water column. However, many of their production values are extrapolated from surface measurements, and, in large areas of the temperate gyres of the North and South Pacific, production is estimated from the avail- able chlorophyll data, or from "oxygen or hy- drogen saturation" values. All values of total production in the water col- umn are strongly dependent upon the assumed (usually) depth of zero productivity. This is traditionally taken to be the depth at which the light intensity has been reduced to 1% of the incident radiation, and this criterion has been used to divide the water column into a euphotic zone and an aphotic zone. Evidence is accumulating that major concen- trations of plant material in the ocean usually occur below the surface, typically within the thermocline and near the bottom of the euphotic zone. Maxima of chlorophyll or phytoplankton as deep as 100 m have been reported from the Indian Ocean ( Yentsch, 1965) , the Sargasso Sea (Menzel and Ryther, 1960), the Gulf of Mexico (Steele, 1964), and the Kuroshio and adjacent regions (Motoda and Marumo, 1963; Saijo, lizuka, and Asaoka, 1969). Shallower maxima are characteristic of the California Current 41 FISHERY BULLETIN: VOL. 71, NO. 1 (Allen, 1939; Lorenzen, 1965, 1967). Initial results of the EASTROPi^ C survey (Love, 1970, 1971) indicate a chlorophyll maximum varying in depth between 50 and 100 m over large areas of the eastern tropical Pacific. Anderson (1969) has studied the chlorophyll maximum layer off the Oregon coast which is present between 50 and 75 m during the summer. The layer is con- tinuous over a broad region in the Eastern Sub- arctic Pacific and maybe transpacific. Chloro- phyll maxima have also been reported from the other major water masses of the Pacific (El- Sayed, 1970; Sorokin, 1970). Different workers have attributed the exis- tence of the maximum layer to different processes including the concentration of detrital chloro- phyll in the pycnocline (Lorenzen, 1965), differ- ential zooplankton grazing (Lorenzen, 1967), an increase in the chlorophyll/carbon ratio in plant / cells, without an accumulation of cells (Steele, 1964) , horizontal advection and layering of dif- ferent water masses and plant populations (Sano, 1966), the sinking of active or senescent cells from shallower depth (Allen, 1932; Steele and Yentsch, 1960) , and in situ production (An- derson, 1969). In short, the tendency has been to consider deep chlorophyll maximum layers as discrete and sporadic phenomena and to inter- pret them strictly according to local conditions. The accuracy with which surface productivity reflects the productivity throughout the water column has been investigated by Koblenz-Mishke et al. (1970) by means of log-log scatter dia- grams. There is a linear trend in their trans- formed data, but the spread of values around the regression line is broad. Lorenzen (1970) showed a significant linear regression, after transformation to logarithms, of total produc- tion on the concentration of surface chlorophyll. The regression, however, removes only half of the variability of the dependent variable, and the author advises that precise values of total production must depend upon direct measure- ments. He also cautions that extrapolations from surface values are based upon averages and will easily miss unexpected events. There have been very few attempts to mea- sure productivity in the deeper maximum layers. Anderson (1969) made one series of in situ mea- surements within the chlorophyll maximum layer off the Oregon coast. There was a peak in pro- duction within the layer and positive photosyn- thesis as deep as 90 m, the 0.1% light level. Im- plicit in most studies to date is the assumption that pigment concentrations below the level of 1% light are nonphotosynthetic and represent a loss of plant material from the "euphotic zone." The two extensive surveys from the Sargasso Sea and the eastern tropical Pacific both adjusted the depth of the lowest sample to the depth of 1% light, and rarely sampled below 100 m, even though the maximum pigment concentrations were frequently obtained from the deepest sam- ple. In the present paper, the authors have sum- marized a large amount of data accumulated over the past 8 years, all of which indicate that a deep chlorophyll maximum layer is a regular and continuous feature of much of the oceanic Pacific. It is frequently observed below the tra- ditionally defined euphotic zone, yet it is dom- inated by photosynthetically active chlorophyll a which is present in concentrations as great as 10 times those at the surface. The development of this maximum layer appears not to be a lo- calized process, but a widespread and regularly occurring phenomenon. Because of its limited vertical extent and great depth, the existence, extent and significance of this maximum layer has been overlooked by most previous surveys of chlorophyll and productivity. Evidence sug- gests that a better understanding of this layer will necessitate revision of existing estimates of total primary production in the ocean. METHODS Since 1964 we have been mapping and study- ing the subsurface chlorophyll maximum in the Pacific on a series of expeditions (Figure 1). In 1964 (URSA MAJOR Expedition: Univer- sity of California, 1967) and 1966 (ZETES Ex- pedition: University of California, 1970), chlo- rophyll pigments were determined with a D U Spectrophotometer; on other expeditions chlo- rophyll a and phaeophytin were assayed with a 42 \ VENRICK, McGOWAN, and MANTVLA : PHOTOSYNTHETIC CHLOROPHYLL 60** -80** Figure 1. — Cruise tracks of seven expeditions from which chlorophyll data was obtained, and the location of Marine Life Research station 100.80 at which chlorophyll was sampled seasonally. 43 FISHERY BULLETIN: VOL. 71. NO. 1 Turner Fluorometer/ Water samples from the same casts were preserved with neutral Forma- lin for subsequent phytoplankton enumeration. Additional measurements have routinely in- cluded temperature and salinity, determined with an STD; oxygen, determined by the Win- kler procedure; and phosphate, silicate, nitrate, and nitrite, measured with the spectrophotom- eter or the autoanalyzer (Strickland and Par- sons, 1968). On CLIMAX II and ARIES III ammonia was assayed (Solorzano, 1968). On stations occupied at noon, the transparency of the water was measured with a secchi disk and the compensation depth was estimated by mul- tiplying the terminal depth by three. A wide variety of zooplankton samples were collected on all expeditions. On the expeditions CLIMAX I, CLIMAX II, and ARIES III, observations were concentrated in two areas of the North and South Pacific, near the axes of the Central Pacific Gyres. In these regions, in addition to the above measure- ments, productivity was routinely measured by the uptake of C-14 by samples incubated on deck in simulated in situ incubators (Owen and Zeit- zschel, 1970) and less frequently by samples in- cubated in situ (Steeman Nielsen, 1952). Pen- etration of light into the ocean was measured with a submarine photometer (Austin and Lou- dermilk, 1968). Coincident secchi disk determi- nations tended to overestimate the depth of 1% light, though the agreement was usually within 6 m. A submersible pump with a deck mounted filtering system was used to obtain stratified samples for determination of biomass (dry weight) of three size categories of zooplankton (333 fx and greater, 332-103 /jl, and 102-35 /i). The same pump supplied water to the fluorom- eter, equipped with a flow-through door, and to the autoanalyzer for continuous vertical profiles of chlorophyll and nutrients (Beers, Stewart, and Strickland, 1967). In September 1968 (CLIMAX I Expedition), a pair of parachute drogues, set at 10 m depth, were released at lat 27°N, long 155°W, and fol- lowed for 10 days, during which time they moved ' Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. northwest approximately 150 miles. Physical, chemical, and biological properties were sampled continuously on a 24-hr schedule. In September- October 1969 (CLIMAX II Expedition), a grid of ten 24-hr stations was occupied along long 155°W between lat 27°10' and 28°30'N, and a grid of six 24-hr stations was occupied along the same meridian between lat 24°40' and 25°20'S. This latter pattern was repeated in March 1971 (ARIES III Expedition) . The sam- pling routine was similar on each of these ex- peditions. Each 24-hr station included four casts for nutrients and chlorophyll, day and night samples for biomass of micro- and macrozoo- plankton, and a simulated in situ productivity experiment. In 1969 and 1971 a single in situ productivity experiment followed the routine sta- tion plan. In addition to data collected on S.I.O. (Scripps Institution of Oceanography) expeditions, we have available data from the California Current collected by institutions participating in the CalCOFI (California Cooperative Oceanic Fish- eries Investigations) program. We have made use of data from station 100.80 which is located near the western edge of the California Current. GEOGRAPHICAL DISTRIBUTION Chlorophyll data collected on several expedi- tions have been combined and contoured in Fig- ure 2. It is clear from these that a subsurface layer of high chlorophyll concentration is present across vast areas of the Pacific Ocean during many months of the year. South of lat 46°N, the maximum concentrations of chlorophyll are frequently observed at greater depths than the estimated 1% light level. The depth of the maximum along the meridianal transects shows no relationship with either temperature, salinity, or density. The chlorophyll maximum layer shoals near land, and in regions of general upwelling such as the North Subarctic Gyre and the Equatorial belt. It deepens near the axes of the Central Pacific Gyres. The meridianal continuity of the layer is especially remarkable considering that it passes through three major epipelagic envi- ronments: the Subarctic, where it is likely to 44 VENRICK, McGOWAN, and MANTYLA: PHOTOSYNTHETIC CHLOROPHYLL -ARIES I NOV -DEC 1970 KIUUERO 31 MARCH 1969- FlGURE 2. — Vertical sections of chlorophyll a concentrations in the Pacific Ocean; vertical exaggeration 5020. be confluent with that discussed by Anderson (1969) , the Central, and the Equatorial environ- ments. The east-west section indicates that the max- imum layer is well developed over most of the middle latitudes of the South Pacific. The max- imum layer in the South Central Gyre tends to be deeper, and the chlorophyll concentrations throughout the water column tend to be lower than in corresponding areas of the North Pacific. The greatest depth so far observed by us was at lat 20°09'S, long 118°18'W, during December (ARIES I Expedition) where a layer containing 0.05 mg/m^ occurred between 200 and 245 m depth; chlorophyll values at the surface were less than 0.01 mg/m^ Portions of all three sections have been re- peated by different expeditions. The depth and concentration of chlorophyll in the maximum layer vary somewhat, but the general features remain the same. In 1969, the Ocean Research Institute, University of Tokyo, ran a transect along long 155°W (Ocean Research Institute, University of Tokyo, 1970) . The results of their chlorophyll measurements compare well with ours. VERTICAL DISTRIBUTION We have supplemented our discrete, quantita- tive chlorophyll samples with in vivo profiles of fluorescence which provide continuous, but qual- itative pictures of the fine scale structure of the maximum layer (Figure 3) . Although the major RELATIVE FLUORESCENCE Stpt 21, 1968 aT'OtfH 155'50'W March 19, 1971 24'36'S I59*00'W Stpt 19. 1968 26'58'N IS5* 24'W Figure 3. — Continuous vertical profiles of in vivo fluor- escence of chlorophyll, a) a simple maximum layer in the North Central Pacific; surface chlorophyll concen- tration 0.02 mg/m^; b) a simple maximum layer in the South Central Pacific; surface chlorophyll concen- tration 0.01 mg/m^; c) a double maximum layer in the North Central Pacific; surface chlorophyll concen- tration 0.02 mg/m3. 45 FISHERY BULLETIN: VOL. 71, NO. 1 s o r .,^ n 1^^^^---^..^^ ^""^^^--.^^ Z 1 1 1 1 1 1 1 1 IN 1 z <2 o uj ! ge O o> O 6 H I O 7- - < O r 1 ' ■-- [ tL 1 1 1 1 1 1 1 1 1 1 1 g (Uf) Hid3a ( UJ ) Hid3a (Ui) HldBO OS ^ to C3 C5 X! i-H »H n o o o u O o rC c >. 43 T3 J* T) > (U fi -u (D ?3 M u 73 r/i .s .2 "rt (h > 0) »H P, i) o ^ s a '^ ;j a ,_r -i-> CS j= u bo CO >> -a 1-1 ^ fl) IV ,£3 ^ -u o ^ a c 4) -n ■s. T-l ,£; 4) Pi -u o rt 4) C!3 a o ° O lO o bo o bo tC O I C -^ 10 (O o u fa.S C C<3 (UJ) HidBQ 46 VENRICK, McGOWAN, and MANTYLA: PHOTOSYNTHETIC CHLOROPHYLL accumulation of pigment generally occupies a layer 50-75 m thick, the core of the layer may be very abrupt. From closely spaced water sam- ples (Table 1) we have found that the highest concentrations of chlorophyll may be contained in a layer less than 5 m thick and may exceed by 109^ to 50% the concentrations in the adjacent samples. Occasionally maximum concentrations are found in more than one layer within the re- gion of chlorophyll accumulation (Figure 3c, Table 1, 23 August 1967). It is very difficult to sample such a narrow core with discrete water samplers, A routine cast, in which samples are usually spaced at least 15 m apart, is likely to miss the peak concen- trations, and underestimate the chlorophyll con- tent of the maximum layer. Moreover, because of the rapid vertical changes in chlorophyll con- centration, slight variations in the position of the samples within the layer may appear as hor- izontal discontinuities of the layer. Discrete chlorophyll data, including those presented in this paper, must be interpreted accordingly. SPECIES COMPOSITION The numbers and species of diatoms in water samples collected on expeditions URSA MAJOR and ZETES have been enumerated (University of California, 1967, 1970; Venrick, 1969). The increase in chlorophyll concentration in the max- imum layer is accompanied by a significant increase in the number of diatom cells. Further- more, the maximum layer is composed of dif- ferent assemblages of species within the Sub- arctic Pacific, the Transition Domain, and the Central Pacific (Venrick, 1971). In August, north of lat 40°N the species within the maxi- mum layer were the same as those occupying the overlying water mass. South of lat 38°N, however, samples from the maximum layer were dominated by species which were not observed in shallower samples. During the winter when the maximum layer had been eroded by increased turbulence, the same species were present, but they were distributed randomly through the water column. More recent studies were undertaken in 1968 at lat 26°57'N, long 155°10'W with a series of 19 replicate casts over a distance of 10.5 miles. Phytoplankton samples were collected from 25 m, 50 m, 75 m, and from the chlorophyll maximum layer at 125 m. A total of 80 species of diatoms were identified, of which 24, 36, and 37 were observed in the samples collected from 25 m, 50 m, and 75 m, respectively. A total of 64 spe- Table 1. — Fine scale structure of the chlorophyll maximum layer. URSA MAJOR CLIMAX 11 lot 43° 2 Sept. 1964 49'N, long 154°44'W 26 Aug. 1969 lot 27°09'N, long 155°18'W 23 Aug. 1969 lot 28°29'N, long 155°16'W Depth (m) Chlorophyll a (mg/m3) Depth (m) Chlorophyll a (mg/m3) Depth (m) Chlorophyll a (mg/m3) 0.10 1 0.06 0.06 20 0.11 20 0.06 20 0.06 50 0.55 39 0.07 40 0.06 60 1.19 58 0.06 60 0.06 65 0.94 77 0.06 80 0.04 70 0.73 96 0.08 90 0.08 75 0.56 101 0.07 100 0.13 80 0.44 105 0.11 105 0.13 85 0.38 110 0.08 110 0.09 90 0.24 114 0.08 114 0.13 100 0.10 124 0.06 118 0.12 141 0.02 122 126 130 135 140 150 165 180 200 0.10 0.07 0.07 0.06 0.04 0.04 0.03 0.02 0.02 47 FISHERY BULLETIN: VOL. 71, NO. 1 cies were found at 125 m, and of these 64, 22 occurred only in samples from that depth. Thus, the chlorophyll maximum layer, which has a higher species diversity (as measured by the number of diatom species) and which may con- tain species not found at shallower depths ap- pears to offer unique features as a biological habitat. THE CENTRAL PACIFIC Studies conducted on Expeditions CLIMAX I, CLIMAX II, and ARIES III near the axes of the North and South Central Pacific Gyres have pro- vided us with a large amount of data concerning the vertical distribution of chlorophyll and pro- ductivity in the water column and their rela- tionship with other physical, chemical, and biological parameters. Comparison with data collected over much wider areas on other expe- ditions leads us to believe that the relations ob- served in the Central Gyres may be pertinent to much of the oceanic Pacic. The vertical distribution of chlorophyll and net production observed during these four stud- ies have been summarized in Table 2. All studies show a well-defined subsurface accumulation of chlorophyll which varies in width from 50 to 75 m and contains maximum concentrations of chlorophyll in excess of 0.10 mg/m^. The core of the layer always occurred below the depth penetrated by 1% of the surface radiation. In fact, more than half of the total chlorophyll within the water column was observed below this depth. The rate of production per unit chlorophyll decreases with depth from a maximum at about 20 m, but this is partially offset by the increase in the amount of chlorophyJl, and production rates as high as 0.13 mg C/m^/hr have been observed in the maximum layer in the South Pa- cific. Our in situ experiments indicate that 7% to 20 % of the total production in the water col- umn occurs below the 1 % light level. These are minimum values since our in situ studies did not reach the level of no productivity. The total rate of production throughout the water column is variable on rather small spatial and temporal scales, but appears to be considerably greater than maximum estimate of 100 mg C/mVday (8.3 mg C/m^/hr) estimated by Koblentz- Mishke et al. (1970). The vertical distribution of chlorophyll and several relevant properties are illustrated in Figure 4. Data points are mean values of ob- servations made in replicate on six 24-hr stations in the South Central Pacific (CLIMAX II Ex- pedition) and represent an area of 60 square miles and a time span of 6 days. Above 200 m there was an average of 12.35 mg chlorophyll per square meter sea surface. Of this, over half occurred below the estimated depth of 1 % light. We estimated the light intensity at the core of the maximum layer to lie between 0.10% and 0.26% of incident radiation. The vertical dis- tribution of phaeophj^in is similar to that of chlorophyll. The accumulation of both chlorophyll and phaeophytin occurs within the pycnocline. On a local scale, these layers may move up and down with the pycnocline, for instance in response to Table 2. — Mean value and 95% confidence limits of the mean for data relative to the vertical distribution of light, chlorophyll a, and productivity at two stations in the North and South Central Pacific Ocean. Data Depth of 1% light m Chlorophyll a Product! vlty Posi- tion Depth of maximum m Surface concen- tration mg/m3 Concen- tration at (2) mg/m^ Wafer col- umn total 0-200 m mg/m^ % of (5) below (1) % Total above (1) mg C/m^/kr Total below (1) mg C/nfilht (1) (2) (3) (4) (5) Lot 27°N, long 155°W Sept. 1958 Sept. 1959 79± 5 104 ± 8 0.03 ±0.01 0.16 ±0.03 11.92 ±2.62 73.5 ± 8.3 16.26 ±2.25 73 ± 5 in ± 9 0.09 ± 0.03 0.11 ±0.02 11.67± 1.32 53.7 ± 4.6 31.74 ±7.35 >1.'1Z Lot 25° S, long Oct. 1969 Mar. 100 ± 13 122 ± 11 0.03 ±0.01 0.13 ±0.02 12.35 ±3.22 58.5 ± 6.4 12.87 ±9.08 >3.28 155°W 1971 132 ± 14 140± 9 0.01 ±0.00 0.11 ±0.05 8.13 ± 1.47 58.3 ± 9.5 11.80 > 1.20 48 VENRICK, McGOWAN, and MANTYLA: PHOTOSYNTHETIC CHLOROPHYLL internal waves. However, on a wider scale, there appears to be no relationship between the depth of the maximum layer and any one isoline of temperature, salinity, or density. Plant nutrients are present in very low con- centrations in the upper 100 m. Phosphate val- ues were less than 1.5 /xg at./liter in the North Central Pacific and less than 0.2 /xg at./liter in the South Central Pacific. Nitrate was less than 0.6 fxg at./liter in the North and less than 0.8 fxg at./liter in the South, while corresponding values of silicate ranged between 1 and 7 fig at./liter in the North and between and 3 /xg at./liter in the South. The concentrations of these three nutrients increase systematically and significantly at, or just below, the level of the chlorophyll maximum. Concentrations of am- monia are low and irregular throughout the up- per 200 m, showing no pattern with depth. In contrast, high values of nitrite in the upper 200 m (occasionally as high as 4.5 /xg at./liter) were observed only within or just below the chlorophyll maximum, and may indicate recent phytoplankton assimilation of nitrate-nitrogen (Vaccaro and Ryther, 1959). In all of our stud- ies, the relationship between the vertical distri- butions of chlorophyll and nutrients was far more predictable than the relationship between chlorophyll and any of the physical properties. We have found no evidence of any accumula- tion of zooplankton within the chlorophyll max- imum layer. Total zooplankton biomass (ani- mals greater than 35 /x) was greatest above the maximum layer during both day and night. This appears to be true for all size categories. THE SEASONAL CYCLE Seasonal samples from the western edge of the California Current (station 100.80 at lat 30°00'N, long 120°07'W) during 1969 demon- strate a seasonal change in the vertical distribu- tion of chlorophyll a (Figure 5). We have evi- dence that this maximum layer is continuous with that observed within the Central Pacific Gyre and we expect their seasonal cycles to be comparable. For a short period in February, chlorophyll is essentially homogeneous through the upper 50 m. This corresponds in time to the maximum development of the mixed layer. When the water column begins to stratify in March, chlorophyll concentrations at the surface decrease abruptly and a subsurface maximum layer develops. As the summer progresses, the maximum decreases in concentration and the layer subsidies, reaching its maximum depth just prior to the breakdown of the density stratifi-|i cation in December. Figure 5b illustrates the' lack of temporal relationship between the depth of the chlorophyll maximum and any one iso- pleth of density. This would seem to preclude the formation of the maximum layer from the accumulation of cells regulated solely by their physical density. The vertical distribution of chlorophyll dur- ing August along long 155°W is presented in Figure 2. This may be compared with the MLR STA. 100.80 30°00'N I20*'07'W CHLOROPHYLL - a 100 - 150 200 SONDJFMAMJJASONDJFMA 1969 25.50 26.00 SONDJFMAMJJASONDJF MA 1969 Figure 5. — Annual development of the subsurface chlo- rophyll maximum layer at lat 30°00'N, long 120°07'W. Chlorophyll a concentration is contoured with respect to depth (A) and density (B). 49 FISHERY BULLETIN: VOL. 71, NO. 1 distribution along the same transect observed during January (ZETES Expedition) when a similar program of chlorophyll measurement was carried out. In January, north of lat 32°N, the mixed layer extends below 100 m. Concen- trations of chlorophyll are uniform throughout this layer, decreasing abruptly below the mixed layer. Between lat 26°N and 32°N a weak chlorophyll maximum is still present near 120 m below the mixed layer which does not reach its greatest depth of 200 m until February (Rob- inson, 1951). The evidence accumulated to date suggests that a subsurface concentration of plant material can persist only in the presence of a density gradient which isolates the layer from the ef- fects of wind-driven turbulence. Thus, any sea- sonal fluctuations in the strength or depth of the pycnocline may be expected to affect the presence of the deep maximum layer. We can postulate with some assurance that in any environment in which the winter mixed layer regularly ex- ceeds the depth of the chlorophyll maximum layer, the maximum layer must be a seasonal phenomenon. At any locality, the duration of the maximum layer will be determined by the duration of seasonal stratification of the water column and thus will be progressively shorter at higher latitudes. This observation has important implications. Over most, if not all, of the ocean, the phyto- plankton within the maximum layer do not rep- resent a permanent loss to the epipelagic com- munity. Neither need there be a balanced energy budget within the maximum layer. Sufficient energy may be produced and stored in a brief period prior to stratification of the water column or the depletion of nutrients from the surface waters to maintain the population within the maximum layer for considerable periods of time, even though photosynthesis may be depressed or absent. scale. In September 1969, the standing crop and productivity were higher and more variable throughout the water column than in the same month of the previous year. As a result, in 1969, the chlorophyll maximum layer was less sharply defined and was occasionally obscured by high chlorophyll concentrations in the overlying water. These fluctuations are of considerable interest. Namias (1971) has investigated the meteorological and oceanographic conditions ac- companying a vast pool of abnormally warm water in the southern portions of the North Pa- cific during the summer and fall of 1968. He concludes: The abrupt and extensive anomalous warming of the southeastern quarter of the Pacific Ocean north of 20°N from May-June, 1968, appears to have been due largely to increased isolation and horizontal con- vergence of the surface layers of the sea and associated downwelling, .... These warming factors in the heat budget were associated with the development and maintenance of a strong and deep Pacific anticyclone in June which appears to have been persistently re- generated by an unusually strong mean jet around 40°N. This period of stronger subsidence was ac- companied by a clear sharpening of the maxi- mum layer and by reduced standing crop of phy- toplankton and productivity above the layer. The observations of Namias suggest that the gener- alized downwelling in the Central North Pacific anticyclone, which is an important factor inhib- iting the vertical diffusion of nutrients into the euphotic zone, is also closely related to the depth and concentration of photosynthetic material be- low the mixed layer. Extrapolation leads us to expect to find similar chlorophyll maxima well developed in other large, persistent temperate gyres, such as the South Atlantic and the south- ern Indian Ocean. DISCUSSION LARGE-SCALE TEMPORAL FLUCTUATIONS In the Central Gyre of the North Pacific we have recorded temporal fluctuations on a larger It is evident that a deep chlorophyll maximum layer is a well-developed and consistent feature of the major gyres of both the North and South Pacific. In view of its geographic continuity, we must reevaluate the mechanisms postulated for its development, and seek a single explana- 50 VENRICK, McGOWAN, and MANTYLA : PHOTOSYNTHETIC CHLOROPHYLL tion to account for the presence of a chlorophyll maximum layer in very different environmental regimes. Of the numerous hypotheses which have been put forward, several may be relevant. The increase of chlorophyll with depth corres- ponds to an increase in the number of phyto- plankton (diatom) cells, but this may be accentuated by an increase in the amount of chlorophyll per cell. Zooplankton have been shown to be concentrated above the maximum layer and differential grazing pressure may help to maintain the abrupt gradient at the top of the maximum layer. In situ production has been demonstrated within the maximum layer at very low light intensities, and this will contribute to the formation and maintenance of the layer. The strong development of a deep maximum within the oligotrophic environments of the Central Gyres, the effect of fluctuations of the rate of downwelling on the depth of the max- imum layer and the productivity in the overly- ing water, and the consistent relationship be- tween the depth of the maximum layer and the depth of the nutricline and the nitrite maximum suggest that the nutrient regime may be a crit- ical factor in the development and maintenance of the chlorophyll maximum layer. Our obser- vations to date support the theory of Steele and Yentsch (1960) that depletion of nutrients above the summer thermocline leads to a reduction in the buoyancy of phytoplankton, and that a sub- surface maximum results from the accumulation of impoverished cells at the top of the nutricline where the absorption of nutrients decreases the sinking velocity (Eppley, Holmes, and Strick- land, 1967). The maximum layer may continue to subside slowly as the nutrients at the top of the nutricline are depleted, and thus it may be that the depth of the maximum layer is ulti- mately determined by the nutrient regime, rather than ambient light intensity. As long as the depth does not exceed the maximum depth of the winter mixed layer the cells will be returned to higher light levels during the winter. It may be that the chlorophyll maximum layer repre- sents a senescent stage in the annual cycle of oceanic phytoplankton which is analogous to the formation of resting spores by many neritic species. It is evident that the chlorophyll maximum layer, which may account for a major portion of the standing crop of plant material and for a substantial portion of the primary productivity, is not restricted to the traditionally defined "eu- photic" zone, the zone above the ISr light level. There is no justification for limiting samples for chlorophyll or productivity measurements to this zone. These programs must be extended below the chlorophyll maximum layer. We expect this will result in a substantial increase in the es- timates of total production within the water column. ACKNOWLEDGMENTS The work was supported in part by National Science Foundation Grant GB-12413 and in part by the Marine Life Research Program, the Scripps Institution of Oceanography's part of the California Cooperative Oceanic Fisheries In- vestigations, which are sponsored by the Marine Research Committee of the State of California, and by the National Science Foundation Grant GB-2861. We thank Gary B. Smith for assistance in many phases of this program and members of the Scripps Institution of Oceanography Data Collection and Processing Group for the collec- tion and processing of the hydrographic and chemical data. The seasonal data from station 100.80 was supplied by R. W. Owen and M. G. Kruse of the National Marine Fisheries Service. LITERATURE CITED Allen, W. E. 1932. Problems of flotation and deposition of ma- rine plankton diatoms. Trans. Am. Microsc. Soc. 51:1-7. 1939. Summary of results of twenty years' re- searches on marine phytoplankton. Proc. 6th Pac. Sci. Congr. 3:577-588. Anderson, G. C. 1969. Subsurface chlorophyll maximum in the Northeast Pacific Ocean. Limnol. Oceanogr. 14: 386-391. Austin, R. W., and R. W. Loudermilk. 1968. An oceanographic illuminometer for light penetration and reflection studies. S.I.O. (Univ. Calif., Scripps Inst. Oceanogr.) Ref. 68-11, 10 p. 51 FISHERY BULLETIN: VOL. 71, NO. 1 Beers, J. R., G. L. Stewart, and J. D. H. Strickland. 1967. A pumping system for sampling small plank- ton. J. Fish. Res. Board Can. 24:1811-1818. Doty, M. S., and L. R. A. Capurro. 1961. Productivity measurements in the world ocean. IGY (Int. Geophys. Year) Oceanogr. Rep. 4, Part I, 625 p. El-Sayed, S. Z. 1970. Phytoplankton production of the South Pa- cific and the Pacific sector of the Antarctic. In W. S. Wooster (editor), Scientific exploration of the South Pacific, p. 194-210. Natl. Acad. Sci. Eppley, R. W., R. W. Holmes, and J. D. H. Strickland. 1967. Sinking rates of marine phytoplankton mea- sured with a fluorometer. J. Exp. Mar. Biol. Ecol. 1:191-208. Koblentz-Mishke, 0. J., V. V. Volkovinsky, and J. G. Kabanova. 1970. Plankton primary production of the world ocean. In W. S. Wooster (editor). Scientific ex- ploration of the South Pacific, p. 183-193. Natl. Acad. Sci. Lorenzen, C. J. ' 1965. A note on the chlorophyll and phaeophytin content of the chlorophyll maximum. Limnol. Oceanogr. 10:482-483. 1967. Vertical distribution of chlorophyll and phaeo- pigments: Baja California. Deep-Sea Res. 14: 735-745. 1970. Surface chlorophyll as an index of the depth, chlorophyll content and primary productivity of the euphotic layer. Limnol. Oceanogr. 15:479- 480. Love, C. M. (editor). 1970. EASTROPAC atlas. Vol. 4. U.S. Dep. Com- mer., Natl. Mar. Fish. Serv., Circ. 330. 1971. EASTROPAC atlas, Vol. 2. U.S. Dep. Com- mer., Natl. Mar. Fish. Serv., Circ. 330. Menzel, D. W., and J. H. Ryther. 1960. The annual cycle of primary production in the Sargasso Sea off Bermuda. Deep-Sea Res. 6:351-367. Motoda, S., and R. Marumo. 1963. Plankton of the Kuroshio water. Proceed- ings of Symposium on the Kuroshio, p. 40-61. Oceanographical Society of Japan and UNESCO, Tokyo, 29 Oct. 1963. Namias, J. 1971. The 1968-69 winter as an outgrowth of sea and air coupling during antecedent seasons. J. Phys. Oceanogr. 1:65-81. Ocean Research Institute, University of Tokyo. 1970. Preliminary report of the Hakuho Maru cruise KH-69-4. Univ. Tokyo, Ocean Res. Inst., 68 p. Owen, R. W., and B. Zeitzschel. 1970. Phytoplankton production: seasonal change in the oceanic eastern tropical Pacific. Mar. Biol. (Berl.) 7:32-36. Robinson, M. K. 1951. Sea temperatures in the North Pacific area, 20°-40°N, 125°-155°W. S.I.O. (Univ. Calif., Scripps Inst. Oceanogr.) Ref. 51-20, 14 p. Saijo, Y., S. Iizuka, and O. Asaoka. 1969. Chlorophyll maxima in Kuroshio and adja- cent area. Mar. Biol. (Berl.) 4:190-196. Sano, a. 1966. Distribution of microplankton on a vertical section along 39°30'N, 142°-150°E. La Mer. Bull. Soc. Fr.-Jap. Oceanogr. 4:4-12. SOLORZANO, L. 1969. Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Ocean- ogr. 14:799-801. Sorokin, Yu. I. 1970. Some data on primary production in the cen- tral Pacific. [In Russian.] Okeanologya 10:691- 694. (Transl. in Oceanology 10:538-542, issued by Am. Geophys. Union.) Steele, J. H. 1964. A study of production in the Gulf of Mexico. J. Mar. Res. 22:211-222. Steele, J. H., and C. S. Yentsch. 1960. The vertical distribution of chlorophyll. J. Mar. Biol. Assoc. U.K. 39:217-226. Steeman Nielsen, E. 1952. The use of radio-active carbon (C^^) for measuring organic production in the sea. J. Cons. 18:117-140. Strickland, J. D. H., and T. R. Parsons. 1968. A practical handbook of seawater analysis. Fish. Res. Board Can., Bull. 167, 311 p. University of California. 1967. Physical, chemical and biological data, URSA MAJOR Expedition, 4 August-4 October 1965. S.I.O. (Scripps Inst. Oceanogr.) Ref. 67-5, 43 p. 1970. Physical, chemical and biological data, ZETES Expedition, Leg I. S.I.O. (Scripps Inst. Oceanogr.) Ref. 70-5, 67 p. Vaccaro, R. F., and J. H. Ryther. 1959. Marine phytoplankton and the distribution of nitrite in the sea. J. Cons. 25:260-271. Venrick, E. L. 1969. The distribution and ecology of oceanic dia- toms in the North Pacific. Ph.D. Thesis, Univ. California, San Diego, 684 p. 1971. Recurrent groups of diatom species in the North Pacific. Ecology 52:614-625. Yentsch, C. S. 1965. Distribution of chlorophyll and phaeophytin in the open ocean. Deep-Sea Res. 12:653-666. 52 THE NAUPLIUS II, METANAUPLIUS, AND CALYPTOPIS STAGES OF THYSANOPODA TRICUSPID AT A MILNE-EDWARDS (EUPHAUSIACEAJ Margaret D. Knight' ABSTRACT A large, spinose metanauplius, a nauplius II, and calyptopis I, found in the Indian Ocean and Equatorial Pacific, are referred to Thysanopoda triciispidata. The identifications are based on the distinctive morpholog-ical features shared by these larval stages and by the calyptopes II and III of T. tricuspidata identified by Sars (1885), and on the observed distribution of T. tricuspidata and the metanauplius in the Indian Ocean. Calyptopes II and III are redescribed to present the complete calyptopis phase of larval development in one account. During a survey of euphausiids in the Indian Ocean (Brinton and Gopalakrishnan, in press), many specimens of a relatively large and very ornate metanauplius were found. There was conjecture that the curious, apparently unde- scribed form might be the larva of a species of the genus Thysanopoda, and it was sought for next in plankton from the Equatorial Pacific. The metanauplius was found in these waters as were specimens of a seemingly related nau- plius II and calyptopis I together with the calyp- topis II, calyptopis III, furcilia, and juvenile stages 0/ Thysanopoda tricuspidata identified by Sars (1885). When individuals of each of the larval stages were placed together, they appeared to form a natural developmental series; their relative size, the distinctive shape of developing eyes, telson, and carapace all suggested that the larvae were progressive stages of the same spe- cies. Evidence of their specific relationship was found in a detailed study of these features and of the morphology of larval appendages, and there seemed to be sufficient justification for re- ferral of the three unidentified early stages to T, tricuspidata. Redescriptions of the calyptopes II and III of T. tricuspidata are included in this paper with ' Scripps Institution of Oceanography, University of California at San Diego, La JoUa, CA 92038. Manuscript accepted June 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. identification and description of the nauplius II, metanauplius, and calyptopis I, in order to pre- sent the complete calyptopis phase of the spe- cies in one account and to illustrate the setation of appendages more fully. METHODS AND MATERIALS Specimens of the metanauplius were observed in the standard collections, approximately 200-0 m depth, obtained during the International Indian Ocean Expedition (IIOE), 1962-65. About 100 metanauplii were removed for study. The distributions of T. tricuspidata and the metanauplius based on the data of Brinton and Gopalakrishnan are shown in Figure 9. Selected samples taken during EQUAPAC Expedition by RV Stranger of the Scripps In- stitution of Oceanography in August-September 1956 between long 165°-175°W and lat 6°S-10°N by oblique tow in the top 200 m (Snyder and Fleminger, 1965) were sorted for the metanaup- lius and calyptopes. Positions of the samples yielding larvae and the developmental stages found in each sample are given in Table 1. The distribution of T. tricuspidata in the Pacific is described by Brinton (1962). For measurement with an ocular micrometer, the larvae were straightened in a few drops of 53 FISHERY BULLETIN: VOL. 71, NO. I Table 1. — Location of samples collected by RV Stranger during EQUAPAC Expedition which contained larvae of Thysanopoda tricuspidata and the developmental stages found. Detailed station data for samples are given by Snyder and Fleminger (1965). Position Sample (net) Developmental stage Station Nouplius Metonou plius Calyptopis No. Lot Long 1 II III IS 5°59'N I66''40'W 45 cm — + + + _ 21 QOOO'N 166°55'W 45 cm — + .« 25 4°00'S 167°03'W 45 cm — + + + + 25 4°00'S 167°03'W 1 m — + + + + 26 S-OO'S 167''08'W 45 cm + + + + + 26 S'OO'S 167°08'W 1 m + + — + + 26 5°58'S i75°02'W 45 cm + + + + + 28 5°58'S 175°02'W 1 m + + + — + 28 5°58'S 175°02'W 1 m + + + + 29 S'OVS 174''59'W 45 cm — + — + + 2% formaldehyde in seawater on a slide. Total length (TL) was measured in dorsal view be- tween center of anterior margin of carapace (excluding spines in metanauplius) or rostrum and distal point on posterior margin of telson excluding spines. Other measurements are ex- plained by stage: nauplius II, width (W) at widest point in dorsal view; metanauplius, car- apace length (CL) between midpoints of anter- ior and posterior margins excluding spines, car- apace width (CW) at widest point between anterolateral margins excluding spines, both measured in dorsal view; calyptopes I and II, carapace length (CL) between midpoints of an- terior and posterior margins measured in lateral view; calyptopis III, carapace length (CL) from rostrum to distal point on posterior margin measured in lateral view. The range (r) and mean (m) of each measurement and number (n) of specimens measured are given by stage. Approximately equal numbers of the meta- nauplius stage from the Indian and Pacific Oceans were measured. The measurements given for nauplius II and calyptopes I-III, how- ever, are based only on larvae from the Equa- torial Pacific and, as calyptopis larvae of a single species have been shown to vary in size in dif- ferent areas of the oceans (Mauchline and Fish- er, 1969) , it should be emphasized that the larvae measured for this study were collected during one season in one area of the Pacific. Measure- ments of some nauplius II and calyptopis stages sorted from Indian Ocean samples did fall well within the size ranges of Pacific larvae in equiv- alent developmental stages. Specimens of a nauplius I definitely referable to T. tricuspidata were not found. For detailed study and dissection of append- ages, larvae were placed in glycerine. Some were stained with Chlorazol Black E to clarify appendage setation. Fourteen nauplii and at least 20 individuals of each of the metanauplius and calyptopis stages I-III were examined in detail. At least 10 specimens of each stage were dissected for study of appendages. In a study of the larval development of Nematoscelis difficilis based on both larvae reared in the lab- oratory and larvae from the plankton, Gopala- krishnan (in press) found no variability in form or setation of appendages among individuals at the same stage of development. This also ap- pears to be true of T. tricuspidata larvae, in the stages described, with respect to the mouthparts where setation is usually intact in preserved specimens. On antennules, however, the ter- minal setae, spines, and aesthetascs (sensory set- ae) were frequently broken; in calyptopis II, for instance, only 1 of 26 antennules examined had the third seta intact on the inner and outer fla- gella. An estimate of variability in the fragile setation in this species will require a detailed study of larvae either reared in the laboratory or collected specifically for the purpose. Drawings of both whole larvae and append- ages were prepared with the Wild M20' com- ' Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. 54 KNIGHT; STAGES OF THYSANOPODA TRICUSPWATA pound microscope equipped with drawing at- tachment. Nomenclature for description of appendages is based on that of Gurney (1942) . For a review of the literature on larval development of the Euphausiacea and the nomenclature of their larval phases, the reader is referred to the papers by Gopalakrishnan (in press) and by Mauchline and Fisher (1969). RESULTS DESCRIPTION OF DEVELOPMENTAL STAGES Nauplius II (Figure la, b) Measurements: TL, r = 1.00-1.12 mm, m = { 1.06 mm; W, r = 0.48-0.56 mm, m = 0.53 mm; n = 43. jj Body oval, about 2 times as long as wide, an- ' terior pointed, posterior truncate; posterior margin armed with 10 spines — 3 pairs of pos- terolateral spines and 2 pairs of small to rudi- mentary medial terminal spines, posterolateral i spine 1 (outer) is small, spines 2 and 3 are rel- atively large, only spine 3 (inner) bears spinules. Antennule (Figure 4a) uniramous; with 2 terminal setae, 1 subterminal seta situated me- dioventrally, and small spiny prominences at base of each seta — that below largest terminal seta is like small lobe; spinules are distributed on surface as figured. Antenna (Figure 5a) biramous; protopod may be constricted near middle and appear weak- ly segmented; endopod unsegmented with 3 term- inal plumose setae, a rudiment of 4th terminal seta, 1 subterminal seta on inner margin, and rows of spinules at bases of setae; exopod with outer margin divided into about 10-11 segments (the segmentation was often indistinct in most distal and proximal parts of exopod) , the 5 distal segments bear plumose setae — the terminal seg- ment has 2 setae with a few spinules and the remaining 4 segments bear 1 seta each. Mandible (Figure 6a) biramous and unseg- mented ; both rami bear 3 plumose setae with spinules at base of setae. In well-developed nauplii nearing molt to metanauplius, the carapace of the metanauplius with its distinctive ornamentation could be seen inside the cuticle of the nauplius (Figure la), both the large long spines around the anterior margin which fold up and back around the body and the 4 large medial and smaller posterolateral spines on posterior margin may be visible and may be partially dissected out. Metanauplius (Figure Ic, d) Measurements: Equatorial Pacific larvae - TL, r = 1.36-1.50 mm, m = 1.43 mm; CL, r = 1.02-1.10 mm, m = 1.05 mm; CW, r = 0.60- 0.70 mm, m = 0.64 mm; n — 39. Indian Ocean larvae - TL, r = 1.36-1.52 mm, m = 1.45; CL, r = 0.98-1.10 mm, m = 1.05 mm; CW, r = 0.61-0.68 mm, m = 0.65, n = 43. Carapace with rounded frontal and anterolat- eral margins produced into long spines (the num- ber of spines, counted in 25 individuals, ranged from 21 to 23 with 23 larvae having 22 spines), there may be tiny spines or "hairs" posterior to the posteriorly directed last large spine; pos- terolaterally deep winglike extensions of car- apace curve ventrolaterally with margins pro- duced into strong posteriorly directed spines which diminish in size around posterior margin where they are separated by small spines; the 4 large medial spines on posterior margin are usually relatively long and they project up dor- sally away from body of larva. A faint outline of developing eyes is visible. Tail long and taper- ing with rounded posterolateral margins and median indentation, there is now a pair of lat- eral spines in addition to 3 pairs of posterolateral spines and 2 pairs of medial terminal spines, a small rudiment of one or both of inner (third) pair of terminal spines may be present. In one well-developed metanauplius near molt, the telson of calyptopis I with invaginated ter- minal and lateral spines was visible beneath the cuticle (Figure le). As can be seen, postero- lateral spine 3, although shorter than spine 2 in the metanauplius, is more deeply invaginated and longer than spine 2 in the developing calyp- topis, and when extruded, it will have the greater relative length observed in the calyptopis stages of T. tricuspidata. 55 FISHERY BULLETIN: VOL. 71, NO. 1 a Q-d Figure 1. — Nauplius: a-b, dorsal and lateral views. Metanauplius: c-d, dorsal and lateral views; e, posterior enlarged showing invaginated spines of calyptopis I beneath cuticle. Figure 2. — Calyptopis: a-c, stages I-III, dorsal views. 56 KNIGHT: STAGES OF THYS.4NOPODA TRICUSPID.4TA a-b 0.5 mm c 57 FISHERY BULLETIN: VOL. 71, NO. 1 fir^, v. Figure 3. — Calyptopis: a-c, stages I-III, lateral views. 58 KNIGHT: STAGES OF THYSANOPODA TRICUSPIDATA 0, 1 mm I 1 a-d Figure 4. — Antennule, right, dorsal view: a, nauplius; b, metanauplius; c-e, calyptopes I-III. Antennule (Figure 4b) now with 3 terminal processes, there is a small seta or sensory fil- ament in place of spiny lobe; surface spinules appear to be organized into fewer simpler rows. Antenna (Figure 5b) with protopod segment- ed into coxa and basis, there are a few spinules on inner distal margin of each; endopod with 4 terminal setae — 1 seta is relatively short, and 2 setae on inner margin — the proximal seta is short to rudimentary; exopod with 5 short term- inal segments but without proximal segmenta- tion of outer margin seen in nauplius, the seta- tion is unchanged although there may be a spin- ous rudiment of third seta on terminal segment. Mandible (Figure 6b) reduced to rounded lobe bearing pointed lateral process, Maxillule, maxilla, and maxilliped are repre- sented only be rounded prominences. In spe- cimens nearing molt, the rudimentary spines on endopod and endites of developing maxillules and 59 FISHERY BULLETIN: VOL. 71, NO. 1 0. I mm a C Figure 5. — Antenna, right, anterior view: a, nauplius; b, metanauplius ; c, calyptopis I. maxillae and setae of biramous maxilliped of calyptopis I may be seen through the cuticle. Figure 7a and d show such a maxillule and max- illa dissected from a metanauplius providing evi- dence of its relationship with the calyptopis I described. Calyptopis I (Figures 2a, 3a) Measurements: TL, r = 1.90-2.16 mm, m = 2.07 mm; CL, r ^ 1.38-1.48 mm, m = 1.44 mm; n = 34. Carapace long and slender, without spines, anterior margin forming narrow hood over de- veloping eyes, posterior margin pointed and curving dorsally. Abdomen unsegmented, telson with a pair of lateral spines, 3 pairs of postero- lateral spines, and 3 pairs of medial terminal spines, posterolateral spine 3 is now longest; posterior margin curves in medially. 60 KNIGHT: STAGES OF THYSANOPODA TRICUSPJDATA a f 0.1 mm Figure 6. — Mandible: a, nauplius, right, anterior view; b, metanauplius, left, posterior view. Mandibles, posterior view: c-e, calyptopes I-III; f, right mandible of calyptopis II rotated to show relative length of triangular lateral process. Antennule (Figure 4c) 2-segTnented ; proto- pod long and slender with small terminal seg- ment forming outer flagellum which bears about 9 terminal processes including 2 long setae, 2 aesthetascs (one of these is situated slightly subterminally on outer margin), and about 5 spinous processes of varying sizes; there is rudi- ment of inner flagellum bearing 1 long seta and 3 spinous processes; a small spine is situated at base of inner ramus, and there is 1 seta and 1 or 2 small spines dorsally on distal margin of protopod at base of outer flagellum. Antenna (Figure 5c) now with form found in calyptopis stages I-III; endopod with 4 long terminal setae and 2 setae on inner margin, prox- imal seta still relatively short; exopod with 7 plumose setae, the terminal segment now bears 3 setae; coxa and basis without spinules. Mandibles (Figure 6c) rudimentary, with large lateral process, medial margins smooth ex- cept for 1 small incisor tooth on each mandible. Maxillule (Figure 7b) armed only with rudi- mentary small spines; endopod of 1 segment with 3 spines; exopod a very small lobe bearing 61 FISHERY BULLETIN: VOL. 71, NO. 1 a 0.1 mm f Figure 7. — Maxillule, left, posterior view: a, developing appendage of calyptopis I dissected from metanauplius ; b-c, calyptopes I and II. Maxilla, left, posterior view: d, developing appendage of calyptopis I dissected from metanauplius; e-f, calyptopes I and II. 2 plumose setae; basal endite with 2 spines, there may be a tiny third spine between large spines; coxal endite with about 5 spines. Maxilla (Figure 7e) with rudimentary seta- tion except for 1 plumose seta arising from small finely setose lobe on lateral margin and represent- ing exopod ; endopod of 1 segment with 2 spines; medial lobes of endites discernible with small spines on medial margin. Maxilliped (Figure 8a) biramous; exopod with 4 terminal plumose setae and 1 subterminal seta on outer margin, also a small stout seta at base of exopod near articulation with basis; en- dopod of 2 segments, terminal segment with 4 setae distally, 3 terminal and 1 subterminal; there are a few weak setae and rudiments of setae on medial margins of both coxa and basis and of proximal segment of endopod. 62 KNIGHT: STAGES OF THYSANOPODA TRICVSPIDATA I Figure 8. — Maxilliped, left, posterior view: a-c, calyptopes I-III. Uropod, left, ventral view: d, caljiitopis III. 63 FISHERY BULLETIN: VOL. 71, NO. I Calyptopis II (Figures 2b, 3b) Measurements: TL, r = 2.50-2,74 mm, m = 2.63 mm; CL, r = 1.42-1.54 mm, m = 1.49 mm; n = 18. Carapace with frontal margin produced into small triangular spine and posterior margin more pointed than in preceding stage; develop- ing eyes may contain some pigment. Thoracic segments forming; abdomen segmented, sixth segment not separate from telson; posterior margin of telson with small median 7th ter- minal spine. Antennule (Figure 4d) with protopod divided into 3 peduncular segments, there is stout seta distally on inner margin of second segment and a small dorsal lobe bearing 2 setae and a few small spines on distal margin of third segment at base of outer flagellum; outer flagellum with about 9 terminal processes including 2 setae, 2 aesthetascs, and about 4-6 spinous processes; inner ramus with about 6 terminal processes in- cluding 1 seta and usually 5 spines, there is 1 subterminal seta on inner margin. Antenna as in calyptopis I. Mandibles (Figure 6d) asymmetrical, now dif- ferentiated into incisor and molar areas, right mandible with slender articulated spine with spinule situated near molar area; right mandible rotated in Figure 6f to show relative length of lateral process. Maxillule (Figure 7c) with setae and spines fully formed; endopod of 1 segment with 3 setae; exopod with 3 plumose setae; basal endite with 4 stout spines armed with spinules; coxal endite with 6 setae — 2 are small smooth setae, 4 are setose and the largest bears strong spinules dis- tally. Maxilla (Figure 7f) with full setation; en- dopod of 1 segment with 2 setae; exopod rep- resented by a single plumose seta on small setose lobe; basal endite with 3 medial lobes, coxal en- dite bilobed, lobes 1-5 with setation of 5-4-4-3-1 progressing distally, 1 seta on each of lobes 1-3 is situated on posterior face of maxilla, 1 mar- ginal seta on lobe 2 is quite small. Maxilliped (Figure 8b) now with full medial setation; coxa with 4 plumose setae, 1 seta is relatively long; basis with 5 setae; proximal segment of endopod with 3 setae; 1 distal seta on basis and 1 distal seta on first segment of endopod are situated slightly submarginally on posterior face, both are small and frequently dif- ficult to locate; setation of exopod and terminal segment of endopod is unchanged. Calyptopis III (Figures 2c, 3c) Measurements: TL, r = 2.90-3.40 mm, m = 3.14 mm; CL, r = 1.30-1.42 mm, m = 1.35 mm; n = 34. Larva now appears more slender for its length; carapace considerably altered, still forming nar- row hood over eyes but in other respects more like carapace of furcilia, frontal margin pro- duced into small triangular rostrum, lateral margins with small anterolateral spine below eye and large posterolateral denticle, posterior dorsal margin no longer tapering to point but indented medially. Eye with 7 well-developed facets arranged in a circle of 6 with seventh central facet, and ommatidia with pigment. Thoracic segmentation more distinct. Abdomen with 6 segments, there is dorsal ridge or fold around segment 1 and segment 6 carries bira- mous uropods. Setation of telson unchanged. Antennule (Figure 4e) with distal lateral mar- gin of basal peduncular segment produced into strong lateral spine which extends to or beyond midpoint of distal segment of peduncle; there are about 5 groups of 2 setae each along inner margin of this spine with spinules between 3 distal groups and a seta at base of spine on both outer and inner margins; basal segment of pe- duncle dorsoventrally flattened; the peduncular segments bear plumose setae along medial mar- gins with 2-2-3 setae on segments 1-3 respective- ly, there are 3 small setae around distal margin of segment 2, and 3 setae and setules on dorsal lobe below outer ramus on distal margin of seg- ment 3; outer flagellum with 2 aesthetascs, 3 setae, and about 4 small spines ; inner flagellum with 3 terminal setae and about 3 spinous pro- cesses. Antenna as in calyptopis I. Mandibles (Figure 6e) similar to calyptopis II, medial teeth somewhat flattened. Maxillule and maxilla as in calj^Dtopis II. 64 KNIGHT: STAGES OF THYSANOPODA TRICVSPIDATA Maxilliped (Figure 8c) with 5 setae on me- dial margin of coxa; there is no other change in setation. No rudiment of the second thoracic appendage was observed. Uropod (Figure 8d) biramous; protopod with stout ventral spine; exopod produced into pos- terolateral spine and bearing 7 plumose setae around posterior and medial margins, the seta near posterolateral spine is relatively small; en- dopod with 5 distal plumose setae, 1 seta is sit- uated submarginally and projects dorsally, IDENTIFICATION OF EARLY STAGES The morphological evidence on which the identification of the larval series is based may be summarized as follows: 1) the nauplius II is linked to the metanauplius by dissection of the spinose carapace of the metanauplius from well- developed nauplii; 2) the metanauplius and ca- lyptopes I-III are related by the setation of mouthparts, particularly the endopods of max- illule and maxilla; 3) the third calyptopis is identified with the larva described by Sars (1885) by the long and slender body, the dis- tinctive 7-facetted eyes, and the setation of the exopod and 1-segmented endopod of the max- illule which bear 3 setae each. There is additional evidence to support the identification of the metanauplius in the way in which the observed distribution of the larva cor- responds with that of T. tricuspidata in the In- dian Ocean as shown in Figure 9, and in the oc- currence of the larvae within the range of T. tricuspidata in the Pacific. DISCUSSION The only description of T. tricuspidata larvae found which deals with the calyptopis stages is that of Sars (1885); other authors referring to larvae of the species (i.e., Tattersall, 1936; : Gurney, 1947; Lebour, 1950; Pillai, 1957) dis- cuss the furcilia stages only. Sars provides some details of setation with his general descriptions I and figures the mandible, maxillule, maxilla, and maxilliped of the third calyptopis (1885, Plate 21, Figures 13-16). The mouthparts of the ca- lyptopis III described in this study agree with those figured by Sars in the dentition of left mandible, in segmentation and setation of endo- pod and exopod of maxillule, in rudimentary ex- opod of maxilla, and in setation of exopod and terminal segment of endopod of maxilliped. The carapace of Sars' calyptopis III appears to be indented medially on the posterior margin rather than pointed as in calyptopis II, but it is not fig- ured with a lateral denticle. The descriptions of larvae of other species of the genus Thysanopoda are also almost entirely limited to the furcilia phase; only two excep- tions were found. Einarsson (1945) described the calyptopes II and III of T. acutifrons and Lebour (1950) the calyptopis III of T. cristata, but figures of the appendages and the details of setation are not given. The described larvae of T. acutifrons are larger than those of T. tricuspidata in equivalent stages; calyptopes II and III measure 3.4 and 3.8 mm in total length respectively while T. tri- cuspidata averages 2.6 and 3.1 mm (Sars' spec- imens measured 2.5 and 3.5 mm) . The carapace of the third calyptopis of T. acutifrons is like that of the second calyptopis with "character- istic pointed end," and the lateral denticle is sometimes discernible although very small. Einarsson notes that the maxillule has a palp of 2 segments and an inner lobe with 7 bristles. Frost (1939) figures the appendages of the first furcilia of T. aciitifrons showing the maxillule with 6 setae on the endopod and 4 setae on the exopod, and the endopod of the maxilla with 3 setae. This setation is probably also found on the calyptopis III of the species she describes. [Einarsson (1945) suggested that, based on the shape of the eye. Frost's larvae may instead be- long to T. microphthahna; he notes, however, that the species are otherwise alike in develop- ment.] Lebour (1950) describes and figures the car- apace of the calyptopis III of T. cristata as long and "pointed behind," noting that the larva closely resembles the calyptopis III of T. acuti- frons described by Einarsson. It is also very large, measuring 4.2 mm in length. Gurney (1947). in his description of the first furcilia 65 FISHERY BULLETIN: VOL. 71, NO. 1 30' 20° 30° 40° 50° 60° 70° 80° 90° 100° llQ' 120° 130° 140° 150 ' .,_ • - - — jr-TT-T I —r-- 1 ' ' \ \ <; ''"° Thysanopoda tricuspidata //// distribution of species metonouplius ^jqo 0° — 10° 20°\ 30° 20° 40° 60° 80° 100° 120° 140° \ Figure 9. — The distribution of Thysanopoda tricuspidata and the metanauplius larva in the Indian Ocean based on the analyses of Brinton and Gopalakrishnan (in press). of T. cristata, notes that the maxillule has an endopod of 2 segments, and again it seems likely that this is also found in the third calyptopis. Both T. acutifrons and T. cristata, then, differ from T. tricuspidata in length of described stages, in shape of carapace in calyptopis III, and probably in details of segmentation and se- tation of maxillule and maxilla at least. Calyptopes I and II of T. monacantha (iden- tified by E. Brinton) were dissected to compare the endopods of the maxillule and maxilla with those of calyptopes I and II of T. tricuspidata. The calyptopis I had full setation of mouthparts and, in both stages, the maxillule, like that of Frost's furcilia, had an endopod of 2 segments with 6 setae and exopod with 4 setae, and there were 3 setae on the endopod of the maxilla. In fact, more setae were found on all of the mouth- parts of the T. monacantha larvae with the ex- ception of the endopod and exopod of the maxil- liped which were like those of the T. tricuspidata calyptopes. Information in these few accounts from the literature and from personal observation sug- gest that T. tricuspidata larvae may prove to differ from larvae of other species of the genus 66 KNIGHT: STAGES OF THYSASOPODA TRICUSPIDATA in many respects. The partial segmentation of antennal exopod in the nauplius II and the form and armature of carapace and telson of the meta- nauplius may be distinctive, the rudimentary setation of mouthparts in calyptopis I appears to be unusual — indeed the larva seems ill- equipped to feed, there seems to be a reduction in dentition of mandibles and in numbers of setae on mouthparts in calyptopes II and III, and the carapace of calyptopis III is transitional between the usual calyptopis and furcilia forms. In ad- dition, the larvae are known to deviate from trends within the genus in development of ab- dominal pleopods during the furcilia phase. Ac- cording to Lebour (1950), T. tricuspidata is the most variable in pleopod succession of any Thysanopoda species, indeed of any euphausiid known and, as it has been demonstrated that there is a correlation between a more rigidly de- fined number of furciliar stages and a more oceanic distribution within the genus Thysan- opoda (Mauchline and Fisher, 1969), such var- iability in the oceanic species T. tricuspidata is surprising. Although there is too little information avail- able at this time for speculation as to the sig- nificance of the unusual morphological features observed in this study, the details found in the literature did support the identification of the larvae in that the combination of setation of endopod and exopod of the maxillule and endopod of the maxilla of T. tricuspidata calyptopes was not noted or figured in descriptions of the larvae either of other species of Thysanopoda or of other genera of the family. ACKNOWLEDGMENTS I wish to thank E. Brinton for his assistance and criticism of the manuscript, and we both wish it to be known that the plankton sorting staff at the Indian Ocean Biological Centre first identified the unique metanauplius as a euphau- siid. This work was supported by the Marine Life Research Program, the Scripps Institution of Oceanography's component of the California Cooperative Oceanic Fisheries Investigations, a project sponsored by the Marine Research Committee of the State of California, and by the Oceanography Section, National Science Foundation, NSF Grant GA-31783. LITERATURE CITED Brinton, E. 1962. The distribution of Pacific euphausiids. Bull. Scripps Inst. Oceanogr., Univ. Calif. 8:51-270. Brinton, E., and K. Gopalakrishnan. In press. The distribution of Indian Ocean euphau- siids. EiNARSSON, H. 1945. Euphausiacea. 1. North Atlantic species. Dana Rep. Carlsberg Found. 27, 185 p. Frost, W. E. 1939. Larval stages of the euphausiid Thysanopoda acutifrons (Holt and Tattersall) taken off the southwest coast of Ireland. Proc. R. Ir. Acad. Sect. B 45:301-319. Gopalakrishnan, K. In press. Developmental and growth studies of the euphausiid Nematoscelis difficilis (Crustacea) based on rearing. Gurney, R. 1942. Larvae of decapod Crustacea. Ray Soc. (Lond.) Publ. 129, 306 p. 1947. Some notes on the development of the Eu- phausiacea. Proc. Zool. Soc. Lond. 117:49-64. Lebour, M. V. 1950. Some euphausiids from Bermuda. Proc. Zool. Soc. Lond. 119:823-837. Mauchline, J., and L. R. Fisher. 1969. The biology of euphausiids. Adv. Mar. Biol. 7, 454 p. PiLLAI, N. K. 1957. Schizopoda. Bull. Cent. Res. Inst., Univ. Travencore, Ser. C, 5:1-28. Sars, G. O. 1885. Report on the Schizopoda collected by H. M. S. "Challenger" during the years 1873-76. Chal- lenger Rep., Zool. 13(3):l-225. Snyder, H. G., and A. Fleminger. 1965. A catalog of zooplankton samples in the ma- rine invertebrate collections of Scripps Institution of Oceanography. S.I.O. (Univ. Calif., Scripps Inst. Oceanogr.) Ref. 65-14, 140 p. Tattersall, W. M. 1936. Mysidacea and Euphausiacea. Sci. Rep. Great Barrier Reef Exped., 1928-29 5:143-176. 67 THE LARVAL STAGES OF THE DEEP SEA RED CRAB, GERYON QUINQUEDENS SMITH, REARED UNDER LABORATORY CONDITIONS (DECAPODA: BRACHYRHYNCHA) Herbert C. Perkins^ ABSTRACT A prezoeal stage, four zoeal stages, and one megalopa stage were obtained from eggs of Geryon quinquedens Smith hatched in the laboratory. Each zoeal stage and the megalopa are discussed and illustrated. The commercial potential and abundance of the deep sea red crab, Geryon quinquedens Smith, are discussed by Schroeder (1959), McRae (1961), and Holmsen (1968). The red crab is obviously an important constituent of the deep- water benthic fauna found on the continental shelf off New England and the middle Atlantic states, and its larvae should therefore occur in considerable numbers in the plankton of that region. Knowledge of the larval stages of this species is apparently totally lacking. Brattegard and Sankarankutty (1967) have described the prezoea and the first zoea of Geryon tridens Kroyer from Norway, but I can find no other reference to the larval stages of this genus. It is the purpose of this paper to describe the larval stages of Geryon quinquedens so that they may be identified in plankton collections and thus facilitate the understanding of the early life his- tory of this species and hopefully to shed light on its apparently tenuous taxonomic status with- in the Brachyrhyncha. METHODS AND MATERIALS In February 1971, several berried females of Geryon quinquedens were captured in 300 fm of water (bottom temperature was 5.9°C) in the Baltimore Canyon area of the continental shelf (lat 37°56'N, long 73°55'W) off Delaware Bay. ' Northeast Fisheries Center, National Marine Fish- eries Service, NOAA, West Boothbay Harbor, ME 04575. Manuscript accepted July 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. Three of the berried crabs were returned to the Boothbay Harbor Laboratory and maintained in shallow tanks at temperatures that ranged from 5° to 12°C. On 29 April 1971, one of the fe- males died and some of her eggs were removed and placed in beakers of filtered seawater. Tem- perature in the beakers was 15°C; prezoeae were observed the next day and on 1 May first zoeae were apparent. On 10 May, another female's eggs started hatching in one of the tanks. Water temperature in this tank was 10°C. A prezoea stage was noted in the tank also but lasted less than 1 hr. First zoeae from each of these batches were maintained separately in beakers contain- ing 1,000 ml of filtered seawater and 50,000 units of penicillin plus a small amount of streptomycin (some larvae were raised without the antibiotics with no differences noted). Newly hatched Artemia nauplii obtained from California eggs were given as food. Small amounts of algae (Dimaliella) were also added to sustain the Artemia nauplii. Water and food were changed every other day. At the start the zoeae were maintained at a constant temperature of 15°C, but fouling organisms grew on many of the zoeae and it was necessary to maintain them at room temperature (18°-21°C) to accelerate develop- ment. Salinity ranged 30 to SVu during the study. Zoeae were also put into compartmented plastic trays, one to a compartment, and were maintained as those in the beakers. The zoeae in the compartments were used for the devel- opmental studies. When a zoea molted in a 69 FISHERY BULLETIN: VOL. 71, NO. 1 compartment it was preserved, along with its molt, the following day. Measurements of the larvae were made with an ocular micrometer and are given in milli- meters; drawings were made with the aid of a camera lucida. Ten individuals, of each stage, and their molts were examined and measured for the developmental series. Carapace length mea- surements were made from the anterior edge of the carapace (posterior edge of eye sockets) to the posterior margin, along the midline. Total length represents the distance from the tip of the rostral spine to the tip of the telson along the curves of the dorsal midline. The setae on some appendages have been shortened or deleted in some of the illustrations to ensure clarity, but are given full descriptions in the text. RESULTS A prezoea stage of short duration, four zoeal stages, and a megalopa stage were obtained. The prezoeae measured about 2.8 mm in total length. The following are the average number of days from one stage to the next at temperatures of 18° to 21°C; first stage zoea to second, 7 days; second to third, 6; third to fourth, 5; fourth to megalopa, 7; and megalopa to first crab, 14. Two abnormalities were observed; a first stage zoea with the protopodite of one antenna forked from the base, and a second stage zoea with the large lateral spine on one side of the telson forked from its base. The larvae are usually light red- dish brown in gross appearance. Some indi- viduals are quite light in appearance but both phases exhibit numerous melanophores scattered over the entire body, particularly on the cephalo- thorax and along the outer margins of the ab- domen. Small melanophores are often scattered on the basipodite on the maxillipeds, DESCRIPTION OF THE LARVAE ZOEA I (Figure lA) Carapace length 0.83 mm (0.81-0.86); total length 3.71 mm (3.67-3.73). Carapace with rostral, lateral, and dorsal spines. Rostral spine strongly depressed; slightly longer than anten- nae and nearly three quarters the length of the dorsal spine. Lateral spines flexed slightly ven- trad, broad based, and nearly as long as the rostral spine. Width of the carapace from tip to tip of lateral spines about half the total length of the body. Dorsal spine long and curved slightly posteriad; its length about one-third the total length of the body. A short slender seta is pre- sent on each side of the carapace, in line with the posterior margin of the dorsal spine's base. The eyes are not stalked. Abdomen with five somites and the telson (Figure 2H). Abdominal somites two through five with lateral spines (fifth somite in some individuals without spines) ; those on second somite fairly blunt and directed somewhat anteriad; spines on somites three through five sharper and hooked posteriad, decreasing in size posterially. Posterior lateral margin of somites three through five produced into long, sharp spines. Second somite with two setae on middorsal surface; somites three through five with two setae each on posterodorsal margin. Telson bifurcate with three pairs of setae on the inner side. Each furca with two lateral spines, one long and strong, the other much smaller, and a small dorsal spine. Anten- nule (Figure 2B) with four unequal aesthetes and a small seta terminally. The antenna (Fig- ure 2A) bears a long protopodite with a row of spinules on the outer margins; the exopodite is about half the length of the protopodite and terminates in a spine and one seta. Mandible (Figure 2C) with two large teeth anteriorly, a medial blade, and a toothed edge posteriorly. Themaxillule (Figure 2D) bears a two-segment- ed endopodite, the short proximal segment bears one long plumose seta, the distal segment with six long plumose setae, two of which are subter- minal; the basal and coxal endites each bear six spinous setae. The scaphognathite of the max- illa (Figure 2E) has seven marginal plumose setae and a plumose apical tip; the endopodite is bilobed with five spinous setae on the distal lobe and three on the proximal; basal and coxal endites each bilobed with 5 + 5 and 3 + 3 spinous setae respectively. There is a suggestion of two segments in the exopodite of the first maxilliped (Figure 2F) which bears four, one- jointed, natatory setae terminally; endopodite 70 PERKINS : LARVAL STAGES OF DEEP SEA RED CRAB 1.0 mm Figure 1. — Geryon quinquedens. A. Zoea I, B. Zoea II, C. Zoea III, D. Zoea IV. 71 FISHERY BULLETIN: VOL. 71, NO. 1 0. 1 mm 0. 1 mm Figure 2. — Geryon qidnquedens, Zoea I. A. antenna, B. antennule, C. mandible, D. maxillule, E. maxilla, F. first maxilliped, G. second maxilliped, H. dorsal view of abdomen and telson. 72 PERKINS: LARVAL STAGES OF DEEP SEA RED CRAB five-segrmented with 2, 2, 1, 2, 5 setae on the seg- ments (proximally to distally) ; basipodite with 10 setae, coxa with 1 seta. The exopodite of the second maxilliped (Figure 2G) bears four, one- jointed natatory setae terminally; endopodite three-segmented with 1, 1, 5 setae; basipodite with four setae; coxa naked. ZOEA II (Figure IB) Carapace length 1.07 mm (1.00-1.13), total length 4.97 mm (4.89-5.02) . Spines on carapace as in Zoea I. The rostral spine is nearly as long as the dorsal spine. Width of carapace from tip to tip of lateral spines about four-tenths the total length of the body. A short, slender seta is present at each side of the posterior margin of the dorsal spine's base. One to four small setae are scattered along the anterior edge of the proximal half of the dorsal spine; a small seta is present slightly above and in line with i the base of each eye. Eyes stalked. Postero- ventral edge of carapace finely serrate with three of four short, slender setae on the inner margin. . Abdomen with five free somites (sixth fused with ! telson) (Figure 3H). Spination and setation as in Zoea I except an additional pair of short setae has been added proximally on the inner margin of the telson. The spines on the abdomen are somewhat more pronounced than in Zoea I. Theantennule (Figure 3A) bears four aesthetes, plus one short and one long setae terminally. Protopodite of the antenna (Figure 3B) as in Zoea I; the exopodite terminates in two unequal setae and the endopodite is evident as a bud. Mandible as in Figure 3C. Endopodite of the maxillule (Figure 3D) as in Zoea I; a single, long plumose seta is present on the protopodite; basal endite with 12 spinous setae, coxal endite with 11. The scaphognathite of the maxilla (Figure 3E) bears 22 marginal plumose setae; endopodite bilobed with five setae on the distal lobe, three on the proximal ; basal endite with seven setae on the distal lobe, five on proximal; coxal endite with four setae on each lobe. The two-segmented exopodite of the first maxilliped (Figure 3F) now bears 10 or 11 articulated natatory setae terminally; endopodite, basipo- dite, and coxa as in Zoea I. The exopodite of the second maxilliped (Figure 3G) bears 11 ar- ticulated natatory setae; endopodite, basipodite, and coxa as in Zoea I. The third maxillipeds, chelipeds, and pereiopods are evident as minute buds. ZOEA III (Figure IC) Carapace length 1.42 mm (1.35-1.48) ; total length 6.13 mm (5.75-6.48). The lateral spines are more ventrally flexed and the number of small setae on the anterior edge of the dorsal spine has increased from the previous stage. There are 15 small setae along the inner margin of each side of the posteroventral and posterior edge of the carapace. Abdomen with six somites and the telson (Figure 41). The spination and setation of somites two through five is the same as in the previous stage. The first somite now bears two small setae on the middorsal surface; sixth segment naked. Telson with five pairs of setae on the inner portion. The pleopods are evident as buds on somites 2 through 5; uropods as buds on the sixth. Antennule (Figure 4A) with four aesthetes and three setae terminally, plus three aesthetes subterminally; basal portion swollen and the endopod occurs as a small bud. The protopodite and exopodite of the antenna (Figure 4B) as in the previous stage; the en- dopodite is now about the same length as the exopodite. The mandible (Figure 4C) now bears a simple palp, evident as a bud. The endopodite of the maxillule (Figure 4D) has five or six long plumose setae on the distal segments; the basal and coxal endites each bear about 17 spinous setae. The scaphognathite of the maxilla (Fig- ure 4E) bears about 31 marginal plumose setae; endopodite and basal endite as in previous stage; coxal endite with five spinous setae on distal lobe, and nine on the proximal lobe. The exopodite of the first maxilliped (Figure 4F) bears about 14 articulated natatory setae terminally; the distal segment of the endopodite now bears six setae; setation of other segments as in the pre- vious stages, as is that of the basipodite; coxa with three setae. The exopodite of the second maxilliped (Figure 40) bears about 14 articu- lated natatory setae; setation of endopodital seg- ments as in the previous stages; basipodite with 73 FISHERY BULLETIN: VOL. 71, NO. 1 0.1 mm Figure 3. — Geryon quinquedens, Zoea II. A. antennule, B. antenna, C. mandible, D. maxillule, E. maxilla, F. first maxilliped, G. seconu maxilliped, H. dorsal view of abdomen and telson. 74 PERKINS: LARVAL STAGES OF DEEP SEA RED CRAB O.Imm 0.1 mm B 0.1 mm Figure 4. — Geryon quinquedens, Zoea III. A. antennule, B. antenna, C. mandible, D. maxillule, E. maxilla, F. first maxilliped, G. second maxilliped, H. third maxilliped, I. dorsal view of abdomen and telson. 