^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 clas