^y,^''^°''c-0. ^'■xrES o* *■ Fishery Bulletin Vol. 83, No. 1 January 1985 PEREZ FARFANTE, ISABEL. The rock shrimp genus Sicyonia (Crustacea: Decapoda: Penaeoidea) in the eastern Pacific 1 NEILSON, JOHN D., GLEN H. GEEN, and BRIAN CHAN. Variability in dimen- sions of salmonid otolith nuclei: implications for stock identification and micro- structure interpretation 81 NEILSON, JOHN D., and GLEN H. GEEN. Effects of feeding regimes and diel temperature cycles on otolith increment formation in juvenile chinook salmon, Oncorhynchus tshawytscha 91 1 m(m m^^mcd Ubcralory [ LIBHABY DEC 12 1985 Woods Hole, Mass. Seattle, Washington U.S. DEPARTMENT OF COMMERCE Malcolm Baldrige, Secretary NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION John V. Byrne, Administrator NATIONAL MARINE FISHERIES SERVICE William G. Gordon, Assistant Administrator Fishery Bulletin The Fishery Bulletin carries original research reports and te(^nical notes on investigations in fishery science, engineering, and economics. The Bulletin of the United States Fish Commission wks 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. Separates 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, nvunber 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. Government Printing Office, Washington, DC 20402. It is also available free in limited numbers to libraries, research institutions. State and Federal agencies, and in exchange for other scientific publications. SCIENTIFIC EDITOR, Fishery Bulletin Dr. William J. Richards Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA Miami, FL 33149-1099 Editorial Committee Dr Bruce B. Collette National Marine Fisheries Service Dr. Edward D. Houde Chesapeake Biological Laboratory Dr Merton C. Ingham National Marine Fisheries Service Dr. Reuben Lasker National Marine Fisheries Service Dr Donald C. Malins National Marine Fisheries Service Dr. Jerome J. Pella National Marine Fisheries Service Dr. Jay C. Quasi National Marine Fisheries Service Dr. Carl J. Sindermann National Marine Fisheries Service Mary S. Fukuyama, Managing Editor The Fishery Bulletin (ISSN 0090-0656) is published quarterly by the Scientific Publications Office, National Marine Fisheries Service, NOAA, 7600 Sand Point Way NE, BIN C15700, Seattle, WA 98115. Second class postage is paid at Seattle, Wash., and additional offices. POSTMASTER send address changes for subscriptions to Superintendent of Documents, U.S. Government Printing Office, Washington, IX; 20402. Although the contents have not been copyrighted and may be reprinted entirely, reference to source is appreciated. 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 1 April 1985. Fishery Bulletin CONTENTS Vol. 83, No. 1 January 1985 PtREZ FARFANTE, ISABEL. The rock shrimp genus Sicyonia (Crustacea: Decapoda: Penaeoidea) in the eastern Pacific 1 NEILSON, JOHN D., GLEN H. GEEN, and BRIAN CHAN. Variability in dimen- sions of salmonid otolith nuclei: implications for stock identification and micro- structure interpretation 81 NEILSON, JOHN D., and GLEN H. GEEN. Effects of feeding regimes and diel temperature cycles on otolith increment formation in juvenile chinook salmon, Oncorhynchus tshawytscha 91 jVlsrine Biste^isal Ufeoratory | LIBRARY DEC 12 19B5 Woods Hole, Mas s. Seattle, Washington 1985 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washing- ton DC 20402 — Subscription price per year: $21.00 domestic and $26.25 foreign. Cost per single issue: $7.00 domestic and $8.75 foreign. The National Marine Fisheries Service (NMFS) does not approve, recommend or endorse any proprietary product or proprietary material mentioned in this pub- lication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales promotion which would indicate or imply that NMFS approves, recommends or endorses any proprietary product or proprietary material mentioned herein, or which has as its purpose an intent to cause directly or indirectly the advertised product to be used or purchased because of this NMFS publication. THE ROCK SHRIMP GENUS SICYONIA (CRUSTACEA: DECAPODA: PENAEOIDEA) IN THE EASTERN PACIFIC Isabel Perez Farfante^ ABSTRACT The genus Sicyonia is redefined and the 12 species occurring between Monterey Bay, California, and off Pisco, Peru, are treated in detail. A key to species is followed by illustrated species accounts including descriptions, ranges of intraspecific variation with analyses of morphometric data (rostrum to carapace ratio graphically represented for 10 species), and color notes. The size ranges at which males and the minimum sizes at which females attain adulthood are summarized, and ecological notes together with maps illustrating the ranges of the species (six of which have been extended beyond limits previously reported) are included. Sicyonia disparri seems to be restricted to the south and gulf coasts of Baja California and waters off Nayarit, Mexico; S. affinis to waters off Costa Rica, Panama, and Colombia; and S. penicillata occurs on the ocean side of Baja California Sur, Mexico, and from the Gulf of California to Costa Rica. Sicyonia ingentis ranges from Monterey Bay to Nayarit, including the Gulf of California. Sicyonia disedwardsi and S. martini occur along the ocean side of Baja California Sur, in the Gulf of California, and southward to Panama, and four others, S. aliaffinis,S. disdorsalis,S. mixta, and S. picta, frequent the same waters, but also reach as far south as Peru. Sicyonia laevigata and S. brevirostris are found on both sides of the Continent, the former at the southern end of the Gulf of California and from off Costa Rica to the Golfo de Panama in the Pacific, and from North Carolina to Santa Catarina, Brazil, in the Atlantic. Sicyonia brevirostris has been recorded from the Golfo de Tehuantepec and from Virginia southward through the Gulf of Mexico to Yucatan, and also from the Bahamas and Cuba. In addition, there is an unverified literature record from Guyana. The treatment of each species is concluded with a comparison of its diagnostic features with those of the most closely allied congeners and a note on its present or potential economic value. Until a few years ago, members of the genus Sicyonia, "rock shrimps", were discarded from the large commercial catches of panaeoid shrimps made in tropical and subtropical waters of the eastern Pacific and western Atlantic. It was com- monly thought that because of their hard, stony exoskeleton, they would be rejected by both con- sumers and the processing industry; however, in- creased demand for shrimp encouraged the fishermen and dealers to bring the larger species to market, and now production is not only readily absorbed, but some prefer rock shrimps to the thinner shelled species. The exploitation and comparatively recent rec- ognition of the commercial potential of Sicyonia, the most distinctive group within the superfamily Penaeoidea, have motivated this review of mem- bers of the genus found in the American Pacific (the western Atlantic species have already been the object of a number of studies, e.g., Chace 1972; Huff and Cobb 1979). For the most part, the infor- Systematics Laboratory, National Marine Fisheries Service, NOAA, National Museum of Natural History, Washington, DC 20560. mation available is limited to the original descrip- tions of the species, which are scattered in works published between 1871 and 1945, and to a limited number of locality records. Of the 12 species occur- ring in the region, 4 had been recognized prior to the close of the century. No other species were reported from these waters until Burkenroad made his invaluable studies (1934-45) which re- sulted in the recognition of five new species plus two others previously known to occur only in the western Atlantic. Recently, Perez Farfante and Boothe (1981) described the 12th member of the genus frequenting the eastern Pacific. Two works have been helpful in the identification of the American Sicyonia: one by Anderson and Lindner (1945) which provided a key to the then known species; the other by Arana Espina and Mendez G. (1978) in which was presented an illustrated key, diagnoses, and ranges of the species found in Peruvian and Ecuadorean waters, and an analysis of morphometric relations, with data on the growth and molting pattern of one of the species. The extensive collections (515 lots containing 4,672 specimens) of Sicyonia available from Mon- terey Bay, Calif., to off Pisco, Peru, and the oppor- Manuscript accepted April 1984. FISHERY BULLETIN: VOL. 83, NO. 1, 1985. l'^ o. "H t^ o o CJ Q s < c o CO o c "a CO Sm e X «, c r! 01 ^ ex ,o o. CO ^-* ^*-< n U 0> «' "•> m m o 1) (-1 -i (i> CO n o J t-H r o CO T3 CO 0) O 13 <1) CO s c CO CO to U "o ^ F (U E > CO CO ^3 ^ n> Tt" > CO 05 T-H 03 "^ CO r ^ o P -t^ q) a. -i«: u CJ 1 l/J m _C « T3 S- 0) CO "C « c n ^ 73 lO w OS D O £1 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 6. — Thelycum of Sicyonia showing terms used in descriptions. Posterior thoracic ridge ^Sternite XI Sternite XII iSternite XIII ^Median plate ii^uerior component Sternite XIV Plat^ of sternite XIV Figure 7. — Sicyonia disedwardsi, 9 30 mm cl, Golfo de Panama, Panama. Dorsal view of sperm receptacles. Scale = 1 mm. ^'■**^ consisting of trilobed membranous sacs: large, longitudinally disposed, mesial lobe, extending to posterior margin of sternite XIV, and two small anterior lobes, one directed anteriorly and the other laterally; receptacles opening by transverse slits at anterolateral margins of plate of sternite XIV. Median plate of sternite XIII (supported by buttress of sternite XII) large, triangular or flask-shaped, tapering anteriorly in long, sharp 8 spine; plate bearing well-defined lateral constric- tions setting off posterior component. Sternite XI armed posteriorly with paired spines of variable size. Branchial components consisting of pleu- robranchia on somite IX; single arthrobranchia on somite VII (lamellar rudiment; Fig. 4D-b); an- terior and posterior arthrobranchiae on somites VIII through XIII, anterior member of VIII very small and that of XIII vestigial; and podobranchia PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA on second maxilliped. Epipod on first and second maxillipeds and on first through third pereopods. [Modified from Burkenroad (1934a, b) and Kubo (1949).] The species of this genus (about 55), the only one encompassed in the family Sicyoniidae, occur in tropical, subtropical, and temperate waters. They are marine, some invading brackish waters, and abundant at shallow to moderate depths; a number of them also range across the continental shelf onto the upper slope, reaching depths of several hundred meters. Members of Sicyonia exhibit a wide range of intraspecific variation. The number and position of the dorsal and apical rostral teeth and the ar- rangement of teeth on the postrostral carina vary as do the depth, extension, and form (continuous or interrupted) of the abdominal sulci. Although the number of the latter almost always provide a reli- able diagnostic character for specific identifica- tion, the posterior pleural sulcus may be present or absent in some species. Furthermore, in females the contour and sculpture of the thelycal plate of sternite XIV and the shape of the posterior margin of that of XIII also vary as pointed out in the accounts for most of the species. In contrast, the petasma of each of the members of the genus is virtually invariable and thus useful for specific recognition, and the appendix masculina is almost constant in shape in all of the species. In the females, sternite XI is armed with a pair of spines which vary considerably in length, rang- ing from quite small, as in all members oi Sicyonia occurring in the eastern Pacific, to extremely long, as in the western Atlantic iS. olgae Perez Farfante 1980. The posterior thoracic ridge varies but within a range that does not prevent its having at least limited diagnostic value. Species of Sicyonia also exhibit various mor- phological changes with increasing size. The shape of the anteroventral margin of the pleuron of the first abdominal somite may change gradu- ally from straight in juveniles to pronouncedly concave in adults, and the anteroventral ex- tremities of the pleura of the first three or four abdominal somites, from rounded to sharply acute. The pleural armature, too, undergoes mod- ifications with age; as Burkenroad (1934a) stated, "A rounded angle usually precedes an unarmed but acute angle, and this a veritable tooth, in the course of individual development"; likewise, the abdominal sculpture is altered, the sulci usually becoming deeper as the animal grows. The forego- ing statements indicate that abdominal features which are diagnostic for the identification of adults have scant systematic usefulness in iden- tify ing juveniles. Another characteristic of the genus is the wide range in size among the species; whereas some are small (the eastern Pacific S. disparri reaches a maximum total length of about 44 mm), others are quite large iS. ingentis and S. brevirostris attain a total length of about 133 and 153 mm, respec- tively). In the genus Sicyonia there are two series of species the contrasting characters of which would appear to justify their separation into two genera. The members of one series (occurring on both sides of the Atlantic, in the Mediterranean and Indo- Pacific, and represented in the eastern Pacific by S. disparri and S. laevigata) lack or occasionally exhibit a very minute antennal spine, are armed with basial and ischial spines on the first pair of pereopods, bear an incision or a notch on the dor- somedian carina of the second abdominal somite, and usually display a conspicuous notch on the lateral margin of the petasma. The members of the other series (restricted to American waters) have a well-developed, buttressed, antennal spine, lack basial and ischial spines on the first pair of pereopods, have no incision on the carina of the second abdominal somite, and never bear a notch on the lateral margin of the petasma. These two series, representing extremes of the range of vari- ation in members of the genus, were first recog- nized by Burkenroad (1934a) as Division I and Division II, respectively. Both in 1934a and 1945, he stated that there are species or series of species in which some of these characters are inter- changed. Certain species (all from the Indo- Pacific) that lack an antennal spine and in which the first pair of pereopods are armed exhibit an entire carina on the second abdominal somite (the first two are characters of Division I, and the last of Division II). At least one species (also occurring in the Indo-Pacific) lacks an antennal spine and has armed first pereopods (both characters of Division I) but bears an unnotched carina (a feature of Division II). Another species (the eastern Pacific S. mixta) that possesses an antennal spine and has armed first pereopods exhibits a clearly distinct depression on the carina of the second abdominal somite which seems to correspond to the notch characteristic of Division I. Burkenroad (1934a) also divided his Division II into species-groups, each named for one of the species belonging to it. They were characterized by the number, size, and position of the teeth on the FISHERY BULLETIN: VOL. 83, NO. 1 postrostral carina. The complex intergradation of the characters that have been used to recognize these "Divisions" and "groups" of Sicyonia seems to demonstrate that there are no superspecific dis- junctions that will justify their being accorded subgeneric or generic rank. In the present work, occasionally, reference is made to these subdivi- sions. Key to the American Pacific species of Sicyonia la. First abdominal somite lacking tooth on dorsomedian carina S. mixta lb. First abdominal somite bearing anterior tooth on dorsomedian carina 2 2a. Antennal spine absent or exceedingly weak and without buttress; second abdom- inal somite with perpendicular incision in anterior half of dorsal carina; first pereopod with short distomesial spine on basis and ischium 3 2b. Antennal spine well developed and but- tressed; second abdominal somite not incised; first pereopod with basis and ischium unarmed 4 3a. Postrostral carina armed with 3 sub- equal teeth, anterior (epigastric) one as large, or almost as large, as posterior 2 teeth; anteromedian sulcus of first abdom- inal somite well marked to near ventral margin of pleuron; posteroventral extrem- ity of fourth abdominal somite dis- tinctly angular S. disparri 3b. Postrostral carina armed with 3 unequal teeth, anterior one conspicuously smaller than posterior 2 teeth; anteromedian sul- cus of first abdominal somite short, often obscure, ending well above ventral mar- gin of pleuron; posteroventral extremity of fourth abdominal somite never distinct- ly angular S. laevigata 4a. Postrostral carina with 2 or 3 teeth pos- terior to level of hepatic spine 5 4b. Postrostral carina with 1 tooth posterior to level of hepatic spine 7 5a. Postrostral carina almost always with 3 teeth posterior to level of hepatic spine, occasionally anterior one of these at level of or slightly anterior to hepatic spine S. brevirostris 5b. Postrostral carina with 2 teeth posterior to level of hepatic spine, never with tooth at level or slightly anterior to hepatic spine 6 6a. Rostrum bearing 2 dorsal teeth; petasma with distal projections short and stout; thelycum with plate of sternite XIV raised in low (sometimes indistinct) bulges and with posterior component of median plate traversed by weak suture; branchiostegite with large ocellus consist- ing of well-defined yellow center sur- rounded by purplish brown ring S. disedwardsi 6b. Rostrum usually bearing 1 dorsal tooth (rarely 2); petasma with distal projections extremely long and slender; thelycum with plate of sternite XIV raised in strong bulges and with posterior component of median plate traversed by deep groove; branchiostegite with moderately large, purplish brown spot sometimes bearing poorly defined but diffuse yellow center S. penicillata 7a. Postrostral carina behind posterior tooth high, conspicuously elevated in arched crest 8 7b. Postrostral carina behind posterior tooth low, not elevated in high crest 11 8a. Fifth abdominal somite without tooth or sharp angle at posterior end of dorsome- dian carina; anteroventral extremity of pleuron of fourth abdominal somite broad- ly obtuse and unarmed S. affinis 8b. Fifth abdominal somite with tooth or sharp angle at posterior end of dorsome- dian carina; anteroventral extremity of pleuron of fourth abdominal somite sharp- ly angular or armed with spine 9 9a. Rostrum long, conspicuously surpassing distal margin of eye; anteroventral angle of second through fourth abdominal somites unarmed, lacking spine; petasma with projection of dorsolateral lobule dis- tinctly bifurcate apically S. martini 9b. Rostrum short, falling short of, or infre- quently barely surpassing distal margin of eye; anteroventral angle of second 10 PfiREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA through fourth abdominal somites armed with curved spine; petasma with projec- tions of dorsolateral lobule simple 10 10a. Rostrum armed with 2 dorsal teeth; ab- domen heavily tuberculate; anteromedian pleural sulcus of first abdominal somite well marked to near ventral margin of pleuron; anteromedian pleural sulcus of second and third somites reaching dor- sally posteromedian pleural sulcus; bran- chiostegite with horizontally disposed 9-shaped, brown mark S. aliaffinis 10b. Rostrum armed with 3 or 4 dorsal teeth; abdomen with very few tubercles; an- teromedian pleural sulcus of first abdom- inal somite well marked only dorsally; an- teromedian pleural sulcus of second and third somites not reaching dorsally pos- teromedian pleural sulcus; branchioste- gite with ocellate mark, consisting of red center surrounded by yellow ring S. picta 11a. Dorsomedian carina of first abdominal somite produced in tooth considerably larger than posterior tooth on carapace, that of fifth truncate or produced in spine at posterior end; petasma with distal pro- jections of dorsolateral lobules turned mesially and lacking terminal filament; thelycum with posterior component of median plate flat or slightly raised pos- terolaterally, not bearing short bosses cut by transverse suture S. disdorsalis lib. Dorsomedian carina on first abdominal somite produced in tooth smaller or only slightly larger than posterior tooth on carapace, that of fifth sloping gradually to posterior end; petasma with distal projec- tions of dorsolateral lobules divergent and produced in short filament; thelycum with posterior component of median plate bear- ing paired short lateral bosses cut by transverse suture S. ingentis Sicyonia laevigata Stimpson 1871 Figures 8-12 Sicyonia laevigata Stimpson 1871: 131 [type not extant; type-locality: Charleston, S.C.]