75 FISHERY BULLETIN: VOL. 71, NO. 1 three or four setae; coxa with one seta. The rudimentary third maxilliped as in Figure 4H. Chelipeds and pereiopods about twice as large as in previous stage. ZOEA IV (Figure ID) Carapace length 1.94 mm (1.89-1.97); total length 8.31 mm (7.62-8.95). There is an indi- cation that the considerable range in total and dorsal spine length is due to the nearness of a particular animal to its next molt. In some in- dividuals a relatively short, blunt dorsal spine was observed; these individuals showed the greatest total length, while on shorter individu- als a relatively long, slender dorsal spine was observed. Dorsal spine from 20 to 30% of the total length, with small setae scattered along its entire anterior edge; rostral spine usually as long or longer than the dorsal spine. Lateral spines flexed ventrally; carapace width, from tip to tip of lateral spines, about one-third of the total length of the body. A few additional setae occur between the anterior edge of the dorsal spine and the base of the rostral spine; 20 to 25 slender setae mid-ventrally to posteriorly on the inner margin of the carapace. Abdomen with six somites and the telson (Figure 51). The blunt lateral spines on the second somite now directed slightly posteriad; spines on other so- mites more pronounced than in the previous stage. The first somite bears four setae on the middorsal surface; the second somite with two setae anterior and four setae posterior to the midline; the third somite with two setae mid- dorsally and two on the posterodorsal margin; somites 4 and 5 each with a pair of setae on the posterodorsal margin, sixth somite naked; tel- son as in previous stage but with the proximal setae on the inner portion larger; pleopods and uropods considerably enlarged from the previous stage. Antennule (Figure 5A) with four aes- thetes and two setae terminally, one group of six aesthetes and another of two subterminally ; bud of endopod enlarged from previous stage; basal portion with five setae. Antenna (Figure 5B) as in previous stage but with the endopodite con- siderably enlarged and longer than the protop- odite. Mandibular palp (Figure 5C) simple and much enlarged from the previous stage. The endopodite of the maxillule (Figure 5D) the same as in the previous stages; basal endite with about 22 spinous setae, coxal endite with 17. Scaphognathite of the maxilla (Figure 5E) with about 54 marginal plumose setae; endopodite the same as in the previous stages; distal lobe of the basal endite with 12 spinous setae, proximal lobe with nine; coxal endite as in the previous stage. The exopodite of the first maxilliped (Figure 5F) bears 17 setae; endopodite and basipodite as in the previous stage; coxa with six setae. The exopodite of the second maxilliped (Figure 5G) with 19 setae; the terminal segment of the en- dopodite with six setae, other segments as in the previous stages; basipodite with three setae; coxa with one. Exopodite of the third max- illiped (Figure 5H) with slight articulation; endopodite faintly five-segmented, the two distal segments each with one spine. Chelipeds and pereiopods considerably enlarged from previous stage. MEGALOPA (Figures 6A and 7A) Carapace length 3.16 mm (3.02-3.26); total length 6.46 mm (6.32-6.60). Rostrum one-fifth the length of the carapace, strongly depressed and bifid at the tip; a medial groove present from interorbital position nearly to the distal end of the rostrum. Eye stalks with a few small setae on the anterior and dorsal surfaces. A carina is present on each side of the mesogastric mid- line (highest points on carapace) and another prominence is present in the cardiac region of the carapace; hepatic and branchial lobes round- ed; setation sparse, occurring along the margins of the rostrum, a few in the postorbital region, and a few tufts on the mesogastric prominences; numerous setae are present along the mid-ventral to posterior margin of the carapace. Abdomen with six somites and the telson; setation sparse and as figured; pleopods with about 28 long natatory setae each (Figure 7D); uropods with about 15 each (Figure 7H). The peduncle of the antennule (Figure 6B) is three-segmented, the proximal segment with one plumose seta, the middle segment with five setae subterminally, and the distal segment has two setae; inner 76 PERKINS: LARVAL STAGES OF DEEP SEA RED CRAB 0.4 mm Figure 5. — Geryon quinquedens, Zoea IV. A. antennule, B. antenna, C. mandible, D. maxillule, E. maxilla, F. first maxilliped, G. second maxilliped, H. third maxilliped, I. ventral view of abdomen and telson. 77 FISHERY BULLETIN: VOL. 71, NO. 1 0.2 mm FiGUKE 6. — Geryon quinquedens, megalopa. A. dorsal view, B. antennule, C. antenna, D. mandible, E. maxillule, F. maxilla, G. first maxilliped. 78 PERKINS: LARVAL STAGES OF DEEP SEA RED CRAB 0.5mm 0.4 mm 0.5 mm H 0.4mm 0. 1 mm Figure 7. — Geryon quinq^iedens, megalopa. A. lateral view, B. second maxilliped, C. third maxil- liped, D. pleopod of second abdominal somite, E. cheliped, F. last pereiopod, G. dactyl of last pereiopod, H. ventral view of telson and uropods. 79 FISHERY BULLETIN: VOL. 71, NO. 1 flagellum two-segmented with four setae ter- minally and two subterminally; outer flagellum with four segments; proximal segment naked, the antepenultimate segments with six aesthetes, penultimate with five aesthetes and one seta, dis- tal segment with five aesthetes near its base and one seta subterminally, terminating in a long, plumose seta. The basal portion of the antenna (Figure 6C) is two-segmented with small setae scattered on the distal segment; peduncle two- segmented with three setae on each segment; the flagellum is eight-segmented, the setation as figured. Mandibular palp two-segmented with 16 setae on the distal segment (Figure 6D) . The endopodite of the maxillule (Figure 6E) is un- segmented, has one lateral and two subterminal setae, and terminates in a spine; the basal en- dite bears about 35 spinous setae; the coxal en- dite with about 25. The scaphognathite of the maxilla (Figure 6F) with about 100 marginal plumose setae, with a few setae scattered on the dorsal and ventral surfaces; endopodite pro- duced into a narrow lobe, with three setae on the distal margin of the base, eight setae lat- erally on the same margin, and one long setae on the proximate lateral edge; basal endite with about 14 spinous setae on the distal lobe, 11 on the proximal; coxal endite with eight spinous setae on the distal lobe and 16 on the proximal. The exopodite of the first maxilliped (Figure 6G) now terminates in but five plumose setae and one naked seta; the endopodite is unsegment- ed and bladelike; basal endite with about 37 spi- nous setae along the margin; coxal endite with 12; epipodite with about 16 nonplumose hairs and 1 seta; setation of other portions as figured. A well-developed epipodite is present on the second maxilliped (Figure 7B) and bears about 14 nonplumose hairs and 2 setae; the exopodite terminates in four plumose and one nonplumose setae, and there are four short setae on the outer lateral margin; the endopodite with four segments, the distal segment with about 13 spinous setae terminally; other setation as figured. The exopodite of the third maxilliped (Figure 7C) terminates in six plumose setae; the endopodite is five-segmented, its spination and setation variable as is that of the epipodite and is approximately as figured. Chelipeds (Fig- ure 7E) with a strong hooked spine on ventral portion of ischum. Spines on coxa of pereiopods one through three and another blunt spine sub- terminally on the posterior margin of the same articulation, decreasing in size posteriorly. Dactyl of last pereiopod with two curved, toothed setae (Figure 7F and G), DISCUSSION The prezoea and first zoea stages of Geryon qiiinquedens appear to be quite similar in struc- ture to the corresponding stages of G. tridens described by Brattegard and Sankarankutty (1967) . The most trenchant diff"erences are the larger size of the zoea of G. quinqiiedens and the lack of posterolateral spines on the fifth abdom- inal somite in G. tridens. Brattegard and Sank- arankutty give the length of the first zoea as 2.0 mm but no mention is made as to how they arrived at this measurement. The large size of the larvae of G. quinqiiedens should help to distinguish them from other sympatric Brachy- rhyncha, with the possible exception of its congener, G. affinis. The family Geryonidae was erected in 1930 by Beurlen (original reference not obtained, in- formation from Christiansen, 1969), but since then Geryon has been placed in various families: Rathbun (1937) places Geryon quinqiiedens in the subfamily Carcinoplacinae of the family Goneplacidae; Bouvier (1940) placed Ger?/on in the family Xanthidae; Gurney (1939) lists the genus under the subfamily Menippinae of the family Xanthidae; and more recently Christian- sen (1969) reassigned the genus to the family Geryonidae, I have found few references dealing with the larvae of goneplacid genera. Lebour (1928) dis- cussed the larval stages of Gonoplax rhomboides; Kurata (1968) described the larvae of Carcino- plax longimanus. Both of these species agree generally with Geryon in the number of larval stages, the spination of the carapace and abdo- men, and the spination and setation of the telson. The relative length of the exopodite to the pro- topodite of the antenna in Geryon is apparently similar to that in Carcinoplax but diff'erent from that in Gonoplax. Megalopa of both Gonoplax 80 PERKINS: LARVAL STAGES OF DEEP SEA RED CRAB and Carcinoplax bear spines on the carapace, while megalopa of Geryon do not. The exopodite of the antenna in zoeae of Ger- yon quinquedens is about one-half the length of the protopodite; in Gouoplax (Lebour, 1928) the exopodite is about the same length as the protopodite, and bears two short setae medially, rather than terminally as in Geryon. On the basis of the relative length of these two struc- tures, Geryon would not align with Gorioplax in Lebour's key to the zoeae. The configuration and relative length of the antennal exopodite and protopodite in Geryon are more like those of Cancer (Poole, 1966), some portunids (Le- bour, 1928; Roberts, 1969) and certain grapsids (Diaz and Ewald, 1968). Boyden (1943) and Leone (1951) discuss the serological relation- ships of Geryon quinquedens to other members of the Brachyura. Each found Geryon to be closer to the Xanthidae than to other families tested, with certain affinities noted to the Can- cridae and Portunidae. However, neither of these workers tested other members of the Go- neplacidae against Geryon. The zoeae of Geryon quinquedens are similar to most xanthid zoeae (Lebour, 1928; Costlow and Bookhout, 1968) in the number of zoeal stages, the spines on the carapace, and the armature of the telson. How- ever, there are differences in the latter two characters within the Xanthidae alone (Costlow and Bookhout, 1966). The structure of the zoeal antennae of Geryon is decidedly different from the antennae of xanthid zoeae; in xanthid zoeae the exopodite of the antenna is very short in relation to the length of protopodite. The number of terminal setae on the exopodite of the first and second maxillipeds of Zoea H through IV apparently distinguish the larvae of Geryon quinquedens from other members of the Brachyrhyncha. In this group the exopodite of the first maxillipeds consistently bear four terminal setae in the first zoeal stage and six in the second stage. The same number is usu- ally associated with the second maxilliped. G. quinquedens bears 4 setae in the first stage and 10 or 11 in the second. Knight (1968) reports that the raninid species, Raninoides benedicti Rathbun, has nine setae on the exopodite of the second maxilliped (six on the first) of Zoea II. Only members of the rather diverse and remote Anomura apparently bear as many terminal se- tae on the maxillipeds of stages subsequent to Zoea I as does G. quittquedens. The lithodid spe- cies, Cryptolithodes ty pious Brandt, bears four setae on the exopodite of the first maxilliped in the first zoeal stage and eight in the second stage (Hart, 1965) . The porcellanid genera Poly onyx (Knight, 1966; Gore, 1968), Pachycheles, and Petrolisthes (Greenwood, 1965) bear from 11 to 14 terminal setae on the exopodite of the first maxillipeds in Zoea II, All bear four setae on this structure in Zoea I. ACKNOWLEDGMENTS I wish to thank James A. Rollins for making the drawings of the larval stages (Figures 1, 6A, and 7A), Warren Rathjen for supplying me with the female crabs, Mary Elizabeth Joralemon and Margaret S. Kelly for assistance in rearing the crab larvae, and Gareth W. Coffin for his reproductions of the illustrations. LITERATURE CITED BouviER, E.-L. 1940. Decapodes Marcheurs. Faune Fr. 37, 389 p. Boyden, A. 1948. Serology and animal systematics. Am. Nat. 77:234-255. Brattegard, T., AND C. Sankarankutty. 1967. On prezoea and zoea of Geryon tridens Kroyer (Crustacea Decapoda). Sarsia 26:7-12. Christiansen, M. E. 1969. Marine invertebrates of Scandinavia. No. 2 Crustacea Decapoda Brachyura. Universitets- forlaget, Oslo, 143 p. Costlow, J. D., Jr., and C. G. Bookhout. 1966. Larval development of the crab, Hexapono- peus angustifrons. Chesapeake Sci. 7:148-156. 1968. Larval development of the crab, Leptodins agassizii A. Milne Edwards, in the laboratory (Brachyura, Xanthidae). Crustaceana, Suppl. 2:203-213. Diaz, H., and J. J. Ewald. 1968. A comparison of the larval development of Metasesarma rubripes (Rathbun) and Sesarma ricordi H. Milne Edwards (Brachyura, Grapsidae) reared under similar laboratory conditions. Crus- taceana, Suppl. 2:225-248. 81 FISHERY BULLETIN: VOL. 7!, NO. 1 Gore, R. H. 1968. The larval development of the commensal crab, Poly onyx gibbesi Haig, 1956 (Crustacea: Decapoda). Biol. Bull. (Woods Hole) 135:111-129. Greenwood, J. G. 1965. The larval development of Petrolisthes elon- gatus (H. Milne Edwards) and Petrolisthes no- vaezelandiae Filhol (Anomura, Porcellanidae) with notes on breeding. Crustaceana 8:285-307. GURNEY, R. 1939. Bibliography of the larvae of decapod Crus- tacea. Ray Soc, Lond., 123 p. Hart, J. F. L. 1965. Life history and larval development of Cryptolithodes typicus Brandt (Decapoda, Ano- mura) from British Columbia. Crustaceana 8: 255-276. HOLMSEN, A. 1968. The commercial potential of the deep sea red crab. Univ. R.I., Dep. Food Resour. Econ., Occas. Pap. 68-138:1-17. Knight, M. D. 1966. The larval development of Polyonyx quadri- ungulatus Glassell and Pachycheles rudis Stimp- son (Decapoda, Porcellanidae) cultured in the laboratory. Crustaceana 10:75-97. 1968. The larval development of Raninoides ben- edicti Rathbun (Brachyura, Raninidae), with notes on the Pacific records of Raninoides laevis (Latreille). Crustaceana, Suppl. 2:145-169. KURATA, H. 1968. Larvae of Decapoda Brachyura of Arasaki, Sagami Bay — IH. Carcinoplax longimanus (De Haan) (Goneplacidae). [In Japanese, English abstr.] Bull. Tokai Reg. Fish. Res. Lab. 56:167- 171. Lebour, M. V. 1928. The larval stages of the Plymouth Brachyura. Proc. Zool. Soc. Lond. 1928:473-560. Leone, C. A. 1951. A serological analysis of the systematic re- lationship of the brachyuran crab, Geryon quin- quedens. Biol. Bull. (Woods Hole) 100:44-48. McRae, E. D., Jr. 1961. Red crab explorations off the Northeastern coast of the United States. Commer. Fish. Rev. 23(5) :5-10. Poole, R. L. 1966. A description of laboratory-reared zoeae of Cancer magister Dana, and megalopae taken under natural conditions (Decapoda, Brachyura). Crustaceana 11:83-97. Rathbun, M. J. 1937. The oxystomatous and allied crabs of Amer- ica. U.S. Natl. Mus. Bull. 166:1-278. Roberts, M. H., Jr. 1969. Larval development of Bathynectes superba (Costa) reared in the laboratory. Biol. Bull. (Woods Hole) 137:338-351. Schroeder, W. C. 1959. The lobster, Homarns americanus, and the red crab, Geryon quinquedens, in the offshore waters of the Western Atlantic. Deep-Sea Res. 5:266-282. 82 THE SYSTEMATIC STATUS OF MERLUCCIUS IN THE TROPICAL WESTERN ATLANTIC OCEAN INCLUDING THE GULF OF MEXICO Charles Karnella' ABSTRACT Several morphometric and meristic characters are used to compare populations of Merluccius from the Gulf of Mexico and Atlantic Ocean. Both populations are shown to have similar values for all characters studied. As a result M. magnoculus Ginsburg is relegated to the synonymy of M. albidus (Mitchill). Geog^raphical variation is noted in many of the characters investigated. The widely distributed gadoid fish genus Mer- luccius contains an indeterminate number of commercially fished species. There are 11 nom- inal species (Grinols and Tillman, 1970), known variously in the United States as either whiting or hake. The object of this paper is to deter- mine the number of species living in the tropical western Atlantic (including the Gulf of Mexico and Caribbean). Ginsburg (1954) recognized three species from the western North Atlantic. One of these, M. hilinearis (Mitchill), is distinct from the other two nominal forms in having more gill rakers on the first arch (15-22 vs. 9-12). This species will not be considered further as it does not occur south of Cape Fear, N.C. M. magnoculus Ginsburg was described as new mainly on the basis of its having a longer head and shorter paired fins than its closest rel- ative, M. albidus (Mitchill) . M. albidus is found in the tropical western Atlantic, although not exclusively so, as it is known to occur sympatric- ally with M. bilinearis in the north. Ginsburg further noted that M. magnoculus and M. albidus were also moderately to slightly divergent in the following characters: maxillary length, snout ' Formerly National Systematica Laboratory, National Marine Fisheries Service, NOAA ; present address : Di- vision of Fishes, U.S. National Museum, Washington, DC 20560, and Department of Biological Sciences, George Washington University, Washington, DC 20006. M»nu5cript accepted July 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. length, eye diameter, and number of first dorsal, second dorsal, pectoral, and anal fin rays. More- over, M. magnoculus was confined to the Gulf of Mexico while M. albidus occurred off the east- ern coast of North America from Georges Bank to the Tortugas off the west coast of Florida. The lack of comparative material of equivalent size from the Gulf, the doubtful systematic status of two specimens from Savannah, Ga., and of a single specimen from off Cape Canaveral, Fla., make uncertain Ginsburg's tentative assignation of these specimens to M. albidus. Difficulty in identifying subsequent material from the Gulf of Mexico and Caribbean has necessitated a re- assessment of the taxonomic status of M. albidus and M. magnoculus, especially since the stated differences between the two are slight and there is at least some overlap in all characters used to separate them. Throughout the body of this paper the At- lantic population is taken to include specimens from the Caribbean also. MATERIAL A total of 253 specimens was examined; 86 from the Gulf of Mexico and 167 from along the eastern coast of the Americas, lat 41*'30'N south to lat 7°26'N (Figure 1). This total included Ginsburg's material whenever possible. How- 83 FISHERY BULLETIN: VOL. 71, NO. 1 ever, some of his specimens were in poor state of preservation and too fragile to be handled. The list of specimens is as follows: ATLANTIC OCEAN AND CARIBBEAN SEA U.S. National Museum (USNM): 25769-1 specimen; 26049-1; 26073-1; 31630-1; 31677-1 31686-1; 31739-1; 31741-1; 31822-1; 31842-2 31844-1; 31863-2; 32791-1; 33032-1; 44264-1 45920-1; 155475-2; 159214-1; 159230-1 186294-5; 186299-1; 186302-1; 190356-1 205223-1; 205224-1; 205230-1; 205231-1 205233-1; 205235-1; 205237-4; 205240-2 205241-1; 205242-1; 205243-1; 205244-1 205245-1; 205246-1; 250247-1; 205248-1 205249-1; 205251-1; 205252-1; 205253-1 205255-1; 205257-2; 205259-1; 205260-1 205262-1; 205263-2; 206190-1; 206191-1 206192-2; 206194-1; 206195-7; 206196-1 206197-1; 206198-1; 206199-1; 206200-1 206201-2; 206202-5; 206203-4; 206204-1 206205-30; 206207-2; 206208-5; 207187-3 207188-1; 207189-1. University of Miami Marine Laboratory (UMML) : 3418-2 specimens; 3696-2; 4431-2; 22957-2; 29224-5; 29506-4; 29508-1. Museum of Comparative Zoology (MCZ): 37754-2 specimens; 38086-3; 38130-3; 38324-1; 38333-1; 38338-2; 38350-1; 38395-1; 38399-2. GULF OF MEXICO Figure 1. — Distribution of samples, western North At- lantic Ocean; equator to lat 45°N. METHODS All counts and measurements were made as described in Ginsburg (1954), so that the data from both studies would be directly comparable. The median fin rays, except the caudal, pectoral rays on both sides, and gill rakers of the outer arch on both sides were counted on all speci- mens that were not damaged. Total vertebral counts were made on selected specimens. Stan- dard length, head length, snout length, maxillary length, eye diameter, pectoral fin length, and pelvic fin length were measured on all specimens when possible. U.S. National Museum (USNM): 92045-2 specimens; 157757-1; 157758-2; 157759-5 ; 157761-4; 157762-2; 157763-10 187134-5; 187136-1; 205225-1 205227-2; 205228-1 ; 205229-3 205234-2; 205236-1; 205238-1 205250-1; 205254-3; 205256-1 205261-1; 206187-1; 206188-2 206193-4; 206206-3; 207153-2 157760-5 186331-2; 205226-1; 205232-3 ; 205239-1; 205258-1; 206189-2; 207190-1. University of Miami Marine (UMML) : 29507-9 specimens. Laboratory RESULTS AND DISCUSSION Inspection of the counts and measurements indicates that the Gulf and Atlantic populations are similar in all characters studied. Within each area there are local differences in most of the characters; however, these differences are minor. The Gulf and Atlantic populations have identical or nearly identical ranges for all char- acters investigated, and the average values for both are generally only slightly divergent. Differences in the relative head length and the 84 KARNELLA: SYSTEMATIC STATUS OF MERLUCCIUS relative length of the paired fins were the main criteria used by Ginsburg (1954) for recognizing the Gulf population as a distinct species, M. magnoculus. These differences, however, were minor. More importantly the material used in his study did not adequately represent either the Atlantic or Gulf population. Thirty of thirty- eight specimens from the Atlantic were taken off Long Island, N.Y., and all 32 of the speci- mens from the Gulf of Mexico came from north of lat 26°N. Ginsburg was not able to make a valid comparison of the Atlantic and Gulf pop- ulations with the limited material available to him. HEAD LENGTH Ginsburg (1954) listed the range of head length taken as a percent of standard length as 27.3-81.3 for M. albidus and 29.6-31.3 for M. magnoculus but gave no mean values. Average values calculated from the data in Table 8 in Ginsburg (1954) are 28.7 for M. albidus and 30.6 for M. magnoculus. The spec- imens from the Atlantic and Caribbean popu- lations examined in this study, had a range of 26.4-32.9 and a mean of 29.0, while the Gulf population had a range of 27.3-31.3 with a mean of 29.7. As can be seen from Table 1 the head length expressed as a percent of standard length is fairly uniform over the entire geographic area represented in this study. Table 1. — Head length as a percent of standard length for the Atlantic and Gulf populations. Population N Range Mean Atlantic 7°-20°N 48 27.9-31.8 29.8 2r-4rN 114 26.4-32.9 28.7 7''-4rN 162 26.4-32.9 29.0 Gulf 19''-25°N 42 27.3-31.2 29.4 26°-29''N 43 28.6-31.3 30.0 19°-29°N 85 27.3-31.3 29.7 Although the Gulf population does have a slightly larger head, degree of difference between the two populations reported by Ginsburg is un- supported by the present data. The two popu- lations are not separable on the basis of relative length. Ginsburg also stated that growth of the head was allometric. The present data indicate that growth of the head is isometric (see Figure 2). 200 r 150 z iOO- 55 = 8 •!• .%•- A'^^ •••• D<9>V •• 250 300 350 400 450 STANDARD LENGTH MM 200 500 600 Figure 2. — Gulf (squares) and Atlantic (circles) populations: relation of head length to standard length. 85 FISHERY BULLETIN: VOL. 71, NO. 1 PAIRED FIN LENGTH Ginsburg (1954) reported the range of pec- toral fin length taken as a percent of standard length to be 18.0-21.5 and 15.5-19.0 for M. alhidus and M. magnoculus, respectively. The Atlantic specimens examined in the study have a similar maximum value to that of M. alhidus (21.7, see Table 2) but the minimum value obtained, 13.7, is much lower. The minimum value obtained from the Gulf population, 13.7, is somewhat lower than the minimum value recorded for M. magnoculus, while the maximum value obtained, 19.4, is similar to that given by Ginsburg for M. magnoculus. Average values calculated from the data in Table 10 in Ginsburg (1954) are 19.8 for M. alhidus and 17.0 for M. magnoculus. These compare fairly well with the values ob- tained for the Atlantic and Gulf populations 18.3 and 16.8, respectively. Table 2. — Pectoral fin length as a percent of standard length for the Atlantic and Gulf populations. Table 3. — Pelvic fin length as a percent of standard length for the Gulf and Atlantic populations. Population N Ranga Mean Atlantic 7''-20°N 48 13.7-19.5 17.3 2r-4rN 113 15.8-21.7 18.8 7°-4rN 161 13.7-21.7 18.3 Gulf 19''-25''N 41 13.7-19.2 17.2 26''-29°N 42 13.8-19.4 16.4 19<'-29°N 83 13.7-19.4 16.8 The range of values for the pelvic fin length expressed as a percent of standard length is 12.8- 19.2 with an average value of 15.6 for the At- lantic population and 11.6-17.0 with a mean of 14.3 for the Gulf population (Table 3). Ginsburg also reported a range of 13.5-19.5 for M. alhidus and 12.0-16.0 for M. magnoculus, and the aver- ages computed from data contained in his Table 9 are 16.6 and 14.0 for M. alhidus and M. mugnoc- ulus, respectively. The present data indicate that the Gulf pop- ulation does have proportionally smaller paired fins than the Atlantic population, however, the differences are much smaller than indicated by Ginsburg. The relative length of the paired fins is similar in both populations and is clearly of no value in separating the two. Population N Range Mean Atlantic 7''-20°N 48 12.8-19.2 14.7 21M1''N 114 12.8-17.7 15.9 7''-4rN 162 12.8-19.2 15.6 Gulf 19°-25''N 41 12.7-17.0 15.1 26°-29°N 43. 11.6-16.2 13.6 19°-29°N 84 11.6-17.0 14J Ginsburg (1954) stated that the growth of the pelvic fin was allometric and that the relative pectoral fin length changed little if any with growth. To compensate for this he arranged his material into several size classes and com- pared similar sizes for both populations. How- ever, he gave no average standard length for the classes, and it is impossible to determine if the size composition of the classes he compared was similar. Figure 3 indicates that growth of the pectoral fin is allometric and not isometric as reported by Ginsburg (1954). The pelvic fin does undergo allometric growth as stated by Ginsburg (see Figure 4). Since the material examined from both areas is not of the same size composition (the average standard length of the specimens from the At- lantic population is 283 mm while the average standard length of the specimens from the Gulf population is 323 mm) at least some of the dif- ference in paired fin length is due to allometric growth. Figures 3 and 4 indicate that for some of the Gulf material the paired fins are relatively smal- ler than in other specimens of similar sizes. The majority of specimens with the smaller fins were collected north of lat 26°N. Most of the speci- mens with the higher values were collected north of lat 21 °N in the Atlantic. Many specimens ex- amined from the northern Gulf have fins of the same size as specimens from the southern Gulf and Atlantic populations. Hence, not all of the northern Gulf material can be distinguished by relative fin size. The paired fins are poor characters to use in Merhiccius because they are generally damaged to some degree. It is often impossible to deter- mine if the fine ends of the rays are broken off. 86 KARNELLA: SYSTEMATIC STATUS OF MERLUCCWS I20r Z z X I- V) : 90- < X O60 »- o a. 30 oB fl G OO "■ea. 180 200 250 300 350 400 450 STANDARD LENGTH MM 500 600 Figure 3. — Gulf (squares) and Atlantic (circles) populations: relation of pectoral fin length to standard length. 100 80- 60- '40- LlJ 20 o Q o □ o • • 'Ji. ^^ " • • • o 1 1 t 1 180 200 250 300 350 400 450 STANDARD LENGTH MM 500 550 Figure 4. — Gulf (squares) and Atlantic (circles) populations: relation of pelvic fin length to a standard length. Although the proportion and degree of damaged fins should be the same for both populations, a slight error will be introduced, and values pre- sented for these measurements should be con- sidered only as approximations of the real values. EYE DIAMETER, SNOUT LENGTH, AND MAXILLARY LENGTH The values obtained for these characters were similar in the Atlantic and Gulf populations, with the Gulf population having a slightly larger average value for all three characters (Tables 4, 5, 6) ; these values agree well with those of Ginsburg (1954). All differences in these characters reported by Ginsburg (1954) may be explained by his limited material. Material from other areas ex- amined in the present investigation indicate there are no differences between the two popu- lations in any of the above characters (Figures 5, 6, 7). Table 4. — Eye diameter as a percent of standard length for the Gulf and Atlantic populations. Population A^ Range Mean Atlantic 7''-20''N 48 4.6-8.4 5.6 21''-41''N 114 4.8-8.4 5.9 7°-41°N 162 4.4-8.4 5.9 Gulf 19">-25°N 42 5.2-7.0 6J0 26''-29°N 43 4.8-7.1 6.1 19°-29°N 85 4.8-7.1 6.0 87 FISHERY BULLETIN: VOL. 71, NO. 1 Table 5. — Snout length as a percent of standard length for the Gulf and Atlantic populations. Population N Ranga Mean Atlantic 7°-20°N 44 8.8-10.7 9.7 2r-*rN 114 8.1-11.1 9.2 7°-4rN 158 8.1-11.1 9.4 Gulf 19°-25°N 42 8.7-11.2 9.8 26''-29°N 43 9.2-10.8 10.2 19''-29°N 85 8.7-11.2 10.0 Table 6. — Maxillary length as a percent of standard length for the Gulf and Atlantic populations. Population N Range Mean Atlantic 7'>-20°N 48 13.6-16.8 15J0 21°-4rN 114 13.3-17.7 14.4 7°-4rN 162 13.3-17.7 14.6 Gulf 19°-25''N 42 13.6-15.9 15.0 26''-29°N 43 14.7-16.2 15.3 I9°-29°N 85 13.6-16.2 15.2 Eye diameter is quite variable and several workers have noted that there are big eyed and small eyed forms in the Caribbean and Gulf of Mexico (D. M. Cohen, National Systematics Lab- oratory, National Marine Fisheries Service, NOAA, Washing-ton, DC 20560, pers. comm.). Figure 5 indicates that the eye size is quite var- iable and there is no division between the big eyed and small eyed forms. Eye size does not appear to be related to sex. Females (73 specimens) with small, interme- diate, and large eyes were noted. Only two males were found, both with eyes of intermediate size. MERISTIC CHARACTERS Values obtained for meristic characters (Tables 7, 8, 9, 10, 11) are in agreement with those given by Ginsburg (1954) for both pop- 3Z • • • 28 a • a a • • a a o • o S24 * D Z O D a ° • • K UJ 1- O •c* O . . ••• liJZO • • fcai.- • i z < K • o • t a"' • ^Ifi • >- u e o. 7^.9 * • . • • i;> o • • 10 1 J 1 1 » 1 180 200 350 400 450 STANDARD LENGTH MM 500 Figure 5. — Gulf (squares) and Atlantic (circles) populations: relation of eye diameter to standard length. 60r 50 40 30 3 2 20 ,8«^#. 180 200 qD do •■ . ■*°.*».» >- 60 X < -J -I <40 S 20 ' 300 350 STANDARD 400 LENGTH 600 MM Figure 7. — Gulf (squares) and Atlantic (circles) populations: relation of maxillary length to standard length. Table 7. — Frequency distribution of the number of gill rakers on the first gill arch for the Gulf and Atlantic populations. Population Number of gill ra kers 8 9 10 11 12 Mean Atlantic 7''-20°N 3 22 46 24 3 10.0 21MI°N 1 14 157 56 3 10.2 7°-4rN 4 36 203 80 6 10.1 Gulf 19°-25°N 2 21 54 5 __ 9.8 26°-29°N 4 13 63 8 _. 9.9 !9°-29''N 6 34 117 13 — 9.8 Table 8. — Frequency distribution of the number of first dorsal rays for the Gulf and Atlantic populations. Population Number of first dorsal rays 10 11 12 13 Mean Atlantic 7''-20''N __ 14 32 3 11.8 2r-4rN 3 63 49 1 11.4 7°-4rN 3 77 81 4 11.5 Gulf 19°-25''N _^ 16 24 2 11.7 26°-29''N __ 4 32 8 12.1 19°-29°N ~ 20 56 10 11.9 ulations. However, for all characters but the number of first dorsal rays there was an increase in the range of one to three elements. In gen- eral, the average values computed from data presented in Ginsburg (1954) for M. albidus and M. magnoculus agree well with the average values calculated for the Atlantic and Gulf pop- ulations respectively. Total vertebral counts for the Atlantic and Gulf populations were similar in both ranges and averages (Table 12). Geographic variation in most meristic characters is slight. Vertebral elements, pectoral fin rays, and anal fin rays are more variable than other meristic characters examined. The ranges for all meristic characters studied are identical or nearly so for both the Gulf and Atlantic populations. For all characters there is a difference of less than one element in the average value between the two populations. Within each population there is variation in some or all of the meristic characters studied. The Table 9. — Frequency distribution of the number of second dorsal rays for the Gulf and Atlantic populations. Population Number of second dorsal rays 35 36 37 38 39 40 41 Mean Atlantic 7''-20°N 2 14 20 8 3 _^ _^ 36.9 2r-4rN 3 22 42 39 9 1 38.3 7°-4rN 2 17 42 50 42 9 1 37.9 Gulf 19''-25°N _^ 6 14 12 8 2 __ 37.7 26°-29°N _^ 2 10 12 14 5 1 38.3 19°-29°N — 8 24 24 22 7 1 38.0 89 FISHERY BULLETIN: VOL. 71, NO. 1 Table 10. — Frequency distribution of the number of anal rays for the Gulf and Atlantic populations. Population Number of ona'l rays 35 36 37 38 39 40 41 42 Mean Atlantic 7''-20°N 2 6 22 15 2 _« _. -. 37.2 21MI°N 1 8 42 32 23 2 2 __ 37.8 7"'-41°N 3 14 64 47 30 2 2 — 37.6 Gulf 19''-25'=N 2 9 12 7 6 4 2 .. 37.6 26°-29°N __ 2 2 7 13 15 4 1 39.2 19°-29''N 2 11 14 14 19 19 6 1 38.4 Table 11. — Frequency distribution of the number of pectoral rays for the Gulf and Atlantic populations. Population Number of pectoral rays 12 13 14 15 16 17 Mean Atlantic 7"'-20°N 6 24 57 6 2 13.7 2r-4rN __ 1 28 124 72 4 15.2 7°-41°N 6 25 85 130 74 4 14.8 Gulf 19°-25°N __ 13 27 25 15 2 14.6 26°-29°N 6 42 28 11 __ _. 13.5 19°-29°N 6 55 55 36 15 2 14.0 northern Gulf population has a slightly higher average value than the southern Gulf popula- tion for all meristic characters except pectoral fin rays and vertebrae. The southern Gulf has on the average a greater number of pectoral fin rays and vertebrae (Tables 7, 8, 9, 10, 11, 12). In the Atlantic the more southerly populations have fewer vertebrae, pectoral rays, second dor- sal rays, and anal rays and more first dorsal rays than the northern populations. Material collected between lat 7° and 20°N in the Atlantic has on the average between two and three (2.5) fewer vertebrae than the ma- terial collected north of lat 21°N. There is very little overlap in the range of vertebrae in the northern and southern Atlantic populations. Only 1 of 41 specimens from south of lat 20°N has more than 53 vertebrae and only 13 of 87 specimens north of lat 20°N have less than 54 vertebrae (Table 12). However, the relatively few specimens collected between lat 16° and 20 °N may not be representative of the popula- tion residing there due to sampling error and hence, not represent the true range of vertebrae for that population. CONCLUSIONS The above data suggest that there is but a single species of Merluccius in the tropical west- ern Atlantic, including the Caribbean and Gulf of Mexico. The Gulf population as a whole can- not be distinguished from the Atlantic popula- tion by means of any of the characters examined. For all of the characters examined differences between both populations are small. Within each area there are local differences in most of the characters ; however, these differences are minor. The Gulf and Atlantic populations have identical Table 12. — Frequency distribution of the number of vertebrae for the Gulf and Atlantic populations. Population Number of vertebrae 50 51 52 53 54 55 56 Mean Atlantic 7°-20''N 6 7 19 8 I ._ 51.8 2r-4rN __ __ _^ 13 41 31 2 54.3 7°-41°N 6 7 19 21 42 31 2 53.5 Gulf 19°-25''N 1 5 14 10 2 1 53.3 26°-29°N __ 1 15 9 4 1 __ 52.6 19°-29°N — 2 20 23 14 3 1 S3.0 90 KARNELLA: SYSTEMATIC STATUS OF MERLUCCIUS or nearly identical ranges for all characters in- vestigated, and the average values for both are generally only slightly divergent. The northern Gulf population is, in many char- acters, divergent from the northern Atlantic population, which led Ginsburg (1954) to de- scribe this population as a distinct species. How- ever, the northern Gulf population is also some- what divergent from the southern Gulf and Atlantic populations and, in both cases, the di- vergence is clearly not great enough to warrant recognition at the specific level. Furthermore, the amount of overlap in all characters is of such magnitude that individuals of the northern Gulf population cannot always be distinguished from individuals from other areas. Hence, M. mag- noculus Ginsburg should be considered a junior synonym of M. albidus (Mitchill). guidance throughout this study. The Southeast Fisheries Center, Pascagoula Laboratory, Na- tional Marine Fisheries Service provided the bulk of the material from the Gulf of Mexico and Caribbean; special thanks are due Bennie A Rohr. Tomio Iwamoto of the University of Miami searched through the University of Miami Marine Laboratory and Tropical Atlantic Bio- logical Laboratory collections to find valuable material. Myvanwy M. Dick provided material from the Museum of Comparative Zoology. Keiko H. Moore of the National Marine Fisheries Service prepared the figures. My especial thanks go to George E. Clipper of the National Marine Fisheries Service for his help and many valuable suggestions. LITERATURE CITED ACKNOWLEDGMENTS Daniel M. Cohen and Bruce B. Collette of the National Systematics Laboratory, National Ma- rine Fisheries Service, NOAA reviewed the man- uscript and made valuable suggestions for im- proving it. I thank them for their advice and Ginsburg, I. 1954. Whitings on the coasts of the American con- tinents. U.S. Fish Wildl. Serv., Fish. Bull. 56: 187-208. Grinols, R. B., and M. F. Tillman. 1970. Importance of the worldwide hake, Merluc- cius, resource. In Pacific hake, p. 1-21. U.S. Fish Wildl. Serv., Circ. 332. 91 REGIONAL DISTRIBUTION OF THYROID STIMULATING HORMONE ACTIVITY IN THE PITUITARY GLAND OF THE ATLANTIC STINGRAY, DASYATIS SABINA Rodney G. Jackson and Martin Sage' ABSTRACT The possibility that the elasmobranch pituitary contains thyroid stimulating hormone (TSH) activity was investigated by measuring the increase in the release of thyroxine from thyroid glands of the Atlantic stingray, Dasyatis sabina, incubated with homogenates of various pituitary regions. The ventral lobe of the pars distalis contained most of the TSH activity, with lesser amounts in the neurointermediate lobe. Histological tech- niques were not sensitive enough to detect changes in the thyroid associated with the increase in thyroxine release. It is concluded that the elasmobranch pituitary contains TSH activity but its functional significance remains to be determined. Few studies have been conducted to examine the functional relationship between the pituitary and the thyroid gland of elasmobranchs. Dodd and Goddard (unpublished but cited by Dent and Dodd, 1961) hypophysectomized adult dog- fish, Scyliorhinus caniculus, but found no his- tological changes in the thyroid after 2 years, whereas Vivien (1964) found that after decapi- tation of Scyliorhimis embryos the thyroid failed to complete its differentiation. The latter result is, of course, open to several interpretations since decapitation removes more than the pitu- itary . Injection of homoplastic pituitary homo- genates into Scyliorhinus resulted in histological signs of stimulation of the thyroid gland (Vivien, 1941; Olivereau, 1954). Unfortunately, histo- logical methods of assessing thyroid activity are frequently both insensitive (Sage and Robins, 1970) and unreliable (Swift, 1960). Ferguson, Dodd, Hunter, and Dodd (unpub- lished data summarized by Dodd et al. (1963)) using the McKenzie mouse assay found thyroid stimulating hormone (TSH) activity in all parts of the S. caniculus pituitary, most of it being ' The University of Texas, Marine Science Institute, Port Aransas, TX 78373. Manuscript accepted May 1972. ■"ISHERY BULLETIN: VOL. 71, NO. 1, 1973. in the ventral lobe. However, the highest ac- tivity found was much less than that found in the posterior lobe of the mouse pituitary, which presumably does not contain TSH. Their re- sults could be interpreted as suggesting that the small amount of TSH activity found in the dog- fish pituitary was of no significance. The inter- pretation of assays of lower vertebrate TSH on mammalian assay systems is further complicat- ed by the probability of phylogenetic specificitj'' of hormone action. It is known that teleost TSH is relatively inactive on the mammalian thyroid (Fontaine, 1969); similarly it is possible that if there is an elasmobranch TSH it may have low activity on mammalian tissues. In a recent re- view Gorbman (1969) states that "definite proof of a TSH-like principle in elasmobranch pitui- taries remains to be provided." In an attempt to elucidate this problem we investigated the stimulatory eflFects of homogenates of the dif- ferent regions of the pituitary of Dasyatis sabina on thyroxine release from the animal's own thy- roid gland in vitro. This technique eliminates the problem of phylogenetic specificity, and, by measuring thyroxine release, avoids the problems of interpretation associated with histological assessment of thyroid activity. 93 FISHERY BULLETIN: VOL. 71, NO. 1 MATERIALS AND METHODS ANIMALS Dasyatis sabina (Lesueur) were collected in otter trawls. In the fall and winter stingrays are most abundant in the shallow waters in the Gulf of Mexico adjacent to Port Aransas, Tex. In late spring the stingrays migrate into the bays behind the line of barrier islands where they were caught during the summer (Sage et al, 1972). INCUBATION TECHNIQUE Animals were killed by cutting across the hind brain. The compact thyroid is located ventral to the anterior end of the ventral aorta. The thyroid was removed and divided into experi- mental and control halves, and further divided where necessary so that no piece of tissue was larger than 5 mg. Preliminary experiments in- dicated that the elasmobranch thyroid was slow in responding to stimulation. Thyroid tissue was therefore incubated for 3 days in 2 ml of elasmobranch saline (Nicoll and Bern, 1964). Antibiotics were added (Bakke et al., 1957) in order to inhibit bacterial growth which might result in the breakdown of the thyroxine re- leased into the medium. The addition of anti- biotics does not interfere with the ability of thyroid glands to respond to TSH (Bakke et al., 1957; Sage, 1968a) , The incubation flasks were gassed with 95% oxygen and 5% carbon dioxide and shaken at 120 strokes/min at 30°C. This temperature is within the normal environmental range of D. sabina. Modification of the incu- bation medium by the addition of 0.5 mg/ml lactalbumin hydrolysate was found to increase control rates but reduce the variability of the response and was used in later experiments as described in the text. Homogenates of whole pituitaries or various regions of the stingray pituitary were made in a glass homogenizer and added to the incuba- tion media at a concentration of one pituitary gland or region per thyroid gland. The homo- genates were added immediately prior to gassing and adding of the thyroid tissue. THYROXINE ANALYSIS At the end of the 3-day incubation period thyroid tissue was removed for histological ex- amination, and the incubation media was centri- fuged at 10,000 rcf for 10 min to remove cell debris. Incubations were then stored below 0°C until analyzed. Thyroxine was isolated by ion exchange chromatography (Galton and Pitt- Rivers, 1959) . The catalytic effect of iodine in reducing eerie ions was used to quantify thy- roxine iodine (Pileggi et al., 1961; Pileggi and Kessler, 1968). Oxford Laboratories' (San Ma- teo, Calif.) kit^ of reagents was used in the de- terminations. The results were converted to rates of thyroxine release per unit thyroid weight per incubation, and the responses of treated halves of the gland were then expressed as a percentage of the matched control incubated halves. Additives to the incubation media were routinely analyzed but were invariably devoid of thyroxine. HISTOLOGICAL METHODS At the end of the incubation period, thyroid tissue was removed and fixed in mercuric formol (90 parts saturated mercuric chloride to 10 parts formaldehyde solution). Sections were cut in polyester wax and stained with hematoxylin and light green. The image of the thyroid follicles was projected onto a sheet of paper, and a plan- imeter was used to determine the percentage of the area of follicle occupied by epithelium. Unstained thyroid sections were used for in- terferometric determinations of mass per unit area of the colloid (Bromage and Sage, 1968; Sage, 1968b) . Such methods are very sensitive in detecting changes in thyroid activity in both teleosts (Bromage and Sage, 1968) and mam- mals (Sage and Robins, 1970). ^ Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. 94 JACKSON and SAGE: TSH IN AN ELASMOBRANCH RESULTS A preliminary study was carried out to deter- mine responsiveness of the thyroid to homoge- nates of the various regions of pituitary and to control material. The results (Table 1) indi- cated that the addition of large amounts of pro- tein or protein hydrolysate resulted in a stim- ulation of the gland, thus suggesting an inade- quate culture medium. The medium was there- fore modified by the addition of 0.5 mg lactal- bumin hydrolysate/ml, and the response to var- ious regions of the pituitary reexamined (Table 2). TSH activity was greatest in the ventral lobe of the proximal pars distalis, but significant activity was also found in the neurointermediate lobe. The latter is not due to the presence of thyroxine in this pituitary lobe since the thy- roxine content of the homogenates was unde- tectable. In this respect the elasmobranch is unlike the mammal where the neural lobe does concentrate thyroxine (see review by Pitt-Rivers and Tata, 1959). Histological methods have previously been used to assay the state of thyroid activity. In Table 1. — Percentage increase in release of thyroxine from Dasyatis thyroid tissue produced by adding homo- genates of various regions of the Dasystis pituitary or lactalbumin hydrolysate to a medium containing salts, urea, and glucose. Item N Mean ± SE Rostral pars distalis (1 lobe/tlr/roid) 7 43 ±22 Neurointermediate lobe (1 lobe/thyroid) 8 *31 ± 9 Proximal pars distalis: Dorsal lobe (1 lobe/thyroid) 6 40 ± 18 Ventral lobe (1 lobe/thyroid) 7 47 ±27 Lactalbumin hydrolysate (2 mg/thyroid) 5 35 ±24 * Significantly differs from zero, P<0.01. Table 2. — Percentage increase in the release of thy- roxine from Dasyatis thyroid tissue produced by adding homogenates of various regions of the Dasyatis pituitary to a medium containing salts, urea, glucose plus lactal- bumin hydrolysate. Item N Mean SE Rostral pars distalis 9 30 ± 19 Neurointermediate lobe 9 ♦55 ± 20 Proximal pars distalis: Dorsal lobe 9 19± 15 Ventral lobe 9 **123 ±40 order to determine whether such techniques would detect stimulation resulting from incuba- tion of thyroids with whole pituitary homoge- nates, interferometric measurements on the col- loid were made together with a determination of the percentage of the follicular area occupied by epithelium. Neither technique was sensitive enough to detect the stimulation observed by measuring changes in the release of thyroxine (Table 3). Table 3. — A comparison of the effectiveness of various techniques for determining the response of Dasyatis thy- roid glands to 3-day stimulation in vitro by homogenates of whole Dasyatis pituitaries (1 pituitary/thyroid). Mean percentage Item N of control values ± SE Increase in release of thyroxine Increase in area of follicles occupied by epithelium Decrease in interferometric measure of dry wt/unit area of colloid 12 ♦51±16 7 4.4 ±5.1 12 63 ±38 Significantly differs from zero, /'<0.01. DISCUSSION Significantly differs from zero, P-CO.OS. Significantly differs from zero, /'<0.02. The present work confirms the unpublished but frequently quoted work of Ferguson et al. (Dodd et al., 1963) in that there is TSH activity in the elasmobranch pituitary and that the great- est concentration is found in the ventral lobe where gonadotropic activity has also been found (Dodd, Evennett, and Goddard, 1960) . The find- ing of lesser amounts of TSH activity in the neurointermediate lobe is in agreement with the finding of Goddard and Dodd (unpublished but quoted by Dodd et al, 1960). However, their suggestion that the activity is due to a thyro- tropin releasing factor cannot explain the pre- sent results obtained in vitro with thyroid tis- sue. The nature of the neurointermediate thy- roid stimulating substance is unknown. Dodd et al. (1963) reported that it is heat stable, whereas the activity of the dogfish ventral lobe is not. However, it is not possible to argue that the activity in the neurointermediate lobe is non- protein since all the activity present in the frog {Rana tempor^aria) pituitary is heat stable and some at least of this is presumed to be the protein TSH. 95 FISHERY BULLETIN: VOL, 71, NO. 1 The comparison of techniques for the demon- stration of TSH activity of the pituitary homoge- nates on the thyroid clearly indicates the inad- equacy of histological methods. In spite of a highly significant increase in the release of thy- roxine there was no change observed in the follicular epithelium nor in measurements on the colloid weight per unit area. This method is capable of detecting the response of teleost thy- roid follicles to a 24-hr incubation with mam- malian TSH (Bromage and Sage, 1968). While the results of incubations of thyroid with pituitary homogenates indicate TSH ac- tivity is present in the pituitary, they do not indicate whether it is of functional significance. An obvious followup to these experiments would be the removal of the ventral lobe and the mea- surement of blood thyroxine levels. However, removal of the ventral lobe in this species has not so far been possible due to the close associ- ation of this region with the carotid anastomosis. Furthermore, the analysis of thyroxine in elasmobranch blood by the present methods is difficult due to unknown factors in the blood which interfere with thyroxine analysis. From this study we conclude that TSH activity is pre- sent in the elasmobranch pituitary and that most of this activity is in the ventral lobe. However the functional significance of this activity re- mains to be determined. ACKNOWLEDGMENTS We are indebted to Mr. J. Thompson and his staff at the Marine Science Institute, especially Boat Captains E. Wingfield and J. Shanklin for help in collecting Atlantic stingrays. We also thank Dr. V. de Vlaming and L. C. Sage for their comments on the manuscript. Supported by NSF grant GB 22995. LITERATURE CITED Bakke, J. L., M. L. Heideman, N. L. Lawrence, and C. Wiberg. 1957. Bioassay of thyrotropic hormone by weight response of bovine thyroid slices. Endocrinology 61:352-367. Bromage, N. R., and M. Sage. 1968. The activity of the thyroid gland of Poecilia during the gestation cycle. J. Endocrinol. 41 : 303- 311. Dent, J. N., and J. M. Dodd. 1961. Some effects of mammalian thyroid stim- ulating hormone, elasmobranch pituitary gland extracts and temperature on thyroidal activity in newly hatched dogfish (Scyliorhinus caniculus) . J. Endocrinol. 22:395-402. Dodd, J. M., P. J. Evennett, and C. K. Goddard. 1960. Reproductive endocrinology in cyclostomes and elasmobranchs. Symp. Zool. Soc. Lond. 1:77- 103. Dodd, J. M., K. M. Ferguson, M. H. I. Dodd, and R. B. Hunter. 1963. The comparative biology of thyrotropin se- cretion. In S. C. Werner (editor). Thyrotropin, p. 3-38. Charles Thomas, Springfield. Fontaine, Y. A. 1969. Studies on the heterothyrotropic activity of preparations of mammalian gonadotropins of tel- eost fish. Gen. Comp. Endocrinol. Suppl. 2:417- 424. Galton, V. A., AND R. Pitt-Rivers. 1959. A quantitative method for the separation of thyroid hormones and related compounds from serum and tissues with an anion-exchange resin. Biochem. J. 72:310-313. GORBMAN, A. 1969. Thyroid function and its control in fishes. In W. S. Hoar and D. J. Randall (editors). Fish physiology. Vol. 2, p. 241-274. Academic Press, N.Y. NicoLL, C. S., AND H. A. Bern. 1964. Prolactin and the pituitary glands of fishes. Gen. Comp. Endocrinol. 4:457-471. Olivereau, M. 1954. Hypophyse et glande thyroide chez les pois- sons. Etude histophysiologique de quelques cor- relations endocriniennes en particulier chez Sal- mo salar L. Ann. Inst. Oceanogr. Monaco 29 : 95-296. PiLEGGI, V. J., AND G. KeSSLER. 1968. Determination of organic iodine compounds in serum. IV. A new nonincineration technic for serum thyroxine. Clin. Chem. 14:339-347. PiLEGGi, V. J., D. N. Lee, 0. J. Golub, and R. J. Henry. 1961. Determination of iodine compounds in serum. I. Serum thyroxine in the presence of some iodine contaminants. J. Clin. Endocrinol. Metab. 21 : 1272-1279. Pitt-Rivers, R., and J. R. Tata. 1959. The thyroid hormones. Pergamon Press, Lond., 247 p. Sage, M. 1968a. Assay of mammalian and fish TSH. J. En- docrinol. 41:xv. 1968b. Responses to osmotic stimuli of Xiphophorus prolactin cells in organ culture. Gen. Comp. Endocrinol. 10:70-74. 96 JACKSON and SAGE: TSH IN AN ELASMOBRANCH Sage, M., R. G. Jackson, W. L. Klesch, and V. L. DE Vlaming. 1972. Growth and seasonal distribution of the elasmobranch Dasyatis sabina. Contrib. Mar. Sci. 16:71-74. Sage, M., and P. C. Robins. 1970. The quantitative relationship between the concentration of TSH and interferometric mea- surements on the thyroid colloid. Gen. Comp. Endocrinol. 14:601-603. Swift, D. R. 1960. Cyclical activity of the thyroid gland of fish in relation to environmental changes. Symp. Zool. Soc. Lend. 2:17-27. Vivien, J. H. 1941. Contribution a I'etude de la physiologie hypophysaire dans ses relations avec I'appareil genitale, la thyroide et les corps surrenales chez les poissons selaciens et teleosteens. Bull. Biol. Fr. Belg. 75:257-309. 1964. Influence de la decapitation sur le develop- pement de I'ebauche thyroidienne de I'embryon de Scylliorhinus caniculus L. C. R. Seances Soc. Biol. Fil. 157:2068-2070. 97 EFFECT OF DRYING AND DESOLVENTIZING ON THE FUNCTIONAL PROPERTIES OF FISH PROTEIN CONCENTRATE (FPCJ David L. Dubrow' ABSTRACT Experiments were performed to determine the effects of drying and steam desolventizing on the functional properties of fish protein concentrate (FPC). The FPC's were pro- duced by a room temperature extraction of either red hake or menhaden with azeotropic isopropyl alcohol. FPC's thus produced contained about 36% soluble protein and, when dried at ambient temperature and pressure, showed very little loss in protein solubility. Drying the extracted wet solids at 40° to 50°, 60° to 70°, 90° to 100°, or 140° to 150°C for 30 or 120 min produced decreased protein solubility, i.e., 30.7% (40° to 50°C) to 12.5% (100° to 120°C). Emulsion stability of an FPC-water-oil system was satisfactory with all samples except those dried at 140° to 150°C. Desolventizing dry solids or alcohol wet solids by steam stripping produced a dramatic loss in soluble protein and emulsion stability. There was also a significant darkening in color of the FPC's desolventized as wet solids as compared to FPC's desolventized as dry solids. Food protein additives are used because of their nutritional and/or functional properties. Func- tional properties include solubility, dispersibility, water holding capacity, and emulsifying capacity (Johnson, 1969, 1970). FPC (fish protein con- centrate) can have a range of functional proper- ties depending upon the processing methods used. It is necessary, however, to control certain processing parameters in order to retain func- tionality. Extraction of fish with IP A (isopropyl alco- hol) at 20° to 30°C produces an FPC with better functional properties than extraction at 50°C (Dubrow, 1971). Similar results have been ob- tained by extracting chicken protein with IPA (Toledo, 1970).^ Although low temperature ex- tracted FPC retains a certain degree of protein solubility and emulsifying capacity, these prop- ' College Park Fishery Products Technology Labora- tory, National Marine Fisheries Service, NOAA, College Park, MD 20740. ' Toledo, R. T. 1970. Design data for a low tem- perature continuous countercurrent extraction process for protein concentrate production. Paper presented at the Institute of Food Technologists, 30th Annual Meet- ing, San Francisco, Calif. Manuscript accepted July 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. erties may be lost during subsequent drying and desolventizing of the wet solids. Drying and desolventizing is necessary to reduce the residual IPA to 250 ppm to meet FDA (Food and Drug Administration) regulations (Federal Register, 1967). The purpose of the present studies, therefore, was to determine the effect of time and temperature of drying and desolventizing on the functional properties of FPC. EXPERIMENT I: EFFECT OF TIME AND TEMPERATURE OF DRYING ON FUNCTIONAL PROPERTIES MATERIALS AND METHODS Preparation of Samples Whole red hake {Urophycis chuss) were ob- tained from Block Island off the coast of Pt. Judith, R.I. They were iced on board the fish- ing vessel and then frozen at dockside. The ex- traction process consisted of a five-stage cross- current batch extraction at 22° to 27°C. The solvent to raw fish ratio was 2:1 w/w. Each 99 FISHERY BULLETIN: VOL. 71, NO. 1 extraction stage was limited to 10 to 15 min followed by centrifugation of the solids from liquid. Under these conditions of extraction, the residual lipid in the FPC is reduced to less than 0.5%. Approximately 10 lb. from the last stage centrifuged wet solids were used for dry- ing experiments. To dry the wet solids, approximately 454 g of wet solids were placed in an aluminum foil dish and spread evenly to a depth of about 6.5 mm. Thermocouples were inserted into the bed of solids for temperature recording. The sample was then placed into a vacuum oven and subjected to drying temperatures of (1) 40° to 50°C, (2) 60° to 70°C, (3) 90° to 100°C, (4) 110° to 120°C, or (5) 140° to 150°C for either 30 or 120 min. Residence time was from the time the sample reached temperature. The sam- ple, after drying, was milled in a Wiley milF and passed through a 40-mesh screen. The samples were then placed into polyethylene bags for storage, and subsequent analysis. A control was dried overnight at ambient temperature and pressure. Methods of Analysis The following properties were determined: Salt soluble protein. — Two grams of FPC were added to 50 ml of cold 5% NaCl (in 0.02 M NaHCOa) and magnetically stirred for 3 hr (Dubrow, 1971). The slurry was filtered through Whatman *1 filter paper. The filtrate was analyzed for nitrogen by Kjeldahl method (Horwitz, 1965). Protein was calculated as N X 6.25. Emulsion stability. — Two grams of FPC were blended (Waring blender, Model *1083) in a pint-size jar with 20 ml of 5% NaCl (in 0.02 M NaHCOs) for 3 min at low speed. Twenty ml of corn oil were added to the blender and the entire mixture blended for 1.5 min at low speed. Ten ml portions of the mix were then poured ^ Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. into three graduated test tubes. The tubes were placed in a water bath (at about 98°C) for 30 min and were then cooled in an ice water bath. Since FPC is more lipophylic than hydrophylic, measurements were taken of the volume of water separated. If oil separated at the same time, measurements were also taken of this phase. Emulsion stability was calculated as the per- centage of water (total) that separated from the system. Residual isopropyl alcohol. — Residual IPA was determined according to the method of Smith and Brown (1969). Total volatiles. — Total volatiles were deter- mined by placing a weighed sample in a 103°C oven overnight; cooling in a dessicator and re- weighing. RESULTS AND DISCUSSION Table 1 shows the results of drying temper- ature and time upon the protein solubility of the FPC solids. The wet solids, prior to drying, had about 36.6% soluble protein. In compar- ison, wet solids produced by extraction at 70° to 80°C prior to drying contained only 3% sol- uble protein. Drying overnight under ambient conditions resulted in very little loss in solubility (36.4% ) . Vacuum drying at 40° to 50°C showed a 15-18% decrease in soluble protein over the ambient dried sample. Variable and unexplain- able results were obtained by drying at 60° to 70°C: the soluble nitrogen was less after 30 min drying than after 120 min. Increasing the drying temperature to 90° to 100°C or to 110° to 120 °C produced a further decrease in protein solubility. Drying at 140° to 150°C resulted in about a 65% decrease in solubility from the starting wet solids. The emulsifying stability of the dried FPC's produced under the various conditions of drying, showed that ail treatments, except the FPC's dried at 140° to 150°C formed stable oil: water emulsions (Table 1). Separation of oil and water occurred with the FPC's dried at 140° to 150°C. 100 DUBROW: FISH PROTEIN CONCENTRATE Table 1. — Effect of drying temperature and time on the salt soluble protein and emulsi- fying capacity of FPC. Tempera- ture Time Kieldahl soluble nitrogen' Soluble protein (N X 6.25) Emulsifying capacity °c hr mg N/ml % dry wt X % water separated Wet solids — 1.17 ±0,08 36.56 36.56 Ambient 16.0 2.10 ±0.04 36.43 36.43 40-50 0.5 1.80 ±0.08 31.33 2.0 1.76 ±0.08 29.99 30.66 60-70 0.5 1.44 ±0.03 24.06 2.0 I.ai ±0.11 31.46 27.76 90-100 0.5 1.76 ±0.09 28.92 2.0 1.62 ±0.10 26.88 27.90 110-120 0.5 1.31 ± 0.04 21.64 2.0 1.29 ±0.03 21.42 21.53 — 140-150 0.5 0.84 ±0.01 13.55 100 1.0 0.77 ±0.01 12.19 100 2.0 0.73 ± 0.02 11.70 12.48 too 1 Mean ± standard deviation. The effects of drying times and temperatures on the residual IPA and total volatiles are shown in Table 2. The sample of FPC dried overnight, under ambient conditions, had a residual IPA content of 2.0%, The samples dried at 40° to 50°C averaged 2.76% IPA; 60° to 70°C aver- aged 2.62%; 90° to 100°C averaged 2.53%; 110° to 120°C averaged 2.42 %r ; and 140° to 150° av- eraged 1.45%, Retention of alcohol residues of about 1 to 2% has been obtained even under prolonged drying for up to 4 hr at 70° to 80°C. Table 2. — Effect of drying temperature and time on the total volatile and residual isopropyl alcohol contents of FPC. Temperature Time Total volatiles Residual isopropyl alcohol' Wet solids Ambient 40-50 60-70 90-100 110-120 140-150 hr % % 50.00 1 - // y^ y ^ X y _ / -^ f I" ^' ■" J! .-■ y J y - / - • - 1 1 1 1 III! 400 500 600 700 WAVE LENGTH mii Figure 1. — Reflectance spectra of FPC's steam desolven- tized, as either dry solids or alcohol wet solids, at 2 to 3 psi for or 10 min. protein solubility as compared to the nonsteamed sample. The emulsifying capacity and stability of the treated solids was affected in a manner similar to that for protein solubility. Both the non- steamed and the 0-min treated samples of FPC's emulsified in oil and water systems. On the other hand, the solids steamed for 5 and 10 min showed a decrease in emulsion stability. Steam desolventization of the dry solids re- duced the residual IPA with each increment of exposure time. The initial residual content was 55,000' ppm, and after 10 min the level was found to be 367 ppm. The total volatiles of the treated solids ranged from 6.4% (0 min) to 4.9% (10 min). The color of the FPC's after steaming showed only a slight darkening. The color changed from a grayish tan to a slight yellowish tan with re- spect to time of exposure. Hunter L, a, and b values are presented in Table 3. Figure 1 illus- trates the reflectance spectra for the 0-min and 10-min FPC's and shows that the 10-min steamed dry solid sample was similar in its reflectance to the 0-min steamed wet solid sample; whereas the 0-min steamed dry solid was much lighter. CONCLUSIONS Two critical steps in the preparation of FPC by low temperature extraction with IPA are the drying and the desolventizing. Both stages in- volve heating: either dry or moist heat, or both. The results of this study showed that drying alcohol wet FPC solids at ambient conditions resulted in negligible loss in soluble protein and emulsifying capacity. Hot air drying of the wet solids, under vacuum, at temperatures ranging from 40° to 50°C to 110° to 120°C for 30 or 120 min resulted in a temperature dependent de- crease in protein solubility. The dry FPC's, however, still retained emulsifying capacity. Under these drying conditions, the residual IPA was reduced to 2 to 3%. Higher drying tem- peratures of 140° to 150°C resulted in further loss of protein solubility and a complete loss in emulsifying capacity. Removal of residual IPA, to a level of less than 250 ppm, by steam desolventization, was faster for wet solids than for dry solids. This procedure, however, brought about a 70% loss in protein solubility, a complete loss in emulsion stability, and a significant darkening of the product as compared to steam dry solids. A similar loss in functionality, but at a slower rate and with less darkening of the FPC's, re- sulted from steaming dry solids. Low temperature extraction coupled with low temperature drying produced FPC with greater functional properties than that produced by high temperature drying. To retain this function- ality, methods other than steaming appear to be necessary. ACKNOWLEDGMENT The author wishes to extend his deepest ap- preciation to Thomas BroAvn for his valuable as- sistance in the course of this work. 103 FISHERY BULLETIN: VOL. 71, NO. 1 LITERATURE CITED DUBROW, D. L. 1971. Effect of processing variables on lipid ex- traction and functional properties of fish protein concentrate (FPC). Ph.D. Thesis, Univ. Mary- land, College Park, 91 p. Federal Register. 1967. Whole fish protein concentrate. Fed. Regist. 32:1173-1175. HoRWiTz, William (chairman and editor). 1965. Official methods of analysis of the Associa- tion of Oflicial Agricultural Chemists. 10th ed. Association of Official Agricultural Chemists, Wash., D.C., XX -I- 957 p. Sections 22.003, 22.010, and 22.011. Johnson, D. W. 1969. Functional aspects involved in the use of oil- seed protein products for foods. In Conference on Protein Rich Food Products from Oilseeds. U.S. Dep. Agric, Agric. Res. Serv., ARS 72-71. 1970. Oilseed proteins-properties and applications. Food Prod. Dev., December-January, p. 78, 82, 84, 87. Smith, P., and N. L. Brown. 1969. Determination of isopropyl alcohol in solid fish protein concentrate by gas-liquid chroma- tography. J. Agric. Food Chem. 17:34-37. 104 FISH LARVAE OF THE ESTUARIES AND COAST OF CENTRAL MAINE Stanley B. Chenoweth^ ABSTRACT Seasonal sampling of fish larvae in the central Maine coast took 22 kinds of larvae; 17 were identified to species, 3 to family, and 2 were not identified. Larvae of a few highly abundant species were present in the winter and early spring. These hatched from demersal eggs and were concentrated ih the upper estuaries. The remaining species were less abundant and were present during the spring and summer. Most of these larvae hatched from pelagic eggs and were not greatly concentrated in the upper estu- aries. The larvae of only one commercially important species, Clupea harengus harengus, were found abundantly in the region. I There is little information on the species com- position and abundance of larval fishes in the numerous estuaries and bays of the coast of Maine. During the past 10 years (1961-70) samples of larval herring have been taken in the central area of the Maine coast for a program of research on the prerecruit stage of the her- ring. In three of those years (1961, 1968, and 1970) other fish larvae also were identified. An examination of the first year's catch was reported by Graham and Boyar (1965). This paper re- ports on the 1968 and 1970 identifications and gives a more complete picture of the seasonal abundance and spatial distribution of the larvae; it also compares the results with surveys in other adjacent areas. The area sampled is a system of drowned river valleys and bays typical of the Maine coast. It is bounded on the west by the Sheepscot estuary and on the east by the Damariscotta estuary, extends offshore approximately 4 miles to lat 43°45'N, and will be referred to in this report as the Boothbay region. The general ecology of the Sheepscot estuary was described by Stick- ney (1959) , and the hydrography of the area was reported by Graham and Boyar (1965). The portion of the Sheepscot estuary sampled during this study is 14 miles long, has a drainage area of 148 square miles, varies from 20 to 60 m in depth. ' Northeast Fisheries Center, National Marine Fish- eries Service, NOAA, West Boothbay Harbor, ME 04575. Manuscript accepted June 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. and is more typical of a long, narrow bay than an estuary. The portion of the Damariscotta estuary sampled is 11 miles long, has more fresh- water dilution in its upper portion than the Sheepscot, and has a smaller drainage basin. The bay separating the two estuaries is a typical coastal indentation with relatively deep water, steep rocky shores, and very little freshwater dilution. Other surveys of fish eggs and larvae from areas close to the coastal Gulf of Maine are pertinent to this study. Perlmutter (1939) and Wheatland (1956) identified the larvae from Long Island Sound, and Merriman and Sclar (1952) from Block Island Sound. Herman (1963) reported on the fish eggs and larvae of Narragansett Bay, R.I., and Pearcy and Rich- ards (1962) on those of the Mystic River estu- ary, Conn. Marak and Colton (1961), Marak, Colton, and Foster (1962), and Marak, Colton, Foster, and Miller (1962) have reported for the oflfshore area of Georges Bank and the Gulf of Maine, and Fish and Johnson (1937) for the Gulf of Maine and Bay of Fundy. METHODS Eight stations were sampled twice a month from January through August 1968 and from November 1969 through October 1970 (Figure 1). Additional information was available from occasional sampling in 1971. The larvae were 105 FISHERY BULLETIN: VOL. 71, NO. 1 44*00' 69*45' Figure 1. — Sampling stations in the Boothbay region, January through August 1968 and November 1969 through October 1970. of the net and the distance towed. The mesh opening of the trawl net (2 mm) was larger than that of the meter net (0.51 mm) used by Graham and Boyar (1965), Small larvae (<2 mm) probably escaped through the larger mesh, but the species composition of the larvae caught in both the Boothbay Depressor Trawl and meter net was similar. Larval identification was based on known spawning time and on previously reported iden- tifications. References used most often in identi- fication were Colton and Marak (1969), Bigelow and Schroeder (1953), and Graham and Boyar (1965). RESULTS Twenty-two kinds of larvae were represented in the collections in the Boothbay region during January to August 1968 and November 1969 to October 1970 (Table 1); 17 kinds were iden- tified to species, 3 to family, and 2 were not iden- tified. All of the species were boreal with centers of abundance north of the mid-Atlantic coast, and many of the more abundant larvae do not occur south of New England. Most of the larvae, particularly the more abundant ones, hatch from demersal eggs. SPECIES ABUNDANCE AND COMPOSITION preserved in 5% Formalin' and identified in the laboratory. The stations were grouped accord- ing to general location within the sampling area and are termed upper estuarine stations (1, 2, and 3), lower estuarine stations (4, 5, and 6), and outer stations (7 and 8) . The outer stations were approximately 4 miles from the headlands. Larvae were collected with a Boothbay De- pressor Trawl (Graham and Vaughan, 1966) using a 3-stepped oblique tow (10 min each level) from bottom to surface. The trawl was towed at 4 knots for 30 min. The amount of water strained was determined by using the opening ^ Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. Larvae were most abundant during late winter to early spring ( Figure 2 ) . The dominant larvae at this time were Pholis gunnellus, Liparis sp., Cryptacanthodes maculatus, Lumpenus lumpre- taeformis, and the Cottidae; they represented 91% of the total catch. In addition, Anguilla rostrata and a species of Gadidae (probably Gadus morhua) occurred in small numbers (0.1% and 0.01% of the total catch) . The com- position of the dominant kinds of larvae differed between years. The Cottidae was dominant in 1968 and P. gunnellus in 1970; C. maculatus, L. lumpretaeformis, and the Liparis sp. were more numerous in 1968 than in 1970. Catches of larvae in the winter and early spring were large between February and April 106 CHENOWETH: FISH LARVAE OF CENTRAL MAINE Table 1. — Larval fish taken in the central Maine coast region January to August 1968 and November 1969 to October 1970. Scientific name Common name Number of larvae Egg Jan.-Aug. Nov.-Oct. deposition 6,222 4,053 Demersal 2^336 5,531 Demersal 1,610 106 Demersal 19 11 Unknown 164 96 Unknown 328 225 Demersal 3 Pelagic 139 29 Unkr>own 4 Pelagic 2 Demersal 4 1 Ovoviviparous 4 6 Demersal 22 17 Pelagic 12 Demersal 15 2 Oviparous 23 8 Pelagic 127 62 Unknown 21 3 Pelagic 619 792 Demersal 8 11 Pelagic 5 Unknown 117 22 Unknown Cottidae Pholis gunnettus (Linnaeus) Liparis sp. Anguilla rostrata (Lesueur) Cryptacanthodes maculatus Storer Lumpenus lumpretaejormis (Walbaum) Gadidoe Asipidophoroides monopttrygius (Bloch) Mtrluccius bilinearis {Mitchitll) A^mmodytes americanvs DeKay Stbastes marinus (Linnaeus) Cyclopterus lumpus Linnaeus Lymanda ferruginea (Storer) Osmerui mordax (MitchilJ) Syngnathus juscus (Storer) Scophthalmus aquosus (Mitchill) Vivaria subhijurcata (Storer) Enchelyopus cimbrius (Linnaeus) Clupea harengus kartngui Linnaeus Tautogolabrus adspersus (Walbaum) Species A Species B Sculpins Rock gunnel Sea snail American eel Wrymouth Snakeblenny Codfishes Alligatorfish Silver hake American sand lance Redfi^ Lumpfish Yeliowtail flounder Rainbow smelt Northern pipefish Windowpone Radiated shanny Fourbeard rockling Atlantic herring Cunner UJ I- tlJ Z o m D O q: UJ 0. u 0.01 - < > < O (t UJ GO Z 0.001 «°~' # / ^"^ *^*/ *>^ / s^^ /..v' ^^ Figure 2. — Seasonal abundance of fish larvae in the Boothbay region, January through August 1968 and November 1969 through October 1970. and largest during the first half of March, In both years the catches were similar. The av- erage for the March peak was 0.18 per m* in 1968 and 0.14 per m^ in 1970 and for the period be- tween February and April, 0.08 per m^ in 1968 and 0.09 per m^ in 1970. The larvae were abun- dant longer in 1968 than in 1970; the large catches in 1968 extended into April and May. In spring the numbers of larvae in the catch declined sharply with the end of the larval stage of the dominant species and continued gradually to decline to a low point in July and August. Most of the remaining larvae were taken in the spring and summer and, although fewer in num- bers, more species were present. Species in this group were: Aspidophoroides monopterygius, Merluccius bilinearis, Ammodytes americanus, Sehastes marinus, Cyclopterus lumpiLS, Limanda ferruginea, Osmerus mordax, Syngnathus fus- cus, Scophthalmus aquosus, Ulvaria suhbifur- cata, Enchelyopus cimbrius, Tautogolabrus ad- spersvs. Of these, U. subbifurcata and E. cim- brius were obtained as larvae into the fall. Clupea harengus harengus hatched in September and October and was present in the area as larvae through May. The increased larval abundance in September and October was due to the hatching 107 FISHERY BULLETIN: VOL. 71, NO. 1 of Clupea harengus harengus which was the only species abundant in the autumn. The distribution of the larvae from offshore to the upper estuaries changed seasonally. Catches from the upper estuarine, lower estu- arine, and outer stations (Figure 3) showed that the larvae in the winter and early spring were i 1968 0.1 0.01 UJ o m U Ui < > ce. < o K. Ul OD Z 3 0.001 UPPER LOWER OUTER I I t ' I ' « I ' - 1969 - 70 0.01 0.001. concentrated in the upper estuaries, while the larvae in the summer were more evenly distrib- uted. The upper estuaries are probably important as nursery areas for the winter-early spring larvae. Most of this group were captured within the estuaries. From January to May the three upper stations contributed 68 ^r of the catch in 1968 and 70% of the catch in 1970. Station 2 in the upper Damariscotta estuary produced the highest catches, accounting for 40% of all the larvae taken in 1968 and 65% in 1970. The distribution of the winter-early spring group of larvae was different within the estuaries between years. In 1968 the larvae were more evenly distributed among the upper stations than in 1970. The seasonal abundance for each kind of larvae taken in the Boothbay region is shown in Figures 4 and 5. The more common kinds are discussed below. Cottidae Cottid larvae were present from January to July and their abundance reached a peak in March. Their distribution was upper estuarine and they were most abundant at station 2 (50% of all cottids in 1968, 74% in 1970). The total abundance of these larvae differed between years (6,222 in 1968 and 4,053 in 1970) because more cottids were taken at the other stations in 1968 (Figure 4A). Spawning probably occurred in the upper estuaries, inasmuch as cottids lay demersal eggs which do not drift, and yolk sac larvae were taken at the upper stations. All cottid larvae were not identified to species, but Nuzrat Khan, Department of Biology, Uni- versity of Ottawa, Ottawa, Canada (personal communication) recognized five species from 1,387 specimens of cottids that I sent to him. Of these, 689 were Myoxocephalus scorpius; 456, Myoxocephahis octodecemspinosus ; 183, Myox- ocephalus aenaeus; and 59, Triglops sp. Figure 3. — The seasonal abundance of fish larvae in three areas of the Boothbay region; the upper estuary, the lower estuary, and outside the headlands. 108 CHENOWETH: FISH LARVAE OF CENTRAL MAINE < > < - t - k 1000 z n I \ . 100 ; 1 i - 10 - - Cottidoe 1; " 1 1 1 ] 1 r T"T 1 1 1 lOOOr 100: 10: NOJ FMAMJ J ASO 20 r 10- UJ z Figure 4. — Seasonal abundance of the following kinds of fish larvae in the Boothbay region, January through August 1968 and November 1969 through October 1970: A. Cottidae, B. Pholis gunnellus, C. Liparidae, D. Anguilla rostrata, E. Cryptacanthodes maculatus, F. Lumpenus lumpretae- formis, G. Gadidae, H. Aspidophoroides monopterygius. r- I - /u 1000 - 1 ' - i\\ B ' 1 1 - 1 \\ 100 ; : V , V - l' t 1 \ , 1 10 : I - ' 1\ - PM/is gunnellus \ " 1 \ \ Mill 1 1 1 J 100 10 Lumpenus lumpretoeformis I ' I 1 I I I 10 N DJ F MAMJ J ASO Gadidae I I I I I I lOOc NOJ F MAMJ J ASO Cryptacanthodes maculatus 100 N OJ FMAMJ J ASO - Aspidophoroides monopterygius 10 N OJ FMAMJJ ASO I I I I J N OJ FMAMJJ ASO NDJFMAMJ J ASO h Pholis gunnellus The eggs of P. gunnellus are demersal, and yolk sac larvae were found in the upper estuaries during this study, suggesting that they were spawned there. The larvae appeared in the catches from January to July and reached peak I abundance in February and March (Figure 4B) . They represented 20% of the catch in 1968 and 1 50% of the catches in 1970. Their distribu- Ition was upper estuarine and, like cottids, they were most abundant at station 2 (28% of the catch in 1968 and 59% in 1970). Liparis sp. Liparid larvae were common in our catches in 1968 (14% of the catch) but less so in 1970 (1%) (Figure 4C). They probably spawn within the estuaries as they lay demersal eggs, and yolk sac larvae were found in the upper estuaries. The greatest number were taken in 109 FISHERY BULLETIN: VOL. 71, NO. 1 lOc - Merluccius bit mean's I ' ' ' I ' I ' I NDJ TMAMJ JASO z Ammodytes amtricanus I I I Jv -i_l NDJ F MAMJ JA SO z Sebastes marinus < > tr. < t 1 > > I I N DJ F MAMJ J ASO O 1 p Cyclopterus lumpus c tlJ m NDJ F MAMJ J ASO 30 1- iimando ferruginea 10- III J 20 r lOz Osmerus mordax X n M M I I I I I • I I I J I I I ■■111 j-j N DJ FMAMJJ ASO lOc Syngnathus fuscus > ' *» / I I I - I I I I I I I I i/Ni r N OJFMAMJ J ASO 20 10 p Scophthalmus aquesus 1 : H - 1 ' 1 ~ 1 f 1 \ 1 NDJ FMAMJJ ASO \0Orz yiyaria subbffureata 10- NDJ FMAMJJ ASO NDJ FMAMJJ ASO 20 10 r Enctttlyopus cimbrius 1000 I I 1 i I .T I N DJFMAMJ J ASO Clupea harengus harengus 100- NDJFMAMJJ A SO lOc Tautogolabrus adspertus NDJFMAMJ JASO Figure 5. — Seasonal abundance of the following kinds of fish larvae in the Boothbay region, January through August 1968 and November 1969 through October 1970: A. Merluccius bilinearis, B. Ammodytes ame'Hcanus, C. Sebastes m,arinu^, D. Cyclopterus lum,pus, E. Limanda ferruginea, F. Osmeriis mordax, G. Syngnathus fuscus, H. Scophthalmus aquosus, I. Ulvaria subbifurcata, J. Enchelyopus cimbrius, K. Clupea harengus harengus, L. Tautogolabrus adspersus. the upper estuaries, and the difference in abundance between years suggests that there was considerably less spawning in 1970 than 1968. Lumpenus lumpretaeformis L. lumpretaeformis probably spawns in the upper estuaries because yolk sac larvae were 110 CHENOWETH: FISH LARVAE OF CENTRAL MAINE found there and egg deposition, although not known, is probably demersal (it is for closely related forms). They were captured from Jan- uary to April with a peak abundance in March (Figure 4F). Aspidophoroides monopterygius A. monopterygiiis larvae were abundant a little later than the winter-early spring group, ranging from April to July with a peak in April or May (Figure 4H). Their distribution was more lower estuarine than upper. Ammodytes americanus and Cyclopterus lumpus The larvae of A. americaniis (Figure 5B) and C. lumpus (Figure 5D) were only rarely taken in this study but Graham and Boyar (1965) re- ported them abundant. However, these authors reexamined some of their specimens identified as Cyclopterus lumpus and Ammodytes ameri- canus and found that most identified as C. lumpus were Liparis sp. and many identified as A. americanus were Pholis gunnellus. Gadidae Several kinds of gadids spawn in our sampling area. Enchelyopus cimbrius (Figure 5J) was one of the two dominant species from June until October in the lower estuaries and outer areas. A few larvae of Merluccius bilinearis (Figure 5A) were taken in May 1970. Three specimens of what was probably Gadus morhua (Figure 4G) were taken in March 1968. Subsequent sampling (1971) took a few more G. morhuxi in December as yolk sac larvae and also later stage larvae in February in the Sheepscot estuary. Ulvarts subbifurcata This was the other dominant species in the spring and summer (Figure 51) . It was present from April until September in the lower estu- aries and outer areas. Clupea harengus harengus This is a pelagic species that lays demersal eggs and uses both the estuaries and bays as nursery areas during its larval stage from Oc- tober to May (Figure 5K). It was the only commercially important species to do so and I would consider these areas important to the population density of the species. Species A and B At present we are attempting to identify these species. Species A is probably one of the Stich- aeidae, possibly Lumpenus maculatus. Species B has been tentatively identified as Hemitripter- us americanus but needs confirmation. DISCUSSION LARVAL NURSERY AREAS Most of the fishes whose larvae were present in the Boothbay region may be placed in one of two groups: those that use the estuaries as primary spawning and nursery areas and those that do not. The larvae found in the region during the winter and early spring {Pholis gunnellus, Li- paris sp., Cryptacanthodes maculatus (Figure 4E), Lumpenus lumpretaeformis, and the Cot- tidae) belong to the first group. They were the most abundant species and their greatest con- centration was in the upper estuaries. They are larvae of resident demersal fish that are not commercially important but are extremely abun- dant in the area. They use the bays and estu- aries as nursery areas, depending to a large extent on these areas for their reproductive suc- cess. These species lay demersal eggs in the estuaries. Pearcy and Richards (1962) dis- cussed the possibility that the larvae of demersal species in the Mystic River estuary maintained themselves there by concentrating in the counter currents near the bottom. The stepped oblique tow that was used in my study was not suitable for an analysis of the depth distribution of the larvae. The winter-early spring group of larvae, 111 FISHERY BULLETIN: VOL. 71, NO. 1 however, were most abundant in the upper estu- aries throughout their larval life, and therefore probably maintained themselves there by adapt- ing to the circulation of water within the estu- aries. These larvae disappeared from the catches very rapidly during April and May, which con- tributed to the rapidly declining spring catch. By this time they were approaching the juvenile stage and, being benthic fish, probably settled to the bottom and were not available to the sam- pling gear. The remaining species (Merluccius bilinearis, Sebastes marintis (Figure 5C), Cyclopterus lumpus, Limanda ferruginea (Figure 5E) , Syng- nathus fuscus, Scopthalmus aquosus (Figure 5H), Ulvaria subbifurcata, Enchelyopus cimbri- us, and Tautogolabrus adspersus (Figure 5L) ) were present but not abundant in the spring and summer, suggesting that the estuaries were not their primary nursery areas. Possibly the num- bers of spawning adults of these species were low in the bays and estuaries, or, as most of these species lay pelagic eggs, the eggs were dispersed before the larvae hatched. Some species did not belong to either of the two above-mentioned groups: Anguilla rostrata (Figure 4D), a catadromous, and Osmerus mordax (Figure 5F), an anadromous species; Aspidophoroides monopterygius, which spawns later than the winter-early spring group, was not as common in the upper estuaries; Ammo- dytes ame7-icanus; and the Gadidae. COMPARISON WITH OTHER AREAS OF THE NORTHWEST ATLANTIC The results of surveys in other areas of the northwest Atlantic indicate the overall distri- bution of the three more abundant larvae of the Boothbay region. Cottid larvae occurred throughout the surveyed areas. Myoxocephalus aenaeus was dominant in the Mystic River estu- ary (Pearcy and Richards, 1962), Block Island Sound (Merriman and Sclar, 1952), and Long Island Sound (Wheatland, 1956) ; M. octodecem- spinosus occurred in the oflfshore areas (Marak and Colton, 1961: Marak, Colton, and Foster, 1962; Marak, Colton, Foster, and Miller, 1962) ; and M. scorpius occurred in the Gulf of Maine (Fish and Johnson, 1937) and appears from my survey to be dominant along the central Maine coast. The larvae of Pholis gunnellus appear to be more abundant in the estuaries than off- shore. They were one of the most abundant species in the Mystic estuary (Pearcy and Rich- ards, 1961) where they also concentrated in the upper estuaries. They were less abundant in the more open Narragansett Bay (Herman, 1963), rare offshore (Marak and Colton, 1961; Marak, Colton, and Foster, 1962 ; Marak, Colton, Foster, and Miller, 1962) , and absent from Long Island (Wheatland, 1956) or Block Island Sounds (Merriman and Sclar, 1952). Larvae of the Liparidae were taken in small numbers off- shore (Marak and Colton, 1961 ; Marak. Colton, and Foster, 1962; Marak, Colton, Foster, and Miller, 1962) and in the Gulf of Maine (Fish and Johnson, 1937) but not at all south of Cape Cod. Pearcy and Richards (1962) found a dominant winter-early spring group of larvae in the Mystic estuary, but the more abundant species differed from those in the central Maine coast. The dom- inant species in the Mystic estuary were Pseudo- pleuro7iectes americanus, Microgadus tomcod, and Myoxocephalus aenaeus. In Narragansett Bay (Herman, 1963) the demersal winter-early spring group of larvae was less evident with only Myoxocephalus sp. dominant, and with many more pelagic forms. An abundance of pelagic forms might be expected in Narragansett Bay because the Bay is characteristically more like the open ocean than the smaller estuaries. The spring and summer species of larvae were abundant enough in southern New England (Pearcy and Richards, 1962; Herman, 1963) to create a second summer peak in larval abun- dance that was absent in my survey of the Booth- bay region. This was probably due to the absence of larvae of such species as Stenotovius chrysops, Anchoa mitchilli, Cynoscion regalis, and Tautoga onitis which have a more southern distribution and are only occasionally taken as adults along the Maine coast (Bigelow and Schroeder, 1953). 112 CHENOWETH : FISH LARVAE OF CENTRAL MAINE LITERATURE CITED BiGELOW, H. B., AND W. SCHROEDER. 1953. Fishes of the Gulf of Maine. U.S. Fish Wildl. Serv., Fish. Bull. 53:1-577. CoLTON, J. B., Jr., and R. R. Marak. 1969. Guide for identifying the common plank- tonic fish eggs and larvae of continental shelf waters, Cape Sable to Block Island. Bur. Com- mer. Fish. Biol. Lab., Woods Hole, Mass., Ref. No. 69-9. Fish, C. J., and M. W. Johnson. 1937. The biology of the zooplankton population in the Bay of Fundy and Gulf of Maine with special reference to production and distribution. J. Biol. Board Can. 3:189-322. Graham, J. J., and H. C. Boyar. 1965. Ecology of herring larvae in coastal waters of Maine. Int. Comm. Northwest Atl. Fish., Spec. Publ. 6:625-634. Graham, J. J., and G. B. Vaughan. 1966. A new depressor design. Limnol. Oceanogr. 11:130-135. Herman, S. S. 1963. Planktonic fish eggs and larvae of Narra- gansett Bay. Limnol. Oceanogr. 8:103-109. Marak, R. R., and J. B. Colton, Jr. 1961. Distribution of fish eggs and larvae, temper- ature, and salinity in the Georges Bank-Gulf of Maine area, 1953. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 398, 61 p. Marak, R. R., J. B. Colton, Jr., and D. B. Foster. 1962. Distribution of fish eggs and larvae, tem- perature, and salinity in the Georges Bank-Gulr of Maine area, 1955. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 411, 66 p. Marak, R. R., J. B. Colton, Jr., D. B. Foster, and D. Miller. 1962. Distribution of fish eggs and larvae, tem- perature, and salinity in the Georges Bank-Gulf of Maine area, 1956. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 412, 95 p. Merriman, D., and R. C. Sclar, 1952. The pelagic fish eggs and larvae of Block Island Sound. In Hydrographic and biological studies of Block Island Sound, p. 165-219. Bull. Bingham Oceanogr. Collect. Yale Univ. 13(3). Pearcy, W. G., and S. W. Richards. 1962. Distribution and ecology of fishes of the Mystic River estuary, Connecticut. Ecology 43: 248-259. Perlmutter, a. 1939. Section I. An ecological survey of young fish and eggs identified from tow-net collections. In A biological survey of the salt waters of Long Island, 1938, Part II, p. 11-71. N.Y. Conserv. Dep., Suppl. 28th Annu. Rep., 1938, Salt-water Surv. 15. Stickney, a. p. 1959. Ecology of the Sheepscot River estuary. U.S. Fish. Wildl. Serv., Spec. Sci. Rep. Fish. 309, 21 p. Wheatland, S. B. 1956. Oceanography of Long Island Sound, 1952- 1954. VII. Pelagic fish eggs and larvae. Bull. Bingham Oceanogr. Collect. Yale Univ. 15:234- 314. 113 I EFFECTS OF TEMPERATURE AND SALINITY ON LARVAL DEVELOPMENT OF GRASS SHRIMP, PALAEMONETES VULGARIS (DECAPODA, CARIDEA)" Paul A. Sandifer* ABSTRACT Larvae of Palaemonetes vulgaris were reared in the laboratory in a factorial experiment employing three temperatures (20°, 25°, and 30°C) and six salinities (5, 10, 15, 20, 25, and 30^!^^). Temperature and salinity exerted significant effects at the 1% level on sur- vival of larvae through metamorphosis. The temperature-salinity interaction was also significant, but at the 5% level. Lowest survival occurred in 5%,; at all temperatures. In higher salinities, survival at 20° and 25°C was similar (>60%) but was significantly less at 30°C in most salinities. Temperature and salinity also influenced the rate of larval development. Development at 20 °C required nearly twice the time as that at 25° and 30°C, but a retarding influence of salinity was slight and evident only at low salinities (5 and lO^r) . Considerable variation in the number of larval instars was observed among animals which survived to the postlarval stage. Metamorphosis occurred as early as the fifth molt and as late as the twelfth. Salinity and temperature-salinity interaction had no detectable influence on the number of instars, but the effect of temperature was sig- nificant at the 1% level. Larvae reared at 25 °C passed through fewer molts prior to metamorphosis than did those reared at 20° and 30°C. Comparing survival, rate of development and number of instars, optimal conditions for larval development occurred at a moderate temperature of about- 25° C over a wide range of salinity (10 to 30^o). The grass shrimp, Palaemonetes vulgaris (Say) , ranges at least from Barnstable County, Mass., to Cameron County, Tex., (Williams, 1965) and is one of the most abundant estuarine decapods in this range. In the laboratory, Nagabhushanam (1961) found the species to be nearly euryhaline, tolerating salinities from 3 to 35%f. More re- cently, Bowler and Seidenberg (1971) found P. vulgaris to be less tolerant of low salinities (^3^f) but more tolerant of high salinities (36 and 40%c) than its congener, P. pugio. In the ' Contribution No. 511 from the Virginia Institute of Marine Science, Gloucester Point, Va. ' This study was supported in part by the Sea Grant Program of the Virginia Institute of Marine Science, under contract GH67 from the National Oceanic and Atmospheric Administration, U.S. Department of Com- merce. This paper is based on part of a dissertation to be presented to the Department of Marine Science, University of Virginia, in partial fulfillment of the re- quirements for the Doctor of Philosophy degree. ^ South Carolina Marine Research Laboratory 217 Ft. Johnson Road, P.O. Box 12559, Charleston. SC 29412. Manuscript accepted May 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. York River, Va., these authors found that the percentage of the Palaemonetes population made up by P. vulgaris decreased markedly with de- creasing salinity, and in North Carolina, Knowl- ton and Williams (1970) found P. vulgaris only in waters of 15 to 35^f salinity. Only Knowlton (1965, 1970) has studied the effects of temperature and salinity on P. vulgaris larvae, and his results were limited by the small number of experimental animals he used. The objectives of the present study were to determine the effects of temperature and salinity on sur- vival and development of P. vulgaris larvae reared through metamorphosis in the laboratory. MATERIALS AND METHODS The experimental design was a 3 X 6 factorial using temperatures of 20°, 25°, and 30°C and salinities of 5, 10, 15, 20, 25, and 30^^. Test media were prepared by diluting seawater with 115 FISHERY BULLETIN: VOL. 71, NO. I distilled water, and temperature control baths were modified from Reed (1969). Each bath was equipped with a thermostat, two 125-w heaters, a maximum-minimum thermometer, and an air stone to circulate the water. A grate sus- pended in each bath supported the culture ves- sels. The baths were placed inside a cold room maintained at 11°C, where a 60-w bulb controlled by a timer provided 14 hr of light every 24 hr, approximately coincident with times of natural daylight. Although the temperature regimes are referred to above and throughout the paper as 20°, 25°, and 30°C, the observed temperatures (mean ± one standard deviation) were 20.3°C ±: 0.7°C (range, 18.3° to 21.1°C), 25.4°C + 1.0°C (range, 22.8° to 26.7°C), and 30.6°C ± 0.5°C (range, 29.4° to 31.7°C), respectively. An ovigerous female was collected near Wach- apreague, Va., on 12 June 1970. Salinity at the collection site was approximately 30^f. The shrimp was maintained in a glass bowl at 30^f salinity and 25°C in the laboratory, and larvae were obtained on the day following collection. Active larvae were first placed in mass cultures at room temperature and fed newly hatched Ar- temia nauplii (California Brine Shrimp, Inc., Menlo Park, Calif.). Zoeae to be reared in 5, 10, and 15%fi salinity were acclimated in 15/^f for 4 hr, and those to be reared in higher salin- ities were maintained in 30/^f for 4 hr. Larvae were then transferred with a large-bore medi- cine dropper to test media in compartmented plastic boxes. Each box contained 18 compart- ments in rows of six, and one zoea in 50 ml of media was placed in every compartment. Three salinities were tested per box (i.e., each row of six compartments was a replicate of a par- ticular temperature-salinity combination), and there were six boxes in each of the three water baths. Thus, there were three replicates (one in each of three boxes) of each temperature- salinity combination, and a total of 18 larvae was reared at each condition. Larvae were transferred to clean boxes with fresh media and fed an abundance of newly hatched Artemia nauplii once daily. Molts, deaths, and maximum and minimum tempera- tures were recorded at this time. Mean temper- atures and standard deviations were calculated from the maximum and minimum temperatures. The experiment was terminated after 40 days, when all survivors were in postlarval stages. RESULTS A detailed presentation of survival and devel- opmental history of each larva reared in the pre- sent study is given in Appendix Table 1. SURVIVAL In general, survival was similar (>60%) at 20° and 25°C but was lower at 30°C in nearly all salinities. Survival in 5/^r salinity occurred only at 25°C, where 13 zoeae successfully com- pleted the first molt, and two survived through metamorphosis; in contrast, at 20° and 30°C only two zoeae molted once, and none survived to molt again. An analysis of variance on arcsin transfor- mations (Steel and Torrie, 1960) of the per- centage survival data showed difi["erences in sur- vival between temperatures and between salin- ities at the 1% level, and the temperature-sa- linity interaction was significant at the 5 /f level (Table 1). Student-Newman-Keuls' multiple range tests (Steel and Torrie, 1960) were used to explain the significant differences (Table 2). Perhaps the simplest way of looking at these differences in Table 2 is to compare survival in each salinity under each of the different tem- peratures, as is shown graphically in Figure 1. Thus, between 20° and 25°C there were signifi- cant differences in survival only in 5 and 30%c salinity. Survival at 25°C, ^Vk was significantly greater than that at 20°C, hVu, while at 20°C. 30%f survival was significantly greater than at 25°C, 30^/^r. Comparing 20° and 30°C, survival at 20 °C was significantly greater than that at 30°C in 10, 15, and 257ff . Finally, comparing 25° and 30°C, survival at 25°C was significantly greater than that at 30°C in 5, 10, 15, and 25^^. Highest overall percentage survival (88.99^) occurred at the combination 20°C, 20^. (Table 2, Figure 1). 116 SAN'DIFER: LARVAL DEVELOPMENT OF GRASS SHRLMP Table 1. — Analysis of variance for differences in survival of Palaemo- netes vulgaris larvae through metamorphosis under different conditions of temperature and salinity. Source of variation Degrees of freedom Sum of squares Mean square F Temperature Salinity Temperature X salinity Error 2 5 10 36 4,912.0345 21,212.5667 3,842.1588 4,950.1734 2,456.0172 4,242.5133 384.2158 137.5048 17.8613** 30.8535** 2.7941* Total 53 34,916.9304 ** Significant at 1% * Significant at 5% level level. Table 2. — Summary of Student-Newman-Keuls' multiple range tests to explain differences in survival of Palae- monetes vulgaris larvae at different temperature and salinity conditions. 20 "C Experimental conditions Mean {transformed % survival) Means not overlapped by the same line ore °C %, ent at the 1% level 30 5 0.0 20 5 0.0 25 5 16.1 30 10 24.1 30 15 31.5 30 25 38.5 25 30 52.0 30 20 58.5 30 30 58.5 20 25 58.5 25 10 62.2 25 20 62.2 20 30 65.9 20 10 66.5 25 25 67.0 25 15 70.2 20 16 70.2 20 20 73.9 r 10 15 20 SALINITY (%o) 25 30 Figure 1. — Comparison of survival to postlarvae for Palaemonetes vulgaris zoeae reared at different temper- atures and salinities. RATE OF DEVELOPMENT The effects of temperature and salinity (ex- cluding h'/(c) on the rate of larval development are shown in Figure 2. The effect of temper- ature was pronounced; development at 20°C was much slower than at 25° or 30°C. Mean dur- ation of development (days) ±: one standard de- viation was 30.2 ± 3.8 (range, 23 to 39) at 20°C, 16.6 ± 2.7 (range, 14 to 25) at 25°C, and 15.7 ± 1.8 (range, 13 to 21) at 30°C. Salinity in- fluenced the rate of development much less than did temperature. Survival in h'/ic salinity oc- curred only at 25°C, where the larvae in ^%, gen- erally required about 1 to 4 more days to pass a given stage than did larvae in higher salinities at the same temperature. Development in lO'/ic also tended to be slightly slower than in higher salinities, regardless of the temperature (Fig- ure 2). There was little apparent difference among developmental rates in 15 to Z0'/,(. In general, a Qio (20° and 30°C) of about 1.8 was typical of larval development. Mean duration of instars (Table 3) was in- versely related to temperature, reflecting devel- opmental rate. Duration of successive instars tended to increase slightly at 20°C. The second instar was markedly short at 25° and 30°C, and the final instar was of longest duration at all temperatures. Overall mean instar duration (days) ± one standard deviation for animals which completed development was 3.6 ±: 0.8 (range, 3 to 7) at 20°C, 2.2 ± 0.7 (range, 1 to 7) at 25 °C, and 1.9 ± 0.6 (range, 1 to 4) at 30 °C. 117 FISHERY BULLETIN: VOL, 71, NO. I 20 •€ 25 'C 30* C 40-1 38- 36 34- 32- 30- 28- 26- 24- 22 20 I8i 16 14 H 12 15 16 — 1 1 — I 1 r 10 15 20 25 30 'M} r4 '* \t — I — I — I — I — I 10 15 20 25 30 SALINITY (%.) I ttl ■J^-+- I r I l|^ 10 15 20 25 30 Figure 2. — Mean ± one standard deviation and range of days required for Palae-monetes vulgaris larvae to reach the postlarval stage at different temperatures and salinities {5%o excluded) (numbers at the lower end of each range line indicate the number of animals which reached the postlarval stage at those conditions of tem- perature and salinity). Table 3. — Mean duration (days) of Palaemonetes vul- garis larval instars at different temperatures. Final in- star treated separately. Temperature {°C) 20 25 30 5 Days Days Days 1- 1 3.0 2.0 2.0 o II 3.1 L4 1.0 2 III 3.1 2.0 1.2 IV 3.2 2.0 1,9 V) V 3.5 2.1 1.9 _l VI 3.8 2.1 2.0 < VII 3.9 2.1 2.0 VIM 3.5 2.4 2.0 2 IX 3.7 12.2 11.8 < X M 12 __ XI — 12 — u. O Final 5.0 3.3 2.9 ^ Based on five or fewer larvae. VARIATION IN NUMBER OF INSTARS Most larvae metamorphosed at the 7th, 8th, or 9th molt, but there was much variation in number of instars. Metamorphosis occurred at the 5th through the 12th molts, and one zoea passed through 12 zoeal instars but never reached the postlarval stage. The effects of temperature and salinity (ex- cluding 5/^f because only two postlarvae were obtained there) on the number of larval stages are shown in Figures 3 and 4. Sample sizes were unequal, so an approximate method, the analysis of unweighted means (Snedecor, 1956) was em- ployed to indicate significant effects (Table 4). The effect of salinity was not significant, al- though there appeared to be a slight tendency I- o o < UJ s < -I \- V) O a. o h- o UJ o z UJ o q: UJ Q. 20%. ys 40- ^^^^ \ / X^.»***'^ // •'^X" 20- /' ./^a::::^.. 0- Figure 3. — Percentage of animals molting to postlarva at each molt under different conditions of temperature and salinity. 118 SANDIFER: LARVAL DEVELOPMENT OF GRASS SHRIMP 50 n 30*C MOLTS Figure 4. — Percentage of animals molting to postlarva at each molt under different temperatures. for fewer instars in 15/rf than in other salinities. The temperature-salinity interaction also was not significant. However, the influence of tem- perature was significant at the 19c level, and mean numbers of instars ±: one standard devi- ation were 8.5 ± 1.0 at 20°C, 7.8 ± 1.1 at 25°C, and 8.4 ± 1.0 at 30°C. A multiple mean test showed no difference between the mean numbers of larval instars passed at 20° and 30°C but indicated that animals reared at 25°C passed through significantly fewer instars in larval de- velopment. DISCUSSION Few previous studies have been concerned with the eflJ'ects of temperature and salinity on Palaemonetes larvae. Sollaud (1919) reared larvae of P. varians microgenitor in the labora- tory and found, as I did for P. vulgaris, that de- velopment was retarded at low temperatures and in low salinities and that more instars were passed at the lower than at the more moderate temperature tested. According to Broad and Hubschman (1962), development of larvae of P. intermedms, P. pugio, and P. vulgaris was un- affected by salinity above 20V.V, but below 10%o survival was poor. In the present study, sur- vival in 5',( was very poor, but in salinities of 10 to S0'/(( at low and moderate temperatures (20° and 25°C), survival was high. More re- cently, Knowlton (1970) conducted a factorial experiment similar to mine, but he used only five larvae in each temperature-salinity combi- nation. Knowlton (1970) found that at 20° and 25°C P. vulgaris larvae seemed to tolerate the entire range of salinity tested (15 to So.'/r ) equal- ly well, with highest survival among larvae reared at 25°C. Lowest survival occurred among larvae reared at 30°C, where no larvae exposed to the low salinities (15 and 20^'^) completed development. The results of the present study were fairly similar, except that some larvae sur- vived through metamorphosis at 30°C in all sa- linities but 5%c. However, Knowlton's (1970) values for mean duration of larval life (37.3 ± 2.0 days at 20°C, 30.7 ± 2.0 days at 25°C, and 31.1 ± 4.3 days at 30°C) were considerably greater than corresponding values in the present study (30.2 ± 3.8 days, 16.6 ± 2.7 days, and 15.7 ± 1.8 days, respectively). Similarly, his values for mean instar duration were greater than val- ues determined here. The number of larval instars varied from 8 to 16 in Knowlton's (1970) study, while in the present study the observed range was 5 to 12. Knowlton (1965,1970) also found that the num- ber of larval instars increased with increasing Table 4. — Summary of analysis of variance for differences in number of larval molts for Palaemonetes vulgaris larvae at different temper- atures and salinities. Source of variation Degrees of freedom Sum of squares Mean square F Total 14 4.0364 Temperature 2 2.2770 1.1385 10.8017** Salinity 4 0.5398 0.1349 1.2798 n.s. Temperature X salin ty 8 1.2196 0.1524 1.4459 n.s. Error M67 10.1054 ** Significant a\ 1% level. n.s. Not significant. 1 See Snedecor (1956) for computation of the error mean square in the method of un- weighted means. 119 FISHERY BULLETIN: VOL. 71, NO. 1 temperature. In contrast, larvae in my study passed through fewer instars at the moderate temperature (25°C) than at higher or lower tem- peratures, and at each temperature larvae re~ quired fewer molts to reach the postlarval stage in my study than in Knowlton's (1970). Simi- larly, Ewald (1969) found that Tozeuma ca^'ol- inense larvae passed through fewer instars at 25°C than at 15° and 20°C. He also reported that there were marked differences in the num- bers of instars among T. carolinense larvae from different populations. Perhaps a similar effect was partially responsible for differences between the numbers of P. vulgaris larval instars ob- served by Knowlton (1970) and by me. The final zoeal instar was of greater duration than the other instars in both Knowlton's (1970) study and in mine, but the reason for the delay of this molt is not known. However, Hubschman (1963) reported that the X organ-sinus gland complex does not become functional as the pri- mary molt regulator in Palaemonetes until after metamorphosis. He suggested that perhaps the rapid larval molting cycle was under the hor- monal control of some type of larval molting gland, the existence of which remains specu- lative. The longer duration of the final zoeal instar thus may reflect transfer of control over molting from some unknown larval molt-regu- lating mechanism to the X organ-sinus gland complex, or breakdown of the larval regulatory mechanism prior to assumption of molt-regu- lating function by the X organ-sinus gland com- plex, or other internal reorganization prior to metamorphosis. Because of the characteristic variability of temperature and salinity in estuaries, success of a particular decapod species may depend on the ability of the larvae to survive frequent expo- sure to suboptimal temperature-salinity condi- tions, to settle and/or metamorphose only under those conditions which are suitable for survival of the adult form, and to remain within, be car- ried into, or return to a given area to replenish the parental population. The number of larval instars may also be important, since ecdyses are critical periods in larval life, and highest mor- tality of cultured decapod larvae often occurs then (Ong, 1966; Knowlton, 1970; Roberts, 1971). Reduction of the number of premeta- morphic molts thus may increase larval survival. So, considering survival, rate of development, and number of instars, it appears that optimal conditions for larval development of P. vulgaris occur at a moderate temperature of about 25°C in salinities of 10 to 30^/f. Knowlton (1970) also concluded that a temperature of 25 °C was optimal over the salinity range tested (15 to 35'/w) in his experiment. ACKNOWLEDGMENTS I would like to thank my graduate committee (Drs. G. C. Grant, W. G. Maclntyre, W. C. Pin- schmidt, Jr., and M. L. Wass, and especially Mr. W. A. Van Engel, Chairman) and my wife, Betty, for constant help and encouragement, and Drs. M. E. Chittenden and J. Loesch for advice re- garding the design and analysis of the exper- iment and for critical review of the manuscript. I was the recipient of a National Defense Edu- cation Act Title IV Graduate Fellowship during the study. LITERATURE CITED Bowler, M. W., and A. J. Seidenberg. 1971. Salinity tolerance of the prawns, Palaemone- tes vulgaris and P. pugio, and its relationship to the distribution of these species in nature. Va. J. Sci. 22:94. Broad, A. C, and J. H. Hubschman. 1962. A comparison of larvae and larval develop- ment of species of Eastern U.S. Palaemonetes with special reference to the development of Palaemonetes intermedius Holthuis. Am. Zool. 2:394-395. EWALD, J. J. 1969. Observations on the biology of Tozeuma carolinense (Decapoda, Hippolytidae) from Flor- ida, with special reference to larval development. Bull. Mar. Sci. 19:510-549. Hubschman, J. H. 1963. Development and function of neurosecretory sites in the eyestalks of larval Palaemonetes (Decapoda: Natantia). Biol. Bull. (Woods Hole) 125:96-113. Knowlton, R. E. 1965. Effects of some environmental factors on larval development of Palaemonetes vulgaris (Say). J. Elisha Mitchell Sci. Soc. 81:87. 120 SANDIFER: LARVAL DEVELOPMENT OF GRASS SHRIMP 1970. Effects of environmental factors on the larval development of Alpheiis heterochaelis Say and Palaemonetes vulgaris (Say) (Crustacea Decapoda Caridea), with ecological notes on larval and adult Alpheidae and Palaemonidae. Ph.D. Thesis. Univ. North Carolina (Libr. Congr. Card No. Mic. 71-3573) 544 p. Univ. Microfilms, Inc., Ann Arbor, Mich. (Diss. Abstr. 31 :5076-B) . Knowlton, R. E., and A. B. Williams. 1970. The life history of Palaemonetes vulgaris (Say) and P. pugio Holthuis in coastal North Carolina. J. Elisha Mitchell Sci. Soc. 86:185. Nagabhushanam, R. 1961. Tolerance of the prawn, Palaemonetes vul- garis (Say), to waters of low salinity. Sci. Cult. 27:43. Ong, K. S. 1966. The early developmental stages of Scylla serrata Forskal (Crustacea Portunidae), reared in the laboratory. Indo-Pac. Fish. Counc. Proc. 11th Sess., Sect. 2, p. 135-146. Reed, P. H. 1969. Culture methods and effects of temperature and salinity on survival and growth of Dungeness crab (Cancer magister) larvae in the laboratory. J. Fish. Res. Board Can. 26:389-397. Roberts, M. H., Jr. 1971. Larval development of Pagurus longicarpus Say reared in the laboratory. II. Effects of re- duced salinity on larval development. Biol. Bull. (Woods Hole) 140:104-116. Snedecor, G. W. 1956. Statistical methods, applied to experiments in agriculture and biology. 5th ed. Iowa State College Press, Ames, Iowa, 534 p. SOLLAUD, E. 1919. Influence des conditions du milieu sur les larves du Palaemonetes variants microgenitor Boas. C. R. Acad. Sci. 169:735-737. Steel, R. G. D., and J. H. Torrie, 1960. Principles and procedures of statistics with special reference to the biological sciences. Mc- Graw-Hill, N.Y., 481 p. Williams, A. B. 1965. Marine decapod crustaceans of the Carolinas. U.S. Fish Wildl. Serv., Fish. Bull. 65:1-298. I 121 FISHERY BULLETIN: VOL. 71, NO. 1 APPENDIX TABLE 1. --Comparison of survival and developmental rates of Palaemonetes vulgaris larvae reared at different temperatures and salinities. Tem- Sa- pera- lin- ity Survival Age (days) Survival Age (days) Survival Age (days) Survival Age (days) ture (°C) (°/oo) % No. Mean Range 'L No. Mean Range 7o No. Mean Range 7. No. Mean Range Molt No. 1 Molt No. 2 «olt No. : Molt No. 4 Zoea I to zoea II Zoea II to zoea III Zoea III to zoea IV Zoea IV to zoea V 20 5 11.1 2 4.0 -- 0.0 -- -- 0.0 -- -- 0.0 -- -- 10 100.0 18 3.0 -- 100.0 18 6.5 6-9 88.9 16 9.8 9-12 88.9 16 13.5 12-16 15 100.0 18 3.0 -- 100.0 18 6.0 -- 100.0 18 9.1 9-10 100.0 18 12.3 12-17 20 100.0 18 3.0 -- 100.0 18 6.0 -- 100.0 18 9.0 -- 100.0 18 12.0 -- 25 100.0 18 3.0 -- 100.0 18 6.0 -- 100.0 18 9.0 -- 100.0 18 12.0 -- 30 100.0 18 3.0 -- 100.0 18 6.0 -- 100.0 18 9.0 -- 100.0 18 12.0 -- 25 5 72.3 13 2.8 2-7 27.8 5 6.2 4-9 22.2 4 9.8 8-13 22.2 4 11.8 10-15 10 100.0 18 2.0 -- 100.0 18 3.8 3-4 100.0 18 5.8 4-6 100.0 18 7.8 7-8 15 100.0 18 2.0 -- 100.0 18 3.1 3-4 100.0 18 5.1 5-6 100.0 18 7.1 7-8 20 100.0 18 2.0 -- 100.0 18 3.2 3-5 100.0 18 5.2 5-7 100.0 18 7.2 7-9 25 100.0 18 2.0 -- 100.0 18 3.4 3-4 100.0 18 5.3 5-6 100.0 18 7.3 7-8 30 100.0 18 2.0 — 100.0 18 3.3 3-4 100.0 18 5.3 5-6 88.9 16 7.3 7-9 30 5 11.1 2 2.0 __ 0.0 _. 0.0 __ 0.0 __ ._ 10 100.0 18 2.0 -- 100.0 18 3.0 -- 100.0 18 4.9 4-5 94.5 17 6.9 6-7 15 94.5 18 2.0 -- 88.9 16 3.0 -- 83.4 15 4.3 4-5 83.4 15 6.3 6-7 20 100.0 18 2.0 -- 100.0 18 3.0 -- 94.5 17 4.1 4-5 94.5 17 6.1 6-7 25 100.0 18 2.0 -- 100.0 18 3.0 -- 100.0 18 4.2 4-5 100.0 18 6.1 6-7 30 100.0 18 2.0 Molt No. 100.0 5 18 3.0 100.0 18 4.2 4-5 Molt No. 94.5 6 17 6.0 Zoea V to zoea VI Zoea V to post larva Zoea VI to zoea VII Zoea VI to pos t larva 20 5 0.0 -- -- 0.0 -- -- 0.0 -- -- 0.0 -- -- 10 88.9 16 17.4 16-20 0.0 -- — 83.4 15 21.4 20-25 0.0 -- -- 15 100.0 18 16.2 15-21 0.0 — — 94.5 17 20.1 18-25 0.0 -- -- 20 100.0 18 15.4 15-19 0.0 -- — 94.5 17 19.1 18-23 0.0 -- -- 25 100.0 18 15.2 15-16 0.0 -- — 94.5 17 19.1 18-20 0.0 -- -- 30 100.0 18 15.1 15-16 0.0 -- -- 100.0 18 18.7 18-19 0.0 -- -- 25 5 16.7 3 12.7 12-13 5.6 1 22.0 __ 16.7 3 14.7 14-15 0.0 -. .. 10 100.0 18 10.0 9-11 0.0 -- — 94.5 17 12.4 11-13 5.6 1 15.0 -- 15 100.0 18 9.2 9-10 0.0 -- -- 100.0 18 11.2 11-12 0.0 -- -- 20 100.0 18 9.2 9-11 0.0 -- — 94.5 17 11.1 11-12 0.0 -- — 25 94.5 17 9.4 9-10 0.0 -- — 94.5 17 11.5 11-13 0.0 -- -- 30 88.9 16 9.3 9-11 0.0 -- -- 83.4 15 11.2 11-12 5.6 1 15.0 — 30 5 0.0 __ _- 0.0 __ „_ 0.0 __ 0.0 10 94.5 17 9.0 8-11 0.0 -- -- 66.7 12 10.8 9-11 0.0 -- -- 15 66.7 12 8.3 8-9 0.0 -- -- 55.6 10 10.6 10-11 0.0 -- -- 20 88.9 16 8.1 8-9 0.0 -- -- 83.4 15 10.0 9-11 0.0 -- -- 25 100.0 18 7.9 7-9 0.0 -- -- 94.5 17 9.9 9-11 0.0 -- -- 30 94.5 17 7.7 7-8 Molt No. 0.0 7 88.9 16 9.7 9-10 Molt No. 0.0 8 Zoea VII to zoea VIII Zoea VII to pos t larva Zoea VIII to zoea IX Zoea VIII to postlarva 1 20 5 0.0 -- — 0.0 -- -- 0.0 — — 0.0 -- 1 10 83.4 15 25.4 24-29 0.0 -- -- 38.9 7 29.0 28-30 44.5 8 30.9 29-35 15 50.0 9 23.3 23-24 33.4 6 26.7 24-30 16.7 3 27.3 27-28 33.4 6 28.3 27-30 20 83.4 15 22.8 22-24 11.1 2 26.0 24-28 44.5 8 26.4 25-27 33.4 6 27.5 27-28 25 83.4 15 22.9 22-24 11.1 2 25.5 25-26 50.0 9 26.6 25-28 22.2 4 27.5 27-28 30 94.5 17 22.1 21-23 5.6 1 23.0 -- 55.6 10 25.3 24-27 33.4 6 27.2 26-28 25 5 5.6 1 18.0 -- 0.0 ._ ._ 0.0 __ .. 5.6 1 21.0 .. 10 50.0 9 14.7 14-16 33.4 6 16.3 14-17 27.8 5 17.4 16-18 16.7 3 17.3 17-18 15 50.0 9 13.3 13-14 38.9 7 14.6 14-15 5.6 1 15.0 -- 38.9 7 17.0 16-18 20 55.6 10 13.5 13-16 33.4 6 14.0 -- 22.2 4 15.3 15-16 22.2 4 16.8 16-17 25 50.0 9 13.8 13-16 44.5 8 14.8 14-16 33.4 6 16.2 15-18 16.7 3 16.0 -- 30 27.8 5 13.4 13-14 33.4 6 14.2 14-15 16.7 3 16.0 -- 11.1 2 16.5 16-17 30 5 0.0 -- _. 0.0 ._ 0.0 .. .. 0.0 .. __ 10 61.2 11 12.9 12-15 0.0 -- -- 44.5 8 14.6 13-15 0.0 — — 15 33.4 6 12.7 12-14 5.6 1 15.0 — 22.2 4 14.5 14-16 5.6 1 16.0 -. 20 55.6 10 11.9 11-13 22.2 4 13.0 — 16.7 3 14.0 -. 33.4 6 15.0 14-16 25 66.7 12 11.8 11-13 5.6 1 13.0 — 38.9 7 13.6 13-14 5.6 1 15.0 -- 30 77.8 14 11.6 10-12 11.1 2 13.5 13-14 44.5 8 13.3 12-14 22.2 4 15.3 15-16 122 SANDIFER: LARVAL DEVELOPMENT OF GRASS SHRIMP APPENDIX TABLE 1 .--Comparison of survival and developmental rates of Palaemonetes vulgaris larvae reared at different temperatures and salinities — Continued . Tem- Sa- pera- lin- Survival Age (days) Survival Age (days) S urvivs 1 Age (days) Survival Age (days) ture (°C) ity (°/oo) Z No. Mean Range 7. No. Mean Range X No . Mean Range ?o No. Mean Range Molt No. 9 Molt No. 10 Zoea IX to zoea X Zoea IX to postlarva Zoea X to zoea XI Zoea X to postlarva 20 5 0.0 -- -- 0.0 -- -- 0.0 -- -- 0.0 -. __ 10 11.1 2 33.0 32-34 22.2 4 34.8 33-36 0.0 -- -- 11.1 2 38.0 37-39 15 5.6 1 32.0 — 11.1 2 32.0 — 0.0 — — 5.6 1 37.0 — 20 22.2 4 30.8 30-31 22.2 4 30.8 30-32 5.6 1 34.0 — 16.7 3 35.0 34-36 25 27.8 5 29.6 28-31 16.7 3 32.3 31-33 5.6 1 35.0 -- 22.2 4 34.0 32-35 30 27.8 5 28.4 27-29 22.2 4 29.5 28-31 0.0 — -- 22.2 4 33.0 31-34 25 "5 0.0 __ __ 0.0 ._ __ 0.0 __ ._ 0.0 __ 10 5.6 1 19.0 — 16.7 3 21.7 20-23 0.0 — -- 5.6 1 23.0 -- 15 0.0 -- -- 5.6 1 18.0 — 0.0 — — 0.0 — -- 20 11.1 2 17.0 -- 11.1 2 18.5 18-19 5.6 1 19.0 -- 5.6 1 20.0 .. 25 16.7 3 18.7 17-20 11.1 2 19.5 18-21 5.6 1 23.0 — 5.6 1 20.0 -- 30 11.1 2 18.0 — 5.6 1 19.0 -- 5.6 1 21.0 -- 5.6 1 21.0 -- 30 5 0.0 _» __ 0.0 _. _. 0.0 __ __ 0.0 __ _. 10 16.7 3 16.7 16-17 11.1 2 16.5 16-17 5.6 1 19.0 — 5.6 1 19.0 -- 15 11.1 2 17.0 16-18 11.1 2 17.0 — 0.0 — — 5.6 1 21.0 -- 20 5.6 1 16.0 — 11.1 2 16.5 16-17 0.0 -- — 5.6 1 18.0 -- 25 11.1 2 15.5 15-16 16.7 3 16.3 16-17 0.0 ._ 11.1 2 18.0 17-19 30 5,6 1 15.0 Molt No. 38.9 11 7 16.3 15-17 0.0 — — Molt No. 0.0 12 — — — Zoea XI to zoea XII Zoea XI to postlarva Zoea XII to zoea XIII Zoes XII to postlarva 20 5 0.0 -- — 0.0 — -- 0.0 -- -_ 0.0 — .. 10 0.0 -- -- 0.0 -- -- 0.0 -- -- 0.0 .. -- 15 0.0 — -- 0.0 -- -- 0.0 -- -- 0.0 — -- 20 0.0 -- -- 5.6 1 39.0 — 0.0 -- — 0.0 — -- 25 5.6 1 39.0 — 0.0 — -- 0.0 -- -- 0.0 -- -- 30 0.0 -- -- 0.0 -- — 0.0 -- -- 0.0 — -- 25 5 0.0 .- 0.0 __ -.. 0.0 __ __ 0.0 ,„ __ 10 0.0 — -- 0.0 -- -- 0.0 — -- 0.0 .- — 15 0.0 — — 0.0 — — 0.0 -- -. 0.0 -. -- 20 5.6 1 21.0 — 0.0 — -- 0.0 — — 5.6 1 25.0 -- 25 5.6 1 26.0 -- 0.0 — — 5.6 1 28.0 — 0.0 -- .- 30 0.0 -- — 0.0 — -- 0.0 -- -- 0.0 -- -- 30 5 0.0 _. __ 0.0 _. __ 0.0 __ __ 0.0 __ __ 10 0.0 -- — 0.0 -- -- 0.0 — — 0.0 — -- 15 0.0 -- — 0.0 -- — 0.0 — — 0.0 — -- 20 0.0 -- -- 0.0 -- — 0.0 — -- 0.0 — — 25 0.0 — -- 0.0 -- — 0.0 — -- 0.0 — -- 30 0.0 — -- 0.0 — — 0.0 -- -- 0.0 — -- 123 ERYTHROCYTE DEGENERATION IN THE ATLANTIC HERRING, CLUPEA HARENGUS HARENGUS L. Stuart W. Sherburne^ ABSTRACT Cytoplasmic inclusions, associated with erythrocytic degeneration, were found in the circulating blood of herring from Boothbay Harbor, Maine, and from Passamaquoddy Bay at Deer Island, N.B., Canada, in 1969. Except in one instance, when inclusions occurred in herring from water of 2°C, all herring from Boothbay Harbor having in- clusions were taken from seawater temperatures of 13.8°C or above. A relationship appears to exist between inclusions in herring erythrocytes and stress factors, especially temperature extremes. At a temperature of 16°C, 96% of a sample of herring were affected with inclusions. Herring sampled at the highest temperature (16°C) were markedly different from all other samples in their blood morphology and had the highest incidence of inclusions. Inclusions were found in the Passamaquoddy Bay area in 2 of the 50 herring sampled from a seawater temperature of 9.8°C, the highest temperature sampled in that area. Inclusions rarely occurred more than one to a red cell and varied in size from 1.3 to 3.9 /I. In herring containing a high incidence of inclusions, the larger inclusions were usually in the youngest red cells. Cells containing inclusions generally appeared rounded and swollen. Either an abnormally high percentage of up to 90% immature red cells or a low of 1 to 5% immature red cells generally characterized herring containing in- clusions. The blood of herring has been studied at the National Marine Fisheries Service Laboratory at Boothbay Harbor to find physiological indi- cators of environmental stress that may help us to determine causes of fluctuations in success of year classes. During this investigation I ob- served inclusion bodies in the cytoplasm of the red cells in many of the herring. In this report I describe these inclusion bodies, their incidence, and the abnormal blood cell morphology asso- ciated with these bodies. Nonspecific cytoplasmic inclusions have been reported in Fundulus sp. (Gardner and Yevich, 1969) occurring in wet smears in May and July prior to, and at the beginning of the new breed- ing season, but not evident in fixed smears. The cytoplasm of erythrocytes from chinook salmon, Oncorhynchiis tshaivytscha, sockeye salmon, Oncorhynchus nerka, and adult rainbow trout, Salmo gairdneri, contained granular material ^ Northeast Fisheries Center, National Marine Fish- eries Service, NOAA, West Boothbay Harbor, ME 04575. Manuscript accepted August 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. following fixation procedures (Ridgway, 1956) that the author thought were of mitochondrial origin. Laird and Bullock (1969) reported finding a distinctive inclusion body formed in the cyto- plasm of infected cells associated with piscine erythrocytic necrosis which is responsible for massive red blood cell destruction in Gadus mor- hua from Passamaquoddy Bay. Liparis atlanti- cus from Kent Island, N.B., Canada, and Myox- ocephaliis octodecemspinosus from Portsmouth Harbor, N.H. were lightly infected. MATERIALS AND METHODS The 355 herring examined in this study from February through October 1969 consisted of 201 wild herring and 154 captive herring in 12 sam- ples. The herring ranged in length from 12.5 to 30.4 cm and in weight from 10.6 to 214.5 g. The wild herring were taken from four fisher- men's catches between central Maine and Can- ada. Three categories of herring are considered 125 FISHERY BULLETIN: VOL. 71. NO. 1 in this report: 1) long-term captive herring held 6 months before sampling began in Feb- ruary and terminated in June when the supply of test fish was exhausted, 2) short-term captives which consisted of herring held 2 weeks before being bled, and 3) wild herring that were taken when available. The captive herring were held in seawater which was pumped from the ocean through the tanks and which approximated the temperature of natural seawater. The water temperature was recorded at the site of capture in each instance. A blood sample was taken from the heart of each herring and preserved in a modified Alsev- er's solution for serological studies; a microhe- matocrit was determined and a morphology slide made for each herring. The herring were mea- sured for total length, weighed, sexed, marked, and frozen for reference. All herring were ex- amined for gross parasitism. The hematocrits and morphology slides were made of blood taken by direct heart puncture with a heparinized 75 mm x 1.3-1.5 mm outside diameter capillary tube. A small drop of blood from the tube was placed on a microscope slide, the tube sealed with plastic clay, and the smear made. The tubes were centrifuged in a micro- hematocrit centrifuge for 31/2 min at 11,000 rpm and read in a microcapillary reader. Slides were air-dried and stained by either the Wright's or Wright-Giemsa staining method. Distilled wa- ter was used as a diluent for the Wright's and Giemsa stains. Cells were examined under oil immersion and photographed at 800 and 1250 powers. Hematocrits were measured as the vol- ume percent of packed red cells to the total blood column. (The term "hematocrit" is used in this paper, although Widmark (1970) has suggested the term be replaced with "packed cell volume") . I classify herring erjrthrocytes according to the stage of development in the peripheral blood as erj^hroblasts, early polychromatics, middle polychromatics, late polychromatics or mature cells, depending upon their size and the amount of polychromasia present. These stages are de- scribed in Table 1. Reticulocytes cannot be iden- tified readily without vital staining so are not included in Table 1. There are variations in individual herring in the size and shape between and within cell stages and the amount of poly- chromasia present is the best indicator as to the series to which the cell belongs. RESULTS The sample source, date of sampling, inci- dence of inclusion bodies, mean length, standard deviation and range in lengths, mean weight, Table 1. — The developmental stages and the average size of erythrocytes in the peripheral blood of wild herring. Stage Description Cell measurements! (microns) Cytosome Erythroblast Early polychromatic Middle polychromatic Late polychromatic Mature erythrocyte Nucleus Round, slightly forger cell than the early polychromatic. Has 7.8 X 7.3 5.9 X 6.2 a dork blue staining cytoplastn with lightly stained spaces. The round purple-red staining nucleus takes up most of the cell. Erythroblosts ore scarce in normal samples. The smallest immature red cell that is normally seen in any 7.8X7.1 4.6X3.3 quantity. Has a light blue to gray staining cytoplasm and appears round. The nucleus takes up most of the cell. Round to slightly oval cell with a gray to light gray-orange 9.5 X 7.0 4.8 X 3.0 staining cytoplasm. Cell is larger than the early polychro- matic. Slightly ovol, has a larger cytoplasm and a smaller nucleus 10.0 X 7.7 4.6 X 2.9 than the middle polychromatic. The cytoplasm appears light orange-yellow. Oval, has a slig'htly larger cytoplasm and a slightly smaller 10.3 X 7.7 4.2 X 2.8 nucleus than the late fjolychromotic. The cytoplosm appears orange-yellow to reddish. Late polychromatic and mature cells have essentially the some appearance with Wright's stain. ! Measurements based on 25 cells in each stage from a normal wild herring in March. '♦'I 126 SHERBURNE: ERYTHROCYTE DEGENERATION IN HERRING Table 2. — The occurrence of inclusion bodies in the cytoplasm of herring erythrocytes, 25 February-30 October 1969. Sample source and cafegoryi Date Incidence in sample Percent Incidence Water Mean length, SD, temp. and range of sample CC) (cm) Mean weight, SD and range of sample (g) Long-term captives Wild, Sheepscot River, Boothbay Hartxir Long-term captives Long-term captives Wild, Eostport, Maine Long-term coiptives Wild, Spruce Point, Boothbay Harbor Wild, Deer Island, N.B., Canada Short-term captives Short-term captives Short-term captives Short term captives 25 Feb. 0/25 13 Mar. 1/35 24 Mar. 0/25 21 Apr. 0/20 10 June 0/40 23 June 2/12 8 July 5/76 16 July 2/50 22 July 24/25 21 Aug. 3/25 25 Aug. 0/10 30 Oct. 0/12 0.0 2.9 0.0 0.0 0.0 16.7 6.6 4.0 96.0 12.0 0.0 0.0 1.3 2.0 3.3 4.9 7.7 15.2 13.8 9.8 16.0 14.0 15.5 9.2 15.3 ±0.79(14.0-17.2) 16.4 ±2.0 (13.5-19.0) 16.0 ±0.99(13.2-17.5) 16.3 ±0.99(14.3-17.7) 22.2 ±3.0 (14.5-30.4) 16.2 ± 1.4 (13.1-18.0) 15.5 ± 1.4 (12.5-18.5) 21.0 ± 1.9 16.0 ± 1.2 16.1 ± 1.3 17.7 ± 1.2 16.2 ± 1.1 (13.9-25.2) (14.4-18.0) (13.2-18.6) (15.3-20X)) (15.1-19J3) 20.2 ± 4.3(14.1- 29.0) 26.2 ± 9.7(14.0- 42.0) 20.7 ± 4.5C10.6- 29.4) 23.0 ± 5.0C14.9- 33.1) 80.3 ±41.3(18.0-214.5) 22.8 ± 5.8(14.1- 34.7) 23.7 ± 6.6(12.8- 43.4) 81.0 ±21.1(13.6-133.0) 25.3 ± 5.7(17.6- 36.4) 22.1 ± 6.2(11.7- 40.9) 29.9 ± 7.4(18.5- 48.1) 20.6 ± 4.2(16.3- 27.6) * Long-term captives— Boothbay Harbor herring held 6 months before Short-term captives— herring from wild 8 July sample held 2 weeks and standard deviation and range in weights of all herring included in this study are given in Table 2. DESCRIPTION OF INCLUSION BODIES The inclusions are round, granular, intracyto- plasmic and appear acidophilic with Wright's stain. The inclusions generally occur singly in the affected cells and vary in size with the largest inclusions usually in the youngest cells. A few red cells contained two inclusions. The bodies characteristically range in size from 2.3 to 3.3 /it in early polychromatics, 1.7 to 1.9 /a in middle polychromatics, and 1.3 to 1.6 jx in late poly- chromatics and mature erjrthrocytes. The in- clusions vary from bright red to reddish-purple in contrast with the blue-gray cytoplasm of the young cells and the dull orange-yellow cytoplasm of the mature cells. Many inclusions have a dark-purple periphery with a light central zone; other inclusions are the same color throughout. Some of the larger inclusions appear to have at least four small, dense-staining particles within or along the periphery of the inclusion. Inclusions were not found outside the red cells, nor were inclusions observed in any white cells of the 355 herring examined in this study. MORPHOLOGY Wild herring that did not contain inclusions ranged from 3 to 35% with an average of 20% being bled, before being bled. immature erythrocytes, while captive herring without inclusions ranged from 2 to 25% with an average of 14% immature erythrocytes in their peripheral blood. Two types of morphology usually character- ized the blood of herring that contained inclu- sions: either upward to 90% immature red cells or a low of 1 to 5% immature red cells. The single herring with inclusions in March had the highest percentage of immature erythrocytes I had found in wild herring to that date. Eighty percent of the red cells were immature, with 12% of the im- mature and 90 % of the mature cells affected with inclusions. Erythroblasts, rare in a normal blood sample, were abundant on this slide. The inclu- sions occurred singly in the cytoplasm and varied in size; the largest were in the youngest cells. The bodies ranged in size from 2.3 to 3.1 /i in early polychromatics, 1.7 to 1.9 jx in middle poly- chromatics, and 1.3 to 1.6 /a in late polychromat- ics and mature erythrocytes. The nucleus of the affected cells exhibited vacuolization and pyk- nosis. Abnormally large immature red cells (macrocytes) were evident with atypical cells present in all developmental stages (Figure 1). The remaining 34 herring in the sample had normal red cell morphology (Figure 2). Inclusions first appeared in long-term captive herring in June in 2 out of 12 specimens. These two herring had the lowest hematocrits of the sample. The blood morphology of the two af- fected herring differed. One herring had 60% 127 FISHERY BULLETIN: VOL. 71, NO. 1 « % f- ^iP^ • • Figure 1.— 13 March 1969. Photo- micrograph of wild herring blood showing macrocytosis of the young cells. Early polychromatics are prev- alent. Arrows point to an inclusion in a middle polychromatic and in a mature red cell. # # • ••# • ^ # M •* W ^ MP % ^ EP .# » # ^ Figure 2.— 13 March 1969. Photo- micrograph of normal wild herring blood showing the absence of in- clusions. EP — early polychromatic eryth- rocyte MP — middle polychromatic erythrocyte M — mature erythrocyte N — neutrophil Th — thrombocyte immature red cells with inclusions found in only 6% of the mature red cells; the other affected herring had 12% immature red cells with inclu- sions in 50% of the immature and 20% of the mature cells. Nearly 7% (5/76) of the wild herring sam- pled on 8 July from Boothbay Harbor contained inclusions, and a few cells in several herring contained two inclusions. Four of the five af- fected herring contained over 70% immature red cells, the other 15%. Both abnormally large and small erythrocytes and many disintegrated cells were present. Anisopoikilocytosis (abnor- mal cell sizes and shapes) of all red cell devel- opmental stages was evident. The nuclei of many affected erythrocytes contained two or 128 SHERBURNE: ERYTHROCYTE DEGENERATION IN HERRING three large vacuoles. The affected mature cells were rounded instead of the usual oval (Figure 3) ; a typical rounded mature cell measured 10.1 X 9.4 [x for the cytosome, 3.7 x 3.4 jx for the nucleus, and 1.2 x 1.5 yu, for the inclusion body. Vacuolization of the cjrtoplasm was evident in many red cells. Inclusions were present in some microcytic mature erythrocytes as small as 4 X 4 /i for the cytosome (less than one-half normal size). Inclusions in a few early poly- chromatics were larger than usual. One of the largest inclusions in a young cell was nearly as large as the cell nucleus — the cytosome measured 9.2 X 8.0 /Lt, the nucleus 4.7 X 3,7 ix, and the inclusion 3.9 x 3.6 jx (Figure 4). Otherwise inclusions in the wild herring of March and July were of the same size. A relationship appears to exist in the occur- rence of inclusions and abnormal red cell mor- phology with temperature extremes. The short- term captive herring sampled on 22 July at 16°C, the highest temperature at which samples were taken, were markedly different from all other samples in their morphology and incidence of inclusions. Ninety-six percent (24/25) of the herring had inclusions, and of those over half had inclusions in at least 90% of their red cells. A majority of the smears in this sample showed 5% or less intact immature red cells. Anucleat- ed "balloon" cells were evident in all smears in this sample, some smears had up to 50% of these cells (Figure 5). The balloon cells appear pale red with Wright's stain, are similar in size, and range from 9.4 x 9.4 fx to 10.9 X 10:9 /a. Some of the cells appear to show diffusion of nuclear material into the cytoplasm. The smears with the greatest incidence of inclusions generally had the most balloon cells. The most heavily affected herring from the 8 July sample also showed these cells. In the smear free of in- clusions a few balloon cells were seen, the intact cells appeared normal and 10% immature red cells were present (Figure 6). Such balloon cells are seen in apparently normal blood samples only occasionally and in very low frequency. The short-term captive herring sampled on 21 August at 14°C showed a substantial decrease in inclusions with 12% of the sample affected, but many nonaffected fish had abnormal cells (Figure 7) . Higher than normal seawater tem- peratures of up to 20.5°C (68.9°F) during Au- gust may account for the abnormal cells in her- ring without inclusions. Inclusions were found in 2 of the 50 herring Figure 3.-8 July 1969. Photomi- crograph of wild herring blood show- ing intracytoplasmic inclusions asso- ciated with nuclear degeneration and a ballooning of the red cells. 129 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 4. — 8 July 1969. Photomi- crograph of wild herring blood show- ing one of the largest inclusions seen in this study. The inclusion mea- sures 3.9 X 3.6 II, the cell nucleus 4.7 X 3.7 li, and the cytosome 9.2 X 8.0 M- I • • ii %f • I * #. >t:.^ % # r • # % (ft < Figure 5.-22 July 1969. Photomi- crograph of herring blood from a short-term captive, 2 weeks after placing wild fish from the 8 July sample in the tanks, showing nearly all of the red cells affected with in- clusions, abnormal nuclei, and anu- cleated "balloon" cells. sampled on 16 July from Deer Island^ N.B., Can- ada. One herring had 25% immature red cells with inclusions in less than 1% of the imma- tures; the other affected herring had 90% im- mature red cells with inclusions in 1% of the immature and 90% of the mature red cells. The morphology and size of inclusions were similar to that of the 8 July samples from Boothbay Harbor. The smear with the greatest incidence of inclusions showed approximately 20% bal- loon cells. HEMATOCRITS The hematocrit mean, standard deviation, and range for each sample and hematocrit values of the males and females in each sample are shown in Table 3. The lowest hematocrit for an indi- 130 SHERBURNE: ERYTHROCYTE DEGENERATION IN HERRING r ^ 9 9 i 9 ^ Figure 6. — 22 July 1969. Photomi- crograph of normal red cells from the only herring not affected with inclu- sions from a sample of 25 short-term captives. # % Figure 7. — 21 August 1969. Photo- micrograph of abnormal cells in short-term captive herring. Higher than normal natural seawater tem- peratures of up to 20.5°C (68.9°F) during August may account for the abnormal cells in herring not affected with inclusions. This herring had one of the lowest hematocrits of the sample (21 volumes percent) ; the scarcity of cells on the slide reflects this finding. % ■#* I i vidual herring in this study was 17 volumes per- cent; the highest, 54.5 volumes percent. The lowest mean hematocrit for a sample was 28.7 volumes percent for the long-term captives in March; the highest mean hematocrit was 41.4 volumes percent for a sample of wild herring in July. The i-test analysis revealed no significant differences in hematocrit values between sexes in these immature herring. A consistent decrease is evident in the mean hematocrit values of the wild herring from the time they were placed in captivity on 8 July 131 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — Hematocrits of herring samples and sexes within each sample, 25 February- 30 October 1969. Water temp. (°C) Herring sampled (Number) Hematocrits of samples Mean lematocrits of: Standard Date Range (vol %) Mean (vol %) Males (vol %) Females (vol %) deviation Long-term captives*: 25 Feb. 1.3 23 22.5-38.0 29.7 4.1 24 Mar. 3.3 25 22.5-3<5.0 28.7 3.6 21 Apr. 4.9 20 23.0-42.5 31.2 4.3 23 June 15.2 5 7 22.042.0 22.5-43.5 34.9 36.3 7.8 7.9 12 22.0-43.5 35.7 7.5 Wild, Spruce Point, Boot-hbay Harbor: 8 July 13.8 27 49 27.0-54.5 31.0-49.5 42.2 40.9 6X) 4.5 Ih 27.0^4.5 41.4 5.0 Short-term captives*: 22 July 16.0 13 12 25.0-47.0 34.0-53.0 40.5 39.6 6.1 5.4 25 25.0-53.0 40.1 5.7 21 Aug. 14.0 12 13 21.0-46.5 17.0-52.5 3^.5 35.4 6.9 8.9 25 17.0-52.5 36.0 7.9 25 Aug. 15.5 4 6 23.0-31.5 24.0^9.0 28.6 33.5 3.8 6.2 10 23.0-39.0 31.6 5.7 30 Oct. 9.2 12 23.0^9.0 30.3 5.2 1 Boothbay Harbor herring held 6 months before being bled. 2 Herring from the wild 8 July sample held 2 weeks before being bled. until the final bleeding on 30 October. Seawater temperatures from 30 July to 22 August were higher than normal with the captive herring ex- posed to temperatures of up to 20.5°C (68.9°F). The physiology of the short-term captives was undoubtedly affected as evidenced by the many disintegrated red cells and abnormal cell types seen in the blood of herring not containing in- clusion bodies. The marked variation in cell sizes and shapes, teardrop cells and bizarre forms are rarely seen in normal herring blood. In 1965 I noted a close correlation between hematocrit values in herring and hemoglobin concentrations measured by the cyanmethemo- globin method. I have found no references on hematocrit values of the Atlantic herring, so I include the relations I found between hematocrit values and hemoglobin concentrations here. The herring sampled in 1965 were long-term cap- tive herring 12.7-25.4 cm in length. Hemato- crits were taken as described in the present study. Blood for hemoglobin measurements was obtained from the heart and placed in a small test tube to which a drop of liquid heparin had been added. Hemoglobins were measured as grams per 100 ml. Regression analysis gave a correlation coefficient of 0.9333. The regression line with the confidence limits of Y at the 0.05 level are shown in Figure 8. DISCUSSION Boyar (1962) reported that mature red cells constitute 97-100% of all blood cells in herring blood, and the immature red cells plus white cells made up less than 3% of the total cells in the herring he examined. However, I found an average of 20% immature erythrocytes in the blood of normal wild herring and 14% immature erythrocytes in the blood of normal captive herring. The occurrence of cytoplasmic inclusions had no apparent relationship to sex, length, weight, or hematocrits, nor did herring with inclusions show, on cursory examination, more than the usual parasites observed in samples without in- clusions. The occurrence of inclusions is asso- ciated with other hematological abnormalities in the peripheral blood including upward to 90% immature red cells or a low of 1 to 5% immature 132 SHERBURNE: ERYTHROCYTE DEGENERATION IN HERRING red cells in contrast to the 20% immature red cells normal for wild herring; microcytic eryth- rocytes less than one-half normal size; and varying degrees of anisocytosis and poikilocy- tosis. The affected red cells have some charac- teristics of piscine erythrocytic necrosis (PEN) as described by Laird and Bullock (1969), in a cod, Gddus morhua, from Passamaquoddy Bay. These authors associated the PEN in cod with viruslike particles. Walker (1971; pers. comm., July 1972) has confirmed the viral nature of PEN in cod by electron microscopy. He also confirmed the correlation of nuclear lesions as described by Laird and Bullock with the pre- sence of cytoplasmic viroplasm and virions. Al- though I believe the inclusion bodies in herring can be explained as a physiological response to environmental stress, the possibility of their viral nature has not been ruled out and requires further investigation. 5 10 15 20 25 30 35 40 45 50 55 60 HEMATOCRIT, VOLUMES PERCENT Figure 8. — Relation of hematocrit values to hemoglobin concentrations in captive herring during late winter, 1965. A relationship appears to exist between inclu- sions in herring erythrocyte? and stress factors, especially temperature extremes. Except in one instance when inclusions occurred in herring from water of 2°C, all herring from Boothbay Harbor (lat 43°50'N, long 69°40'W) having in- clusions were taken from seawater temperatures of 13.8°C or above. At a temperature of 16°C, 96% of a sample of herring were affected with inclusions. Inclusions were found in 2 of 90 herring sampled from the Passamaquoddy Bay area (lat 45°00'N, long 67°00'W). These her- ring were taken from a seawater temperature of 9.8°C, the highest temperature sampled in that area. During the months of June and July water temperatures in the Passamaquoddy Bay area have, over a number of years, averaged approximately 4°C lower than in the Boothbay Harbor area (Colton and Stoddard, 1972). The incidence of inclusions within a popula- tion can change rapidly, apparently with chang- ing environmental conditions, and they are ca- pable of affecting a high percentage of herring within a population in a very short time. As an example, the wild herring on 8 July from Booth- bay Harbor had a 6.6% incidence of inclusions (5/76); however, 2 .weeks after herring from this population were placed in the laboratory tanks, 96% of the herring sampled (24/25) were affected with inclusions, and over 90% of the red cells in individual herring contained these bodies. These bodies, associated with erythrocytic de- generation characterized by necrotic nuclei, a ballooning degeneration of the red cells and the appearance of unusual cells in the blood, may be indicative of stress situations for immature her- ring in the wild. If the stress factors causing these inclusion bodies affect enough herring, they could conceivably have an adverse affect on the population structure endemic to certain areas. The erythrocytic degeneration found in herring may be due to a viral infection as de- scribed in other fishes by Laird and Bullock (1969) and confirmed by Walker (1971). The occurrence of such a viral infection in epidemic frequency would certainly be no less important to our understanding of fluctuations in abun- dance of herring populations. 133 FISHERY BULLETIN: VOL. 71, NO. 1 ACKNOWLEDGMENTS I wish to express my appreciation to George J. Ridgway and John E. Watson of the Northeast Fisheries Center, Boothbay Harbor Laboratory, National Marine Fisheries Service and to Roland Walker of the Rensselaer Polytechnic Institute who critically reviewed the manuscript and made suggestions to improve clarity of presentation. I thank Gareth W. Coffin of the Boothbay Harbor Laboratory for his excellent photomicrographic work. LITERATURE CITED BOYAR, H. C. 1962. Blood cell types and differential cell counts in Atlantic herring, Clupea harengus harengus. Copeia 1962:463-465. CoLTON, J. B., Jr., and R. R. Stoddard. 1972. Average monthly sea water temperatures, Nova Scotia to Long Island, 1940-1959. Ser. Atlas Mar. Environ., Am. Geogr. Soc. Folio 22. Gardner, G. R., and P. P. Yevich. 1969. Studies on the blood morphology of three estuarine cyprlnodontiform fishes. J. Fish. Res. Board Can. 26:4.33-447. Laird, M., and W. L. Bullock. 1969. Marine fish haematozoa from New Bruns- wick and New England. J. Fish. Res. Board Can. 26:1075-1102, RiDGWAY, G. J. 1956. Some cytological observations on fish eryth- rocytes. Progr. Fish-Cult. 18:67-69. Walker, R. 1971. PEN, a viral lesion of fish erythrocytes. (Abstr.) Am. Zool. 11:707. WiDMARK, R. M. 1970. How reliable are red cell indices? Lab. Med. 1(12) :37. 134 FOOD OF TUNAS AND DOLPHINS (PISCES: SCOMBRIDAE AND CORYPHAENIDAE) WITH EMPHASIS ON THE DISTRIBUTION AND BIOLOGY OF THEIR PREY STOLEPHORUS BUCCANEERI (ENGRAULIDAE) Thomas S. Hida' ABSTRACT The results of examining the stomach contents of skipjack tuna (Katsuwonus pelamis), bigeye tuna (T/iMnnMS ofeesMs), yellowfin tuna (Thunnus albacares) ,ka-wa.ka\va (Euthyn- mis af finis) , common dolphin (Coryphaena hippxtnis) , and the little dolphin (Coryphaena equiselis) caught by live bait pole-and-line fishing and trolling in the equatorial eastern Pacific and around the Samoa Islands are presented. Fishes, crustaceans, and molluscs were found to be important food items. The presence of the anchovy, Stolephorus buccaneeri^ among the stomach contents was of particular interest, and information gained on their distribution, size frequency, fecundity, and food habits is presented. This report is based mainly on observations and stomach sample collections that were made during Charles H. Gilbert cruise 116 to the equa- torial eastern Pacific in October-November 1969 (Hida, 1970a) and cruise 117 to the Samoa Is- lands in February-April 1970 (Hida, 1970b). In this study, Stolephonis hiiccaneeri was first found in the stomach contents of bigeye and skip- jack tunas caught in the equatorial eastern Pa- cific and again in the stomach contents of tunas caught around the Samoa Islands. Since there has been no food study made of tunas and dol- phins from these areas and very few reports on the distribution and biology of S. huccaneeri, it is the intent of this paper to (1) describe the food items of the tunas and dolphins caught in these two geographically distant and environ- mentally diverse — oceanic versus insular — areas, (2) extend the known distributional range of S. hiLccaneeri, (3) report on biological informa- tion obtained from the anchovy specimens. Charles H. Gilbert is a U.S. Department of Com- merce, NOAA research vessel assigned to the Southwest Fisheries Center, Honolulu Labora- ^ Southwest Fisheries Center, National Marine Fish- eries Service, NOAA, Honolulu, HI 96812. tory. National Marine Fisheries Service. Ob- jectives of the cruises were to assess the distri- bution and abundance of surface swimming- tunas, to tag and release skipjack tuna {Katsu- ivonus pelamis) and yellowfin tuna (Thiinnus albacares) for migration and growth studies, and to collect olood samples of these tunas for subpopulation studies. Tunas and dolphins were caught by live bait pole-and-line fishing and by trolling. Threadfin shad, Dorosoma petenense, were transported from Honolulu in baitwells on both cruises and used as chum for the fishing operation. It was the exclusive baitfish used on cruise 116, while on cruise 117 supplementary baitfishes, mostly sardines, Sardinella melaniira and Herklotsichthys punctatus, and a mackerel, Rastrelliger kanagurta, were caught in Pago Pago Harbor and used. Since anchovies were not used as live bait on either cruise, the occur- rence of 5. buccmieeri in the stomachs of the tunas examined indicates that this species is a natural food item in this area. Many studies have been made on the food and feeding habits of tunas in the Pacific. Ronquillo (1953) examined the stomach contents of yel- lowfin tuna, skipjack tuna, kawakawa {Euthyn- nus af finis), and the common dolphin {Cory- phaena hipjnirus) caught in Philippine seas. Manuscript accepted July 1972. FISHERY BULLETIN: VOL. 71, NO. I, 1973. 135 FISHERY BULLETIN: VOL. 71, NO. t He found that juvenile fish, especially the acron- urus larvae of Acanthuridae, were most impor- tant in their diet. Also of importance were members of the fish families Trichiuridae, Scom- bridae, Triacanthidae, Holocentridae, Balistidae, and Monacanthidae, and invertebrates such as squids, larval and juvenile stomatopods, larval crabs and shrimp. Hotta and Ogawa (1955) ex- amined the stomach contents of skipjack tuna caught to the east and south of the main Jap- anese islands and reported that Scombridae, Engraulidae, Exocoetidae, and Holocentridae were major dietary items. Important inverte- brates included squids, crab larvae, euphausiids, and shrimp. Alverson (1963) examined the stomach contents of skipjack and yellowfin tunas caught in the eastern tropical Pacific. He found euphausiids to be the main food items for skip- jack tuna, followed by Gonostomatidae, Exocoeti- idae, and the "red crab," Pleuroncodes planipes', and for yellowfin tuna, the "red crab," the swim- ming crab (Portunidae), Thunnidae, Ostraci- dae, Exocoetidae, and Tetraodontidae. Wald- ron and King (1963) studied the stomach contents of skipjack tuna taken around the Ha- waiian, Line, and Phoenix Islands and found that common dietary items were Gempylidae, Scom- bridae, Mullidae, Chaetodontidae, and Holocen- tridae. Larval and juvenile skipjack tuna, stomatopod larvae, shrimp, and crab megalops were also important. E. Nakamura (1965), up- on examination of the stomach contents of skip- jack tuna from the Marquesas and Tuamotu Islands, reported that scombrids, with skipjack tuna constituting a high percentage, were com- mon food items. Serranidae, Lutjanidae, and Gempylidae were of importance as were stomato- pods, crab megalops, and squids. It was con- cluded by Hotta and Ogawa (1955) that the tunas were nonselective in their feeding habits and ate whatever was available in the area. Although these previous observations covered broad areas of the Pacific, no mention was made of any anchovy that may have been S. buccaneeri occurring in the stomach contents. An exception is a report by H. Nakamura (1936) on the food of yellowfin tuna caught in the Celebes Sea that mentioned an anchovy as one of the common food items. The area in which he found tunas con- taining the anchovy, the frequency of occurrence of the anchovy in tuna stomachs, and the num- bers in which it occurred lead me to believe that it may have been S. buccaneeri. METHODS The stomachs of the troll and pole-and-line caught fish were removed after they were mea- sured and sexed. Stomachs that appeared empty and those of most male tunas were examined in the field and their contents recorded. The rest were placed in muslin bags and preserved in 10% Formalin.' One of the objectives of the cruises was to collect 50 skipjack tuna and/or 50 yellowfin tuna blood samples from each school. Therefore, there were four occasions on which 50 stomach samples per school were collected. In the laboratory, counts were made of the organisms in the stomachs whenever possible. Many of the partially digested fishes were iden- tified by their vertebrae which were prepared by teasing away the muscles when necessary and staining with alizarin red. Skipjack tuna re- mains were identifiable by skeletons. Enough of the external characters of the anchovy usually remained for identification. Most of the other fishes were identifiable only to family. The stomach contents of the anchovy, which included many crustaceans, were identified by staining the organisms with methylene blue. Many of the copepods were identified to species but other invertebrates were identifiable only to major groups such as the Chaetognatha, Amphipoda, and shrimp, STOMACH CONTENTS EQUATORIAL EASTERN PACIFIC The results of examining 268 skipjack tuna, 44 bigeye tuna (Thunnus obesus) , 45 yellowfin tuna, 2 common dolphin, and 7 little dolphin (Coryphaena equiselis) caught on cruise 116 of the Charles H. Gilbert are presented in Table 1. The presence of S. buccaneeri in the stomach ' Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. 136 HIDA: FOOD OF TUNAS AND DOLPHINS Table 1. — Frequency occurrence of organisms in the stomachs of 268 skipjack tuna, 44 bigeye tuna, 45 yellowfin tuna, and 9 dolphin (2 com- mon and 7 little) examined from cruise 116 of the Charles H. Gilbert. Predators Food items Sk ipjack B igeye Ye llowfin tuna tuna tuna No. % No. % No. % No. % Fislies: Alepisauridae 1 0.7 __ __ __ Bromidoe 3 I.l 1 2.3 1 2.2 __ Chaetodontidae 1 2.3 Diodontidae 1 0.4 Engraulidae: Stolephorus buccaneeri 35 13.1 20 45.4 __ __ Exocoetidae 3 1.1 1 2.2 3 33.3 Gempylidae 4 1.5 4 9.1 4 8.9 Nomeidae 1 2.3 1 2.2 __ Scombridae: Auxis rochei 1 0.4 Katsuwonus pflamis 5 1.9 „_ Sternoptychidae 1 2.3 _, Zeidoe 1 2.3 __ __ Unidentified 9 3.4 1 2.3 7 15.6 2 22.2 Chum 100 37.3 20 45.4 22 48.9 — — Crustacea: Amphipoda 2 0.7 2 4.4 __ Euphousiacea 1 0.4 __ Shrimp 2 0.7 1 2.3 — — — — Mollusca: Argonauta 6 2.2 __ 1 2.2 Heteropoda 1 0.4 _. 3 6.7 _^ Squids 22 8.2 6 13.6 8 17.8 2 22.2 Chaetognatha 1 0.4 -- — — — — — Stomach empty 136 50.7 15 34.1 17 37.8 5 55.6 contents of 13.1 9f of the skipjack tuna and 45.4% of the bigeye tuna examined was of par- ticular interest. Of the invertebrates, squids were most frequently found in the contents. Many of the stomachs examined were empty. This study revealed that only a few varieties of organisms were eaten in the oceanic environ- ment, which contrasted markedly with Ronquil- lo's (1953) work showing a great diversity of organisms eaten in an environment influenced by land. The fact that 5. buccaneeri was found only in the stomachs of tunas from two schools that were close to each other suggests that it was not widespread in this area. SAMOA ISLANDS Table 2 shows the results of examining 205 skipjack tuna, 23 kawakawa, 24 yellowfin tuna, and 1 common dolphin which were caught on cruise 117 of the Charles H. Gilbert. S. bucca- neeri occurred very frequently in the stomachs examined. Other fishes occurring frequently belonged to the families Acanthuridae and Holo- centridae. Stomatopod larvae, of the inverte- brates, occurred most frequently in the contents. Many of the stomachs examined were empty. The variety of organisms eaten around the Samoa Islands was limited. However, a com- parison of the studies shows a greater diversity ingested around Samoa than in the equatorial eastern Pacific, probably because of the proxim- ity to the islands. The distribution of S. bucca^ neeri was found to be widespread in this area. Their frequency of occurrence in the stomachs suggested that they were an important forage for the tunas and dolphins here. 137 FISHERY BULLETIN: VOL. 71, NO. I Table 2. — Frequency occurrence of organisms in the stomachs of 205 skipjack tuna, 23 kawakawa, 24 yellowfin tuna and 1 common dolphin, examined from cruise 117 of the Charles H. Gilbert. Predators Food items Skipjack Bigeye Yellowfin Common tuno tuna tuna dolphin No. % No. % No. % No. % Fishes: Aconthuridae 45 22.0 1 4.3 4 16.7 Balistidae 13 6.3 2 8.7 1 4.2 Bramidae -- — — — ' ^-^ Corongidae 3 1.5 Chaetodontidae 1 1 5.4 — — 1 4.2 Dactylopteridae 1 0.5 Engraulidae: StoUphorus buccaneeri 38 18.5 4 17.4 6 25.0 1 100 Exocoetidae 5 2.4 Gempylidaa 13 6.3 Holocentridae 61 29.8 2 8.7 2 8.3 Molidae 1 0.5 Monacanthidae 3 1.5 I 4.3 — — 1 100 Mullidae 2 1.0 Ostraciidae 1 0.5 Pomacentridae — — 1 4.2 Scombridae: Katsuwonus petamis 19 9.3 Unidentified 8 3.9 Siganidaa 8 3.9 Synodontidaa (?) 