. Kingsley 1878:69, 1880:426. Rathbun 1901: 103 [part]. De Man 1911:11. Bouvier 1918:6. Hay and Shore 1918:379, pi. 25, fig. 1. Rath- bun 1920:319. Burkenroad 1945:5. Lunz 1945:4, fig. 1. Pearse and Williams 1951: 143. Wass 1955:142. Menzel 1956:41. Hut- ton et al. 1959:6. Wells 1961:248. Williams 1965:33. Fausto Filho 1966a:32, 1966b:47, fig. 8. Eldred et al. 1965:32. Joyce and El- dred 1966:24. Fausto Filho 1968:73. Rouse 1969:136. Bayer et al. 1970:41. Lyons et al. 1971:28. Garcia Pinto 1971:5. Chace 1972:11. Camp et al. 1977:23. Rodriguez de la Cruz 1977:11. Huff and Cobb 1979:67, fig. 38a-d. Rodriguez 1980:70. Perez Farfante 1980:773. Greening and Living- ston 1982:151. Coen and Heck 1983:206. Williams 1984:47. Sicyonia sculpta var. americana De Man 1907:450. Sicyonia carinata De Man 1907:451. [Not Cancer carinatus Briinnich 1768 = Sicyonia carinata.] Sicyonia carinata var. americana De Man 1911:10. Eusicyonia laevigata. Burkenroad 1934a:76, fig. 21, 26, 32, 1934b:117. Schmitt 1935:132. Burkenroad 1938:80. Lunz 1945:4, fig. 1. Sicyonia carinata americana. Burkenroad 1934a:76. Vernacular names: rock shrimp, hardback, coral shrimp (United States); camaron de piedra, cacahuete (Mexico); camaron conchiduro (Mexico, Panama); camarao-da-pedra (Brazil). Diagnosis. — Antennal spine absent or barely dis- tinct and lacking buttress. Second abdominal so- mite with perpendicular incision on dorsomedian carina. First pereopod armed with short spine on basis and ischium. Postrostral carina bearing three unequal teeth, epigastric one considerably smaller than posterior two teeth. Anteromedian sulcus of first abdominal somite, if distinct, short, ending well above margin of pleuron; posteroven- tral extremity of fourth abdominal somite not dis- tinctly angular. Petasma with distal projection of dorsolateral lobule almost straight but with apical part curved dorsally. Thelycum with plate of ster- nite XIV produced in elongate anterolateral lobules, their anteromesial borders strongly di- vergent. Description. — Body moderately robust (Fig. 8) and lacking tubercles. Carapace with patches of short setae on dorsum, as well as ventral and an- terior to posterodorsal part of branchiocardiac carina; extremely long setae flanking base of, and between, teeth of postrostral carina. Rostrum long, reaching as far as distal end of 11 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 8. — Sicyonia laevigata, 9 9 mm cl, Punta Paitilla, Panama. Lateral view. Scale = 5 mm. antennular peduncle, its length (0.40-0.77 cl) in- creasing linearly with carapace length (Fig. 9); relatively deep, usually with dorsal and ventral margins straight and subparallel; subhorizontal or upturned to as much as 35°; armed with two (25%), three (72%), or four (3%) dorsal teeth and two (5%), three (33%), or four (62%) apical teeth, additional rudimentary one present between two ventral ones; ventralmost tooth, largest of apical cluster, subterminal, directed anteriorly or curved upward, and distinctly removed from adjacent one. Position of first dorsal tooth ranging between 0.09 and 0.30 (mean 0.18) rl from level of orbital mar- gin, and that of second tooth between 0.42 and 0.60 (mean 0.45) rl; third tooth, when present, between 0.76 rl and tip of rostrum (mean 0.87); and fourth, rarely present, located at tip. Two, occasionally one or three, minute but rather heavy movable spinules (often absent in adults) flanking ventral margin of rostrum, just posterior to apical teeth. Adrostral carina, subparallel to ventral margin of rostrum, extending to base of apical cluster. Carapace with postrostral carina, barely dis- tinct between teeth, bearing three unequal, acutely produced teeth: 1) epigastric tooth, situated between 0.06 and 0.16 (mean 0.12) cl from orbital margin, subequal to or only slightly larger than first rostral and conspicuously smaller than more posterior teeth; 2) middle tooth, placed be- tween 0.35 and 0.50 (mean 0.45) cl from orbital margin; and 3) posterior tooth positioned be- tween 0.68 and 0.80 (mean 0.72) cl from orbital margin, well in advance of posterior margin of carapace. Antennal spine absent or barely per- ceptible, antennal angle 90° to broadly obtuse, lacking buttress; hepatic spine long, sharply pointed, projecting from low buttress, and situated * 4 6 9 carapace length (mm) 12 15 Figure 9. — Sicyonia laevigata. Relationship between rostrum length and carapace length (regression equation, >> = -0.39153 + 0.64127X). between 0.22 and 0.30 (mean 0.25) cl posterior to orbital margin. Postocular sulcus short; hepatic sulcus shallow, subhorizontal; hepatic carina lack- ing; branchiocardiac carina barely evident. Ocular calathus broad and bearing conspicuous tuft of setae on dorsolateral extremity; ratio of length of lateral margin to width (across base of cornea) 0.54-0.65 (mean 0.60) and ratio length of lateral margin to cl 0.11-0.15 (mean 0.13). Antennular peduncle with stylocerite produced in acute spine reaching 0.65-0.75 of distance be- tween lateral base of first antennular article and mesial base of distolateral spine; latter extending only to about proximal 0.40 of second antennular article. Antennular flagella short, mesial one, about 0.7 as long as lateral; latter, about 0.4 cl. Scaphocerite almost reaching or slightly sur- 12 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA passing distal end of third antennular article; lat- eral rib produced in long, slender spine slightly overreaching lamella. Antennal flagellum as much as 3 times as long as carapace. Third maxilliped much stouter than pereopods. Basis and ischium of first pereopod armed with short but well-developed sharp spine projecting from distomesial margin. Abdomen with dorsomedian carina low an- teriorly, increasing in height posteriorly; carina on first somite produced in rather small, anteriorly directed tooth, smaller than posterior tooth on carapace; carina on second somite conspicuously incised (just dorsal to juncture of tergal sulci) and that on sixth terminating in short, sharp tooth. Antero ventral extremity of pleuron of first three somites rounded; pleuron of fourth with postero- ventral extremity angular or subangular, always unarmed; postero ventral extremity of fifth and sixth somites produced in very small, caudally directed, sharp tooth. First somite with short anteromedian pleural sulcus (usually well marked but sometimes obscure dorsally, obsolete, or represented by shal- low depression ventrally) and long, united pos- terior tergal-posteromedian pleural sulci. Second and third somites marked by 1) long anterior ter- gal (extending to base of pleuron) and short pos- terior tergal sulci; 2) anteromedian pleural sulcus, extending to near ventral margin on second somite but restricted to dorsal part on third, in both so- mites delimiting anterior shallow depression set- ting off weak prominence dorsally; and 3) pos- teromedian pleural sulcus, its dorsal extremity curving anteriorly ventral to (not joining) pos- terior tergal sulcus. Fourth and fifth somites with anterior tergal sulcus (that of fourth obliterated about midlength), merging with united posterior tergal-posteromedian pleural sulci. Sixth somite marked by weak, arched posteromedian pleural sulcus and bearing shallow longitudinal depres- sion between low dorsolateral ridge and elongate cicatrix. Telson with pair of short but strong fixed spines and two longitudinal rows of movable spinules on each side of median sulcus. Rami of uropod reach- ing or barely overreaching apex of telson. Petasma (Fig. lOA, B) with rigid distal projec- tion of dorsolateral lobule bulbous proximally, al- most straight but with terminal part strongly curved dorsally; fleshy distal projection of ven- trolateral lobule directed distolaterally, broad basally, and with slender but blunt terminal part curved proximally. Lateral margin of petasma Figure lO. —Sicyonia laevigata, 6 5.7 mm cl, Isla Taboga, Panama. A, Petasma, dorsal view; fi, ventral view; C, right appendix masculina, dorsolateral view. Scale = 0.5 mm. 13 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 11. — Sicyonia laevigata. Thelyca. A,? 7 mm cl, Isla San Lucas, Costa Rica; B, J 9 mm cl, Punta Paitilla, Panama. Scales = 0.5 mm. slightly to conspicuously notched near midlength. Petasmal endopods coupled in males with carapace length as little as 2.9 mm, about 13 mm tl, petasma joined in all larger males. Appendix masculina as illustrated in Figure IOC. Thelycum (Fig. ILA, B) with plate of sternite XIV produced in elongate, subalate anterolateral lobules, their anteromesial borders strongly di- vergent; plate flat except for deep, broad, antero- median depression, limited posteriorly by raised margin of posterior thoracic ridge. Median plate of sternite XIII narrow, lanceolate, tapering gently into sharp spine reaching as far as midlength of coxae of extended second pereopods; plate con- stricted and excavate at level of fourth pereopods; posterior component of plate with posterior mar- gin entire to deeply emarginate and traversed by weak suture. Paired conspicuous spines projecting anteroventrally from posterior margin of sternite XI. Posterior thoracic ridge insensibly fused later- ally with plate of sternite XIV. The smallest impregnated female observed has a carapace length of 5 mm, about 18 mm tl. Co/or.— Huff and Cobb (1979) presented a detailed 14 account of the color pattern of this species based on specimens collected on Florida's west central shelf Maximum size. — Males 7.3 mm cl, about 34 mm tl; females 15 mm cl, about 53 mm tl. Geographic and bathymetric ranges. — In the American Pacific, S. laevigata is known from Mazatlan (lat. 23°13'N, long. 106°25'W), Mexico (Rodriguez de la Cruz 1977); Isla San Lucas (9°56'N, 84°54'W), Golfo de Nicoya, Costa Rica; and the Golfo de Panama, in the latter as far as Punta Paitilla (8°58'N, 79°31'W), Panama. In the western Atlantic this species ranges from off Cape Hatteras (35°08'30"N, 75°10'00"W), N.C., south- ward and into the Gulf of Mexico to northwest Florida, and off Yucatan. Also, it occurs through the Antilles and around the Caribbean coast of Mexico, Central America, and South America, and along the Atlantic coast of South America to Anse de Zimbros (27°13'S, 48°31'W), Santa Catarina, Brazil (Fig. 12). In the Pacific, this shrimp has been taken from tide pools to a depth of 4-9 m, but in the Atlantic it occurs from the shore to as deep as 90 m. It occurs PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA on corals or other firm, rocky or shelly substrates, but occasionally, as reported by Wass (1955), on soft mud. On the basis of their samplings and the observa- tions made by Lyons et al. (1971), Huff and Cobb (1979) concluded that S. laevigata "shows some preference for grass habitats in estuarine and nearshore environments, associating with coarse substrates further offshore where seagrasses are absent." This species has been found at salinities be- tween 22 and 37%o (Lyons et al. 1971 and Menzel 1956, respectively) and temperatures between 17° and 32°C (Lyons et al. 1971 and Camp et al. 1977, respectively). Discussion. — Sicyonia laevigata is most similar to the Pacific S. disparri and the western Atlantic S. parri, but it may be readily distinguished from them by the following features. The epigastric tooth in S. laevigata is smaller than, instead of almost as large as, the other teeth on the postros- tral carina. Also in S. laevigata this tooth is al- ways situated anterior to the hepatic spine, be- tween 0.06 and 0.16 (mean 0.12) cl from the orbital margin, whereas in S. disparri it is often located opposite or posterior to the hepatic spine but, if anterior, usually farther from the orbital margin, between 0.12 and 0.25 (mean 0.17) cl. The an- teromedian pleural sulcus of the first abdominal somite in S. laevigata is short, commonly well defined (sometimes obscure) dorsally, but obsolete, cr represented by a shallow depression ventrally; in contrast, in S. disparri and S. parri it is long, extending to near the ventral margin of the pleuron, although sometimes it is interrupted dor- sal to midlength. In S. laevigata the posterior pleural sulci are lacking, as they usually are in S. disparri, whereas in S. parri they are present. Furthermore, in the petasma of S. laevigata the distal projection of the dorsolateral lobule is di- rected distally, its tip curved dorsally, whereas in S. disparri and S. parri it is strongly curved mesially — but in the latter the tip is bent slightly dorsally. It should be pointed out that S. parri , like the other two species, lacks an antennal spine, possesses a spine on the basis and ischium of the first pereopod, and bears a dorsal incision on the dorsomedian carina of the second abdominal so- mite. Remarks. — In 1980, 1 stated that the dorsal mar- gin of the rostrum of S. laevigata is armed with three teeth, an opinion that was contrary to that of Figure 12. — CJeographic distribution of Sicyonia laevigata, S. disparri, and S. mixta. 15 FISHERY BULLETIN: VOL. 83, NO. 1 previous authors who noted that it bears only two; 1 was convinced at the time that they considered the variably situated third tooth, when located subterminally, an apical instead of a dorsal tooth. The study of the extensive collection available to me for the present project has indicated that this species sometimes (25% of the specimens) bears only two teeth, and occasionally four (3% of the individuals). It should be mentioned that the holotype of this species was lost during the Chicago fire of 1871. Material. — 73 specimens from 41 lots. Eastern Pacific — 12 specimens from 6 lots. Costa Rica — 15, USNM, Isla San Lucas, 5 January 1930, M. Valerio. Panama— Id 19, AHF, Isla Taboga, 4-9 m, 2 May 1939. 39, SIO, Isla Taboga, 3 m, 30 March 1967, R. Rosenblatt. 19, USNM, Punta Paitilla, intertidal, 1 July 1969, L. G. Abele and J. Gra- ham. 19, USNM, Punta Paitilla, intertidal, 17 June 1969, J. Graham. 49, USNM, Punta Pai- tilla, 8 m, 13 April 1972, C. E. Dawson and party Western Atlantic — 61 specimens from 34 lots. United States— North Carolina: 19, USNM, off Cape Hatteras, 90 m, 17 October 1885, Alba- tross stn 2596. 29 , USNM, off Morehead City, 14 July 1913, Fish Hawk. 39, USNM, Boque Sound (off Morehead City), August 1912, Fish Hawk . 19 , USNM, Black Rocks (off New River), 13 m, 8 Au- gust 1949, A. S. Pearse. South Carolina: 19, USNM, mouth of Bulls Creek, 1891, Fish Hawk . 16 , USNM, Charleston Harbor, 14.5 m, 13 March 1891, Fish Hawk stn 1659. 19 , USNM, off S end of May River, Calibogue Sound, 18 m, 16 January 1891, Fish Hawk stn 1651. Flori- da: 19, USNM, Biscayne Bay, 7 July 1960, B. Petskin. 19 , MCZ, off Key West, J. R. Miller. 16 49, USNM, Marco, 2-5.5 m, H. Hemphill. 16, USNM, Punta Rassa, 2 m, February 1884, H. Hemphill. 19, USNM, Charlotte Harbor, March 1887, W H. Dall. 26 29, USNM, Sarasota Bay H. Hemphill. 16, USNM, St Martin, 5.5 m, 15 January 1902, Fish Hawk stn 7222. 16 39, USNM, off St Martin's Reef, 5 m, G. F Moser 26 , USNM, off NW end St Martin's Reef, 1887, G. R Moser. 19, USNM, Aucilla, 9 m, 6 November 1901, Fish Hawk stn 7148. Cuba— Id 19, USNM, off Cayo Levisa, 4-5.5 m, 2 June 1914, Tomas Barrera Exped stn 14. Jamaica— Id, USNM, SE of Great Pedro Bluff, 27-29 m, 6 July 1970, Pillsbury stn 1223. Haiti— 19, AMNH, Port-au-Prince, W Beebe. Puerto Rico— 29, USNM, Mayagiiez, 23 m, 20 January 1899, Fish Hawk stn 6093. Id 39, USNM, Mayagiiez Harbor, 19-20 January 1899, Fish Hawk . 29 , USNM, off Humacao, 23 m. Fish Hawk. 26 29, USNM, off Isla Culebra, 27 m, 8 February 1899, Fish Hawk stn 6093. Virgin Islands— St. Thomas: 19 , USNM, 1884, Albatross. Mexico — Quintana Roo: Id, USNM, off Isla Mujeres, 29.5 m, 11 June 1962, Oregon stn 3638. Id , USNM, SE of Isla Mujeres, 101-275 m, 9 September 1967, Gerda stn 880. 19, USNM, Bahia de la Ascension, 17 April 1960, Smith- sonian-Bredin Caribbean Exped stn 85-60. Nicaragua — 29, USNM, off NE Nicaragua, 55 m, 8 June 1964, Oregon stn 4930. 29 , USNM, off Prinzapolca, 27 m, 28 January 1971, Pillsbury stn 1335. 19, USNM, E of Isla del Venado, 24 m, 28 January 1971, Pillsbury stn 1330. Id 19 , USNM, SE of Punta de Perlas, 27 m, 28 January 1981, Pillsbury stn 1334. Colombia— 19, USNM, Sabanilla, 1884, Alba- tross. 