6 2.9 Tetraodontidae 2 1.0 Chum 66 32.2 Unidentified 36 17.6 2 8.7 2 8.3 1 100 Crustacea: Amphipoda: Phronima sp. I 4.2 Crab megalops 2 1.0 2 8.7 Phyllasoma larvae I 0.5 Shrimp 1 4.3 Stomatopod larvae 7 3.4 3 13.0 1 4.2 Mollusca: Squids 20 9.8 -- — 1 4.2 Stomach empty 64 31.0 14 60.8 10 41.7 NOTES ON STOLEPHORUS BUCCANEERI DISTRIBUTION Strasburg (1960) described S. buccaneeri from Hawaii and proposed the common name, roundhead. His holotype was a specimen taken in a nearshore bait seine haul close to Lehua Island. He also found a few specimens in the stomach contents of kawakawa caught about a mile offshore from Oahu. Matsui (1963) found this species in the bait samples he obtained around the island of Maui. The abundance of 5. buccaneeri in Hawaiian waters is not known. This is largely because the Hawaiian skipjack tuna fishermen use the anchovy, Stolephorus purpureus, as their prin- cipal baitfish. These two fish are almost identical and therefore difficult to distinguish from one another. Anchovies regurgitated on deck and found in the stomach contents of tunas are as- sumed to be their baitfish. At times, however, Hawaiian skipjack tuna fishermen have reported seeing skipjack tuna feeding on what they refer to as "oflfshore nehu" (liberal translation of the Japanese term used), which more than likely is S. buccaneeri. The distribution of S. pui^pureus is inshore while that of the S. buccaneeri seems to be generally off"shore. It is therefore pro- posed that another common name of S. bucca- neeri might be offshore nehu. Besides Hawaii, Whitehead (1967) gave the distribution of the S. buccaneeri as the Red Sea, 138 HIDA: FOOD OF TUNAS AND DOLPHINS Persian Gulf, Comoro Islands, east coast of Afri- ca, Formosa, Hong Kong, Japan, the Philippines, Palau, southern India, and Singapore. He stated that they were very common in Hong Kong, Japan, and Hawaii. Additional notes on the distribution of S. buc- caneeri are given below; occurrences discussed are shown in Figure 1. S. huccaneeri was first noticed on cruise 116 in the stomach contents of 4- to 12-kg bigeye tuna that were caught from a "boiling" school (see Scott, 1969) at lat 4°N and long 119°W. It was found again the next day in the stomach contents of skipjack tuna caught at lat 5°N on long 119°W, about 700 miles from Clipperton Island, the closest land. This occurrence is of interest because this species previously had been recorded only near land masses. On cruise 117, S. huccaneeri was observed to be a common organism eaten by skipjack and yel- lowfin tunas, kawakawa, and dolphin caught around the Samoa Islands. Although it was very often eaten by tunas close to shore, it was neither seen nor caught while baiting in the inshore areas. Similarly, Robert E. K. D. Lee (pers. comm.) has found it eaten by yellowfin tuna and kawakawa caught near shore in the Fiji area but has not observed it during baiting operations in inshore waters. In May of 1971 on cruise 53 of the Toivnsend Crormvell, S. huccaneeri juveniles were collected under a night light while the vessel was anchored in a depth of 25 m on Condor Reef in the Caroline Islands. An estimated 20 kg of S. huccaneeri were caught in a close-to-surface haul made with a modified Cobb pelagic trawl (see Higgins, 1970 for a description of this trawl) 160 miles east of Agrihan Island in the Mariana Islands on cruise 55 of the Cromwell in November 1971. It was present in five other trawl hauls, in the stomach contents of a wahoo, Acanthocyhium solandri, caught northwest of Ponape, and in several skipjack tuna caught by trolling north of Namorik during the same cruise. John Naughton, National Marine Fisheries Service, Honolulu, informed me that several schools of yellowfin and skipjack tunas fished by the Hawaiian fishing vessel Anela around Majuro and Arno Atolls in April 1972 were feed- ing on schools of an anchovy resembling S. huc- caneeri. Wilson' cited that two Palauans trolling be- tween Angaur and Peleliu Islands observed and sampled a school of kawakawa feeding on S. huccaneeri. The occurrence of S. hiiccaneeri as discussed here in the equatorial eastern Pacific, Samoa Islands, Caroline Islands, Mariana Islands, Palau Islands, Marshall Islands, and Fiji in conjunc- tion with previous records shows it to be a wide- spread Indo-Pacific (including eastern Pacific) species. Because it occurs in great abundance locally, such as at Fiji and the Samoa Islands, it is to be expected that details of its occurrence will be more likely noted. SIZE Most of the anchovies found in the stomach contents were in poor condition. The caudal fin and snout of many specimens were so badly digested that their standard lengths could only be estimated. The S. huccaneeri found in the bigeye tuna stomachs ranged from 30 to 57 mm in standard length (SL). Those found eaten by the skipjack tuna ranged from 20 to 58 mm. Those caught on cruise 117 of the Charles H. Gilhert near Samoa ranged from 23 to 78 mm. The samples from Condor Reef measured 15 to 30 mm while those from the trawl hauls caught close to the Mariana Islands ranged from 14 to 70 mm. The small postlarvae were semi- transparent when alive and turned whitish when preserved in Formalin. They were identified by their exposed urohyal plate and posterior ex- tent of their maxilla. The presence of large numbers of postlarvae more than 100 miles from land, and adults as far as 700 miles from land, strongly suggests that this species is capable of completing its life cvcle in an oceanic environment. ^ Wilson P. T. Observations of various tuna bait species and their habitats in the Palau Islands. Un- published manuscript. Marine Resources Division, Trust Territory of the Pacific Islands, Saipan, Marianas 96950. 139 FISHERY BULLETIN: VOL. 71, NO. I 160° ISO" HAWAI I —f> — 140"" 130* A • M 5» E • r>n» ^ C\J • • N > MARIANA ISLANDS • • i«^<» f ^ (fcUAM B PALAU ISLANDS N -7" C^PELELIU I 'TkNGAUR I I 134° 20 -10° 4^. ^^' CAROLINE ISLANDS •:'A '■■■■-J TRUK E 150° SAVAI I ^-^ ,-VypoLu TUTUILA -15° S MANUA IS. • ROSE I SAMOA ISLANDS 170° W F FIJI islands' VANUA LEVU 180° CVITA LEVU -20°- S n 'T°' ^ io°- U 1 V N MARSHALL^ ISLANDS MAJURO^ ^- ■>ARNO • ,— . '-''JALUIT 1 CLIPPERTON I. ■10° N 120° I 110° w Figure 1. — The distribution of Stolephorus buccaneeri in the Pacific covered in this study. 140 HIDA: FOOD OF TUNAS AND DOLPHINS FECUNDITY Ova of 32 specimens of S. buccaneeri obtained from tuna stomach contents were measured: Specimens were 38-55 mm SL. From each sub- sample, diameters of 30 or more of the ova from the most advanced mode were taken. Their dis- tribution ranged from 0.4 to 0.8 mm and peaked at 0.5 mm. The ova were opaque, granulated, and classified as maturing. Since there are no previous estimates of fe- cundity, ova from the most advanced mode from two S. buccaneeri ovaries were counted. This method was based on the assumption that all of the ova in this mode constituted a single spawning. A 44-mm specimen contained 595 ova in her left ovary and 830 in her right, a total of 1,398. A 39-mm individual had 340 ova in her left ovary and 454 in her right, a total of 794. and in much better condition for identification purposes. Hiatt (1951) examined the stomach contents of the nehu, S. pitrpnreus, caught from five major baiting areas in Hawaii and concluded that nehu were selective feeders in that they took the crustacean elements in the plankton. He found important food items to be copepods, ghost shrimp (Lucifer), barnacle nauplii, shrimp lar- vae, and crab larvae. ACKNOWLEDGMENTS I am indebted to Peter Whitehead of the Brit- ish Museum (Natural History) for his identifi- cation and verification of S. buccaneeri samples. Thanks are also due to the crew members and scientific personnel of Charles H. Gilbert who were instrumental in collecting the samples. FOOD STUDY The examination of 58 stomach contents of S. buccaneeri recovered from tuna stomachs showed that crustaceans w^ere important in their diet, as shown in Table 3. Only one stomach was found empty. The stomachs of S. buccaneeri in this study contained primarily calanoid cope- pods and other organisms. The copepods that were abundant in one or more anchovy stomachs from the equatorial eastern Pacific were Can- dacia truncata and Euchaeta marina. The cy- clopoid copepod, Oncaea vemista, was common in one stomach. Copepods found in abundance in one or more anchovy stomachs taken from tunas caught from the Samoa Islands were Can- dacia bispinosa ( ?) C. cahila, C. truncata, Cen- tropages gracilis, Euchaeta marina and Temora discaudata. C. truncata and E. marina were the only two species that were abundant in both areas. Not unexpectedly, the close-to-shore sam- ples from Samoa were represented by more spe- cies than those of the oceanic equatorial eastern Pacific. It should be noted, however, that there were many copepodites and badly digested spe- cimens in the equatorial eastern Pacific samples, while those from the Samoa Islands were larger LITERATURE CITED Alverson, F. G. 1963. The food of yellowfin and skipjack tunas in the Eastern Tropical Pacific Ocean. Bull. Inter- Am. Trop. Tuna Comm. 7:293-396. Hiatt, R. W. 1951. Food and feeding habits of the nehu, Stoleph- orus purpiirens Fowler. Pac. Sci. 5:347-358. HiDA, T. S. 1970a. Surface tuna schools located & fished in equatorial eastern Pacific. Commer. Fish. Rev. 32(4) :34-37. 1970b. Surface tuna-school fishing & baiting around Samoa Islands. Commer. Fish. Rev. 32(12) :37- 41. HiGGINS, B. E. 1970. Juvenile tunas collected by midwater trawl- ing in Hawaiian waters, July-September 1967. Trans. Am. Fish. Soc. 99:60-69. HOTTA, H., AND T. OGAWA. 1955. On the stomach contents of the skipjack, Katsuwomis pelamis. Bull. Tohoku Reg. Fish. Res. Lab. 4:62-82. Matsui, T. 1963. Population of anchovy baitfish (Stolephorus) in the vicinity of Maui, Hawaiian Islands. M.S. Thesis, Univ. Hawaii, Honolulu, 98 p. Nakamura, E. L. 1965. Food and feeding habits of skipjack tuna (Katsuu'07ius pelamis) from the Marquesas and Tuamotu Islands. Trans. Am. Fish. Soc. 94:236- 242. 141 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — The stomach contents of Stolephonis biiccaneeri found in tuna stomachs in the equatorial eastern Pacific and the Samoa Islands (A = abundant, C = common, P = present). [The numbers ex- amined are in parentheses.] Samoa Islands Equatorial eastern Pacific Food items Skipjack tuna (10) Yellowfin |^^ tuna (6) vvakawa (4) Bigeye tuna (23) Skipjack tuna (15) Copepodo: Calanoids: Candacia bispinosa (?) A Candacia catula A P — Candacia simplex P _^ Candacia truncata A A Candacia sp. __ _.. P — Centropages gracilis A — _- — — Centropages sp. — — P — — Eucalanus sp. — — P — — Euchaeta concinna P __ Euchaeta marina A __ „_ A Euchaeta sp. P P P Lucicutia flavicornis P — — — Nannocalanus minor (?) P — — Pleuromamma xiphias P Scolecithricella ctenopus P __ Scolecithrix danae (?) C __ Temora discaudata A P .. -. Temora sp. _« __ P — Undinula darwini P __ __ Unidentified calanoids P P P c A Cy<;lopoids: Copilia mirabilis P — — ~ — Copitia sp. .__ P — Corycaeus limbatus (?) __ P __ Corycaeus spesiosus P — . p — Corycaeus vitreus (?) P .— — . — — Corycaeus sp. P P p P Farranula concinna (?) P ^_ ^_ Farranula gibbula (?) P P __ __ Farranula sp. __ P P p Microsetella rosea P P __ __ Microsetella norvegica (?) -_ P — Oncaea conifera p Oncaea venusta -- _„ c Oncaea sp. P P __ p P Sapphirina gastrica (?) P __ __ .. Sapphirina sp. P «_ P p Unidentified cyclopoids P P A C Amphiipoda P __ Mysidacea -- P — Shrimp juvenile C P P P P Crab megolops P — Chaetognattia A P P P P Gastropod larvae P P Heteropodo: Atlanta inclinata P __ __ „_ Atlanta sp. P — — Bivalve larvae P __ — Ostracoda P P Polychaeta P — Pteropodo: Creseis virgula — _« P Unidentified fish P ~ P ~ — 142 hida: food of tunas and dolphins Nakamura, H. Strasburg, D. W. 1936. The food habits of yellowfin tuna Neo<;iz«2- I960. A new Hawaiian engraulid fish. Pac. Sci. 14: nus macropteriis (Schlegel) from the Celebes Sea. 395-399. Trans. Nat. Hist. Soc. Formosa 26(148) :l-8. (English transl. in U.S. Fish Wildl. Serv., Spec. Waldron, K. D., and J. E. King. Sci. Rep. Fish. 23, 8 p., 1950). 1963. Food of skipjack in the central Pacific. FAO RONQUILLO, I. A. (Food Agric. Organ. U.N.) Fish Rep. 6:1431-1457. 1953. Food habits of tunas and dolphins based up- on the examination of their stomach contents. Tf "i^"^' , , . , ^ ^ ^ ,^ , T11.-1- T -m- u o/i\ r7i oo 1967. The clupeoid fishes of Malaya. A synopsis Philipp. J. Fish. 2(l):71-83. ^ ^ j t- SroTT T M with keys to all Indo-Pacific genera. J. Mar. Biol. 1969. Tuna schooling terminology. Calif. Fish Assoc. India 9(2) :223-280. Game 55:136-140. 143 HARVEST AND REGROWTH OF TURTLE GRASS (THALASSIA TESTUDINUM) IN TAMPA BAY, FLORIDA' John L. Taylor,^ Carl H. Saloman,' and Kenneth W. Prest, Jr.' ABSTRACT A comparison of leaf growth and new leaf production in plots of cut and uncut turtle grass, Thalassia testudinum, indicated that plants suffered no damage when harvested twice during a 6-month growing season in Boca Ciega Bay (Tampa Bay), Fla. In deeper or warmer waters where the growing season is protracted, three or more cuttings per year may prove practical. One of the environmental catastrophes to occur in the past 30 years is the destruction of vast beds of turtle grass through dredge-fill opera- tions, other types of coastal engineering, and pollution in its many forms (McNulty, 1961; Taylor and Saloman, 1968; McNulty, Lindall, and Sykes, in press). The most recent devel- opment that may affect turtle grass is the pos- sibility of its harvest for use as a food supple- ment for livestock. Interest in the nutrient content of turtle grass was first stimulated by Burkholder, Burkholder, and Rivero ( 1959) , who showed that turtle grass leaves contain about 13% protein. Their anal- ysis was substantiated by Bauersfeld et al. (1969), who further found that turtle grass in pellet form significantly increased the weight gain and feed utilization of experimental sheep over that of control animals when added to nor- mal rations as a replacement for alfalfa at a level of about 10%. One of the many questions raised by the success of these feeding trials is whether or not beds of turtle grass can survive and regrow after harvest. This report presents ^ Contribution No. 76, Gulf Coastal Fisheries Center, St. Petersburg Beach Laboratory, National Marine Fish- eries Service. ' Gulf Coastal Fisheries Center, National Marine Fish- eries Service, NOAA, St. Petersburg Beach, FL 33706; present address: 1307 Pass-A-Grille Way, St. Peters- burg Beach, FL 33706. * Gulf Coastal Fisheries Center, National Marine Fish- eries Service, NOAA, 75 33d Avenue, St. Petersburg Beach, FL 33706. Manuscript accepted June 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. results of a study in which leaves in an exper- imental plot of turtle grass were repeatedly cut, measured, and compared with those taken from a control area between August 1968 and Novem- ber 1969. In the Gulf of Mexico and Caribbean Sea, the dominant sea grass is turtle grass, Thalassia testudinum Koenig and Sims. Generally, it flourishes in estuaries and coastal waters from the level of low water to depths of 10 m or more depending on water clarity. Throughout its range in the central, western Atlantic, turtle grass meadows attain maximum development in muddy sands where average salinity is between 25 and 39^/f (Phillips, 1960; Hartog, 1970). Morphological features of turtle grass have been reported by Tomlinson and Vargo (1966) and Tomlinson (1969a, b), who showed that new grass beds are established from seeds, which mature during spring and summer months, or vegetatively by rhizome fragments that are broken off and relocated by storm action and currents. Kelly, Fuss, and Hall (1971) demon- strated that the normally slow and uncertain spread of turtle grass can be accelerated by transplanting and securing sprigs treated with naphthelene acetic acid. This procedure may prove useful in establishing and replacing turtle grass in unvegetated areas — especially through the northern part of its range where apparently there is little or no seed production (Phillips, 1960). 145 FISHERY BULLETIN: VOL. 71, NO. I Ecologists have shown that the turtle grass community exhibits great biological diversity and forms the basis of an extremely stable and productive ecosystem (Margelef, 1962; Odum, 1967). Its roots and rhizomes penetrate the bottom down to 25 cm or more in a matlike network that effectively binds and holds sedi- ments and detritus against erosion, and provides a unique habitat for many benthic invertebrates (Bernatowicz, 1952; Voss and Voss, 1955 Ginsburg and Lowenstam, 1958; Phillips, 1960 Strawn, 1961; Thomas, Moore, and Work, 1961 O'Gower and Wacasey, 1967; Hartog, 1970). The broad, elongate leaves of turtle grass have a surface area of about 18 m^ for each square meter of sediment they occupy, and usually rep- resent a standing crop in excess of 1 metric ton (dry weight) per acre (Phillips, 1960; Gessner, 1971 ) . Furthermore, leaves of turtle grass mod- erate water movements, offer attachment sites for various algae and sessile invertebrates, and serve as a feeding ground, shelter, and nesting area for many fishes and motile invertebrates (Humm, 1964; Stephens, 1966). The rich mi- crobial biota that reduces and recycles much of the organic production from turtle grass beds has been recently described by Fenchel (1970). PROCEDURE Turtle grass leaves were harvested in August and October 1968 and in July and September 1969. The cutting was done within a 30 m^ ex- perimental plot in lower Boca Ciega Bay (Tampa Bay), Fla., where the standing crop of turtle grass on a dry, whole weight basis was 1,198 g/m- (Taylor and Saloman, 1968) . The harvest- ing machine was designed and constructed by personnel at the Fisheries Service laboratory in College Park, Md., and consisted of an adjustable, motor-driven sickle bar mounted on a small, sty- rofoam barge. The cutting head was set about 10 cm above the bay bottom, and the barge was directed by hand as water depth was little more than 1 m at high tide. Between harvests, weekly samples of at least 100 leaves were picked from plants dug by shovel within the experimental plot and from uncut plants that served as controls in the surrounding area. The point of leaf removal was at the leaf node. Leaf length was measured from both sam- ple sets, and as an additional measurement of plant vigor, the number of new shoots per leaf cluster was also recorded from each set. 1968 1969 Figure 1. — Average monthly length of turtle grass leaves from cut and uncut plants, and related water temperatures in Boca Ciega Bay (Tampa Bay), Fla., between August 1968 and November 1969. 146 TAYLOR. SALOMAN. and PREST: REGROWTH OF TURTLE GRASS LEAF GROWTH AND REGROWTH AFTER HARVEST Growth of turtle grass foliage and ultimate leaf length are largely controlled by water tem- perature and depth (Phillips, 1960; Strawn, 1961 ) . In Tampa Bay, turtle grass normally ex- hibits a seasonal growth cycle in which leaves elongate rapidly from April to July and die back to short stubble between October and March. During the period of maximum leaf growth, blades develop at a rate of 5 cm per month or more and reach a total length of about 30 cm (Figure 1). Leaves harvested in the growing season had an equivalent or greater rate of regrowth and reached the height of uncut plants in about 2 months ( Figure 1 ) . Observed growth rates of both cut and uncut leaves were compar- able to figures previously reported from southern Florida by Thomas et al. (1961) and Zieman (1968). Furthermore, harvesting had no ap- parent influence on production of new leaves. For each month, the average number of shoots produced by both cut and uncut plants was nearly the same (Figure 2). Thus, from a comparison of leaf growth and new leaf production among cut and uncut plants, it seems likely that turtle grass in the Tampa Bay 8> i O UNCUT 1 n r, ,... < [ % >^ \ K j\ N \. ^ \ > - "x ^ ^ r^] — ^ Figure 2. — Average monthly number of new shoots per leaf cluster for cut and uncut turtle grass plants sam- pled in Boca Ciega Bay (Tampa Bay), Fla., between August 1968 and November 1969. area can be harvested twice each year without adversely influencing plant vigor. DISCUSSION Our findings show that turtle grass beds can sustain periodic cutting without apparent dam- age at intervals of about 2 months in the growing season. In deeper or warmer waters of the Gulf and Caribbean where turtle grass has a longer growing season, it may be practical to harvest leaves more than twice per year. Inherent, tech- nical problems presented by off"shore harvesting would probably be offset by the fact that turtle grass in deep water generally has longer leaves and greater biomass than plants growing in shal- low areas (Burkholder et al., 1959; Phillips, 1960). Offshore along the west coast of Florida esti- mates show that turtle grass grows over about 4 million acres of the sea floor, and in the Car- ibbean, turtle grass resources are even greater. Thus, the tonnage of turtle grass available for harvest is very large (Bauersfeld et al., 1969). However, from the standpoint of resource man- agement, there are a number of questions that must be resolved before the harvest of turtle grass can be seriously considered by commercial enterprises. Principal queries include: (1) can turtle grass leaves regrow normally after more than two seasons of harvesting; (2) how are other plant and animal members of the turtle grass community influenced by harvesting op- erations; (3) what would be the consequences of removing vast amounts of primary production from the food webs in coastal waters; (4) would removal of foliage cause serious erosion of sed- iments in and around turtle grass beds; and (5) how would harvesting methods alter water clar- ity, and thereby influence populations of phyto- plankton and pelagic fishes, and water recre- ation ? LITERATURE CITED Bauersfeld, P., R. R. Kifer, N. W. Durrant, and J. E. Sykes. 1969. Nutrient content of turtle grass {Thalassia testudinum). Proc. Sixth Int. Seaweed Symp., Madrid, p. 637-645. 147 FISHERY BULLETIN: VOL. 71, NO. 1 Bernatowicz, a. J. 1952. Marine monocotyledonous plants of Bermuda. Bull. Mar. Sci. Gulf Caribb. 2:338-345. BURKHOLDER, P. R., L. M. BURKHOLDER, AND J. A. RiVERO. 1959. Some chemical constituents of turtle grass, Thalassia testudinum. Bull. Torrey Bot. Club 86:88-93. Fenchel, T. 1970. Studies on the decomposition of organic detritus derived from the turtle grass {Thalassia testudinum). Limnol. Oceanogr. 15:14-20. Gessner, F. 1971. The water economy of the sea grass Thalassia testudinum. Mar. Biol. (Berl.) 10:258-260. Ginsburg, R. N., and H. A. Lowenstam. 1958. The influence of marine bottom communities on the depositional environment of sediments. J. Geol. 66:310-318. Hartog, C. D. 1970. The sea grasses of the world. Verh. K. Ned. Akad. Wet., Afd. Naturkd., Tweede Reeks 59(1), 275 p. HUMM, H. J. 1964. Epiphytes of the sea grass, Thalassia testud- inum, in Florida. Bull. Mar. Sci. Gulf Caribb. 14:306-341. Kelly, J. A., Jr., C. M. Fuss, Jr., and J. R. Hall. 1971. The transplanting and survival of turtle grass, Thalassia testudinum,, in Boca Ciega Bay, Florida. Fish. Bull., U.S. 69:273-280. Margalef, R. 1962. Communidades naturales. Publ. Espec. Inst. Biol. Mar., Univ. Puerto Rico, Mayaguez vii + 469 p. McNULTY, J. K. 1961. Ecological effects of sewage pollution in Biscayne Bay, Florida: sediments and the dis- tribution of benthic and fouling micro-organisms. Bull. Mar. Sci. Gulf Caribb. 11:394-447. McNULTY, J. K., W. N. LiNDALL, JR., AND J. E. SYKES. In Press. Cooperative Gulf of Mexico estuarine in- ventory and study: Phase I, area description. Odum, H. T. 1967. Biological circuits and the marine systems of Texas. In T. A. Olson and F. J. Burgess (ed- itors), Pollution and marine ecology, p. 99-157. Interscience Publishers. N.Y. O'Gower, a. K., and J. W. Wacasey. 1967. Animal communities associated with Thalas- sia, Diplanthera, and sand beds in Biscayne Bay I. Analysis of communities in relation to water move- ments. Bull. Mar. Sci. 17:175-210. Phillips, R. C. 1960. Observations on the ecology and distribu- tion of the Florida seagrasses. Fla. State Board Conserv. Mar. Lab., Prof. Pap. Ser. 2, 72 p. Stephens, W. M. 1966. Life in the turtle grass. Sea Front. 12: 264- 275. Strawn, K. 1961. Factors influencing the zonation of sub- merged monocotyledons at Cedar Key, Florida. J. Wildl. Manage. 25:178-189. Taylor, J. L., and C. H. Saloman. 1968. Some effects of hydraulic dredging and coastal development in Boca Ciega Bay, Florida. U.S. Fish Wildl. Serv., Fish. Bull. 67:213-241. Thomas, L. P., D. R. Moore, and R. C. Work. 1961. Effects of Hurricane Donna on the turtle grass beds of Biscayne Bay, Florida. Bull. Mar. Sci. Gulf Caribb. 11:191-197. Tomlinson, p. B. 1969a. On the morphology and anatomy of turtle grass, Thalassia testudinum (Hydrocharitaceae). II. Anatomy and development of the root in re- lation to function. Bull. Mar. Sci. 19:57-71. 1969b. On the morphology and anatomy of turtle grass, Thalassia testudinum (Hydrocharitaceae). III. Floral morphology and anatomy. Bull. Mar. Sci. 19:286-305. Tomlinson, P. B., and G. A. Vargo. 1966. On the morphology and anatomy of turtle grass, Thalassia testudinum (Hydrocharitaceae). I. Vegetative morphology. Bull. Mar. Sci. 16:748- 761. Voss, G. L., and N. a. Voss. 1955. An ecological survey of Soldier Key, Bis- cayne Bay, Florida. Bull. Mar. Sci. Gulf Caribb. 5:203-229. Zieman, j. C. 1968. A study of the growth and decomposition of the seagrass Thalassia testudinum. M.S. Thesis. Univ. Miami, Coral Gables, Fla., 50 p. 148 THE INFLUENCE OF TEMPERATURE AND SALINITY ON THE TOXICITY OF CADMIUM TO THE FIDDLER CRAB, UCA PUGILATOR James O'Hara^ ABSTRACT The concentrations of cadmium lethal to the fiddler crab, Uca pugilator, were determined for various environmental regimes of temperature and salinity. Mortality was greatest in high temperatures and low salinities when tested for 240 hr. Concentrations of cad- mium were greatest in green gland followed by gill, hepatopancreas, and muscle. The waste discharge of electroplating plants, lead and zinc mines, and chemical plants fre- quently contains toxic cadmium salts which contribute to the widespread environmental pol- lution (McKee and Wolf, 1963), and the impor- tance of this pollutant has been stressed by its relationship with the crippling "itai-itai" disease of Japan (Kobayashi, 1971). The effects of cadmium on aquatic organisms have been in- vestigated for numerous freshwater organisms (Doudoroff and Katz, 1953; Ball, 1967; Mount and Stephan, 1967), and while the cadmium is normally flushed down to the estuarine and ma- rine environments, only Gardner and Yevich (1970), Jackim, Hamlin, and Sonis (1970), and recently Eisler (1971) have examined the ef- fects of cadmium on estuarine forms. Eisler alone has reported the effects of normal varia- tions in salinity and temperature on the toxic effect of cadmium on mummichogs. The present report is part of a program to examine the effects of chronic exposure of cad- mium to fiddler crab, Uca pugilator. This study examines the synergistic role of salinity and thermal stress on the acute toxicity of cadmium to the crabs. ^ Belle W. Baruch Coastal Research Institute, Uni- versity of South Carolina, Columbia, SC 29208. Manuscript accepted April 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. METHODS Fifteen adult male {x = 2.2 g) and 10 adult female (x = 1.5 g) fiddler crabs were placed in 23 X 30 cm plastic boxes along with 250 ml of dilute filtered seawater. The containers were slightly tilted in the incubator so that the crabs could freely select total or partial immersion. No avoidance of the toxic solution was noted. Desired salinities were obtained by the addition of distilled water. The cadmium stock for all experiments was reagent grade CdCl2 • 2-1/2 H2O made up to a stock solution of 1 mg Cd"^"^ per ml water. Aliquots of this stock were added to each test chamber to bring the cadmium con- centration to the desired levels of 1.0, 5.0, 10.0, 25.0, and 35.0 ppm Cd"^"^. All crabs were kept in constant temperature boxes on a 12-hr light- dark photoperiod for the 10-day duration of the experiment. The water was changed every third day to reduce the buildup of metabolic wastes and to keep the concentration of cadmium near the nominal level. Preliminary tests indicated no loss of cadmium from the test medium with- out organisms. Eisler (1971) showed less than 5 /f loss from similar concentrations of cadmium in dilute seawater. Dead organisms were re- moved every 24 hr during the tests. To determine the synergistic effects of envi- ronmental stresses on the toxicity of cadmium, crabs were exposed- to the different cadmium 149 FISHERY BULLETIN: VOL. 71, NO. 1 concentrations in water of 10, 20, and 30%r sa- linity maintained at 10°, 20°, or 30°C. Each experimental group had a control maintained in uncontaminated water, but subjected to the sa- linity and temperature stresses. Cadmium concentrations in the tissues of crabs exposed to lethal concentrations were de- termined by use of radioactive cadmium (^"''Cd) using the following procedure. Fifteen male and 15 female crabs were placed in 300 ml of filtered seawater of 20^f salinity at 30°C. Each of three test chambers received 2.3 fxc '"^Cd and an aliquot of stock soultion to bring the cadmium level to 5, 15, or 25 ppm Cd"^"^. These thermosaline regimes and cadmium concentrations were chosen because the acute toxicity tests show that they cause relatively high mortality rates. Four active animals (two males, two females) were sacrificed from each chamber at 0, 12, 24, 36, 48, and 60 hr. The animals were frozen until dis- section of the tissues could be accomplished. Four tissues were digested and analyzed: he- patopancreas, gill, green gland, and thoracic muscle. Individual crabs were analyzed; since results from males and females showed no mea- surable difference, the results were pooled. Con- centrations of ^"^Cd were determined by liquid scintillation on a Packard Tricarb Model 3320 counter." Since each 2.3 /uc represented 1.5, 4.5, or 7.5 mg of cadmium in the test water, a simple ratio of counts per minute to microgram of cad- mium was determined from spiked samples and used to calculate the amount of cadmium in the tissues. Concentrations of cadmium are ex- pressed as parts per million wet weight of tissue. RESULTS Table 1. — Cadmium concentrations (Cd ''■■'" in ppm) lethal to 50% of test organisms (TLm) at different salinities, times, and temperatures. Salinity Time Temperature ICC 20°C sec %, hr ppm ppm ppm 10 48 _» __ 11.0 96 __ 325 6.8 144 51.0 21.3 4.0 192 28.5 18.0 3j0 240 15.7 11.8 2.9 20 48 __ __ 28j0 96 ^. 46.6 10.4 144 __ 23.0 5.2 192 52.0 16.5 3.7 240 42.0 9.5 3.5 30 48 „ __ 33.3 96 __ 37.0 23.3 144 29.6 7.6 192 _^ 21.0 6.5 240 47.0 17.9 5.7 regime of 30°C and lO^r. The concentration fatal to 50% of the organisms in 240 hr (TLm- 240 hr) was calculated to be 2.9 ppm Cd*"^ (American Public Health Association, 1971). At higher cadmium concentrations, the time re- quired to kill 50% of the crabs was considerably reduced. Table 1 shows the influence of temperature, salinity, and cadmium concentration on the level of toxicant which kills 50% of the crabs in dif- ferent time periods. The effect of temperature was extremely pronounced and TLm values were generally more influenced by temperature changes than by salinity levels within a thermal regime. The influence of salinity on the TLm- 240 hr was most pronounced at 10°C and 10%c and at higher cadmium concentrations in shorter times. The combined role of temperature and salinity on cadmium toxicity indicates that tem- perature is less influential at higher salinities. ACUTE TOXICITY In general, the higher temperatures and lower salinities produced the greatest cadmium tox- icity. The susceptibility of fiddler crabs to cad- mium was most pronounced in the thermosaline ^ Reference to trade names in the publication does not imply endorsement of commercial products by the National Marine Fisheries Service, NOAA. TISSUE ACCUMULATION Gills In the first 12 hr of exposure, gill tissue ac- cumulated cadmium in proportion to the ex- posure concentration (Figure 1). Thus, gill tissue from crabs exposed to 25 ppm Cd"""" con- tained 110 ppm; gill tissue from those exposed 150 O'HARA: TOXICITY OF CADMIUM TO FIDDLER CRABS 150 a. Q. 100 Z S 50 a < hepolopancreas gill 12 24 36 48 60 TIME IN HOURS Figure 1. — Concentration of cadmium in gill and he- patopancreas of crabs in 5, 15 and 25 ppm Cd^"^ at 30°C, 20%.. to 15 ppm Cd^^ contained 59 ppm, while such tissue from those exposed to 5 ppm Cd"^"^ con- tained 18 ppm. Each accumulation in gill tissue was about four times the concentration of cad- mium in the surrounding water. Gill tissues from crabs in 25 ppm Cd^^ did not increase their cadmium concentration apprecia- bly over 110 ppm in 24 hr and exhibited a de- cline in tissue concentration at 36 hr. The large mortality rate at 48 hr prevented reliable sam- ples from being obtained. Gill tissue from crabs exposed to 15 ppm Cd^^ showed an increase in cadmium content between 24 and 48 hr with a maximum accumulation of 109 ppm. The sig- nificance of the value around 110 ppm is unclear, but may represent a maximum tissue burden in terms of equilibrium with the external medium. The cadmium concentration in gill tissues from crabs sacrificed at 60 hr showed a marked re- duction in cadmium content. Considering the large mortality of crabs in this concentration, the lower cadmium content in the tissues prob- ably represents a reduced binding of the metal due to the destruction of tissue. Crabs exposed to 5 ppm Cd^"" continually con- centrated cadmium in their gill tissue with a maximum of 39 ppm after 60 hr. No mortality occurred in this concentration, and only one an- imal died during this period in the acute toxicity tests. Significant mortality occurred only after 96 hr. Hepatopancreas The hepatopancreas from crabs in the diflfer- ent cadmium solutions concentrated cadmium about two times exposure level in 12 hr (25 ppm was concentrated to 50 ppm in tissue, 15 to 32 ppm in tissue, and 5 to 11 ppm in tissue). The hepatopancreas tissue in crabs exposed to the highest concentration was almost completely destroyed after 24 hr and precluded samples from these crabs (Figure 1). The hepatopan- creas was changed from a firm glandular tissue to an amorphous and liquified condition. Crabs exposed to 15 ppm Cd""^ showed an increase in cadmium levels to about 116 ppm in 48 hr, fol- lowed by a rapid decline. This decline might be associated with the destruction of the hepato- pancreas tissue. Crabs exposed to water con- taining 5 ppm showed the same general increase in Cd"""^ concentration that was evident in gill tissue with a maximum of 24 ppm after 60 hr. Green Gland The bioaccumulation was highest in the green gland tissue (Figure 2) with maximum concen- trations of 380 ppm in tissue from crabs exposed to 25 ppm, 171 ppm from crabs in 15 ppm, and 118 ppm from crabs in 5 ppm. These values are 12 to 20 times the exposure concentrations. s 3 a. a. Z 2 i a < 100 u 24 36 TIME IN HOURS Figure 2. — Concentration of cadmium in green gland tissue of crabs in 5, 15 and 25 ppm Cd"'* at 30°C, 20%«.. 151 FISHERY BULLETIN: VOL. 71. NO. 1 At all exposure levels the highest tissue accumu- lation occurred in the first 12 hr. At 24 hr, the concentrations in the green glands had shown a considerable decline and then increased steadily with values remaining over 10 times the ex- posure level. The 48-hr determination of 280 ppm is based on only two samples and needs verification. Muscle Muscle tissue remained almost constant over the entire time of the experiment, and tissue levels remained only slightly above the exposure levels with maximum concentrations of 29.3 ppm in crabs exposed to 25 ppm, 17.3 ppm from crabs in 15 ppm, and 8.9 ppm from crabs exposed to 5 ppm. DISCUSSION Cadmium toxicity is related to both temper- ature and salinity. The acute toxicity data for crabs maintained at different temperatures show a time delay in the onset of the lethal eflfect of cadmium. Whether this delay is due to differ- ences in bioaccumulation rates or to differences in a temperature-dependent metabolic response to the metal remains to be examined. Fiddler crabs are often exposed to temperatures well in excess of 30°C, and higher temperatures would further accentuate the toxic effects of small amounts of cadmium. There is a clear relationship of high suscep- tibility of fiddler crabs to cadmium in a low- salinity water. It has not been determined if this is due to interaction between the metal and the variety of salts in the seawater resulting in a nontoxic precipitate forming in proportion to the salinity (Bryan, 1971) or if the direction of the osmotic gradient in the higher salinities reduces the rate of entry of the metal. The rapid accumulation of cadmium from the surrounding water results in considerable tissue destruction in the first 24 hr. High concentra- tions of cadmium were found in the gills and hepatopancreas of fiddler crabs. Similar results have been reported in Crustacea exposed to zinc and mercury (Bryan, 1966; Vernberg and Vern- berg, 1972) although high metal concentrations in the green glands were not reported for these metals. Gardner and Yevich (1970) reported gill tissue destruction in the mummichog begin- ning after 20 hr exposure to cadmium. The data presented here for Cd"""^ concentrations in fiddler crab gills indicate that 24 hr is the time when the cadmium content in crabs exposed to high Cd^^ concentrations is reduced by tissue destruc- tion. Yager and Harry (1966) showed a de- crease in cadmium concentration in the liver of snails exposed to high concentrations of cadmium but attributed this decline to individual varia- tion rather than to tissue destruction. Mount and Stephan (1967) suggested that there is a threshold concentration of cadmium in the gill tissue of fishes and that death occurs when this concentration is exceeded. This threshold may be around 110 ppm for fiddler crabs. The relationship between cadmium toxicity and temperature and salinity variation illus- trates that physiological stresses, even within the usual ecological range experienced by the ani- mals, lowers the tolerance of organisms to en- vironmental pollutants. ACKNOWLEDGMENTS I wish to express thanks to Mrs. Barbara Caldwell and Mrs. Cary Clark for their techni- cal help and to Dr. Winona B. Vernberg for her support and advice on the preparation of the manuscript. This study was supported by the Belle W. Baruch Coastal Research Institute, Uni- versity of South Carolina. LITERATURE CITED American Public Health Association. 1971. Standard methods for the examination of water and wastewater, including bottom sediments and sludges. 14th ed. Am. Public Health Assoc, N.Y., 874 p. Ball, I. R. 1967. The toxicity of cadmium to rainbow trout (Salmo gairdnerii Richardson). Water Res. 1: 805-806. 152 O'HARA: TOXICITY OF CADMIUM TO FIDDLER CRABS Bryan, G. W. 1966. The metabolism of zinc and ^sZn in crabs, lobsters and freshwater crayfish. Symp. Radio- ecological Concentration Processes, Stockholm, Sweden, p. 1005-1016. Pergamon Press, Oxford. 1971. The effects of heavy metals (other than merc- ury) on marine and estuarine organisms. Proc. R. Soc. Lond., Ser. B 177:389-410. DOUDOROFF, P., AND M. KaTZ. 1953. Critical review of literature on the toxicity of industrial wastes and their components to fish. II. The metals, as salts. Sewage Ind. Wastes 25 : 802-839. ElSLER, R. 1971. Cadmium poisoning in Fundulus heteroclitus (Pisces: Cyprinodontidae) and other marine or- ganisms. J. Fish. Res. Board Can. 28:1225-1234. Gardner, G. R., and P. P. Yevich. 1970. Histological and hematological responses of an estuarine teleost to cadmium. J. Fish. Res. Board Can. 27:2185-2196. Jackim, E., J. M. Hamlin, and S. Sonis. 1970. Effects of metal poisoning on five liver en- zymes in the killifish (Fundulus heteroclitus) . J. Fish. Res. Board Can. 27:383-390. Kobayashi, J. 1971. Relation between "Itai-itai" disease and the pollution of river wai,er by cadmium from a mine. In S. H. Jenkins (editor), Advances in water pollution research, 1970. Vol. 1, Pap. 1-25, 7 p. Pergamon Press, Oxford. McKee, J. E., and H. W. Wolf, editors. 1963. Water quality criteria. 2d ed. Calif. State Water Qual. Control Board, Publ. 3-A, 548 p. Mount, D. I., and C. E. Stephan. 1967. A method for detecting cadmium poisoning in fish. J. Wildl. Manage. 31:168-172. Vernberg, W. B., and J. Vernberg. 1972. The synergistic effects of temperature, salin- ity, and mercury on survival and metabolism of the adult fiddler crab, Uca pugilator. Fish. Bull., U.S. 70:415-420. Yager, C. M., and H. W. Harry. 1966. Uptake of heavy metal ions by Taphius glabratus, a snail host of Schistosoma mansoni. Exp. Parasitol. 19:174-182, 153 FISHES, MACROINVERTEBRATES, AND HYDROLOGICAL CONDITIONS OF UPLAND CANALS IN TAMPA BAY, FLORIDA' William N. Lindall, Jr., John R. Hall, and Carl H. Saloman^ ABSTRACT Faced with statutory restraints that prohibit dredging and filling of estuarine bottoms, coastal developers have turned to alternate methods of providing water front property for homesites. One method, recently used in Tampa Bay, Fla., is the construction of access canals that lead from open water to upland acreage. This paper presents biological and hydrological data from new upland canals together with some comparative data from older upland canals and bayfill canals. In all types of canals, as presently engineered, stratified, stagnant water causes low levels of dissolved oxygen in summer months, resulting in mortality or emigration among resident organisms. Means of alleviating the problems are discussed. Among Florida's 322,000 ha of estuarine habitat less than 2 m deep, about 24,000 ha have been filled by coastal developers (Marshall, 1968). Public indignation over indiscriminant and un- regulated exploitation of these areas has stimu- lated legislative action designed to conserve and protect natural resources in estuarine areas that remain (Linton and Cooper, 1971). Faced with statutory restraints, coastal developers have, in some instances, abandoned plans for further bay filling and now seek alternate ways to create pre- mium homesites that will satisfy ever-increasing public demand for waterfront property. One way is the construction of access canals that lead from open water to upland acreage (Barada and Partington, 1972). This method was recently used in Tampa Bay in northeast St. Petersburg, Fla., to connect a housing development with the estuary. Shortly after draglines removed earth plugs between the excavated canal system and the bay, property owners gave this Laboratory permis- sion to monitor the canals so that ecological con- ditions in the manmade waterways could be doc- umented. This report contains ecological data recorded at canal and control stations during the first 13 months after the waterway system was completed. Conditions within the upland canal are compared with those recorded in bayfill ca- nals of Boca Ciega Bay, Fla., and older upland canals. DESCRIPTION OF AREA The study area, known as Tanglewood Estates, is located at the southern end of Old Tampa Bay on a tract of land that was originally drained by a small tidal inlet of approximately 0.5 ha (Figures 1 and 2) . During development, the in- let was dammed and pumped dry. Canals were dug to a depth of approximately 4 m below mean low water and stabilized by concrete seawalls (Figure 3). Bay water was introduced into the ditches in June 1970 creating a canal system of approximately 1.6 ha. Average tidal range in the canal svstem is about 1 m. ' Contribution No. 78, Gulf Coastal Fisheries Center, St. Petersburg Beach Laboratory, National Marine Fish- eries Service. ^ Gulf Coastal Fisheries Center, National Marine Fish- eries Service. NOAA, 75 33d Avenue, St. Petersburg Beach, FL 33706. Manuscript accepted July 1972. FISHERY BULLETIN: VOL. 71. NO. 1, 1971. PROCEDURES Hydrological (five stations) and biological (four stations) samples were collected monthly 155 FISHERY BULLETIN: VOL. 71, NO. 1 GUL 50 OF MEXICO Figure 1. — Tampa Bay, Fla., showing location of study area. Figure 3. — Study area after alteration (hydrologic sta- tion • ; trawl station < >). within the canals between 0830 and 1130 hr from August 1970 through August 1971. Also, a hy- drological control station, Station 1, was estab- lished in the bayou adjacent to the development site to monitor ambient water conditions in a natural area (Figure 3). Hydrological factors were recorded from surface and bottom water and included water temperature, dissolved oxy- gen, and salinity. Temperature was recorded to the nearest tenth of a degree with a handheld mercury thermometer; dissolved oxygen was de- termined by a modified Winkler method (Strick- land and Parsons, 1968); and salinity was de- termined with a Model 10401 TS Goldberg re- fractometer'. Fish and invertebrates were collected with a 4.8-m otter trawl composed of a 2.5-cm stretched ^ Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Figure 2. — Study area before alteration. 156 LINDALL, HALL, and SALOMAN: CONDITIONS OF UPLAND CANALS mesh body fitted with a 0.6-cm mesh inner liner in the cod end. No suitable control station for trawling was established in the adjacent bayou because of numerous snags and oyster beds. Specimens from the trawl were killed in a 10% Formalin-seawater solution and transferred to 50 "^r isopropyl alcohol for preservation. All spec- imens Avere identified to species and enumerated. RESULTS TEMPERATURE Only small diflferences were recorded between surface and bottom water temperature at canal stations or the control station in any sampling period (Table 1). With few exceptions, bottom temperatures were slightly lower than surface temperatures in all months except July and Au- gust 1971 when the situation was reversed. The greatest difference observed was at Station 4 in February when bottom temperature was 1.8°C lower than temperature at the surface. SALINITY Drought conditions prevailed throughout the Tampa Bay area during most of the study period, and, as a result, salinity rose almost steadily from 2S.2'/i, in August 1970 to greater than 30.0:^, by July 1971 (Table 1). During this period salinity at the control station was similar to that in the canals. Heavy rains in August 1971 reduced salinity values considerably. This was the only time during the study when strat- ification occurred at all stations. Bottom sa- linity at the control station was 9.0//^ higher than Table 1. — Monthly hydrologic measurements of surface and bottom water, August 1970-August 1971. Stn. Depth Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Mean Temperature (°C) 1 S 29.8 27.6 24.8 14.8 17.4 14.9 21.0 19.0 26.1 28.0 29.5 28.5 28.5 23.8 B 27.5 24.7 14.7 16.7 14.7 20.8 19.0 26.4 27.8 29.5 29.0 ^9.4 23.4 2 S 29.5 27.4 24.8 14.8 17.2 16.0 20.8 19j2 25.8 27.8 29.0 29.0 28.4 23 .3 B 27.7 25.2 14.5 16.5 14.8 20.2 18.8 25.5 27.4 29.0 30.5 29.5 23.3 3 S 29.0 28.8 25.8 15.7 17.2 15.6 20.6 19.4 25.3 27.8 29.5 29.5 28.4 24.1 B 28.5 24.7 15.5 17.0 15.6 19.7 18.4 25.1 27.2 29.4 30.1 30.0 23.4 4 S 29.5 28.7 25.5 15.2 17.3 16.1 20.6 18.7 25.5 28.0 29.4 29.1 28.5 24.0 B 28.0 24.5 14.2 16.8 15.2 18.8 18.3 25.6 27.0 29.4 29.8 29.8 23.1 5 S 29.5 27.9 25.8 15.8 17.2 15.3 20.9 19.0 25.5 2i8.0 30.5 29.4 28.8 24.1 B 27.8 24.9 15.3 17.0 15.2 20.7 19.1 25.5 27.0 30.0 30.3 29.8 23.6 6 S 29.5 28.1 25.2 15.5 }7.7 16.0 21.0 19.5 25.9 28.2 30.0 29.1 28.8 24.2 B — 28.0 25.2 15.3 17.3 15.2 20.8 19.6 25.9 27.4 30.0 29.4 29.3 23.6 Salinity {%c) 1 S 24.72 25.67 25.67 27.61 29.00 29.06 26.72 29.78 30.106 29.94 31.39 30.17 13.28 27.16 6 __ 25.56 26.00 27.67 29.11 28.94 26.67 29.83 30.06 29.89 31.28 30.00 22.22 28.10 2 S 23.44 25.00 26.56 28.00 28.78 29.00 27.22 29.72 3)0.00 29.17 30.06 29.94 13.33 26.94 B __ 25.28 26.56 27.94 28.72 29.06 27.94 29.67 29.89 29.33 30.72 30.22 27.89 28.60 3 S 23.22 25.11 26.11 27.94 29.00 29.06 27.39 29.44 29.89 29.17 30.61 29.61 13.61 26.94 B 25.28 26.11 27.94 29.00 28.94 27.78 29.56 29.89 29.83 30.39 30.11 28.61 28.62 4 S 23.39 24.89 26.17 28.00 29.06 28.83 27,72 2 9. '36 30.00 29.28 30.33 29.94 13.89 26.97 B __ 24.56 26.44 27.89 29.06 29.06 27.67 30.06 30.11 29.72 30.83 29.00 28.00 28.62 5 S 23.39 25.11 26.39 28.106 28.94 29.00 27.61 29.44 29.89 29.61 30.72 29.28 li3.89 27.03 B __ 25.44 25.83 28.00 28.94 28.94 27.67 29.72 29.89 29.17 30.72 30.06 26.78 28.43 6 S 25.22 24.67 ■26.39 27.94 29.11 29.06 27.28 29.17 29.94 29.44 30.28 28.67 14.06 27.01 B — 24.56 26.28 27.94 28.89 28.S3 27.50 29.33 29.89 29.83 30.44 29.00 25.72 28.23 D issolved ox ygen (ml/liter) 1 S 3.87 3.06 3.46 5.80 4.67 5.64 4.35 4.75 4.03 3.06 2.90 2.58 2.34 3.86 B __ 3.62 3.38 5.64 4.&3 5.23 4.35 4.67 4.11 2.98 3.06 2.66 2.18 3.89 2 S 4.19 3.87 4.43 5.72 4.43 5.07 4.67 4.91 3.22 4,27 3.14 2.18 3.6G 4.13 B __ 3.78 4.03 5.15 3.95 4.75 2.49 4.19 1.37 3.95 3.14 1.86 0.89 3.30 3 S 4.35 2.98 4.03 5.47 4.99 5.07 4.91 3.95 2.74 3.95 4.35 3.14 4.03 4.15 B 2.66 2.42 5.40 4.51 4.35 1.13 3,10 1.62 2.98 3.46 0.00 0.82 2.70 4 S 4.35 2.26 4.27 5.64 4.75 5.31 4.67 3.78 2.66 4.59 3.62 3.30 4.67 4.14 B -_ 2.02 2.42 4.67 3.70 4.59 0.57 3.46 2.22 2.00 2.50 0.66 0.65 2.A6 5 S 3.62 3.06 4.43 5.40 4.85 5.72 3.95 4.27 2.50 3.87 3.87 3.54 4.99 4.16 B __ 2.90 4.35 5.23 4.75 4.59 2.66 3.14 2.34 2.98 3.81 0.08 0.57 3.12 6 S 4.19 2.90 4.51 5.64 5.15 5.23 4.75 4.67 3.30 3.22 3.71 2.98 4.43 4.12 B — 2.82 4.35 5.55 5.14 4.75 4.59 4.27 2.42 3.14 3.38 2.50 0.17 3.59 157 FISHERY BULLETIN: VOL, 71, NO. 1 the surface. The stratification was even more evident at canal stations where the difference between surface and bottom values ranged be- tween n.6%r (Station 6) and IS.O^r (Station 3). In general, the most landward stations exhibited the greatest differences between surface and bot- tom salinities. OXYGEN Daytime concentrations of dissolved oxygen at the surface and bottom for each station are shown in Table 1. Only at the control station were surface and bottom values similar, varying linity at the control station was 9.0'/(< higher than no more than 0.6 ml/liter at any one sampling time throughout the year. At this station the lowest observed concentration was 2.1 ml/liter (August 1971). Surface oxygen values within the canal system were comparable with those at the control station throughout the year. How- ever, bottom oxygen dropped in the canals in February when less than 2.0 ml/liter was re- corded at Stations 3 and 4. These values rose above 3.0 ml/liter in March, but in April and May dissolved oxygen near 2.0 ml/liter or less was recorded at several canal stations. In July less than 1.0 ml/liter was recorded at Stations 3, 4, and 5, and by August less than 1.0 ml/liter of oxygen was recorded at the bottom at all canal stations. To determine the diel changes in oxygen con- centration during the July sampling period, a 24-hr sampling program was conducted at each station. Results showed that surface and bottom values were similar only at the control station (Figure 4). Surface oxygen concentration in the canals corresponded with values recorded at the control station and never fell below 2.0 ml/ liter. However, at all canal stations the bottom was nearly anoxic throughout the 24-hr sam- pling period. FISHES AND MACROINVERTEBRATES Thirty-six species and 10,497 individuals of vertebrates and invertebrates were collected within the canals during the year (Table 2). Of the 36 species, 32 were finfish (23 of sport or commercial value), 1 was the diamondback N 6 E . O >- X O O 2 m i/i 5 STATION 1 (control) STATION 2 STATION 3 STATION 4 STATION 5 -' STATION 6 T*^ ■^**-p I I I I I \ 1 1 r 0800 1000 1200 1400 1600 1800 2000 2200 2400 0200 0400 0600 HOURS Figure 4. — Results of 24-hr oxygen survey in July 1971 (surface • •; bottom • •)• terrapin, Malaclemys terra'pin, and 3 were com- mercially important invertebrates (blue crab, Callinectes sapidus; pink shrimp, Penaeus duo- rarum; and brief squid, LolUguncula brevis). The four species of fish caught in greatest abundance represented 92 Sr of the total number of specimens. They were the bay anchovy {An- choa mitchilli) , spotfin mojarra {Eucinostomus argenteus) , spot {Leiostomus xanthurus), and silver jenny (Eucinostomus gula). The bay an- chovy alone made up nearly 72 9f of the total. The brief squid was by far the most abundant invertebrate (84% of all invertebrates collected) 158 LINDALL, HALL, and SALOMAN: CONDITIONS OF UPLAND CANALS Table 2. — Monthly occurrence and number of individuals of vertebrates and invertebrates collected with otter trawl at all stations from August 1970 through August 1971. No individual collected in July and August 1971. Species Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June Total Number Percent No. No. No. No. No. No. No. No. No. No. No. Vertebrates: Anchoa mitchilli^ 56 539 1,582 26 93 698 1,376 746 16 2,425 7,557 72.0 Eucinostomus argenteuj^,'' 120 135 241 220 153 16 25 6 5 921 8.8 Leiostomus xanthurus^,' 1 26 74 108 661 1 821 7.8 Eucinostomus guta^,'' 18 82 164 99 I 8 372 3.5 Micropogon itndulatus^ ,^ 1 20 7 34 9 71 0.7 Cynoscion arenarius^,' 3 10 8 20 8 2 51 0.5 MentUirrhus americanus^,^ 3 15 13 1 1 1 4 5 43 0.4 Pogonias cromis^,^ 1 3 7 4 4 4 1 24 0.2 Archosargus probatocephalus'^ ,- 2 10 4 2 1 1 2 1 23 0.2 Lagodon rhomboidfs'^,^ 1 4 6 3 6 1 21 0.2 Microgobius gulosus^ 1 11 2 14 0.1 Cynoscion ncbulosus^,- 3 4 4 1 2 14 0.1 Chaetodipterus faber'' 4 1 2 1 1 4 13 0.1 Arius jilis'^,^ 1 3 5 3 12 0.1 Malaclemys terrapin 2 1 5 4 12 0.1 Bairdiella chrysura^,'^ 1 2 5 I 1 10 0.1 Gobiosoma bosci^ 7 1 1 9 0.1 Prionotus tributus^ 1 1 4 1 7 0.1 Orthopristis chrysoptera',^ 1 2 3 1 7 0.1 Sphoeroides nephelui^ 3 1 1 5 0.1 Opisthonema oglinum^ 4 4 0.0 Dasyatis sabina^ 1 2 3 0.0 Mugil cepkalus^." 1 1 2 0.0 CMoroscombrus chrysurus'^ 2 2 0.0 Anchoa kepsetus'' 2 2 0.0 Achirus lincatus^ 2 2 0.0 Synodus loetens^,^ 1 0.0 Opsanus beta^ 1 0.0 Sympkurus plagiusa- 1 0.0 Sciaenops ocellata^," 1 0.0 Paralichthys albigutta^," 1 0.0 Chasmodes saburrae^ 1 0.0 Gymnura micrura- 1 0.0 Invertebrates: Caltinectes sapidus^,^ 8 2 2 1 5 3 9 12 4 1 47 0.5 Penaeus duorartim'','' 4 2 7 2 4 5 1 1 26 0.3 Lolliguncula brevis^ 45 55 93 43 4 15 6 29 87 20 395 3.8 Total species 1 8 15 15 18 19 18 19 16 9 17 36 Total individuals 56 721 1,811 489 506 417 883 2,153 821 154 2,485 10,497 100.0 1 Of commercial or sport value. " Demersal or bottom feeder. and made up nearly 4% of the total number of animals collected during the year. The first trawl sample was made in August 1970, 2 months after bay water was introduced in the canal system. At that time the only spe- cies of fish found in the canals was the bay an- chovy, and 98% of the specimens were taken at Station 4 (Figure 3; Tables 3, 4, 5, 6) . By Sep- tember, the blue crab, pink shrimp, brief squid, and four additional species of finfish were caught within the canal system. Station 4 contained seven species of vertebrates and invertebrates while Stations 2 and 3 contained five species each. No specimens were yet found at Station 1. In October, 5 months after water was introduced, fishes and invertebrates were found in all canals, and a total of 15 species were collected. Station 4, with 11 species, still contained the greatest faunal diversity. Stations 1, 2, and 3 contained 10, 5, and 4 species, respectively. Through the winter, spring, and early summer months (November through June) an average of 16 species per month was collected throughout the area. The number of species and individuals at each station declined in April and May cor- responding to reduction in dissolved oxygen, but in June the number of species and individuals increased again at all stations. 159 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — Monthly occurrence and number of individuals of vertebrates and invertebrates collected with otter trawl at Station I from August 1970 through August 1971. No individual collected in August and September 1970 and in July and August 1971. Total Species Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June Nunnber Percent No. No. No. No. No. No. No. No. No. Vertebrates: Anchoa mitchilti 118 16 1 77 31 9 412 664 58.9 Eucinoitomus argenteus 49 1 6 51 13 9 5 4 138 12.3 Leiostomus xanthurus 2 3 41 50 96 8.5 Eucinostomus gula 10 3 41 39 93 8.3 Micropogon undulatus 1 21 22 2.0 Mentkirrhus americanus 1 4 1 6 0.5 Archosargus probatocephalus 1 2 1 1 5 0.4 MalacUmys terrapin 2 1 i2 5 0.4 Cynoscion arenarius 1 1 2 4 0.4 Lagodon rhomboides 2 1 1 4 0.4 Opisthonema oglinum 4 4 0.4 Microgobius gulosus 2 2 4 0.4 Chaetodipterus faber 1 1 1 3 0.3 Prionotus tribulus 1 2 3 0.3 Cynoscion nebulosus 2 2 0.2 Pogonias cromis 2 2 0.2 Orthoprislis chrysoptera 1 1 2 0.2 Sphoeroides nephelus 2 2 0.2 Bairdiella chrysura 1 1 0.1 Chloroscombrus chryiurus 1 1 0.1 Gobiosoma bosci 1 1 OJ Invertebrates: Lolliguncula brtvis 5 3 6 1 4 1100 mm Percentage of total catch Commercial Berried females 80 17 3 11 29 60 Because lobsters ranging in carapace length from 81 to 90 mm constituted approximately 80 Sr of the commercial catch and only 11 "^r of the berried fem.ale sample, it is apparent that only a very small percentage of females mature below 90-mm carapace length. Certainly the most marked disparity in size composition was for carapace lengths greater than 100 mm (3Sr commercial and 60 /r berried females). This would seem to validate the previous conclusion that most females above 100-mm carapace length are mature. SUMMARY This study which is concerned with the anal- yses of data collected from 1968 through 1970 on the natural population of American lobsters along the Maine coast has yielded the following information: 1. Sex ratio was 1:1, thus suggesting that differences in growth rate and molting frequency do not exist between mature males and immature females. 2. Molting was concurrent for males and fe- males, with shedding reaching a peak in late summer. 3. The length-weight relationship for sexes combined was log W = —2.9052 + 2.9013 logL. 4. Length-frequency histograms revealed the high rate of exploitation by the commercial fish- ery and an increase in unavailability of lobsters progressively smaller than 70-mm carapace length. 5. Male lobsters begin maturing at relatively small sizes (509^ mature at about 44-mm cara- pace length); however, because native Maine females rarely mature below the minimum legal size of 81-mm carapace length and males must approximate females' size to successfully mate, it is doubtful that prerecruit males contribute reproductively to the natural population. 6. Female maturity was assessed by the fol- lowing methods: 1) classification of ovaries by color and ovum diameter to three stages of development; 2) examination of seminal recep- tacles for spermatophores ; 3) morphometric re- lationship of abdominal width: carapace length ratio to carapace length; and 4) length-fre- quency distribution of native Maine berried fe- males. From estimates by these four independ- ent methods, I concluded that females seldom become sexually mature at a size less than 81-mm carapace length, and then only a small fraction of those females between 81 and 90 mm attain maturity, whereas, at carapace lengths greater than 100 mm, nearly all females are assumed to be mature. Bearing in mind the minimum legal size regulation of 81-mm carapace length, I demonstrated in this study that the majority of females are harvested commercially prior to their first opportunity to spawn. An obvious change in management suggested by the results of this study would be to increase the minimum size limit to insure successful spawning by a sizeable portion of the population. Based on results of this study and those from the com- mercial sampling phase of the Maine lobsters project, Thomas (1971, see footnote 3) deals specifically with minimum size limit increases as a means to achieve a maximum sustainable yield. ACKNOWLEDGMENTS I am indebted to James C. Thomas for his guidance and support throughout the course of this study and his review of the manuscript. Richard Hanley, Gerald Brackett, Robert Nunan, Stephen Ham, and Andrew Dolloff" assisted with field collections and data compilations. Gareth Coffin of the Northeast Fisheries Center, Nation- al Marine Fisheries Service, West Boothbay Harbor, Maine, performed the photographic work. 172 KROUSE: SIZE OF AMERICAN LOBSTERS LITERATURE CITED Cooper, R. A. 1970. Retention of marks and their effects on growth, behavior, and migrations of the Amer- ican lobster, Homarus americanus. Trans. Am. Fish. Soc. 99:409-417. Ennis, G. p. 1972. Growth per moult of tagged lobsters (Ho- manis americanus) in Bonavista Bay, Newfound- land. J. Fish. Res. Board Can. 29:143-148. Harding, J. P. 1949. The use of probability paper for the graph- ical analysis of polymodal frequency distributions. J. Mar. Biol. Assoc. U.K. 28:141-153. Herrick, F. H. 1911. Natural history of the American lobster. Bull. [U.S.] Bur. Fish. 29:149-408. Skud, B. E., and H. C. Perkins. 1969. Size composition, sex ratio, and size at ma- turity of offshore northern lobsters. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 598, 10 p. Squires, H. J. 1970. Lobster {Homarus americanus) fishery and ecology in Port au Port Bay, Newfoundland, 1960- 65, Proc. Natl. Shellfish. Assoc. 60:22-39. Steel, R. G. D., and J. H. Torrie. 1960. Principles and procedures of statistics with special references to the biological sciences. Mc- Graw-Hill, N.Y., 481 p. Templeman, W. 1932. Investigator's summaries. Biol. Board Can., Annu. Rep., p. 38-41. 1934. Mating in the American lobster. Contrib. Can. Biol. Fish., New Ser. 8:421-432. 1935. Local differences in the body proportions of the lobsters, Homarus americanus. Biol. Board Can. J. 1:213-226. 1939. Investigations into the life history of the lobster (Homarus am,ericanus) on the west coast of Newfoundland, 1938. Newfoundland Dep. Nat. Resour., Res. Bull. (Fish.) 7, 52 p. 1944. Abdominal width and sexual maturity of fe- male lobsters on Canadian Atlantic coast. J. Fish. Res. Board Can. 6:281-290. Wilder, D. G. 1953. The growth rate of the American lobster (Homarus americanus) . J. Fish. Res. Board Can. 10:371-412. 1963. Movements, growth and survival of marked and tagged lobsters liberated in Egmont Bay, Prince Edward Island. J. Fish. Res. Board Can. 20:305-318. 173 APPARENT GROWTH OF YELLOWFIN TUNA FROM THE EASTERN ATLANTIC OCEAN J. C. Le Guen'-^ and Gary T. Sakagawa'-' ABSTRACT Apparent growth of yellowfin tuna from the eastern Atlantic Ocean was estimated from modal progression of length-frequency distributions by two methods. One was to use fish of unknown age, which gave estimates of parameters of the von Bertalanffy growth function of L „ ^ 194.8 cm and A' ^ 0.035, on a monthly basis. The other was to u.se fish of apparent known age, which resulted in L^ = 175.2 cm and K = 0.044. Although the parameter estimates were different, estimates of length at ages 1.5-4.5 years were quite similar with both approaches. A comparison of growth estimates of yellowfin tuna was made. Estimates from anal- ysis of length-frequency distributions appeared to be superior to those from analysis of scales because they were based on a larger range of fish sizes. However, observed lengths at ages 1.5-5 years were similar for both types of analysis and for yellowfin tuna from both the Atlantic and Pacific Oceans. It is recommended that observed sizes at age rather than the estimated sizes at age from the von Bertalanff"y function be used in estimating yield per recruitment of yellowfin tuna. There have been several studies (e.g., Le Guen, Baudin-Laurencin, and Champagnat, 1969; Yang, Nose, and Hiyama, 1969) on growth of Atlantic yellowfin tuna (Thunmis albacares) but little agreement among them. The disagreement can be traced to at least three sources: first, the kinds of data, e.g., length-frequency distri- butions and scale readings have been different; second, the method of fitting the von Bertalanffy growth function has varied; and third, the range of fish sizes employed has been different. Be- cause an accurate estimate of growth is impor- tant for estimating yield per recruitment by the Beverton and Holt approach (Schaefer and Beverton, 1963), one method that can provide information for rational management of the re- source, a study was initiated to estimate growth from the best series of data available and, hope- fully, to resolve the disagreement. In this report ^ Alphabetical authorship since the paper is based on independent studies of both authors. - Office de la Recherche Scientifique et Technique Outre-Mer, Centre de Pointe-Noire, P.O. Box 1286, Pointe-Noire, Republic of Congo. ^ Southwest Fisheries Center, National Marine Fish- eries Service, NOAA, La JoUa, CA 92037. Manuscript accepted May 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. the results of that study on apparent growth of yellowfin tuna from modal progression of length- frequency distributions are presented and com- pared to growth estimates derived from pub- lished data and computed by standardized pro- cedures. PLAN OF ANALYSIS Length frequency samples from commercial landings were employed in our study (Table 1). The fish were caught off Africa by baitboats and purse seiners and were sampled by French and Inter-American Tropical Tuna Commission (lATTC) scientists. (Scientists of the lATTC sampled the Atlantic tuna catch of U.S. vessels under a contract from the National Marine Fish- eries Service.) The French scientists sampled the French catches, which were from three gen- eral regions — Abidjan, Ivory Coast; Dakar, Senegal; and Pointe-Noire, Congo — and the lATTC scientists sampled the American catches, which were from primarily the Gulf of Guinea. The lATTC samples were caught in both the Abidjan and Pointe-Noire regions, but because 175 FISHERY BULLETIN: VOL, 71. NO. 1 Table 1. — Sources of length data of Atlantic yellowfin tuna caught in the surface fishery off Africa. Abidjan Dakar Gulf of Guinea Polnte- Noire Year Type of length measurement Type of vessel sampled Bait- boat Small seiner^ Large seiner- Source 1966 Predorsal 1966-69 Predorsal X 1970 Predorsal X 1968 Fork X 1969 Predorsal X 1970 Predorsal X 1968-70 Fork 1965^ Predorsal X 1967-68 Predorsal X 1969 Predorsal X 1970 Predorsal X 1971 Predorsal X X X X X X X X X X X O.R.S.T.O.M., 1971 O.R.S.T.O.M., 1971 X O.R.S.T.O.M., 1971 Champagno-t and Lhomme, 1970 Champagnot and Lhomme, 197C X O.R.S.T.O.M., 1971 X Staff, Tuna Population Dynamics Project, 1971' O.R.S.T.O.M., 1971 Lo Guen et al., 1969 O.R.S.T.O.M., 1971 O.R.S.T.O.M., 1971 Unpublished data (Le Guen) 1- Small purse seiner = less than 500 metric tons capacity. ^ Large purse seiner = larger than 503 metric tons capacity. * Staff, Tuna Population Dynamics Project. 1971. Size composition of the yellowfin and skipjack tuna purse seine fishery off the west coast of Africa 1968-1970. Unpublished manuscript, 28 p. Southwest Fisheries Center, National Marine Fisheries Service, NOAA, La Jolla, CA 92037. they could not be separated as such, they were treated separately from the French data. Two methods were employed in our analysis. One approach ("age unknown") was based on all samples from the four regions, for years 1965- 70 and with age of size groups unknown. The second approach ("apparent age known") was slightly different. Only fish that were caught in an area from Sao Tome to southern Angola, 1967- 71, and with the apparent age of each size group known, were employed. Growth was estimated with the von Berta- lanffy growth function. This function is often expressed as, Lt = L, [1 — exp — K {t - UU, where Lt = length at age t, L^ = asymptotic length, K = growth rate, and to = theoretical ■age when Lt = 0. It is fitted to growth data by various procedures (e.g., Walford, 1946; Abramson, 1963; Ricklefs, 1967; Gulland, 1969; Knight, 1969), most of them require data on size at known age. A least-squares procedure that does not contain this limitation was de- scribed by Fabens (1965) . He fitted a von Bert- alanflfy function of the form Lt + A = Lt + (L„ — Lt) (1 — exp - K) to tag-return data, but his procedure is equally applicable to length observations of untagged fish made at t and again at a later date, t + A, when the age of the fish is unknown. For tuna, Rothschild (1967) and Joseph and Calkins (1969) employed Fabens' procedure to estimate growth of skipjack tuna {Katsuivomis pelamis) from tagging data. We used the Fabens' pro- cedure with monthly mean lengths for individ- ual year classes to estimate growth of yellowfin tuna of unknown age. A computer program written by Tomlinson (Abramson, 1971) was employed to estimate L„ in centimeters and K, expressed on a monthly basis. For growth estimates based on apparent known age fish, we used a computer program written by Abramson (1963) and modified by Psaropulos (1966) of a least-squares procedure described by Tomlinson and Abramson (1961). ANALYSIS WITH UNKNOWN AGE FISH METHODS Fish landed at Abidjan, Dakar, and Pointe- Noire (Figure 1) were measured for predorsal length (tip of snout to anterior base of the dor- sal fin) by French scientists; fish were measured for fork length by lATTC scientists. In order to standardize the length measurements, we em- 176 Le GUEN and SAKAGAWA : APPARENT GROWTH OF YELLOWFIN TUNA POINTE-NOIRE Figure 1. -Area off Africa where the surface fishery for yellowfin tuna operates. ployed the relation, log Lf = 0.273 + 1.175 logLd to convert samples with predorsal length in centi- meters (Ld) to fork length in centimeters (Lf) . This relation is based on 508 observations and differs from Lf = (3.624 + 0.212 La)-, which was employed by Poinsard (1969). It has a slightly better correlation coefficient (r — 0.9943) than Poinsard's equation (r = 0.9940) (Lenarz, 1971).' Calculated fork lengths based on either equation are accurate only to 1-4 cm. Monthly length-frequency distributions were tabulated by 4-cm-fork-length groups for sam- ples from each region. Modes were identified and assumed to represent age classes within which lengths were normally distributed. Nor- mal distributions were then fitted to the length frequencies of samples in which two or more modes were present, and the mean length of each age class was estimated (Table 2). A computer program for separating size classes in a mix- ture that was written and described by Hassel- * Lenarz, W. H. 1971. Length-weight relations for five Atlantic scombrids. Unpublished manuscript, 9 p. Southwest Fisheries Center, National Marine Fisheries Service, NOAA, La Jolla, CA 92037. blad (1966) and modified by Tomlinson (1970)' was used to separate the age classes and estimate the mean lengths. For samples with only one prominent mode, the modal length, or midpoint of length interval of maximum frequency was considered the "mean length." Representative length-frequency distributions are shown in Fig- ure 2. Mean lengths for each sample are plotted in Figures 3-6. For each region a serial succession of increasing mean lengths with time was desig- nated a year class, with only one recruited per year although two groups appear to be recruited ^ Tomlinson, P. K. 1970. Program for separating mixture of normal distributions. Unpublished manu- script, 2 p. California Department of Fish and Game, Operations Research Branch, Long Beach, CA 90802. JULT N.309 AUGUST N-(55 240 J SEPTEMBER 200 N=556 40 riL ': r f^ OCTOBER N=76 NOVEMBER N=224 DECEMBER N=592 80 lOO 120 140 160 160 200 20 40 60 80 100 IZO I40 160 IBO 200 FORK LENGTH (cm) Figure 2. — Length-frequency distributions of samples from Pointe-Noire, 1970. Arrows indicate mean lengths of modal groups that were identified by curve fitting. o ,,-«.l963 4 1 1 FMAMJJ«SONDJFH*MJJASOMOJFMAMJJ*SONDjrH*MJJASON0JFHAHJJ*SOND 1966 1967 1961 1»69 I9T0 Figure 3. — Mean length of size groups of yellowfin tuna as a function of sampling date at Abidjan. Growth of the 1963-69 year classes are indicated 177 FISHERY BULLETIN: VOL. 71, NO. 1 Table 2. — Mean lengths (cm) of modal groups identified in samples from Abidjan, Dakar, Gulf of Guinea, Pointe-Noire. Values that were not used in the analysis of growth are shown in parentheses. and Region Year Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Abidjan 1966 80.7 97.2 (56.8) (70.8) 89.4 59.2 94.2 (140.9) 112.0 73.5 (155.2) 1967 87,3 84.3 92.0 (48.5) 92.8 (56.0) 62.9 65.3 124.6 118.2 (115.4) 129.7 146.6 103.8 114.9 122.9 114.8 125,3 149.7 144.5 156.8 1968 (71.7) 86.2 121.9 88.0 69.4 82.3 76.3 94.8 1969 76.2 (82.3) 81.9 105.4 87.9 112.2 (125.5) 153.6 118.2 145.6 140.0 (50.1) 67.9 122.4 1970 (54.8) 69,5 76.9 124.0 140.0 (55.6) 58.1 (48.0) (43.1) 73.8 125.3 (92.1) 105.6 143.0 (58.4) 60.6 123.0 129.4 143.7 148.4 Dakar 1968 (44.0) (56.3) (61.2) (63,5) 72.0 (55.3) 76.0 (63,0) (58.4) 62.3 66.9 64.0 117.9 122.3 120,7 70.1 (118.6) 81.2 88.4 89.6 100.1 1969 (70.1) 75.3 75.5 79.9 (43.6) (49.0) (49.1) (64.8) (69.4) 62.4 53.2 (61.9) 79.4 102.6 105.0 113.9 (57.4) (60.1) 90.9 95.7 102.5 (75.7) (78.5) 74.0 102.3 85,4 112,2 93.7 107.8 108.2 118.5 123.8 124.5 119.9 1970 64.8 (49.6) (49.4) (51.4) (50,6) (53.4) (55.3) (56.3) (39.7) (42.9) (30.9) (46.4) 124.8 71.8 73.5 72.5 76,2 (70.5) (72.3) (74.3) (56.4) 59.2 (45.0) (57.4) (88.9) (101.7) (106.0) 132.4 90.6 92.4 (72.3) 63.0 69.9 126.0 124.3 145.1 126.4 147.2 133.7 100.8 145.2 108.8 146.3 Gulf of 1968 62.9 57.8 70.0 Guinea 1969 1970 (54.8) 87.1 141.2 (55.5) 82.6 (106.8) 143.3 165.7 136.0 (160.2) 56.8 112.7 124.7 (56.0) 77.0 135.8 (52.6) 120.3 132.6 151.3 66.8 140.6 84.4 (59.4) 121.5 153.4 65.6 (103.9) 156.0 Pointe-Noire 1965 1966 (57.7) (76.1) 113.0 (57.4) (80.2) 116.4 59.1 104.0 (64.0) 99.1 145.3 60.3 (78.4) 121.1 60.0 63.0 102.2 1967 (50.5) (52.9) (58,0) (53.4) (53.6) (59.3) (61.4) 112.0 (68.0) 86.9 90.6 113.8 155.9 88,8 102.1 104.0 113.0 105.1 108.4 1968 (60.7) (68.0) (56.4) 59.7 62.2 64.6 65.9 (75.0) n.7 122.7 (73.4) (71.1) (73.6) (75.3) (76.2) (89.5) (154.6) 120.9 (85.4) 134.0 140.0 (90.9) 130.8 152.7 (87.0) 143.8 (156.6) (91.9) 136.8 (101.5) 1969 76.0 86.3 49.6 lOO.O (56.9) (55.7) (45.8) (57.7) (57.4) 59.9 56.6 (111.1) 94.3 (117.7) 148.4 164.8 104.2 (112.9) 154.5 (56.3) (117.5) 156.0 112.7 114.6 (127.7) 127.1 1970 (51,5) (48.5) (48.8) (48.9) (49.9) (50.7) (51.3) (53.7) (56.0) (55.3) (57.6) (56.0) 69.0 65.5 (57.5) 76.8 82.8 131.8 132.2 62.2 93.5 113.5 129.7 134.7 75.6 134.1 178 Le GUEN and SAKAGAWA : APPARENT GROWTH OF YELLOWFIN TUNA I50r 1969 JFMAMJ JASONDJFMAMJ JASONDJFMAMJJASOND 1968 1969 1970 Figure 4. — Mean length of size groups of yellowfin tuna as a function of sampling date at Dakar. Growth of the 1965-69 year classes are indicated. I75r 150- E ■« 125 100 75 - 50 JFMAMJJASONOJFMAMJJASONDJFMAMJJASOND 1968 1969 1970 Figure 5. — Mean length of size groups of yellowfin tuna from the Gulf of Guinea. Growth of the 1965-69 year classes are indicated. o __ , , ^ 1965 _ ,..^' o-o- ' ' r..- ^V- ^ ■OI967 ■T f o PI968 / . /:--' O ' i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 or' o 1 1 1 1 1 J 1 1 <:>-^>l969 1 1 I 1 in some years. Recruitment is assumed to be completed in the second year of life (Le Guen et al, 1969). RESULTS Recruitment Yellowfin tuna are recruited into the surface fishery when about 60 cm long. Recruitment is year-round but most pronounced during June to December. Two groups of yellowfin tuna appear to be recruited in some years. For example, in 1968 at Pointe-Noire (Figure 6) one group en- tered in January and another in August-Septem- ber. The January group was of low relative abundance and persisted up to a length of about 90 cm, while the August-September group was of high relative abundance and discernible up to a length of about 140 cm. A similar phenomenon was described by Hennemuth (1961) and later verified by Davidoff (1963) for yellowfin tuna of the eastern Pacific Ocean. Hennemuth sug- gested that sampling bias, differential growth in a year class, and multiple spawning were some possible causes of the phenomenon. Variation in the seasonal distribution of fishing effort can be added as another possible cause. Year Class Difference in Apparent Growth Estimates of apparent growth for individual year classes for each region are shown in Table 3. 75 ^ -OI963 ^ o &■ 0-' o ' I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ■' I I I I I I I I I I I [ I I ^I96S JFMAMJJASONDJFMAMJJASONDJFMAMJJASONDJFMAMJJASONDJFMAMJJASONOJFMAMJJASONO 1965 1966 1967 1968 1969 1970 Figure 6. — Mean length of size groups of yellowfin tuna as a function of sampling date at Pointe- Noire. Growth of the 1963-69 year classes are indicated. 179 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — Estimates of parameters of the von Bert- alanffy growth function for yellowfin tuna of unknown age from the eastern Atlantic Ocean. Region Year class K No. of obser- vations Range of lengths (cm) Abidian 1963 Linear 2 97.2-156.6 1964 138.6 0.137 9 80.7-149.7 1965 275.2 0.016 n 59.2-153.6 1966 Linear 9 62.9-148.4 1967 155.1 0.086 13 69.4-143.7 1968 Linear 4 67.9-105.6 All years 185.0 0.043 48 59.2-156.6 Dakar 1965 Linear 2 117,9-122.3 1966 Linear 18 70.1-147.2 1967 201.5 0.038 20 62.3-146.3 1968 557.2 0.008 11 53.2-108.8 All years 307.9 0.017 51 53.2-147.2 Gulf of 1965 677.4 0.002 5 135.8-165.7 Guinea 1966 497.5 0.009 3 77.0-132.6 1967 174.4 0.052 8 57.8-143.3 1968 Linear 2 56.8- 87.1 All years 185.0 0.041 13 56.8-165.7 Pointe-Noire 1963 168.3 0.067 5 104.0-155.9 1964 273.9 0.017 10 59.1-164.8 1965 162.9 0.059 16 63.0-156.0 1966 191.9 0.024 11 61.4-127.7 1967 160.2 0.05!1 17 59.7-134.7 1968 177.8 0.033 3 56.6-113.5 All years 210.1 0.027 67 56.6-164.8 All regions 1963 158.5 0.136 7 97.2-156.6 1964 237.4 0.023 19 59.1-164.8 1965 19-1.0 0.064 34 59.2-165.7 1966 895.7 0.O03 41 61.4-148.4 1967 172.6 0.054 58 57.8-146.3 1968 502.4 0.009 25 53.2-113.5 All years 194.8 0.035 184 53.2-165.7 In some instances, the estimates of K and L„ are unexpectedly too high or too low, indicating that the estimates are inappropriate for the entire life span of the species. According to Knight (1968) and Le Guen (1971), a possible cause of variation in K and L« is lack of size measure- ments for the entire life span of the species. This appears to be the case in some instances for our data. Length measurements for Dakar, for ex- ample, were from catches made predominantly by pole-and-line, or baitboats that generally catch small fish, a characteristic that is well document- ed (Pianet and Le Hir, 1971). Consequently, large fish were underrepresented in the samples, resulting in heavier weight on the lower size groups. Estimates of L^ were therefore unrea- sonably high, while those of K were unreason- ably low. It should be noted that generally L„ and K are inversely correlated (Beverton and Holt, 1959). For some year classes, apparent growth ap- pears to be exceptionally faster than for others. Apparent growth of yellowfin tuna from Pointe- Noire can best illustrate this point (Figure 6). The 1965 and 1967 year classes grew at a faster rate than the 1964 or 1966 year class. The re- sult was an apparent convergence of the growth curve for the 1964 year class with that for the 1965 year class, and the 1966 year class with the 1967 year class. In each case, there appears to be no relation between the time of recruitment and the rate of growth. Regional Differences in Apparent Growth For each region, the von Bertalanffy equation was fitted to data for all year classes combined. Apparent growth of yellowfin tuna from Abid- jan, Gulf of Guinea, and Pointe-Noire was quite similar for sizes (ranged from about 60 to 160 cm long) observed in the samples (Figure 7). Apparent growth of Dakar fish, on the other hand, seemed exceptionally faster, which is at- 180 r 160 140 - 120 - O 100 - 80 60 ABIDJAN- GULF OF GUINEA ALL REGIONS DAKAF! POINTE-NOIRE 2 3 4 5 APPARENT AGE (YEARS) Figure 7. — Growth of yellowfin tuna from the eastern Atlantic Ocean. 