19, USNM, off Tucuracas, 9 m, 30 July 1968, Pillsbury stn 778. Brazil — Santa Catarina: 19, MP, Anse de Zimbros, 5-0 m, 16 December 1961, Calypso stn 148. Sicyonia disparri (Burkenroad 1934) Figures 5, 12-17 Eusicyonia disparri Burkenroad 1934a:83, fig. 27 [holotype: 9, YPM 4392, Bahia San Luis Gon- zaga, Baja California Norte, Mexico, 17 May 1926, Pawnee]. Burkenroad 1938:80. Anderson and Lindner 1945:315. Sicyonia disparri. Brusca 1980:256. Vernacular names: rock shrimp (United States); camaron conchiduro, camaron de piedra, cacahuete (Mexico). Diagnosis. — Antennal spine absent. Second ab- dominal somite with perpendicular incision on dorsomedian carina. First pereopod armed with short spine on basis and ischium. Postrostral carina bearing three subequal teeth, epigastric one as large, or almost as large, as posterior two teeth. Anteromedian sulcus of first abdominal so- mite clearly distinct to near ventral margin of pleuron; posteroventral extremity of fourth ab- dominal somite markedly angular Petasma with distal projection of dorsolateral lobule curved me- 16 PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA sially to apex. Thelycum with plate of sternite XIV produced in elongate anterolateral lobules, their anteromesial borders strongly divergent. Description. — Body moderately robust (Fig. 13) and lacking tubercles. Carapace with short setae scattered over surface and extremely long ones between and flanking base of teeth on postrostral carina. larger adults) flanking ventral margin of rostrum posterior to apical teeth. Adrostral carina, sub- parallel and distinctly dorsal to ventral margin, extending almost to base of apical cluster Carapace with weak postrostral carina bearing three subequal, acutely produced teeth: 1) epigastric tooth, often larger than first rostral and nearly as large as posterior teeth, situated be- tween 0.12 and 0.25 (mean 0.17) cl from orbital Figure 13. — Skyonia disparri, holotype 9 10.4 mm cl, Bahia San Luis Gonzaga, Baja California Norte, Mexico. Lateral view. Scale = 5 mm. Rostrum long, reaching as far as distal margin of second antennular article, its length (0.40-0.70 cl) increasing linearly with carapace length (Fig. 14); relatively deep, with dorsal and ventral mar- gins subparallel; upturned to angle between 15° and 45°; armed with three dorsal teeth and two (7%), three (34%), four (57%), or five (2%) apical teeth, additional rudimentary one present be- tween two ventral ones; ventralmost tooth largest of apical cluster, subterminal, directed anteriorly or curved upward, and distinctly removed from adjacent tooth. Position of first dorsal tooth rang- ing from slightly posterior to level of orbital mar- gin, to as much as 0.30 (mean 0.18) rl anterior to it; that of second tooth between 0.40 and 0.65 (mean 0.50) rl from orbital margin, and that of third between 0.65 and 0.90 (mean 0.75) rl. Paired mi- nute, movable spinules (sometimes absent in 3 6 9 12 carapace length (mm) \5 Figure 14. — Sicyonia disparri. Relationship between rostrum length and carapace length (regression equation, y = -0.03809 + 0.51152X). 17 FISHERY BULLETIN: VOL. 83, NO. 1 margin, i.e., anterior, opposite, or posterior to hepatic spine; 2) middle tooth, largest of three, placed between 0.34 and 0.48 (mean 0.42) cl from orbital margin; and 3) posterior tooth, positioned well in advance of posterior margin of carapace, between 0.64 and 0.73 (mean 0.70) cl from orbital margin. Antennal spine absent, antennal angle broadly obtuse or rounded; hepatic spine long, sharply pointed, projecting from low buttress, and positioned between 0.20 and 0.30 (mean 0.25) cl posterior to orbital margin. Postocular sulcus short; hepatic sulcus shallow, subhorizontal; hepa- tic carina lacking; branchiocardiac carina at most barely evident. Eye (Fig. 15A) with ocular calathus broad and bearing conspicuous tuft of setae on dorsolateral extremity; ratio of length of lateral margin to width (across base of cornea) 0.50-0.60 (mean 0.54) and ratio length of lateral margin to carapace length 0.08-0.13 (mean 0.11). Antennular peduncle with stylocerite produced in acute spine reaching 0.65-0.85 of distance be- tween lateral base of first antennular article and mesial base of distolateral spine; latter extending to proximal 0.45 of second antennular article. An- tennular flagella short, mesial one about 0.8 as long as lateral; latter about 0.4 cl. Scaphocerite reaching between midlength and distal end of third antennular article; lateral rib produced in long, slender spine conspicuously overreaching lamella. Antennal flagellum as much as 3 times as long as carapace. Third maxilliped much stouter than pereopods. Basis and ischium of first pereopod each armed with well-developed sharp spine. Figure 15.— Eyes. A, Sicyonia disparri, ? 9 mm cl, NW of Mantanchen, Nayarit, Mexico. B,S. parri (Burkenroad), j 10.5 mm cl, east side of Cocoa Point, Barbuda. Dorsal views. Scale = 1 mm. Abdomen with dorsomedian carina low an- teriorly, increasing in height posteriorly; carina on first somite produced in rather small, an- teriorly directed tooth, usually smaller than pos- terior tooth on carapace; carina on second somite conspicuously incised, and that on sixth terminat- ing in short, sharp tooth. Anteroventral extremity of pleura of first three somites rounded; pleuron of fourth with postero- ventral margin straight to concave, its postero- ventral extremity distinctly angular, occasionally armed with minute tooth; posteroventral extrem- ity of fifth and sixth somites produced in very small, caudally directed, sharp tooth. First somite traversed by long, sometimes inter- rupted, deep anteromedian pleural sulcus ending well above ventral margin without meeting long, united posterior tergal-posteromedian pleural sulci. Second and third somites marked by 1) long anterior tergal sulcus and relatively short pos- terior tergal sulcus; 2) anteromedian pleural sul- cus, extending almost to ventral margin on second somite but restricted to dorsal part on third, in both somites delimiting anterior shallow depres- sion setting off rounded prominence dorsally; and 3) posteromedian pleural sulcus, its dorsal ex- tremity curving anteriorly, ventral to (not joining) posterior tergal sulcus. Fourth and fifth somites with anterior tergal sulcus (that of fourth usually obliterated at about midlength) fused with united posterior tergal-posteromedian pleural sulci. Sixth somite marked by strongly arched pos- teromedian pleural sulcus and bearing shallow, longitudinal depression between dorsolateral ridge and elongate, often ill-defined cicatrix. Telson with pair of short but well-developed fixed spines and two longitudinal rows of movable spinules on either side of densely setose median sulcus — mesial row extending almost to base of spine. Both rami of uropod almost reaching or barely overreaching apex of telson. Petasma (Figs. 5, 16A, B) with cornified distal projection of dorsolateral lobule bulbous prox- imodorsally, curved mesially, and minutely bifid distally; fleshy distal projection of ventrolateral lobule directed distolaterally, expanded basally and with slender but blunt terminal part slightly curved proximally. Lateral margin of petasma conspicuously notched just proximal to mid- length, forming shoulder immediately proximal to notch. Petasmal endopods coupled in males with carapace length as little as 3 mm (about 13 mm tl) but sometimes unjoined in individuals with 18 PEREZ FARFANTE: ROCK SHRIMP GE^^JS SICYONIA Figure 16. — Sicyonia disparri,6 6 mm cl, Canal de San Lorenzo, Baja California Sur, Mexico. A , Petasma, dorsal view; B , ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 0.5 mm. carapace length as much as 4.9 mm (about 21 mm tl). Appendix masculina as illustrated in Figure 16C. Thelycum (Fig. 17) with plate of sternite XIV, produced in conspicuous anterolateral lobules, flat except for deep, broad, median depression. Median plate of sternite XIII narrow, lanceolate, tapering gently into long, sharp spine reaching as far as proximal 0.25 of basis of extended second pereopods; plate constricted and deeply excavate at level of coxae of fourth pereopods; posterior component of plate with shallow, broad postero- median emargination and well-marked transverse suture. Paired conspicuous spines projecting an- teromesially from posterior margin of sternite XI. Posterior thoracic ridge with sharp, concave an- teromedian margin but areas immediately lateral to concavity flush with plate of sternite XIV. The smallest impregnated female encountered has a carapace length of 5 mm, about 21 mm tl. Maximum size. — Males 6.9 mm cl, about 30 mm tl; females 11 mm cl, about 44 mm tl. Figure 17. — Sicyonia disparri, holotype 5 10.4 mm cl, Bahia San Luis Gonzaga, Baja California Norte, Mexico. Thelycum. Scale = 1 mm. Geographic and bathymetric ranges. — In the Gulf of California, from Bahia San Luis Gonzaga (29°48'N, 114°22'W), Baja California Norte, and southward along the east coast of the peninsula, to Isla Santa Magdalena (24°55'N, 112°15'W), Islas Tres Marias, Nayarit, Mexico; also off Cabo San 19 FISHERY BULLETIN: VOL. 83, NO. 1 Lucas (22°52'23"N, 109°53'23"W), Baja California Sur (Fig. 12). This species occurs at depths between 0.2 and 82 m, mostly at <24 m, and on sandy bottoms: sand, sand and shell, sand and gravel, and a mixture of sand mud, and coral. Among the eastern Pacific rock shrimps, S. disparri appears to have one of the most restricted distributions, being virtually confined to the Gulf of California and waters off Nayarit. Its presence south of the Gulf of California is reported here for the first time. Discussion. — This shrimp is very similar to the geminate western Atlantic S. parri both in mor- phology and size. Burkenroad (1934a) first distin- guished S. disparri by the absence of posterior pleural sulci on the anterior two abdominal so- mites and by the shorter, deeper, and more up- turned rostrum. A few years later (1938), on the basis of three additional specimens, he pointed out other features in which S. disparri differed from his two specimens of S. parri: the shape of the posteroventral margin and extremity of the fourth abdominal somite, the presence of one or two pairs of movable spinules dorsal to the ventral margin of the rostrum near its anterior end, and the presence of four instead of three teeth on the dorsal margin of the rostrum. He stated that these characters are probably subject to variation and in so doing indi- cated that they might not be diagnostic. He noted, however, that the size and shape of the ocular calathus and the size and disposition of the cornea might prove to be diagnostic. My examination of a relatively large collection of S. disparri has demonstrated that among the various features that Burkenroad (1934a, 1938) suggested to distinguish this species from S. parri , three are diagnostic: 1) the disposition of the ros- trum, which is upturned between 15° and 45° in the former, is subhorizontal or inclined not more than 13° in the latter; 2) the shape of the postero- ventral extremity of the pleuron of the fourth ab- dominal somite, which is angular in S. disparri and rounded in S. parri; and 3) the shape of the ocular calathus and the breath and disposition of the cornea. In the Pacific shrimp the calathus is broader than in the Atlantic species, the lateral margin ranges from 0.50 to 0.60 (mean 0.54) its width at the base of the cornea and the latter is obliquely disposed. In S. parri the lateral margin of the calathus (Fig. 15B) varies from 0.80 to 0.91 (mean 0.85) its width at the base of the cornea, and the latter is almost horizontally disposed. I have confirmed that the ratio of the lateral margin of the calathus to the length of the carapace is usu- ally smaller in .S. disparri than in S. parri, rang- ing from 0.08 to 0.13 (mean 0.11) in the former and from 0.13 to 0.17 (mean 0.13) in the latter, but sometimes overlapping. The absence of posterior pleural sulci is a character that, although not infallible, serves al- most always to separate S. disparri from S. parri, lacking in all specimens of the former except in two small individuals, in one of which traces of them are present in the first three somites, and in the other, in the second somite. In contrast, all individuals of S. parri bear such sulci. As Burken- road anticipated, the shape of the posteroventral margin of the pleuron of the fourth somite, which is usually concave in S. disparri and convex in S. parri, is variable, sometimes straight in both shrimps. The length of the rostrum is not a reliable diag- nostic character, as previously suggested, only tending to be slightly longer in S. disparri than in S. parri — the ratio rl/cl ranges from 0.43 to 0.59 in the former and 0.36 to 0.55 in the latter. It does tend to be deeper, but not consistently, in the Pacific than in the Atlantic species. Burkenroad (1938) also pointed out the presence of a fourth tooth on the dorsal margin of the ros- trum in four of his specimens of S. disparri, but he considered this tooth to belong to the apical cluster (as I have in the meristic data presented here) when discussing differences in number of apical teeth between his smaller male and the remaining four shrimps. The number of apical teeth vary ii* both species; however, more tend to be present in S. disparri, 59% of the specimens possess more than three teeth (57% four, 2% five), whereas in S. parri 90% of them bear two or three (80% three, 10% two) and only 10% bear four teeth. In S. dis- parri the rostrum seems always to be armed with submarginal, movable spinules; their absence in a few adults is probably due to loss by accident. But among the specimens of S. parri I have examined, only one from south of Joao Pessoa, Paraiba, Brazil, bears a pair of such spinules. Another from Varadero, Cuba, possesses a single, very minute spinule located on the ventral margin of the ros- trum, near the base of the ventralmost apical tooth. The shape of the posteroventral margin of the pleuron of the fourth abdominal somite is vari- able, as Burkenroad predicted for S. disparri, sometimes straight in both species, but, as stated above, the posteroventral extremity is always an- gular in S. disparri and rounded in S. parri. In the 20 PfiREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA holotype of the former this extremity is sharply angular but unarmed, contrary to what was stated in the original description of the species; however, I have examined a few specimens in which the angle is produced in a small spine. Sicyonia disparri is also quite similar to S. laevigata but the epigastric tooth is larger than that in the latter, usually almost as large as the other two teeth on the postrostral carina, and is located closer to the level of the hepatic spine, i.e., farther from the orbital margin, 0.12-0.25 (mean 0.17) cl from it rather than 0.6-0.16 (mean 0.12) cl. The anteromedian pleural sulcus of the first ab- dominal somite is always well defined in S. dis- parri whereas it is absent or, infrequently, rudimentary in S. laevigata; and the posteroven- tral extremity of the fourth abdominal somite in S. disparri is angular rather than rounded. Fur- thermore, whereas the distal projection of the dor- solateral lobule of the petasma is conspicuously curved mesially in S. disparri, it is directed distally and curved dorsally at its tip in S. laevi- gata. The discussion of S. parri is based on 34 speci- mens, including the holotype (YPM 4395) and one male from the Bermudas, which represents the second record of the species from this area. Bur- kenroad (1938) reported it from the Bermudas, but his record has not been cited by subsequent au- thors, including me in my 1980 paper on the west- ern Atlantic Sicyonia. Material. — 62 specimens from 20 lots. Mexico — Baja California Norte: 9, holotype, YPM 4392, Bahia San Luis Gonzaga, 17 May 1926, Pawnee. 19, paratype, YPM, Bahia San Luis Gonzaga, 17 May 1926, Pawnee. 19, AHF, off Isla Partida, 82 m, 8 March 1936. Baja California Sur: Id 19, YPM, Bahia Santa Ines, 55 m, 11 April 1936, Zaca stn 142D-1. 66 49, AHF, 1.6 km WSW of Punta Perico, Isla Carmen, 13-20 m, 21 March 1949, Velero TV. 16 , AHF, Bahia Salinas, Isla Carmen, 13 m, 20 March 1940, Velero IV. 19 , SIO, Bahia Salinas, Isla Carmen, 0.3 m, 13 July 1965, B. W. Walker. 19, CAS, Bahia Balandra, Isla Carmen, 22 May 1921, F Baker 16, AHF, Bahia Agua Verde, 18 m, 12 February 1940, Velero III. 66 89 , SIO, NW of Isla Santa Cruz, 0-3 m, 10 July 1960, B. W. Walker. 26 29, SIO, Isla San Jose, 3-5 m, 29 March 1967. Id 29 , AHF, Canal de San Lorenzo, 11-24 m, 14 February 1940. 4d 99, USNM, off Isla del Espiritu Santo, 15 m, 30 April 1888, Albatross stn 2824. 29, SIO, off Punta Lobos, 18 m, 26 June 1961, R. Rosenblatt. Id 19 , USNM, Bahia La Ventana, 24-27 m, 20 April 1939, Stranger stn 38. 19, SIO, Ensenada de los Muer- tos, 9 m, 20 June 1961, R. Rosenblatt. Id , YPM, off Punta Arena, "Arena Bank," 64 m, Zaca stn 136D-30. Id , SIO, E of Cabo San Lucas, 0-6 m, 12 June 1961, R. Rosenblatt. Nayarit: 19, AHF, 1.6-3 km NW of Mantanchen, 21 December 1961. 19 , AHF, Isla Santa Magdalena, Islas Tres Marias, 5.5-9 m, 9 May 1939. Sicyonia mixta Burkenroad, 1946 Figures 12, 18-20 Sicyonia mixta Burkenroad, 1946:3, fig. 1-4 [holotype, d, NR (Stockholm) 2527; type- locality: "St. Joseph (probably San Jose, Lower California). Swedish Eugenie Expedition #818"]. Rodriguez de la Cruz, 1977:11. Diagnosis. — Antennal spine well developed, pro- jecting from short but strong buttress. First ab- dominal somite with dorsomedian carina uniquely lacking anterior tooth; second abdominal somite with anterior depression on dorsomedian carina limited posteriorly by subvertical wall. First pereopod with basis and ischium unarmed. Rostrum very short, not surpassing midlength of eye, bifid and bearing one dorsal tooth. Petasma with distal projection of dorsolateral lobule straight, but with tip curved dorsally; distal pro- jection of ventrolateral lobule bifid, arms curved inwardly. Thelycum with plate of sternite XIV convex laterally and with deep median depression. Description. — Body moderately robust (Fig. 18) and lacking tubercles. Carapace with long setae on dorsum, arc anterior to hepatic spine, and patch accompanying hepatic sulcus ventrally Rostrum very short, not surpassing midlength of eye, its length 0.13-0.16 cl; straight; armed with only one dorsal tooth situated almost at midlength of rostrum, and two minute apical teeth (tip bifid); ventral apical tooth located at same level or dis- tinctly anterior to dorsal apical tooth. Conspicu- ous adrostral carina, close and subparallel to ven- tral margin, extending to base of ventral apical tooth. Carapace with low postrostral carina ending markedly anterior to posterior margin of carapace and bearing three teeth: 1) epigastric tooth, small- est of three but larger than dorsal rostral tooth, placed well in advance of hepatic spine, between 0.05 and 0.08 cl from orbital margin; 2) middle tooth, as large as or slightly larger than posterior 21 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 18. — Sicyonia mixta Burkenroad, 1946, 6 12 mm cl, off Cabo San Lucas, Baja California Sur, Mexico. Lateral view. Scale = 5 mm. one, but about twice as high as anterior, situated posterior to hepatic spine, between 0.30 and 0.33 cl from orbital margin; and 3) posterior tooth, be- tween 0.60 and 0.63 cl. Tuft of setae present at anterior base of each tooth. Antennal spine well developed, projecting from short buttress; hepatic spine long, sharp, borne on prominent buttress arising from swollen hepatic region; branchiocar- diac carina barely distinct, only for short distance behind latter region. Postocular sulcus short and deep anteriorly, continuing posteriorly as well- defined groove; hepatic sulcus well marked, long, extending caudally to about level of apex of pos- terior tooth. Antennular peduncle with stylocerite produced in long, acute spine, its length almost or quite equal to distance between lateral base of first an- tennular article and mesial base of distolateral spine; latter sharp, long, reaching as far as distal margin of second antennular article; flagella rela- tively elongate, mesial one about 0.20 cl, shorter than lateral, latter about 0.30 cl. Scaphocerite reaching or surpassing antennular peduncle, sometimes by as much as 0.10 its own length; lateral rib produced distally in long, sharp spine overreaching lamella; antennal flagella in- complete in all specimens examined. 