180 Le GUEN and SAKAGAWA: APPARENT GROWTH OF YELLOWFIN TUNA tributed to lack of data on older fish as was dis- cussed earlier. The range of mean length is 50 to 150 cm. Estimated Length at Age An estimate of L« --= 194.8 cm, and K = 0.035, on a monthly basis, was derived from data for all year classes and regions combined (Table 3). We believe these estimates are the "best," on the average, for yellowfin tuna of the eastern At- lantic Ocean, because they were based on data from a broad geographic area ofl^ Africa and a wide range of sizes. The estimated growth curve is quite similar to that for Abidjan, Gulf of Guinea, and Pointe-Noire (Figure?), For in- dividual year classes, however, estimates of L^ and K can be expected to deviate from the av- erage, since there is an apparent diflference in apparent growth among year classes (Table 3) and considerable scatter of observed mean lengths around the average curve (Figure 8). Data on growth of tagged yellowfin tuna in the eastern Pacific and on the time of spawning in the Atlantic were used to estimate U in months. The above best estimates of parameters of the von Bertalanfl'y equation were then used to esti- mate the length of yellowfin tuna at particular ages. A tacit assumption of this method of esti- mating length at age is that the von BertalanflFy function is a valid growth model for yellowfin tuna, and the date of birth is constant. Schaefer, Chatwin, and Broadhead (1961) re- ported that yellowfin tuna of 40-49 cm long at tagging grew at a rate of 33 cm /year. They in- dicated that growth was probably adversely af- fected by tagging, implying that their estimate was too low. Le Guen et al. (1969) reviewed the literature on time of spawning of yellowfin tuna in the At- lantic Ocean. They concluded that spawning oc- curred primarily at temperatures greater than 26°C and salinity of about 33. 5^;^. From seasonal distributions of temperature, salinity, and tuna larvae captured off" Africa, they estimated that spawning peaked on about March 1 off Pointe- Noire and on about July 1 oflF Dakar. From the above information together with the fact that recruitment into the surface fishery is APMRENT AGE IMOejTHS) Figure 8. — Mean length of year classes of yellowfin tuna as a function of age. The curv^e is for all year classes and regions combined and is estimated by a von Bert- alanffy growth function. Average of observed mean lengths (circles) and the range of mean lengths (vertical line) at various estimated ages are shown. generally from June to December, we estimated that yellowfin tuna were, on the average, 18 months old at recruitment, about 60 cm long, and ^0 = 7.48. Estimates of length at age were cal- culated with Zoo = 194.8, K = 0.035, and to = 7.48, employed in the von Bertalanflfy function (Table 4). The estimates are graphed in Fig- ure 8, together with monthly mean lengths of individual year classes of each region. There is considerable scatter of the data about the line and an indication that lengths at age 50 months and older are overestimated. Estimated Weight at Age Length can be converted to weight with a weight-length relation. Lenarz (see footnote 4) reported that the weight-length relation for yel- lowfin tuna from the eastern Atlantic is W = 0.0000214 L/^•«■^^ where W = weight in kilo- grams and Lf = fork length in centimeters. This equation was employed to convert estimates of length at age to weight at age (Table 4) . ANALYSIS WITH APPARENT KNOWN AGE FISH METHODS The method of analysis with apparent known 181 FISHERY BULLETIN: VOL. 71, NO. 1 Table 4. — Observed and estimated size at various ages of yellowfin tuna from the Atlantic and Pacific Oceans. Length (cm) is shown for most ages, and weight (kg) in parentheses for a few ages. Estimated length is based on the von BertalanfFy growth function. Source of data Age (years) Yang et a!., 1969 Atlantic Ocean Le Guen el- a!., 1969 Present study Eastern Atlantic Eastern Atlantic Davidoff, 1963 Eastern Pacific Pointe-Noire All regions Sao To me-Angola All regions Observed Estimated Observed Estimated Observed Estimated Observed Estimated Observed Estimated Observed Estimated 1.0 .. 54.0 .. 33.2 __ 32.2 17.3 (0.1) 28.5 (0.4) 34.6 1.5 66.1 75.8 64.6 62.9 63.8 60.0 61.5 53.9 (3.0) 62.2 60.O (4.2) __ 61.9 2.0 86.1 94.9 84.6 86.6 79.5 83.1 77.6 82.0(10.5) 82.3 85.5(11.9) 83.0 84.7 2.5 104.1 111.5 108.3 105.6 103.9 102.0 111.0 103.6(21.1) 105.0 106.2(22.7) las.o 103.8 3.0 120.0 125.9 __ 120.9 124.0 117.7 116.1 120.2(82.8) 125.0 123.0(35.1) 122.0 119.7 3.5 132.9 138.5 132.2 133.1 132.2 130.6 132.2 133.0(44.3) 140.6 136.6(48.0) 136.0 132.9 4.0 149.4 142.9 141.3 135.6 142.8(54.7) __ 147.6(60.5) 141.0 144.0 4.5 158.9 147.0 150.7 143.6 150.1 147.0 150.3(63.8) 153.4 156.6(72.0) ._ 153.3 5.0 — 167.2 152.0 157.0 152.0 157.4 153.7 156.1(71.3) 164.8 163.8(82.4) — 161.0 age fish were similar to those described by Le Guen et al. (1969) and are briefly described as follows. Predorsal length-frequency distribu- tions were tabulated for monthly samples col- lected in 1967-71 off Pointe-Noire in an area from Sao Tome to southern Angola. Modes were se- lected by comparison of successive maxima in the length-frequency distributions and mean pre- dorsal length estimated for each size group by a method described by Gheno and Le Guen (1968). Mean dorsal lengths were then con- verted to fork lengths with the aid of Table 5, which was based on fish measured for both pre- dorsal and fork lengths at Pointe-Noire. The data in Table 5 give a fork length-predorsal length relation of log L/ = 0.299 + 1.162 log Ld that is not significantly different from the equa- tion used earlier. An estimated age was assigned to each size group (Table 6) based on: (1) date of birth of yellowfin tuna caught off Point-Noire is on the average March 1 and (2) recruitment occurs in the second year of life (Le Guen et al,. 1969). Estimates of parameters of the von Bertalanffy function were then calculated with Psaropolos' (1966) computer program. Length at age estimates were converted to weight at age with the weight-length relation of Lenarz (see footnote 4) , which was mentioned earlier. Table 5. — Predorsal length and fork length measure- ments of yellowfin tuna landed at Pointe-Noire, 1967-71. Predorsal lengfh (cm) Mean fork length (cm) Number of obser- vations Predorsal length (cm) Mean fork length (cm) Number of obser- vations 12 39.0 11 31 109.5 46 13 40.9 21 32 111.3 33 14 45.0 18 33 116.1 27 15 47.3 37 34 118.8 19 16 50.0 36 35 122.9 26 17 53.9 33 36 132.3 24 18 57.2 58 37 134.7 35 19 59.8 83 38 138.4 25 20 63.1 66 39 143.7 28 21 66.3 43 40 145.7 29 22 71.0 20 41 149.7 29 23 74.6 23 42 152.3 14 24 76.0 18 43 158.8 5 25 61. 1 16 44 164.0 5 26 84.2 16 45 166.3 10 27 89.0 9 46 172.0 6 28 92.8 21 47 175.4 8 29 99.1 28 48 177.8 7 30 104.9 27 49 179.8 2 RESULTS Estimates of parameters of the von Bertalan- flfy function were L^ = 175.17 cm (SE = 3.67), K = 0.044 per month (SE = 0.003), and U = 9.643 months (SE = 0.815). These estimates are quite similar to those derived by Le Guen et al. (1969) for the Pointe-Noire region based on only data from 1967-68 (Table 7) ; but L« is significantly lower and K significantly higher than our best estimates for yellowfin tuna from a larger area of the eastern Atlantic, even when the difference in range of lengths in the data is taken into account. On the other hand, length at age and weight at age estimates for ages 1-5 years are quite similar to those for the entire eastern Atlantic (Table 4). Thus, we conclude 182 Le GUEN and SAKAGAWA : APPARENT GROWTH OF YELLOWFIN TUNA Table 6. — Size classes (cm) of yellowfin tuna identified in samples from Sao Tome to southern Angola. Year classes are separated by horizontal lines. Age (monrtis) 1967-681 1969 1970 1971 18 64.5 58.5 19 59.8 20 61.4 21 70.3 22 85.0 75.3 68.6 58.5 23 85.1 71.0 63.1 24 84.6 32.6 74.6 68.6 25 89.0 90.9 81.1 72.8 26 90.6 104.9 82.6 74.6 27 91.0 107.2 92.8 76.0 28 95.5 107.2 86.6 29 102.6 30 108.3 113.7 31 107.7 116.1 32 109.4 127.5 33 114.5 127.5 34 124.0 35 120.0 111.3 36 116.1 37 122.0 118.8 33 126.7 39 123.5 40 134.7 41 136.8 42 132.2 43 139.0 46 133.5 47 138.6 48 136.5 134.7 52 147.0 141.0 138.4 53 151.3 143.7 54 147.0 55 150.5 60 152.0 155.5 61 149.7 156.0 64 158.8 65 163.4 160.0 66 161.9 73 166.3 74 168.1 75 168.0 170.1 76 170.1 77 170.1 1 Data from Le Guen et (1969). that there is no appreciable difference in the es- timate of apparent growth of yellowfin tuna from the region of the eastern Atlantic, illustrated in Figure 1 or from a smaller region within that part such as off Pointe-Noire. COMPARISON OE GROWTH ESTIMATES Studies on growth of yellowfin tuna have largely been based on two types of data: length- frequency distributions and scale readings. For comparative purposes we chose two studies that were based on scale readings — one each from the Pacific (Yabuta, Yukinawa, and Warashina, 1960) and Atlantic (Yang et al., 1969)— and three studies that were based on modal progres- sion of length-frequency distributions — two from the Pacific (Davidoff, 1963; Moore, 1951) and one from the Atlantic (Le Guen et al., 1969) — for comparison with our best estimates for the eastern Atlantic. The procedure of estimating the parameters of the von Bertalanffy function was standardized with the use of the Fabens' (1965) procedure whenever appropriate data were available. ESTIMATES FROM SCALE READINGS Lengths at mark formation from interpreta- tion of marks on scales were reported by Yabuta et al. (1960). They indicated that mark forma- tion occurs twice a year, in March-April and in September-October, or 6 months apart in the western Pacific. An estimate of growth was calculated from their data with 6 months between marks (Table 7). Growth appears to be sub- stantially slower in the western Pacific than in the eastern Atlantic (Figure 9) . Either growth is indeed slower in the western Pacific or the in- terpretation of scale marks by Yabuta et al. is in error. The latter possibility is suggested by the absence in their data of fish greater than 119 cm long with a designated mark, although fish as large as 161 cm long were reportedly sampled. Moreover, only about 42% of their scales were readable. Other studies made in the western Pacific (see Shomura, 1966; Suzuki, 1971) suggest that growth was underestimated by Yabuta et al. Loo = 222.8 cm and K ^ 0.023, on a monthly basis, were estimated by Yang et al. (1969). Their estimates were based on scale readings of 296 yellowfin tuna caught by the Atlantic long- line fishery. Since Yang et al. used the Walford (1946) procedure to estimate growth, we re- calculated growth with the Fabens' procedure using the data of Yang et al. and the assumption that the scale marks formed every 6 months. The results (Table 7) were not markedly dif- ferent from the estimates by Yang et al. Com- pared to our best estimate of growth rate (K), on the other hand, their estimate is substantially 183 FISHERY BULLETIN: VOL. 71, NO. 1 Table 7. — Estimates of parameters of the von Bertalanffy growth function for yellowfin tuna from the Atlantic and Pacific Oceans. Es- timates are based on data reported in various studies, and were calcu- lated by Fabens' (1965) procedure, except those of Le Guen et al. (1969). Region ioo K Range of length (cm) Source of data Data Atlantic Ocean 223.0 0.023 66-130 Yang et al., 1969 Scale readings; Table 6 Eastern Atlantic Le Guen et al.. Length frequencies; Dakar 206.6 0.026 63-162 1969 estimates report- Pointe-Noire 182.4 0.037 64-162 ed by authors All areas 191.7 0.032 63-162 Eastern Atlantic Present study Length frequencies Abidjan 185.0 0.043 59-157 (Age unknown) Dakar 307.9 0.017 53-147 Gulf of Guinea 185.0 0.041 57-166 Pointe-Noire 210.1 0.027 57-165 All oreas 194.8 0.035 53-166 Eastern Atlantic Sao Tome-Angola 175.2 0.044 58-170 Present study Length frequencies (Age known) Central Pacific 191.9 0.036 47-168 Moore, 1951 Length frequencies; Table H Eastern Pacific 200.3 0.030 69-148 Davidoff, 1963 Length frequencies; Toble 6 Western Pacific Yabuta et al., 1960 Scale readings; Males 202.1 0.023 58-119 Table 5 Females 174.9 0.03 1 57-1 19 All sexesi 188.4 0.027 57-119 1 Estimates were based on weighed average length et al. (1960). Sample size of each sex wos used as for each scale mark reported by Yabuta the weighing factor. smaller. Possiblj^ this smaller K is caused by error in the interpretation of scale marks and the paucity of large fish in their data. The max- imum number of marks observed by Yang et al. was five, with a corresponding mean length of 132.9 cm at time of fifth mark formation, but fish as large as 180 cm long were reportedly sampled. For our study, fish as large as mean length 166 cm were used in the calculations. ESTIMATES FROM MODAL PROGRESSION Davidoff (1963) examined modal progressions of length-frequency distributions of eastern Pa- cific yellowfin tuna caught by baitboats and purse seiners and calculated with the Walford pro- cedure L«, = 167 cm and K = 0.05, on a monthly basis, which he noted were similar to earlier esti- mates reported by Hennemuth (1961). David- oflF's estimates were based on average modal length at each age of all year classes combined. Equal weight was therefore given to each datum point in his calculation. Using the Fabens' procedure and data for each year class reported by Davidoff (his Table 6), we recalculated the growth estimates. The re- sults, L„ = 200.3 and K = 0.030, are consider- ably larger for L„ and smaller for K than Davidofl["'s estimates but similar to our estimates for Atlantic yellowfin tuna (Table 7). Hennemuth (1961) reported that fish 70 cm long in the eastern Pacific were about 20 months old. Entered into the von Bertalanflfy equation, this gives a ^o of 5.67 months with L« = 200.3 and K — 0.030, and a means of estimating length at age for eastern Pacific yellowfin tuna. The results are shown in Table 4. They compared favorably with our estimates for Atlantic yellow- fin tuna, although apparent growth in the east- ern Atlantic is 0.9 to 2.8 "^f faster than that in the eastern Pacific for ages 2 through 5 years. Moore (1951) based his estimates of growth on length-frequency distributions of yellowfin tuna caught primarily by longline gear in the central Pacific. He used the Walford procedure and calculated L« = 190.0 cm and K = 0.037 per month. Because of a limitation of Walford's 184 Le GUEN and SAKAGAWA: APPARENT GROWTH OF YELLOWFIN TUNA 180 r 160 - 140 PRESENT STUDY (APPARENT KNOWN AGE) q: O 120 100 - 3 4 APPARENT AGE (YEARS) Figure 9. — Comparison of growth of yellowfin tuna from the Pacific and Atlantic Oceans. Curves were adjusted to a common base of age 1.5 years = 60 cm long and were estimated, except for that of Le Guen et al. (1969), from data reported in various studies. method — requiring length measurements at equal time intervals — Moore was able to use only 16 out of his 25 observations. We recalcu- lated Loo and K, using the Fabens' procedure and the 25 observations reported by Moore (his Table H) . The estimates, L« = 191.9 and K = 0.036, differ slightly from those of Moore and are very similar to our estimates for Atlantic yellowfin tuna. Le Guen et al. (1969) estimated growth of yel- lowfin tuna from Dakar, Pointe-Noire, and both regions combined, based on modal progression of length-frequency samples (Table 7). Their samples were identical to some used in our study, but their estimate of growth for combined re- gions is slightly lower than ours ; the difference in estimated lengths for ages 2 through 5 years is 2.8 to 4.3% less (Table 4). Part of the dif- ference is in the method of analysis. The esti- mates by Le Guen et al. were based on mode se- lection from predorsal length distributions, and the lengths of size groups were not assumed to be normally distributed. Predorsal lengths were then converted to fork length; whereas in our best estimate predorsal length was converted to fork length by a log function before frequency distributions were analyzed, and the lengths of size groups were assumed to be normally distrib- uted. Furthermore, Le Guen et al. assumed that the date of birth of fish of each year class of a region was the same and accordingly ages were assigned to size classes; such an assumption was not made for our estimate of K and I^ ; but for obtaining to we assumed that yellowfin tuna of 60 cm long are 18 months old. Nevertheless, the difference is insignificant in view of the fact that there is considerable variability in observed mean lengths at age (Figure 7). DISCUSSION It is obvious from the results that estimates of growth of yellowfin tuna are quite variable and largely dependent on the method of anal- ysis. Both the length-frequency and scale meth- ods are based on various assumptions that are not always satisfied. For example, the assump- tion in the length-frequency method that size groups represent age groups, and the age groups are formed once a year, i.e., hatching within a short period, or season, is not completely sat- isfied for yellowfin tuna, since spawning occurs over several months (Matsumoto, 1966; Le Guen et al., 1969; Richards, 1969). Nevertheless, in many areas, as in the eastern Atlantic, there is generally a peak month of spawning (Le Guen et al., 1969) that can create a size group discern- ible in size-frequency distributions in later dates. The scale method assumes that the scale marks are formed at regular intervals. So far, this as- sumption has not been satisfactorily verified for yellowfin tuna, although Yabuta et al. (1960) and Yangetal. (1969) have indicated that the marks formed every 6 months. Furthermore, because yellowfin tuna generally spawn over an extended season, the age at first annulus formation is not the same for all individuals of a year class. The back-calculated length at age I may therefore be questionable. It is surprising, however, that the observed lengths at age are remarkably sim- ilar for studies based on the scale and length- 185 FISHERY BULLETIN: VOL. 71, NO. 1 frequency methods (Table 4). This suggests that the marks on scales of Atlantic yellowfin tuna are indeed layed down at regular intervals and that observed lengths at age rather than esti- mates of parameters of the von Bertalanffy func- tion are more meaningful in comparison of growth of yellowfin tuna. For such a compari- son, the average growth rate of Atlantic yellow- fin tuna is 17 cm/6 months, based on the scale method, and 18 cm/6 months, based on the length-frequency method for ages 1.5-3.5 years. The comparison of observed lengths at age also indicates that there is little difference be- tween growth of Atlantic and Pacific yellowfin tuna (Table 4) . Yang et al. (1969) , on the other hand, suggested that growth is faster in the At- lantic than in the Pacific. We analyzed their data with analysis of covariance and found that their Walford curves for the Atlantic and Pacific yellowfin tuna were not significantly different from a common line {F2. 5 = 0.474) nor from parallel lines (Fi, 5 = 0.904). Thus the sug- gestion by Yang et al. was not demonstrated by their data, but in fact, growth of yellowfin tuna appears to be similar in the two oceans. Finally, since the parameters of the von Bert- alanffy growth function are sensitive to the method of analysis and range of sizes used to estimate them, we recommend that the observed size at age rather than the estimated size at age from the von Bertalanffy growth function be used in estimating yield per recruitment. The Ricker (1958) model of yield per recruitment, for example, is appropriate for observed values, ACKNOWLEDGMENT We thank W. H. Bayliff , D. Kramer, and W. H. Lenarz for reading the manuscript and offering helpful suggestions. LITERATURE CITED Abramson, N. J. 1963. Computer programs for fisheries problems. Trans. Am. Fish. Soc. 92:310. 1971. Computer programs for fish stock assess- ment. FAO (Food Agric. Organ. U.N.) Fish Tech. Pap. 101, 149 p. Beverton, R. J. H., AND S. J. Holt. 1959. A review of the lifespans and mortality rates * of fish in nature, and their relation to growth and other physiological characteristics. In G. E. W. Wolstenholm and M. O'Conner (editors). Ciba Foundation colloquia on ageing, Vol. 5, p. 142-180. J. & A. Churchill Ltd., Lond. Champagnat, C, and F. Lhomme. 1970. La peche thoniere a Dakar de 1966 a 1969. Centre de Recherches Oceanographiques de Dakar- Thiaroye (Senegal), DSP 27, 24 p. Davidoff, E. B. 1963. Size and year class composition of catch, age and growth of yellowfin tuna in the Eastern Trop- ical Pacific Ocean, 1951-1961. [In English and Spanish.] Inter-Am. Trop. Tuna Comm., Bull. 8:199-251. Fabens, a. J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth 29:265-289. Gheno, Y., and J. C. Le Guen. 1968. Determination de I'age et croissance de Sardinella eba (Val.) dans la region de Pointe- Noire. [English summ.] Cah. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer), Ser. Oceanogr. 6(2) : 69-82. Gulland, J. A. 1969. Manual of methods for fish stock assessment. Part 1. Fish population analysis. FAO (Food Agric. Organ. U.N.) Man. Fish. Sci. 4, 154 p. Hasselblad, V. 1966. Estimation of parameters for a mixture of normal distributions. Technometrics 8:431-444. Hennemuth, R. C. 1961. Size and year class composition of catch, age and growth of yellowfin tuna in the Eastern Trop- ical Pacific Ocean for the years 1954-1958. [In English and Spanish.] Inter-Am. Trop. Tuna Comm., Bull. 5:1-112. Joseph, J., and T. P. Calkins. 1969. Population dynamics of the skipjack tuna (Katsuwonus pelamis) of the eastern Pacific Ocean. [In English and Spanish.] Inter-Am. Trop. Tuna Comm., Bull. 13:1-273. Knight, W. 1968. Asymptotic growth : an example of nonsense disguised as mathematics. J. Fish. Res. Board Can. 25:1303-1307. 1969. A formulation of the von Bertalanffy growth curve when the growth rate is roughly constant. J. Fish. Res. Board Can. 26:3069-3072. Le Guen, J.-C. 1971. Dynamique des populations de Pseudotolithus (Fonticulus) elongatus (Bowd. 1825). Poissons - Sciaenidae. [English abstr.] Cah. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer), Ser. Oceanogr. 9:3-84. 186 Le GUEN and SAKAGAWA : APPARENT GROWTH OF YELLOWFIN TUNA Le Guen, J. C, F. Baudin-Laurencin, and C. Champagnat. 1969. Croissance de I'albacore (Thunnus albacares) dans les regions de Pointe-Noire et de Dakar. [English summ.] Cah. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer) , Ser. Oceanogr. 7(1) : 19-40. Matsumoto, W. M. 1966. Distribution and abundance of tuna larvae in the Pacific Ocean. In T. A. Manar (editor), Proceedings, Governor's Conference on Central Pacific Fishery Resources, State of Hawaii, p. 221- 230. Moore, H. L. 1951. Estimation of age and growth of yellowfin tuna (Neoth^innus macropterus) in Hawaiian waters by size frequencies. U.S. Fish Wildl. Serv., Fish. Bull. 52:133-149. O.R.S.T.O.M. 1971. Les mensurations d'albacores {Thunnus al- bacares) et de listaos (Katsuwonus pelamys) faites a Dakar, Abidjan et Pointe-Noire entre 1965 et 1970. [English abstr.] Doc. Sci. Cent. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer) Pointe-Noire, Nouv. Ser., 11, 9 p, 49 tables. Pianet, R., and Y. Le Hir. 1971. Evolution de la peche thoniere dans le sud du Golfe de Guinee de 1964 a 1970. [English abstr.] Doc. Sci. Cent. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer) Pointe-Noire, Nouv. Ser., 17:16- 48. Poinsard, F. 1969. Relations entre longueur predorsale, longueur a la fourche et poids des albacores Thunnus alba- cares (Bonnaterre) peches dans le sud du Golfe du Guinee. Cah. O.R.S.T.O.M. (Off. Rech. Sci. Tech. Outre-Mer), Ser. Oceanogr. 7(2) :89-94. PSAROPULOS, C. T. 1966. Computer program manual. Inter- Am. Trop. Tuna Comm., Intern. Rep. 1, 59 p. Richards, W. J. 1969. Distribution and relative apparent abundance of larval tunas collected in the tropical Atlantic during Equalant surveys I and II. In Proceedings of the Symposium on the Oceanography and Fish- eries Resources of the Tropical Atlantic, Abidjan, 1966, p. 289-315. UNESCO, Paris. RiCKER, W. E. 1958. Handbook of computations for biological sta- tistics of fish populations. Fish. Res. Board Can., Bull. 119, 300 p. RiCKLEFS, R. E. 1967. A graphical method of fitting equations to growth curves. Ecology 48:978-983. Rothschild, B. J. 1967. Estimates of the growth of skipjack tuna (Katsuivonus pelamis) in the Hawaiian Islands. Indo-Pac. Fish. Counc, Proc. 12th Sess., Sect. 2: 100-111. SCHAEFER, M. B., AND R. J. H. BeVERTON. 1963. Fishery dynamics — their analysis and inter- pretation. In M. N. Hill (editor). The sea. Vol. 2, p. 464-483. Wiley, N.Y. Schaefer, M. B., B. M. Chatwin, and G. C. Broadhead. 1961. Tagging and recovery of tropical tunas, 1955- 1959. [In English and Spanish.] Inter-Am. Trop. Tuna Comm., Bull. 5:341-455. Shomura, R. S. 1966. Age and growth studies of four species of tunas in the Pacific Ocean. In T. A. Manar (editor), Proceedings, Governor's Conference on Central Fishery Resources, State of Hawaii, p. 203-219. Suzuki, Z. 1971. Comparison of growth parameters estimated for the yellowfin tuna in the Pacific Ocean. Far Seas Fish. Lab. (Shimizu), Bull. 5:89-105. Tomlinson, p. K., and N. J. Abramson. 1961. Fitting a von Bertalanffy growth curve by least squares. Calif. Dep. Fish Game, Fish Bull. 116, 69 p. Yabuta, Y., M. Yukinawa, and Y. Warashina. 1960. Growth and age of yellowfin tuna. I. Age determination (scale method) . Rep. Nankai Reg. Fish. Res. Lab. 12:63-74. Yang, R. T., Y. Nose, and Y. Hiyama. 1969. A comparative study on the age and growth of yellowfin tunas from the Pacific and Atlantic Oceans. Far Seas Fish. Res. Lab. (Shimizu), Bull. 2:1-21. Walford, L. a. 1946. A new graphic method of describing the growth of animals. Biol. Bull. (Woods Hole) 90: 141-147. 187 DESCRIPTIONS OF THE LARVAE OF FOUR NORTH PACIFIC PORCELLANIDAE (CRUSTACEA: ANOMURA)' S. L. Conor and J. J. Gonor^ ABSTRACT Complete descriptions are given of the zoea and megalopa stages of four porcelain crabs common in the rocky intertidal regions of the Pacific coast of North America. Both larvae reared in the laboratory and larvae taken from the plankton were available for the species Petrolisthes cinctipes (Randall), Petrolisthefs eriomerus Stimpson, Pachy- cheles pubescens Holmes, and Pachycheles rudis Stimpson. All four species have a short prezoea stage, two zoeae, and a planktonic megalopa stage. Extensive variation was found in setal numbers between left and right members of appendage pairs and between indi- viduals in these larvae. Setal counts characteristics of species cannot be obtained from one or a few larvae, and ranges in the counts overlap considerably between species. The criteria of Lebour for grouping zoeal types in the Porcellanidae are applied to these four species and comparisons made between all available descriptions of porcellanid larvae. Intermolt growth of appendage buds occur in these four species, and it is con- cluded that this type of growth is the cause of most of the differences described previously as larval substages. Tabulations of species showing stage variation and intermolt growth are given. All larval stages of four common Pacific coast rocky intertidal porcelain crabs have been reared in this laboratory. Material obtained from cul- tures and from the plankton was used for the descriptions given here of the larvae of each species. One of these species, Pachycheles rudis, has been described previously (Knight, 1966), and in this case the available material permitted this description to be expanded. Attention was also directed to variation in larval characteris- tics since adequate numbers of specimens were available for examination. Gravid female porcelain crabs of the species Petrolisthes cinctipes (Randall), Petrolisthes eriomerus Stimpson, Pachycheles rudis Stimp- son, and Pachycheles pubescens Holmes were ' Research supported in part by a traineeship from Crant Nos. 5T1-WP-111-02, -03, and -04, Federal Wate>- Quality Administration, and by National Oceanic and Atmospheric Administration (maintained by the U.S. Department of Commerce) Institutional Sea Grant 2-35187. ^ Department of Oceanography and Marine Science Center, Oregon State University, Newport, OR 97365. Manuscript accepted January 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. collected from rocky intertidal shores at Boiler Bay and Yaquina Head on the central Oregon coast. P. rudis was collected from the lower intertidal ( — 0.8 to — 2.5 ft below mean lower low water) beneath Phyllospadix root mats on the seaward sides of large boulders. Pachycheles pubescens was collected from burrows and crev- ices in rock at the same tide level. The Petrol- isthes species were gathered somewhat higher in the intertidal regions of the two sites beneath loosely bedded rocks and rubble. All four spe- cies occur in various other habitats as well (Haig, 1960) but were taken from these areas for convenience. Egg-carrying Pachycheles rudis were collected in May and June of 1968. In the same months of 1969, egg-bearing females of the other species were obtained for larval studies. Gravid females were also noted dur- ing other months of the year, in agreement with Knudsen's (1964) observations on the repro- ductive cycles of these species. The specific identity of the females used was confirmed by comparing them to the descriptions given by Haig (1690). 189 FISHERY BULLETIN: VOL. 71, NO. 1 Female crabs were held individually in 1-liter beakers containing 600 ml of aerated, unfiltered seawater, standing in trays of running seawater ranging in temperature from 10.8° to 15.2°C. The water in the beakers was changed every 2 days. Larvae released by the females were re- moved from the beakers as soon as a hatch was discovered. Only actively swimming larvae of normal appearance were used. Glassware for handling and culturing larvae was washed in fresh water, rinsed with distilled water, and steam autoclaved to minimize mi- crobial contamination of the cultures. Seawater used for culturing was collected at high tide from the laboratory seawater system, passed through a sintered glass filter of pore size 40-60 m/x, and stored at 9°C in darkness. The salinity of this water ranged from 32.8 to 33.7^f. Laboratory-hatched larvae of Pachycheles rudis and Petrolisthes eriomerus were cultured in flasks, while P. cinctipes larvae were reared in mass cultures only. Larvae from a single Pachycheles pubescent female were divided be- tween flask and mass cultures. Light received by the larvae was not controlled. In all flask cultures larvae were maintained without aer- ation, initially with six to eight zoea I larvae per Erlenmeyer flask. The seawater volume allow- ance ranged from 25 ml per larva for early stages to 50 ml per larva for the larger later stages. Flasks were maintained at either 12°C or 15°C (± O.OrC) in refrigerated baths of circulating water. All flask cultures were ex- amined daily, mortality recorded, and newly molted larvae transferred to new cultures. Plankton tows were made intermittently from May through September of 1968 in Yaquina Bay and 1 mile outside the estuary along a rocky reef paralleling the shore. The catch was immedi- ately resuspended in cold seawater and kept under refrigeration while being taken to the lab- oratory. Live porcellanid larvae obtained in these hauls were removed to fresh seawater, sorted by zoeal stage, and further separated into four groups on the basis of differences in the patterns of the primary red chromatophores. Larvae within each group were further divided and placed in flask cultures. Mass cultures of laboratory-hatched Petrolis- thes cinctipes larvae were maintained in 5-gaI carboys filled with filtered seawater, with 150 or fewer larvae per carboy. Aerated mass cultures were immersed in baths of running seawater and experienced a temperature range of 10.8° to 15.2°C during the culture period and a max- imum daily fluctuation of 2.5°C due to tidal in- fluence. Larvae in mass culture were trans- ferred to clean containers and fresh seawater every 14 days. Zoea larvae were fed Artemia salina nauplii not more than 3 days old daily if required so that excess food was always present. The meg- alopa larvae required suspended algal material as food. Several monoalgal {Tetraselmis sp., Isochrysis sp.) and diatom (unidentified) cul- tures were tried singly and in combinations as a food source. Nutrient culture medium inoc- ulated with raw seawater was also used as a source of food organisms. Small stones were introduced into the flasks to provide the mega- lopae with a surface for settling. The larvae of the two Petrolisthes species and Pachycheles rudis were reared to the megalopa stage in cultures from embryos obtained from females held in the laboratory. Live larvae of these species taken from the plankton were reared to confirm stages obtained from the lab- oratory-hatched broods and to supplement labo- ratory-grown material for descriptive purposes. The larval history of P. pubescens was first de- scribed from larvae captured in the plankton and placed in laboratory cultures. Three gravid females of this species were collected later, and the identity of the planktonic larvae was con- firmed by comparison with larvae released by one of the females and reared to the second zoeal stage. Larvae in various stages of development were preserved for dissection primarily in 70 9^ eth- anol, but a few larval specimens were preserved in a mixture of 50 9<^ glycerine and 509^ ethanol for chromatophore examinations. Both tem- porary and permanent slides were made of the dissected materials. Temporary mounts were made in glycerine, and permanent mounts were made in Zeiss hardening phase medium as well 190 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE as in Turtox CMC-9AB and CMC-10 media.' Drawings were made from both types of ma- terial with the aid of a Wild M-20 compound microscope and camera lucida attachment. Since the length of the rostral and posterior spines of zoeae is highly variable and dependent on a number of factors, only the carapace proper was measured to serve as an indication of zoeal size. This measurement was made from the point of junction of the posterior spines to the posterior margin of the orbital arch, using an ocular micrometer. The term "seta" is used to indicate any firm- walled process, located on an appendage or other body surface, which is distinctly articulate at its base or attachment point (Figure 1). The figures are in part schematic and represent typ- ical setal counts only. Setules are not represent- ed in actual numbers and are often omitted en- tirely for clarity. Only one member of each setal pair present is figured for the exopodites of zoeal maxillipeds. Only the right member of each appendage pair is figured except for man- dibles, which are sometimes drawn in pairs from the dorsal view. Sexually mature females of the two genera considered in this study produce eggs of dis- tinctly diflferent types. Pachycheles rudis and P. pubescens females carry eggs which are 0.50 to 0.58 mm in size and brilliant yellow-orange when they are newly extruded. The eggs of P. rudis, as Knudsen (1964) and Knight (1966) observed, gradually change to a translucent am- ber color as the embryos develop and the yolk is absorbed. The color change with development is similar in P. pubescens. The age of the egg masses of the individual females collected in the field was unknown; however, the maximum length of time any female P. rudis carried eggs was 47 days, and two others carried eggs for 43 days before releasing larvae. This time may approach the length of time eggs are carried in this species. In both Pachycheles species, eggs hatched and all the larvae were released in a period of 10 to 20 hr. A hatch from a large Reference to trade names in the publication does not imply endorsement of commercial products by the Na- tional Marine Fisheries Service. female (carapace width 14.2-15.0 mm) may yield between 2,000 and 3,000 larvae but brood size varies (Knudsen, 1964). In contrast, the eggs of Petrolisthes cinctipes and P. eriomerus are somewhat larger, 0.80 to 0.84 mm, and are deep scarlet to maroon in color, when newly extruded, similar to those of other species in the genus (Wear, r965b; Greenwood, 1965). As Boolootian et al. (1959) also ob- served, the eggs gradually change to a translu- cent brownish red color as the embryos develop and the yolk is absorbed. Females of these two species were deliberately collected close to the hatching time; consequently, no information on the length of the brooding period was obtained. The time required for a female to complete a hatch is similar in these two Petrolisthes species but diflfers considerably from that observed for the Pachycheles species. Female Petrolisthes require 40 to 70 hr to release an entire group of larvae in the laboratory. Release of the lar- vae in these two species occurs in spurts with "resting" periods between each period of con- centrated release. The embryonic and larval histories of all four species observed in the laboratory have a num- ber of characteristics in common. Fully devel- oped embryos examined prior to eclosion pos- sessed the full complement of primary red chromatophores found throughout the active larval life of these crabs. Although occasional minor variations were noted both in laboratory hatches and in larvae from the plankton, the occurrence and placement of the chromatophores is generally stable and is species specific, as var- ious workers have noted in other species (Wear, 1964a, b, 1965a, b; Greenwood, 1965; Gore, 1968; Gurney, 1942). The primary chromatophores can thus be used to identify a larva to the species level at any stage of development. In the hatching of all four species, females re- leased larvae in the form of prezoeae. This has been observed in other porcellanid species, for example, by Lebour (1943), Greenwood (1965), Wear (1965b, 1966), and Gore (1968). The duration of this stage in the laboratory varies considerably and lasts from 10 min to about 1 hr in the species studied here. Prezoeae were never collected in the plankton. 191 FISHERY BULLETIN: VOL. 71, NO. 1 / 0.05 0.03 Figure 1. — Setal types in four porcellanid species studied: A-E (Scale 1), zoeal telson processes, distal portions: A - processes 3 and 4, Pachycheles spp. ; B - processes 5, 6, and 7, Pachycheles spp.; C - median spine, Zoea II, Pachycheles spp.; D - processes 3 through 7, Petrolisthes spp. ; E - median spine, Zoea II, Petrolisthes spp. F-K (Scale 2), zoeal maxillary setae, all four species. L, N (Scale 2), scythelike cleaning setae, straight and hooked, megalopa, pereiopod 5, all four species. M, O (Scale 2), "spikelike" and "brushlike" setae, megalopa, maxilliped III, all four species. Scales in millimeter. 192 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE After the larva has spent a period of time as a prezoea, the cuticle over the carapace is rup- tured dorsally along the midline and the prezoea molts to the first true zoeal stage. The rostrum is the first portion of the body to be freed, fol- lowed by the functional mouth parts and nata- tory maxillipeds. The abdomen and telson are the last portions to be withdrawn from the cu- ticle. When the molt is completed, the natatory setae and setae of the telson, previously com- pacted and confined by the prezoeal cuticle, be- come fully extended and serve to keep the larva afloat and propel it through the water. The rostral spine, partially invaginated into the carapace in the prezoea, straightens out almost immediately following the molt. The posterior carapace spines which often, but not always, be- come tightly coiled immediately after the cuticle is shed may take several minutes to several hours to uncoil. The cuticle is shed virtually intact except for the original dorsal split. Often in later molts, the delicate elastic cuticle covering the pereiopod and pleopod buds is shriveled or damaged in the cast exoskeleton. Larvae of all four species always pass through two true zoeal stages and one megalopal stage (terminology of Williamson, 1957) in the lab- oratory as do certain other porcellanids (Le Roux, 1961, 1966; Knight, 1966; Boschi, Scelzo, and Goldstein, 1967; Gore, 1968, 1970, 1971a, b, 1972) . This was true of larvae hatched in the laboratory as well as those taken from the plank- ton. In these four species, both zoeal stages dem- onstrate the remarkable property of intermolt growth in which certain of the larval appendages increase in size, apparently throughout the dur- ation of a stage, while the larval cuticle remains intact. This phenomenon has been reported for various other porcellanid larvae (Le Roux, 1961, 1966; Kurata, 1964; Knight, 1966; Gore, 1968, 1970, 1971a, b, 1972) but has not yet been thoroughly investigated. Appendages may in- crease in size as much as threefold between molts as a result of this growth pattern. The number of specimens of each stage dis- sected and examined is given in Table 1. The ranges given in the descriptions refer only to variations found within these specimens. Many other specimens were examined but not dis- sected. LARVAL DEVELOPMENT OF PETROLISTHES CINCTIPES (RANDALL) Larvae of P. cincti'pes were reared using the mass culture method. The single megalopa thus obtained as well as one megalopa from the plankton were preserved and dissected for purposes of description, PREZOEA (Figure 2) The prezoea of P. cinctipes is virtually spine- less and hairless. The carapace has a generally rounded appearance because the rostral and posterior spines are curved downward and in- ward toward the center of the body. These spines are further compacted by being tele- scoped and invaginated into their respective portions of the carapace. The natatory setae are nonfunctional, withdrawn into the ends of the maxillipeds, and held in place by the prezoeal cuticle. The primary red chromatophores ap- pear in the following locations: one on either side of the mouth; one distally on the basipodite of each second maxilliped; one distally in ab- dominal segment number two or between seg- ments two and three; and one on either side of the body between the bases of maxillipeds one and two. The rostrum and posterior spines are tipped with red, and an additional red band appears on the rostrum proximal to the red tip and separated from it by a white or colorless band. With the exception of the chromatophore numbers and arrangement, the prezoeal stages of the four species do not diflfer significantly. For this reason, a prezoea will be figured only Table 1. — Number of porcelain crab larvae dissected and examined for this study. Species Zoea I Zoea II Megalopa Petrolisthes cirutipes Pi'troIistkfS friomerus Pachychelfs pubescens Pachycheles rudis 11 4 2 16 4 9 9 6 9 12 12 7 193 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 2. — Prezoea, Zoea I, Zoea II, entire: A - prezoea, PetrolL.. inctipes with pri- mary chromatophores (Scale 3) ; B - prezoeal telson with modified cuticle, Pachycheles pubescens (Scale 4) ; C - Petrolisthes eriomerus, Zoea I with chromatophores (Scale 2) ; D - Pachycheles pubescens, Zoea II with chromatophores (Scale 1) ; E - Petrolisthes spp., Zoea II, ventral rostral bulge (Scale 3). Scales in millimeter. 194 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE once, and specific variations will be indicated in the text. Prezoeae were not dissected. ZOEA I (Figure 3) Antennule unsegmented with five or six ter- minal processes including three aesthetascs and two or three setae. Antenna biramous; endopodite fused with protopodite and bearing a terminal point and subterminal tubercle with a fine seta between; exopodite mobile, about li/> times as long as endopodite, with three or four prominent distal spines and one long seta proximal to the spines. Mandibles strongly sclerotized, heavily- toothed, asymmetrical appendages. Maxilla I with unsegmented endopodite bear- ing three or four terminal setae and a number of fine hairs along the anterior margin; basal endite with 9 or 10 stout spinous processes; coxal endite with 9 or 10 stout setae. Maxilla II with unsegmented endopodite bear- ing eight or nine setae, grouped 3, 1-2, 3-4 prox- imal to distal, 3-2-4 being the most common grouping. Basal endite bilobed: distal lobe with 7 to 10 setae; proximal lobe with seven to nine setae. Coxal endite bilobed: distal lobe with four to six setae; proximal lobe with 8 to 11 setae. Scaphognathite with six or seven long plumose marginal setae, and one apical seta; numerous fine hairs occur on scaphognathite margin between plumose setae and elsewhere on appendage as figured. Maxilliped I biramous. Coxopodite with one or two distal setae. Basipodite with 9 to 11 setae most commonly grouped 2, 2, 3, 3 proximal to distal. Endopodite four-segmented: segment 1 with one to three distal setae, inner margin; segment 2 with two or three distal setae, inner margin ; segment 3 with four to seven setae in- cluding two or three medial, one or two distal, inner margin, occasionally one long medial seta, outer margin; segment 4 with five to seven ter- minal setae and one proximal on outer margin. Exopodite two-segmented with four terminal natatory setae on distal segment. Additional groups of very fine hairs appear on endopodite segments 2, 3, and 4 and on segment 2 of exop- odite as figured. Maxilliped II biramous. Coxopodite lacking setae. Basipodite with three setae on inner margin grouped 1, 2 proximal to distal. Endop- odite four-segmented: segment 1 with one or two distal setae, inner margin; segment 2 with one or two distal setae, inner margin; segment 3 with one medial and one or two distal setae, inner margin; segment 4 with three to five (usu- ally five) terminal setae and one proximal seta, outer margin. Exopodite two-segmented with four terminal natatory setae on distal segment. Groups of very fine hairs appear on endopodite segments 2, 3, and 4 and on both segments of exopodite as figured. Maxilliped III present as small bilobed bud which undergoes growth throughout stage. Pereiopods present as five pairs of short limb buds, all without setae; none chelate; undergo growth during entire stage. Abdominal somites numbering five; segments 3, 4, and 5 with serrated dorsal posterior mar- gins; segments 4 and 5 each with a pair of strong ventrolateral spines on posterior margin. Telson with seven symmetrically arranged pairs of processes, with central pair located on central prominence ; outer margin to center line: one heavy, articulated lateral process with few spinules (not figured), one short fine seta, and five long plumose articulating setae; all long plumose setae armed distally with fixed curved spines (Figure ID); fine hairs (not figured) on margin of telson between all major plumose setae and on central prominence; anal spine present. Chromatophores as described in the prezoea; carapace 1.23 to 1.39 mm in length. ZOEA II (Figure 4) Antennule biramous; exopodite with six or seven terminal processes including four aesthet- ascs and two or three setae, followed by five tiers of subterminal aesthetascs grouped 2, 3, 3, 4, 3 progressing proximally; three or four fine setae present on distal margin of protopodite. Antenna biramous; endopodite same form as in zoea I; exopodite similar to zoea I, with three spines and one seta distally; exopodite and en- dopodite approximately equal in length. 195 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 3. — Petrolisthes cinctipes, Zoea I: A - antennule; B - antenna; C - mandible; D - maxilla I; E - maxilla II (A-E, Scale 2) ; F - maxilliped I; G - maxilliped II (F and G, Scale 1) ; H - maxilliped III and pereiopod buds (Scale 2) ; I - abdomen and telson, ventral (Scale 1). Scales in millimeter. 196 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE FiGT'RE 4. — Petrolisthes cinctipes, Zoea II: A - antennule with only one aesthetasc shown for each of the five subterminal tiers; B - antenna; C - mandible; D - maxilla I; E - maxilla II (A-E, Scale 2) ; F - maxilliped I; G - maxilliped II; H - maxilliped III and pereiopods; I - abdomen and telson (F-I, Scale 2). Scales in millimeter. 197 FISHERY BULLETIN: VOL. 71, NO. 1 Mandibles larger than in zoea I and with prominent palp bud. Maxilla I with unsegmented endopodite bear- ing four or five terminal setae and a number of fine hairs along anterior margin; basal endite with 10 or 11 setae; coxal endite with 11 to 13 setae. Maxilla II with unsegmented endopodite bear- ing nine setae grouped 3, 2, 4 progressing dis- tally. Basal endite bilobed; both distal and prox- imal lobes with 11 or 12 setae each. Coxal endite bilobed: distal lobe with seven setae; proximal lobe with 11 to 15 setae. Scaphognathite with 16 to 18 outer marginal setae, three apical setae and one on internal margin of posterior lobe (total 20 to 22 setae), all plumose. Fine hairs occur on scaphognathite margin between plu- mose setae and elsewhere on appendage as fig- ured. Maxilliped I biramous. Coxopodite with one or two distal setae. Basipodite with setae grouped 2, 2, 3, 3 proximal to distal. Endopodite four-segmented: segment 1 with three distal setae, inner margin; segment 2 with three distal setae, inner margin and one distal seta, outer margin; segment 3 with two medial and three distal setae, inner margin and one medial seta, outer margin; segment 4 with five or six ter- minal setae and one proximal seta, outer margin. Exopodite two-segmented with 14 natatory setae on distal segment. Maxilliped II biramous. Coxopodite lacks setae. Basipodite with three setae grouped 1, 2 proximal to distal. Endopodite four-segmented: segments 1 and 2 each with two distal setae, inner margin and one distal, outer margin; seg- ment 3 with one medial, two distal, inner margin and one medial seta, outer margin ; segment 4 with four to six (usually five) terminal setae and one proximal seta on outer margin. Ex- opodite two-segmented with 14 natatory setae on distal segment. Maxilliped III biramous, larger than in zoea I; endopodite curved anteriad; appendage grows and additional segments become defined during stage. Pereiopods present as five pairs of limb buds, pairs one and five distinctly chelate; buds ex- pand in size throughout stage and become dis- tinctly segmented in a late zoea II. Abdominal somites larger but otherwise sim- ilar to those in zoea I; pairs of pleopods, unequal in length, occur ventrally on segments 2, 3, 4, and 5; pleopods increase in length throughout stage. Telson similar to form in zoea I but with single median seta (Figures IE and 41) added to cen- tral prominence and size increased; anal spine present. Basic pattern of primary red chromatophores same as in prezoea and zoea I; however, new chromatophores appear in the late second stage zoea on the growing pereiopods and on body be- neath carapace. Carapace 1.72 to 1.98 mm in length. Rostrum bears distinct ventral bulge just anterior to eyes (Figure 2E). MEGALOPA (Figures 5 and 6) Antennule biramous with three-segmented peduncle. Dorsal ramus with six segments; segments 2, 3, and 4 each bear two tiers of aesthetascs and segment 5 bears one tier; num- bers of aesthetascs per tier are approximately 3-4, 2-3, 3, 2, 2-3, 2, 2 progressing distally (spec- imens damaged) ; one long seta present on each of segments 3 and 4 associated with the distal aesthetasc tier. Ventral ramus with three dis- tinct segments, the distal segment itself appear- ing indistinctly divided; setation as figured. Antennal flagellum usually composed of 17 segments; peduncle three-segmented; hairs and bristles associated with the distal end of each segment along the length of the flagellum vary in number and pattern of distribution. Mandible strongly sclerotized, partially cup- shaped appendage with bladelike leading edge which may be somewhat irregular but is not strongly toothed as in the zoea; three-segmented palp present with about six spinelike setae on terminal segment. Maxilla I with probably two-segmented endop- odite (specimens damaged) bearing one or more setae; basal endite with a total of 21 or 22 spines and setae arranged predominantly along the distal margin; coxal endite with approx- imately 26 processes, most of them slender setae in contrast to the spinelike processes of the basal 198 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Figure 5. — Petrolisthea cinctipes megalopa: A - antennule; B - antenna; C - mandible D - maxilla I; E - maxilla II; F - maxilliped I; G - maxilliped II; H - maxilliped III I - pereiopod III; J - pereiopod V (B, G, H, J, Scale 1; A, C, E, F, Scale 2; D Scale 3 I, Scale 4). Scales in millimeter. 199 FISHERY BULLETIN: VOL. 70. NO. 4 O CN O I- CN .S. to 15 pq E •- ftl 0) ^ tl I o •- , CO >i m ill X £ S *-> o ^- " <« O) +3 -— & c ?> £ o ^ 2 a § a. "i* I §o a: s" < ^ ^ V > M P.O - 9 O ^ o < E o .2 5 ^ "^ -^ ca 51..J-, CO H.^ ^ u a, s i. u o ^ a; a, >H •- 1 ^'-S I. T3 O a " u « C 3 ^ <= — U to 'C fe -2s £8 200 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE endite; a small fanlike exite also present, rimmed with very fine hairs. Maxilla II with unsegmented endopodite bear- ing five setae. Basal endite bilobed: distal lobe with approximately 26 setae; proximal lobe with 13 to 15 setae. Coxal endite bilobed: distal lobe with nine marginal setae and six submarginal setae, dorsal surface ; proximal lobe with about 20 marginal setae, 13 submarginals, dorsal surface, and 8 submarginal setae, ventral sur- face. Scaphognathite with approximately 50 marginal setae. Maxilliped I biramous. Exopodite devoid of setae or hairs in specimens examined (two) . Endopodite two-segmented in appearance, but indistinctly so; distal segment bears two short setae. Basal endite with 19 or 20 major mar- ginal setae, 5 to 8 submarginal setae, dorsal surface. Coxal endite with about seven major marginal setae and six submarginal setae, three on dorsal surface and three on ventral surface. Maxilliped II biramous. Exopodite consists of two distinct segments, the distal segment it- self appearing indistinctly segmented; there are five to seven plumose setae on the distal segment and five setae on the internal margin of the first segment. Endopodite four-segmented, the distal two segments each bearing dense brushes of ap- proximately 20 to 30 setae; segment 2 with four or five setae along distal margin; segment 1 with approximately 11 setae along the internal margin. Maxilliped III biramous. Exopodite incom- pletely formed with two segments only; segment 2 with one small proximal seta. Endopodite specialized for filter feeding and consists of five segments; segment 2 with 10 to 12 major setae, dorsal internal margin; segment 3 with 9 or 10 major feeding setae, 3 or 4 short, stout brushlike setae, and 2 or 3 very short, spikelike setae all on the dorsal internal margin; segment 4 with nine major feeding setae along the internal ventral margin, six or seven minor feeding setae along dorsal internal margin; four or five major brush setae and two spikes, all along internal dorsal margin ; segment 5 with six pairs of feed- ing setae and two minor brush setae terminally; other setation as pictured. Pereiopods well developed and functional. Chelipeds slender and streamlined, generally dorsoventrally flattened with fine bristles over entire surface; outer margin of chelae produced in a series of spines; anterior margin of carpus with one major spine and a fine bristle associated with it. Pereiopods 2, 3, and 4 similar to each other in shape and setation as figured (Figure 51). Pereiopod 5 chelate and armed with setae and bristles as shown; four to eight scythelike cleaning setae (Figure IL, N) are also present. Abdomen six-segmented; segments 2 through 5 each bearing one pair of pleopods. Pleopods biramous; exopodite with 16 plumose marginal setae; endopodite with two short setae and four hooks on margin. A single primary red chro- matophore located on segment 2 or betw^een 2 and 3. Segment 6 of the abdomen bears a pair of biramous uropods, the outer ramus usually hav- ing 14 and inner ramus with 15 marginal setae. Telson with 14 to 16 major plumose setae and 14 to 16 minor setae located between the major marginal setae. The dorsal surface of the telson bears a number of symmetrically placed pairs of fine hairs. In megalopae of advanced age, the beginning of the division of the telson into five plates can be seen beneath the cuticle. Up- on molting to the first crab stage, the division of the telson is complete and distinct. LARVAL DEVELOPMENT OF PETROLISTHES ERIOMERUS STIMPSON Larvae of P. eriomerus were reared in un- aerated Erlenmeyer flask cultures. PREZOEA The prezoea has essentially the same form as that described for P. cinctipes and diflfers only in one pair of primary chromatophores. P. eriomenis prezoeae lack the chromatophore on either side of the body between the bases of maxillipeds I and II which occurs in P. cinctipes larvae. 201 FISHERY BULLETIN: VOL. 71. NO. 1 ZOEA I (Figure 7) Antennule unsegmented; five to six terminal processes including four aesthetascs and one or two setae. Antenna biramous, with endopodite and pro- topodite fused; endopodite with terminal point, subterminal tubercle and fine seta between; ex- opodite approximately I14 times as long as en- dopodite with one, two, or three prominent distal spines and one long seta proximal to spines. Mandibles strongly sclerotized asymmetrical appendages, heavily armed with teeth and tu- bercles on leading edges. Maxilla I with unsegmented endopodite bear- ing three to six terminal and subterminal setae and a number of fine hairs along anterior mar- gin; basal endite with 9 or 10 setae; coxalendite with 8 to 10 setae. Maxilla II with unsegmented endopodite which bears 7 to 10 setae, grouped 2-3, 2, 3-4 progressing distally. Basal endite bilobed: dis- tal lobe with 8 to 10 setae ; proximal lobe with 6 to 10 setae. Coxal endite bilobed: distal lobe with four to eight setae; proximal lobe with 7 to 11 setae. Scaphognathite with six to eight long plumose marginal setae and one strong apical seta. Numerous fine hairs occur on scaphognathite margin between plumose setae and elsewhere on appendage as figured. Maxilliped I biramous. Coxopodite with two or rarely three distal setae. Basipodite with 10 setae usually grouped 2, 2, 3, 3 proximal to distal. Endopodite four-segmented: segment 1 with two or three distal setae, inner margin; seg- ment 2 with three distal setae, inner margin; segment 3 with usually two medial and three or rarely four distal setae, inner margin; seg- ment 4 with five to seven terminal setae and one proximal seta, outer margin; segments 2 and 3 also bear groups of fine hairs as figured. Ex- opodite two-segmented, the distal segment bear- ing four terminal natatory setae. Maxilliped II biramous. Coxopodite lacking setae. Basipodite with usually two but some- times one distal seta and one medial seta, inner margin. Endopodite four-segmented: segment 1 with usually two but sometimes with one distal seta, inner margin; segment 2 usually with two but rarely one or three distal setae, inner margin; segment 3 with one medial seta or occasionally two setae, and usually two, but sometimes three distal setae, inner margin; segment 4 with three to five terminal setae and one proximal seta, outer margin; segments 2 and 3 also bear groups of fine hairs as figured. Exopodite two-segment- ed, the distal segment bearing four terminal natatory setae. Maxilliped III a small bilobed bud which grows throughout stage. Pereiopod buds simple, five pairs in number, growing throughout the duration of the stage. Abdominal segments five in number; seg- ments 3, 4, and 5 having serrations on dorsal posterior margins; segments 4 and 5 with strong ventrolateral spines on posterior margin. Telson with seven symmetrically placed pairs of processes, with seventh pair on central prom- inence. Outer margin to center line: one heavy lateral spine fused with telson, one short fine seta, and five long plumose articulating setae, all armed distally with a series of short, curved fixed spines (Figure ID) ; numerous fine hairs on telson margin (not figured) between all major plumose setae; anal spine present. Chromatophore pattern as noted for the pre- zoea; extended rostrum with one terminal and one subterminal band of red color; posterior spines each with one terminal red band. Cara- pace 1.31 to 1.48 mm in length. ZOEA II (Figure 8) Antennule biramous; exopodite with six or seven terminal processes including four aesthet- ascs and two or three setae followed by five tiers of subterminal aesthetascs grouped 3, 4, 3, 3, 2 proximal to distal. Antenna biramous; endopodite same form as in zoea I; exopodite similar to that in zoea I with one to three spines, one seta distally; spines somewhat less prominent than in zoea I; exopo- dite approximately three-quarters length of en- dopodite. Mandibles larger than in zoea I and with prominent palp bud. 202 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Figure 7. — Petrolisthes eriomerus zoea I: A - antennule; B - antenna; C - mandibles; D - maxilla I; E - maxilla II; F - maxilliped I; G - maxilliped II; H - maxilliped III and periopod buds; I - abdomen and telson (A-E, H, Scale 2; F, G, I, Scale 1). Scales in millimeter. 203 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 8. — Petrolisthes eviomerus zoea II: A - antennule with only one aesthetasc shown for each of the five subterminal tiers ; B - antenna ; C - mandible ; D - maxilla I ; E - maxilla II; F - maxilliped I; G - maxilliped II; H - maxilliped III and pereiopods; I - maxilliped III, late zoea II; J - abdomen and telson (A, B, F-H, J, Scale 1; C-E, I, Scale 2). Scales in millimeter. 204 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Maxilla I with unsegmented endopodite bear- ing four to seven terminal and subterminal setae and a group of fine hairs along anterior margin; basal endite with 10 or 11 setae; coxal endite with 9 to 11 setae. Maxilla II with unsegmented endopodite bear- ing nine setae grouped 3, 2, 4 progressing dis- tally. Basal endite bilobed: distal lobe with 12 or 13 setae; proximal lobe with 11 or 12 setae. Coxal endite bilobed: distal lobe with five to seven setae; proximal lobe with 12 to 15 setae. Scaphognathite with 13 to 16 marginal, 3 apical, and 1 or 2 internal lateral setae (total 17 to 21) ; fine hairs present on scaphognathite margin be- tween plumose setae and elsewhere on append- age as figured. Maxilliped I biramous, Coxopodite with two distal setae. Basipodite with setae usually- grouped 2, 2, 3, 3 proximal to distal along inner margin. Endopodite four-segmented: segment 1 with two or three distal setae, inner margin; segment 2 with three or four distal setae, inner margin and one distal seta, outer margin; seg- ment 3 usually with two medial and three distal setae, inner margin and one medial, outer mar- gin; segment 4 with four to seven terminal and one proximal seta, outer margin. Exopodite two- segmented, the distal segment bearing 14 nata- tory setae. Maxilliped II biramous. Coxopodite lacking setae. Basipodite with three distal setae grouped 1, 2 proximal to distal. Endopodite four-segmented: segment 1 with two (occasion- ally one) distal setae, inner margin; segment 2 with two distal setae, inner margin and one dis- tal setae, outer margin; segment 3 with one medial, two or rarely three distal, inner margin, and one medial seta, outer margin; segment 4 with five terminal setae and one proximal seta, outer margin. Exopodite with two segments, the distal segment bearing 14 natatory setae. Maxilliped III biramous. Exopodite indis- tinctly divided with two segments; zero to three setae have been observed on endopodite. Endop- odite with up to five segments apparent beneath cuticle, depending on the age of the zoea. Pereiopod buds five pairs in number with first and fifth pairs distinctly chelate; buds increase in length throughout the stage. Abdominal somites numbering five, similar to form in zoea I but larger and with paired ple- opod buds of unequal length on segments 2 through 5; pleopods increase in length through- out the stage. Telson similar to form in zoea I but larger and with one unpaired median seta on the cen- tral prominence (Figures IE and 8J). Pattern of primary chromatophores same as that in zoea I but new ones may be added on the growing pereiopods. Carapace 1.80 to 1.89 mm in length. Rostrum bears distinct ventral bulge just anterior to eyes. MEGALOPA (Figures 6 and 9) Antennule biramous with three-segmented peduncle. Dorsal antennular ramus six-seg- mented; segments 2, 3, 4, and 5 with 2, 2, 2, 1 tiers of aesthetascs respectively; tiers, progres- sing distally, contain approximately 6-7, 6-7, 6-7, 3-4, 3, 2, 3 aesthetascs respectively; one long slender seta associated with the distal tier on each of segments 3 and 4. Ventral ramus with three segments ; setation as figured. Antenna long and slender with three-seg- mented peduncle; flagellum with 28 to 30 seg- ments; most segments armed distally with var- iable numbers of short bristles and hairs as figured. Mandible differs from toothed form in zoeal stages and is partially cup-shaped with slightly irregular bladelike leading edge; three-segment- ed palp has seven to ten spines on distal segment. Maxilla I with indistinctly two-segmented en- dopodite bearing a single terminal seta on the distal segment and a single seta situated at base; basal endite with a total of about 25 stout spines and setae; coxal endite with a total of 32 stout setae. In addition, there is a single long seta proximally on the exopodite and a delicate exite edged with fine hairs associated with coxal por- tion of appendage. Maxilla II with slender, indistinctly segmented endopodite bearing three terminal setae and one 205 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 9. — Petrolisthes eriomerus megalopa : A - antennule ; B - antenna ; C - mandible ; D- maxilla I; E- maxilla II; F - maxilliped I; G - maxilliped II ; H - maxilliped III ; I - pereiopod III; J - pereiopod V (B, G, H, Scale 1; A, C-F, J, Scale 2; I, Scale 3). Scales in millimeter. 206 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE lateral seta. Basal endite bilobed: distal lobe with 32 to 34 setae; proximal lobe with about 19 setae. Coxal endite bilobed: distal lobe with about 19 setae; proximal lobe with about 39 setae. Maxilliped I with exopodite bearing three mar- ginal setae; endopodite with two terminal setae in specimens dissected; basal portion of protop- odite produced into triangular lobe bearing 35 to 42 setae; coxal lobe with numerous setae num- bering 16 to 21 in specimens counted. Maxilliped II with two-segmented exopodite usually bearing six setae on terminal segment and six internal marginal setae on proximal seg- ment. Endopodite composed of four segments the distal two of which are armed with dense setal brushes each containing about 16 to 20 setae. Other setation as figured. Maxilliped III with single-segmented exop- odite bearing two very fine setae. Endopodite with five well-developed segments; segments 2 through 5 armed with filtering setae; segment 2 with 13 major setae along dorsal inner margin; segment 3 with three major ventral and four or five major dorsal internal marginal feeding setae, seven or eight minor submarginal brush- like setae dorsally and five ventral major sub- marginal brush setae ; segment 4 with 10 major ventral marginal feeding setae, 9 minor dorsal marginal feeding setae, protuberance distally and dorsally on segment bearing a row of 5 major brush setae and a subordinate row of 4 spikelike setae and 1 fine seta; segment 5 with five major ventral and four or five dorsal (mixed major and minor) feeding setae, two slender terminal brush setae; three major brushes and two spikes on dorsal, submarginal bulge. Pereiopods well developed and functional. Chelae slender, narrow, dorsoventrally flattened with two fixed spines present on internal margin of carpus. Walking legs all similar in form and setation as figured. Pereiopod 5 chelate with bristles and hooked setae for cleaning as figured. Abdomen segments numbering 6, segments 2 through 5 with paired biramous pleopods; mar- ginal setation of pleopod exopodites varies from 10 to 13, the higher numbers generally being found on the more proximal pleopods; each pleopod endopodite is armed with four small hooks and two setae. Segment 2 bears one chromatophore distally. A pair of biramous uropods articulate with segment 6; the outer ramus with 15 or 16 marginal setae, inner ramus with 10 or 11 marginal setae. Telson with 15 or 16 major marginal plumose setae and usually four minor setae placed ap- proximately symmetrically with respect to the center line of the telson between the major marginal setae; dorsal surface of telson bears about 10 pairs of fine setae in approximately symmetrical positions. LARVAL DEVELOPMENT OF PACHYCHELES PUBESCENS HOLMES Larvae taken from the plankton were reared in unaerated Erlenmeyer flask cultures and identified at a later date by comparison with laboratory hatched larvae. PREZOEA The general body form of Pachycheles puhes- cens is the same as that described for Petrolis- thes cinctipes prezoeae; however, certain details of the prezoeal cuticle of the telson diff"er from P. cinctipes. The prezoeal cuticle of Pachycheles pubescens is produced into flat spines with toothed margins (Figure 2) , an additional adap- tation for swimming by abdominal flexion. Chro- matophores occur as follows: one on either side of the mouth; one each on abdominal segments 1, 2, 3, and 5; and one on the telson. ZOEA I (Figure 10) Antennule unsegmented with five or six ter- minal processes including three or four aesthet- ascs and one or two setae. Antenna biramous with endopodite and pro- topodite fused ; endopodite without subterminal tubercle or fine seta; exopodite 1% to 2 times as long as endopodite; usually three short stout, curved spines along internal lateral margin dis- tally on exopodite, but spines occasionally num- ber one or two and rarely four. 207 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 10. — Pachycheles pubescens, zoea I: A - antennule; B - antenna; C - mandibles; D - maxilla I; E - maxilla II; F - maxilliped I; G - maxilliped II; H - maxilliped III and pereiopod buds; I - abdomen and telson (A-E, H, Scale 2; F, G, I, Scale 1). Scales in millimeter. 208 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Mandibles strongly toothed, heavily sclero- tized asymmetrical appendages. Maxilla I with unsegmiented endopodite bear- ing three to five terminal setae and a number of fine hairs along anterior margin ; basal endite with eight or nine setae; coxal endite with seven to nine setae. Maxilla II with unsegmented endopodite bear- ing eight or nine setae grouped 3, 2, 3-4 pro- gressing distally. Basal endite bilobed: distal lobe with six to eight setae; proximal lobe with six to eight setae. Coxal endite bilobed: distal lobe with three to five setae; proximal lobe with six to nine setae. Scaphognathite with seven long plumose marginal setae and one apical seta on posterior lobe. Numerous fine hairs occur on scaphognathite margin between plumose setae and elsewhere on appendage as figured. Maxilliped I biramous. Coxopodite usually with two distal setae. Basipodite usually with nine setae arranged 2, 2, 2, 3 proximal to distal along inner margin. Endopodite four-segment- ed: segment 1 with two or three distal setae, inner margin; segment 2 with three distal setae, inner margin; segment 3 usually with one, some- times two medial setae, and usually three, some- times two or four distal setae, inner margin; segment 4 with seven to nine terminal setae and one long proximal seta, outer margin. Exopo- dite two-segmented with four natatory setae on distal segment. Maxilliped II biramous. Coxopodite lacking setae. Basipodite with one medial, two distal setae, inner margin. Endopodite four-segment- ed: segment 1 with two or three distal setae, inner margin; segment 2 with two distal setae, inner margin; segment 3 with one medial and one or two distal setae, inner margin; segment 4 with five or six terminal and one proximal seta, outer margin. Exopodite two-segmented with four terminal natatory setae. Maxilliped III present as small bilobed bud which increases in size during the stage. Pereiopod buds numbering five pairs, simple; none chelate; buds increase in size throughout stage. Abdominal somites five in number; segments 4 and 5 with prominent paired ventrolateral spines on posterior margin; segments 2 through 5 with serrated dorsal posterior margins. Telson with seven symmetrically arranged pairs of processes with seventh pair on central prominence. Outer margin to center line: one heavy lateral spine fused with telson; one short, fine seta; five long plumose articulating setae, the outer two armed distally with fixed, curved spines (Figure lA, B); fine hairs on margin of telson (not figured) between all major plu- mose setae and on central prominence. Chromatophore pattern same as that de- scribed for prezoea. Rostrum commonly with three bands of red, one terminal and two sub- terminal, occasionally with only one subter- minal band; both posterior spines tipped with red. Carapace 1.31 to 1.56 mm long. ZOEA II (Figure 11) Antennule biramous; exopodite with six or seven terminal processes including three or four aesthetascs and two or three setae followed by five tiers of aesthetascs grouped 2, 3, 3, 4-5, 4-6 proceeding proximally. One long seta projects from distal portion of protopodite. Antenna biramous; endopodite pointed ter- minally and armed subterminally with one fine seta and two small spines; exopodite unarmed and about two-thirds length of endopodite. Mandibles increased in size; prominent palp bud present. Maxilla I with unsegmented endopodite bear- ing three or four terminal setae and a group of fine hairs along anterior margin; basal en- dite with 9 or 10 heavy spinous setae; coxal endite with 8 to 11 heavy spinous setae. Maxilla II with unsegmented endopodite bear- ing nine setae. Basal endite bilobed: distal and proximal lobes each with 9 to 11 setae. Coxal endite bilobed: distal lobe with four to seven setae; proximal lobe with 10 to 13 setae. Scaph- ognathite with 18 setae on outer margin and 3 strong apical setae on posterior lobe. Numerous fine hairs occur on scaphognathite margin be- tween plumose setae and elsewhere on appendage as figured. 209 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 11. — Pachycheles pubescens, zoea II: A - antennule with only one aesthetasc shown for each of the five subterminal tiers; B - antenna; C - mandible; D - maxilla I; E - maxilla II; F - maxilliped I; G - maxilliped II; H - maxilliped III, late zoea II; I - pereiopods; J - abdomen and telson, early zoea II (A, B, F-J, Scale 1; C-E, Scale 2). Scales in millimeter. 210 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Maxilliped I biramous. Coxopodite usually with two distal setae. Basipodite usually with nine setae along inner margin grouped 2, 2, 2, 3 proximal to distal. Endopodite four-segment- ed: segments 1 and 2 each usually with three distal setae (segment 2 occasionally with four), inner margin and one distal seta, outer margin; segment 3 most commonly with one but occasion- ally two medial, two to four distal setae, inner margin and one medial seta, outer margin; seg- ment 4 with 8 to 11 terminals and 1 proximal seta, outer margin. Exopodite two-segmented with 14 natatory setae on distal segment. Maxilliped II biramous. Coxopodite lacking setae. Basipodite usually with three setae grouped 1, 2 proximal to distal. Endopodite four-segmented: segment 1 with two or three distal setae, inner margin and one distal seta, outer margin; segment 2 with one or two distal setae, inner margin and one distal seta, outer margin; segment 3 usually with one (occasional- ly none) medial, two distal setae, inner margin, and one medial seta, outer margin; segment 4 with five terminals and one proximal seta, outer margin. Exopodite two-segmented with 14 nata- tory setae on distal segment. Maxilliped III a bilobed bud. Exopodite in- creases in size throughout stage; four to six setae present; one or two definite segments present beneath cuticle, depending on age of zoea. Endopodite entire or segmented with up to five segments beneath cuticle, depending on age; setae of megalopa stage visible beneath cuticle of advanced zoea. Pereiopod buds enlarge during stage and be- come well developed; first and fifth pairs dis- tinctly chelate. Spines and claws appear on pereiopods in advanced, premolt larva. Abdominal somites larger than in zoea I but similar in form; segments 2, 3, 4, and 5 each bear a pair of pleopods of unequal length. Telson increased in size; an unpaired median seta added on central prominence (Figures IC and IIJ); anal spine present. Basic color pattern same as in zoea I; red chromatophores added to pereiopods and beneath carapace in zoeae of advanced age. Carapace 2.37 to 2.54 mm in length. MEGALOPA (Figures 12 and 13) Antennule biramous, consisting of a three- segmented peduncle, a dorsal ramus with six seg- ments, and a ventral ramus with three segments. Segments 2, 3, and 4 of dorsal ramus each bear two tiers of aesthetascs; segment 5 with only 1 aesthetasc tier. Numbers of aesthetascs per tier proximal to distal 7-8, 8, 6-7, 5, 4, 2, 3. One long seta associated with distal tier of aesthetascs on segments 2, 3, and 4. Other setation as figured. Antennal flagellum composed of 29 to 30 seg- ments in addition to the three-segmented pe- duncle; most segments bear variable numbers of fine bristles around the distal margin as fig- ured. Mandible highly sclerotized with three-seg- mented palp; terminal segment of palp with about 17 spines; grinding edge only slightly ir- regular with a generally bladelike form. Maxilla I with a two-segmented endopodite bearing a single distal seta on terminal segment. Basal endite with 7 slender setae and 14 heavy spinelike setae along distal margin; also seven short, stout, spinelike setae submarginally on dorsal surface of segment. Coxal endite with four setae on dorsal surface and about 29 mar- ginal and near submarginal setae; delicate fan- like coxal exite rimmed with hairs is present. Maxilla II with unsegmented endopodite which bears five or six setae. Basal endite bi- lobed: distal lobe with 33 to 36 setae; proximal lobe with about 22 setae. Coxal endite bilobed: distal lobe with about 20 setae; proximal lobe with about 38 setae. Scaphognathite with about 68 marginal plumose setae and 5 short fine hairs on the surface of the fan. Maxilliped I with exopodite bearing nine setae. Endopodite indistinctly two-segmented, bearing one seta on distal segment. Protopodite with about 52 setae. Coxal endite with about 23 setae. Maxilliped II biramous; exopodite indistinctly two-segmented with eight or nine terminal setae; endopodite four-segmented, segments 3 and 4 having heavy terminal brushes of about 26 and 30 setae respectively ; other setation as figured. 211 FISHERY BULLETIN: VOL. 71, NQ. 1 Figure 12. — Pachycheles pubescens, megalopa: A - antennule; B - antenna; C - man- dible; D - maxilla I; E - maxilla II; F - maxilliped I; G - maxilliped II; H - max- illiped III; I - pereiopod III; J - pereiopod V (A, C-F, J, Scale 2; B, G, H, Scale 1; I, Scale 3). Scales in millimeter. 212 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Figure 13. — Pachycheles pubescens (A-C) and P. rudis (D-F) megalopae, whole mounts : A - P. pubescens megalopa, whole mount; B - telson and uropods; C - right third pleopod; D - P. rudis megalopa, whole mount; E - telson and uropods; F - right third pleopod (A, D, Scale 1; B, C, E, F, Scale 2). Scales in millimeter. 