22 Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with dorsomedian carina extending from first through sixth somites, carina weak on first and second, increasing progressively in height through sixth; carina on first somite lack- ing anterior tooth; on second interrupted by well-marked depression limited posteriorly by abrupt elevation, situated at level of juncture of tergal sulci; on fifth descending gradually instead of truncate posteriorly; and on sixth ending in large, acute, posterior tooth. Pleuron of first abdominal somite with antero- ventral margin slightly concave, anteroventral ex- tremity rounded and unarmed; posteroventral ex- tremity of first four somites also rounded, that of fifth usually bearing minute spine, and that of sixth always produced in small spine. First somite with very short but deep anterome- dian pleural sulcus and relatively weak (similar to most remaining sulci) but long posterior tergal (extending 0.60-0.66 height of somite). Second and third somites bearing short anterior and quite long posterior tergal sulci. Fourth somite with long posterior tergal sulcus, but anterior tergal almost indistinct. Fifth somite with barely dis- tinct anterior tergal and relatively short posterior pfiREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA FIGURE 19. -Sicyonia mixta , 6 12 mm cl, off Cabo San Lucas, Baja California Sur, Mexico. A , Petasma, dorsal view; B , ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. tergal, and sixth bearing weak anterior tergal and arched posteromedian pleural sulci. Telson with paired fixed spines extremely mi- nute (in the only specimen examined by me in which they are present). Rami of uropod, subequal in length, falling slightly short of apex of telson. Petasma (Fig. 19A, B ) with rigid distal projec- tion of dorsolateral lobule bearing rounded prox- imodorsal prominence, extending almost straight distally but with tip curved dorsally Distal projec- tion of ventrolateral lobule bifid (arms curved), inclined laterally, and falling considerably short of adjacent projection. Appendix masculina as illustrated in Figure 19C. Thelycum (Fig. 20) with plate of sternite XIV, delimited by rounded lateral margins, raised in paired bulges sloping towards deep median de- pression. Median plate of sternite XIII roughly flaskshaped in outline, tapering into long, slender spine reaching proximal extremity of basis of an- teriorly extended second pereopods; posterior component of plate with arched posterior borders flanking shallow to deep, broad median emargina- tion, and traversed by strongly incised suture. Sternite XI armed posteriorly with paired, mod- erately long, acute spines. Posterior thoracic ridge with sharp, concave, anteromedian margin over- FIGURE 20. —Sicyonia mixta, 9 13 mm cl, off Isla T^boga, Golfo de Panama, Panama. Thelycum. Scale = 1 mm. 23 FISHERY BULLETIN: VOL. 83, NO. 1 lapping plate of sternite XIV but areas im- mediately lateral to it flush with preceding plate. Color. — The specimen examined by me from Peru exhibits a dark colored mask in the shape of a "2" (the base situated anteriorly), disposed hori- zontally from the posterolateral part of the carapace onto the anterodorsal part of the first abdominal pleuron. Maximum size. — Male 12.7 mmcl, about 43.5 mm tl; female 20 mm cl, 65.5 mm tl. Geographic and hathymetric ranges. — In addition to the undetermined type-locality, "St. Joseph" (most probably San Jose, Baja California), it has been found in Bahia Aimejas (24°29'18"N, 111°47'24"W) and off Cabo San Lucas, both on the ocean side of Baja California Sur, in Bahia de la Paz, on the eastern side of the latter, off Isla Taboga in the Golfo de Panama, and off Puerto de Eten (6°22'S, 80°47'W), Peru (Fig. 12). Sicyonia mixta occupies shallow water at depths between 11 and 24 m. The discovery of S. mixta first in the Golfo de Panama and more recently northwest of Puerto de Eten, Peru, was surprising and represented a con- siderable extension of the range of the species. F*reviously, it was known only from the waters of Baja California, where the three male types were collected and where, in Bahia de la Paz, five females and a male were found (Rodriguez de la Cruz 1977). The specimen from off Puerto de Eten (a female 20 mm cl, 26.5 mm tl ) was collected by M. Niquen from the RV Humboldt on 27 April 1983, and is deposited in the IMARPE. Discussion. — As Burkenroad (1946) stated, S. mixta is unique in lacking a tooth on the anterior end of the dorsomedian carina of the first abdomi- nal somite. Also this shrimp exhibits two of the basic characters of the species that Burkenroad (1934a) grouped in his Division II (the presence of a well -developed or clearly distinct antennal spine and the absence of basial and ischial spines on the first pereopod) at the same time that the dorsome- dian carina of the second abdominal somite, al- though not incised, is abruptly depressed an- teriorly. This depression seems to represent the deep incision or notch typical of the species of his Division I. Among the American species (excluding those grouped in Division I), S. mixta, S. disedwardsi, S. penicillata, and two western Atlantic species 24 — S. typica (Boeck 1864) and sometimes S. olgae Perez Farfante, 1980 — possess three dorsal teeth on the postrostral carina, two of which are situ- ated posterior to the hepatic spine. In S. mixta, however, the posterior tooth arises on the post- rostral carina considerably in advance of the pos- terior margin of the carapace, and the carina ends markedly anterior to the margin; in the other species the tooth arises nearer the margin where the carina ends. Sicyonia mixta also differs strik- ingly from the other four in the general sculp- ture of the abdomen; whereas in S. mixta it is not tuberculate and lacks all pleviral sulci except the anteromedian on the first somite and the postero- median on the sixth, in these congeners the ab- domen is strongly tuberculate, exhibits deeper sulci, and possesses pleural sulci and all sixth so- mites. The bifurcate shape of the terminal part of the distal projection of the ventrolateral lob- ule of the petasma is another character that dis- tinguishes S. mixta from the just mentioned rela- tives. The relationships of this species are rather puzzling. Except for the distinctly depressed dor- somedian carina of the second abdominal somite, it does not share any other characters of impor- tance with the species of Division I, represented by S. disparri, S. laevigata, and S. parri. Actually, it appears to be much closer to the group represented in the eastern Pacific by S. disedwardsi and S. penicillata, for in addition to possessing an anten- nal spine and lacking spines on the basis and is- chium of the first pereopod, like them, it is armed with two teeth on the postrostral carina posterior to the hepatic spine. It seems to me that S. mixta has had, although remotely, a common origin with the above-mentioned group. Remarks. — Because females of this shrimp have not been known previously, the above description of the thelycum is the first available for this species. In addition to the females cited from the Golfo de Panama and off northern Peru, two other new records are presented here: one represented by a male from off Cabo San Lucas, Baja Califor- nia Sur, and the other by a female from Bahia Aimejas, Gulf of California. These four specimens, the types, and the six reported by Rodriguez de la Cruz (1977) are the only ones that have been re- corded for this shrimp. The holotype of this species and the paratype, in the Naturhistorisches Museum (Vienna), are the only extant types of the Sicyonia treated here that were not examined by me. PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Material. — 5 specimens from 5 lots. Mexico— Id paratype, ZMB 6097, "California. Forrer" (?Ferrer). Baja California Sur: 16, SIO, Bahia Almejas, 11-24 m, 30 November 1961, F. H. Berry. IS , CAS, off Cabo San Lucas, 6 August 1932, Zaca stn D-21R. Panama— 19, USNM, off Isla Taboga, Golfo de Panama, 10 March 1969, H. A. Clarke and A. Rodaniche. Peru— 19, IMARPE, off Puerto de Eten, 27 April 1983, M. Niquen, Humboldt, stn 8304, haul 175. Sicyonia hrevirostris Stimpson 1871 Figures 21-25 Sicyona cristata. de Saussure 1857:306. [Not Hippolyte cristata De Haan 1844:194 = Sicyonia cristata (De Haan 1844).] Sicyonia hrevirostris Stimpson 1871:132 [syntypes not extant; type-locality: S. Florida coast]. Kingsley 1878:69. Faxon 1896:162. De Man 1911:10. Pesta 1915:118. Hay and Shore 1918:380, pi. 25, fig. 4. Hedgpeth 1953:160. Hildebrand 1954:268, 1955:220. Menzel 1956: 41. Lunz 1957:4. Anderson 1958:1, fig. 5. Eldred 1959:5. Gunter and McCaughn 1959: 1194. Anderson 1962:1, fig. 15. Kutkuhn 1962:2. Chapa Saldaha 1964:4. Joyce 1965: 132. Cerame-Vivas and Gray 1966:263. Joyce 1968:254, unnumbered fig. Rouse 1969:136. Bayer et al. 1970:41. Zyznar 1970:87. Brusher et al. 1972:75. Cain 1972:79. Franks etal. 1972:54. Allen 1973:1. Cobb et al. 1973:7, fig. 3, 4A-C. Day et al. 1973:36. Bryan and Cody 1975:1. Brusher and Ogren 1976:158. Hooks et al. 1976:103. Kennedy et al. 1977:1. Perez Farfante 1978: Sicyoniidae. Huff and Cobb 1979:51. Wenner and Boesch 1979:130. Holthuis 1980:58. Perez Farfante 1980:772. Soto 1980a:68, 1980b:84. Castille and Lawrence 1981:519. Wenner and Read 1981:4. Arre- guin Sanchez 1981:21. Perez Farfante 1982: 370. Wenner and Read 1982:181. Williams 1984:43. Eusicyonia edwardsi. Hay and Shore 1918, pi. 25, fig. 2. [Not Sicyonia edwardsii Miers 1881:367 = Sicyonia typica Boeck 1864.] Eusicyonia hrevirostris. Burkenroad 1934a: 84, 1934b:117, 1939:57. Lunz 1945:4. Anderson et al. 1949:16. Anderson 1956:2. McConnell 1960:52. Vernacular names: rock shrimp, hardback (United States); camaron de piedra, camaron conchiduro, camaron de roca (Mexico). FAO names (Holthuis, 1980): rock shrimp (English); camaron de piedra (Spanish); boucot ovetger- nade (French). Cobb et al. (1973) and Huff and Cobb (1979) presented extensive bibliographic references to this species, many of which are omitted from the above synonymy. Diagnosis. — Antennal spine well developed and projecting from strong buttress. Second abdominal somite with dorsomedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing three teeth posterior to level of hepatic spine, rarely anterior one of these at level of or slightly anterior to hepatic spine. Rostrum armed with two dorsal teeth (rarely three). Petasma with distal projection of dorsolat- eral lobule relatively short, its apical part curved dorsally. Thelycum with plate of sternite XIV al- most flat or slightly raised laterally in paired low bulges; posterior component traversed by well- marked suture. Branchiostegite lacking large spot or ocellus. Description. — Body robust (Fig. 21). Carapace with dorsum covered by densely set, short setae; also patches of setae present anterior and ventral to hepatic spine, subjacent to hepatic sulcus and to branchiocardiac carina, and band of smaller setae on ventral part of branchiostegite. Patches also present on terga of abdominal somites. Abdomen heavily or moderately tuberculate, tubercles small, more numerous on first four somites. Rostrum short, in young not overreaching distal end of second antennular article, in adults falling considerably short of distal margin of eye, its length increasing linearly with carapace length (Fig. 22), to about 16 mm cl, then growing little, rarely surpassing 6 mm (proportional length de- creasing with increasing size from as much as 0.49 to as little as 0.15 cl); subhorizontal or upturned as much as 85°, tapering from base to tip, and armed with two (very rarely three) dorsal and two or three (four in one) apical teeth; latter subequal or ventral one extending farther anteriorly than dor- sal and often upturned; when two teeth present, knob usually between them. Adrostral carina strong, bordering ventral margin and reaching end of rostrum. Carapace with postrostral carina high, bearing 25 FISHERY BULLETIN; VOL. 83, NO. 1 Figure 21. — Sicyonia brevirostris Stimpson,(5 18.1 mm cl, off Puerto Madero, Chiapas, Mexico. Lateral view. Scale = 5 mm. Figure 22. — Sicyonia brevirostris. Relationship be- tween rostrum length and carapace length (regression equation for specimens with carapace length less than about 16 mm, V = 0.52372 + 0.33342;t:; regression equa- tion for those larger, >> = 5.06145 + 0.01211x). T* c ^ 4 E 3 • • ■ • • • •• • • • • • • • • /:\- • %r*: ••. • • ; •. • • • • • / / • •• /? 16 24 32 carapace length (mm) 40 four teeth: 1) epigastric tooth only slightly larger than first rostral, situated from anterior to orbital margin to as much as 0.1 cl posterior to it; and 2) three large teeth usually placed posterior to level of hepatic spine, anterior one (level with hepatic spine in only three specimens examined) smallest, situated between 0.20 and 0.28 (mean 0.25) cl from 26 orbital margin, middle tooth between 0.52 and 0.60 (mean 0.55) cl, and posterior one between 0.74 and 0.79 (mean 0.75) cl. Antennal spine moder- ately long, projecting from strong buttress; hepa- tic spine, longer than antennal, acutely pointed and arising from moderately raised area between 0.20 and 0.25 (mean 0.23) cl from orbital margin. P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Postocular sulcus with anterior part deep, con- tinuing posteriorly as shallow groove; hepatic sul- cus well marked; hepatic carina weak; bran- chiocardiac carina conspicuous, extending to or almost to transverse ridge near posterior margin of carapace. Antennular peduncle with stylocerite produced in long, sharp spine, its length 0.85-0.90 distance between lateral base of first antennular article and mesial base of distolateral spine; latter extending as far as distal end of proximal 0.80 of second antennular article; antennular flagella short, me- sial antennular flagellum shorter than lateral, 0.15-0.20 as long as carapace, lateral flagellum 0.19-0.23 as long. Scaphocerite reaching or slightly overreaching distal margin of antennular peduncle; lateral rib ending distally in long, acute spine distinctly overreaching margin of lamella. Antennal flagel- lum about twice as long as carapace. Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with high dorsomedian carina ex- tending from first through sixth somites: on first produced in strong anterior tooth, on fourth usu- ally sloping posteriorly but sometimes abruptly truncate, and on fifth and sixth terminating in sharp tooth, latter longer. Anterolateral margin of pleuron of first abdom- inal somite concave, anteroventral extremity of first through fourth somites ending in spine, that of first directed anterodorsally, those of second through fourth curved posterolaterally; postero- ventral extremity of fourth through sixth somites bearing posteriorly directed, acute spine, that of fifth largest. Pleural spination of first four somites represented by tubercles in juveniles, becoming longer with increasing length of carapace. First abdominal somite with long anteromedian pleural sulcus ending dorsally at anterior margin of tergum and converging with long posterome- dian pleural sulcus ventrally; posterior tergal sul- cus long or short, well marked or weak; posterior pleural sulcus well defined. Second and third so- mites with long anterior and posterior tergal sulci; anteromedian pleural sulcus continuous with an- teroventral depression setting off prominence dor- sally and ridge posteriorly; posterior pleural sul- cus as on first somite. Fourth and fifth somites with anterior tergal joining curved, united pos- terior tergal-posteromedian pleural sulci dorsally; fourth also with shallow but clearly distinct caudodorsal depression, placed close to posterior margin. Sixth somite marked with faint anterior tergal and arched posteromedian pleural sulci, also bearing longitudinal ridge along base of dor- somedian carina and lateral depression (thickly covered with setae) dorsal to interrupted cicatrix. All sulci weak in juveniles, becoming deeper with age. Telson with pair of short, often abraded, fixed subterminal spines. Rami of uropod subequal in length, reaching or slightly overreaching apex of telson. Petasma (Fig. 23A, B) with rigid distal projec- tion of dorsolateral lobule raised proximodorsally in rounded prominence, strongly curved dorsome- sially, its distal part compressed, rounded ven- trally, produced dorsally in short, subacute salient; mesial base of projection bearing short, dentiform prominence. Distal projection of ventrolateral lobule with firm terminal part curved laterally, convex dorsally, plane ventrally, and tapering to ventrally directed apex. Petasmal endopods coupled in males as small as 6.3 mm cl, about 22 mm tl, but in individuals with as much as 13 mm cl, about 48 mm tl, they may not be joined. Appendix masculina as illustrated in Figure 23C. Thelycum (Fig. 24) with plate of sternite XIV produced in anterolateral lobules, almost flat or raised in paired low bulges sloping toward an- teriorly deep median depression. Median plate of sternite XIII sagittiform, tapering gradually into long, slender spine reaching as far as proximal 0.25 of basis of extended second pereopods; plate strongly excavate and bearing shallow, lateral in- cisions at level of fourth pereopods; posterior com- ponent with deep, broad posteromedian emargina- tion separating paired subtriangular processes limited anteriorly by well-marked transverse su- ture. Paired spines projecting anteroventral ly from posterior margin of sternite XI, spines broad basally, often spiculiform apically. Posterior thoracic ridge narrow, with concave anteromedian margin well marked. The smallest impregnated females encountered have a carapace length of 10 mm, about 44 mm tl. Color. — The coloration of this shrimp was de- scribed in detail from live specimens from the Gulf of Mexico by Burkenroad (1939). Williams (1965) presented notes on the color of materials of North Carolina, and Cobb et al. (1973) recorded observa- tions on individuals from the Yucatan shelf. In the latter, the dorsal part of the body is more reddish than in specimens from the northern Gulf of 27 FISHERY BULLETIN: VOL. 83, NO. 1 « Figure 23. — Sicyonia brevirostris , 18.1 mm cl, off Puerto Madero, Chiapas, Mexico. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scales = 1 mm. FIGURE 2A.— Sicyonia brevirostris, 9 26.3 mm cl, off Cape Lookout, North Carolina, USA. Thelycum. Scale = 2 mm. 28 P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Mexico, which Burkenroad described as having the dorsum brown and the sides white, and also differs from shrimp occurring in North Carolina, in which the ground color is off-white. Maximum size. — According to Holthuis (1980), 153 mm tl. Largest individuals examined by me: males 32 mm cl, about 125 mm tl; females 37 mm cl, about 130 mm tl. The only specimen from the Pacific, a male from off Puerto Madero, Chiapas, Mexico, measures 18.1 mm cl, 70 mm tl (latter from Burkenroad 1934a). Geographic and bathymetric ranges. — In the east- ern Pacific, off southern Mexico, from Colima (Chapa Saldaha 1964) to Chiapas (Fig. 25). In the western Atlantic, from Norfolk, Va., along the coast of the United States and the Bahamas to the southern coast of Cuba, and around the Gulf of Mexico from the Florida Keys to off Isla Contoy, Yucatan; perhaps also off Guyana (McConnell 1960). In the western Atlantic it occurs from shal- low water to 329 m (Williams 1965), usually be- tween 10 and 110 m, and rarely at depths >190 m. Exploitable populations are found between 34 and 55 m (Huff and Cobb 1979). This species prefers sand and shell-sand sub- strates, but occasionally it is found on mud bot- toms (Hildebrand 1954, 1955; Menzel 1956; Wil- liams 1965; Cobb et al. 1973; Kennedy et al. 1977; Huff and Cobb 1979). The depths and substrates with which it was associated in the eastern Pacific were not recorded. Discussion. — Sicyonia brevirostris is rather closely related to the western Atlantic S. typica and the eastern Pacific S. disedwardsi . It differs from both in possessing three, instead of two, large teeth on the postrostral carina posterior to the level of the hepatic spine and a caudodorsal de- pression on the fourth abdominal somite. It differs further from S. disedwardsi in 1) exhibiting well-marked posterior pleural sulci on the first three abdominal somites, which are lacking in the latter shrimp or, if present, weak, often distinct on only one or two somites; 2) having the distal pro- jection of the dorsolateral lobule of the petasma turned dorsomesially, compressed distally, and produced apically in a short, subacute, dorsally directed salient — in contrast, in S. disedwardsi the projection is turned mesially and then dor- solaterally, its apical extremity produced in a rela- tively elongate, acutely pointed, laterally directed salient which often bears a crest; and 3) lacking an ocellus on the posterior part of the branchioste- gite rather than bearing a large, brightly colored one such as that in S. disedwardsi. Sicyonia brevirostris occupies water of high sa- 20 10 100 Figure 25. — Geographic distribution of Sicyonia brevirostris. 29 FISHERY BULLETIN: VOL. 83, NO. 1 Unity; recorded off east Florida and off west cen- tral Florida in salinities between 32.00 and 36.75%o and 31.22 and 36.71%o (Kennedy et al. 1977 and Huff and Cobb 1979, respectively) and off Mis- sissippi between 27.8 and 34.6%.. (Franks et al. 1972). Only once, in the Chatham River, Fla., has it been recorded in an estuarine environment, at a salinity of 24%o (Rouse 1969). The specimens on which this record is based consisted of larvae and small juveniles, and might have been misiden- tified, perhaps belonging to one of the other con- geners found in that area. This shrimp, unlike other penaeoids, does not depend upon estuarine waters during its life cycle (Eldred 1959; Joyce 1965). Many investigators (Lunz 1957; Joyce 1965; Brusher et al. 1972; Cobb et al. 1973; Brusher and Ogren 1976; Camp et al. 1977; Kennedy et al. 1977; Huff and Cobb 1979; Wenner and Read 1981) note that this species is predominantly nocturnal. Cobb et al. (1973) suggested that it burrows into the substratum during the day, thereby avoiding pre- dation and capture by trawls. Notes on biology and abundance. — Whereas the other American rock shrimps have been largely neglected, because of its considerable economic value, large size, and ready availability S. bre- virostris has been the subject of a number of inves- tigations. Cobb et al. (1973) and Kennedy et al. (1977) studied the reproductive cycle (including ovarian development in detail) of west central and east Florida populations, respectively They con- cluded, as did Huff and Cobb (1979) who investi- gated the former population, that spawning and recruitment seem to occur throughout the year, with a peak of spawning from October to February off the west coast of Florida and during winter and early spring off the northeast coast. Cobb et al. (1973) suggested that a decrease in the daily photoperiod was responsible for the onset of spawning. Morphometric studies by Kennedy et al. (1977) demonstrated that increase in total length occurs at the same rate in males as in females until they reach 20 mm cl, then the rate of increase of total length in females become less. They also found that the juveniles grow at an average rate of 2-3 mm cl per month whereas the adults grow at 0.5- 0.6 mm cl. It was also estimated by them that the life span of this species is 20-22 mo. More re- cently Arreguin Sanchez (1981) presented biologi- cal fishery statistics (length/weight, growth, mor- tality, etc.) for this species. Density of this shrimp in various populations fluctuates seasonally. Wenner and Read (1981, 1982) found that S. brevirostris is the dominant species of decapod crustacean on the continental shelf between Cape Fear, N.C., and Cape Canav- eral, Fla., and that highest densities occurred in one summer of their 2y2-yr study. Lunz (1957) noted a bimodal seasonal abundance off South Carolina, with peaks occurring from September through December and again in May. Kennedy et al. (1977) observed that peak abundance is reached during the fall in the east Florida population. In the Gulf of Mexico, off central Florida (Cobb et al. 1973; Huff and Cobb 1979), maximum abundance was found to exist from late summer through the fall. Off Apalachicola, Fla., (Allen 1973) highest densities seem to occur from June to October; in coastal water of Louisiana (on the basis of maximum production) and Texas (Brusher et al. 1972), from June through January, and in Bahia de Campeche (Hildebrand 1955) this species was more abundant in February than in July. It thus appears that throughout the range of this species maximum abundance occurs from summer through fall, and, in some areas, into early winter. Kutkuhn (1962) calculated regression equations for predicting "headless" from "whole" weights and vice versa, and Cobb et al. (1973) presented equa- tions for relating carapace length to total length for males and females, as well as others for carapace length and total weight for males and females, and for both sexes combined. Huff and Cobb (1979) also calculated the relationships of carapace length to total length and carapace length to weight for each sex. Commercial importance . — Commercial fishing for S. brevirostris began in the United States in 1970 (Allen 1973) and since then production has in- creased noticeably, amounting to 3,351,000 lb, with a value of $3,222,000 in 1982 (Table 1). The fishing grounds are located off the southeast coast from North Carolina to central Florida (the most Table l. — Landings of Sicyonia brevirostris by areas and their values for 1982.' Area Pounds (heads-off) Dollars Georgia East Florida West Florida Texas Total 369.000 1.980.000 1.001.000 1,000 3,351.000 353,000 1 ,869,000 999.000 1,000 3,222.000 'Data provided by tlie Southeast Fisheries Center Statistical Survey Division, National Marine Fisheries Service, NCAA, Miami, Fla 30 PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA important ones by far being those off Cape Canav- eral and Fort Pierce), and in the Gulf of Mexico off the coast of Florida (those off Apalachicola and Pensacola the most productive of the entire region) and western Texas. The production of the Caroli- nas was insignificant in 1982, last year for which landings are available. This species is also exploited in Mexico off Isla Contoy, Quintana Roo, and in Bahia de Campeche. Arreguin Sanchez (1981) estimated that until 1978-79 the fishery off Isla Contoy, under optimal conditions, could have produced as much as 450 t (992,070 lb) of heads-on shrimp. Remarks. — This species was first recorded from the eastern Pacific by Burkenroad (1934a). His record was based on a single specimen reportedly taken by the Pawnee off southern Mexico, in the Gulf of Tehuantepec, at lat. 14°40'20"N, long. 92°40'30"W. Later, Chapa Saldaha (1964) recorded five lots of this shrimp from the coast of Oaxaca, Golfo de Tehuantepec, which were in the collec- tions of the Institute Nacional de Investigaciones Biologico-Pesqueras, Mexico. Unfortunately, these specimens are no longer extant (Concepcion Ro- driguez de la Cruz^). Presence of S. brevirostris has not been confirmed by subsequent investigations in the region, including that of Sosa Hernandez et al. (1980), who conducted a survey of the decapod crustaceans in the general area from which Bur- kenroad's and Chapa Saldaha's materials were ob- tained. Furthermore, representatives of the species were not found in the large collections of penaeoids from the Pacific coast of Mexico and Central America examined by me. Because of the large size of this shrimp (it may reach 153 mm), it should have been retained by the commercial shrimp trawls or other gear used off southern Mexico. The surprising fact that it has not been recorded from commercial catches since the ap- pearance of Chapa Saldana's report nor from col- lections resulting from exploratory work raises the possibility that the specimens cited above either bear incorrect data (as might be true of the correctly identified specimen examined by Bur- kenroad) or were misidentified. McConnell (1960) reported this species from Guyana, but because it has not been recorded in studies (including my own) made of large collec- tions of penaeoids obtained by research vessels and "Concepcion Rodriguez de la Cruz, Institute National de Pesca, Secretaria de Pesca, Mexico, D.F., Mexico, pers. commun.. May 1983. shrimp trawlers in the waters of that country or in any others south of Cuba, its presence there needs confirmation. This is the only species for which full biblio- graphic references are not given. Because of its abundance, accessibility, and economic value, the literature on this shrimp is extensive. As stated above, much of it is cited in the works of Cobb et al. (1973) and Huff and Cobb (1979), consequently, I am including those references in which synonyms were created, articles not cited by them, others which appeared subsequent to their contributions, and all of those cited in the treatment below. Material. — 281 specimens from 66 lots. Eastern Pacific — 1 specimen. Mexico — Chiapas: Id, YPM, off Puerto Madero, 9 April 1926, Pawnee. Western Atlantic — 280 specimens from 64 lots. United States — North Carolina: 45 29, USNM, off Rodanthe, 49 m, 20 October 1884, Al- batross stn 2296. 14(? 129 , USNM, off Cape Hat- teras, 64 m, 21 June 1957, Combat stn 396. 19, USNM, NE of Cape Hatteras, 55 m, 26 July 1969, Oregon II stn 10697. 36 39, USNM, off Raleigh Bay, 26 m, 30 July 1969, Oregon II stn 10738. 26 69, USNM, off Raleigh Bay, 33 m, 27 May 1962, Silver Bay stn 4028. Id, USNM, SE of Cape Lookout, 37 m, 12 March 1961, Silver Bay stn 2913. 39, USNM, off Cape Lookout, 43 m, 21 June 1957, Combat stn 397. Id 39, USNM, Onslow Bay, 46 m, 2 August 1962, Silver Bay stn 4196. South Carolina: 2d , USNM, off Port Royal Sound, 51-44 m, 25 June 1957, Combat stn 428. Id 19, USNM, off Hilton Head Island, 40-46 m, 7 October 1957, Combat stn 514. 39, USNM, off Hilton Head Island, 64 m, 14 December 1961, Silver Bay stn 3657. Georgia: 3d, USNM, off Cape Romain, 1941, J. Oney 19, USNM, off Sapelo Island, 42 m, January 1940, Pelican. 49, USNM, NE of Savannah Beach, 40 m, 4 February 1940, Pelican . 6d 99 , USNM, off Jekyll Island, 73 m, 15 March 1940, Pelican. Florida: Id 29, USNM, NE of Fernandina, 31 m, 2 October 1951, Combat stn 505. Id 29, USNM, off Fernandina, 42 m, 10 March 1976, George M. Bowers stn 37. 3d 39, USNM, off Ponte Vedra Beach, 24 m, 23 April 1956, Pelican stn 32. 4d 19, USNM, off St. Augustine, 329 m, 16 September 1956, Combat stn 119. Id , USNM, off Matanzas Inlet, 183 m, 18 November 1965, Oregon stn 5741. 13d 229, 31 FISHERY BULLETIN: VOL. 83, NO. 1 USNM, off Flagler Beach, 40 m, 7 November 1963, Silver Bay stn 5201. 25 89, USNM, off Edge- water, 22 m, 1 December 1961, Silver Bay stn 3588. 16, USNM, off Cape Canaveral, 25 January 1962, Silver Bay stn 3704. 19, USNM, off Cape Canaveral, 70 m, 16 January 1966, Ore- gon stn 5860. 36, USNM, off Melbourne Beach, 40 m, 23 March 1956, Pelican stn 14. 56 59, USNM, off Hutchinsons Island, 63 m, 11 November 1963, Silver Bay stn 5267. 26 69, USNM, NE of St Lucie Inlet, 38-42 m, 21 May 1968, Gerda stn 1002. 26 29, USNM, off Key Largo, 110-113 m, 26 January 1966, Gerda stn 769. 16, USNM, S of Islamorada, 49-40 m, 15 September 1965, Gerda stn 756. 26 59 , USNM, S of Pine Island, 170-194 m, 25 February 1969, Gerda stn 1023. 16 29, USNM, S of Marathon, 57-69 m, 26 February 1969, Gerda stn 1030. 16 19, USNM, NW of Marquesas Keys, 27 m, 28 May 1952, Oregon stn 562. 16 39, USNM, SE of Dry Tortugas Islands, 59 m, 13 April 1965, Gerda stn 574. 16 89 , USNM, S of Dry Tortugas Islands, 68 m, 13 April 1954, Oregon stn 1004. 26 69 , USNM, S of Dry Tortugas Islands, 64 m, 12 April 1965, Gerda stn 566. 16 , USNM, SW of Dry Tortugas Islands, 91 m, 8 March 1970, Gerda stn 1241. Id 49, USNM, NW of Dry Tortugas Islands, 55 m, 18 June 1956, Oregon stn 1553. Id, W of Dry Tor- tugas Islands, 37 m, 7 March 1970, Gerda stn 1235. 16, USNM, off Appalachee Bay, 27 m, 9 March 1954, Oregon stn 905. 19, USNM, off St George Island, 37-35 m, 26 July 1957, Silver Bay stn 88. 29 , USNM, off St George Island, 68 m, 10 March 1954, Oregon stn 916. 19, USNM, S of Apalachicola Bay, 64 m, 7 March 1954, Oregon stn 896. 16, USNM, off Grayton Beach, 43 m, 16 November 1952, Oregon stn 707. Alabama: 16, USNM, off Orange Beach, 37 m, 24 January 1957, Oregon stn 1651. 56 49, USNM, off Dauphin Island, 70 m, 9 August 1950, Oregon stn 82. Louisiana: 16 , USNM, off Beeton Sound, 40 m, 21 August 1962, Oregon stn 3713. 19, USNM, S of Mississippi Delta, 84 m, 12 September 1950, Oregon stn 101. 29, USNM, off Southwest Pass, Mississippi Delta, 60 m, 14 September 1980, Oregon stn 110. 16 19, USNM, off Ship Shoal Lighthouse, 37-40 m, 12 September 1962, Oregon stn 3186. 19 , USNM, off Pelican Island, 22 m, 8 March 1957, Oregon stn 1755. Texas: 66 62, USNM, SE of Galveston, 20 m, 6 May 1938, Peli- can. 59, USNM, S of Galveston, 18 m, 5 May 1938, Pelican. 69 , SW of Galveston, 16 m, 5 May 1938, Pelican. 19 , USNM, NE of Brownsville, 26 m, 5 February 1939, Pelican. 32 Mexico— Tabasco: 16 19, USNM, off Paraiso, 35 m, 16 May 1954, Oregon stn 1057. 19 , USNM, off Laguna Machona, 64 m, 16 May 1954, Oregon stn 1060. 19 , USNM, NW of Punta Frontera, 60 m, 8 June 1970, Oregon II stn 10981. 36 39, USNM, NW of Punta Frontera, 66 m, 9 June 1970, Oregon II stn 10982. Campeche: 26 , USNM, N of Arrecifes Triangulos, 64 m, 17 August 1951, Oregon stn 411. 16 , USNM, W of Cayos Areas, 66 m, 16 June 1970, Oregon II stn 11005. 16 , USNM, E of Cayos Areas, 37 m, 11 December 1952, Oregon stn 720. 19, USNM, NE of Cayos Areas, R. Ramirez and M. Flores, 48 m, 30 April 1959. 36 19, USNM, NE of Cayos Areas, 42 m, 10 December 1952, Oregon stn 719. Id 29, USNM, 16 km, NE of Ciudad del Carmen, R. Ramirez and M. Flores, 29-37 m, 29 April 1959. 19, USNM, off Cam- peche, 13 m, 2 May 1959, R. Ramirez and M. Flores. Sicyonia disedwardsi (Burkenroad 1934) Figures 3, 7, 26-30 Eusicyonia disedwardsi Burkenroad 1934a:86, fig. 23, 29, 34 [holotype: d, YPM 4394; type-local- ity: Bahia Concepcion, Baja California Sur, 3 May 1926, Pawnee]. Burkenroad 1938:82. Anderson and Lindner 1945:317. Castro 1966:17. Sicyonia disedwardsi. Brusca 1973:219. Rosales Juarez 1976:41. Rodriguez de la Cruz 1977:11. Anonymous 1980:6. Brusca 1980: 256. Perez Farfante 1982:371. Vernacular names: rock shrimp, target shrimp, Japanese shrimp (United States); cacahuete, camaron de piedra, camaron de roca, camaron japones (Mexico); camaron conchiduro (Mexico, Panama). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing two teeth posterior to level of hepatic spine. Rostrum armed with two dorsal teeth. Petasma with distal projection of dorso- lateral lobule short, stout, curved dorsomesially then laterally. Thelycum with plate of sternite XIV raised in relatively low, sometimes indistinct bulges; posterior component traversed by faint su- ture. Branchiostegite with large ocellus consisting of well-defined yellow center surrounded by purplish brown ring. P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Figure 26.— Sicyonia disedwardsi (Burkenroad 1934),? 34 mm cl, Golfo de Panama, Panama. Lateral view. Scale = 10 mm. Description. — Body robust (Fig. 26). Carapace studded with long setae anterior to hepatic spine and in pterygostomian region; patches of densely set shorter setae present on dorsum, in depression anterior to posterodorsal part of branchiocardiac carina, on branchiostegite, and subjacent to hepat- ic sulcus; patches also on tergum of abdominal somites and in depression just ventral to dorsal ridge of sixth somite. Abdomen bearing numerous tubercles on all somites (especially first three), except few or none on sixth. Rostrum short, reaching distal margin of eye at most, its length increasing linearly with carapace length (Fig. 27) to about 20 mm cl, then growing little, not surpassing 6 mm (proportional length decreasing with increasing size from as much as 0.43 to as little as 0.13 cl); subhorizontal or up- turned to as much as 45° (in young), tapering con- siderably from base to truncate, sometimes de- curved tip; armed with two dorsal and two (96%) or three (4%) apical teeth. Adrostral carina, situated close to ventral margin, extending to end of ros- trum. Carapace with well-marked postrostral carina bearing three teeth: 1) epigastric small, only slightly larger than first rostral, situated from slightly anterior to orbital margin to 0.07 cl be- hind it; 2) middle tooth, larger than epigastric, placed well posterior to hepatic spine, between 0.29 and 0.38 (mean 0.33) cl from orbital margin; and 3) posterior tooth, larger than middle one, acutely pointed (usually abraded in larger adults), rising almost vertically before turning anteriorly or, more often, strongly inclined anteriorly, and situated between 0.68 and 0.80 (mean 0.72) cl from orbital margin; tuft of setae present at anterior base of all three teeth. Postrostral carina increas- 7 14 21 carapace length (mm) Figure 27. — Sicyonia disedwardsi. Relationship between ros- trum length and carapace length (regression equation for speci- mens with about 20 mm cl or less, y = 0.74318 + 0.2174aac; regression equation for those larger, y = 3.81074 + 0.04939x). 33 FISHERY BULLETIN; VOL. 83, NO. 1 ing in height from low anterior part (between epigastric and middle teeth) to posterior tooth, descending gradually from it to posterior margin of carapace. Antennal spine relatively long, sharp, and projecting from short buttress; hepatic spine acutely pointed, arising from moderately raised area, and placed between 0.19 and 0.26 (mean 0.22) cl from orbital margin. Postocular sulcus short but deep, continuing posteriorly as barely distinct narrow groove; hepatic sulcus shallow; bran- chiocardiac carina broad, long, extending lon- gitudinally from hepatic region almost to posterior margin of carapace where bifurcate: one branch curving dorsally and other, short, disposed ventrally. Eye as illustrated in Figure 3. Ocular stylet with terminal part often bent laterally in young, but usually straight in larger individuals. Antennular peduncle with stylocerite produced in long spine, its length 0.80 to 0.90 distance be- tween lateral base of first antennular article and mesial base of distolateral spine; latter extending as far as distal 0.3 of second antennular article; antennular flagella short, maximum length 0.25- 0.35 cl, mesial flagellum slightly shorter than lat- eral in young but subequal to or slightly longer in larger adults. Scaphocerite nearly or quite overreaching an- tennular peduncle by as much as 0.15 its own length; lateral rib ending distally in long, acute spine conspicuously surpassing margin of lamella. Antennal flagellum as much as 2.2 times as long as carapace. Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with high dorsomedian carina ex- tending from first through sixth somites, carina on first produced in strong tooth, sometimes disposed almost horizontally, but usually elevated as much as 55°, tooth tapering to sharp apex, and consider- ably larger than posterior one on carapace; carina on fifth somite produced in small tooth and that on sixth terminating in strong sharp one. Anterior margin of pleuron of first abdominal somite straight or very slightly concave; antero- ventral extremity of pleuron of first through fourth somites ending in spine, that of first usually directed anteroventrally, spines on second through fourth often curved posterolaterally; posteroven- tral margin of first through third somites rounded, that of fourth broadly angular, sometimes bearing node or minute spine, and that of fifth and sixth sharply angular and armed with small, caudally directed spine. Pleural spination of first four so- mites barely if at all distinct in juveniles, becom- ing stronger with increasing length of carapace. First somite traversed by deep, long anterome- dian pleural sulcus (sometimes obsolete along midlength), latter usually converging with united posterior tergal-posteromedian pleural sulci ven- FlGURE 28. — Sicyonia disedwardsi , S 23.5 mm cl, Golfo de Panama, Panama. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. 34 P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA trally, but its dorsal extremity often not reaching anterior margin. Second and third somites with anterior and often short posterior tergal sulci; an- teromedian pleural sulcus frequently joining posteromedian pleural sulcus (sometimes obso- lete) dorsally, and continuous with anteroventral depression setting off prominence dorsally and ridge posteriorly. Traces of posterior pleural sulcus occasionally present in one or more of first three somites. Fourth and fifth somites with anterior tergal joining curved, united posterior tergal- posteromedian pleural sulci. Sixth somite with shallow, arched posteromedian pleural sulcus; longitudinal ridge along base of dorsomedian carina delimited ventrally by weak depression lying dorsal to ill-defined cicatrix. Telson with pair of small but clearly distinct fixed spines. Rami of uropod subequal in length and falling slightly short of or overreaching apex of telson by as much as 0.15 its own length. Petasma (Fig. 28A, B) with rigid distal projec- tion of dorsolateral lobule mesially inclined (in- stead of erect), strongly curved dorsomesially then laterally; distal part of projection slightly com- pressed, often bearing crest, rounded ventrally and produced laterally in elongate, acutely pointed salient. Distal projection of ventrolateral lobule with terminal part strongly curved later- ally, bulbous dorsally, plane ventrally, with pointed tip curved proximoventrally. Petasmal endopods coupled in males with carapace length as little as 5.5 mm (about 22.5 mm tl), but sometimes unjoined in individuals with carapace length as much as 10.5 mm (about 38.5 mm tl). These observations are similar to those noted by Burkenroad (1938). Appendix masculina as illustrated in Figure 28C. Thelycum (Fig. 29) with plate of sternite XIV raised in paired, low (sometimes indistinct), rela- tively short bulges, sloping toward deep, broad, median depression. Median plate of sternite XIII sagittiform, tapering gradually into long, slender spine reaching as far as proximal 0.25 of basis of extended second pereopods; plate deeply excavate and bearing arched lateral incisions at level of fourth pereopods; posterior component with deep, broad posteromedian emargination forming rela- tively elongate, posterolateral processes marked basally by rather faint transverse suture. Sternite XI armed posteriorly with pair of small, broad based spines. Posterior thoracic ridge either al- most flush with plate of sternite XIV or with only anteromedian margin raised (ventrally). Sperm receptacles as illustrated in Figure 7 (il- lustration based on specimen treated following method by Monod and Cals (1970)). The smallest impregnated female encountered has a carapace of 12 mm, about 42 mm tl. Color. — The only observation on color of fresh material from the Gulf of California (Anonymous 1980) indicates that the typical large spot found in the posterior part of the branchiostegite is purplish brown with a yellow center. The con- spicuous ocellate spot persists in preserved speci- mens, appearing like a broad dark ring surround- ing a light center. Maximum size. — Male, 29.2 mm cl, about 102 mm tl; female 34 mm cl, about 108 mm tl. Geographic and bathy metric ranges. — Southwest of Isla Santa Margarita (24°19'36"N, 111°46'24"W - 24°19'48"N, 111°47'06"W) to Bahia San Lucas, Figure 29. — Skyonia disedwardsi , 9 21.3 mm cl, off Punta Gorda, Baja California Sur, Mexico. Thelycum. Scale = 2 mm. 35 FISHERY BULLETIN: VOL. 83, NO. 1 Baja California Sur, in the Gulf of California along the central and southern parts of both coasts and southward to Bahia Chamela (19°33'42"N, 105°07'24"W), Jalisco, Mexico. Also from Bahia de Culebra (10°37'00"N, 85°40'00"W), Costa Rica, to northwest of Punta Caracoles (7°45'00"N, 78°24'30"W), Darien, Panama (Fig. 30). It has been found at depths between at least 18 (5-18) and 249 m, but seems to be most abundant at 30-60 m. It occurs on bottoms of shell, mud, fine sand, and rocks. The occurrence of this shrimp along the west coast of Baja California Sur has not been previ- ously reported. Discussion. — Sicyonia disedwardsi is most simi- FlGURE 30. — Geographic distribution of Sicyonia disedwardsi and S. penicillata. lar to the western Atlantic S. typica. As stated above, among the American members of the genus lacking an incision or abrupt depression on the middorsal carina of the second abdominal somite, these two, together with S. penicillata and the western Atlantic S. olgae bear two relatively large teeth on the postrostral carina posterior to the level of the hepatic spine. The genitalia of S. dis- edwardsi and S. typica are so similar that they are almost indistinguishable, but in the petasma of S. typica the tip of the projection of the dorsolateral lobule is not so strongly produced and is usually directed dorsally rather than dorsolaterally. These two species, however, differ in the number and extension of abdominal sulci: in S. typica the first three somites exhibit well-marked posterior pleural sulci, which are absent or weak in S. dis- edwardsi , and the dorsal extremity of the united posterior tergal-posteromedian pleural sulci of the first somite reaches the anterior margin of the somite, whereas in S. disedwardsi the dorsal end usually does not reach the margin. Also in S. typica, the anteromedian pleural sulcus of the first somite is unbroken and that of the second joins the posteromedian, whereas in S. disedwardsi the former is often interrupted and the median sulci do not merge; in addition, the posterior tergal sulci of the second and third somites are much longer than in S. disedwardsi, in which they terminate considerably dorsal to the base of the respective pleuron. Furthermore, the rostrum of S. dised- wardsi usually bears two teeth on the dorsal mar- gin posterior to the apical teeth, whereas in S. typica it is often armed with only one. Sicyonia disedwardsi differs from S. penicillata by the same features of the abdomen that distin- guish it from S. typica, except that in S. penicillata the posterior pleural sulci are more frequently present and slightly better marked than in S. dis- edwardsi. Too, the rostrum of S. penicillata, like that of S. typica, commonly bears only one dorsal tooth and is less elevated and usually shorter than in individuals of same size of S. disedwardsi. The two partly sympatric species can be separated readily by the genitalia. The unusually long distal projections of the dorsolateral and ventrolateral lobules in S. penicillata are not exhibited by any other of its congeners. Also, whereas in S. dised- wardsi the thelycal plate of sternite XIV bears paired low (sometimes indistinct) bulges, in S. penicillata it is raised in strongly marked and more striking ones; the posterior component of the median plate of sternite XIII in S. disedwardsi exhibits a broad and deep posteromedian emar- 36 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA gination flanked by elongate processes and is traversed by a faint suture; in contrast, that of S. penicillata possesses a shallow emargination, often bearing a small anteromedian notch, and is traversed by a deep groove. Long after they have been taken from the water, even after years in alcohol, specimens of this species may be readily recognized by a large ocel- lus, consisting of a well-defined yellow center sur- rounded by a broad purplish ring, on the bran- chiostegite. In S. penicillata, as in S. typica, the branchiostegite does not bear a large marking. Burkenroad (1938) indicated that all members of S. disedwardsi may be distinguished from those of S. penicillata by the shape of the ocular stylets which in the former, according to him, diverge at the tip; they are straight in the latter. I have found, however, that although the stylets are laterally inclined distally in most of the young of S. dised- wardsi, they are straight in some young and in many adults. Commercial importance. — Sicyonia disedwardsi, one of the stubbiest of the rock shrimp occurring in the region, is usually taken with other penaeoids by the shrimp fleets operating in various areas from the Gulf of California to the Golfo de Panama. Because of its size, it appears to have the potential of providing a fishery as has the simi- larly heavy S. brevirostris in the western Atlantic. Material. — 242 specimens from 54 lots. Mexico— Baja California Sur: 66 69, SIO, 3 km SW of Isla Santa Margarita, 48-57 m, 13 November 1964, Black Douglas . 5d 69 , SIO, SW of Isla Santa Margarita, 29-40 m, 13 November 1964, Black Douglas. 26 19, SIO, NW of Punta Marquez, 37 m, 4 December 1962, H. Perkins and R. Wisner. 126 119, SIO, NW of Todos Santos, 38 m, 9 November 1964, Black Douglas. 36 19, YPM, Bahia San Lucas, 15-29 m, 6 May 1936, Zaca stn 135 D-1. 19 , YPM, Bahia San Lucas, 11-37 m, 6 May 1936, Zaca stn 135 D-9. 26 29, YPM, off Punta Arena, 92 m, 30 April 1936, Zaca stn 136 D-24. 19, YPM, off Punta Arena, 82 m, 3 April 1936, Zaca stn 136 D-1. 4d 29, YPM, off Punta Arena, 64 m, 1 May 1936, Zaca stn 136 D-30. 36 19, USNM, Bahia La Ventana, 24-27 m, 20 April 1939, Strange stn 38. 2619, SIO, off Punta Gorda, 81-84 m, 2 July 1965, C. Hubbs. 46 39, USNM, Canal de San Lorenzo to Isla del Espiritu Santo, 4 April 1960, R. Mercado and G. Pre- ciado. 26 29, SIO, Bahia de la Paz, 55-79 m, 6 July 1965, R. Rosenblatt. 36 99, SIO, Canal de San Jose, 64 m, 8 July 1965, R. Rosenblatt. 46 109, SIO, Punta San Telmo, 10 July 1965, W Baldwin. 86 69, SIO, off W of Isla Monserrate, 92-73 m, 12 July 1965, R. Rosenblatt. 26 , SIO, Bahia Concepcion, 4 February 1940, D. Rouch. Id , holotype, YPM, Bahia Concepcion, 3 May 1926, Pawnee. 26 29, USNM, 4-6 m, off Punta Concepcion, 12 April 1964, F. Rosales Juarez. 19, YPM, Bahia Santa Ines, 50 m, 13 April 1936, Zaca stn 143 D-1. 26 19 , YPM, Bahia Santa Ines, 37 m, 10 April 1936, Zaca stn 141 D-4. 19, YPM, Bahia Santa Ines, 13-16 m, 10 April 1936, Zaca stn 141 D-1. 26 39 , YPM, Bahia Santa Ines, 50 m, 11 April 1936, Zaca stn 142 D-1. 86, SIO, off Santa Rosalia, 35-26 m, 25 March 1960, R. Parker. 16 , SIO, S arm of Bahia de los Angeles, 22-37 m, 26 April 1962, R. Rosenblatt. 26 39, AHF, Puerto Refugio, Isla Angel de la Guarda, 38 m, 27 January 1940. Sonora: 16 19, USNM, off Estero de Lobos, 47 m, 3 April 1978, Toral Garcia. 19, USNM, 8 km off Guaymas, 26 m, April 1980, M. Hatziolos. 3d 59, INP, off Punta Rosa, 56 m, 2 April 1978, Toral Garcia. 66 49, USNM, SE of Punta Rosa, 54 m, 1 April 1978, Toral Garcia. Sinaloa: Id, USNM, off San Ignacio, 25 May 1962, R. Bush M. 2d, SIO, Isla de Altamura, 22-31 m, 26 May 1965, El Golfo II stn 50-6. 19, AHF, off Rio San Lorenzo, 11-24 m, 14 February 1940. Nayarit: Id 59, AHF, off Isla Isabela, 27-46 m, 9 May 1939. Jalisco: Id, SIO, Bahia Cha- mela, 27-18 m, 2 April 1973, Agassiz. Colima: 19, CAS, off Manzanillo, 17 July 1932, Zaca. Costa Rica— Id, AHF, Bahia de Culebra, 5-18 m, 24 February 1934. 29, AHF, S of Bahia de Culebra, 18 m, 25 February 1934. 19, SIO, Cabo Blanco, 60 m, 18 April 1973, Agassiz. Id, SIO, Cabo Blanco, 137-145 m, 19 April 1973, C. Hubbs and S. Luke. 19 , UCR, near Cabo Blanco, 245 m, 28 April 1973, Enriqueta. Id, UCR, near Cabo Blanco, 249 m, 28 April 1973, Enriqueta. Id, UCR, near Puerto Quedos, 242 m, 26 April 1973, Enriqueta . Id , AHF, 5 km off Isla Manuelita, 146 m, 3 June 1973, Velero IV. 19 , AHF, Golfo Dulce, 35-88 m, 26 March 1939, Velero IV. Panama— 2d 29, USNM, Golfo de Panama, Canopus stn 670. 5d 119, USNM, Golfo de Panama, Canopus stn 126. 2d 29, UP, Ar- chipielago de las Perlas, 11 December 1970, J. M. del Rosario. Id 19, USNM, Isla San Jose, 64 m, 23 February 1973, fishermen. 3d 19 , USNM, S of Isla del Rey 44-42 m, 7 May 1967, Pillsbury stn 551. 19, AHF, Islas Secas, 46-48 m, 27 March 1939. 19, AHF, off Bahia Honda, 55-64 m, 1 37 FISHERY BULLETIN: VOL. 83, NO. 1 March 1938. M 49, UP, between Punta Ave Maria and Ensenada Guayabo, 14 December 1969, staff Dep. Biol. Mar, UP 3d 19, USNM, 12 km NW of Punta Caracoles, staff Dep. Biol. Mar., UP Sicyonia pentcillata Lockington 1879 Figures 30-34 Sicyonia penicillata Lockington 1879:164 [syn- types (not extant): "Bolinas Bay (?Bahia de Ballenas), Lower California", 14 fm (fathom) (25.6 m); Angeles Bay (Bahia de los Angeles), Gulf of California, W. J. Fisher]. De Man 1911:112. Pesta 1915:118, fig. 7. Schmitt 1924:387. Brusca 1973:219. Rosales Juarez 1976:41. Rodriguez de la Cruz 1977:10. Anonymous 1980:7. Brusca 1980:256. Rodriguez de la Cruz 1981:1. Mathews 1981: 329. Eusicyonia penicillata. Boone 1930:115 [part], pi. 36. Burkenroad 1934a:88, figs. 30, 31, 33, 1938:93. Steinbeck and Ricketts 1941:444. Castro 1966:17 [part]. Word and Charwat 1976:22, 2 figs. Eusicyonia sp. Castro 1966:16, 17 [part], fig. 4. Vernacular names: rock shrimp, target shrimp, Japanese shrimp (United States); cacahuete, camaron de piedra, camaron de roca, camaron japones (Mexico). FAO names: peanut rock shrimp (English), camaron cacahuete (Spanish), boucot cacahouette (French). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing two teeth posterior to level of he- patic spine. Rostrum armed with two dorsal teeth (rarely one). Petasma with distal projection of dor- solateral lobule acicular, long and slender. Thelycum with plate of sternite XIV raised in strong bulges; posterior component traversed by deep groove. Branchiostegite with moderately large purplish brown spot bearing poorly defined yellow center. Description. — Body robust (Fig. 31). Carapace studded with long setae anteroventral to hepat- ic spine and pterygostomian region; patches of densely set shorter setae present on dorsum, in depression anterior to posterodorsal part of bran- chiocardiac carina, on branchiostegite and along hepatic sulcus; patches also on tergum of abdomi- nal somites and in lateral depression of sixth so- mite. Abdomen with numerous tubercles on first three somites, fewer on last three. Rostrum short, not surpassing distal margin of eye, its length increasing linearly with carapace length (Fig. 32) but proportionately longer in young (0.30-0.12 cl); usually straight but occasion- ally curved, subhorizontal or elevated as much as 45°; tapering, sometimes considerably, from base to truncate tip; and armed with one dorsal (rarely 2) and two (96%) or three (4%) apical teeth. Adros- FlGURE 31. — Sicyonia penicillata Lockington, 23 mm cl, west of Punta T^sca, Isla Santa Margarita, Baja California Sur, Mexico. Lateral view. Scale = 5 mm. 38 PfcREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA tral carina, situated distinctly dorsal to ventral margin, strong and extending to end of rostrum. Carapace with well-marked postrostral carina bearing three teeth: 1) epigastric tooth small, only slightly larger than first rostral, situated opposite (rarely) to <0.1 cl posterior to orbital margin; 2) middle tooth, larger than epigastric, sharp, placed well posterior to hepatic spine, between 0.33 and 0.40 (mean 0.36) cl from orbital margin; and 3) posterior tooth, larger than middle tooth, acutely pointed, rising almost vertically with only apical part inclined anteriorly or so inclined throughout, and situated between 0.70 and 0.77 (mean 0.74) cl from orbital margin; tuft of setae present at an- terior base of all three teeth. Postrostral carina low anterior to middle tooth, high between middle and posterior one, and descending gradually from latter to posterior margin of carapace. Antennal spine relatively long, sharp, and projecting from short buttress; hepatic spine acutely pointed, larger than antennal, arising from moderately raised area between 0.20 and 0.26 (mean 0.22) cl from orbital margin. Postocular sulcus short but deep, continuing as shallow groove; hepatic sulcus well marked; branchiocardiac carina usually low (sometimes barely distinct) but occasionally quite prominent and long, extending to bifurcation near posterior margin of carapace, short branch curv- ing dorsally and longer one ventrally. Ocular stylet with terminal part straight or, occasionally, turned laterally. Antennular peduncle with stylocerite produced in long spine nearly or quite reaching mesial base 7 14 21 28 carapace length (mm) 35 Figure 32. — Sicyonia penicillata. Relationship between ros- trum length and carapace length (regression equation, y = 0.65537 + 0.13963;c). of distolateral spine; latter slender and sharp, ex- tending as far as proximal 0.70 of second antennu- lar article; antennular flagella short, with maximum length of 0.20-0.30 cl, in juveniles and young adults mesial flagellum slightly shorter than lateral one but in larger adults subequal to or slightly longer. Scaphocerite reaching distal margin of anten- nular peduncle or overreaching it by no more than 0.10 of its own length; lateral rib ending distally in sharp spine distinctly surpassing margin of lamella. Antennal flagellum 2.4-2.7 times cl in young, and as much as 2.0 times in larger adults. Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with high dorsomedian carina ex- tending from first through sixth somites, carina on first produced in strong, sharp, anterodorsally di- rected tooth, slightly to considerably larger than posterior tooth on carapace; carina on fifth somite produced in conspicuous tooth, and that on sixth terminating in strong acute one. Anterior margin of pleuron of first abdominal somite almost straight; anteroventral extremity of pleuron of first through fourth somites ending in spine, that of first directed anteroventral ly but that of second usually curved posterolaterally, re- sembling strong spines on third and fourth so- mites; posteroventral margin of first through third somites convex, on fourth broadly angular, and on fifth and sixth sharply so and armed with small caudally directed spine. Pleural spination of first four somites barely, if at all, distinct in juveniles, becoming stronger with increasing length of carapace. First somite marked by long, anteromedian pleural sulcus converging with united posterior tergal-posteromedian pleural sulci ventrally, dor- sal extremity of tergal reaching anterior margin of somite; posterior pleural sulcus weak, but usually clearly distinct. Second and third somites with anterior and posterior tergal sulci long, almost reaching base of pleuron; anteromedian pleural sulcus deep, continuous with anteroventral de- pression setting off elongate prominence dorsally and ridge posteriorly; posteromedian pleural sul- cus also long, extending anterodorsally subparal- lel to posterior tergal sulcus; shallow posterior pleural sulcus commonly present in both somites. Fourth and fifth somites with anterior tergal and curved, united posterior tergal-posteromedian pleural sulci merging dorsally. Sixth somite marked by arched posteromedian pleural sulcus and bearing longitudinal ridge along base of dor- 39 FISHERY BULLETIN: VOL. 83, NO. 1 somedian carina delimited ventrally by depres- sion lying just dorsal to well-defined cicatrix. Telson with pair of small but well-developed fixed spines. Rami of uropod subequal in length and falling slightly short of or overreaching apex of telson by no more than 0.10 its own length. Petasma (Fig. 33A , B ) with short distal plate of dorsomedian lobule bearing distolaterally small, scalelike process bent inwardly (posteroven- trally). Projection of distolateral lobule acicular, extremely long, about 0.75 as long as body of lobule, with heavily sclerotized triangular plate proximodorsally and flexible flagellum arising from ventrolateral surface; flagellum long, reach- ing between 0.60 and 0.75 length of projection from level of apex of triangular plate. Projection of ventrolateral lobule also long, about half length of acicular projection, bladelike and flexible, except for sclerotized tip curving proximally. Petasmal endopods coupled in males with carapace length as little as 5.5 mm, about 22 mm tl, but sometimes unjoined in individuals with carapace length as much as 11 mm, about 39 mm tl. Appendix masculina as illustrated in Figure 33C. Thelycum (Fig. 34) with plate of sternite XIV raised in paired, well-marked, elongate bulges sloping toward deep, narrow, median depression. Median plate of sternite XIII lanceolate, tapering gradually into slender spine reaching as far as proximal 0.20 of basis of extended second pereopods; posterior component with relatively shallow posteromedian emargination (occasion- ally replaced by longitudinal incision) often form- ing small notch anteriorly, and flanked by short, rounded, posterolateral processes marked basally by deep transverse suture. Sternite XI armed posteriorly with pair of broad based, acute spines. Posterior thoracic ridge with anteromedian mar- gin sharp and raised (ventrally), its lateral mar- gins usually well marked, occasionally flush with plate of sternite XIV. The smallest impregnated female encountered has a carapace length of 8.5 mm, about 33 mm tl. Color. — Available information based on speci- mens that had been recently caught in the Gulf of California is limited to a purplish brown spot, with a yellow center not sharply defined, postero- ventral to the hepatic spine; sometimes the entire spot is purplish brown (Anonymous 1980). Lock- FlGURE 33. — Sicyonia penicillata, 6 21.5 mm cl, west of Punta T^sca, Isla Santa Margarita, Baja California Sur, Mexico. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. 40 P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Figure 34. — Sicyonia penicillata, 9 23 mm cl, west of Punta Tasca, Isla Santa Margarita, Baja California Sur, Mexico. Thelycum. Scale = 1 mm. ington (1879) noted that "Color after two weeks exposure to alcohol, bright red; with a dark red- brown ocellated spot on each side of the carapace. Antennae bluish." In many of the preserved speci- mens that I have studied, the ocellus appears as a dark circle. Maximum size. — Males 32 mm cl, about 103 mm tl; females 35 mm cl, about 110 mm tl. Geographic and bathymetric ranges. — From southwest of Punta Canoas (29°20'N, 115°02'W), Baja California Norte, Mexico (Fig. 30), south- ward to Bahia San Lucas, and in the Gulf of California, from the northern end to Bahia Con- cepcion on the west and northern Sinaloa on the east; it seems to be absent from the southernmost part of the Gulf. It has also been found off Pun- tarenas (9'58'N, 84°50'W), Costa Rica (Boone 1930). This species occurs at depths between 0.60 and 180 m (latter by Boone 1930), mostly at 35-70 m. It has been recorded on sand (fine or coarse) and mud substrates that are sometimes densely cov- ered with algae. Discussion. — The closest affinities of S. penicil- lata seem to be with S. disedwardsi from which it differs most conspicuously in features of the petasma. The extremely long, slender, distal pro- jections of both the dorsolateral (which bears a lateral fiagellum) and ventrolateral lobules are far different from the short, stout, and simple projec- tions of the petasma of S. disedwardsi, as well as from those of all other species of Sicyonia. These two species also exhibit significant thelycal differ- ences. In the thelycum of S. penicillata the lateral bulges of the plate of sternite XIV are well defined; the posteromedian emargination of the median plate of sternite XIII is shallow (occasionally re- placed by a longitudinal incision), often bears an anterior notch, and is flanked by short posterolat- eral processes marked basally by a conspicuous transverse groove. In the thelycum of S. dised- wardsi the bulges are low, often indistinct, the posteromedian emargination of the median plate of sternite XIII is broad and deep, lacks a notch, and is flanked by relatively elongate posterolat- eral processes which are delimited anteriorly by a weak suture. The following characters are also helpful but somewhat less reliable for distinguishing between the two species. In S. penicillata the rostrum is armed with only one tooth (rarely two) on the dorsal margin, instead of two as in S. disedwardsi; the stylocerite reaches the mesial base of the disto- lateral spine of the first antennular article, whereas in the latter species it often falls short of the base; the dorsal extremity of the united pos- terior tergal-posteromedian pleural sulci of the first abdominal somite reaches the anterior mar- gin of the somite, whereas in S. disedwardsi often it does not; and the clearly defined although shal- low posterior pleural sulci are usually present on the first three abdominal somites of S. penicillata but are quite weak or, more often, lacking on some or all of the latter Fresh material of this shrimp may be identified by a purplish brown spot on the branchiostegite, sometimes bearing a yellow center with diffuse border. Burkenroad (1934a) presented a detailed discus- sion of the differences between S. penicillata and the western Atlantic S. typica (as Sicyonia ed- wardsii Miers, 1881). These species, which share among other characters three teeth on the post- rostral carina and usually one dorsal and two api- 41 FISHERY BULLETIN: VOL. 83, NO. 1 cal rostral teeth, differ strikingly in other fea- tures. In S. typica the sculpture of the abdomen is stronger than that in the eastern Pacific shrimp, exhibiting a long and deep posterior pleural sulcus on the first four somites instead of ones that are weak or even obsolete, and the telsonic spines are quite inconspicuous, rather than being well devel- oped as they are in S. penicillata .Also, in S. typica the petasma lacks long slender distal projections as well as accessory flagella, the thelycal plate of sternite XIV is almost flat laterally instead of ele- vated in strong bosses, and the posteromedian emargination of the posterior component of the median plate is quite broad rather than narrow or even reduced to a longitudinal incision as it is in S. penicillata. Remarks. — The types of this species were de- stroyed in the San Francisco earthquake and fire of April 1906 as were all of Lockington's types which had been deposited at the California Acad- emy of Sciences (Dunn 1982). The locality of one of the syntypes, the one Lockington described in de- tail, "Bolinas Bay, Lower California," is uncertain. The NIS Gazetteer (Office of Geography, Depart- ment of the Interior, 1956) does not include any place or geographic feature under "Bolinas". On the west coast of Baja California Sur is Bahia de Ballenas or "Ballenas Bay" (NIS Gazetteer, p. 50), at 26°45'N, 113°26'W, and it is quite possible that the name of this locality was misspelled on the label accompanying the sjnitj^e or that Lock- ington misread and transcribed it as "Bolinas Bay". There is a bay by this name at 37°53'36"N, 122°39'54"W, in Marin County, California; how- ever, I am inclined to think that Bahia de Ballenas actually is the place where the specimen was ob- tained because it is well within the range of the species, whereas Bolinas Bay is not only outside "Lower California" but also far beyond the known northern limit of this shrimp — southwest of Punta Canoas, Baja California Norte. Commercial importance. — There is a fishery for rock shrimp in the northern half of the Gulf of California, and the catches are believed to consist largely of S. penicillata a very abundant species in that area. This fishery in 1979-80 produced 1,426,541 kg, but in 1981-82 (data recorded in Guaymas by the Instituto Nacional de Pesca, Mexico), the last year for which landings are available, the production declined sharply to 187,786 kg; fishing for rock shrimp is only sea- sonal, from February to June, with maximum catches being obtained during March and April (Concepcion Rodriguez de la Cruz see footnote 2). Material. — 939 specimens from 56 lots. Mexico— Baja California Norte: 29, SIO, SW of Punta Canoas (29°20'N, 115°02'W), 40 m, 6 Sep- tember 1952, K. S. Norris. 16 29, SIO, Bahia Playa Maria, 11 m, 1 April 1952, K. S. Norris. 39 , SIO, Bahia Sebastian Vizcaino, surface, 17 Au- gust 1952, Spencer F. Baird. 16 , YPM, E of Isla Cedros, 1-73 m, Zaca stn 126D-3. 49 , SIO, Bahia Sebastian Vizcaino, 0-2 m, 14 August 1952, K. S. Norris. Baja California Sur: 3d , SIO, Bahia Sebastian Vizcaino, 55 m, 11 August 1952, K.S. Norris. 16 , SIO, E of entrance to Laguna Ojo de Liebre, 2 m, 16 August 1952, K. S. Norris. 76 59, SIO, Bahia Tortolo, 31 March 1962, H. C. Per- kins. Id 29, SIO, E of Punta Asuncion, Bahia Asuncion, 15 m, 24 March 1951, R. Wisner and K. S. Norris. 66 99, SIO, Bahia Asuncion, 40-44 m, 17 November 1964, Black Douglas . 136 119 , SIO, Bahia Asuncion, 68-64 m, 17 November 1964, Black Douglas. 116 149, SIO, Laguna San Ig- nacio, 1.5 m, 11/12 February 1950, C. Hubbs. 3d 89, SIO, Bahia de Ballenas, 18 m, 14 February 1948, Scripps. 43d 509, SIO, SE of Punta Abre- ojos, 55-59 m, 17 November 1964, Black Doug- las. 246 229, SIO, W of Punta Pequeha, 37-40 m, 16 November 1964, Black Douglas. 266 219 , SIO, off Punta Pequeiia, 55-51 m, 16 November 1964, Black Douglas. 46 109, SIO, WSW of Punta Pequeha, 68-73 m, 16 November 1964, Black Doug- las. 16 59, SIO, NW of Santo Domingo del Pacifico, 45-40 m, 19 April 1969, D. Dock- ins. 26d 399, SIO, 15 km WSW of Boca de las Animas, 55-57 m, 16 November 1964, Black Doug- las. 26 39, SIO, S of Boca de las Animas, 137 m, 29 January 1964, C. Hubbs. 13d 79 , SIO, between Boca de Santo Domingo and Boca de Sole- dad, 12 m, 5 February 1964, A. Stover and B. Zahuranec. 19, USNM, Boca de Soledad, 26 April 1964, H. Chapa. 19, SIO, channel N of Bahia Magdalena, 6 m, 30 January 1964, A. Stover and B. Zahuranec. 48d 449, SIO, NW of Isla Santa Magdalena, 73 m, 15 November 1964, Black Douglas. 3d 39, SIO, Bahia Santa Maria, 0-36 m, 8 December 1962, H. C. Perkins. Id 19, SIO, Bahia Magdalena, 42-44 m, 29 November 1962, F H. Berry. 19, USNM, Bahia Magdalena, surface, 10 July 1953. 14d 99, SIO, Bahia Magdalena, 37-40 m, 24 August 1960, F H. Berry 8d 69, YPM, Bahia Magdalena, 0.6-0.9 m, 1936, Zaca [unnumbered stn]. 19, AHF, Bahia Magdalena, between mainland and Punta Redonda, 15 m, 5 42 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA February 1974, H. G. Moser stn 13. 16, SIO, off Bahia Magdalena, 88 m, 3 February 1964, C. Hubbs. 5c? 69, SIO, Bahia Almejas, 21-24 m, 11 November 1964, Black Douglas. 506 509, SIO, SW of Isla Santa Margarita, 29-40 m, 13 November 1964, Black Douglas . 406 409 , SIO, 3 km SW of Isla Santa Margarita, 46-57 m, 13 November 1964, Black Douglas. 186 89, SIO, SW of Isla Santa Margarita, 75-80.5 m, 13 November 1964, Black Douglas. 16 59, SIO, SW of Isla Santa Mar- garita, 88-90 m, 13 November 1964, Black Doug- las. 36 29, SIO, W of Punta Tasca, Isla Santa Margarita, 7 July 1955, Andres stn 143. 19, SIO, 11 km NW of Punta Marquez, 55 m, F H. Ber- ry 29 , SIO, 14.5 km WNM of Punta Marquez, 92 m, 4 December 1962, F H. Berry 16, YPM, Bahia San Lucas, 11-37 m, 6 May 1936, Zaca stn 135D-11, 12. 16, SIO, Bahia Santa Ines, 40-82 m, 14 July 1965, C. Hubbs. 34c5 229, YPM, Bahia Concepcion, 3 May 1926, Pawnee. 16 29, SIO, off Santa Rosalia, 35-36 m, 25 March 1960, R. Parker Baja California Norte: 16 59, SIO, S arm of Bahia de los Angeles, 22-37 m, 26 April 1962, R. Rosenblatt. b6 159, YPM, Bahia de los Angeles, 31-46 m, 13 May 1926, Pawnee. 16 69, SIO, off San Felipe, 2 April 1973, C. Farwell. 4<5 29, YPM, Bahia San Felipe, 19 May 1926, Pawnee. &6 59, USNM, near northern end of Gulf of California, 9-18 m, February 1949, B. W. Walker. Sonora: Id 19, USNM, Bahia de Adair, 46 m, 5 April 1968, Toral-Garcia. 26 39, USNM, Bahia de Adair, 29 m, 5 April 1978, Toral-Garcia. 56 139, USNM, off Bahia de San Jorge, 26 March 1967, shrimp trawler. 26 19, INP, N of Guaymas, 26 April 1961, H. Chapa. 16, AHF, Bahia de Guaymas, 4-6 m, 22 January 1940. 19, SIO, Bahia de Guaymas, 22 March 1939, M. W Johnson. 26 , SIO, Bahia de Guasimas, 32 km S of Guaymas, April 1968, D. Hoese. Sinaloa: 16 19, USNM, off Sinaloa, H. Chapa. Sicyonia affinis Faxon 1893 Figures 35-38 Sicyonia affinis Faxon 1893:209 [syntypes: 16 19, MCZ 4637, off Isla del Coco, Costa Rica, 5°31'30"N, 86°52'30"W, 100 fm (183 m), 28 Feb- ruary 1891, Albatross stn 3367; 16, USNM 21169, off Isla del Coco, Costa Rica, 5°32'45"N, 86°55'20"W, 52 fm (95 m), 28 February 1891, Albatross stn 3369; 29, USNM 21170, W of Isla de Malpelo, Colombia, 3°58'20"N, 81°36'00"W, 112 fm (205 m), 5 March 1891, Albatross stn 3378. 19 , MCZ 4638, W of Isla de Malpelo, Co- lombia, 3°59'40"N, 81°35'00"W, 52 fm (95 m), 5 March 1891, Albatross stn 3379]. Faxon 1895:179, pi. 46, fig. 1, la-c. A. Milne Edwards and Bouvier 1909:244. De Man 1911:112. ?Chapa Saldaha 1964:9. Chirichigno Fonseca 1970:7, fig. 5. ?Rodriguez de la Cruz 1977: 12. Arana Espina and Mendez G. 1978:23, fig. 1-5. Mendez G. 1981:47, pi. 9, fig. 75-77. Perez Farfante and Boothe 1981:424. Eusicyonia affinis. Burkenroad 1934a:93, 1934b:126, 1938:84, fig. 24. Anderson and Lindner 1945:317. Sicyonia penicillata Boone 1930:115 [part]. [Not Sicyonia penicillata Lockington 1879.] Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing one tooth posterior to level of hepatic spine and forming high crest behind pos- terior tooth. Rostrum short, not overreaching distal margin of eye. Abdomen lacking tubercles; second and third somites lacking inverted V-shaped ridges laterally; fifth somite with dor- somedian carina lacking tooth or sharp angle at posterior end. Petasma with distal projection of dorsolateral lobule compressed distally its trun- cate tip produced dorsally in simple, minute spine. Thelycum with plate of sternite XIV without an- teromedian tubercle and raised in paired low, but well-defined bulges. Branchiostegite lacking large mark. Description. — Body relatively slender (Fig. 35) and lacking tubercles. Carapace studded with numerous short setae, those on anterior part of dorsum forming dense patches. First five abdomi- nal somites with paired broad patches of short setae flanking dorsomedian carina; sixth with one in dorsolateral depression. Rostrum short, reaching only as far as distal margin of eye, its length 0.25-0.35 cl; upturned to angle between 10° and 40°; armed with two dorsal teeth and three minute apical teeth; latter dis- posed on truncate apex with ventralmost one either terminal or subterminal; first dorsal tooth located distinctly anterior to orbital margin, sec- ond tooth situated at about anterior 0.25 cl of ros- trum. Conspicuous adrostral carina, subparallel and distinctly dorsal to ventral margin, extending almost to end of rostrum. Carapace with well-marked postrostral carina bearing two teeth; epigastric tooth, situated 43 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 35. — Sicyonia affinis Faxon, ? 