213 FISHERY BULLETIN: VOL. 71, NO. 1 Maxilliped III biramous; exopodite two-seg- mented, terminal segment being incomplete and lacking setae. Endopodite with five segments; segment 2 with 13 major marginal setae in double row (7, 6); segment 3 with five or six slender minor setae, outer margin, six major in- ner marginals, and nine brush setae, inner mar- gin; segment 4 with five major and three spike- like brush setae, inner margin, 11 outer major marginals, and 7 inner minor auxiliaries; seg- ment 5 with five pairs of major feeding setae, one or two pairs of minor terminal setae. Pereiopods well developed and functional. Chelipeds large, swollen and bristly; two or three prominent fixed spines on the anterior margin of carpus. Three pairs of walking legs similar in form and setation as figured. Perei- opod 5 with 8 to 10 sickle-shaped setae and a number of other bristles. Abdominal segments six in number, segments 2 through 5 with paired biramous pleopods bear- ing setae as follows: 16 marginal setae on ex- opodite; endopodite armed with four small hooks and two setae. Chromatophores appear poster- iorly in segments 1, 2, 3, 5, and 6. Segment 6 bears a pair of biramous uropods, the outer rami each with about 19 plumose marginal setae and the inner with 17 or 18 such setae. The telson has 15 or 16 major marginal setae. Minor marginal setae are present between al- most all major setae. The dorsal surface of the telson is equipped with a number of very fine short hairs arranged in approximate symmetry as figured. Frontal margin of carapace as fig- ured (Figure 13A). LARVAL DEVELOPMENT OF PACHYCHELES RUDIS STIMPSON In the laboratory, only a single P. rudis larva reared at 15°C and 33;^f salinity reached the megalopa stage. Laboratory-reared material was extensively supplemented with larvae from the plankton for purposes of examination. Knight (1966) has dissected and accurately described the two true zoeal stages of P. rudis. She also adequately described the gross external morphology of the megalopa stage. Owing to high mortality in her cultures, she was forced to describe the larvae of this species on the basis of very few specimens. In this study, numbers of laboratory-hatched larvae surviving each stage in cultures were comparable to numbers obtained by Knight (1966), but the various stages were extremely abundant in the plankton. Knight's description of the species is expanded here using this more plentiful material. PREZOEA Body form is the same as that described for Petrolisthes cinctipes (Figure 2). Chromato- phores occur as follows: one on either side of the mouth; one posteriorly in abdominal seg- ment 2 or between segments 2 and 3. This is the least colorful of the species discussed here. Knight (1966) does not mention a prezoeal stage. ZOEA I AND II Table 2 indicates points in which the larvae studied here diff"er from those described by Knight (1966). Diff"erences are minor but are included to establish a more accurate range of variability for this species. MEGALOPA (Figures 13 and 14) Antennule biramous with three-segmented peduncle. Dorsal ramus with six segments, seg- ments 2 through 5 bearing 2, 2, 2, and 1 tier of aesthetascs respectively. Aesthetascs are ar- ranged 6, 6-7, 5-6, 3-4, 3-4, 2, 3 in tiers pro- gressing distally. A single long plumose seta is associated with the distal tier on segments 2, 3, and 4. Ventral ramus distinctly three-seg- mented, the most distal segment being indis- tinctly divided. Other setation and spination as figured. Antenna long, slender, with three-segmented peduncle; flagellum with 20 to 21 segments ; var- iable numbers of fine hairs and bristles arranged distally on most segments as figured. Mandible strongly sclerotized, partially cup- shaped appendage with three-segmented palp; distal segment of palp with about 13 terminal spines as figured. 214 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE Table 2. — Comparison of species descriptions of Pachycheles rudis. S^age Appendage Knight (1966) This study Zoea 1 Antennule 6 processes including 6-7 processes including 3 aesthetascs, 2 setae 4 aesthetascs, 1-2 setae Zoea 11 Antennule 7 processes, including 6-7 processes including 3 aesthetascs, 2 setae 4 aesthetascs, 2-3 setae Zoea II Maxilla 1: Coxal endite 5 spines, 6 setae 10 setae Zoea II MaxiMa II: Basal endite, distal lobe 10-12 setae 11-12 setae Basol endite, proximal lobe 10 setae 10-11 setae Coxal endite, distal lobe 6 setae 6- 7 setae Coxal endite, proximal lobe 10-12 setae 9-10 setae Zoea II Maxlilliped 1: Bosipodite Setae grouped 2, 2, 3, 3 Setaa grouped 2, 2, 2, 3 Endopodite; segment 3 2 medial setae 1-2 medial setae Endopoditei segment 3 3 distal setae 3-4 distal setaei Endopodite; segment 4 7-8 terminal setae^ 8-9 terminal setae^ Zoea II Maxilliped II: Endopodite; segment 1 2 distal setae. Inner margin 2-3 distal setae, inner margin Zoea II Maxilliped III: 2 slender setae 2 slender setae, and 2-3 fine Exopodite setae All larval stages Abdominal segment 2 or between 2 and 3 — 1 chromatophore Either side of mouth — 1 chromatopihore ^ Most common occurrence in bold face. Maxilla I with indistinctly two-segmented endopodite bearing two stout setae on distal seg- ment. Basal endite with 8 long, slender and 16 stout, spinous setae on margin and 6 short setae on ventral surface. Coxal endite with about 26 marginal setae and 4 fine setae on ventral surface, Coxal exite present as delicate fleshy lobe rimmed with fine hairs. One proximal seta on appendage near articulation with gnathal skeleton. Maxilla II with large numbers of setae. En- dopodite unsegmented with five setae. Basal endite bilobed: distal lobe with about 35 setae; proximal lobe with 9 or 10 marginal and 4 sub- marginal setae. Coxal endite bilobed: distal lobe with 20 to 21 setae arranged as figured; proximal lobe with 30 to 35 setae arranged as figured. Scaphognathite with about 58 setae around the margin and 2 fine submarginal setae on the anterior lobe. Maxilliped I with numerous setae. Exopodite fleshy with one seta on margin. Endopodite in- distinctly two-segmented, bearing a single seta proximally. Protopodite produced into trian- gular lobe with about 35 to 45 marginal setae and 3 proximal setae as figured. Coxal lobe with a total of about 18 setae arranged as fig- ured. Maxilliped II biramous. Setation of coxopo- dite and basipodite variable, approximately as figured. Exopodite two-segmented with setae as figured, with eight or nine setae on terminal segment. Endopodite with four segments; seg- ments 3 and 4 having distal brushes of short setae numbering about 24 to 28 and 16 to 20 respectively. Other setation as figured. Maxilliped III biramous. Exopodite two-seg- mented but incompletely formed. Segments 2 through 5 of endopodite equipped with major feeding setae; segment 2 with 11 or 12 feeding setae of two lengths, dorsal margin; segment 3 with five or six feeding setae on ventral margin, seven feeding setae or dorsal margin, and seven brush setae of varying lengths; no spikelike setae on dorsal protuberance; segment 4 with 2 short setae, 5 major brushes and 4 spike setae on dorsal protuberance, 10 long feeding setae on ventral margin, and 9 scattered minor sub- marginal setae on ventral surface; segment 5 with 10 or 11 marginal setae and 2 or 3 short terminal setae. Other setation as figured. Details of setation and form of the walking 215 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 14. — Pachycheles rudis megalopa: A - antennule; B - antenna; C - mandible; D- maxilla I; E- maxilla II; F - maxilliped I ; G - maxilliped II; H - maxilliped III; I - pereiopod III; J - pereiopod V (A, C-F, Scale 2; B, G, H, J, Scale 1; I, Scale 3). Scales in millimeter. 216 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE legs, the fifth pereiopods, the telson and sixth segment, and the frontal margin of the carapace are as figured. DISCUSSION The four species whose larvae are described here are the only porcellanids known (Haig, 1960) on the Pacific coast of North America north of Bodega Head, Calif., and we have found their larvae simultaneously in the plankton off the central Oregon coast. Both preserved and live zoea larvae of these four species can be read- ily assigned to genus on the basis of the number of long telson setae bearing conspicuous distal spines and the nature of these spines (Table 3) . Once live or freshly killed zoeae are separated by genus using telson characters, individuals of the congeneric species can be separated to species on the basis of the distribution of the primary red chromatophores. Zoea larvae from pre- served plankton samples cannot be identified to species easily once the chromatophores have faded. The only other nonvariable character found which distinguishes the species was max- illipedal setal counts, and it is possible that further study will indicate that these counts are also variable. The two Pachycheles species diflFer in both zoeal stages by a single seta on the inner margin of segment two of the endopodite on maxilliped I (Table 4). In addition, second zoea larvae of these two species differ by a single seta on the outer margin of endopodite segment 1 on max- illipeds I and II (Table 4). Similar characters separate the larvae of the two Petrolisthes spe- cies. First zoeae differ by one seta on the inner margin of segment two of the endopodite of both maxillipeds I and II (Table 5). Second zoeae of these Petrolisthes differ by a single seta on the outer margin of the first segment of the endopodite on maxilliped II (Table 5). Mega- lopae of all four species can be readily differen- tiated on the basis of cheliped form and other finer characters (Figures 12-14). Available knowledge of the larvae of Pach- ycheles species does not support Haig (1960) who, studying adult animals, suggested that P. rudis is most closely related to P. stevensii. A comprehensive comparison (Table 4) of the four most similar larvae of the Pachycheles species studied thus far does not definitely indicate such a relationship. On the basis of larval form and setal numbers, all four of the species are equally similar. At this time too few porcellanid larvae have been described to use larval characters to draw firm conclusions about generic and specific rela- tionships. There is however a need for means to identify as closely as possible porcellanid zoea larvae taken from the plankton in regions where larvae of all of the porcellanid species have not been reared. For all known porcellanid larvae of established specific identity, we have listed in Table 3 three morphological characteristics which appear to assign porcellanid zoeae to ge- nera or generic groups. The following discus- sion evaluates the usefulness of these charac- teristics in grouping presently known larvae systematically. Planktonic larvae of uncertain adult origin are included and grouped with known larvae they most closely resemble, but are not discussed further. Two New Zealand porcellanids for which larvae have been de- scribed, Petrolisthes novaezelandiae and P. elon- gatus, have some unusual characteristics in both the larval and adult forms which distinguish them from all other porcellanids known. For this evaluation, they are conditionally placed with groups with which they share the larval characters listed in Table 3. All presently known porcellanid zoeae can be placed in one of three groups on the basis of telson form. Lebour (1943) originally estab- lished two of these groups (A and B, Table 3) by distinguishing larvae of the genus Porcellana from larvae of the genus Petrolisthes as the gen- era were known at that time. Lebour's orig- inal Petrolisthes type telson group (B, Table 3) now includes all described species of the genera Pachycheles and Megalobrachium and all but one of the described Petrolisthes larvae, P. nov- aezelandiae. Wear (1964a) has suggested that the exceptional P. novaezelandiae be placed in the genus Pisidia on the basis of both larval and adult characters. Lebour's original Porcellana type telson group (A, Table 3) now includes all described larvae of the genera Euceramus, Poly- 217 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — Comparison of porcellanid larvae on the basis of distal armature of telson processes and telson form.^ Positiorv of armed Species telsoiT processes Type of armature Telson form Author Zoea 1 Zoea II t/ll ^tsiuc Pachycheles rudis 3,4 3,4 All Pachycheles spp. have B Knight, 1966 PachycheUs pubtscens 3,4 3,4 equal number of spines on inner and outer margins. B This study PachychtUs stevensii 3,4 3,4 B Kurata, 1964 PachycheUs haigae 3,4 3,4 B Boschi et al., 1967 Pachycheles natalensis 3-7 ? Spines most prominent on 5tK process. B Sankolli, 1967 Megalobrachium poeyi B Gore, 1971 Petrotisthes cinctipes 3-7 3-7 Spines, equal number B This study Petrolisthes eriomerus 3-7 3-7 inner and outer margins (first 5 species listed). B This study Petrolisthes armatus 3-7 3-7 B Labour, 1943, 1930 Gurney, 1938 Gore, 1970, 1972 Petrolisthes lamarckii 3-7 ? Spines most prominent on 5t4i process. B Sankolli, 1967 Petrolisthes boscii 3-7 3-7 Spines most prominent on 5th process. B Shenoy and Sankolli, 1967 Petrolisthes rujescens ? ? Not mentioned. B Gohar and Al-Kholy, 1957 "Porcellanella" sp. ? ? I^ot mentioned. B Menon, 1937 Petrolisthes elongatus 3-7 3-7 6 prominent spines, inner margin; less prominent spines, outer margin. B Wear, 1964b Greenwood, 1965 Petrocheles spinosus 4.5 4,5 6-8 spines, inner margin. C Wear, 1966 Petrolisthes novaexe- 3 3,4 5-6 spines, inner margin. A Wear, 1964ci landiae Greenwood, 1965 t Petrolisthes sp. 1 ? ? Not mentioned. A Menon, 1937 ^Petrolisthes sp. 1 1 ? ? Not mentioned. A Menon, 1937 PoTCellana platychetes 3 3 Long fine spines inner margin only (?) A Lebour, 1943 Porcellana ornata 3-7 ? "Hooks" minute, most prominent on 3rd process. A Sankolli, 1967 Porcellana sigsbeiana 3-7 3-7 Finely dentate A Gore, 1971b Porcellana sayana 3-7 ? According to figures, tips Brooks and Wilson, 1883 (as P. ocellata) ore similar, smooth, or finely serrated(?) Pisidia tongicomis 3 3 Well-defined spines, equal number inner and A Lebour, 1943 Sars, 1869 Pisidia bluteli 3 3 outer margins (first 3 apecies 'listed). A Bourdillon-Casanova, 1956 Pisidia inaequalis 3 3 A Gurney, 1938 Pisidia spinulijrons 3 ? Minute "hooks", more developed on outer than on inner margin. A Sankolli, 1967 Polyonyx gibbesi A Gore, 1968 Polyonyx guadriungutatus 3 3 More spines on outer than on inner margin. A Knight, 1966 Polyonyx hendersoni 3-7 ? Minute spines, most prominent on 3rd process. A Sankolli, 1967 Euceramus praetongus 3,4 3,4 Well-defined spines, equal number inner and outer margins. A Roberts, 1968 1 A = Lebour's (1943) definition of Porcellana telson. Telson about IV2 times as wide as long with seventh pair of processes situated outside and below central prominence in stage I. In stage II, an eighth pair of processes is added on the central prominence. B = Leb>our's (1943) definition of Petrolisthes telson. Tefson about as long as wide with seventh pair of processes stiuated on central prominence in stage I. In stage II, a single median spine (or setae) added C = Fits neither A nor B exactly. onyx, Porcellana, and Pisidia. A third telson type (C, Table 3) has since been described and to date includes only the larvae of Petrocheles spinosus from New Zealand. Among larvae possessing a similar telson form, further systematic separation following generic lines appears to be possible on the basis of the position of the long telson setae (processes 3 through 7) armed distally with conspicuous spines. This is particularly true of 13 larvae of known adult origin possessing the Petrolisthes type telson. Of these 13 species, only the larvae attributed to Pachycheles natalensis differ from other larvae of the same genus. The known larvae possessing the Porcellana type telson are more difficult to group using tel- son seta armature alone. Only larvae of the single known species of Euceramus can be dis- tinguished from other known larvae of this telson type on the basis of setal armature. In 218 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE _c — (X. « ■<§ ^ c UCO E o * D "lO o 2 £V o 01 D ■D Q. a < o ■^CM CO — ■D o E K d) II) O c C — O - G£ O a o O z t^ ~z il a o O ^> "2 •o 3 -0 TJ D 3 a O — ^ 00 C>> CO o o "? o CM o Q o 0) , c lU p Z lU O UJ D CM CO ■* o CO- CO o -■■ o CM ■* ■O CM- CM -- CM ■D % V) CM 1 -6 >o ■o c o o SI Jl u •; p> a lo a "? CM o- CM h-T _- b |-8 •o 00 N. N. o CM ■>f — CM CO- ^'■ '5 a ^8 2 t< "O •o CO 00 CN CM CM 2 cm" — o — ■o V to a CM r^ Ul D CO ^ . o- ?j CM CM CM ■<)■ CO CO - o CM 2 CM -- _o '5 a c D o o> a 0) o in a ,2 o CM o o ^ c CO c>. ^ £ o N. ,. lO — CM \t C u» e^ — s S CM CM - CM CO cm" — - •D 1 T5 c o c7)"a S8 to 2 a. >6 2 O (S 6 CM CM CM ■ CO <) ^ CO IV -<*■ rx rv o 1 3 0) o a; o O 5 o — _ a o o c U m lu O. Q Q. Q ^ ^ ^ ^ * 4> ID S £ "S a a o n o o X X b; i; T) o o o o c tj U CO CO UJ O z £CM o . C CM Tt Z o I- z UJ 't Q- CM- O •o _ CO CO CM t3 3 tN Q CO O CO C3 "9 « •o T3 ■o 3 3 ^ J3 CO CM-^^ "8 L- JD ^'"^ _0 a CM O 15 lO u "9 'a CO cm" + CM CM- d ■o CM CM Tf co" d -o ■D 3 3 lO — ^ ^ CO d "S o J3 '5 CM O a •- * o lll o a a X "^ o O o X (J tia UJ c £ 'a P fe 5 c .•S E E = a o c - O = •^ (D If (5 a X ';;; o □ U CD T) o a o ■D I = c ° H t o (D a "o to c 10 "O Eo 0) a I/) 0) ^ •D c 3 a a E o o ^ .0 'x o "5 2 5 a. < a 2 u SI ° oi S} -5-2 E- xS 12 aT EQ 219 FISHERY BULLETIN: VOL. 71, NO. 1 Table 5. — Comparison of maxilliped setation in Petrolisthes cinctipes and P. eriomerus. (All serial listings are numbers of setae, arranged proximally to distally.) Zoea 1 1 Zoea II Appendage P. cinctipes P. friomerus P. cinctipes P. eriomerus Maxiilliped 1: Coxopodite 1-2 2 1-2 2 Basipodite 2, 2, 3, 3 2, 2, 3, 3 2, 2, 3, 3 2, 2, 3, 3 Exopodite 4 4 M 14 Endopodite: Inner margin 1-3, 2-3, 4-7, 5-7 2-3, 3, 5-6, 5-7 3, 3, 5, 5^5 2-3, 3-4, 5, 4-7 Outer margin 0, 0, 0-1, 1 0, 0, 0, 1 0, 1, 1, 1 0, 1, I, 1 Maxilliped II: Coxopodite Basipodite 1, 2 1, 1^ 1, 2 I, 2 Exopodite 4 4 14 14 Endopodite: Inner margin 1-2, 1-2, 2^, 3-5 1-2, 1-3, 3-4, 3-5 2, 2, 3, 4-6 1-2, 2, 3-4, 5 Outer margin 0, 0, 0, 1 0, 0, 0, 1 1, 1, 1, 1 0, 1, 1, 1 contrast, larvae belonging to species of the present genera Poixellana, Pisidia, and Polyonx exhibit several different patterns of armed tel- son setae, a fact which, when coupled with the confusion of adults of these genera in the past, suggests that systematic problems still exist among these groups. The only point worth not- ing at this time is that second zoeae of two of the three species currently bearing the genus name of Pisidia and the second zoea of Petrolis- thes novaezelandiae, which Wear suggests is actually a Pisidia, all have similar setal arma- tures and possess three pairs of pleopod buds on the abdomen. All other porcellanid zoea II lar- vae known to species have four pairs of pleopod buds. In order to determine whether these two characters are of any value in distinguishing the larvae of the genus Pisidia from other larvae with the Porcellana type telson, the second zoea of Pisidia spinulifrons must be described, the identity of Menon's (1937) "Petrolisthes" spe- cies I and II, with three pleopod pairs, resolved and Pisidia inaequalis, with four pleopod pairs, reexamined as a member of the genus. In the course of this study, a number of spec- imens of each stage were examined, and par- ticular attention paid to morphological varia- tion, especially in setal counts for the larval ap- pendages, since larval variability has caused a great deal of confusion in the literature on por- cellanid and other anomuran larvae. Morpho- logical variations in setal numbers, spine lengths, etc. were found within individuals from the left to the right sides, between individuals of the same species, and between individuals of the different species studied. These findings em- phasize the necessity of basing larval descrip- tions, especially setal counts, on examination of adequate numbers of larvae of each stage. Throughout the descriptions given here, only the range of setal numbers found is indicated. Be- cause of the importance placed on setation for- mulae in descriptions of crustacean larvae, some of the setal count variations found and their implications will be analyzed in a separate paper. Variability in the number of larval molts re- quired to reach a specific point in development, apparently in response to varying environmental conditions, occurs in many Anomura (Table 6) and other decapod groups as well (e.g., Broad, 1957; Costlow, 1965). This type of variation has caused difl!iculties among larval systematists for some time and has in part given rise to the idea that either larvae produced under labora- tory conditions are abnormal and should be dis- counted altogether or that such stages are ex- traneous and should be subgrouped under larval stages most commonly noted. This concept of substages or extra stages has served to confuse the developmental picture (Efford, 1970), par- ticularly among the porcellanids, where another type of variability has been discovered which, unfortunately, has lent further support to the "substage" concept. Intermolt growth in which certain appendages increase in size within a single larval stage with- out molt is a type of variability prevalent in por- cellanids. All four species described here show 220 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE this type of growth. Lebour (1943, 1950) was probably the first author to notice specimens showing this phenomenon, Lebour observed some substage molting and concluded that each substage she found in plankton collections must necessarily have been separated from less ad- vanced substages by a discrete molt. She sub- sequently proposed that Pisidia longicornis and Porcellana platycheles probably possessed a var- iable number of stages with two "essential" stages. She assumed that a molt always sep- arated the "alternative" stages, or substages from each other. The variable molting sequence may indeed occur in these species under certain circumstances, as it does in other Anomura (Table 6) , but it is probable that lb and lie sub- stages in Pisidia longicornis and lie in Porcel- lana platycheles, which were taken in plankton samples but never obtained by molt in the lab- oratory, were products of intermolt growth. LeRoux (1961, 1966) subsequently reared the larvae of both species and reported only two "true" zoeal stages each of which underwent growth in certain appendage buds without molt- ing. It is likely that these species have the po- tential for both intermolt growth and stage number variability and that Lebour found a com- bination of both while LeRoux did not. A sim- ilar case might also exist in the Petrolisthes lar- vae studied by Wear (1964a, b), although he reports definite ecdysis between each of his sub- stages. Lai^ae of these species, like the Pisidia and Porcellana species mentioned above, may show both intermolt growth and variable stage numbers, depending on environmental circum- Table 6. — Species of Anomura other than Porcellanidae for which stage number variability has been reported.' Species of Anomura Source Pleuroncodes planipes Emerita tatpoida Emerita analoga Emerita rathbunae Hippo cubensis Blepharipoda occidentalis Calcinus tibicen Cotnbita clypeatus Triiopagurus magnificus Pftrockirus dioginfS Birgus latro Paralithodfs camtschatica Boyd ond Johnson, 1963 Rees, 1959 Johnson and Lewis, 1942; Efford, 1970 Knight, 1967 Hanson, 1969 Knight, 1968; Johnson and Lewis, 1942 Provenzono, 1962a Provenzano, 1962b Provenzono, 1967 Provenzano, 1968 Reese and Kinzie, 1966 Kurato, 1960 stances, a possibility not considered for example, by Roberts (1968) and Gore (1970) in discus- sing substages. Other authors have subsequently supported Lebour's original substage designation of growth forms even though no molting was observed (e.g., Boschi et al., 1967). This is a needless and confusing subdivision of an apparently con- tinuous process and it obscures the nature of the flexibility of the animals. The term substage is now used in such a variety of ways that it should be abandoned altogether, as (]k)re (1970) has suggested, and attention directed to the pos- sible occurrence of variation in both molting and development rates as observed by Costlow (1965) in the portunid CalUnectes sapidus. Intermolt growth to date has been recorded directly or indirectly in the larvae of 70% of the porcellanids studied for which two or more zoeal stages are known (Table 7), All these species have demonstrated abbreviated development (fewer number of molts necessary to attain ju- venile adult form) compared to most other Table 7, — Species of Porcellanidae for which intermolt growth or stage number variability have been reported. Includes only species with both zoeae known. Species Intermolt Stage growth^ variability^ Autfior Petrolisthts cinctipes + Petrolisthes eriomerus + Petrolisthes armatus — + Petrolisthes elongatus O Petrolisthes novaezelandiae O Petrolisthes rujescens O Petrolisthes sp. I Q Petrolisthes sp. II O Porcellana sp. Q Porcellana platycheles + Porcellana sigsbeiana + Pisidia longicornis + Pisidia bluteti Q Pisidia inaequalis Q Polyonyx gibbesi 4" Polyonyx guadriungulatus + Pachycheles pubescens -f- Pachycheles rudis + Pachycheles haigae + Pachycheles stevensii + Euceramus praelongus + Petrocheles spinosus — Megalobrachium poeyi + — This study — This study + Gurney, 1938 + Lebour, 1943, 1950 — Gore, 1970, 1972 + Wear, 1964b + Wear; 1964a — Gohar and Al-Kholy, 1957 — Merron, 19G7 — Menon, 1937 — Menon, 1937 — Le Roux, 1961 + Lebour, 1943 — Gore, 1971b — Le Roux, 1966 + Lebour, 1943 — Bourdillon-Casanova, 1956 — Gurney, 1938 — Gore, 1968 — Knight, 1966 ^ This study — Knight, 1966; this study — Boschi et al., 1967 — Kurata, 1964 — Roberts, 1968 — Wear, 1965o — Gore, 1971a ^ Intermolt growth has not been reported in the larvae of any of the above species. -f- = Occurrence reported. — = Occurrence not reported. O = Occurrence probable but not directly reported. 221 FISHERY BULLETIN: VOL. 71, NO. 1 Anomura studied but not necessarily reduced time spent in the plankton. This suggests that intermolt growth might permit larvae of these species to develop to metamorphosis under a va- riety of conditions without passing through a large number of molts. This mechanism might thus be of survival value in reducing larval loss at molt by effectively decreasing the number of molts required under various conditions to reach the juvenile form. As more larval histories are studied, it may be found that intermolt growth occurs in Ano- mura other than the Porcellanidae. If the func- tion and value of intermolt growth has been cor- rectly interpreted here, this form of variability might be expected among species with low num- bers of larval molts required to attain the sub- adult. Only careful studies of more life histories can reveal the possible relationship of intermolt growth and variable stage numbers. LITERATURE CITED BOOLOOTIAN, R. A., A. C. GlESE, A. Farmanfarmaian, AND J. Tucker. 1959. Reproductive cycles of five West Coast crabs. Physiol. Zool. 32:213-220. BoscHi, E. E., M. A. SCELZO, and B. Goldstein. 1967. DesarroUo larval de dos especies de Crusta- ceos Decapodos en el laboratorio. Pachycheles haigae Rodrigues Da Costa (Porcellanidae) y Chasmagnathtis granulata Dana (Grapsidae). Bol. Inst. Biol. Mar. Mar del Plata 12, 46 p. Bourdillon-Casanova, L. 1956. Note sur la presence de Porcellana bluteli (Risso) Alvarez dans le Golfe de Marseille et sur le developpement larvaire de cette espece. Rapp. P.-V. Reun. Comm. Int. Explor. Sci. Mer Medi- terr., Nouv. Ser. 13:225-232. Boyd, C. M., and M. W. Johnson. 1963. Variations in the larval stages of a decapod crustacean, Pleuroncodes planipes Stimpson (Galatheidae). Biol. Bull. (Woods Hole) 124: 141-152. Broad, A. C. 1957. The relationship between diet and larval de- velopment of Palaemonetes. Biol. Bull. (Woods Hole) 112:162-170. Brooks, W. K., and E. B. Wilson. 1883. The first zoea of Porcellana. Johns Hopkins Univ. Stud. Biol. Lab. 2:58-64. COSTLOW, J. D., Jr. 1965. Variability in larval stages of the blue crab Callinectes sapidus. Biol. Bull. (Woods Hole) 128:58-66. Efford, I. E. 1970. Recruitment to sedentary marine populations as exemplified by the sand crab, Emerita analoga (Decapoda, Hippidae). Crustaceana 18:293-308. Gohar, H. a. F., and a. a. Al-Kholy. 1957. The larvae of four decapod Crustacea (from the Red Sea). Publ. Mar. Biol. Stn. Al-Ghardaqa (Red Sea) 9:177-202. Gore, R. H. 1968. The larval development of the commensal crab Polyonyx gibbesi Haig, 1956 (Crustacea: Decapoda). Biol. Bull. (Woods Hole) 135:111- 129. 1970. Petrolisthes armatus: A redescription of larval development under laboratory conditions (Decapoda, Porcellanidae). Crustaceana 18:75- 89. 1971a. Megalobrachmm poeyi (Crustacea, Deca- poda, Porcellanidae) : Comparison between larval development in Atlantic and Pacific specimens reared in the laboratory. Pac. Sci. 25:404-425. 1971b. The complete larval development of Por- cellana sigsbeiana (Crustacea: Decapoda) under laboratory conditions. Mar. Biol. (Berl.) 11:344- 355. 1972. Petrolisthes armatus (Gibbes, 1850) : the de- velopment under laboratory conditions of larvae from a Pacific specimen (Decapoda, Porcellani- dae). Crustaceana 22:67-83. Greenwood, J. G. 1965. The larval development of Petrolisthes elon- gatus (H. Milne Edwards) and Petrolisthes novaezelandiae Filhol (Anomura, Porcellanidae) with notes on breeding. Crustaceana 8:285-307. Gurney, R. 1938. Notes on some decapod Crustacea from the Red Sea.— VI.-VIII. Proc. Zool. Soc. Lond., Ser. B 108:73-84. 1942. Larvae of decapod Crustacea. Ray Soc. (Lond.), Publ. 129, 306 p. Haig, J. 1960. The Porcellanidae (Crustacea Anomura) of the eastern Pacific. Allan Hancock Pac. Exped. 24, 440 p. Hanson, A. J. 1969. The larval development of the sand crab Hippa cubensis (de Saussure) in the laboratory (Decapoda, Anomura). Crustaceana 16:143-157. Johnson, M. W., and W. M. Lewis. 1942. Pelagic larval stages of the sand crabs Emerita analoga (Stimpson), Blepharipoda occi- dentalis Randall, and Lepidopa myops Stimpson. Biol. Bull. (Woods Hole) 83:67-87. Knight, M. D. 1966. The larval development of Polyonyx quadri- ungulatus Glassell and Pachycheles rudis Stimp- 222 CONOR and CONOR: LARVAE OF FOUR PORCELLANIDAE son (Decapoda, Porcellanidae) cultured in the laboratory. Crustaceana 10:75-97. 1967. The larval development of the sand crab Emerita rathbunae Schmitt (Decapoda, Hippi- dae). Pac. Sci. 21:58-76. 1968. The larval development of Blepharipoda oc- cidentalis Randall and B. spinimana (Philippi) (Decapoda, Albuneidae). Proc Calif. Acad. Sci., Ser. 4, 35:337-369. Knudsen, J. W. 1964. Observations of the reproductive cycles and ecology of the common Brachyura and crablike Anomura of Puget Sound, Washington. Pac. Sci. 18:3-33. KURATA, H. 1960. Last stage zoea of Paralithodes with inter- mediate form between normal last stage zoea and glaucothoe. [In Japanese, English synopsis.] Bull. Hokkaido Reg. Fish. Res. Lab. 22:49-56. 1964. Larvae of decapod Crustacea of Hokkaido, 7. Porcellanidae (Anomura). [In Japanese, English synopsis.] Bull. Hokkaido Reg. Fish. Res. Lab. 29:66-70. Lebour, M. V. 1943. The larvae of the genus Porcellana (Crus- tacea Decapoda) and related forms. J. Mar. Biol. Assoc. U.K. 25:721-737. 1950. Notes on some larval decapods (Crustacea) from Bermuda. Proc. Zool. Soc. Lond. 120:369- 379. Le Roux, A. 1961. Contribution a I'etude du developpement larvaire de Porcellana platycheles Pennant (Crustace Decapode). C. R. Hebd. Seances Acad. Sci. (Paris) 253:2146-2148. 1966. Le developpement larvaire de Porcellana. longicomis Pennant (Crustace Decapode Ano- moure Galatheide). Cah. Biol. Mar. 7:69-78. Menon, M. K. 1937. Decapod larvae from Madras plankton. I. Bull. Madras Gov. Mus. (Nat. Hist. Sect.), New Ser. 3(5): 1-56. Provenzano, a. J., Jr. 1962a. The larval development of Calcinus tibicen (Herbst) (Crustacea, Anomura) in the labora- tory. Biol. Bull. (Woods Hole) 123:179-202. 1962b. The larval development of the tropical land hermit Coenobita clypeatus (Herbst) in the lab- oratory. Crustaceana 4:207-228. 1967. The zoeal stages and glaucothoe of the trop- ical eastern Pacific hermit crab Trizopagurus magnificus (Bouvier, 1898) (Decapoda, Diogeni- dae), reared in the laboratory. Pac. Sci. 21:457- 473. 1968. The complete larval development of the West Indian hermit crab Petrochirus diogenes (L.) (Decapoda, Diogenidae) reared in the laboratory. Bull. Mar. Sci. 18:143-181. Rees, G. H. 1959. Larval development of the sand crab Emerita talpoida (Say) in the laboratory. Biol. Bull. (Woods Hole) 117:356-370. Reese, E. S., and R. A. Kinzie III. 1968. The larval development of the coconut or robber crab Birgus latro (L.) in the laboratory (Anomura, Paguridea). Crustaceana (Suppl. 2) : 117-144. Roberts, M. H., Jr. 1968. Larval development of the decapod Eucer- amus praelongiis in laboratory culture. Chesa- peake Sci. 9:121-130. Sankolli, K. N. 1967. Studies on larval development in Anomura (Crustacea, Decapoda) — I. Proceedings of the Symposium on Crustacea. Mar. Biol. Assoc. India, Symp. Ser. 2, 2:744-776. Sars, G. 0. 1889. Bidrag til Kundskaben om Decapodernes Forvandlinger. II. Lithodes, Eupagurus, Sphir- opaguriis, Galathodes, Galathea, Munida, Porcel- lana, (Nephrops). Arch. Math. Naturvidensk. 3:133-201. Shenoy, S., and K. N. Sankolli. 1967. Studies on larval development in Anomura (Crustacea, Decapoda) — III. Proceedings of the Symposium on Crustacea. Mar. Biol. Assoc. India, Symp. Ser. 2, 2:805-814. Wear, R. G. 1964a. Larvae of Petrolisthes novaezelandiae Fil- hol, 1885 (Crustacea, Decapoda, Anomura). Trans. R. Soc. N.Z., Zool. 4:229-244. 1964b. Larvae of Petrolisthes elongattis (Milne Edwards, 1837). (Crustacea, Decapoda, Ano- mura). Trans. R. Soc. N.Z., Zool. 5:39-53. 1965a. Larvae of Petrocheles spinosus Miers, 1876 (Crustacea, Decapoda, Anomura) with keys to the New Zealand porcellanid larvae. Trans. R. Soc. N.Z., Zool. 5:147-168. 1965b. Breeding cycles and pre-zoea larva of Pet- rolisthes elongatus (Milne Edwards, 1837). (Crustacea, Decapoda). Trans. R. Soc. N.Z., Zool. 5:169-175. 1966. Pre-zoea larva of Petrocheles spinosus Miers, 1876 (Crustacea, Decapoda, Anomura). Trans. R. Soc. N.Z., Zool. 8:119-124. Williamson, D. I. 1957. Crustacea, Decapoda: Larvae. I. General. Cons. Perm. Int. Explor. Mer, Fiches Identifica- tion Zooplancton 67, 7 p. 223 FEEDING, CLEANING, AND SWIMMING BEHAVIOR IN LARVAL STAGES OF PORCELLANID CRABS (CRUSTACEA: ANOMURA)' S. L. Conor and J. J. Conor* ABSTRACT Pachycheles rtidis, Pachycheles pubescens, Petrolisthes eriomerus, and Petrolisthes cinctipes have a swimming prezoeal stage of short duration. The prezoeal form is prob- ably related to escapement of the larvae from the confined adult habitat. Both zoeal stages are strong swimmers, moving both forward and backward by means of the maxillipedal exopodites, aided by the telson. Zoeae have a well-defined cleaning behavioral sequence for removal with the mouth parts of particles from the telson and the maxillipedal endopodites. The two zoeal stages are carnivorous and capture live prey upon contact but do not appear to locate prey visually. Prey are caught with the maxillipedal endopodites and held with the flexed abdomen and telson. At the molt to megalopa, the larva becomes a filter-feeding herbivore like the adult. Other adultlike behavior of the megalopa includes use of the fifth legs for cleaning and attempts by older megalopae at swimming by clapping the abdomen. Megalopae filter feed only when suspended algal food is present; water movement enhances feeding but is insuflficient alone to induce feeding. At first, using their abdominal pleopods, megalopae swim continuously, later swim intermittently, and finally settle permanently if a substrate suitable for clinging is available. A large amount of information is available about the comparative larval morphology of most groups of decapod Crustacea. Further under- standing of the adaptive significance of larval morphology and its changes during larval life requires comparable information, now largely lacking, about larval behavior and ecology. The present state of knowledge on the larvae of porcellanid crabs is typical in this respect, with a growing descriptive literature available but little known about other aspects of larval biology. A few observations on various features of porcel- lanid larval behavior have been given by Russell (1925), Spooner (1933), Foxon (1934), Gurney (1942), Lebour (1943), Greenwood (1965), and Knight (1966). None of these studies present a complete account of any behavior pattern ^ This research was supported in part by a trainee- ship from Crant Nos. 5T1-WP-111-02, -03, and -04, Federal Water Quality Administration and by National Oceanic and Atmospheric Administration (maintained by the U.S. Department of Conrmerce) Institutional Sea Crant 2-35187. " Department of Oceanography and Marine Science Center, Oregon State University, Newport, OR 97365. Manuscript accepted February 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. throughout development. Information on food and feeding is incomplete and somewhat incon- sistent. The detailed studies on feeding and respiration of adults by Nicol (1932) and Knud- sen (1964) indicate that adult porcellanids are specialized filter feeders. The available infor- mation indicates that zoeae are carnivorous or possibly also detrital feeders. Since it is now known that decapod zoeae are typically strict carnivores capturing live prey, a change in diet and mode of feeding during the development of porcellanids would be expected, but has not been clearly described. Large numbers of live larvae of all develop- mental stages and ages were available in lab- oratory cultures during a study of the larval development of four species of porcellanid crabs (Gonor and Gonor, 1973). Several aspects of the behavior of the larvae being cultured were studied closely throughout development. Special attention was given to the ontogeny of normal respiratory, feeding, cleaning, and locomotory behavior patterns from hatching through devel- opment to the first juvenile crab stage. These 225 FISHERY BULLETIN: VOL. 71, NO. 1 observations are presented here to clarify some aspects of the ecology of both the larval and adult stages. for at least 2 days, and available for observations on behavior at some time during the 2-year pe- riod of the study. METHODS The species used were Petrolisthes cinctipes (Randall), Petrolisthes eriomerus Stimpson, Pachycheles rudis Stimpson, and Pachycheles pubescens Holmes. Adults of all of these species occur in the rocky intertidal or the shallow sub- tidal zone of the northeastern Pacific but their habitats differ (Haig, 1960). The Petrolisthes species occur in the upper intertidal beneath loosely bedded rocks and boulders, whereas the two Pachycheles species are crevice and burrow dwellers. P. rudis was collected from beneath root mats of the surf grass Phyllospadix in the lower intertidal zone and P. pubescens was col- lected from burrows and crevices in rocks. Both larvae captured alive in the plankton and larvae reared from eggs candied by females in the lab- oratory were available. Gravid females were maintained in the lab- oratory at ambient sea temperatures until hatch- ing occurred. Thereafter, larvae were cultured in filtered seawater kept at several constant temperatures and fed Artemia nauplii. Larvae captured in the plankton were similarly main- tained in the laboratory. A detailed account of methods used in rearing and handling the larvae is given elsewhere (Conor and Conor, 1978). RESULTS In laboratory cultures, larvae of each of the four species showed similar basic patterns of locomotion, feeding, respiration, and cleaning, as well as changes in these patterns through time. Consequently, separate accounts for each species will not be given. Basic body and ap- pendage form is suflficiently similar in the first and second zoeae for Figures 2 and 3, of second zoeal appendages, to be referred to in the text for both stages. The descriptions given below are summarized from observations repeated throughout the rearing study as each of the cul- tures was examined daily. Table 1 gives the number of larvae of each stage kept in culture PREZOEA Females of all four species studied here release their larvae in the form of a prezoea (Figure 1) . Figure 1. — Prezoea of Pachycheles pubescens, showing rounded form. Cravid females were maintained in the labora- tory for as long as 35 days, showing normal locomotion, feeding, and cleaning behavior be- fore their eggs hatched. The following numbers of broods were hatched in the laboratory: Pachycheles rudis, 35; P. pubescens, 2; Petro- listhes eriomerus, 11; P. cinctipes, 10. In each case female behavior during hatching was of the normal type described elsewhere (Conor and Conor, 1973), and the larvae released were pre- zoeae which molted to viable first zoeae. The behavior patterns of the prezoea are simpler than those of later stages. Respiratory motions in the prezoea are effected by the fanlike sca- phognathite of the second maxilla. The type of Table 1. — Number and source of larvae of different stages available for behavioral observations. Species Zoea 1 Zoea II Megalopa Petrolisthes cinctipes (reared) 727 19 1 Petrolisthes eriomerus (reared) 603 58 14 Pachycheles rudis (reared) 757 21 1 Pachycheles rudis (plankton) 76 58 28 Pachycheles pubescens (reared) 100 18 — Pachycheles pubescens (plankton) 26 81 36 Total 2,2S9 255 80 226 CONOR and CONOR: BEHAVIOR IN LARVAL PORCELLANID CRABS cleaning behavior described below^ for the zoeal stages has not been observed in this stage. It probably is not present since the stage is short- lived and the major natatory and telson setae are confined beneath the larval cuticle. Prezoeal locomotion consists of violent spas- modic flexions of the abdomen, and there is no feeding during this stage. The larva appears to be only slightly photopositive. The strong abdominal flexions finally split the cuticle long- itudinally along the dorsal midline of the cara- pace just behind the point where the rostrum joins the carapace, and a first zoea emerges from the thin prezoeal cuticle. ZOEAL STAGES The behavior patterns of the first true zoea are considerably more complex than those of the prezoeal stage. Respiratory currents are still produced by the beating of the maxillae and are aided by swimming, as was also noted by Foxon (1934). Detrital particles often become en- tangled in the plumose setae of the maxillipeds and telson. Cleaning behavior is simple but well defined in this stage. The telson and the plumose setae along its posterior margin are cleaned by curving the abdomen under the thorax and drag- ging the telson posteriorly across the functional mouth parts, thus loosening and scraping oflf clinging debris. This scraping also serves to remove large pieces of detritus from the periph- eral setae on the mouth parts and from the setae of the endopodites on both sets of maxillipeds. The endopodites are used in feeding and require frequent cleaning. Direct cleaning of the exop- odites and natatory setae and of the carapace spines was never observed, however. The clean- ing observed was indirect and probably acci- dental, since it consisted simply of freeing the body surface of entangled debris by larval mo- tion and reversal of swimming direction. The first zoea swims primarily by beating the maxillipedal exopodites (Figure 2) which have long natatory setae, but this may be augmented by motions of the telson. The maxilliped endop- odites are held extended ventrally and do not function in swimming. The zoea are strong swimmers and due to the configuration of the Figure 2. — Maxillipeds of Zoea II, Pachycheles pubes- cens, with setules omitted from most plumose setae. Ex - exopodite; en - endopodite; A - Maxilliped I; B - Maxilliped II; C - Maxilliped III. 0.5 mm Figure 3. — Zoea II of Pachycheles pubescens, in forward swimming posture. Maxillipeds I and II rotated slightly for clarity. carapace spines (Figure 3), swim predominant- ly backward and forward. Forward motion is accomplished by synchro- nously beating both sets of exopodites downward and posteriorly, with the telson extended poster- iorly. Reversal of direction can be achieved very quickly by a forward snap of the telson and a 227 FISHERY BULLETIN: VOL. 71, NO. 1 simultaneous change in the pitch of the max- illipeds so that the exopodites beat downward and anteriorly. This rapid adjustment sends the larva on its way backward with its posterior spines leading. Unlike some larvae of the Gal- atheidae (Foxon, 1934), which swim only back- ward with the telson leading, the zoeae of por- celain crabs swim almost equally well in either direction. Normally, zoeae swim forward, re- versing direction on contact with an object. Zoea larva are voracious predators and are cannibalistic if not well supplied with other types of food. However, true hunting behavior of the kind reported for certain other predatory zoeae (Knudsen, 1960) was not observed. In fact, when a potential food organism touches any portion of the rostrum or posterior spines, or the top or sides of the carapace, the zoea imme- diately moves away from the point of contact, apparently in an act of avoidance. If, however, the prey approaches the thoracic region of a zoea closely enough to touch the setae of the ventrally extended maxillipedal endopodites (Figures 2, 3) or the ventral surface of the first two segments of the abdomen, the zoea reacts immediately. The prey is clutched between the endopodites, and the telson is used to force it forward and upward within reach of the func- tional mouth parts. This sequence has been ob- served many times, with no indication that sight is involved in locating and capturing the prey. The zoea also demonstrates what appears to be sensitivity to water movements. If a prey or- ganism passes near the thoracic or abdominal region of the zoea but does not touch the sensitive setae, the zoea, without the stimulus of direct contact, will go through all the motions normally associated with the capture of prey. In these cases, the prey is usually out of reach, and the attempts to capture the passing animal fail. The second zoeal stage is only slightly more complex morphologically than the first stage, and behavior patterns remain essentially the same throughout both zoeal stages. Metamorphosis to the megalopa (Figure 4), however, brings about immediate and drastic changes in struc- tures and habits of the larva. Basic behavior patterns which were stable in the zoeal stages undergo progressive changes throughout the megalopa stage, gradually becoming more adult- like in nature. Figure 4, — Megalopa of Pachycheles pubescens, in for- ward pleopodal swimming posture, except antennae not extended posteriad. PI, detail view of pleopod. MEGALOPA In the megalopa respiratory currents are pro- duced primarily by fanning motions of the max- illary scaphognathites (Figure 5C) and are aided by the outward flicking of the setose ex- opodite of maxilliped II (Figure 5B). The out- ward flicking causes an exhalent current by "spooning out" the branchial cavity. The ex- opodite of maxilliped III is nonfunctional at this stage. In addition, a newly settled larva ele- vates its body above the substrate and concur- rently vibrates its pleopods (Figure 4), thus in- creasing circulation of the surrounding water. Cleaning behavior becomes highly complex in the megalopa. The fifth pereiopods (Figure 4), which are mobile, chelate, and armed with spe- cialized hooked setae and dense clusters of short bristles, are carried folded along the sides of the carapace when not in use. These structures serve to clean all portions of the abdomen and telson, the three pairs of functional walking legs, 228 CONOR and CONOR: BEHAVIOR IN LARVAL PORCELLANID CRABS Figure 5. — Megalopa respiratory and feeding structures Pachycheles pubescens. Setules omitted from all plumose setae. Ex - Exopodite; en - endopodite; arrow - detail of a plumose seta. A - Maxilliped III ; B - Maxilliped II ; C - Maxilla II; D - mandible, anterior view, palp folded in place. and the under side of the carapace in the branch- ial chamber. The foremost one-fourth of the carapace cannot be reached by these legs, and in dense algal suspensions, fine hairs on this por- tion of the megalopa may become entangled with algae and other detritus. The chelipeds are freed of foreign matter by rubbing the dorsal surface of one on the ventral surface of the other in a simple lateral scraping motion. The antennules are cleaned singly or simultaneously in the following manner. An antennule is lowered to the level of a raised third maxilliped and inserted into a notch on the max- illiped formed by a long and a short group of dense setae. The setae and aesthetes of the antennule are then effectively combed free of particles as the maxilliped is drawn forward and down and the antennule passes between the setal brushes. The grooming of the feeding mechanism is the most complex cleaning behavior. This behavior can be seen in actively feeding animals as well as in nonfeeding megalopae which have been placed in a dense suspension of algal cells. The long feeding setae of the third maxilliped (Fig- ure 5A) are combed clean by the short dense setal brushes located on the two terminal seg- ments of the second maxilliped (Figure 5B). The short setal brush is inserted at the bases of the feeding setae and rolled downward and inward toward the mouth following the long curve of the setae being combed. When the en- tire length of the filtering setae has been combed free of particles, the brushes of the second max- illiped are then combed out by the first max- illiped and inner mouth parts. Undesirable particles are rejected into the exhalent current and swept out by the flicking of the exopodite on maxilliped II. The diet and feeding behavior of the megalopa are drastically different from those of the zoeal stages. Predatory habits are replaced imme- diately by filter-feeding habits when the molt is completed and the megalopal skeleton has be- come hard enough to permit motion. The mega- lopae of all four species rejected newly hatched Artemia nauplii as food and would feed only on suspended phjrtoplankton. Several monoalgal and diatom cultures (Tetraselmis sp., Isochrysis sp., and several unidentified diatoms) and nu- trient culture medium inoculated with raw sea- water were tried singly and in various combi- nations as food. In each case the megalopae fed on the suspended organisms in the cultures in the same manner. The behavioral change from prey capture to suspension feeding is reflected in changes in mandibular and maxilliped form. The natatory second maxilliped and the functionless third maxilliped of the zoea become highly specialized parts of a complex feeding mechanism in the megalopa. The feeding pattern most commonly observed in the laboratory is composed of the fol- lowing sequence of events. The endopodite of the highly setose third max- illiped is extended and a "setal net" spread open. After a moment, the maxilliped is lowered and swung in toward the body, where the setae are 229 FISHERY BULLETIN: VOL. 71, NO. 1 combed out by the process described earlier. The terminal brushes of the second maxilliped are subsequently cleaned by the first maxilliped and so on, until the particles initially trapped in the extended net finally reach the mandibles. Somewhere in this chain of events particles are sorted, and undesirable portions of the catch are ejected in the exhalent respiratory stream. Par- ticles which are acceptable as food are ground up between the curved bladelike edges of the mandibles (Figure 5D) and passed into the mouth with the aid of the mandibular palp. Most often the maxillipeds work rapidly and alternately, with one extended while the other is being combed. However, when a strong cur- rent of water runs steadily from a single direc- tion, megalopae often extend only one maxilliped and leave it out for a time. When the setae have gathered a sufficient quantity of particles, the maxilliped is withdrawn and cleaned. At this point, the free maxilliped may be extended, or the same maxilliped may be extended again after it has been cleaned. The appearance of variable feeding behavior under varying conditions led to the question of the importance of water move- ment to megalopal feeding. A 36-hr experiment was conducted in an eff'ort to determine the effect of turbulence on feeding. Eighteen healthy megalopae, nine each of Pachy- cheles rudis and P. pubescens, were used. Each of the megalopae and a stone for it to cling to were placed in a flask of filtered seawater and starved for the first 13 hr of the experiment. A mixed algal suspension of Isocrysis sp. and Tetraselmis sp., previously found to be accept- able to megalopae as food, was supplied to the same 18 larvae for the remaining 24 hr of the experiment. At intervals of 2 hr, throughout the 36-hr period, the number of animals showing feeding motions was recorded before and after stirring the water. The flasks with the mega- lopae were held at 12°C in a water bath. The ex- periment was started at 0100 on one day and continued to 1300 the next before food was in- troduced, so that observations were made for both starved and fed conditions at night and during daylight hours. Figure 6 summarizes the observations on the number of animals feeding under each condi- tion, before and after stirring the water. When no food was present and the water was still, virtually no feeding activity was observed. STARVED FED I 13 \3 15 23 7 8 10 n 12 13 TIME Figure 6. — Histogram showing effect of water movement on feeding behavior in megalopae of Pachycheles pubescens and P. rudis under starved and fed conditions. Left bars, P. pubescejis ; lower, open part of bar, number of larvae feeding in still water; cross-hatched part of bar, number feeding after agitation of water. Right bars, P. rudis ; open part of bar, number of larvae feeding in still water; solid part of bar, number feeding after agi- tation of water. 230 CONOR and CONOR: BEHAVIOR IN LARVAL PORCELLANID CRABS Turbulence caused a few animals to undergo feeding behavior even when food was absent. When food was present, there usually was some "feeding activity in still water. When the water was agitated and food was present, feeding be- havior greatly increased. In the presence of food, water movement, although not essential for the initiation of feeding, favors both the initiation and maintenance of feeding activity, probably because turbulence suspends the par- ticles so that they can readily be filtered. The results suggest that there is a periodicity to feed- ing activity, but the experiment was too brief to clearly demonstrate this. No effect of daylight or darkness on feeding behavior was observed. Locomotory behavior gradually changed throughout the megalopa stage. The numbers of megalopae available for observation at dif- ferent ages are indicated in Table 2, which gives the age of the megalopae in days since the molt from the second zoeal stage. Adultlike locomo- tory behavior slowly evolved as the megalopa grew older. Newly metamorphosed megalopae were strongly planktonic, swimming almost con- tinuously by means of the pleopods, with the walking legs and chelipeds extended forward. After spending some time (1-4 days) as truly planktonic animals, the megalopae became more quiescent and began to demonstrate clinging tendencies. Small stones, to which the settling megalopae could cling, were introduced into the culture flasks at this point. The megalopae which are just beginning to settle show no signs of recognizing a substrate suitable for settling. If a larva encounters a rock while swimming forward, it continues swimming motions, pushing against the obstacle with the extended chelipeds but not moving for- Table 2. — Number and age in days of megalopae avail- able for behavioral observations. Species Age in days 5 10 15 20 30 40 45 50 Pachychfles pubfscens 34 32 27 23 18 9 4 Pachychelts rudis 21 19 17 17 11 4 3 1 Petrolisthes friomerus 14 8 7 1 Petrolistkes cinctipes 1 Tota< 70 59 51 41 29 18 7 1 ward. However, if megalopae of this age are artificially introduced to a rough surface, walk- ing legs first, by means of turbulence or by di- rect placement, they cling readily and will usu- ally remain on that surface unless disturbed. As the megalopae increase in age, they appear to develop the ability to recognize a suitable sub- stratum. Larvae of 2 weeks or older were seen to collide head on with a rock, stop swimming, and put the walking legs down. They then turned around and backed onto the piece of gravel. In the laboratory, recently settled megalopae can be induced to leave an apparently suitable substrate if the water is stirred vigorously or if the larvae are touched. In addition, when a settling megalopa encounters a stone that is al- ready occupied, it will remain there only if it can do so without contacting the other occupant. If the stone is too small to allow this, the settling megalopa will usually cling only momentarily and then resume swimming. In a few cases, however, the settling megalopa forced the ori- ginal occupant to leave its stone and begin swim- ming. A specialized form of behavior was observed in advanced megalopae of Petrolisthes eriomer- iis. After locating a stone, a swimming meg- alopa will settle on it and then elevate and lower both chelipeds simultaneously several times. If the megalopa then moves a short distance over the rock, the cheliped elevation sequence is often repeated. This activity was observed only in individuals that had just arrived on a substrate and was exhibited whether or not another meg- alopa was. present on the rock. An advanced behavior was observed in still older animals (30-33 days in Pachycheles pubes- cens) . These megalopae could not be induced to use their pleopods when swimming, even when they were so disturbed that they would finally leave their rocks. Instead, a disturbed animal would bob up and down in the water, clumsily clapping the abdomen to the thorax. The presence of the four pairs of fully developed pleopods prevented effective swimming by this action. Adult porcelain crabs, whose pleopods are reduced in size and often in number, swim only by clapping the abdomen to the thorax. The 231 FISHERY BULLETIN: VOL. 71, NO. 1 megalopae therefore show in this situation the behavior of adults. Further development along the lines already established by the megalopa is shown by the suc- ceeding juvenile stages. Respiratory currents become much stronger and well defined and are now aided by the tandem motion of the two sets of maxillipedal exopodites. Feeding movements are the same as those described here for the megalopa and for the adult by Nicol (1932). In adults, food preferences vary according to species. Knudsen (1964) reports that pelagic diatoms are the preferred food for Petrolisthes eriomerus but does not state the preferences of Pachycheles rudis. In the laboratory, adults of all species readily fed in unfiltered seawater, but only Petrolisthes females accept and ingest frag- ments of mantle and adductor muscle of Mytihis. The preferred adult locomotion is pereipodal walking, but swimming by abdominal clapping is also used, especially in Petrolisthes. DISCUSSION PREZOEA Adults of both Petrolisthes species live where the larvae are released into turbulence caused by waves and currents. These conditions favor the presence of a short-lived, rounded initial larva with few body projections, that would more readily escape entangling algae and debris in the intertidal zone. The habitats of the two Pachycheles species (Haig, 1960) more strongly favor a compact, rounded, initially spineless larval form. In many cases adults become so large that they are unable to pass out through the openings of the burrows or crevices they inhabit. Larvae are released within the adult burrow and must escape this confinement to survive. The prezoeal cuticle covering the telson in P. piibescens larvae is well modified for swimming. This was described by Gurney (1942) for other decapods and by Le- bour (1943), Wear (1965), and Greenwood (1965) for other porcellanids. These observations suggest that in the four species considered here, and probably in the family as a whole, the prezoea is a short-lived natural stage and is not a laboratory artifact as has often been suggested. Its existence as a transport stage is ecologically consistent with the natural habitat of the adults. Similar argu- ments have been put forth by Gore (1968) in defense of the interpretation of the prezoea as a natural stage in the commensal porcellanid Polyonyx gihhesi, which releases larvae from in- side the tube of the polychaete Chaetopterus. Another observation supporting this argument is that, under laboratory conditions, true zoeae, with their long spines, respond very unfavorably to collisions, spine breakage, and collection of detrital material on the spines, all of which would be likely to occur in nature if full zoeae emerged from the eggs and were released into the adult environment. Photopositive swim- ming behavior would also prove useful to pre- zoeae released in burrows and crevices, and the larvae studied showed a photopositive response. This response was, however, weak under lab- oratory conditions. ZOEA With the passage of the larva through the prezoeal molt, the first true zoea emerges and becomes an actively swimming planktonic car- nivore. Despite the good swimming ability and well-developed eyes of the zoea, no evidence of true hunting behavior was found. Instead, the larvae appear to rely entirely on chance en- counters with prey, with capture behavior ini- tiated by direct contact or by vibrations stimu- lating the maxillipedal endopodites and setae and the ventral surface of the abdomen. Similar stimulation of other parts of the body elicits an escape response by the zoea. Survival of these zoea in the plankton is probably highly depen- dent upon suitable prey density. The method of prey capture used by these larvae, involving the use of the telson to scoop up the prey and hold it from below, appears to be a feeding method used by zoea throughout the Decapoda. Knudsen (1960), for example, describes this method of feeding in xanthid Brachyura. MEGALOPA Many of the adult behavioral features de- scribed by Nicol (1932) appear in the megalopa 232 CONOR and CONOR: BEHAVIOR IN LARVAL PORCELLANID CRABS when the structures used first resemble those of the adult, but before the adult mode of life is adopted. Attempted swimming by abdominal clapping is an example of the development of an adult behavior pattern before the adult structure is completed. Megalopae of these four species clean the body with the fifth legs in the adult manner, a feature also observed for other por- cellanid megalopae by Lebour (1943) and Knight (1966). At the molt to megalopa there is an abrupt change from carnivorous to a filter-feeding, herbivorous habit. As Nicol (1932) first noted, adult Porcellanidae are specialized for filter feed- ing on suspended material. Adults of the two Petrolisthes species studied here will also some- times accept pieces of mussel as food, but the two Pachycheles species will not. The observations on the megalopae of these species and those of Lebour on Porcellana species, indicate that the acquisition of morphological and behavioral adaptations to filter feeding in the Porcellanidae involve both the adult and megalopa stages. Although many other anomuran adults feed on particulate material by some mode of filter feed- ing, it is not known whether this also involves the late larval stages. No information could be found in the literature on feeding by their post- zoeal stages. In hermit crabs, which are more generalized detrital-feeding Anomura, both zoea and postzoeal stages can be reared on Artemia nauplii, as found for example by Provenzano (1962). Since population and species success depends upon the megalopae locating a suitable adult habitat, settling behavior is a critical part of later larval development. Settling behavior of barnacles, bryozoa, and some other forms has been studied; however, similar settling behavior in decapod Crustacea has not been studied, with the exception of shell selection by the glaucothoe stage of hermit crabs (Reese, 1962; Hazlett, 1971) and the coconut crab Birgus latro (Reese, 1968). For the megalopae studied here, the behavioral sequence of the true planktonic pe- riod, the settling and swimming period, and the period of final settlement seems to be highly spe- cialized for substrate selection. Possession of this behavioral mechanism would permit, under na- tural conditions of turbulent water movement, older megalopae to select or reject substrates encountered by random contact. No information is available on how settled megalopae reach the final adult habitat after initial settlement, , but postsettling megalopae and very young juveniles have been found clinging to the base of surf grass and under stones with adults. LITERATURE CITED FoxoN, G. E. H. 1934. Notes on the swimming methods and habits of certain crustacean larvae. J. Mar. Biol. Assoc. U.K., New Ser. 19:829-849. GONOR, S. L., AND J. J. GONOR. 1973. Descriptions of the larvae of four North Pacific Porcellanidae (Crustacea: Anomura). Fishery Bull., U.S. 71: Gore, R. H. 1968. The larval development of the commensal crab Polyonyx gibbesi Haig, 1956 (Crustacea: Decapoda). Biol. Bull. (Woods Hole) 135:111- 129. Greenwood, J. G. 1965. The larval development of Petrolisthes elon- gatiLS (H. Milne Edwards) and Petrolisthes novaezelandiae Filhol (Anomura, Porcellanidae) with notes on breeding. Crustaceana 8:285-307. GURNEY, R. 1942. Larvae of decapod Crustacea. Ray Soc. (Lond.) Publ. 129, 306 p. Haig, J. 1960. The Porcellanidae of the eastern Pacific (Crustacea Anomura). Allan Hancock Pac. Ex- ped. 24, 440 p. Hazlett, B. A. 1971. Influence of rearing conditions on initial shell entering behavior of a hermit crab (Deca- poda, Paguridae). Crustaceana 20:167-170, Knight, M. D. 1966. The larval development of Polyonyx quad- riungulatus Glassell and Pachycheles rudis Stimp- son (Decapoda, Porcellanidae) cultured in the laboratory. Crustaceana 10:75-97. Knudsen, J. W. 1960. Reproduction, life history and larval ecology of the California Xanthidae, the pebble crabs. Pac. Sci. 14:3-17. 1964. Observations of the reproductive cycles and ecology of the common Brachyura and crablike Anomura of Puget Sound, Washington. Pac. Sci. 18:3-33. Lebour, M. V. 1943. The larvae of the genus Porcellana (Crusta- cea Decapoda) and related forms. J. Marine Biol. Assoc. U.K. 25:721-737. 233 FISHERY BULLETIN: VOL. 71, NO. I NiCOL, E. A. T. 1932. The feeding habits of the Galatheidea. J. Mar. Biol. Assoc. U.K., New Ser. 18:87-106. Provenzano, a. J., Jr. 1962. The larval development of Calcinus tibicen (Herbst) (Crustacea, Anomura) in the labora- tory. Biol. Bull. (Woods Hole) 123:179-202. Reese, E. S. 1962. Shell selection behavior of hermit crabs. Anim. Behav. 10:347-360. 1968. Shell use: An adaptation for emigration from the sea by the coconut crab. Science (Wash., D.C.) 161:385-386. Russell, F. S. 1925. The vertical distribution of marine macro- plankton. An observation on diurnal changes. J. Mar. Biol. Assoc. U.K., New Ser. 13:769-809, Spooner, G. M. 1933. Observations on the reactions of marine plankton to light. J. Mar. Biol. Assoc. U.K., New Ser. 19:385-438. Wear, R. G. 1965. Larvae of Petrocheles spinosus Miers, 1876 (Crustacea, Decapoda, Anomura) with keys to the New Zealand porcellanid larvae. Trans. R. Soc. N.Z., Zool. 5:147-168. 234 BIOLOGY OF THE PYGMY SEA BASS, SERRANICULUS PUMILIO (PISCES: SERRANIDAE) Robert W. Hastings^ ABSTRACT During the period from 1968 to 1971, numerous specimens of Serraniculus pumilio, were collected in shallow waters of the northern Gulf of Mexico. This paper presents biological data accumulated from these and other specimens in the fish collection of Florida State University and from scattered literature references regarding the species. The range of S. pumilio extends from North Carolina along the continental margin of the western Atlantic Ocean to Guyana, but it apparently does not occur in the West Indies. It has been collected at depths from 1 to 117 m, usually over sand or shell bottoms near coral or rock reefs or accumulations of mollusk shells. Individuals move about considerably, although they spend much time resting on the bottom. S. pumilio is a synchronous hermaphrodite, but pairs mate to exchange gametes and self-fertilization probably never occurs. Spawning occurs between March and August or September in the northern Gulf of Mexico. A length-frequency distribution of specimens collected in the northern Gulf is presented to show the growth rate of first year fish. Juveniles (15-20 mm SL) which appear inshore in September reach a size of 50-55 mm by the following June. Most fish move offshore to deeper water for the winter (January and February) and individuals larger than 55 mm apparently never appear inshore. Small crustaceans are the most important food items. Since the pygmy sea bass, Serraniculus pumilio, was described in 1952 by Ginsburg, virtually nothing has been added to our knowledge of the species except for a few brief notes in general surveys of fishes. During a study of reef fishes in the northern Gulf of Mexico from 1968 to 1971, numerous specimens of Serraniculus were col- lected and notes were taken on their biology. The present paper is a synopsis of the information scattered in the literature and the new data ac- cumulated during this study. The generic allocation of the species used in this paper is tentative. Robins and Starck (1961) briefly discussed the characteristics of the genus Serraniculus but did not include the species in their review of the genus Serranus. Ginsburg (1952) noted that Serraniculus and Dules differ from Serranus in having six rather ' Department of Biological Science, Florida State Uni- versity, Tallahassee, FL 32306; present address: De- partment of Biology, Rutgers University, Camden, NJ 08102. Manuscript accepted February 1972. FISHERY BULLETIN: VOL. 71, NO. I, 1973. than seven branchiostegal rays, and that Dules differs from the other two genera in having the third dorsal spine greatly prolonged. After sub- sequent study Robins (personal communica- tion) believes that Serraniculus is inseparable from Dules, but because Dules itself is so close to Serranus, he recommends no change until a morphological study of a wider range of ser- ranid genera can be completed. According to Robins, Serraniculus pumilio and Dules auriga are distinct species. METHODS During the present study, 106 specimens of Serraniculus pumilio [15.8-54.1 mm standard length (SL)] were collected. Most of these (75) were collected at East Pass of Choctawhatchee Bay near Destin, Fla, where biweekly or monthly observations were made between June 1968 and December 1970. Others were collected from St. Andrew Bay near Panama City, from Apalachi- cola Bay, and from AlHgator Harbor, Fla. A 235 FISHERY BULLETIN: VOL. 71, NO. 1 few were taken short distances offshore from these locations. Numerous field observations were made while diving, and small groups of specimens (2-3) were maintained and casually observed for a few months in 20-gal aquaria. Eighty-seven specimens were examined for gonad development. Most were examined super- ficially using a low-power ( 10-30 x) dissecting microscope, and the stage of development of the ovarian portion of the gonad was estimated, fol- lowing the definitions of Smith (1965) and Moe ( 1969) . Sexual maturity (the presence of Stage 4 oocytes) was determined on the basis of ovar- ian tissue only, since very small individuals often contained some mature sperm, even though the ovarian tissue was immature. For 22 specimens the gonads were removed, embedded in paraffin, and sectioned. Most specimens had been origi- nally preserved in 10% Formalin' and then transferred to 40% isopropyl alcohol. Gonads to be sectioned were removed from the fish, placed in Bouin's fluid for several days, dehy- drated in ethyl alcohol, celloidin-methyl salicy- late, and xylene, and then embedded in paraffin. Gonads were sectioned at 10 /a, mounted on mi- croscope slides, and stained. Mallory-Heiden- hain stain was used for most slides. This stain was easy to use and yielded good contrast be- tween various tissues within the gonad. Stomachs of 31 specimens were removed and their contents examined under a low-power dis- secting microscope. Food items were identified to major group (usually class or order) and counted. The importance of each group was determined by calculating its frequency based upon the total number of food items counted and upon the number of fish containing each food type. DISTRIBUTION Briggs (1958) gave the range of Serraniculus pumilio as North Carolina to Florida, and the southwestern Gulf of Mexico in shore areas. Subsequent references (Bullis and Thompson, " Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. 1965; Cervigon, 1966) indicate that it occurs along the coast of the western Atlantic Ocean from North Carolina to Guyana. Most collec- tions have come from the Atlantic coast of the southeastern United States and from the eastern Gulf of Mexico as a result of the numerous fish- eries surveys in these areas (Reid, 1954; Spring- er and Bullis, 1956; Springer and Woodburn, 1960; Moe and Martin, 1965; Moe et al, 1966; Starck, 1968; Struhsaker, 1969). The species has also been collected in the western Gulf of Mexico off Texas (Ginsburg, 1952; Hildebrand, 1954) and in Campeche Bay off Mexico (Hilde- brand, 1955) . More recent collecting by the RV Oregon and RV Pillsbury off the coast of Co- lombia indicates that the species is also common in the Caribbean Sea (C. R. Robins, pers, comm.). Two specimens have been taken by the Pillsbury from off the coast of Honduras. Serraniculus is unrecorded from the Bahamas (Bohlke and Chaplin, 1968) and apparently also absent from the other islands of the West Indies. Numerous species of fishes have similar conti- nental distributions in the western Atlantic, while other species are restricted to the coral reef areas of the islands and portions of the Central and South American coast where the continental shelf is narrow. The factors which prevent the distribution of continental species in the islands are not completely understood, but probably include differing ecological condi- tions, as well as competition from closely related and better adapted island species (Robins, 1971) . Physical barriers are undoubtedly not important since many of the species have pelagic larval stages, and some free-swimming species such as Scomberomorus maculatus are also absent from most of the West Indies. Ecological parameters which may be important are temperature, bot- tom type, salinity, and turbidity. Temperature may not be important for Serraniculus pumilio since it is distributed well into the tropics, but it is apparently more tolerant of low temperatures than are most of the coral reef fishes. Serranic- ulus may prefer continental sediments rather than coral reef debris, but accurate descriptions of substrates where it has been collected are un- available. In general, the continental species, including S. pumilio, are more tolerant of vary- 236 HASTINGS: BIOLOGY OF PYGMY SEA BASS ing conditions, but whether they require such changes is unknown. For species such as S. pumilio, which are territorial, competition by other species with similar habitat requirements may be important. Two potential competitors of S. pumilio are noted in the following section. HABITAT Serraniculus pumilio apparently occurs typi- cally at moderate depths (10-70 m) over the con- tinental shelf but may occasionally occur in shal- low coastal waters less than 1 m deep. It has been recorded as deep as 117 m (Bullis and Thompson, 1965). During this study, the species was often com- mon in the shallow waters along the jetties at East Pass of Choctawhatchee Bay in depths of about 1-10 m but was usually absent during Jan- uary and February. Inshore populations of Ser- raniculus apparently move offshore to deeper water during the winter. The lowest temper- atures at which the species was recorded at East Pass were 13°-14°C. Shallowwater tempera- tures in the northern Gulf often drop below 10 °C during the winter, while the temperature in water deeper than about 18 m usually remains above about 15°C. The minimum temperature extreme at which S. pumilio can survive may be about 13°C. Like other small serranids, this species is most common over sand or shell bottoms near irreg- ularities such as coral or rock outcrops. Springer and Woodburn (1960) noted that it was similar in this respect to Centropristis ocyurus, another species restricted to the continental shelf. Ser- ranvs tigrinus and S. baldwini (Robins and Starck, 1961) also occupy such habitats at about the same depth range as Serraniculus pumilio, but these two species are restricted primarily to coral reef areas of the West Indies where Serraniculus is not found. A few specimens of Serraniculus have been collected in grass beds (mostly Thalassia and Syringodium) at the mouth of Alligator Harbor. However, in such areas, numerous bare patches of sand bottom are present, and there are accu- mulations of shell debris in places which provide suitable habitat for the species. Apparently Serraniculus pumilio does not have a particularly restricted home range. The num- ber of individuals observed on the East Pass jetties varied considerably, indicating that in- dividuals were often moving into and out of the area. These observations were based upon adults or advanced juveniles; hence recruitment of populations on the jetties resulted from move- ment of adults, not from the immigration of re- cently spawned young. The pattern of move- ment of adults is not known, but more stable populations may occur on the numerous lime- stone reefs which lie short distances offshore in the area. Accumulations of mollusk shells occur over most of the sandy bottom in the area and small fishes suoh as Se7'raniculus could use such accumulations for shelter when moving over open bottoms. In this way Serraniculus could easily move the few miles from the offshore reefs to the jetties. BEHAVIOR Springer and Woodburn (1960) described Serraniculus pumilio as sedentary, but individu- als apparently move about considerably. The fish rests on the bottom using its pelvic fins as props and moves about in short "hops" over the bottom. It appears to be territorial and protects its temporary abode (near a large shell, rock, or coral ledge) from intrusion by other fishes. The size of the area defended is not known, but when two or more individuals were placed in a 20-gal aquarium, one became dom- inant and forced the other fish to remain off the bottom, even when several rock piles were provided as hiding places. During agonistic displays, Serraniciilus spreads its dorsal and caudal fins and gill covers, and presents the lateral side of the body to the opponent (Figure 1). During these displays, the dark and light bands on the side of the body become more distinct and series of small, pale spots are clearly visible along the rays of the dorsal and caudal fins. These displays are fol- lowed by the fish beating the side of its opponent with its caudal fin and the posterior part of its body. When one individual proves subordinate, it retreats with its dorsal fin depressed and usu- 237 FISHERY BULLETIN: VOL. 71, NO. 1 Figure 1. — Agonistic display of aquarium-held Serrani- Ik i cuius pumilio. ally with the victor in pursuit. Quite similar ag'onistic behavior involving fin-spreading and tail-beating has been described in Serranus scriba (Kirchshofer, 1954). Subsequently, the subordinate individuals in an aquarium remained in precarious positions in the corners of the aquarium some distance above the bottom. When these fish left such positions, except during pe- riods of feeding, they were soon attacked by the dominant individual and returned to their "sanc- tuaries" near the surface of the water. Once, agonistic behavior was observed be- tween two fish on the East Pass jetties, but in this case the retreating individual had its dor- sal fin greatly expanded. REPRODUCTION AND DEVELOPMENT Microscopic examination of, sectioned gonads (Figure 2) proves that Serraniculus pumilio is a synchronous hermaphrodite, as Ginsburg (1952) originally supposed. His statement that the testicular tissue is "interspersed with the masses of ripe roe" is misleading, however. The gonads of Serraniculus are identical to those of Serranus in that the testicular portion is well separated from the ovarian tissue and is re- stricted to narrow bands which lie along the ventral surface of each gonad (Reinboth, 1962; Smith, 1965). The smallest and the largest specimens ex- amined had both ovarian and testicular tissue in the gonad and many were seen with both mature eggs and sperm present. As with Ser- ranus, internal self-fertilization is undoubtedly impossible since separate ducts are present to carry eggs and sperm. Clark (1959, 1965) found that under aquarium conditions, Serranus subligariu^ could fertilize its own eggs, but mates with another individual and exchanges gametes under normal conditions. Reinboth (1962) also induced self-fertilization in Serranus scriba but suggested that paired spawning normally occurs. The same is probably true for Serraniculus. On 10 May 1968, several pairs of Serraniculus pumilio apparently involved in reproductive be- havior were seen on the sand bottom at the base of the St. Andrew Bay jetties, on the Gulf side of the west jetty in water about 2 m deep. The water temperature was 23°C, Pairs would move slowly about in close proximity with one indi- vidual following behind the other. The trailing individual would repeatedly nudge the anal re- gion of its mate. Clark (1959) noted such nudg- 238 HASTINGS: BIOLOGY OF PYGMY SEA BASS ing behavior in spawning Serraniis. Although no spawning clasps were observed, six specimens were collected, and subsequent examination re- vealed that all had large numbers of ovulated eggs within the lumen of their gonads. These specimens were 41.7-48.6 mm SL. Based upon examination of gonads of pre- served specimens, Serraniculus pumilio spawns between March and August or September. The number of small individuals collected during these months is limited, so the size at which maturity of the ovarian tissue is reached was not determined. There is some indication that it may be about 40 mm SL. Seven specimens (34,1-41.2 mm) collected in March were imma- ture, while two (43.0 and 44.2 mm) were ma- ture. One specimen (36.6 mm) collected in April had mostly immature gonads but had a few Stage 4 oocytes. Twelve specimens (40.6- 49.5 mm) collected in May were mature, while two (40.5 and 40.8 mm) were mostly immature, with only a few Stage 4 oocytes. A few indi- viduals as small as 23 mm SL contained mature sperm, even though the ovarian tissue was im- mature, but it is not known whether such in- dividuals could successfully spawn as males. Reinboth (1962) noted a similar condition in Serranus scriba and S. cabrilla, in which the testicular portion of the gonads matured earlier in the annual reproductive cycle and also at a smaller size. He speculated that S. scriba in their first year of sexual maturity (120-140 mm body length) function only as males and that the ovarian tissue matures only in fish over 160 mm in length. He did not demonstrate actual spawning in such first-year males, however. All 38 specimens (15.8-48.8 mm SL) collected in February, September, October, November, and December had immature gonads, although one specimen (42.7 mm) collected in October had a few mature oocytes in the posterior region of its gonad. All specimens larger than 43.0 mm collected from March through August had ma- ture gonads. Specimens with ovulated eggs were collected in May, June, and July. Nothing is known of the embryonic and larval development of Serraniculus pumilio. The eggs of Serranus are buoyant (Clark, 1959) and eggs and larvae of Epinephelu^ (family Serranidae) are pelagic (Moe, 1969), but it is not known if the same condition exists in Serraniculus. The smallest individual collected during this study (15.8 mm SL) has the general body shape and pigmentation of adults. Figure 2. — Cross section through the basal (posterior) portion of the gonads of Serraniculus pumilio showing mature ovarian and testicular tissues. / 289 FISHERY BULLETIN: VOL. 71, NO. 1 GROWTH A length-frequency distribution (Figure 3) of S. pumilio occurring in inshore waters indicates that all belong to the same year class and that no second year fish occur inshore. Small fish which were apparently spawned in the summer appear inshore in September, and some may remain throughout the winter. Most probably move to deeper water as the temperature drops 8 z — 1 SEPT r" 1 1 6 4 2 14 ^ OCT 12 10 8 6 4 2 I — 4 NOV 2 1 4 2 DEC mmm 2 IAN 2 - FEB ■■■;■:... 6 MAR y * 3 2 - :::::::::::«::: :■:■:■:■:■:■:■:-:-: u> 2 APR kk. I '-' 8 |6 1 MAY - % 4 :,::■■ ■■r:: ' 2 /{ J UN OWgWgfl 2 — aaSSSSSs 6 JUL 4 2 : ■:■:■:■:■:■:■:■:■:■: 6 AUG 4 2 _ 15 20 25 30 35 40 45 STANDARD LENGTH (in mm) I 1 EatI Pats - 1968 M3 East Past - 1969 ^ East Pass - 1970 I I Others 50 55 Figure 3. — Monthly length-frequency distributions of Serraniculus pumilio collected inshore at East Pass, Choctawhatchee Bay, Fla., and at other locations in the northern Gulf of Mexico. and move inshore again in March. No adults were found after August, The largest specimen collected during this study (54.1 mm) is consid- erably smaller than the maximum size attained (80 mm SL — Ginsburg, 1952) so possibly the larger, second year fish remain offshore in deeper water. FOOD HABITS AND PREDATION Thirty-one specimens were examined for stomach contents, but two were empty. Results of stomach analyses are shown in Table 1. S. pumilio feeds predominantly upon crustaceans, which made up 91% of the total number of food items. Numerically, amphipods are the most common group, but shrimps and crabs comprise a larger volume of the stomach contents when present and may be the most important food items. Serraniculus appears to be an indiscrim- inate carnivore, feeding upon any small organ- ism which it discovers, but showing preference for small crustaceans. The extent of predation by other fishes on S. pumilio is not known. It seems strange that D. S. Jordan never found this species in his ex- tensive studies of the stomach contents of snap- pers and groupers taken in the northern Gulf of Mexico (Jordan, 1884, 1886; Jordan and Gilbert, 1882, 1883; Jordan and Swain, 1885). Stephen A. Bortone (personal communication) found one specimen in the stomach of a Lagodon rhom- boides collected in Apalachee Bay, 11 October 1969, about 2 miles south of the St. Marks light- house, Wakulla County, Fla., at a depth of about 3 m. The general cryptic coloration and seden- tary habits of the pygmy sea bass may conceal it from most predators. ACKNOWLEDGMENTS The author was partially supported during this study by a grant from the Sport Fishing Insti- tute. This grant also provided funds to purchase a boat for Florida State University which was used extensively during this study. I thank the many persons who assisted in the collection of specimens examined during this study, especially 240 HASTINGS: BIOLOGY OF PYGMY SEA BASS Table 1. — Food habits of Serraniculus pumilio based on stomach analysis of 29 specimens (15.8-54.1 mm SL) collected in inshore waters of the northern Gulf of Mex- ico, 1968-71. Food types Number of items Percent of total number Number of fish containing eoch Percent frequency of occurrence Crustacea 262 91.0 28 96.5 Amphipods no 38.2 17 58.6 Caprellid amphipods 12 4.2 6 20.7 Isopods 32 11.1 8 27.6 Copepods 20 6.9 7 24-1 Tonaidoceans 4 1.4 3 103 Cumaceans 1 0.3 1 3.4 Shrimps (adults) 18 6.2 10 34.5 Shrimp zoea (?) 23 8.0 7 24.1 Crabs (adults) 21 7.3 9 31.0 Crab mega lops 14 4.9 2 6.9 Unidentified crustacean 7 2.4 5 17.2 Polychaetes 9 3.1 9 31.0 Unidentified wormlike organisms 13 4.5 I 3.4 Gastropods 1 0.3 1 3.4 Algae (1 mass) 0.3 1 3.4 Fishes 2 0.7 2 6.9 Dr. Camm C. Swift and Stephen A. Bortone, both formerly of Florida State University. I also thank Dr. P. P. Graziadei of Florida State University who allowed me to use facilities in his histology laboratory in preparing microscope slides of sectioned gonads. I greatly appreciate the comments of Dr. C. Richard Robins of the Rosenstiel School of Marine and Atmospheric Science at the University of Miami (Florida) and his critical re\ iew of this manuscript. Spe- cial thanks are expressed to Dr. Ralph W, Yerger of Florida State University for his continued encouragement in my studies of marine fishes of the northern Gulf of Mexico and for reviewing and editing the original manuscript. I am in- debted to my wife, Diana, for her encouragement and patience and for her assistance in typing. LITERATURE CITED BOHLKE, J. E., AND C. C. G. ChAPLIN. 1968. Fishes of the Bahamas and adjacent tropical waters. Livingston Publ., Wynnewood, Pa. 771 p. Briggs, J. C. 1958. A list of Florida fishes and their distribution. Bull. Fla. State Mus., Biol. Sci. 2:223-318. Bullis, H. R., Jr., and J. R. Thompson. 1965. Collections by the exploratory fishing vessels Oregon, Silver Bay, Combat, and Pelican made during 1956 to 1960 in the southwestern North Atlantic. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 510, 130 p. Cervigon M., F. 1966. Los peces marinos de Venezuela. Estac. Invest. Marinas de Margarita, Fund. La Salle Cienc. Nat., Monogr. 11, 951 p. Clark, E. 1959. Functional hermaphroditism and self-fertili- zation in a serranid fish. Science (Wash., D.C.) 129:215-216. 1965. Mating of groupers. Nat. Hist. 74(6) : 22-25, GiNSBURG, I. 1952. Eight new fishes from the Gulf coast of the United States, with two new genera and notes on geographic distribution. J. Wash. Acad. Sci. 42:84-101. Hildebrand, H. H. 1954. A study of the fauna of the brown shrimp (Penaeits aztecus Ives) grounds in the western Gulf of Mexico. Publ. Inst. Mar. Sci. Univ. Tex. 3:231-366. 1955. A study of the fauna of the pink shrimp (Penaeus duorarum Burkenroad) grounds in the Gulf of Campeche. Publ. Inst. Mar. Sci. Univ. Tex. 4:169-232. Jordan, D. S. 1884. Notes on a collection of fishes from Pensacola, Florida, obtained by Silas Stearns, with descrip- tions of two new species (Exocoetus volador and Gnathypops mystacinus) . Proc. U.S. Natl. Mus. 7:33-40. 1886. Notes on some fishes collected at Pensacola by Mr. Silas Stearns, with descriptions of one new species (Chaetodon aya) . Proc. U.S. Natl. Mus. 9:225-229. Jordan, D. S., and C. H. Gilbert. 1882. Notes on fishes observed about Pensacola, Florida, and Galveston, Texas, with description of new species. Proc. U.S. Natl. Mus. 5:241-307. 1883. Description of two new species of fishes (Aprion ariommus and Ophidium beani) from Pensacola, Florida. Proc. U.S. Natl. Mus. 6:142- 144. Jordan, D. S., and J. Swain. 1885. Description of three new species of fishes (Prionotus stearnsi, Prionotus ophryas, and An- thias vivanus) collected at Pensacola, Florida, by Mr. Silas Stearns. Proc. U.S. Natl. Mus. 7:541- 545. KiRCHSHOFER, R. 1954. Okologie und Revierverhaltnisse beim Schriftbarsch (Serranus scriba Cuv.). Osterr. Zool. Z. 5:329-349. MoE, M. A., Jr. 1969. Biology of the red grouper, Epinephelus morio (Valenciennes), from the eastern Gulf of Mexico. Fla. Dep. Nat. Resour. Mar. Res. Lab., Prof. Pap. Ser. 10, 95 p. 241 FISHERY BULLETIN: VOL. 71, NO. 1 MoE, M. A., Jr., P. C. Heemstra, J. E. Tyler, and H. Wahlquist. 1966. An annotated listing of the fish reference collection at the Florida Board of Conservation Marine Laboratory. Fla. State Board Conserv. Mar. Lab., Spec. Sci. Rep. 10, 121 p. MoE, M. A., Jr., and G. T. Martin. 1965. Fishes taken in monthly trawl samples off- shore of Pinellas County, Florida, with new addi- tions to the fish fauna of the Tampa Bay area. Tulane Stud. Zool. 12:129-151. Reid, G. K., Jr. 1954. An ecological study of the Gulf of Mexico fishes, in the vicinity of Cedar Key, Florida. Bull. Mar. Sci. Gulf Caribb. 4:1-94. Reinboth, R. 1962. Morphologische und funktionelle Zweigesch- lechtlichkeit bei marinen Teleostiern (Serranidae, Sparidae, Centracanthidae, Labridae). Zool. Jahrb. Abt. Zool. Physiol. Tiere 69:405-480. Robins, C. R. 1971. Distributional patterns of fishes from coastal and shelf waters of the tropical western Atlantic. In Symposium on Investigations and Resources of the Caribbean Sea and Adjacent Regions, p. 249- 255. FAO, Fish. Rep. 71.2. Robins, C. R., and W. A. Starck, IL 1961. Materials for a revision of Serranus and re- lated fish genera. Proc. Acad. Nat. Sci. Phila. 113:259-314. Smith, C. L. 1965. The patterns of sexuality and the classifica- tion of serranid fishes. Am. Mus. Novit. 2207, 20 p. Springer, S., and H. R. Bullis, Jr. 1956. Collections by the Oregon in the Gulf of Mexico. U.S. Fish Wildl. Serv., Spec. Sci. Rep. Fish. 196, 134 p. Springer, V. G., and K. D. Woodburn. 1960. An ecological study of the fishes of the Tampa Bay area. Fla. State Board Conserv. Mar. Lab., Prof. Pap. Ser. 1, 104 p. Starck, W. A., IL 1968. A list of fishes of Alligator Reef, Florida, with comments on the nature of the Florida reef fish fauna. Undersea Biol. l(l):4-40. Struhsaker, p. 1969. Demersal fish resources: Composition, dis- tribution, and commercial potential of the Conti- nental Shelf stocks off Southeastern United States. U.S. Fish Wildl. Serv., Fish. Ind. Res. 4:261- 300. 242 DESIGN AND EVALUATION OF A SAMPLER FOR MEASURING THE NEAR-BOTTOM VERTICAL DISTRIBUTION OF PINK SHRIMP, PANDALUS JORDANI Alan J. Beardsley^ ABSTRACT A shrimp sampler was constructed as one portion of a research effort dealing with the development of a fish-shrimp separator trawl. The sampler segregated shrimp caught in a series of 1-ft high vertical openings positioned between the seabed and a height of 13 ft above the seabed. Knowledge of the vertical distribution of shrimp was considered essential in the design of an efficient shrimp trawl. Results indicated that vertical dis- tributions of shrimp vary, and the amount of light striking the seabed is suggested as the triggering stimulus. Auxiliary investigations conducted with the sampler dealt with evaluations of mesh size and tickler chain. Experiments indicated that mesh sizes smaller than 2 inches restrict the passage of shrimp. The weight of shrimp caught was nearly doubled when a tickler chain was used. The sampler may have application to both shrimp biologists and commercial fishermen. Research was begun in our laboratory on trawls capable of separating pink shrimp, Pandalus jor- dani, from other marine organisms and debris while the net is being towed over the seabed (High, Ellis, and Lusz, 1969; Beardsley and High, 1970). Knowledge of the near-bottom vertical distribution of shrimp was considered essential to the design of an effective shrimp trawl since the vertical height of any bottom trawl should approximate the off-bottom distri- bution of the target species. Subsequently a multipurpose shrimp sampler was designed to facilitate investigation of shrimp distributions above the seabed by 1-ft intervals. Auxiliary investigations conducted with the sam- pler included evaluating the effects of: (1) diel or circadian movements on the abundance of shrimp near the seabed; (2) light on shrimp vertical distribution; (3) mesh size on the re- tention of shrimp and other marine organisms; and (4) a tickler chain on shrimp catch rate and vertical distribution. ' Northwest Fisheries Center, National Marine Fish- eries Service, NOAA, 2725 Montlake Boulevard East, Seattle, WA 98102. Manuscript accepted July 1972. FISHERY BULLETIN: VOL. 71. NO. 1, 1973. METHODS AND MATERIALS The shrimp sampler was designed for towing on the seabed either attached in the mouth of a conventional 57-ft Gulf shrimp trawl or directly to dandy lines without the net. All data pre- sented in this paper resulted from tows without an attached net. This fishing configuration made the sampler easier to set and retrieve; it also eliminated the variability in shrimp catches caused by differences in fishing modes. Com- mercial trawling conditions were simulated using 5- X 7-ft otterboards, 15-fm dandy lines, and a towing speed of 21/? knots. The shrimp vertical distribution sampler (Figure 1) consists of an aluminum frame par- titioned into 18 openings each measuring 1 ft vertically by 2 ft horizontally. The sampler openings are positioned in six horizontal rows (1 ft high) and three vertical columns (2 ft wide), resulting in a triplicate series of vertical samples. A vertical extension was bolted with %-inch bolts behind the sampler, permitting sampling as high as 12 to 13 ft above the seabed. 243 FISHERY BULLETIN: VOL. 71, NO. I Individual collector bags of 3/t-inch mesh' were lashed behind the sampler, thereby segregating and retaining the shrimp that entered the sep- arate sampler openings. The dandy lines were attached to the sampler at the four center pad eyes and led freely through shackles at the four corner pad eyes. This method of attachment distributed the pull of the dandy lines across the face of the sampler in the event it hit a heavy object during a tow. Aluminum trawl floats (8- inch diameter) buoyed the sampler upright dur- ing setting and retrieving, whereas water pres- sure maintained this posture while towing. During tickler chain (a device used to stim- ulate shrimp off of the seabed and into the trawl) evaluations, the chain was attached directly to the lower dandy lines with cable clamps and shackles. The length of chain was adjusted so it maintained a position 21/2 ft in front of the bottom center of the sampler as determined by scuba diver observations. Several mesh sizes' were evaluated using rec- tangular aluminum frames (separator panels) placed over the sampler openings (Figure 1). Aluminum teeth separated the panels on the front of the sampler. Although each frame had mesh of uniform size, the mesh size between frames varied between I14 and 3 inches. Com- parisons of different mesh sizes were facilitated by placing the meshes to be compared on the lateral two columns of vertical openings on the sampler. Two-inch mesh webbing was placed over the center column of the sampler during mesh size comparisons so this column could act as a gross index of relative shrimp abundance on the fishing grounds. The tow direction was reversed after each tow, and the lateral two panels were exchanged on alternate tows to re- duce any difference in catch efficiency by one side of the sampler or the other. During fishing trials on shrimp beds off the Washington coast, adequate sample sizes (50- 2,000 g of shrimp per collector bag) were ob- SIDE VIEW HK^ TVPICAL RloeTE I 8 NEEDED r— l'/2 6061 -Te PIPE scHED ao e" 0IAM6TER ALUMINUM TRAWL FLOATS 20' OF 1/2 POLYPROPYLENE LINE Vi 5' STAINLESS *OLT,j, 6 NEEDED !• B* DIAMETER ALUMINUM TRAWL FLOAT -. ^ 2^^ 606I-T6 PtFl-U- W. lVJ60U-T»^ 5/s' S061 - T6 ROO ^ Figure 1. — Construction specifications for the shrimp sampler and extension. tained from tows of 10-min duration without separator panels and 15 min with separator panels. The individual collector bags were emptied at the conclusion of each tow and the contents la- beled and frozen. In the laboratory shrimp were thawed and the weight, the carapace length (base of eyestalk to the dorsal posterior margin of the carapace), and sex recorded. PRELIMINARY EVALUATIONS OF THE SAMPLER ' All mesh sizes in this paper are stretched measure. ' The following mesh and thread sizes and materials were evaluated: 1%-inch mesh, 9 thread, nylon and acetate ; 1 V2-inch mesh, 15 thread, nylon ; 1 % -inch mesh, 18 thread, nylon ; 2-inch mesh, 12 thread, nylon ; 2 '/2-inch mesh, 21 thread, nylon; 3-inch mesh, 18 thread, nylon. Scientist-divers first appraised the shrimp sampler in Puget Sound as it was being towed on bottom in 8 fm of water at 21/2 knots. On each of eight tows the sampler was reported to 244 BEARDSLEY: DESIGN AND EVALUATION OF A SAMPLER be uprig-ht, stable, less than 3 inches off the sea- bed, and perpendicular to the direction of tow. Fishing trials were then conducted on popu- lations of pink shrimp and spot shrimp, Pandalus platyceros, in 40-60 fm of water in Dabob Bay, Wash. Most shrimp were taken near the seabed (Figure 2), but substantial numbers of pink shrimp were taken as high as 5 to 6 ft off bottom. 6r A 2 O O m I 3 Spot shrimp Pink shrimp J i_ _i_ J i_ _i_ _i_ _i_ _!_ 200 400 600 800 1,000 1,200 NUMBER OF SHRIMP Figure 2. — Vertical distribution of pink and spot shrimp taken in four tows in Dabob Bay, Wash. Numbers of shrimp are totals for the 1-ft intervals. whereas no spot shrimp were found higher than. 3 to 4 ft. This phenomenon was interpreted as a behavioral difference between the two species rather than a difference due to physical size as one might expect the larger spot shrimp to jump higher off the seabed. Subsequent fishing trials with the shrimp sam- pler were conducted on pink shrimp beds of com- mercial importance off Grays Harbor, Wash. A total of nine tows were made without separator panels and nine with 2-inch mesh across the front of the sampler. Total weights and carapace lengths for shrimp in all sampler openings in each of the vertical columns for each fishing mode were pooled (54 samples) . The means for these data are presented in Table 1. These data were further analyzed using a three-way factorial analysis of variance with the three main effects being vertical columns and horizontal rows of the sampler and repetitive tows (Table 1). In tows without separator panels the weights of shrimp in vertical columns were significantly different (F = 3.24), but a comparison of starboard and port vertical col- umns revealed their difference was not signifi- cant (F = 0.430) at the 5% significance level. The average of starboard and port colmns for shrimp weight was significantly different from weights for the center column (F = 6.06) . Anal- ysis of the data for carapace length of shrimp without separator panels and for both shrimp weight and carapace length with 2-inch sepa- rator panels indicated no significant difference Table 1. — Comparison of pink shrimp caught with and without 2-inch mesh separator panels on the front of the shrimp vertical distribution sampler. Means and F values were computed from 54 samples (i.e., six vertical openings over a nine-tow repetition). Type of Degrees or Critical F value at 5% Without panels With 2- nch panels comparison freedom significance level Weight Length Weight Length Mean' Starboard column 176.0 18.2 81.2 18.1 Middle column 156.0 17.9 S3 .3 17.8 Port column 172.0 i8.3 86.6 19.9 F value All vertical columns 2,80 3.11 3.24 21.20 20.529 20.699 Middle vs. V2(starboard ■+ port) 1,80 3.96 6.06 »2.32 "0.017 20.831 Starboard vs. port 1,i80 3.96 20.430 20.074 2 1.04 20.56 Horizontal rovrs 5,80 2J3 10.-4 6.28 4.65 3.78 Repetitive tov/s 8,80 2.05 52.0 5.81 15.0 4.71 * Weight in grams,- length in millimeters. 2 Not significantly different at 5% significance level. 245 FISHERY BULLETIN: VOL. 71, NO. 1 between vertical columns. Since the differences in shrimp catches between port and starboard vertical columns were not significant during tows either with or without separator panels, later comparisons of the effect of mesh size on shrimp catches were made between the starboard and port vertical columns. As might be expected, significant differences were evident for both horizontal rows on the sampler and repetitive tows. The differences in shrimp vertical distribution (horizontal rows) are discussed later in the paper. Differences in catch during repetitive tows were anticipated as shrimp samples were collected over several months on several shrimp beds. DAYTIME VERTICAL DISTRIBUTION OF SHRIMP In March 1969 a series of eight tows were made with the shrimp sampler in 71 fm off Destruction 6r 5 - §4 o CD 6r 3 - o z < a " < 1 _ Pinheads ( Marketable shrimp / ' \ ' f 1 1 1 1 I - 10 20 30 40 50 60 70 80 90 100 PERCENT Figure 3. — Percentages of pinheads (shrimp less than 15-mm carapace length) and marketable shrimp in catches off Destruction Island, Wash., taken by the shrimp sampler without separator panels (four tows). The percentages indicate the percent composition at each 1-ft interval. Total weights for both pinhead and mar- ketable shrimp increased near the seabed. o 4 I- t- o o z < I- o MarKetable shrimp _i_ _!_ _!_ -L. _!_ _!_ 10 20 30 40 50 60 70 80 90 100 PERCENT Figure 4. — Percentages of pinheads (shrimp less than 15-mm carapace length) and marketable shrimp in catch- es off Destruction Island, Wash., taken in four tows of the shrimp sampler with separator panels. The per- centages indicate the percent composition at each 1-ft interval. Total weights for both pinhead and marketable shrimp increased near the seabed. Island, Wash. Principal objectives were to de- termine if differences in number and size com- position of shrimp would occur with increasing height off bottom. Four hauls were made with- out separator panels and four hauls with uni- form 2-inch mesh across the front of the sam- pler. The greatest number of unmarketable (less than 15-mm carapace length) and mar- ketable (15-mm carapace length or greater) shrimp were taken in sampler openings near the seabed. The ratio of marketable to unmarket- able (pinhead) shrimp increased with distance off bottom (Figures 3 and 4). The relationship between distance off bottom and length-frequen- cy of shrimp is s-hown in Figure 5 for the same series of tows. This figure indicates that shrimp were most abundant near the seabed and a high percentage were small. Additional tows with the sampler under a va- riety of conditions (weather, time of day, season) 246 BEARDSLEY: DESIGN AND EVALUATION OF A SAMPLER NO SEPARATOR PANELS TWO- INCH MESH SEPARATOR PANELS 8 12 16 20 24 8 12 16 20 24 CARAPACE LENGTH (mm) Figure 5. — Relation of length frequency of shrimp to distance off bottom for shrimp sampler tows off Destruc- tion Island, Wash. Four tows were made with no sep- arator panels and four with 2-inch separator panels. N is the total number of shrimp taken in the sampler openings with the cumulative totals for four tows shown at the top of the figxire. indicated the vertical distribution of shrimp is not static but subject to dynamic changes, often in a brief period of time. Representative mid- morning tows for three different days are pre- sented in Figure 6, Widely divergent shrimp distributions are evident in this figure. The most rapid alteration in shrimp distribution was observed between midmorning tows on 22 and 23 June (Figure 6). Over the same time pe- riod the weather changed from partly cloudy to complete overcast with rain. Commercial shrimp fishermen trawling on the same shrimp grounds reported a reduction in shrimp catches accom- panying the change in cloud cover. As commer- cial shrimpers drag their trawls on bottom re- gardless of weather, it is conceivable that their nets were passing under large numbers of shrimp. During sunny weather with relatively clear water, shrimp were found concentrated near the seabed as on 9 August in Figure 6. DIEL MOVEMENTS OF SHRIMP Diel variations in shrimp distribution were investigated with the sampler during three day- night cycles using the shrimp sampler. All tows were 15 min in length and were made with 2- inch separator panels on the sampler. During any single diel cycle, all tows were made in the same direction and in the same location as indi- cated by the vessel's compass, depth sounder, and loran. Vertical distributions for pink shrimp collected during three diel cycles are presented in Figure 7. Tows on 24 March were conducted under a cloudless sky with the greatest quantity of shrimp taken early in the day. Greatly reduced catches were taken at night. In general, shrimp appeared to be higher above the seabed during early morning and evening tows. The weather preceeding 28 March was over- cast with rain. Tows during 28 March (Figure 7) indicated shrimp were not plentiful near the 8-1 7- 6- 5- 4- 3- 2- I- 8- 7- 6- 5- 4- 3- 2- I- Overcast (90% cloud cover) June 22, 1969 off Grays Horbor, Wash. Overcost (100% cloud cover) June 23, 1969 off Groys Horbor, Wo$h. T 1 1 Clear (10% cloud cover) August 9, 1969 off Destruction Island, Wash. 100 200 300 400 WEIGHT OF SHRIMP (g) Figure 6. — Examples of three widely divergent shrimp vertical distribution patterns in the shrimp sampler on three different tows. 247 FISHERY BULLETIN: VOL. 71, NO. 1 10,000-1 ■=^ 5,000- o V E I- -S 5,000- 8C O o £ I- 61 2 »5 5,000- o li Is (A O 5-° Cleor (20% cloud cover) Off Destruction Islond in 7l-72fm Worch 24,1969 T 1 1 1 1 r Sunrise Overcost ( 100 % cloud cover) Off Groys Horbor in 7l-73fm March 28, 1969 Sunset T — I 1 r T r Sunrise Overcast (90% cloud cover) Off Groys Horbor in 70-71 fm June 23, 1969 Sunset T r T 1 1 1 r — I r 0000 0600 T r T r T 1 1 r 800 T 1 0000 Figure 7. — Three different diel distributions of pink shrimp collected with the shrimp sampler. The vary- ing width of each diagram represents the percentage proportion of the catch taken at various distances from the bottom of the sampler. seabed. The greatest quantity of shrimp were taken during the day. The tows on 23 June presented a valuable opportunity to observe the changes in shrimp vertical distribution accompanying a change in cloud cover. The 2 days preceeding 23 June were sunny. Tows with the shrimp sampler on these days indicated shrimp were concentrated near the seabed during daylight hours. Com- mercial fishermen working near the sample sta- tion were making good catches of shrimp. On the evening of 23 June a zone of low baro- metric pressure moved into the area with result- ing cloud cover and rain squalls on 23 June. Catches by commercial fishermen dropped to a fraction of those taken on previous days. The vertical distribution sampler indicated shrimp were well off bottom (Figure 7) with reduced shrimp abundance, compared with tows made on previous days. A most interesting situation oc- curred during tows made in twilight and after dark as the catch rates were much higher than in tows made in the preceeding daylight hours. MESH SIZE EVALUATIONS The effect of mesh size on the movement of shrimp was investigated by placing web of two different sizes on separator panels over the port and starboard vertical columns. A total of six different mesh sizes were evaluated in this man- ner. Two-inch mesh was placed over the center 248 BEARDSLEY: DESIGN AND EVALUATION OF A SAMPLER Table 2. — Comparison of pink shrimp caught behind separator panels of differing mesh size. Means and F values were computed from 24 samples (i.e., six vertical openings over a four-tow repetition). F value Item Mesh size (inches) 1.0 Mean weight of shrimp (g) 27.6 Mean length of shrimp (mm) 14.7 Mesh size (inches) 1.75 Mean weight of shrimp (g) 166.8 Mean length of shrimp (mm) 19.1 Mesh size (inches) 2.5 Mean weight of shrimp (g) 231.3 Mean length of shrimp (mm) 19.4 All vertical columns^ Middle s^. V2(5tarboard + port) 2 Port vs. starboard^ 2.0 1.5 329.3 112.7 190.0 202.0 16.3 16.9 17.6 13.2 32.03 24.3 2.0 2.0 225.7 243.7 11.4 3 1.98 20.9 19.1 19.2 30.605 30.330 30.870 2,0 3.0 222.0 260.7 9.67 32.60 4.71 19.3 19.2 30.948 30.013 31.83 ^ The F value at the 5% significance level is 3.32 with 2 and 30 degrees of freedom. - The F value at fhe 5% significance level is 4.17 with 1 and 30 degrees of freedom. 3 Not significantly different at 5% significance level. vertical column as a control or index of the rel- ative abundance of shrimp available.' The po- sitions of the separator panels being evaluated (port and starboard) were switched after each set of two tows, and tow direction was altered 180° after each' haul. This was done to reduce any sampling bias attributable to one side of the sampler or the other. Two different mesh sizes were compared over four consecutive tows. The relationship between mesh size and the catch of shrimp as determined by tows with the shrimp sampler is shown in Table 2. The mean sample weight of shrimp catches increased di- rectly with the mesh size, but the rate of increase diminished in mesh sizes above 2 inches. Dif- ferences between the total weight of catches taken with 21/2-inch mesh and 3-inch mesh were significant {F = 4.71), but difference in the mean weights (231.3 and 260.7 g) for these mesh sizes was less than the differences between the means for 1-inch and 1 1/2-inch mesh (27.6 and 112.7 g) and for 134-inch and 2-inch mesh (166.8 and 243.7 g). These differences are also sug- * In the situation where small mesh sizes (i.e., 1% and l'/2 inches) were being compared on the lateral vert- ical openings of the sampler, there is the possibility that shrimp catches in the center vertical opening (2-inch mesh) would be proportionally greater than when larger mesh sizes (i.e., .21/2 and 3 inches) were being compared. The likelihood of shrimp avoiding the small mesh was considered remote on the basis of diver observations of fish pinned against the web in the separator panels and the presence of large numbers of gilled shrimp in small- mesh web on the separator panels at the conclusion of a tow. gested by the computed F values in Table 2. Fig- ure 8 shows this relation between mesh size and catch from the data in Table 2. In this figure mean weight and carapace length (means) of shrimp catches for the six different mesh sizes have been converted into a percentage of the weight and length (means) taken by the center vertical column (control) with its standard 2- inch mesh separator panel. In contrast, the mean carapace length of shrimp did not differ significantly with increas- ing mesh sizes above II/2 inches (Table 2 and Fig- ure 8). This indicates that the larger shrimp are able to pass readily through the 1 1/2-inch mesh if the meshes are open and held at right angles to the current. Meshes in the interme- diate and cod end of conventional west coast shrimp trawls are II/2- to 1%-inch mesh. How- ever, shrimp do not pass through these meshes in large quantities because the meshes are nor- mally partially closed and nearly parallel with the current (High et al., 1969). EFFECT OF TICKLER CHAIN ON SHRIMP CATCHES Commercial shrimp fishermen in the north- eastern Pacific commonly employ a contrivance called a "tickler chain" to excite shrimp off bot- tom in an effort to increase their catches. This chain is several feet shorter than the trawl foot- rope with the ends attached to the ends of the foot rope. In this configuration the tickler chain 249 FISHERY BULLETIN: VOL. 71, NO. 1 Table 3. — Shrimp catches with and without a tickler chain attached to the shrimp vertical distribution sampler. Location Date Off Grays Harbor, Wash. 6-23-69 Off Destruction Island, Wash. 6-24-69 8-29-69 11-13-69 Depth (fm) Shrimp catch' (g) Time Without tickler chain With tickler chain 1 150-1205 73 2,270 1256-1311 73 4,561 1345-1400 73 3,208 1425-1440 73 8,198 1515-1530 73 3,696 1610-1625 73 4X)8I 09150930 76 221 0958-1012 76 S89 1040-1055 76 4,064 1125-1140 76 1,276 0655-0710 66 3,376 0820^835 66 9,983 0915-0930 66 3,451 1000-1015 66 8,900 0900-0918 69 1,887 1O0O-1O15 69 5,364 1103-1118 69 1,942 1232- 1247 69 3,906 1327-1342 69 1,230 1415-1430 69 912 1508-1523 69 1,569 1605-1620 69 3,574 Total 26,914 51,644 precedes the footrope as the net is being towed on bottom. Even though the tickler chain is accepted and used by most shrimp fishermen under the assumption that it increases catches, very little is known about the effect of this chain on either the quantity of shrimp caught or the vertical distribution of shrimp as they enter the trawl. Utilizing the unique features of the shrimp sampler, a brief examination of the effects of tickler chain on shrimp height distribution was undertaken. Unwanted fish and debris were eliminated from the shrimp catches by placing 21/2-inch mesh webbing on separator panels over the entire face of the sampler. On alternate tows a length of %-inch tickler chain was placed 21/2 ft in front of the bottom of the sampler. All comparative tows were made in the same direction and over the same track line as deter- mined by loran, compass, and depth sounder. Over a 4-month period 22 comparative hauls were made in this manner. Table 3 shows the catch of shrimp by weight for these tows. Aggregate catches with the tick- ler chain were nearly twice as great as those without the chain. Summation of the total shrimp weights for each sampler height revealed that when the tickler chain was used, a greater proportion of shrimp were taken in the lower sampler openings in contrast to higher distribu- tion without the chain (Table 4). The effects of a tickler chain on the size distribution of shrimp were also examined. The carapace lengths from shrimp in 200-g samples from each sampler opening were measured. These data indicate that at all distances off the bottom, there Table 4. — Comparison of shrimp catches at various distances off the bottom in the shrimp vertical distribu- tion sampler with ( + ) or without ( — ) the tickler chain. These data were computed from the 22 tows presented in Table 3. Distance bottom off (ft) Weight (g) Percent of catch for each height interval - + +/- + 1 5,476 15,594 2.8 13.2 30.2 2 6,363 17,064 2.8 11.9 33.0 3 4,859 8.047 1.7 12.8 15.6 4 3,445 4,159 1.2 18.1 8.0 5 3,211 3,375 1.1 23.7 6.5 6 3,555 3,465 1.0 Total 20.3 6.7 100.0 100.0 250 BEARDSLEY: DESIGN AND EVALUATION OF A SAMPLER was no significant difference between sample means for shrimp carapace lengths in catches made with or without the tickler chain (Table 5) . Table 5. — Comparison of mean carapace length of shrimp (computed from 200-g samples) taken in the shrimp vertical distribution sampler with ( + ) or with- out ( — ) the tickler chain. These data were computed from the 22 tows presented in Table 3. The F value at the 10% significance level is 2.79 with 1 and 64 degrees of freedom. Distance off (ft) Mean carapace lengt+i (mm) F value bottom — + 1 1<5.2 16.7 0.856 2 16.7 17.0 0.616 3 17.1 17.3 0.299 4 17.2 17.6 0.830 5 17.3 17.1 0.092 6 17.5 17.3 0.133 DISCUSSION Confidence in the shrimp sampler as a research tool capable of indicating the height of shrimp near the seabed resulted from diver observa- tions, trial tows in Dabob Bay, and statistical evaluations of catches made on offshore shrimp grounds. Neither daylight distributions of shrimp off the seabed nor abrupt changes in ver- tical distributions were expected. The param- eters responsible for changes in distribution were not investigated, but other papers (Schaef- ers and Johnson, 1957; Schaefers and Powell 1958; Alverson, McNeely, and Johnson, 1960; Pearcy, 1970) indicate pink shrimp follow diel or circadian movements which may be triggered by changes in light levels. Similar differences in illumination during daylight hours may be the cause of more subtle alterations in shrimp dis- tribution near the seabed. Tows with the shrimp sampler indicated shrimp were further from the bottom during daytime tows under overcast skies. Commercial fishermen using bottom trawls commonly report reductions in shrimp catches when they encounter turbid water and overcast weather, or both. This evidence suggests changes in catch rates for pink shrimp are not due entirely to endog- enous factors as their distribution appears to change in direct response to light intensity re- gardless of the time of day. Exogenous rhyth- micity in crustaceans has been demonstrated previously by Skud (1968) where during a total eclipse of the sun over Maine in July 1963, cru- staceans exhibited a variety of responses from no response to movements toward or away from the surface at totality. Data collected with the shrimp sampler suggest pink shrimp will move off bottom during the day (probably to feed) if the light intensity is reduced enough to present some protection from predators. These move- ments may also be directly associated with mi- grations of prey. One explanation for relatively dense concentrations of shrimp near the seabed at night during overcast weather is that these shrimp were able to come off the bottom during the day to feed and returned to the seabed when they became satiated. The shrimp sampler indicated the highest proportion of small shrimp occur near the sea- bed. Commercial fishermen could avoid these shrimp by "flying" their nets 2 ft off bottom, but this strategy would also eliminate significant numbers of large shrimp from the catch. Sep- aration and release of small shrimp once they enter the trawl is a difficult if not impossible task. Experiments with the shrimp sampler indicated that the mesh size necessary to segregate shrimp by size is extremely critical even in fully opened meshes at right angles to the current (Figure 8) . The problem is compounded in a trawl where few meshes are fully opened and their orientation to the current varies. Moreover large numbers of shrimp are swept in mass into the trawl cod end and may never encounter trawl meshes. These factors reveal the futility of shrimp trawl mesh size restrictions when the design of the trawl is not considered. Mesh size comparisons indicate that the op- timum mesh size for separator panels in shrimp separator trawls should be 2 inches (Figure 8). Smaller mesh would reduce the catch of shrimp, and a larger mesh size would not appreciably increase the size or catch of shrimp captured but could introduce more contaminating organ- isms into the catch. These results compare fav- orably with fishing trials of shrimp separator trawls with different mesh sizes in the separator panel. Two-inch separator panels are now stan- dard in this gear. 251 FISHERY BULLETIN: VOL. 71, NO. 1 I40r MESH SIZE (inches) Figure 8. — Relationship between mesh size of separator panel and the means of shrimp length and total weight. The values for the mean carapace length and mean sam- ple weight are compared with the shrimp catches (means) taken behind the 2-inch mesh separator panels (control). The tickler chain caused a significant increase in the number of shrimp entering the lower por- tion of the shrimp sampler. This phenomenon suggests that the tickler chain either excites shrimp into the water column that normally would pass under the sampler or confuses shrimp so that they are unable to avoid the sampler. The fact that pink shrimp can be readily taken in a plankton sampler with an opening 15 cm. in diameter indicates that avoidance may not be an important consideration. The major effect of the tickler chain is probably to move shrimp ver- tically where they are vulnerable to capture by the trawl. The utility of the shrimp sampler in describing the vertical distribution of shrimp suggests that it would be a valuable tool for providing sup- portive evidence necessary for sound manage- ment decisions regarding shrimp resources. For example, in California, trawl surveys of the shrimp beds are used to estimate the quantity and quality of shrimp available and hence estab- lish the quota size ( Abramson, 1968) . These an- alytical estimations of the standing stock of a shrimp bed are based on the assumption that catch per unit of effort is a function of stock density in the area being surveyed and that changes in catch per effort are directly propor- tional to changes in density (Ricker, 1958; Gul- land, 1964). Equations relating population size to catch per effort for trawl gear (Alverson and Pereyra, 1969) require some knowledge of the vulnerability of shrimp within the influence of the trawl and the proportion of the total shrimp stock in the water column sampled by the trawl. Estimates for vulnerability of shrimp have tra- ditionally been placed near 1 largely because of a lack of knowledge regarding the behavior of shrimp to trawls. With consideration of the size of the trawl opening and the erratic escape movements of shrimp, a vulnerability coefficient of 1 may be relatively accurate. However, use of the shrimp sampler has shown that the co- efficient of catchability for a shrimp trawl which is towed a constant distance off bottom may vary dramatically from day to day. Often the catch coefficient would not approach 1 for a trawl having a 4-ft vertical opening. Thus the over- all coefficient, which is a product of the vulner- ability and catchability coefficient, may at times be considerably less than 1 if shrimp are being sampled with a conventional shrimp trawl. Moreover, the value for the overall coefficient would vary from day to day and reduce the ac- curacy of population size estimates. The tickler chain must also have an important effect on the vulnerability coefficient, but this relation has not been explored. The greatest utility of the shrimp sampler in providing estimates of standing stock may be realized when the sampler is towed alternately with a standard trawl. This approach has been taken by scientists at the Auke Bay Fisheries Laboratory, National Marine Fisheries Service, NOAA, Auke Bay, Alaska (James Olson, pers. comm.) where estimates were being made of the standing stock of shrimp in Kachemak Bay, Alaska. In this instance the sampler was used to determine the catchability coefficient of the standardized shrimp trawl used in the standing stock estimates. The sampler has also been towed alternately with trawls in an experiment to measure the fishing power of four dissimilar shrimp trawls near Kodiak, Alaska (Lael L. Ronholt, North- 252 BEARDSLEY: DESIGN AND EVALUATION OF A SAMPLER west Fisheries Center, National Marine Fisher- ies Service, NOAA, Kodiak, Alaska, pers. comm.) Biologists from the Fish Commission of Oregon (Jack G. Robinson, per. comm.) now insert a modified sampler, 2 ft wide by 9 ft tall, in the mouth of a conventional 41-ft Gulf shrimp trawl to achieve estimates of trawl efficiency. ACKNOWLEDGMENTS The enthusiastic support and advice of Rich- ard M. McNeeley, Manager of the Conservation Engineering Program in the Division of Marine Fish and Shellfish, Northwest Fisheries Center, Seattle, Wash., are gratefully acknowledged. Donald D. Worlund, Division of Fisheries Data and Management Systems, Northwest Fisheries Center, Seattle, helped in the statistical evalu- ation. I also thank Robert P. Larsen, Skipper, and the crew of the NMFS research vessel John N. Cobb and Olaf Engel, Skipper of the chartered vessel Baron, for aid in the collection of shrimp samples. LITERATURE CITED Abramson, N. J. 1968. A probability sea survey plan for estimating relative abundance of ocean shrimp. Calif. Fish Game 54:257-269. Alverson, D. L., R. L. Mc^eely, and H. C. Johnson. 1960. Results of exploratory shrimp fishing off Washington and Oregon (1958). Commer. Fish. Rev. 22(1) :1-11. Alverson, D. L., and W. T. Pereyra. 1969. Demersal fish explorations in the northeast- ern Pacific Ocean — an evaluation of exploratory fishing methods and analytical approaches to stock size and yield forecasts. J. Fish. Res. Board Can. 26:1985-2001. Beardsley, a. J., AND W. L. High. 1970. Shrimp sorting trawls in the Pacific North- west. Natl. Fisherman 50(13) :49, 51-52. Gulland, J. A. 1964. Catch per unit effort as a measure of abun- dance. Cons. Perm. Int. Explor. Mer, Rapp. P.-V. Reun. 155:8-14. High, W. L., I. E. Ellis, and L. D. Lusz. 1969. A progress report on the development of a shrimp trawl to separate shrimp from fish and bottom-dwelling animals. Commer. Fish. Rev. 31(3):20-33. Pearcy, W. G. 1970. Vertical migration of the ocean shrimp Pandalus jordani: a feeding and dispersal mech- anism. Calif. Fish Game 56:125-129. RiCKER, W. E. 1958. Handbook of computations for biological sta- tistics of fish populations. Fish. Res. Board Can., Bull. 119, 300 p. SCHAEFERS, E. A., AND H. C. JOHNSON. 1957. Shrimp explorations off the Washington Coast, fall 1955 and spring 1956. Commer. Fish. Rev. 19(l):9-25. SCHAEFERS, E. A., AND D. E. PoWELL. 1958. Correlation of midwater trawl catches with echo recordings in the northeastern Pacific. Com- mer. Fish. Rev. 20(2): 7-15. Skud, B. E. 1967. Responses of marine organisms during the solar eclipse of July 1963. U.S. Fish Wildl. Serv., Fish. Bull. 66:259-271. 253 A REVIEW OF THE CHEMICAL AND NUTRITIVE PROPERTIES OF CONDENSED FISH SOLUBLES Joseph Scares, Jr.,' David Miller,' Susan Cuppett,^' and Paul Bauersfeld, Jr.'' ABSTRACT This paper attempts to review some of the pertinent data available on the chemical and nutritive properties of condensed fish solubles with special reference to condensed men- haden solubles. Information concerning other kinds of fish solubles from various published sources is tabulated for comparison. The results of three analytical and three biological studies of menhaden solubles are reported. Data show that menhaden stickwater contains an average of 4.8% protein, 93.4% water, and about 1.25% lipid. The solids are concentrated about seven times by vacuum evap- oration to yield condensed solubles. Data show that this product contains about 32% protein, 49% water, and 11% fat. Additional data are presented on the content of amino acids in fish solubles, which show that the indispensable amino acid tryptophan is rel- atively low and. on the content of essential inorganic elements which are shown to be present, generally in acceptable levels. Fish solubles are shown to be rich sources of choline, niacin, pantothenic acid, riboflavin, biotin, and vitamin Bj2. An experiment with broiler chicks showed that the average metabolizable energy con- tent of 10 samples of condensed menhaden solubles from commercial plants located along the Atlantic and Gulf coasts was 2.03 kcal/g on an "as fed" basis. Average protein and fat digestibility was 89 and 57%, respectively. Further studies with broiler chicks showed that significant unidentified growth factor responses could be obtained when 5% menhaden solubles was used in diets. A significant response was obtained even when the chicks were reared in a gnotobiotic environment. This indicates that the presence of bacteria is not necessary for solubles to produce a growth response in chicks. This review attempts to discuss some of the more pertinent recent published and unpublished data available on the chemical and nutritive proper- ties of condensed fish solubles. For many years stickwater was considered as a waste effluent and was not used in animal feeds or in other manufacturing processes. However, with the postwar surge in production of fish meal, this by-product began to be recognized for its nu- tritional value. In the presynthetic nutrient era, condensed fish solubles was recognized as an out- standing source of B vitamins and minerals for poultry and swine. Later, solubles was the prin- ' College Park Fishery Products Technology Labora- tory, National Marine Fisheries Service, NOAA; present address: Department of Poultry Science, University of Maryland, College Park, MD 20740. " College Park Fishery Products Technology Labora- tory National Marine Fisheries Service, NOAA, College Park, MD 20740. ciple source of the "animal protein factor" or vitamin B12. Today we still consider condensed fish solubles to be valuable for these purposes, but now more emphasis is put on its unidentified growth factor content. There are various sources and kinds of con- densed fish solubles available in the United States. However, about 80 9r of the solubles used in this country are produced by the men- haden reduction industry located along the At- lantic and Gulf coasts. Since 1963, the quantity of solubles available for animal feedstuff sup- plementation in the United States has averaged about 90,000 tons annually. However, because of the difficulty in handling, storing, and mixing the viscous condensed fish solubles, a considerable quantity of this product is dehydrated on a carrier to produce a free- flowing dry product. One such product is "full Manuscript accepted September 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. 255 FISHERY BULLETIN: VOL. 71, NO. I meal" containing 1 part condensed solubles to 3.5 parts of fish meal. This recombination of the two fish products is a logical approach as it is really equivalent to the original fish minus a major part of the original oil. Numerous other feedstuffs, such as soybean meal, alfalfa meal, or wheat middlings are also used to produce a free-flowing dry product consisting of one-half carrier and one-half dried condensed solubles. PROCESS METHOD Condensed fish solubles is one of two by- products produced by the wet reduction of whole fish to fish meal. Briefly, the production of con- densed fish solubles by the "wet reduction" pro- cess involves cooking whole fish (e.g., menhaden) and subjecting the cooked fish to screw-type presses. The liquid fraction which passes through the screens of the presses is collected and separated by first passing through a vibrat- ing screen to remove suspended material. The remaining liquor is either pumped into settling tanks for gravity separation or, more commonly, centrifuged to separate the oil and aqueous por- tion. Usually the centrifugation system consists of a three-phase centrifuge designed to separate oil, water, and suspended material. The aqueous portion in these centrifuges is collected in large storage tanks. During this storage process usu- ally a quantity of sulfuric acid, or occasionally phosphoric acid, is added to the stickwater until a pH of 4.5 is reached (normal pH 6-7.0). The storage tanks are heated by steam coils to a tem- perature of approximately 150°F for 30 min to coagulate the proteins. The suspended and co- agulated proteins are removed by centrifuges, and the free oil is removed by oil-separating cen- trifuges. The solid materials are dried on press cake and the liquid is called "stickwater." Table 1 shows the composition of several samples of stickwater produced in 1970. These samples con- tained an average of about l^/r total solids and about 5% protein. The amino acid profile in- dicates that this protein is gelatinous in nature because of the relatively high glycine, proline, and alanine content. Stickwater is then concentrated by means of vacuum evaporation, generally operated in units of three or more. Thus, the terms "triple effect" or "quadruple eflFect" are applied to the evap- oration procedure. The evaporators may be hor- izontal or vertical, depending on the position of the steam tubes used for heat. Each evaporator consists of a series of tubes, usually vertical, hold- ing the liquor and surrounded by steam. The tubes passing through the evaporator allow the vaporized water to escape at the top, and the partially evaporated product passes to the next unit at a lower position in the evaporator. These evaporators are subjected to reduced pressure. The first unit is usually heated by low pressure exhaust steam from the boilers and is generally maintained at about 24 inches of mer- cury pressure (or under a slight vacuum). As the liquor reaches a predetermined concentra- tion, it is drawn into the second unit. The vapors from the boiling liquor in the first unit are passed through the steam tubes of the second, and the liquor continues to boil by increasing the vac- uum to about 28 inches of mercury. In the third stage the process is either com- pleted under a vacuum of about 30 inches of mercury or continued depending on the number of stages employed. When the liquor reaches approximately 50% solids, the process is fin- ished and the solubles are stored. Thus, during the process the original material containing about 93% water is concentrated by a factor of at least seven. Other methods of condensing fish solubles have been and are used but not exten- sively enough to warrant further discussion here. Table 1. — Chemical composition of Atlantic menhaden stickwater.' Item Amount Item Amount % % Protein (N X 6.25) 4,82 Valine 2.36 Moisture 93.43 Methionine 1.37 Ash 1.12 Isoleucine 1.59 Total fat 1.26 Leucine 3.34 Ethei* fat 0.83 Tyrosine 1.02 Phenylalanine 2.12 % nitrogen Tryptophan 0.31 Ammonia 0.034 Taurine 3.55 Asparctia acid 5.20 % of protein Serine 2.31 Lysine 4.45 Glutamic acid 8.22 Histidine 4.19 Proline 5.10 Arginine 1.89 Glycine 10.45 Threonine 2.06 Alanine 5.86 ^ Values are averages of data from three analyses. 256 SCARES ET AL.; CONDENSED FISH SOLUBLES Table 2. — Analytical data on condensed solubles. Menhaden solubles Pacific sardine solubles Item March Lassen et at. Lea (1956) (1962) (1951) Dry solids, % 47.6 ±3.5 50.3 50.0 Crude protein, % (N X 6.25) 33.5 ±2.7 33.7 32.0 Ammonia, % 1.3 ± 1.1 _. _^ Corrected protein, ^ % 27.2 ±3.8 _^ _^ Fat, % 6.4 ±3.2 4.3 7.0 TotafI ash, % 9.2 ±2.0 9.2 12.0 Water-insoluble matter. % 4.5 ± 1.6 _^ __ Water-insoluble ash, % 0.28 ± 0.21 »« _^ pH 4.64 ± 0.54 4.5 _. Specific gravity 1.19 ±0.03 1.2 — Number of samples 32 ? ? ^ Corrected for ammonia-nitrogen 5.15. COMPOSITION OF CONDENSED FISH SOLUBLES Analytical data (Table 2) published by Lee (1956) showed that condensed fish solubles con- tained approximately 50% dry matter, which in- cluded protein (Kjeldahl nitrogen x 6.25), fat, minerals, and vitamins. According to the above author the major variable factors affecting com- position were species and age of the fish, season and location of the catch, handling techniques, and type of plant equipment used during pro- cessing. Data describing other types of fish sol- ubles produced in the United States are very scanty. However, Table 2 lists some average results for herring solubles reported by March (1962) and Lassen, Bacon, and Dunn (1951), and these appear very similar to the menhaden data. Table 3 shows the proximate, total fat, and ammoniacal nitrogen composition of 24 con- Table 3. — Proximate composition, total fat, and ammo- niacal nitrogen content of menhaden fish solubles.^ Analyses Protein (N X 6.25) Ash Ether fat Total fat Moisture .Ammonia as % N ^ Scares, MiHer, and Ambrose (1970). Average of 24 samples % 31.8 ±2.52 7.8 ± 0.93 8.9 ±2.13 11.2 ±2.09 48.7 ±2.18 0.5 ±0.23 densed menhaden fish solubles samples collected during 10 months of the 1969 fishing season (Scares, Miller, and Ambrose, 1970). These samples were obtained from seven different fish meal plants located along the Middle Atlantic and Gulf coasts, and they represent regular com- mercial production. The methods for proximate analyses were all standard AOAC (Association of Official Analytical Chemists) methods except total fat, which is a chloroform-methanol extrac- tion method of Bligh and Dyer that was mod- ified by Smith, Ambrose, and Knobl (1964). Since the data from the Atlantic and Gulf sam- ples did not differ significantly (although they are made from two different species of men- haden), it is presented together as an average. In general, these analyses indicate that the solids content is slightly more than 51 % . The average concentration of protein (Kjeldahl N X 6.25) was 31.8% and that of ash was 7.8%. If these values are converted to a dry basis, they are very similar to those for menhaden fish meal reported by Kifer and Payne (1968) and Kifer, Payne, and Ambrose (1969). This is not to say that dried fish solubles is equivalent to fish meal in protein quality, but this only points out the rel- atively high protein content of this material even after allowing for nonprotein nitrogen (Table 3) . The fat content, expressed on a dry basis is some- what higher, however, than that found in fish meal. Ammoniacal nitrogen (which included volatile amines, etc.) makes up about 0.5% of the total nitrogen. 257 FISHERY BULLETIN: VOL. 71, NO. 1 PROTEIN Biologically, the protein in solubles is consid- ered inferior in quality because of its high con- tent of gelatin, derived from fish collagen, which is low in the sulfur amino acids and almost com- pletely devoid of tryptophan. Therefore, the essential amino acids content of the protein in condensed fish solubles is not adequate as a sole source of protein for the growth of chicks be- cause tryptophan is deficient and the content of the other essential amino acids in condensed fish solubles is lower than that found in fish meal. Nevertheless, the protein may serve very well in a diet as a minor supplementary protein or in a diet otherwise adequate in amino acids since a 3% level of fish solubles contributes only 1% protein to the total diet. AMINO ACIDS Table 4 lists the amino acid composition of menhaden fish solubles published by Soares et al. (1970) and shows that solubles do contain some of the amino acids in reasonably high amount. All amino acids except available lysine, trypto- phan, and cystine were analyzed by ion exchange column chromatography. Available lysine was determined by the method of Carpenter (1960). Tryptophan was determined by the method of Spies and Chambers (1948), and cystine was analyzed by a microbiological assay. It is noteworthy that cystine averaged 1.05% of protein, which is slightly higher than the 0.9% of protein found in menhaden fish meal (Kifer and Payne, 1968) . Glycine, an important amino acid in poultry rations since it is necessary for maximum growth of chicks, is the only other essential amino acid found in fish solubles at levels higher than in fish meal. As mentioned earlier, tryptophan is the first limiting amino acid based on chemical composition. Available lysine averaged about 83% of total lysine which is similar to the fish meal values that we have obtained. The taurine content is also shown, inasmuch as a recent report by Monson (1969) indicates that this amino acid may be one of the unknown growth factors. Table 4 also shows the amino acid composition of sardine and herring solubles. For the most part these data are simi- lar to menhaden. However, some large differ- ences exist in the histidine levels which are not explainable. Table 4. — Amino acid content (% of protein) of various kinds of con- densed fish solubles. Amino Menhaden solubles Herring solubles' Pacific herring solubles Pacific sardine solubles acid Soares et al. (1970) Loksesvela (1954) Ewing (1963:291) Lassen et al. (1951) — . % of Pfotfin — — — Arginine 3.9 ±0.65 5,2 4.5 4.3 Histidine 2.5 ± 0.80 0.7 6.0 5.8 Lysine 4.8 ±0.56 4.3 4.3 4.9 Tyrosine 1.1 ±0.20 0.7 __ _^ Tryptxjphan 0.3 ± 0.07 0.1 0.4 0.4 Phenylalanine 2.1 ±0.27 1.5 3.2 2.3 Metihionine 1.6±0.17 \A 1.5 1.5 Cystine I.l ±0.28 0.1 03 _^ Threonine 2.1 ±0.25 2.1 1.8 2.4 Leucine 3.7 ±0.44 3.2 3.0 4.7 Isoleucine 1.9 ±0.28 1.9 2.3 2.7 Valine 2.7 ±0.35 2.5 3.4 3.0 Glutamic acid 8.1 ±0.94 7.5 _^ 8.4 Glycine 8.7 ± 1.50 10.9 10.9 _^ Prdline 4.5 ±0.60 __ _. 6.7 Taurine 3.2 ±0.50 ? _^ _^ Available lysine 4.0 ± 0.62 _^ M ^^ No. of samples 24 — ? ? 1 Unknown origin. 258 SCARES ET AL.: CONDENSED FISH SOLUBLES MINERALS The ash content of condensed solubles aver- ages about 99^ and occasionally is as high as 13% of total weight (Scares et al. 1970). The major inorganic elements (Table 5) are P, Na, and K. Table 5. — Inorganic elements content of menhaden fish solubles.' Inorganic Average Range elements Low High Percent — — . — Percent K 1.57 ± 0.18 1.06 1.87 Na 1.14± 0.28 0.70 1.67 P 0.56 ± 0.14 0.20 0.80 Mg 0.11 ± 0.03 0.08 0.17 Ca 0.06 ± 0.02 0.03 0.11 ppm _ - ppm . Fe 573 ±: 216.00 210. 1,000.0 Al 194.5 ±224.00 16.0 640.0 Cu 44.3 ± 49.86 1.0 120.0 Zn 18.1 ± 7.93 4.0 31.5 Ba 5.25 ± 3.85 2.0 15.2 Mn 5.22 It 3.37 2.0 14.2 Sr 3.73 ± 1.49 1.0 7.2 B 3.05 ± 1.64 LO 6A Cr 2.S6± 0.30 2.2 3.0 So 2.4 ± 0.44 1.4 3.6 1 Scares, Miller, and Ambrose (1970) and Scares and Miller (1970). Chloride which is not included in Table 5 is also present in amounts about equal to Na. Others that are present in lower amounts include: Mg, Ca, Fe, Al, Zn, and Mn. The trace elements that are found at even lower levels include: Co, I, Ag, B, Ba, Cr, Li, Ni, Rb, Sr, Si, Se, and perhaps others not yet identified. Calcium and phosphor- us analyses were determined by the method of Kingsley and Robnett (1958) as adapted for the Technicon. All other elements were determined by emission spectographic analysis. These data show that the calcium and phos- phorus levels are expectedly very low compared to fish meal since the major portion of these two elements are found in the bone which remains with the meal. Most of the other minerals in fish solubles also occur in lower concentrations than in fish meal. Selenium, however, tends to be found in concentrations that are equal to or greater than fish meal. Three minerals — alumi- num, copper, and iron — were detected in quite variable concentrations. The average concen- tration of auminum in all solubles samples is somewhat lower than that reported by Kif er and Payne (1968) for fish meals from both the At- lantic and the Gulf coasts. The aluminum con- tent of solubles from Atlantic coast plants was, however, considerably lower than that of men- haden fish meals, whereas the aluminum content of Gulf coast solubles was somewhat higher. Similar diflferences in aluminum content (unpub- lished data) were found in Gulf and Atlantic menhaden fish protein concentrate (FPC) made at the College Park Laboratory, starting with raw fish and using stainless steel equipment under aseptic conditions. Consequently, the higher aluminum content appears to occur in the Gulf species itself rather than as a contaminant after they are caught. Selenium has become a focal point in trace mineral research since Schwarz and Foltz (1957) and Nesheim and Scott (1958) showed that it is essential for the prevention of dietary liver necrosis in rats and of encephalomalacia, exu- dative diathesis, and muscular dystrophy in chicks. Thompson and Scott (1968) have sug- gested that the range of 0.15 to 0.2 ppm is the desirable concentration for selenium in practical poultry diets. As little as 0.05 ppm selenium in the form of sodium selenite, however, has been found to prevent the physiological abnormalities mentioned above. So far, the Food and Drug Administration has not permitted chemical compounds contain- ing selenium to be added to the rations of live- stock. For this reason, natural sources are cur- rently the only available means of supplementing selenium in the diet. Fish meal has been shown to be a rich source of selenium by Kifer and Payne (1968) and Kifer et al. (1969). A number of samples of fish solubles was ob- tained from commercial menhaden plants throughout the 1969 fishing season and analyzed for selenium. A total of 38 samples were an- alyzed — 20 from fish meal plants along the At- lantic coast and 18 from plants along the Gulf coast. Although neutron activation analysis was used to determine the selenium content of the fish meals reported in the above-mentioned papers, we found that this method did not give reliable results with fish solubles. Consequently, a wet 259 FISHERY BULLETIN: VOL. 71, NO. 1 chemical method (Hoffman, Westerby, and Hidiroglou, 1968) was used. Table 5 shows the average results from all analyses. The data showed that there were higher levels of selenium in the Gulf product than in the Atlantic coast product (2.59 ppm vs. 2.22 ppm) (Soares and Miller, 1970), However, the corresponding fish meals from these two areas (Kifer et al., 1969) did not contain significantly different concentrations of selenium. This ob- servation indicated to us that selenium is present in greater concentrations in the water-soluble portion of fish rather than in the water-insol- uble solids,, and the higher levels of selenium found in Gulf menhaden fish solubles may reflect leaching from the relatively high selenium soils draining into the Mississippi River or other local environmental conditions. Further research, however, may reveal other explanations for these differences. Regardless of whether or not Gulf menhaden solubles or Atlantic menhaden solubles are used in feeds, a level of 2-2.5% will supply the min- imal level of 0.05 ppm selenium in the diet based on total chemical analysis. Availability determinations of selenium in fish meal and solubles are now in progress. To date there is some indication that this selenium source is not as available as sodium selenite (Miller and Soares, 1972; Scott, pers. Comm, to Morris and Levander, 1970). VITAMINS The fat-soluble vitamins are present in small quantities in condensed fish solubles. In con- trast, the contents of the various water-soluble vitamins are at least 50% greater in the con- densed solubles than in fish meal. Fish solubles are considered very rich sources for vitamin B12, choline, niacin, and one or more unknown factors. Table 6 lists data from six reports on the vita- min content of condensed fish solubles. The data are somewhat spotty and are often labeled "con- densed fish solubles" with little or no details on its origin. The data are particularly deficient in regards to the fat soluble vitamins, inositol, pyridoxine, and folic acid. However, there seems to be reasonable reliability in the values given for the remaining vitamins. In particular, each kilogram of solubles contains about 4,400 mg of choline, 270 mg of niacin, 38 mg of pantothenic acid, 11 mg of riboflavin, 0.5 /ug of biotin, and 440 fxg of vitamin B12. NUTRITIVE VALUE METABOLIZABLE ENERGY AND NUTRIENT DIGESTIBILITY Metabolizable energy (ME) is the energy de- rived from a feedstuff after subtracting from the gross energy value the amount of energy lost Table 6. — Vitamin content of condensed fish solubles. Source Vitamin _ A % i^\ ^^m j-m j-fc 1 2 3 4 5 6 Mveruyo Choline, ppm 4,028.0 — M _^ 5,300.0 2,960.0 _^ 4,429.0 Niacin, ppm 169.0 307.0 325.0 230.0 308.0 _« 268.0 Pantothenic acid, ppm 35.0 32.3 40.0 45.0 43.1 33.0 37.7 Folio acid, ppm __ __ __ __ 1.3 0.8 1.1 Inositol, ppm __ __ _^ _^ 352.0 __ 352.0 Riboflavin/ ppm 14.5 6.9 20.0 7.7 _^ 4.0 10.6 Pyriodoxine, ppm __ 8.3 ^^ __ _^ 34.0 21.2 Thiamine, ppm 5.5 2.9 4.0 __ „^ 5.0 4.4 Biotin, ppb 0.2 1.3 __ 0.3 0.1 -.« 0.5 Vitamin Bii, ppb _^ 640.0 ^_ 400.0 550.0 180.0 443.0 Vitamin A, lU/g _^ 360.5 __ __ __ 5.8 183.0 Vitamin D, lU/g _^ 55.0 __ __ _^ _^ 55.0 Vitamin E, lU/g — — — 6j0 — — 6.0 1. National Research Council (1968). 2. Hideo, Murayomo, and Yonose (1961). 3. Lassen, Bacon, and Dunn (1951). 4. Scott, Nesheim, and Young (1969:440). 5. Ewing (1963:291). 6. Murayama and Yanase (1960). 260 SCARES ET AL.: CONDENSED FISH SOLUBLES in the feces and urine. This parameter is uni- versally accepted among poultry nutritionists as a valid estimate of dietary energy content and is very important in formulating least cost poultry rations. Owing to the limited data available in. the lit- erature on the ME value of condensed fish sol- ubles, Cuppett and Soares (in press) have con- ducted investigations concerning this property. Data were gathered on 10 of the menhaden sol- ubles samples used in the analytical studies dis- cussed earlier. The experimental design with a few modifications for convenience was similar to that used by Potter and Matterson (1960) for various common feed ingredients. In these experiments, eight 3-week-old White Rock broiler cockerels were placed in each battery pen. Each diet was fed to two groups of chicks for 5 days and excreta samples were collected for 3 days. Chromic oxide was incorporated in the diet at the 1% level as a marker. Fish solubles were fed at 0, 5, 10, and 15% levels. Analyses for proximate composition and gross energy were made on all diets and excreta samples. The average ME value was 2.03 kcal/g which is somewhat higher than the calculated figures usually presented in feed analysis tables. These results are in good agreement with the data re- ported by Matterson et al. (1965) and Chu and Potter (1969) (Table 7). The ME value of all UNIDENTIFIED GROWTH FACTOR (S) The major contribution of condensed fish sol- ubles is the unknown factor necessary for max- imum growth of poultry and hatchability of eggs. Considerable evidence has been presented to demonstrate the existence of a stable unknown essential in condensed fish solubles. Combs, Arscott, and Jones (1954) indicated that the growth response obtained with arsanilic acid occurs only in the presence of the fish factors. The so-called fish factor is believed by Tamimine (1955) to be two components: one, organic in nature and the second, inorganic. Similar re- sults were obtained by Steinke, Bird, and Strong (1963). Combs et al. (1954) also reported a lack of chick growth response from fish solubles fed with penicillin. Similarly, Barnett and Bird (1956) found a lack of response with high levels of chloretetracycline. The conclusion is that the activity of the fish solubles occurs only in the presence of certain intestinal microorganisms. They also obtained no growth response when chicks were in new uncontaminated buildings. Actually, their bioassay method developed for evaluating the fish factor is dependent upon the ingestion of poultry excreta by the chicks on test. This would seem to confirm that beneficial bac- teria flourish in the intestines of poultry only in Table 7. — Metabolizable energy and protein and fat digestibility eval- uations of fish solubles in diets fed to young broilers and turkeys. Source Experimental Metabolizamie animal energy Fat digestibility Nitrogen digestibility Matterson ef al. (1965) Chicken Chu and Potter (1969) Turkey Cuppett and Soares (in press) Chicken Average kcal/g 2.094 2.223 2.030 ±0.23 2.12 97.7 88.7 ± 2.48 93.2 67.'2 56.2 62.0 3.70 three reports averages 2.12 kcal/g (962 kcal/lb) with fat and protein digestibilities of 93.2 and 62.0%, respectively. It appears that much of the problem with lower calculated values report- ed in the various feedstuflfs tables is due to underestimating the fat content of fish solubles. the presence of the fish factor. Similar evidence exists for the essentiality of the fish factor for the hatchability of chicken and turkey eggs. However, there are some contradictory reports. The reason for the differences in the observa- tions has not yet been elucidated. 261 FISHERY BULLETIN: VOL. 71, NO. I Table 8. — Unidentified growth factor response from various sources of condensed fish solubles. Reference Animal Fish solubles Experimental! period Types of diet Growth response % days of age % ol control Menge and Lillie (1960) Chick 2 8-36 Conventional 105.0 Mason et a\. (1961) Chick 2 0-14 Conventional 122.0* Mason et al. (1961) Chick 2 0-28 Conventional 107.0* Mason et al. (1961) Chick 2 0-14 Purified 107.0 Mason et al. (1961) Chick 4 0-28 Conventional 115.0* Harrison and Coates (1964) Chick 5 0-28 Conventional 113.0* Harrison and Coates (1964) Chick 5 0-28 All vegetables' 107.0* Chu (1968) Turkey 2-5 0-56 Conventional 106.0* Harrison and Coates (1969) Germ-free chick 5 0-28 Sterile con- ventional 104.0 Harrison and Coates (1969) Conventional chick 5 0-28 Conventional 112.0* Bhargova and Sunde (1969) Chick 2 0-7 Purified 109.0* Bhargava and Sunde (1969) Chick 2 0il4 Purified 118.0* 1 Only vegetable sources used. * Statistically significant {P<0.05). A compilation of recent reports (Table 8) in which the phenomena of unidentified growth factors in condensed solubles were studied brings out several interesting points. First, these data indicate that significant unidentified growth re- sponses are still being observed even when every effort is made to supply all known nutrients in the experimental diets. Secondly, growth re- sponse seems to be maximal when the solubles content ranges from 2 to 5 % of the diet and when these diets are fed to chicks for 14 to 21 days beginning with day-old birds. Growth responses tend to diminish as the birds become older. Fur- thermore, growth stimulations are consistently observed with battery-reared chicks fed conven- tional feeds or highly purified rations. There- fore, it is evident that stress, such as that found in field conditions, may not be necessary to ob- tain a significant growth response from feeding condensed fish solubles. An interesting series of experiments conducted by Harrison and Coates (1964) showed that sig- nificant growth responses were obtained from chicks whose parents were maintained on con- ventional rations containing animal protein or all-vegetable rations. In fact, the average growth response was greater when the conven- tional diets were fed. In comparing the effects of feeding 5% solubles to conventional or germ- free chicks, a significant response was observed only with the conventional birds. This indicates that the presence of microbes does influence the magnitude of the growth response. However, these workers observed that the addition of a small amount of sterilized or fresh droppings to the diet caused a significant growth depression which could be counteracted by feeding fish sol- ubles. Miller and Soares (1972) have conducted some detailed experiments designed to study unident- ified growth factors in fish solubles. They used virtually a complete chemically defined diet (Table 9) except for the sources of essential fat- ty acids, which were supplied as purified oils. The amino acid mixture was patterned after that of Dean and Scott (1965). The vitamins, min- erals, and antioxidant were supplied in sucrose- based premixes (Miller and Soares 1972). Table 9. — Basal composition for crystalline amino acid diet used to study unidentified growth factor responses in chicks.^ Ingredient Amount Sucrose Crystalline amino acid^ Antioxidant premix^ Celluilose Safflov^er oil Menhaden oil CaCOa CaHP04 Mineral mix' Vitamin mix'' Choline CI mix glkg 502.3 225.3 2.0 80.0 50.0 10.0 5.0 26.9 28.5 50.0 20.0 1,000.0 1 Miller and Soares (1972). a Dean and Scott (1965). * Santoquin premixed one part with nine ports sucrose. Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. 262 SCARES ET AL.: CONDENSED FISH SOLUBLES The diets in these experiments were balanced for energ^% calcium, phosphorus, and total ni- trogen. When fish solubles was added to the diets, the total nonessential amino acid content was balanced by adjusting the glutamic acid con- tent, and the essential amino acid content was balanced by decreasing the amount of the re- spective crystalline amino acid in the diet. All additions of condensed fish solubles produced greater growth rates in chicks as compared to the controls. One of the above experiments consisted of six treatments with four variables (Table 10) fed to day-old chicks (duplicate groups of 12) for 3 weeks. Diet 1 was our basic crystalline diet with no added protein whereas Diet 2 contained 5.0% fish solubles. A significant growth re- sponse was obtained by the addition of fish sol- ubles. Diet 3 plus 0.5 ppm selenium (as sodium selenite) did not give any growth response over the control. Diet 4 containing 0.5 ppm selenium plus 0.82% proline (which Graber, Allen, and Scott (1970) suggest is an essential amino acid) gave a slight but nonsignificant response in growth. Similarly, Diet 5 with 0.5 ppm seleni- um, 0.82% proline, and 5% gelatin protein re- sulted in only a small growth response. Gelatin was incorporated in the diets because earlier ex- periments had shown that the presence of intact protein was beneficial to the overall growth re- sponse to crystalline amino acid diets. However, when gelatin and fish solubles (Diet 6) were in- corporated in the diet, there again was a sig- nificant growth response that was similar to the response obtained with only fish solubles. In a further experiment by Miller and Soares (1972) Diets 5 and 6 as listed in Table 10 were irradiated with 4.5 Mrad of gamma radiation by the method described by Soares et al. (1971) and were fed to conventional and gnotobiotic White Rock chicks. The results (Table 11) of an initial experiment indicate that condensed fish solubles is capable of exerting a growth re- sponse whether or not bacteria are present. It should be noted that both groups of chicks were housed under conditions making it possible for the ingestion of excreta which most likely oc- curred because of the type feeders used. There- fore, these data may not be in conflict with those of Harrison and Coates (1969) who observed that germ-free chicks consuming sterile diets intentionally contaminated with sterile fecal matter exhibited growth responses when fish solubles was included in the diet whereas similar chicks fed diets with no fecal contamination showed no response to fish solubles. SUMMARY This review shows that condensed fish solubles (particularly those made from menhaden) is quite rich in many essential nutrients needed for livestock production. This is especially true if the composition is expressed on a dry-matter basis, since half of the product consists of water. Most essential minerals and water-soluble vita- mins are found in solubles in relatively high con- centrations when compared with other common feedstuff's. Fish solubles is highly digestible and rich in energy (measured as metabolizable en- ergy) and, if expressed on a dry-matter basis Table 10. — Growth response and efficiency of feed con- version of chicks fed purified diets supplemented with selenium, proline, gelatin, and fish solubles.^ Table 11. — Growth response of chicks for purified diets containing selenium, proline, gelatin, and fish solubles.' Item Diet Environment Menhaden solubles Weight gain Chicks I 2 3 4 5 6 Se, ppm Proline, % Gelatin, % Menhaden fish solubles. Weight goin^ Feed efficiency % 3233 0.62 5.0 *279 0.58 0.5 3236 3 0.64 0.5 0.82 247 0.64 0.5 0.82 5.0 3242 0.67 0.5 0.82 5.0 5.0 «286 0.54 Diet 5 Conventional Gnotobiotic Diet 6 Conventional Gnotobiotic % 5.0 5.0 i 269.6 269.5 398.5 » 107.9 Number \6 6 17 7 2 Average w/eight (g) of 18 day-old chicks. 3, * Means bearing different superscripts are statistically different (P<0.ai). 1 Miller and Soares (1972). -, 3 Means bearing different superscripts are statistically different (P<0.01). 263 FISHERY BULLETIN: VOL. 71, NO. 1 have a higher caloric density than fish meal. Cur- rently the most important nutrient in condensed fish solubles is the unidentified growth factor be- cause significant growth responses can be dem- onstrated with poultry and other species of live- stock under field conditions as well as in the lab- oratory. LITERATURE CITED Barnett, B. D., and H. R. Bird. 1956. Standardization of assay for unidentified growth factors. Poult. Sci. 35:705-710. Bhargava, K. K., and M. L. Sunde. 1969. A short time chick assay for unidentified growth factors. Poult. Sci. 48:694-697. Carpenter, K. J. 1960. The estimation of available lysine in animal protein foods. Biochem. J. 77:604-610. Chu, a. B. 1968. UGF(s), protein and fat digestibility, and metabolizable energy evaluations of fish solubles in diets of young turkeys. Ph.D. Thesis. Vir- ginia Polytech. Inst., Blacksburg. Chu, a. B., and L. M. Potter. 1969. Metabolizable energy, and protein and fat digestibility evaluations of fish solubles in diets of young turkeys. Poult. Sci. 48:1169-1174. Combs, G. F., G. H. Arscott, and H. L. Jones. 1954. UGF required by chicks and poults. 3. Chick studies involving practical-type rations. Poult. Sci. 33:71-79. CuppETT, S. L., AND J. H. Scares. In press. The metabolizable energy values and di- gestibilities of menhaden fish meal, fish solubles, and fish oil. Poult. Sci. Dean, W. F., and H. M. Scott. 1965. The development of an amino acid reference diet for the early growth of chicks. Poult. Sci. 44:803-808. Ewing, W. R. 1963. Poultry nutrition. 5th ed. Ray Ewing Co. Pasadena, Calif., 1475 p. Graber, G., N. K. Allen, and H. M. Scott. 1970. Proline essentiality and weight gain. Poult. Sci. 49:692-697. Harrison, G. F., and M. E. Coates. 1964. Studies of the growth-promoting activity for chicks of fish solubles. Br. J. Nutr. 18:461-466. 1969. The eff'ect of the gut flora on the growth re- sponse of the chick to fish solubles. Proc. Nutr. Soc. 28(2): 70 A. Hideo, H., S. Murayama, and M. Yanase. 1961. Fish solubles. I. Nutritive elements of com- mercial fish solubles. Takaiku Suisan Kenkyusho Kenkyu Hokoku 31:317-322. Hoffman, I., R. J. Westerby, and M. Hidiroglou. 1968. Precise fluorometric microdetermination of selenium in agricultural materials. J. Assoc. Anal. Chem. 51:1039-1042. Kifer, R. R., and W. L. Payne. 1968. Selenium content of fish meal. Feedstuffs 40(35) :32. Kifer, R. R., W. L. Payne, and M. E. Ambrose. 1969. Selenium content of fish meals II. Feedstuffs 41(51) :24. Kingsley, G. R., and 0. Robnett. 1958. Further studies on a new dye method for the direct photometric determination of calcium. Am. J. Clin. Pathol. 29:171-175. Laksesvela, B. 1954. An unidentified chick growth factor in her- ring meal. Meld. S. S. F. 5:94-102. Lassen, S., E. K. Bacon, and H. J. Dunn. 1951. An evaluation of condensed whale solubles as a supplement in poultry nutrition. Poult. Sci. 30:422-425. Lee, C. F. 1956. Preparation of a dry product from condensed menhaden solubles. U.S. Fish Wildl. Serv., Res. Rep. 45. March, B. 1962. Fish meal and condensed fish solubles in poultry and livestock feeding. In G. Borgstrom (editor). Fish as food. Vol. 2, p. 377-434. Aca- demic Press, N.Y. Mason, M. E., J. Sacks, and E. L. Stephenson. 1961. Isolation and nature of an unidentified growth factor (s) in condensed fish solubles. J. Nutr. 75: 253-264. Matterson, L. D., L. M. Potter, M. W. Stutz, and E. P. Singsen. 1965. The metabolizable energy of feed ingredients for chickens. Agric. Exp. Res. Stn. Univ. Conn., Res. Rep. 7. Menge, H., and R. J. Lillie. 1960. Ineffectiveness of antibiotic combination on response of chicks fed fish solubles. Poult. Sci. 39:1188-1190. Miller, David, and J. H. Soares, Jr. 1972. A study of the methodology for assaying un- identified growth factor in fish meal and fish sol- ubles. Poult. Sci. 51:1288-1291. Miller, D., J. H. Soares, Jr., P. Bauersfeld, Jr., and S. L. Cuppett. 1972. Comparative selenium retention by chicks fed sodium selenite, selenomethionine, fish meal and fish solubles. Poult. Sci. 51:1669-1673. Monson, W. J. 1969. Evidence that taurine may be one of the elusive unidentified factors. (Abstr.) Poult. Sci. 48:1846. Morris, V. C, and 0. A. Levander. 1970. Selenium content of foods. J. Nutr. 100 : 1383- 1388. 264 SOARES ET AL.: CONDENSED FISH SOLUBLES MURAYAMA, S., AND M. YaNASE. 1960. Nutritive elements of fish meal and solubles produced in various countries. Bull. Tokai Reg. Fish. Res. Lab. 27:55-59. National Research Council. 1968. Nutrient requirements of swine. Natl. Res. Counc. Publ. 1599. Natl. Acad. Sci., Wash., D.C. Nesheim, M. C, and M. L. Scott. 1958. Studies on the nutritive effects of selenium for chicks. J. Nutr. 65:601-618. Potter, L. M., and L. D. Matterson, 1960. The metabolizable energy of feed ingredients for chickens. Agric. Exp. Res. Stn. Univ. Conn., Prog. Rep. 39. Schwarz, K., and C. M. Foltz. 1957. Selenium as an integral part of factor 3 against dietary necrotic liver degeneration. J. Am. Chem. Soc. 79:3292. Scott, M. L., M. C. Nesheim, and R. J. Young. 1969. Nutrition of the chicken. M. L. Scott and Associates, Ithaca, N.Y. Smith, P., Jr., M. E. Ambrose, and G. N. Knobl, Jr. 1964. Improved rapid method for determining total lipids in fish meal. Commer. Fish. Rev. 26(7) :l-5. Soares, J. H., Jr., and D. Miller. 1970. Selenium content of Atlantic, Gulf menhaden fish solubles. Feedstuffs 42(37) :22. Soares, J. H., Jr., D. Miller, and M. E Ambrose. 1970. Chemical composition of Atlantic and Gulf menhaden fish solubles. Feedstuffs 42 (33) :65, 71. Soares, J. H., Jr., D. Miller, N. Fitz, and M. Sanders. 1971. Some factors affecting the biological avail- ability of amino acids in fish protein. Poult. Sci. 50:1134-1143. Spies, J. R., and D. C. Chambers. 1948. Chemical determination of tryptophan. Anal. Chem. 20:30-39. Steinke, F. H., H. R. Bird, and F. M. Strong. 1963. Unidentified chick growth factor in fish sol- ubles. J. Nutr. 80:60-68. Tamimie, H. S. 1955. The response of chicks to an unidentified growth factor in fish products. (Abstr.) Poult. Sci. 34:1224. Thompson, J. N., and M. L. Scott. 1968. Selenium in practical chicken feeds. Pro- ceedings of the Cornell Nutrition Conference, p. 121-125. 265 HELMINTHS OF SOCKEYE SALMON {ONCORHYNCHUS NERKA) FROM THE KVICHAK RIVER SYSTEM, BRISTOL BAY, ALASKA' David A. Pennell,' C. Dale Becker," and Nora R. Scofield* ABSTRACT A study of helminths infecting juvenile and adult sockeye salmon (Oncorhynchus nerka) leaving and entering the Kvichak River system, Bristol Bay, Alaska, was conducted in 1969. Ten helminths acquired in fresh water were found in smolts: Diplostomulum sp. ; an unidentified trematode; Diphyllobothrium spp. ; Triaenophorus crassus Forel, 1868: Proteocephalus sp. ; Eubothrium salvelini (Schrank, 1790); Neoechinorhynchus rutili (Mueller, 1789) ; Philonema oncorhynchi Kuitunen-Ekbaum, 1933; Rhabdochona sp.; and Contracaeciim sp. In addition to surviving larval stages of freshwater parasites, adults were infected by nine helminths acquired in the sea: Gyrodactyloides strelkowi Bykhovskaya and Polyanskaya, 1953; Lecithaster gibbosus (Rud., 1802) ; Brachyphallus crenatus (Rud., 1802) ; Tubulovesicula lindbergi (Layman, 1930) ; Phyllobothrium. caudatum (Zschokke and Heitz, 1914) ; Echinorhynchus gadi Mueller, 1776; Bolbosoma caenoforme Heitz, 1920; Anisakis sp. ; and Contracaecum sp. Infection incidences and intensities are tabulated where accurate data are available. Information on life histories is assembled from scattered sources, and some ecological aspects of helminths infecting Kvichak sockeye salmon are briefly discussed. The Kvichak River system is the largest pro- ducer of sockeye salmon, Oncorhynchus nerka (Walbaum), among five river systems in Bristol Bay, western Alaska. Its drainage basin covers nearly 8,000 square miles and includes two large lakes, Iliamna Lake (90 miles long and up to 26 miles wide) and Lake Clark (50 miles long and up to 4 miles wide). Intensive studies on sockeye salmon from this system and the other four systems, the Wood River, Naknek River, Egegik River, and Ugashik River, have been underway since 1946 (Thompson, 1962) . Recent research has been directed toward determining the biological basis for annual fluctuations in ^ This study was supported by funds provided by the Washington Sea Grant Program, which is maintained by the National Oceanic and Atmospheric Administra- tion of the U.S. Department of Commerce. Contribution No. 369, College of Fisheries, University of Washington. ' Fisheries Research Institute, College of Fisheries, University of Washington, Seattle, WA 98105; now de- ceased. ^ Ecosystems Department, Battelle Memorial Insti- tute, Pacific Northwest Laboratories, Richland, WA 99352. * Fisheries Research Institute, College of Fisheries, University of Washington, Seattle, WA 98105; present address: University of Idaho, Moscow, Idaho 83843. Manuscript accepted January 1972. FISHERY BULLETIN: VOL. 71, NO. 1, 1973. the number of seaward migrants and subsequent return of adults 2 and 3 years later, and pro- viding a reliable estimate of the optimum escape- ment for each of these river systems (Burgner et al., 1969). The survival rate for Kvichak River sockeye salmon is generally higher in the peak year of the abundance cycle than in other years since the total production from each year class comes in installments 4, 5, or 6 years later, depending on the total age of the adults when they return to spawn. The occurrence of mortality inversely related to density has been ascertained from ob- servations, but its causes remain unidentified. Since the available records show that, generally, smolt production has been proportionate to adult escapement, it appears that losses occur primarily after the smolts leave the nursery areas, but the possibility remains that the pre- disposing conditions are found in fresh water. The adjustments that smolts make to environ- mental changes, their physiological condition, and their resistance to stress have been studied by those who seek the factors that underlie var- iations in survival rate among year classes in 267 FISHERY BULLETIN: VOL. 71, NO. 1 the sea. Parasitic infections are considered a possible factor; albeit their effects may be largely sublethal, nevertheless they might con- tribute to disproportionate survival rates. A field study of the helminth fauna of sockeye salmon in the Kvichak system was commenced in 1968 and expanded in 1969. The immediate tasks were to (1) identify the parasites acquired in fresh water and those acquired in the sea, (2) determine the incidence and intensity of infec- tions, and (3) review the available literature for information on life cycles. Earlier studies by Margolis (1963) concentrated on parasites of sockeye salmon in the North Pacific Ocean as biological indicators of continental origin; he noted that over 50 parasitic species of fresh- water and marine origin are known to infect sockeye salmon occurring in the North Pacific Ocean and adjacent seas. Many of these para- sites occur in other species of salmon that come from Eurasia and North America to feed and mature in these waters (Akhmerov, 1963; Mamaev, et al., 1959; Mamaev and Oshmarin, 1963; Zhukov, 1960). MATERIALS AND METHODS The Kvichak River system consists of Iliamna Lake and Lake Clark, their tributaries, and the Kvichak River, which leads to the sea (Figure 1). Smolts migrate seaward for a few weeks following ice breakup in late May, whereas adults return to spawning grounds in July. Samples were taken from four groups of fish in 1969 (Figure 1): 1) smolts, captured by fyke net at Igiugig, on the Kvichak River (Site I), be- tween May 28 and June 6; 2) adults in Bristol Bay, collected by set nets at Pederson Point ( Site II) during May; 3) adults in fresh water, taken by beach seine in the Newhalen River (Site III) during mid-July; and 4) spawners, collected by beach seine on spawning grounds at Finger Beach (Site IV), Woody Island (Site V), and Porcupine Island (Site VI) during August. No attempt was made to discriminate between smolts originating from Iliamna Lake and Lake Clark. Adults collected at Pederson Point in Bristol Bay were known from past tagging ex- periments to consist almost entirely of Kvichak sockeye. The skin, gills, eyes, gastrointestinal tract, viscera, body cavity, and swim bladder of the fish were examined for helminths. Blood smears were prepared from each fish, gall bladder smears were made from adults, and brain tissue imprints were taken from smolts. Examinations were usually conducted on the day of capture. Thus, most helminths were recovered, tenta- tively identified, and counted while alive and be- fore preservation in hot Formaline-Acetic acid- Alcohol solution (standard solution). In some cases tissues were preserved in hot Bouin's solution (standard solution) for future dissec- tion and examination for parasites. Representa- tive specimens of cestodes and trematodes were stained in Delafield's hematoxylin and mounted for further study; nematodes were either stained ^ Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. SAMPLING SITES 1 LAKE CLARK^ 1 IGIUGIG II COMMERCIAL FISHERY III NEWHALEN RIVER 1 y\/^ IV FINGER BEACH |1 \^^..yC^ V WOODY ISLAND -N- ^"^^S/ VI PORCUPINE ISLAND J ILIAMNA Si ^ ^J'"'^ ■ O 4 - O ^^ .-^X^" n * . ^ \ .y-yy- ° y^ ^ y ..••\y^.-- y < 3 - >- .. >v O y'^ y'^ ^y^y y'^ et Z lU 2 y^ --■ y^y' y^ 1 y 20 40 60 80 ORGANIC MATTER (PERCENT ASH-FREE DRY WEIGHT) 100 Figure 1. — Relation between the caloric and organic content of estuarine organisms. Solid line is the linear regression line; dashed line represents the 95% confidence limits for the prediction equation; dotted line represents th3 95% confidence limits for the regression line. 291 FISHERY BULLETIN: VOL. 71, NO. 1 Table 1. — Summary of changes in ash and caloric content of decalcified and nontreated tissues of some estuarine molluscs. In each case, six organisms were combined and six caloric measurements made for each combined sample. HCl treatments were for 15 min. Organism Collection area Treatment Ash % kcal/g ash-free dry weight ±: Standard deviation Brachidontes recurvus Florida None 11.8 5.232 0.056 HCl 2.5 5.034 0.081 Chione canctllata North Carolina None 10.5 5.817 0.096 HCl 3.7 5.754 0.081 Merifnaria mercenaria North Carolina None 18.3 5.235 0.185 HCl 11.1 5.282 0.045 Modiolus demissus Virginia None 11.9 5.036* 0.010 HCl 3.6 4.783* 0.085 Modiolus demissus North Carolina None 15.7 5.118 0.046 HCl 6.3 4.998 0.087 Modiolus demissus Georgia None 5.4 5.102 0.051 HCl 4.8 5.053 0.095 Mytilus edulis Virginia None 13.5 5.006 0.046 HCl 3.3 5.068 0.045 Tagelus plebeius North Carolina None 16.5 5.471* 0.1O4 HCl 12.6 4.751* 0.166 'Significantly different at the 0X)5 level. Significant differences in energy content be- tween phyla were observed on the basis of ash- free dry weights and live weights. The phyla show an orderly phylogenetic progression of energy content only on the basis of live weight (Table 2). This is partially the result of de- creasing water content of more structurally ad- vanced phyla; we observed significant differen- ces among phylum means for water content which followed a similar trend. Not all phylum means were significantly different, however. The significant overall phylum effect can be attri- Table 2. — Analysis of variance of energy content, in kilocalorie per gram live weight, for all species examined, and listed in Appendix Table 1, and the mean kilocalorie per gram live weight for each taxon. Source of variation Degrees of freedom Mean square F ratio Among phyla V^ithin phyla Total 5 87 92 2.494 0.073 2.567 34.01** Phylum Mean kcal/g live weight Ctenophora Mollusca Echinodermata Annelida Arthropoda Chordota 0.049 0.373 0.393 0.850 1.027 1.156 "Significant at the 0.01 levd. buted to ctenophores' having a significantly low- er energy content than the annelids, arthropods, and chordates, but there are no significant dif- ferences among the foregoing three phyla nor among the ctenophores, echinoderms, and mol- luscs (Table 2). Also, within the phylum Mol- lusca and the class Crustacea the more advanced or more specialized groups (Cephalopoda and Decapoda) had significantly higher caloric con- tents per gram live weight than the less advanced groups. Although the values we obtained in this study may have resulted from a fortuitous selec- tion of organisms, we feel that the species an- alyzed are representative of their phyla and thaP the relation between caloric content, percent dry weight, and percent organic matter, and the phylogenetic position reflects an evolutionary trend for increased caloric content per gram live weight as a result of increased proportion of living tissue with increasing phylogenetic posi- tion. The differences in caloric content may have arisen from variation in the growth or reproduc- tive stage, or the energy content of the food source. The ctenophores and gastropods showed differences of as much as 1 kcal/g between sam- pling periods (Appendix Table 1, columns C and D). The variation in the gastropods may have 292 THAYER ET AL.: CALORIC MEASUREMENTS OF ORGANISMS been the result of increased lipid content prior to spring spawning. We feel that the large var- iation observed for the ctenophores was partially the result of high energy food in the gut contents. February and March are periods of high zoo- plankton abundance in the Newport River estu- ary, and ctenophores may exert heavy predation pressure on zooplankton (Hyman, 1940; Her- man, Mihursky, and McErlean, 1968). Conse- quently, high energy zooplankton (Comita and Schindler, 1963) in the gut may have caused the high caloric values obtained during this period. The Osteichthyes showed trends in their cal- oric content which were associated with life history stages and feeding habits. Adult fishes did not show the temporal variation in caloric values which was observed for many of the other taxonomic groups (Appendix Table 1). Post- larval and juvenile stages generally had higher caloric (per unit dry weight and per unit ash- free dry weight) and lower ash contents than their adult stage. The higher caloric content per unit dry weight is of course partly due to the lower ash content. The higher caloric content per unit ash-free dry weight is probably the re- sult of consumption of high energy food such as zooplankton (Comita and Schindler, 1963) by postlarval and juvenile fishes. Adult Brevoortia tyrannus, a planktonic feeder, tended to have have higher (but not significant) caloric values (mean 6.018 kcal/g ash-free dry weight) than adult carnivore-omnivore feeding types (mean 5.748 kcal/g ash-free dry weight) . The higher values obtained for menhaden are not surprising since up to 28.7% of the wet weight of these fish- es may be fats (Perkins and Dahlberg, 1971). On the basis of limited data Slobodkin and Richman (1961) argued that the frequency dis- tribution of energy (per gram ash-free dry weight) in a "haphazard collection of species" is skewed right, i.e., the modal frequency is at the lower end of the energy range. They explain this distribution by stating that ". , . there has always been selection for maximum number of reproducing progeny, but only sporadic selection for high energy content/gm." Even if there al- ways has been selection for maximum number of reproducing progeny, the analysis of energy con- tent relative to spawning time will afl^ect the re- sults. The argument also ignores the evolution of other life history strategies resulting in fewer offspring with higher survival rates (Mac Arthur and Wilson, 1967; Gadgil and Bossert, 1970) where there may be long-term storage of energy in individual organisms. Cummins and Wuycheck (1970) presented a frequency distribution similar to that of Slobodkin and Richman (1961) and stated that the distribution resulted from the predominance of plant values in their data. We suggest, with Paine (1964), that it may have been premature for Slobodkin and Richman to recognize a particular type of distribution. By combining the results of 15 species of invertebrates with Slobodkin and Rich- man's data on 17 species, Paine observed a more symmetrical distribution. If, as indicated by our samples, there is an evolutionary trend toward increased energy con- tent, the predominance of structurally more ad- vanced species in our sample would result in a frequency distribution skewed left as shown in Figure 2. The predominance of advanced spe- cies in our samples is consistent with available check lists for the Beaufort area (Duke Uni- versity Marine Laboratory, 1953; Turner and 40 « 4 4 8 5 2 S.e 6.0 6.4 6.8 7.2 7.6 8.0 8.4 KCAL/G ASH-FREE DIY WII6HT Figure 2. — Frequency distribution of energy in a system of shallow estuaries. The values are means for the 51 species presented in Appendix Table 1. 293 FISHERY BULLETIN: VOL. 71, NO. 1 Johnson).' The mean of the distribution is 5.57 kcal/g ash-free dry weight, the median is 5.74 and the mode is 5.80. We are continuing to obtain caloric data on these and other species throughout the year to determine whether the frequency distribution of energy in organisms from the Newport River estuary is continually skewed left (Figure 2) or because of seasonal variations follows some other pattern, such as the normal distribution suggested by Paine (1964). ACKNOWLEDGMENTS We wish to thank the numerous individuals at the Atlantic Estuarine Fisheries Center for their assistance in the collection of organisms and for their advice during the preparation of this manuscript. Dr. Kenneth W. Cummins and Dr. David G. Frey were helpful in the initial re- view of the manuscript. LITERATURE CITED American Public Health Association. 1965. Standard methods for the examination of water and, wastewater, including bottom sediments and sludges. 12th ed. Am. Public Health Assoc, N.Y., 769 p. Brawn, V. M., D. L. Peer, and R. J. Bentley. 1968. Caloric content of the standing crop of benthic and epibenthic invertebrates of St. Mar- garet's Bay, Nova Scotia. J. Fish. Res. Board Can. 25:1803-1811. COMITA, G. W., AND D. W. SCHINDLER. 1963. Calorific values of microcrustacea. Science (Wash., D.C.) 140:1394-1396. Cummins, K. W. 1967. Calorific equivalents for studies in ecological energetics. 2d ed. Univ. Pittsburgh, Pittsburgh, 52 p. Cummins, K. W., and J. C. Wuycheck. 1970. Caloric equivalents for investigations in eco- logical energetics. Int. Ver. Theor. Angew. Lim- nol. Verh. 18:1-158. Duke University Marine Laboratory. 1953. Check list of marine invertebrates at Beau- fort, N.C. 2d. ed. Duke Univ. Mar. Lab., 35 p. ^ Turner, W. R., and G. N. Johnson. Distribution and relative abundance of fishes in Newport River, North Carolina. Unpublished manuscript, 52 p. Atlantic Estu- arine Fisheries Center, National Marine Fisheries Ser- vice, NOAA, Beaufort, NC 28516. Gadgil, M., and W. H. Bossert. 1970. Life historical consequences of natural se- lection. Am. Nat. 104:1-24, Golley, F. B. 1961. Energy values of ecological materials. Ecol- ogy 42:581-584. Herman, S. S., J. A. Mihursky, and A. J. McErlean. 1968. Zooplankton and environmental character- istics of the Patuxent River Estuary 1963-1965. Chesapeake Sci. 9:67-82. Hyman, L. H. 1940. The Invertebrates, Protozoa through Cten- ophora. McGraw-Hill, N.Y. 726 p. Mac Arthur, R. H., and E. 0. Wilson. 1967. The theory of island biogeog^aphy. Prince- ton Univ. Press, Princeton, N.J., 203 p. OSTAPENYA, A. p., and A. I. SeRGEEV. 1963. [The caloric content of the dry substance of aquatic invertebrates used as food by fish.] [In Russian.] Vopr. Ikhtiol. 3:177-183. (Transl. 1967, Fish. Res. Board Can., Transl. Ser. 874, 18 p.) Paine, R. T. 1964. Ash and calorie determinations of sponge and opisthobranch tissues. Ecology 45:384-387. 1966. Endothermy in bomb calorimetry. Limnol. Oceanogr. 11:126-129. Parr Instrument Co. 1960. Oxygen bomb calorimetry and combustion methods. Parr Manual No. 130. Moline, 111., 56 p. Perkins, R. J., and M. D. Dahlberg 1971. Fat cycles and condition factors of Altamaha River shads. Ecology 52:359-362. Phillipson, J. 1964. A miniature bomb calorimeter for small bi- ological samples. Oikos 15:130-139. Richards, N. J., and S. W. Richards. 1965. Effect of decalcification procedures on the dry weights of benthic invertebrates. Limnol. Oceanogr. 10:469-471. Slobodkin, L. B., and S. Richman. 1961. Calories/gm. in species of animals. Nature (Lond.) 191:299. Thayer, G. W. 1969. Phytoplankton production and factors influ- encing production in the shallow estuaries near Beaufort, North Carolina. Ph.D. Thesis, North Carolina State Univ., Raleigh. 170 p. Williams, R. B. 1966. Annual phytoplanktonic production in a system of shallow temperate estuaries. In H. Barnes (editor). Some contemporary studies in marine science, p. 699-716. George Allen & Un- win, Lond. Williams, R. B., and M. B. Murdoch. 1966. Phytoplankton production and chlorophyll concentration in the Beaufort Channel, North Carolina. Limnol. Oceanogr. 11:73-82. 294 THAVER ET AL.: CALORIC MEASUREMENTS OF ORGANISMS Appendix Table 1. — Summary of data collected during caloric analysis of estuarine organisms. Classification Collection dote Combustions (Col. A) Ash (Col. B) Dry/ live wt (Col. C) (Col. D) Energy content (Col. E) Number % % kcal/g dry wt kcal/g ash- free dry wt kcal/g live wt Ctenophoro Class Tentaculata Mixed ctenophores 12-10-69 4 73.2 2.3 •0.652 ±0.419 2.428 ± 1.563 0.014 Mixed ctenophores 1-10-70 6 72.5 2.7 0.579 ±0.188 1.934 ±0.546 0.016 Mixed ctenophores 1-22-70 4 72.7 2.8 1.210±0.102 4.439 ± 0.373 0.082 Mixed ctenophores 2-5-70 4 55.5 3.9 2.000 ±0.103 4.493 ± 0.224 0.079 Mixed ctenophores 3-17-70 4 67A 68-3 2.4 2.8 2.178 ±0.344 6.725 ± 1.062 0.054 Mean 1.324 4.033 0.049 Mollusca Class Gastropoda^ Bittium varium 11-20-69 6 15.3 10.2 3.696 ±0.111 4.357 ±0.1 19 0.377 B. varium 12-17-69 5 16.4 9.1 4.414 ±0.199 5.279 ± 0.236 0.402 B. varium 2-11-69 4 23.5 12.2 4.087 ±0.129 5.334 ±0.177 0.499 Mitrdla lunata 11-20-69 5 18.3 14.8 3.274 ±0.105 4.005 ±0.127 0.485 M. lunata 1-14-70 5 7.7 16.1 4.139 ±0.156 4.483 ±0.169 0.666 M. lunata 2-11-70 4 10. 5k- 92 3 30- 37 h 28-57 li 50-51 35-36 18 10 19 10 19 10 19 10 18 10 18 10 16 10 US 23 U7 31 50-52 22 53-5U 22 53-55 37 J7-129 29-31 Ii3-U 22-23 U3 19 U.-li2 18 39-U. 16-17 U8 33 57 21 20 7 23 7 23 7 21 7 18 7 18 7 19 7 18 7 16 7 19 1 6 19 7 19 7 16 7 31 10 1*9-51 2I.-25 32 10 3U n 3I.-35 11 U.-li5 12 26 8 93-95 60-62 71 llU-U5 73-71. U7-U6 56-59 36-37 91 60-62 62 51.-55 13-15 22 16 28-30 31-32 16-18 21 2U 23-2U 23-2U IS 36 13 16 16 LU 11 11 12 11 U 13 12 12 11 25-26 22 23 23-2U 32-33 18 33 26-27 25-26 21-22 29-31 27-28 U 7-8 5 5 5 16-18 12 13 13 13 12 12 12 10-u 12 10 7-8 6-6 7-6 15 26 36-39 hi 33-35 1.1,4.7 27-28 29-30 3I.-36 2U 26-26 30-32 32-31. 27-30 29 21.-25 26 27 36-37 16 23-25 13 15-16 13-11. 23-21. 21-22 9 - 6 - 6- 7 9 U 8 9 u a 9 u 7 9 u 7 9 u 7 9 u 7 9 u 10 36-1.0 9 -n 10 6 UO u 10 16-19 53-55 19 -20 7 26-28 56-59 22 -23 7 82-89 U0-U2 50 35-37 U74.9 2U-25 27-28 30-32 21-22 23-25 31-32 32-3U 27-31 2U-25 38 8-10 32-35 u 33-36 8- 9 16-19 32-35 7- 8 17-16 - - 21-25 35-36 9-11 12 12 12 12 12 12 12 12 12 12 12 12 12 33-35 6- 9 8- 9 21-22 18 2 22 16 2 22.23 le 2 33-3U IS 2 16-17 16 2 26 25-26 5- 7 17-19 22-25 2-5 18-19 2U-25 U-5 lU-15 25-26 5 20-21 28-29 6- 7 20-21 28-30 6. 7 6-7 5 5 5 5 5 6 5 9 9-10 9 10 6 19-20 8 21 5- 7 6-10 6-10 6- 9 U-5 5-6 5 5 5-6 7 7- 6 Chasmod ea bosqulanua Chasmodes gaburrag &itoinflcroduB nlgricana HypleurochlluB ggninatuB HypBoblenr.ius hentzi Hjpsoblenr.lus lonthaa Op hi obi ennlua atlantlcue 52- 70 U 3U-35 10 2U-25 11 16-19 2 18-19 19 3 6 Ul- 60 u 33-3U 10 23-2U n 17-16 2 17-16 17-19 2- 3 6 2- 3 UO- 62 U 3U 11 23 13 lU-15 2 16 26-27 7 6 6- 7 33- U3 U 32-33 10 22-23 12 lU-15 2 16-17 21-23 U- 5 6 U- 5 66- 66 U 32-3U 10 22.2U 12 IS 2 16-17 23-2U 5- 6 6 85 1 33 10 23 12 IS 2 17 23 5 6 59- 65 5 33 10 23 12 20 2 21 27-29 7- 8 6 7- e 303 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FiUULT Size range SI Speci- mens exam- ined VERTEBRATE DORSAL FIN ANAL FIN CAUDAL FIN Genus, species toial Precaudal Caudal Spines Rays Spines Rays Total Dorsal secondary rays Dorsal prljmry rays Ventral prljnary rays Ventral secondary rays BOTHIDAE Ancylopaetta antlllarum Ancylopaetta cyclotdejT Ancylopaetta dilecta i Uicylopaetta kunperae Ancylopaetta mlcrocteaius j ncvlopBetta quadroceUa ta BotflUB 111 alus Bothus macullf erua Bothu3 ocellatua Chascanopgetta prorlgcra CltharictitJiys aiBblybregmatua Cltharlchthya aretlfrona CliJiarlchthya comutua Cltharlchthya dlnoceroB Clthsrichthys gynnorhlnus Citharichthyg macropa fcliharlchthya gpllopterus C yclopaetia chittenJeni Cyclops etta fimbri ai^ Rigyophrya senia 7tropu8 crossotua Etropus microatoinug Etropufl rlmoBUB OaEtropsetta frontalis Hlppoglosalna oblopga " Mlcrolene antlllarum Mlcrolene atrimana Mlcrolene megalepla Mlcrol en e aesBlllcauda ParalichthyB alblguHi Parallchthye deritatua Parallchtyys lethostlgma f'arallchthyB sguamile-ntuj Scophthalmus aquoaua gyaclujn gunterT Syaclum micrunijn Syaclum papllloguin Trlchopaetta carlbbaea Trichopaetta melaama T rlchopsetta orblauleua Trlchopaetta ventral la BRANCHIOSTEOIDAE Malacanthua plunlerl BREOMACEROTIDAE Bregmaceroa atlantlcua CAlXIilNYMIDAE Calllonynrua agaaalzl Draconetta ~ acanthopoma Draconetta oregona CAPROIDAE Antlgonla caproa Antlgonla combatia CARANQIDAE AlectlB crlnitua Caranx bartholomael Caranx crysoa Caranx hlppoa Caranx latus ^aranx ruber Chl oroacombrua ch ryauruB Icecap terij 3 macarellua D ecapteruB punctatuT " ^agatlg bipinnulata Hgtilcaranx amblinrhynchus Naucrates ductor Ollgoplltes sauniB Selar crujnenophthiJjnuB Selene vomer Serlola durierill Serlola rlyollan a Serlola zonata Trachinotua carollnuB TrachinotuB cayennenals 'rrachlnotuB falcatua TrachinotuB good el TrachuruB lathaml Vomer aetaplnnla CARAPIDAE CarapuB bermudenala CENTRISCIDAE Macrorhamphosufl acolopax CEKTROPOKIDAE Centropomua enalferug CentropoiTtufl unldeclJnalla CHAETODOHTIDAB Chaetodon aya Cbaetodon caplstratua uKaetodon ocellatuB Chaetodon sedentarlua Chaetodon strlatus Holacanthus bermuJenalB Holacanthus clllarlB HolacanthuB tricolor PomacanthuB arcuatiia Pomacanthus aureus Prognathodes aculeatus CHAULIODONTIDAE Chaullodua aloanel 111-200 ll 107-153 li 126-15(1 u 110-220 t> 9U-21B ll 69-195 ll 5IJ-232 7 132-217 3 70-1311 ll 133 1 105 1 101-110 u 68-77 h 80-98 U li7 1 100-lli7 ll 87-129 5 162-169 k 152-177 k 69-82 k 109-120 k 83-100 k 80-95 k 112-186 k 156-2U k UO-129 k 80-118 2 78-90 5 91-130 ll 105-221 k 137-173 k 95-213 k 161-195 3 111-120 ll 92-III1 k 97-1511 k 157-228 k eii-156 23 72-207 16 97-U7 2 75-UiJ 18 112-116 102-132 88 70-103 I1I1-87 85-115 71-91 113-liiO 191-212 99-IOI1 51-160 82-137 88-125 180-195 125-155 60-90 ia-5k 2U5 lll0-182 lli7-166 9I1-II12 Lli7-205 li7-235 l87-2li5 125-162 327 39-68 17-60 120-150 U7-160 126-165 68-115 122-280 56-168 56-90 U9-67 99-130 82-99 28-92 98-liiO 62-183 113-173 77-128 236-250 52 Ili5-185 36 10 3I1-35 10 36 10 36 10 35 10 37 11 liO 10 39-I1O 10 35-37 10 55 16 35 10 36-37 10 35-36 9-10 37 10 3U 10 3I1-35 10 3U-35 10 36-37 10 36-37 10 37-38 10 35-36 10 3U-35 10 3U-3S 10 37-38 10 U1-U2 u 1.5-U6 10-11 52-53 u Ii3-lii 10 I16-I17 10-11 37 10 U1-J12 11 37 10 38 10 3I1-36 11 3I1-35 10 3U-35 10 35-36 10 la-U3 10-11 U1-U3 10-11 Uo 10 liO-Ui 10 21 23 23 22 22 2ll 2U 25 2U 21i 2I1 2U 2I1 25 2ll 26 25 26 2U 21t 2U 2a 21l 2U 2U 2U 2b 2I1 2I1 2I1 2li 2U 2U 2U 2U 2U 2U 21i 2U 2I1 2U 2ll 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 11 10 10 10 ID 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 26 211-25 26 26 25 26 30 29-30 25-27 39 25 26-27 25-26 27 2U 21i-25 2U-25 26-27 26-27 27-28 25-26 2I1-25 211-25 27-28 30-31 35-36 Ul-li2 33-31l 36 27 30-31 27 28 23-25 21i-25 2U-25 25-26 31-33 31-33 30 30-31 3I1-36 Ui 15 15 12 12 lii 111 15 U lU 111 Hi 111 15 111 16 15 16 Ui 111 111 111 13 lU 111 lU Ui lii lii 102-106 IS 111 111 111 111 111 Hi lU lii Hi Hi Hi Hi Hi 62.69 65 72-75 76-79 62-71 n 68-73 91-98 91-98 79-85 115 81 76-8U 79-82 9I-9I1 7U 79-8U 75-80 85-87 82-8I1 7U-60 79-82 76-79 77-79 61-6U 76-82 105-108 117-121 91-93 93-IOI1 75-80 87-90 82-90 76-65 6I1-71 7I1-8I 85-87 85-91 95-102 96-IOI1 91-92 89-95 12-Hi 12 -U 12-13 13-Hi 12 Hi Hi 13-Hi 10 9 13 I1I1-J16 k 8 3 Vi 3 Hi 8 33-35 9 28-29 7-8 19 9 25-27 9 23 9 19-20 9 20-21 9 27-28 9 26-27 9 32-33*1 9 31-33*1 7 25-26*1 9 28-29 5 26 6-7 20 9 2li-26 9 21-22 8 28-32 8 29-31 9 35-37 7 211-25 6 26 7 19-20 7 19 9 29-31 9 21-22 9-10 10 17-19 18-20 18-20 21-23 20-21 18-19 20 17-19 28-29 30-31 19 5-6 51-5II 16 U9-51 18 SU-S8 18 60-63 18 51-55 18 5U-57 16 72-75 17 72-76 17 61-6I1 17 86 17 67 17 63-67 17 63-65 17 71-73 17 59 17 62-63 17 57-61 17 65-68 17 62-65 17 60-6U 17 63-66 17 56-61 17 57-63 17 U8-51 17-18 0-1 61-67 18-19 1 80-88 17 98-103 1'. 72-7I1 17 76-83 17 58-61 18 1 66-70 18 1 6I1-7I 18 1 60-6I1 18 1 50-5I1 17 60-63 17 67-70 17 67-72 17 75-81 17 80-85 17 70-73 17 69-75 17 51 39 12 U6-li7 31-32 - 7 15 3 13 16 3 13 16 3 30-32 19 ll 26-27 18-19 3-U 16 35-36 9-10 22-23 33-3I1 8-9 19-20 33-3I1 6-9 16-17 3I1 9 17 33-3U 8-9 21,-26 32-35 8-9 26-27 32-35 6-9 25-26*1 35-36 9 26-28*1 35 9 18*1 3I1-39 7-U 2I1-25 3I1-35 9 16 38 11 19-20 32.35 9-10 21-23 31-33 7-8 17-18 32-33 6-9 20-21 36-a 9-13 29-31 36-li2 U-13 20 3I1-I1O e-12 21-22 32-33 6 27 32 8 16-16 30-31 7 17-18 31-32 7-8 27-28 35-37 9-10 16-18 32-3li 8-9 18-19 27-28 6 36-38 10-11 6 37-I1O 11-12 lJi-15 2ll-25 I1-5 16-17 23 3 16-17 23 3 17-19 23 3 16-17 23-2I1 3-U 18-19 25 ll 20 25 Ii 16-19 25 ll 22-23 2li-25 ll 23-2U 23-25 3-I1 16 2I1 ll 0-1 7-9 7-9 9-10 9 10-U 8-9 10 6-8 7-8 7 10-U 8-12 9-11 7-8 7 7 7 9-10 7-6 8 9-10 8 9-11 3-U 3 3 3 3 U U u 3-U 3-U 3 304 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FUOLr Site range SL Speci- mens exam- Uled VEKTZBRA2 DORSAL FIN ANAL FIN CAUDAL FIN OmuB, species total Precaudal Caudal Spines Rays Spines Ray. Total Dorsal aecondar? rays Dorsal prlnary raya Ventral Ventral prixary secondary rays rays No. - )UbfT CKAUNACIDAE Chaunax plctus chehodipteridaf J^ogon aurol lpe atuB Apogon blno'ta'to's ' Cneilodlpterus afflnia Fpigonua occl^entallg Epigonua pandlonri Pna-'optyx conkllTii Phaeoptyy plf^tnentarla Ph a eoptyx pseudoraculatus Phaeoptyx quad ri squama tu3 Synagrops Delia Synagrops paeudomicrolgplB Synagrops aplnosa CHLOROPHTHALKIDAE Chlorophthalnma agassiai Faraaudia truciiler.ta amiDAE LabrlsoHus guppyl LabriaoBius nuchipinnlB Starkaia y-llneata CLUPFIDAE Alosa aeatlvalls Alosa medlocria Alosa sapldisaljna Brqvoortia funterl Bre^oortia patronua Prevoortla sTrdthl Bre^oortia iyrarinuB 61-123 Clupe 1 Kara ngU3 Dorosoma cepedianum 6oro3oiTia pg^g^gng* Etrumeui aadlna Harengula penaacolaa Jenkins la 1 amp ro taenia OpiathoneiTa cglimjjn S ardin ella ~ an c ho via CORYPHAENIDAE Coryphaena equlaella Coryphama hlppums CORYPKA?3iOIDIDAE CoelorhynchUB carmlnatus Hazumla balrSll llO 1 21. 10 Ik 7 9 2 8 29 6 9 8 6 30-60 3 21i 10 H. 6 8-9 2 7-8 31-32 7-8 9 6 7 39-51i 3 2U 10 IS 7 9 2 7-9 32-31. 7 9 6 5-7 103-136 h 25 10 IS 8 10 2 9 36-39 10-12 9 8 10-11 99-112 h 25 10 IS e 9-10 2 9 35-37 9-11 9 6 7-10 3I.-I16 3 21. 10 111 8 9 1-2 8 27-31 5-7 9 8 5-7 liii 1 2U 10 11. 7 9 2 8 30 7 9 8 6 78-87 U 2U 10 li. 6 9 2 8 30-32 7-6 9 6 6-7 U2 I 2U 10 11. 6 9 2 e 31 7 9 e 7 116-160 li 25 10 15 10 9 2 7 35-36 10 9 8 8-9 83-89 u 25 10 IS 10 10 2 9 la-1.2 12-U 9 s 12-U 80-115 u 25 10 15 10 6-9 2 7 35 9 9 8 9 107-127 7 1.6-1.8 18 28-30 10-12 8 1.3-1.6 13-lU 10 9 11-13 135-UiO k 39 17 22 10 6-9 UO-W 11-13 10 9 11-13 30-57 3 35 U 2U 19 10-11 2 16-19 29 6 7 6 6 90-110 U 31. U 23 18 12 2 18-19 26- lo 6-9 7 6 7-8 12-lU 2 31 10 21 19 7 2 U.-15 - - - - - 137-215 6 a7-51 Ii.-16 33-35 17-18 16-20 32-31. 7-8 10 9 6-7 118 1 51. 17 37 19 22 35 9 10 9 7 103-121 5 56-57 18-19 37-38 16-19 19-21 33-31. 7-8 10 9 7 190 1 Ui lU 30 20 21 36 9 10 9 6 2ll-170 5 1.5-1.6 16 29-30 20-21 21-23 31.-36 8-9 10 9 7-6 72-82 5 1.5-1.6 11.-15 30-31 19-20 21-21. 33-35 8 10 9 6-6 UO-233 3 U8 18-19 29-30 20-22 21-21. 32-35 7-9 10 9 6-7 190-2U7 U 55-57 23-25 32-33 17-19 17-16 37-la 10-13 10 9 6-9 137-215 h U8-50 11-13 35-39 13-15 32-36 35-37 9-11 10 9 7 37-62 7 1.3-ai. 11-13 30-32 11.-16 22-21. 31.-35 9 10 9 6-7 107-125 5 1.8-50 15-17 32-3U 18-21 11-12 31-32 6.7 10 9 6 61,-77 k 1.0-1.2 12-11. 27-29 16-18 17-18 33-35 8-9 10 9 6-7 27Ji7 8 38J.2 19-21 19-21 10-12 13-15 23-2U 3-1. 9 6 3 57-192 9 U5-I.9 12-13 32-36 20-22 20-21. 31.-35 9 10 9 6-7 23-29 6 U5-i7 16 29-31 16-19 16-17 31. 8 10 9 7 77-100 h 33 U. 19 51-51. 21.-27 UO-Ul U-12 9 6 12 91-130 u 31 13-11. 17-16 58-60 27-28 39-1.3 ll-U 9 6 10-U 170-205 2 75-78 12 63-66 . . . . . . . 350 1 - Ik - 2 10 UO - - - - - CYCL0PT15?IDAE Cyclopterua lumpua CTN0GU3SSIDAE Synyburua clrltataa Syrnphurufl diomedianua Syryhurul marglnat u s Syp^huTui minor Syp^hurus plger Sywphurus plaglusa Symphurus plag-ul Ba 5 yrp Hutu's pusilJ'uB Syirphurul a uroapilus CTPRTNODONTIDAE Cyprinodon varlegatua Florldlchthya earplo Fundulus grandla Fundulue heterocli'ma Fundulus luciae FunduluB iT-.aJalla Fundulus simllla 367 29 IB 108-132 U U7-I.9 9 38-UO 90-91 73-75 12 6 6 128-181 1. 1.6-1.9 9 39-1.0 90-92 71.-76 10-U 5 5-6 100-112 1. 52-53 9 U3-U. 95-96 61-61. 12 6 6 1.5-50 2 1.3 9 31. 75-76 60 10-U 5 5-6 91-106 U 1.7-1.8 9 36-39 8U-87 66-72 12 6 6 121-135 1. U64.8 9 37-39 86-89 70-72 10 5 5 U.7-191. U 50-1.2 9 la-U3 91.-96 78-63 U-12 6 5-6 91-116 3 53-51. 9 hk-llS 95-99 63-85 12 6 6 96 1 U. 9 35 65 69 U 5 6 31.-5U 1. 26-27 12 lU-lS 12-13 11-12 28-29 . . _ U8 1 23 9 11. U 9 31 - - - 65-111 1. 33-35 15-16 16-20 11-12 U 36-1.2 - - - 62-72 u 33-31. lU 19-20 U-12 10-11 364.0 - - - 17-32 10 31-33 12-13 18-20 8-9 10-11 33-31. - - - 30-65 1. 3U-35 15 19-20 13-11. 10-12 37-1,0 - - - u.-ei 5 35-36 15 20-21 12-13 11-12 364.0 - - - - DACrmiPTHlIDAI Dactylopterus voUtans DIODONTIDAF ChllonyctenJs echoepfl Diodon holacajithuj Dlodon hystrljt DIRFU'JDAF Dlretmua arggiteua ECHZNEIDAi; Fchenele naucrates Echenels neucratoldes Phthelrlchthys llneatus Rorora remora 60-115 97-125 37-166 62-95 16-20 29-30 lit 10-12 9-U 9 U 5 12-11, 12-15 9 k 5 15 16 9 h 5 25-26 210-268 3 30 11. 16 31.-37 32-35 374.0 10-U 9 6 10-12 103-137 2 30 11. 16 37 33 UO 13 9 8 10 50-61 2 36-39 18 20-21 35 35 39 U 9 6 U 7U-158 6 27 12 IS 2U-26 22-21. 394,3 U-13 9 8 U.13 ELR07RIDAE Dormltator maculatua FI/OFIDAF Flops saurus Hegalops atlantleus mMniCHTHYIDAE Biinellchthyops atlantleus 76-77 IS 32-33 1,0 8-9 60-233 1. 76-60 55-56 2U 26-26 16-19 36-37 9-U 10 9 7-6 75-115 3 55-56 33-3U 22 16 2U-25 32-33 7 10 9 6-7 305 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FAMIIY Size range SL Speci- mens exam- ined VERTEBRAE DORSAL FIN A M A L FIN CAUDAL FIN Genus, speclee Total Pre caudal Caudal Spines Rays Spines Rays Total Dorsal Dorsal secondary primary rays rays Ventral primary rays Ventral secondary rays WJGRAl'LIDAE Anchoa cubana Anchoa hepsetus Anchoa lyolppls Anchoa mitchllll Anchoa spinifer Anchoa tricolor Anchovlella 1 epld en tos tole Anchovlella perfagciata Cetengraulis edentulus Lycaigraulls grossldens EPHIPPIDAE Chaetodipterus faber KIOCOFTIDAE CypseluruB comatus C ypselurus cyanopterue Cypsclurus exgiliens Cyp eel urns furcatus Cypselurus hetenirus E^lepto^^;aI^phus velox Rxocoetus Qbtusirostris Hemiramphus balao HejTtirajnphus brasill ensis Hirundichthys afflnis Hirundichthys rondeleti Hyporhanyhug unifasciatus Oxyporhamphua micropterus Parexocoetus brachypterue Prognichthys gibbifrons FISTULARIDAE Fistularia petimba t'istuiarla labacaria GADIEAF Enchelyopus cimbrius Helanograjnimjs aeglefinus Herluccius albldu s Kerlucclus billnearls Phyc i 3 Chester! Urophycis chuss Urophycis cirratus Urophycis earlll Urophycis regius Urophyci's tenius GEKPYLIDAE Gempylus serpens Lppi^ocybiuji fjavobrunnemn Neoeplnnula orl entails Nesiarcfaus naautus Ruvett us pretiosua OniR^IDAE Diaptems ollsthostowus Diapterus rhomb eus Eucincstowus argenteus Eucinostomus gula Eucinostomus lefroyl Gerres clnereus ue-67 •) 65-75 U 51-82 3 57-67 u U7-133 li 95 1 63-70 k 28-35 U U7 1 Ili0-ii5 2 ue-103 UO-Ui 21-2U Ii2-lj3 21-22 ai-u3 21-22 Uo-U 19 h2-hl 17 U3 22 Uo 19 iii-Uk - Id 20 1)3 20 19-21 20-21 20-21 21-22 25-26 21 21 21 23 li-lS 16 lU-15 Ui-15 15-16 IS lU-lS li 16 15 22 32-3lj 7 10 20-22 33-36 7-9 10 21-22 32-35 7-8 10 26-27 35-36 9 10 38-39 37-39 10-U 10 21 30 6 10 22-21, 31-36 6-9 10 . . - 10 25 35 6 10 27 35 8 10 17-18 27-29 5-6 Ul-182 1) 1,2 26-27 15-16 12 7-9 26-26 5-6 U9-100 u U,4)5 30-31 U) 12-13 9-10 27-29 5-6 U5-166 2 1)1)4,5 30 11,-15 IS 10 27-28 5-6 58-105 3 1)5-1)6 29-31 11,-16 13-11) 10 29-33 6-9 197-228 1, 1,6-1,7 31 15-16 13-11) 9-10 26-29 5-6 190 1 72 1)6 26 22 23 23 1, 157-177 3 lik-hS 26-27 18 11) 11) 26-26 5-6 57-70 1, 51)-56 38-39 16-17 13-11) 11-13 21,-27 5-6 50-90 1) 52-53 35-37 16-16 13-15 12-13 21,-25 5 121-210 3 1,5-47 29 I6.ie 10-11 11-12 30 6 212 1 U5 29 16 12 12 26 6 137-173 h 51-52 33-31, 17-16 Il)-15 16 23-21) 1,-S 100-131) 1) 50 18-19 31-32 11) -15 15-16 23 U 97-115 1) 39 23 16 12 13-11) 23 U 30-73 U 1,3-1)5 27-29 lS-16 9-12 6-9 26-27 1,-5 21)0-420 3 82-63 16 IS 21)5-1)70 2 eu-85 - - 15 1I)-15 26 6 7 137-230 u 51-52 16 35-36 U54)8 374)3 30-31) 265 2 53-55 20-21 33-31) 15,22,22-21) 21,,26,22.23 57-58 . - 165-185 1) 51-52 25 26-27 12,13,36-37 37-36 394,2 - - 21)5-300 k 5U 27-28 26-27 12,13,394)1 384,1 31)-36 - - 198-265 u 1)7-1)9 15 32-31) 10,55-60 1)8-5U 26-31 - - 80-105 h U8 15 33 9-10,53-58 L5-51) 30 . - 317-337 3 51 16 35 10,62 -til) 5U-55 32 - - 282 1 hb 15 31 10,59 53 30 - - 82-177 1) 1)54)6 13-11) 31-33 8-9,1)7-51 1)5-50 30-32 - - ei)-U5 U 56-57 Ili-15 1)2 6,57-59 53-57 - - - 210 1 53 31, 19 32 12*8 3 . 29 b 9 373 1 32 17 15 9 19->5 3 12.1) 37 10 9 115-167 u 32 16 16 17 16 3 18 35-37 9-10 9 215-21)5 3 36 22 111 22-23 22-23 3 18-19 33-31) 7-6 9 270-310 3 32 16 16 13-15 18-19*2 2 17-18*2 31)-37 9-10 9 160-163 2 21) 10 11, 9 10 3 8 36 11 9 80-105 I) 21) 10 11) 9 10 2 9 37-36 10-U 9 123-11,5 t) 21, 10 m 9 10 3 7 37-38 10-11 9 107-U3 u 2L 10 11) 9 10 3 7 37-36 10-U 9 128 1 21, 10 Ik 9 10 2 6 36 10 9 160 1 21) 10 11) 9 10 3 7 33 8 9 6-8 7-8 6-6 7-8 6-9 5 6-6 5-6 6-7 7-6 7 8-9 6.6 U 6-7 l)-« U-5 9 7 U 1) U 7 10 9-10 9 8-10 10 10 10 10 9 OOBIESOCIBAZ Ooblesoy atrumosus 364)9 25-26 OOBIIDAE QobioneUus boleosoioa GobloneUus ahufeldti 5oEI5^ rSosci Gobiosoma ^inaburgi Gob io soma robustum Hicro^oblus gulosus QONOSTOHATIDAE ArgyripnuB atlantlcus Bon^artia pedal iota Gono stoma bathyphilum Gono stoma elongatu m haurolicus muellefl {"olymetme corythaeola Trlplophos hemingl GRAhMCOLEPIDAE Grawmicolepis brachlusculua Xenolepidichthys dalgleiahl ORArniSTIDAE Rypticus blstrisDlmis RypticuB maculatua Rypticus saponaceus HOLOC^KTRIDAE Comlger srinosus Holocpntrus ascengionla Holocentrus bullisl HolocentruB coruseus Holocentrus nifus Holocentrus vexillariua Hyrlpristis jacobus (!)stlchthys " trachyponiua I3TT0PH0RIDAE Tetropturus albldus KYPHOSIDAE Kyphosus incisor 35 1 26 10 16 6 U 12 32 9 6 7 8 51 1 26 10 16 6 12 13 30 8 8 7 7 29-50 1) 27 U 16 7 12-13 U 30-32 8-9 6 7 6-7 28 1 27 U 16 7 12 u 32 9 6 7 6 25-31 1) 27 11 16 7 U-12 10-12 31-32 8-9 6 7 7-9 314)5 1) 27 U 16 7-8 16-17 17-16 30 6 6 7 7 60-73 1, U64,7 16 30-31 10-U 10 9 51-70 2 37-38 17 20-21 19 29 - . 10 9 . 127 2 36-39 17 21-22 13 21) _ _ _ 135-215 1) Ul 17 21) U-ll) 29-31 38-U 10-12 10 9 9-10 U3 1) 32-33 12 20-21 . 122-165 k U5 18 27 U-12 26-31) _ _ 10 9 _ 172-183 2 59 19 1,0 10-U 57 - - - 76-90 1) 36-37 10 26-27 6 27-29 2 27-29 17 1 7 6 1 70-75 1) 37-36 10 -7-^6 5 29 2 28-29 17 1 7 8 1 61-70 1) 25 10 15 2 25-26 15-16 2l)-26 U-5 9 6 34, 82-152 1) 2U 10 U) 2 2U-25 15-16 25-26 l)-5 9 6 1) 50-175 I) 2U 10 11) 3 2U-25 17 21)-25 34) 9 8 1) 103-121, 1) 27 u 16 12-13 13-11) 1) u 26-29 5 10 9 1)-S 11,1-200 1) 27 u 16 U 15 h 10 30 6 10 9 5 71,-108 u 27 u 16 11 U-12 1) 7-6 30 6 10 9 5 105 1 27 u 16 11 U U 7 30 6 10 9 5 1,8-153 2 27 11 16 U 11)-15 1) 9-10 29-30 6 10 9 1,-5 61.65 h 27 u 16 U 13 1) 9 30-31 6 10 9 5-6 68-115 1) 26 u 15 U 13-lS U 12-13 27-28 5 10 9 34) 71)-105 h 26 u 15 12 12-13 u U 26 5 10 9 1) 35-36 8-U KJphQBUs sectatrlx 31)-50 h 26 10 16 10-U 13-11) 3 13 31) 33-53 U 26 10 16 11 12 3 11 33 306 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FAMILT Slie range SL Speci- mens exam- ined VERTEBRAE DORSAL FIN A H A L FIN CAUDAL Fin Ganio, Hpecles Total Precaudal Caudal Spines Rays Spines Rays Total Dorsal secondary rays Dorsal prLioary rays VcDtral prlaary rays Vmtral secondary rays nim Mo. - Nunber - UBRIDAE Bodlanus pulchelluB Clqjtlcug parral Decadon puellaria Kallchoeres bathyphllua Hallchoeres blvlttatua Halichoere5 macullpinna Hallchoer*>8 radlatug Hgrdpteronotus martirilcenBlB Hgnipteronotua novacula LachnolalJius majclmus Tau toga onitis l^al'assoma bifasclatun Thalaasoma nltiJuin L030TIDAE Lobotes surinanensls LOPHIIDAE Lophlus anerlcanufl LL'TJANIDAE Ap3iluE dentatuB Ft ells oc ulatu B Lutjanus analls Lut^anus apodua Lilt] anus aya Lutjanus buccangJla Lutjamis grlseus Lutjanus jocu Liitjanus mahogonl Lutjanua synagrls Lutjarua vlvanus Oc yurua chrysuma Pristipomoldes aqullonarla Prlftipomoldes freemanl Pristipomoldes iracraphthalrus Rhomb oplites aurombens Synphysanodon typus MALACXJSTEIDAE Malacosteus nlger KICFOD^SKIDAE Mlcrodesmus earri 187-215 3 36 1 115-1S2 1> 120-137 2 123-11»0 U 57-83 2 307 1 98 1 116-152 u 112-122 u 363 1 87 1 ia-55 5 19-1.0 k 66-98 2 202 1 95-160 3 165-197 u 32-99 u 9U-172 7 92-U7 k 112-150 h 81-170 li 86-150 It 100-115 k 95-lii3 3 157-173 k 86-178 h 135-155 li 55-173 7 123-UJ. 1. 111-123 li 26 28 26 25 25 25 25 25 25 30 35 25 25 25-26 31-31i 21i 2U 2U 2U 21. 21. 2U 21. 2U 21. 2U 2U 21. 21. 21. 21. 25 U7 67-66 U U 11 10 10 10 10 9 9 13 17 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 17 17 17 IS 15 IS IS 16 16 17 16 15 15 13 U-12 Hi li. Hi lU Hi lU Hi Hi Hi Hi Hi Hi Hi U Hi Hi 15 12 12 11 9 9 9 9 9 8-9 Hi 17 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 12 9 10 10 10 11 11 11 11 u 12-13 11 u 13 12-13 15-16 9 11 Hi Hi Hi 11. Ik H. 11-12 12 Hi U U 11 11 11 10 U-12 32-33 9-10 12 37 12 10 31i-35 10-U 12 26 6 12 26-27 6-7 11 26-27 6-7 12 26 6 12 26 6 12 21-23 U-5 10-11 26-28 6-7 8 28 7 U 25 6 10-11 2I.-26 5-6 11 23-25 3-5 9 9-10 - - - 6 1.2 13 9 6 39-ia 12-13 9 9 35-36 9-10 9 6 31-33 7-8 9 9 35-37 9-10 9 7-8 36-37 10 9 8 31-33 6 9 6 33-31. 6-9 9 6 31.-36 9-10 9 6 32-35 6-9 9 8 37-38 10-11 9 8-9 31i-35 9 9 7-8 39-U U-12 9 6 39-UO 12 9 8 lil-U3 11-13 9 8 36-38 10-11 9 7 Ii0-Ji2 12-13 9 9 U 10-11 6 6 6 6 6 6-7 6 5 5-6 12 10-n 8-9 7-8 8-10 9-XO 6-6 6 6-9 7-9 ID 6-9 10-12 lO-U 12-13 9-10 U-12 MDRIDAE BrosmlculuB imberbia Laemonana barbatulum Ph^'ai cuius fulvus KUGILIDAE AgonostomuB monticola Mug LI CephaluB Hugll curana Kugil incilla Wugil trichodon MULLIDAE tt ull o ldichthys martinlcua Hullus ' aiiratus Pseudupeneus maculatua l'penei-s parvus KTCTOPHIDAE Centrobranchua nlgroocellatua Diaphus ajiteorb Italia Diaphus dumerlli Diaphus laryrossa Diaphus raf inesquei Diaphus termor hilus Oonjjchthys coccol Hy^ophum roacrochlr Lepidophajies guentheri Lepldophanes supralateralls h^tophum afClne Kyctophmn agperum Hyctophom nltidulum Myctophuw obtuslro stria NotoacQpelua el ongatua Synibolopborua rufinum tr^OSCOPELIDAE Neoacopelue macrolepldotus OGCOCEPHALIDAE Dibranchua atlantlcua Halieuthichthy s aculeatua Ogcocephalus nasutu s Ogcocephalus parvua OgcocephaJ.u3 veapertillo OPHIDIIDAE Lepophldium brevibarbe Lepophliium cervlnuai Lepophldlum Jeannae Lepophidium kalllon Lepophidiun marworatum Lepophidium pro^undorun Qphidion tf-ani Op hi 'ii on" grayl Q phidio n holbrookl Qphidion velshl Otophldluin omostigiman Rissola marginata 135-165 1. 90-165 2 U5-122 li 28-30 3 105-155 1. 61i-105 I. 35 2 58-105 II 90-131i 1. 97-126 3 113-155 U 100-131. 1. 2U-27 1. U7-162 U 51-80 1. 68-97 5 68-81 5 1.8-51 1. 20-ii7 li 33 1 lii. I 100 1 62-76 1 26-66 li 55-66 li 1.2-61 2 25 1 51-52 2 U7-136 75-11.2 1 65-87 L UO-127 li 55-95 h 100-130 li 165-260 U 190-2UO li 210-275 I. 155-166 2 iSo-180 2 200-220 1. 210-230 U 190-265 1. 125-260 li 175-210 k 60-122 3 11.5-180 u 50-51 59 UB-li9 25 2li 2U 2U 2U 21i 21. 21. 21. 36 35 35-36 31. 31. 35 39-1.0 35 37 31. 35-37 37 38 35 38 37 17 17 H. 12 12 12 12 12 10 10 10 10 16 16 16 16 16 16 15 16 16 16 15 15 15 IS 17 IS 16 5 17 5 19 - 16 - 17-18 - 72 IS 72-71. IS 71.-75 H. n-72 U. 71-72 U-lS 72 Hi 66-67 16 61i-*S 16 66-67 16 66-67 16 57-59 Hi 68-69 15 33-31. 1.2 3U-35 13 12 12 12 12 H. H. Hi H. 20 19 19-20 18 IB 19 21.-25 19 21 16 20-22 22 23 20 21 22 17-18 U 12 57 57-59 59-61 52-58 57 57 50-51 1.6-1.9 50-51 50-51 U3-U5 53-51. 10,53-56 5-6,59 io-u,5o-S6 9-U 15 13-H. IS 12-13 13-Hi U 12 H. 12 13 12-13 13 13 21 Hi-15 127-129 132-13U 135-139 130-135 128 126-132 115-125 133-HJ. 120-136 136-1U6 102-105 Hi7-156 56-62 67 60-61. 9 9 8-9 16-16 15 U-15 H. H. Ik-lS 21-23 20 11. Ik 16-20 17-16 19-20 16 20-21 106-lU 112 -U7 Uk-116 108-115 106 106-106 96-100 96-105 98-111 llk-121 82-85 U8-12k 3k-37 26 23-2U 32-31. 26-30 26-29 2>-30 28-29 31 33 3k-35 31 30-31 31-32 30-31 31-32 31-32 32-33 31-32 3k 3k 32 3k-36 35-37 35-37 3k-35 kk 36-37 9-10 7-8 7-8 9 8 10 8 8 8 6 10 6-7 10 6 10 6-7 K) 0-7 10 7 10 6 10 8 10 8 10 7 10 8-9 10 6-9 10 6-9 10 8 10 12 10 9 10 7 9-10 7 7-6 7 7-8 7 7-6 7 7-8 7 8 7 9 7 9-10 7 8 9 5-6 9 6 9 5-6 9 6 9 6-7 9 6-7 9 6-7 9 7 9 7 9 6 9 7-8 9 6-9 9 6-9 9 7-6 9 13 9 6-9 307 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FUOLT species Size range SL Speci- mens exam- ined VEHTEBEAE D R S A I F I N ANAL FIN CAUDAL FIN Qenus, total Precaudal CauiJal Spines Rays Spines Rays Total Dorsal secondary rays Dorsal primary rays Ventral primary rays Ventral secondary rays mm No. Number - SCIAFNIDAJ: Bairdiella chrysura Bairdiella ronchus Cynosclon arenarius Cynoscion .lamalcecsis Cynoscion leiarchus Cynoscion nebulosue Cynoscion no thus Cynoscion regalia Cynoscion virescens Equetus acujninatus Equetus lanceolatus Equetus punctatus Equetus ujrj^rosus~ Isoplsthus parvlpinnig Larlmus breviceps Larijiiug fasciatus Leiostomus xanthurus Macrodon ancylodon HenticirrHus ajnerfcanus Henticirrhus llttoralls Menticirrhua martinicensis Henticirrhus sajcatills HicropoRon fumierl MicropoKon undulatus Nebris ma crops Odontoscion dentcx OphiosclQn~costaricen3ls Paralonchurus brasiligiaia Paralonchurus peterai PoRonias cromis Sclaenops ocellata Stein fer lanceolatus Stel n f er rastrlfer Steimer stellifer Umbrina fn-aclllcirrhua SCOKBISESOCIDAE Scomber esox saurus SCOMBRIDAE Auxls thaaard Euthynnus alletteratua Euthynnu? pelamis Sarda sarda ScomFpr japonicus Scomber 3CoTTibrus~ 5comberomoru9 cavalla Scomberomorus maculatus Thunnus albacares Tbunnus atlanticus Tbunnus thynnus SCORPABIIDAE Helicolemis dactylopterus Neomerinthe beanonun Weomerinthe pollux Pontlnus castor Pontinus lonf;l3pinis Pontinus macrolepls Pontlnus rathbuni Scorpaena a^asslzi Scorpaena bergi Scorpaena brasilienslB Scorpaena calcarata Scorpaena dlspar Scorpaena Inermis Scorpaena isthmensis Scorpaena petrlcola Scorpaena plumerT " Setarches guentheri Trachyscorpia cristulata Ut>-lii2 135-153 150-177 iao-215 112-178 31-220 121-130 26-165 90-175 90-I0I1 100-122 77-197 m8-l65 112-173 152-173 27-130 27-182 221-230 171-195 72-lli3 210-233 75-213 lUO-235 101-200 Uj 6-177 Uj5-180 118 112-153 105-155 28-163 20-26 38-108 137-168 1*7 e7-i7U 51-58 12 2 3 U 3 8 U 10 5 6 h 5 li Ii li 10 IS 3 h k 2 li 5 5 U li 1 k h 6 5 7 u 1 li 52-115 5 38 1 370 1 330 1 170-180 U 66 1 120-150 5 155-220 h 580 1 500 1 75 1 186-210 li 75-91 li 222 1 63-69 3 108-135 u 51i-76 u 83-105 3 86-127 k 71-133 1 92-187 h 107-135 li 96-126 U 73-158 3 91-122 u 2U 1 165-197 2 85-110 u lli7-20li 3 25 25 25 25 25 25 27 25 25 25 25 25 25 25 25 25 25 26 25 25 25 25 25 25 25 25 25 29 25 2U 25 25 25 2U 25 66.67 39 39 lil 50 30-31 31 Ii2-li3 52-53 39 39 39 25 2U 2U 2U 2U 2U 2U 2U 2U 2U 2U 2U 21i 2L-25 2U 21i 2U 25 11 11 13 13 13 13 Hi 13 Hi 10 10 10 10 10 10 10 10 lii 10 10 10 10 10 10 u 12 10 11 10 10 ID 10 10 9 ID 20 19 22 25 Hi U 17 21 18 19 18 10 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 10 9 H. 12 lii 11 12 U 12 u 12 11 12 11-12 13 U 12 u U n 15 10-n 15 13-lli 15 13 15 9-U 15 7 15 U 15 11-12 IS 11-12 12 10-11 1? U 15 11 15 11 15 11 15 11 15 11 lii 9 13 12 15 n IB 11 15 u Hi 11 15 11 15 12-13 15 12-13 15 12 15 n 26-27 19-22 23-21i 25-27 21i-26 21-23 2U-27 28-30 21i-28 27-30 36-liO li6-50 liU-U9 36-39 20 2li-26 25-27 29-32 29-30 21i-26 21i-25 23 23-25 26-26 28-29 30-31i 21-23 21 2^-30 31-33 21-23 23-25 21-21i 21-23 20 22-23 10-11*5 19 11-12 11-12»8 20 17 12*7 19 16 Hi*9 25 22 lii*9 16-17 10-U 12.1i-5 17 11 ll-S 25-26 16 lli-16*8-U 31-32 18-19 18-19*8-9 21 13 12*- 20 Hi Hi*8 21 17 12*8 15 12 12 15 12 9 15 12 10 15 12 10-11 15 12 9 15 12 9 IS 12 9 15 12 9 15 12 9 IS 12 8-10 15 12 9 15 12 9 15 12 8-9 15-16 12 9 15 12 9 15 12 9 Hi 12 9 16 12 9 2 8-10 31-31i 8-9 9 2 8 31i 9 9 2 10-12 28-33 6-8 9 2 9 30-32 7-8 9 2 U 31-32 7-6 9 2 10-U 29-33 6-9 9 2 9 30-32 7-« 9 2 10-12 29-33 7-S 9 2 7-8 26-31 6-7 9 2 6-8 30-32 7-6 9 2 6 27-28 6-7 9 2 7-6 29-31 7 9 2 7 31-32 7-6 9 2 19-20 30-35 7-9 9 2 6-7 29-30 6-7 9 2 6 26-31 6-7 9 2 12-13 29-33 6-6 9 2 9-10 30 6-7 9 2 7-6 32-33 6-9 9 2 7 30-31 7-6 9 2 7 31-32 8 9 2 7-8 29-31 6-6 9 2 7-8 31i-36 9-10 9 2 6 33-3U 6-9 9 2 10 31-33 8 9 2 9 35-37 9-11 9 2 9 35 9 9 2 8 26-30 6 9 2 7 27-28 5 9 2 6 32-33 8-9 9 2 7-8 32-36 8-10 9 2 7-9 30-35 7-9 9 2 8-9 33-36 ?-10 9 2 8 31i 9 9 2 7-6 30-33 7-8 9 13*6 21-25 3-I1 7 lli*7-e la 13 9 . 13*7 - - 9 2 13*7 - - 9 1 U*7 - - 9 2 U*li-5 35-36 8-11 9 2 11*5 - . 9 li 13-15*8-9 U-liU 13-Hi 9 li lU-16*7-8 la 12 9 - 12*- - - 9 1 12*7 - . 9 2 12*7 US 15 9 3 5 311-36 10-12 7 3 5 26-29 7-6 7 1 5 27 7 7 3 5 25-27 6-7 7 3 5 27-26 7 7 3 5 27-29 7-8 7 3 5 27 7 7 3 5 25-26 6 7 3 5 27-29 7-8 7 3 5 26-28 6-7 7 1 5 25-27 6-7 7 3 5 26-27 6-7 7 3 5 26 6-7 7 3 5 27-29 7-8 7 3 5 26 7 7 3 5-6 21i-25 5-6 7 3 5 29 e 7 3 5 30 6 7 5-8 6-7 7 5-7 6-6 5-7 5-7 6-7 U-5 5-7 7 6-9 6 U-7 6-8 6-7 7 6 6-7 6 7-9 6 6-6 9 9 5-7 5-6 7 7-9 6-9 7-9 3-6 13 12 9-10 5-7 6 5-6 6-7 6-7 6 5-6 6-7 6-1 5-* 6 5-6 6-7 5 5 7 308 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. fakhi sppcies Size range SL Speci- mens exam- ined v?:rtebrae DORSAL FIN ANAL K r N C A I D A L f • K Genus, Total Precaudal Caudal Spines Rays Snlnes Hays Total Dorsal secondary rays iJorsal primary rays Ventral prinary rays Ventral secondary rays mm No. - Nupter - . OPISTHOGNATHIDAF. Lonc h oplsthus hlgmanl 6pi sTtio gna thus macrognathus bplsthognathufl maxlllosus OSTRACITDAE Lactophrya polygonla Lactophrys yiadricornlB P3IPHain)AE Penyhrls schomburgkl PERCOPHTDIDAE Ben^rops ana tiro atria Bembrops gobialciaa bacibropa macrQiraoa Chriomystajc squamentum PLEUROKECTIDAE Olyptocephalus cynogloaBUS Poecllopaetta albomarglnata Poecllopsetta beanl POBCILirDAE G ambus la afflnla Heterandrla fonnoaa Poecilla latlpinna POLWTXIDAE Polymlxla lowe l Polymixl'a nob Ills POLYireMIBAE Polydactyly 8 octonemua Polydactylua vlrlglnleus POHACENTRIDAE Abudefduf analoe ua Abudefduf saxatllle Chrnmls enchrysusus Chromls ln3olatu3 Pomacentrus fuscue PomacentruB leucost Ictus Pomacentrus planlfrons POMADASYTDAF Anlaotrenus surlnamensls Aniaotremus virginlcua Conodon nobllis Oenyatremus lu teu a Haamilon album H aenrnlon aurollneatum haemulon chrysargyrcum Hacmulon flavollneatum H aemulon melanurum Haenu'lon plumlerl Haemulon stelndachnerl Haemulon striatu m Orthoprlstla ch a lce u s Orthoprlstls ehr'yBo'p'^erua Orthoprlstla ruber PonadasyB corvlnaefomda POMATOMIDAE Pomatomua saltatrlJi PRUCAIITfaDAE Cookeolu s boops ^lacaniTniB arena tu a ^•lacanihuB cruent'a'i^ua PrlstlgenyB alta RACHTCENraiDAE Rachycentron canaduai SCAP.IDAE ptotomua roacfUB holBlna uata ScaruB crolcensla Sparlsoma chrysopterum radians Crypt< Klcho: 5U-81j h 26 10 18 U 18 3 16-17 23 3-li 6 6 3-li 116-131 2 30 10 20 n 17 3 17 23 li 6 6 3 75-105 3 28 10 16 11 IS 3 lu-15 21,-25 5 8 6 3-1, 26-120 1, 13 e 5 10 5 5 1,2-11,6 5 li, 8 6 10 - - 5 5 130-216 1, 28 9 19 6 IL-IS 17-19 31,-36 10 -U 6 7 9-10 163-192 1, 30 9 21 6 16-17 17-18 31,-36 10 -u 6 7 9-10 11,5-160 1, 28 9 16 6 lU 17-18 36-37 11 -12 6 7 10 72-7a 2 28 9 19 6 15 16 UO-lil 13 6 7 12-13 220-230 2 57-58 U-12 1,5-1,7 108-117 91-101 22-23 D-1 U 11 101,-107 II 1,0-1,1 10 30-31 61-61, 53-51, 20 1 9 6 2 90-101, 1, 1,1-1,2 10 31-32 63-66 53-56 20 1 9 6 2 23-26 u 32-33 11, 18-19 7-8 11 23-26 16-20 u 31 U, 17 7-8 9 21, - - - _ 20-21, h 29-30 13-11, 15-17 I3-II1 6 30-32 - - - - 108-127 ll 26 12 16 29-30 h 15 29 6 9 9 5 130-195 3 28 12 16 35-36 k 16-17 29 6 9 9 5 63-66 1, 21l 10 111 11-12 3 13 ui-a3 12 -13 9 6 12-U 22-21, 3 2U 10 11, 8-9 10-12 3 12-13 - 9 6 US 1 26 11 15 13 12 2 10 26 6 8 7 5 55-126 3 26 11 IS 13 13 2 12 26.27 6 6 7 5-6 75-82 u 26 11 IS 13 12 2 12 25 6 7 5 55-60 u 26 11 IS 13 12 2 U 2U-25 6 7 I1-5 55-62 h 26 11 IS 12 15-16 2 13-li 25 8 7 5 1,0-51 1, 26 u 15 12-13 15 2 13 25 8 7 S 38 1 26 11 IS 12 IS 2 13 25 8 7 5 119-165 2 26 10 16 11-1? 17-18 3 9 Ii2-U3 13 9 6 12-13 86-178 3 26-27 10 16-17 12 16-17 3 10-11 38-39 10 -11 9 8 10-11 135-155 1, 26 10 16 12 13 3 7 39-1,0 11 -12 9 8 11 11,2-190 k 26 10 16 13 11-13 3 11 35-37 10 9 8 6-10 161-238 2 26 10 16 12 16-17 3 7-8 39-1)0 12 -13 9 6 9-11 93-133 u 26 10 16 13-li) 11,-15 3 9 3S.J,1 11 -12 9 8 10-11 128 1 26 10 16 12 11, 3 9 Ul 13 9 6 11 95-135 3 26 10 16 12 11, 3 8 33-36 9 -10 9 6 6-9 97-115 U 26 10 16 12 15-16 3 8 37-1,0 10 -12 9 6 10-11 77-227 1, 26 10 16 12 15-16 3 9 37-39 9 -12 9 6 W-U 137-160 u 26 10 16 12 16 3 9 39-U. 11 -12 9 6 11-12 88-102 II 26-27 10 16-17 13 13-15 3 8-9 39-U2 12 -13 9 8 10-12 117-163 1, 26 10 16 12-13 15 3 10-11 Ul 12 -13 9 8 U-12 95-103 1, 26 10 16 12-13 15-16 3 13 UO-1.2 12 -13 9 8 11-12 1S2-185 1, 26 10 16 12 15 3 10 38-1,0 12 9 6 9-11 125-170 1, 26 10 16 12 15-16 3 7 3B-39 11 -12 9 6 10-12 11,2-210 U55 21,-25 35-36 159 1 23 10 13 10 12 3 12 25 5 137-188 ll 23 10 13 10 U 3 1J.-15 26-28 5-6 135 1 23 10 13 10 13 3 lU 2U 1, 89-135 1, 23 10 13 10 U 3 10 2U 1, 21, 1, 5-6 1, U Sparlsoi Sparlsoma rubriprinne Sparlsoma vlrlde 57-67 3 25 10 15 9 10 9 25-27 7 6 5-6 117-11,8 3 25 10 15 9 10 9 26-26 7-6 6 6-7 69-210 3 25 10 15 9 10 9 25-27 6-7 6 6-7 67-95 3 25 10 15 9 10 9 28 6 6 7 81-121 U 25 10 15 9 10 9 25-27 6-7 6 6-7 7U-92 2 25 10 15 9 10 6-9 26-26 7-6 6 6-7 192 1 25 10 15 9 10 9 28 6 6 7 309 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FAMILY Size range SL Speci- mens exam- ined VERTEBRAE DURSAL FIN ANAL FIN C A I D A L FIN Genus, species Total Precaudal Caudal Spines Rays spines Rays Total Dorsal secondary rays Dorsal prunary rays Ventral primary rays Ventral secondary rays STmANTDAF Anthiapicus Icptus Centroprlstis ocyurus Ce ntroprlptis rihiladelphica Centropristis gtriata Cephalopholls fulva Choi'i;tiFtiiir. eukrines Dermatolepis inermis Diplectrum bivittatuin Diplectrum formosum Ddplcctrujn radiale Fpinephplur d nmnncndhayi Epinephelus flavolimViatus Fpinophelus guttatus Fpinephelus irystacinus Epinephelus ni^ritls Epinephelus niveatus Fpinephelus striatus Hemianthias 'ivanu s Hycteroperra bonaci Mycteroperca falcata Wycteroperca interstltialis Mycteroperca microlepis Mycteroperca pbenax Mycteroperca venenosa Ocyanthias martinlcensis Paranthias fu rc ifer Petrometoi''on cn-entatum Fikea cubensis Pronofot-'ranaTais aureombens Schiiltzea IfTa Serraniculus pumil io S err anus SerraniJi annularis atrobranchus ^ err anus chionaraia S err anus maytapi S err anus notospi lus Serran\:r phorbe S err anus subligariua Scrranus tabacarius SerranuE tortugarum SULFIDAE AchlruE infcrlptus Achirus lineatus Gymnachirus mela s T rinectes maciilatus SFAF I DAE Archosargus probatocephalus Archosargus rhomb oidalle piplodus holbrooki Lagodon rhomboides PagruB sedecim St eno tonus caprlnus Stgnotomus crysops SPHRA'^'IDA-^' Sphyraena borealls Sphyraena guachancho S phyraena picu'h\lT ' stk:phanob^ycidaf Stephanoberyx monae ST^lNOPT^iTCHIDAF Argyropelecus aculeatua Argyropelecus aJTlnis Argyropelecus gigas A r_ g y Topelecu8 henilgyinus PolyifnuE a?ti?roide3 Polyipnus laternatus Stemoptyx diaphana STROKAT'^IDAF Cubic eps melanus Cubiceps nlf^riargenteus Homeus ^ronovii Peprilus alepid od u a Peprilus paru Poronotus trl acanthus Psenes cyanophrya Psenes pacificus Psenes rep;ijlu3 Tetragonurus atlantleus STYLFPHORIDAE Stylephorus chorda tus SYNGNATHIDAJ: Hippocampus erectus MlcrogpathUB crinlgerus S yngnathus dunckerl Syngnathus elucens Syngi'iathug floridae Syngnathus fuFCus Syngnathue louigjanae Syngnathus rr^ckayl Syngnathus pelagicus Syngnathus scovelli " Syngnathus springerl 191-295 k 12?-1?5 h 101-167 h 97-160 h 78-23': k 82 1 373 1 105-113 li 133-175 U 92-155 u 27': -265 2 165-205 li 177-205 2 1L8 1 170 1 7U-m6 u 100-202 2 52-102 6 120 1 2ljO 1 127 1 122-205 3 U9-6t' 3 393 1 7U-10li U 65-9U 2 103-125 3 66-107 li 125-195 h 52-57 h 29-U4 3 U5-66 U 75-87 h 33-ii7 k 59-72 k 99-125 h U2-lliO h 60-76 k U8-117 h U9-63 3 70-93 5 9U-118 3 52-117 B 80-92 k 69-U2 h 168-195 3 137-157 h 72-76 h 129-162 h 109-iue h 107-Uili h 122-206 ? 168-220 h 2514 1 50-57 U 51-57 h 52 1 19-21 2 50-67 1. 27-hO 1* 32-li7 5 128-161. It 100-115 U 68-138 1. 61-77 U 178 1 115-120 u 67-123 ll 75-U2 3 83-193 U 160 1 26 10 2U 10 2U 10 2L 10 2L 10 2il 10 21t 10 2U 10 2U 10 21< 10 2a 10 21l 10 2L 10 2k 10 2U 10 21i 10 2h 10 26 10 2U 10 2U 10 21i 10 2U 10 21i 10 2U 10 26 10 21i 10 21i 10 2U 10 26 10 21. 10 21i 10 2U 10 2U 10 21. 10 2U 10 2U 10 2U 10 2L 10 2U 10 21. 10 27-28 9 28 9 35-36 9 28-29 9 21. 10 21. 10 21.-25 10-11 21. 10 21. 10 21. 10 2U 10 2U 12 21. 12 21l 12 35-36 11 36-39 11 39 n 38 11 33 10 33-31. 10 29-30 u 30-31 15-16 31 15 1.1-1.2 16 30 13 30 13 32 12 31 111 31 11. 31 U. U5 23 16 11. 11. 11. 11. li. 11. 11. Ill 11. 11. 11. 11. lU 11. 11. 11. 16 11. li. 11. lU 11. 11. 16 11. 11. U. 16 11. li. lU 11. U. 11. 11. lit 11. 11. 11. 18-19 19 26-27 19-20 11. 11. 11. 11. 11. u 11. 12 12 12 10 10 10 10 9 8 11 10 9-10 10 11 11 11 11 10 u 11 9-10 U 11 11 11 11 11 10 9 9 e 10 10 9-10 10 10 10 10 10 10 10 10 10 12 13 12-13 12 12 12 12 21.-2S 27-28 28 27 2;- 23-21. 16-19 15 12 16 12 25-26 12-13 17 3 17 3 20 3 17 11 17 10-11 17 12-13 22 15 li. 11 11 11 IS 12 19 1? 12 12 16 11. 16 11. m 13-11. 17 13-11. 17 17 16 16-17 16-17 16 15 18-19 11. 13 15 11-12 U 12 11-12 U-12 11-12 12 12 13-11. 12 12 51.-56 53-58 60-70 51-55 10-12 10-11 lU-lS 11 10 12 12 9 7-9 9 11.-15 11.-15 9-11 15-16 IS 27-26 lli.4.6 1.3 U-1.7 25-26 25-28 11.-15 9 105-180 1. 1.9-51 13 36-3E 19 -20 65 1 56 17 39 U7-S2 U 50-53 16 32-35 « Uo 2 U9-50 16 31-32 « 115-122 3 51-52 18 33-31. . 108-295 h 55-60 19-21 36-39 35 175-21.0 k 51-52 19-20 32 ie5-2UO h 50-53 20 30-33 27 -26 72-102 1. 1.9-52 18-19 31-33 28 -29 87-95 1. 1.9-51 17-16 31-33 29 -32 55-263 7 61-62 21.-25 37-36 6 1.0-1.3 13-11. 6 7 31.-35 9-10 9 7 3U-35 9-10 9 7 3U-35 9-10 9 6-9 35-36 9-10 9 8 33 6 9 9 31 7 9 7 39-1.0 12 9 7 38-1.0 11-12 9 7 37 11-12 9 9 37-1.0 10-12 9 9 33-35 6-9 9 8 35-37 9-10 9 9 31. 9 9 9 33 8 9 9 33-3U 8-9 9 8 36 10 9 8 39-1.1 12-13 6 12 38 U 9 U 38 U 9 11 37 10 9 10-12 37 10-11 9 U 37-38 10-11 9 11 37 10 9 7 33 9 8 9 1.0 12 9 8 33-3U 8-9 9 7-8 31.-35 9 9 6 33-31. 9-10 8 7 36-38 10-11 9 7 31.-35 9-10 9 7 32-33 6 9 7 35-36 10 9 7 32 8 9 6-7 37-38 10-11 9 7 38-39 U-12 9 7 36-38 10-11 9 7 31-32 7-8 9 7 3I.-36 9-10 9 7 33-31. 8-9 9 1.0-1.1. 15-16 7-8 1.0-1.3 16 6 U1.-51 16 6 ia-1.2 16 6 9-10 32-33 8-9 9 10 32-33 8 9 13-U 33-31. 6-9 9 11 31.-36 10-11 9 8 35-37 9-10 9 11-12 33-31. 9 9 11 3a-37 9-10 9 9 35 9 9 6 36 10 9 9 35 9 9 11-1? 39-1.1 10-11 10 7*7 32-36 f-11 10 7'5 9-10 10 7*5 - 9 10 15-17 35-36 9-11 10 15-16 36 10 10 11. 33-35 7-6 10 1I.-1S 35-37 9-10 9 15 35-37 9-10 9 26-27 31.-35 8-9 9 1.1-1.3 28 6 9 1.0 29 6 9 39-W. 31-lil. 7-9 9 26-27 33-3U 6-9 9 21.-25 32-31. 6-9 9 1I.-15 36-la 10-12 9 9 36 11 8 12-U. 8 7-9 8 9 6 7 10-11 10-11 8-9 10-U 8-9 9-10 7-6 9 11-13 IC 10 10 9-10 9-10 10 9 11 6 8-9 9 9-10 7-6 7-8 6-9 7 10 10-U 9-10 7 8-9 7 7-8 6 7-10 9-10 7-8 8-10 5-6 6-7 7 7-8 9-10 9-10 9 5 6 7-8 6 7-8 9-12 10 10 10 10 10 10 10 310 Table 1. — Meristic characters of some marine fishes of the western North Atlantic Ocean. — Continued. FAMTLT Genua, species Size range SL Speci- mens cxam- Caudal DORSAL FIN Spines Rays ANAL FIN Spines Rays CAUDAL FIN Dorsal I Dorsal ] Ventral 1 Ventral secondary primary prijaary secondary rays rays rays rays SYNODONTIDA? Saurida braslllensle Saurida caribbaea Saurida ncnriani Saurida Fuspic lo Synodug foettns S ynodus interFiedius Synodus poeyl Synodus 5 _aur u 3 Synodus syriocius Trachijiocephalus inyopa TETRAOtONTIDAE Canthigaater rostrata Lagocephalus laevipatus S phoeroides cutaneus Sphoeroides dorsalis Sphoeroides naculatus Sphoeroides nephelus Sphoeroides sp&ngler i Sphoeroides testudineua IRACHICHTHYIDAE Hoplostethua medlterraneus TniACANTHODTDAF Hollardia hollardla Parahollardia schmidti TRTCFTUPIDAF Benthod esmus sltnonyl Ben.hodesmus tenuis Trichiurus lepturus TP.ICLIDAE Bellator brachychir Bellator egrctta Bellator milltaris Bellator ri beirol iaia tus" Prionotug b eei Prionotus carollnus Prionoti-B evolai is Prionotus ophryas FrlonotuF paralatus Prionotus pect oralis Prio notus pun eta tus Prionotus ro^pus P rionotus rubjo F Vionotus scitulus Prionotus g t earnsl Prionotus tr'ilrulus PeristpdTor. antillarum Perlstedlon brevirosTre Peri=tedlon ecuadorensis Perlstedlon gracile Perlstedlon grt^yae Perlstedion imbpr^e Perlstedlon Perlstedlon longispatha minlatujn Periste 60 min 2— — 120— -^329- -^406— ^ Frequency of high tides In ^study period -^ 386— — 98— — 2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 32 (ml 7 8 9 HEIGHT OF TIDE 10 (ft) Figure 2. — Graphs showing range of flow of the Colum- bia River for various tidal heights at Astoria, Oreg. The number within each bar is the estimated discharge below which flow reversals of 60 min or longer occur at Prescott, Oreg., for a given tidal height. The number below each bar gives the frequency of occurrence of the tide height for the study period. height appear to be independent variables while flow reversal is the dependent variable in the analysis. Discussion During the time interval examined, reversal of direction of current (^ 60 min) at Prescott occurred every month except June 1968 and April and May 1969. It is important to discover that the reversals in flow are frequent as previ- ous work in this area only indicated that they could occur at very low river discharges. Be- cause of the frequent occurrence of flow rever- sals in the region of the lower Columbia River near Prescott, the regulation of waste discharges in this area should receive special attention as these reversals in flow tend to produce pockets of high concentration of wastes in the river system when the waste discharges are constant in time (Clark and Snyder, 1969). Acknowledgment We gratefully acknowledge the helpful sug- gestions of A. C. Duxbury, Department of Ocean- ography, University of Washington, in prepar- ing the manuscript. Literature Cited Clark, S. M., and G. R. Snyder. 1969. Timing and extent of a flow reversal in the lower Columbia River. Limnol. Oceanogr. 14:960- 965. Snyder, G. R., T. H. Blahm, and R. J. McConnell. 1971. Floating laboratory for study of aquatic or- ganisms and their environment. U.S. Dep. Com- mer., Natl. Mar. Fish. Serv., Circ. 356, 16 p. George R. Snyder Northwest Fisheries Center National Marine Fisheries Service, NO A A 2725 Montlake Boulevard East Seattle, WA 98102 Robert J. McConnell Northwest Fisheries Center Biological Field Station National Marine Fisheries Service, NO A A P.O. Box 1051 Longview, WA 98632 314 LONG-TERM OLFACTORY "MEMORY" IN COHO SALMON, ONCORHYNCHUS KISUTCW Many experiments have correlated the impor- tance of olfaction and the precise homing of sex- ually mature salmon. As juveniles, the fish are presumably imprinted on the natural odors of their natal-area water (Hasler, 1966). The odors apparently serve as cues to guide the adult's return. Thus, some type of "odor mem- ory" must persist from the time of the down- stream journey of the smolt to the return of the sexually mature adult. For introduced Lake Michigan coho salmon, Oncorhynchns kisiitch, this is either i/^ year (precocious males) or II/2 years. The existence of long-term olfactory memory persisting over this time period has only been inferred. Idler et al. (1961) and Fagerlund et al. (1963) found that already homed salmon made unconditioned behavioral responses to home water. Hara, Ueda, and Gorbman (1965) , Ueda, Hara, and Gorbman (1967), and Oshima, Hahn, and Gorbman (1969a) found specific EEG (electroencephalographic) responses to home- stream water. Hara ( 1970) in his review of this EEG technique states: "This electric response [from the olfactory bulbs] is specific in the sense that it cannot be evoked by water from spawning sites of other groups of breeding salmon." The EEG and behavioral studies strongly suggest long-term memory of the juvenile's stream ex- perience. However, since these workers used homed adults that had recently experienced home water, the data are evidence only of an odor memory lasting from the time of removal from the home stream to the time of testing. We tested coho salmon that were exposed to a synthetic odoriferous substance for 1 month during smoltification and then removed from any conceivable influence of this substance for 10 f \]^ Yt°^^ '^^^ completed under financial support -Tc^^v. National Science Foundation (Grant No. GB 7bl6X) to Prof. Hasler and the University of Wisconsin bea Grant Program which is a part of the National Sea Grant Program maintained by the National Oceanic and Atmospheric Administration of the U.S. Department of Commerce. months. Ten months later these fish and con- trols were examined for olfactory bulbar EEG responses (after Hara et al., 1965) to the im- printing substance. Materials and Methods On 7 April 1970, approximately 2,500 hatch- ery-raised coho salmon smolts (II/2 years old) were put into each of two contiguous 25-m sec- tions of a raceway at a Wisconsin State fish hatchery at Crystal Springs. We marked the fish to eventually distinguish the upper section control subjects from the lower section exper- imentals. A small drop (1/3 m) prevented water in the lower section from reentering the upper section. Immediately below the drop a dilute concentration of morpholine was introduced by infusion pump at a rate to maintain a steady- rate concentration of 10-^ ppm. This value is one order of magnitude above an avoidance threshold of unconditioned coho salmon finger- lings (Wisby, 1952). On 5 May 1970, 1 month after initiation of the morpholine treatment, all but 50 fish from each raceway section were trucked to Lake Michigan and released as part of another experiment. The 100 remaining fish were moved to a hatchery near Madison, Wis., and held together in a single outside raceway for 10 months prior to EEG tests. Our testing procedure was generally similar to that used by Hara et al. (1965) to examine olfactory bulb responses to home-stream water. The subject was paralyzed with gallamine trie- thiodide (2 mg/kg), restrained, and the gills perfused with tap water. One of the olfactory bulbs was exposed, and an electrode (Transidyne General, model 415') was placed on the surface near the rear margin. The responses evoked by perfusion of the ipsilateral naris were amplified (Bioelectric Instruments, model DS2c) and re- corded on a two-channel oscillograph (Hewlett Packard, model 7712B) for later analysis. This oscillograph was equipped with an integrating preamplifier for efficient quantification of bulbar activity. Therefore, all responses reported later Reference to trade names does not imply endorse- ment by the National Marine Fisheries Service, NOAA. 315 are expressed as the sum of the positive areas under the response wave form. Beginning on 25 February 1971, one fish was examined per day with 1 % and 0.01 % morpho- line stimuli. Fourteen fish were used. Each test was started at approximately 1000 hr. Every subject was tested with the morpholine concen- trations and the responses compared with re- sponses to 0.06 N NaCl. Stimuli were randomly ordered and presented for 10 sec followed by 75 sec of tap water rinse. The stimulus series was then repeated seven times. Results Fish that had been exposed to 10 ~' ppm of morpholine as smolting juveniles evidenced sig- nificantly higher bulbar EEG activity over con- trols when tested with 1% and 0.01% concen- trations of morpholine (Table 1). Responses to 1% morpholine gave a Mann- Whitney value of U = 5 (Siegel, 1956) with probability of 0.006 that the control group and the experimental group were drawn from the same treatment pop- ulation. Responses to the 0.01 % level were less markedly, but still significantly, different (U = 11, P = 0.049). Discussion Exposure to low concentrations of morpholine produced a sensitization which lasted at least 10 months. But, we did not attempt to determine whether this observed sensitization was exclu- sively to morpholine or to other stimulatory products. Casual observation of our data did not reveal the experimental subjects to be more responsive to NaCl than the controls; but this comparison was difficult to make in our experi- mental design because of the changing relation- ship between response amplitudes and back- ground activity levels. (Hence, the necessity of continual comparison of morpholine response with NaCl response, our reference.) Even if overall olfactory responsiveness is increased as a result of pretreatment, sensitization to mor- pholine is still proportionally greater (experi- mental vs. control) 10 months later. We hy- pothesize that exposure to morpholine imprinted the fish during one of the critical periods in the life of the coho salmon, a period when the fish is undergoing physiological changes in prepara- tion for entering a marine environment. Independent evidence indicates the existence of a critical period. The Wisconsin Department of Natural Resources allows approximately 1 month of imprinting during the period of smolt- ification before releasing the fish into the river system. Imprinting for less time or at different stages of the life cycle seems to result in more straying (Peck, 1970). Morpholine was chosen as the imprinting sub- stance since the responses of fingerling coho salmon to it have been investigated (Wisby, 1952). Consequently, the concentration could be chosen with knowledge of the performance parameters of the coho salmon. It was necessary to be above threshold but not so high as to cause enthusiastic avoidance or sublethal damage. Be- cause of the vagaries of the flow measurements Table 1. — Morpholine-elicited EEG responses of morpholine-imprinted coho salmon compared with those of controls. E designated subjects were imprinted with morpholine at a concentration of 10 "^ ppm; C designated were controls. Median EEG responses of each fish to 1% morpholine and 0.01% morpholine stimuli are ranked (ties carry aver- aged ranks) and Mann- Whitney U values and probability values are shown for each treatment level. 1% Group C C C C c C E E E E E C E E Median^ 63.5 100 140 220.5 230 250 255 266.5 287 333 351.5 600 915 917 Rank 1 2 3 4 5 6 U = 5 7 8 P = 0.006 9 10 U 12 13 14 0.01% Group C C E C C C E E C E E C E E Median^ 25.6 38.5 41.5 45 45 50 50 67 77.5 Rank 3 3 3 3 3 6 {/ = 11 7 3 P = 0.049 9.5 9.5 11.5 n.5 13 14 1 Median response = .response morpholine X 100) for eight trials. ^response 0.0