17 mm cl, 4.4 km off Isla Manuelita, Costa Rica. Lateral view. Scale = 5 mm. slightly anterior to level of hepatic spine at about 0.15 cl from orbital margin, small, subequal to, or only slightly larger than first rostral tooth; and large posterior tooth, considerably larger than epigastric, hooklike, with apical portion acutely pointed and slightly curved anteriorly; tooth placed distinctly in advance of posterior margin of carapace, betw^een 0.65 and 0.75 (mean 0.67) cl from orbital margin. Postrostral carina low an- teriorly forming high crest descending gently from posterior tooth to posterior margin of carapace. Tuft of setae present at anterior base of each tooth. Antennal spine small, sharp, projecting from weakly developed buttress; hepatic spine moder- ately long and acutely pointed, situated at 0.20- 0.25 (mean 0.22) cl from orbital margin; hepatic sulcus almost horizontal, accompanying incon- spicuous carina; branchiocardiac carina broad, low, longitudinally disposed except for posterior part curving dorsally near posterior margin of carapace. Antennular peduncle with stylocerite long, al- most reaching level of mesial base of distolateral spine, its length about 0.95 distance between lat- eral base of first antennular article and mesial base of distolateral spine; latter reaching as far as distal 0.25 of second antennular article, antennu- lar flagella short, mesial one more slender and longer, about 0.20 cl, than lateral, 0.16 cl. Scaphocerite extending to distal end or slightly overreaching antennular peduncle; lateral rib produced distally in long, strong spine, surpassing margin of lamella. Antennal flagellum incomplete in specimens examined. Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with strongly marked dorsomedian carina extending from first through sixth somites, carina on first somite produced anteriorly in strong tooth (slightly more elevated than posterior tooth on carapace), its anterior margin subvertical but apical extremity slightly curved anteriorly; carina deeply cleft posteriorly on first five somites, on fourth and fifth not truncate but sloping gradu- ally to apex of cleft, and on that of sixth somite produced in large, acute, posterior tooth. First four somites with rounded or broadly an- gular anteroventral extremity unarmed; fourth somite with posteroventral extremity rounded and lacking spine, and that of fifth and sixth somites bearing minute spine. First somite marked with short, weak, an- teromedian pleural sulcus, its length slightly <0.2 distance from origin (emargination on anterior margin) to ventral margin of pleuron; united pos- terior tergal-posteromedian sulci relatively shal- low. Second and third somites with weak anterior and posterior tergal sulci joining anterodorsally; 44 P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA anteromedian pleural sulcus short, extending from slightly below midheight of somite to near ventral margin; posteromedian pleural sulcus with faint dorsal extension directed anteriorly at about 0.33 height of somite from dorsomedian line and just ventral to weak crescent- shaped ridge (latter delimited dorsally by tergal sulcus). Fourth somite with anterior tergal sulcus shallow and posterior tergal and posteromedian pleural sulci coalescent, extending from near base of dorsal carina to curve anteriorly near ventral margin. Fifth somite with united posterior tergal- posteromedian pleural sulci ending well above ventral margin. Sixth somite with strongly arched posterior pleural sulcus and low longitudinal ridge situated between base of middorsal carina and setose depression lying dorsal to weak cica- trix. Telson with very weak median sulcus and bear- ing pair of small, fixed subterminal spines. Both rami of uropod extending as far as apex of telson or surpassing it by as much as 0.2 of their own lengths. Petasma (Fig. 36A, B) with rigid distal projec- tion of dorsolateral lobule curved mesially, raised proximodorsally in rounded prominence, and compressed distally; its truncate tip with ventral extremity rounded and dorsal extremity acutely produced in acute salient. Distal projection of ven- trolateral lobule fleshy, with terminal part flat- tened (dorsal surface not bulbous) and curving dorsally. Appendix masculina as illustrated in Figure 36C. Thelycum (Fig. 37) with plate of sternite XIV, delimited by arched or straight lateral edges, rounded anteriorly, its surface raised in low, ovoid bulges separated by broad, median depression. Median plate of sternite XIII flask-shaped in out- line, tapering gradually into long, slender spine reaching between midlength of coxae and proxi- mal extremity of bases of second pereopods; plate incised and excavate at level of coxae of fourth pereopods; posterior component of plate with shal- low posteromedian emargination. Sternite XI armed with paired short spines. Posterior thoracic ridge with concave anteromedian margin slightly overlapping plate of sternite XIV, ridge then flush with, or separated by shallow, transverse depres- sion from sternite XIV. Color. — "...light greenish yellow, banded with vermilion on the branchial regions and abdomen. Appendages red, antennary flagellum trans- versely banded with light and dark" (Faxon 1893). Maximum size. — In the meager material avail- FIGURE 36.— Sicyonia affinis,i 21.5 mm cl, 4.4 km off Isla Manuelita, Costa Rica. A , Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scales = 1 mm. 45 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 37.— Sicyonia affinis, 9 17 mm cl, 4.4 km off Isla Ma- nuelita, Costa Rica. Thelycum. Scale = 1 mm. able, larger male 21.5 mm cl, about 46 mm tl; largest female, 17 mm cl, about 62 mm tl. Geographic and bathymetric ranges. — Known with certainty only from a restricted area between Isla Manuelita (5°34'N, 87°00'W), Costa Rica, and Isla de Malpelo (3°58'20"N, 81°36'00"W), Colom- bia (Fig. 38). Chirichigno Fonseca (1970) and Arana Espina and Mendez G. (1978) cited Paita, Peru, as the southern limit of the range of the species. Chirichigno Fonseca did not present a list of her material and Matilde Mendez G.^ found no representative of the species in Peruvian collec- tions, including those of the Instituto del Mar del Peru (IMARPE) from which Chirichigno Fonseca obtained most of her information. Chapa Saldaha (1964) recorded the occurrence of this species in the waters of Chiapas and Sinaloa, Mexico, but again, more recent studies, including the present one based on extensive material, failed to disclose its presence north of Costa Rica. Further investi- gations are necessary to ascertain the limits of the range of this shrimp both south of Isla de Malpelo and north of Isla Manuelita. Sicyonia affinis is one of only four members of the genus that have been recorded from the eastern Pacific off South America. This species has been found at depths between 79-77 and 205 m, on substrates of rocks or broken shells. Discussion . — Sicyonia affinis is one of the three closely related American Pacific species belonging to Burkenroad's (1934a) "affinis group" of his Divi- sion II. He characterized this group (in which he included S. affinis and S. aliaffinis, and to which * S. affinis * S. aliaffinis * S . martini 'Matilde Mendez G., Instituto del Mar del Peru, Callao, Peru, pers. commun., January 1984. 46 Figure 38. — Geographic distribution of Sicyonia affinis, S. aliaffinis, and S. martini. P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA recently Perez Farfante and Boothe (1981) added S. martini^ as possessing two teeth posterior to the orbital margin, of which one, that posterior to the level of the hepatic spine, is large. Like all mem- bers of Division II, in S. affinis the antennal spine is buttressed, but the buttress is barely distinct in contrast to that in S. aliaffinis and S. martini in which it is well developed. In S. affinis the first rostral tooth is situated farther anteriorly, at about the end of the anterior third of the rostral length from the orbital margin, than in the other two species in which it is placed opposite or im- mediately anterior to the orbital margin. In S. affinis the abdominal surface is punctate but otherwise rather smooth, lacking tubercles, conspicuous ridges, and deep sulci. Also, in S. affinis the anteromedian sulcus of the first abdom- inal somite is not only weak but short, whereas in S. aliaffinis and S. martini it is deep (except ven- trally in S. martini in which it is represented by a shallow depression) and long, in S. aliaffinis al- most reaching the ventral margin of the pleuron. The anteromedian sulci of the second and third somites in S. affinis are distinct only ventrally, rather than dorsally as in S. martini, or along most of the height of the somite, as in S. aliaffinis. The posteromedian sulci of these somites in S. affinis extend dorsally only to a point situated at about 0.33 of the height of the somite from the dorsal midline where they turn anteriorly and are marked dorsally by weak longitudinal ridges; these sulci are considerably shorter than the well- incised ones in S. aliaffinis, which extend to about the dorsal 0.25 of the height of the somite, and are not accompanied dorsally by longitudinal ridges. In S. martini the posteromedian sulci curve an- teroventrally at their dorsal ends, defining strong angular ridges. As Burkenroad (1934a) pointed out, in iS. affinis the dorsal carina of the fifth abdominal somite does not end in a sharp angle or acute tooth as it does in the other two species; instead its posterior part slopes gradually to the apex of the caudal cleft. In S. affinis the tip of the tooth on the first abdominal somite is slightly curved anteriorly whereas in S. martini it forms a conspicuous hook, and in S. aliaffinis the entire tooth is straight and projects anterodorsally Furthermore, the antero- ventral extremities of the pleura of the first four somites in S. affinis are unarmed whereas they bear a small spine in S. aliaffinis , and in S. mar- tini, although lacking spines, are strongly angular instead of faintly so or rounded as they are in S. affinis. These three species can also be distinguished readily by petasmal and thelycal characters. In S. affinis the projection of the dorsolateral lobule of the petasma, like that of S. aliaffinis, is truncate or shallowly emarginate distally and produced in a simple, dorsally directed, sharp salient, whereas in S. martini the projection curves gently to a conspicuously bifurcate, mesially directed tip; on the other hand, in S. affinis, as in S. martini, the projection of the ventrolateral lobule is fiattened and curved or concave dorsally rather than being strongly bulbous as it is in S. aliaffinis. In the females of S. affinis and S. martini the thelycal plate of sternite XIV bears a pair of low but well- marked lateral bulges (longitudinally disposed in the former and transversely so in the latter), whereas in S. aliaffinis the plate is almost flat or barely raised in ill-defined elevations. Moreover, in both S. affinis and S. aliaffinis, the posterior emargination of the median plate of sternite XIII does not embrace a tubercle, as it does in S. mar- tini. Material. — 17 specimens from 6 lots. Costa Rica— 3c5 59, AHF, 4.4 km off Isla Ma- nuelita, 146 m, 3 June 1973, Velero IV stn 19044. 16 , syntype, USNM, off Isla del Coco, 95 m, 28 February 1891, Albatross stn 3369. 16 19, syntypes, MCZ, off Isla del Coco, 183 m, 28 Feb- ruary 1891, Albatross stn 3367. Panama— 39, USNM, NE of Isla Iguana, 79-77 m, 4 May 1967, Pillsbury stn 515. Colombia — 19, syntype, MCZ, W of Isla de Mal- pelo, 95 m, 5 March 1891, Albatross stn 3379. 29 , syntypes, USNM, W of Isla de Malpelo, 205 m, 5 March 1891, Albatross stn 3378. Sicyonia aliaffinis (Burkenroad 1934) Figures 38-42 Eusicyonia aliaffinis Burkenroad, 1934a:92, fig. 24 [holotype 6, YPM 4393; type-locality: Pacific coast of southern Mexico (NW of Puerto Ma- dero), 14°48'40"N, 92°54'40"W, 19-30 fm (35- 55 m), 9 April 1926, Pawnee]. Burkenroad 1938:84, fig. 25, 27. Anderson and Lindner, 1945:317. Eusicyonia sp. Castro, 1966:17 [in part, by im- plication]. Sicyonia aliaffinis. Chapa Saldana 1964: 15. Bayer et al. 1970:A97. Chirichigno Fonseca 1970:7, fig. 6. Del Solar 1972: 7. Rodriguez de la Cruz 1977:10. Arana Espina and Mendez G. 1978:25, fig. 6-9. 47 FISHERY BULLETIN: VOL. 83, NO. 1 Anonymous 1980:7. Brusca 1980:256. Sosa Hernandez et al. 1980:12. Mendez G. 1981:47, pi. 9, fig. 78-82. Perez Farfante and Booths 1981:424. Perez Farfante 1982:370. Vernacular names: rock shrimp, target shrimp, Japanese shrimp (United States); camaron de piedra, camaron de roca, camaron japones, cacahuete (Mexico); camaron conchiduro (Mexico, Panama); camaron de mar, camaron cascaradura (Peru). FAO names: hardhusk rock shrimp (English); camaron cascara dura (Spanish); boucot noisette (French). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing one tooth posterior to level of he- patic spine and raised in high crest behind pos- terior tooth. Rostrum short, not overreaching dis- tal margin of eye. Abdomen tuberculate; second and third somites not bearing special inverted V-shaped ridges laterally; fifth somite with dor- somedian carina ending in sharp angle or tooth posteriorly. Petasma with distal projection of dor- solateral lobule compressed distally, its truncate tip produced dorsally in simple, minute spine. Thelycum with plate of sternite XIV lacking an- teromedian tubercle and either flat or barely raised in ill-defined bulges. Branchiostegite bear- ing large, horizontally disposed 9-shaped color pattern. Description. — Body relatively short (Fig. 39). Carapace with irregular patches of longer setae on dorsum, in depression delimiting branchiocardiac carina posterodorsally, and on areas anterior to hepatic spine and ventral to hepatic sulcus; one patch also present in dorsolateral depression of sixth abdominal somite. Abdomen rather heavily granulate on first three abdominal somites, usu- ally slightly so on last somites. Rostrum short, not overreaching distal margin of eye, its length increasing linearly with carapace length (Fig. 40) to about 16 mm cl, then increasing little, not surpassing 6 mm (proportional length decreasing with increasing size from as much as 0.43 to as little as 0.20 cl); subhorizontal or up- turned to 30° in males and to 50° in females; armed with two dorsal teeth and three (occasionally two) apical ones, latter disposed on obliquely truncate apex, upper tooth posterior to level of ventral one (occasionally appearing to be third of dorsal series); first dorsal tooth subequal to or, more often slightly smaller than, epigastric and situated op- posite or immediately anterior to orbital margin; second tooth variably placed between anterior 0.17 and 0.40 (mean 0.30) rl. Conspicuous adrostral carina, subparallel and near ventral margin, ex- FlGURE 39. — Sicyonia aliaffinis (Burkenroad 1934), 9 26 mm cl, west of Puerto Madero, Golfo de Tehuantepec, Mexico. Lateral view. Scale = 5 mm. 48 PfiREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA t) • 1 6 J • • • ■^L-^ ^^ • /T .I'''** 0» c t • , ^•«i *• • J.-- • — 4 * . / . .••« E 3 J^ \ k • *>* M j^ • o 2 yV% • y^9 12 18 24 carapace length (mm) 30 Figure 40. — Sicyonia aliaffinis. Relationship between rostrum length and carapace length (regression equation for specimens with about 16 mm cl or less,^' = 0.83950 + 0.25635x; regression equation for those larger, y = 2.34086 + 0.13665x). tending from orbital margin almost to end of ros- trum. Carapace with strong postrostral carina bear- ing two teeth: 1) epigastric tooth small, subequal to or slightly larger than first rostral tooth, situated opposite or anterior to level of hepatic spine, between 0.15 and 0.22 (mean 0.18) cl from orbital margin; and 2) posterior tooth, much larger, as much as three times higher than epigas- tric, hooklike, its apical portion acutely pointed and strongly curved anteroventrad; tooth placed well in advance of posterior margin of carapace, between 0.66 and 0.76 (mean 0.71) cl from orbital margin. Postrostral carina slightly elevated just in front of posterior tooth and forming high crest from latter descending gently to posterior margin of carapace. Tuft of setae present at anterior base of each tooth. Antennal spine sharp, projecting from well-marked buttress; hepatic spine acute, larger than antennal, arising from raised area, and situated between 0.19 and 0.26 (mean 0.23) cl from orbital margin. Postocular sulcus deep an- teriorly, continuing posteriorly as low groove; hepatic sulcus well marked; branchiocardiac carina distinct but rather low, extending longitu- dinally from hepatic region almost to posterior margin of carapace, there bifurcating: one branch curving dorsally and other disposed ventrally. Antennular peduncle with stylocerite produced in long spine, its length about 0.9 distance be- tween lateral base of first antennular article and mesial base of distolateral spine; latter extending to about midlength of second article. Scaphocerite almost reaching or slightly over- reaching distal margin of antennular peduncle; lateral rib produced distally in long, strong spine surpassing distal margin of lamella. Antennal flagellum as much as 2 times as long as carapace. Abdomen with high dorsomedian carina ex- tending from first through sixth somites, carina on first somite produced in strong triangular tooth as high as, or usually higher (as much as one-third) than, posterior tooth on carapace, its anterior margin straight, subvertical or sloping anterodor- sally; carina on fourth somite obliquely truncate posteriorly forming obtuse (rarely almost right angle); that on fifth strongly truncate forming acute posterior tooth; and that on sixth strongly produced in large acute posterior tooth. First four somites with angular anteroventral extremity bearing small spine; fourth somite with posteroventral extremity broadly angular, occa- sionally armed with minute spine, and fifth and sixth somites with posteroventral extremity bear- ing small sharp spine, that of fifth slightly larger. First somite marked with long anteromedian pleural sulcus joining coalescent posterior tergal-posteromedian pleural sulci near margin of pleuron. Second and third somites with deep an- terior and posterior tergal and long, well-incised anteromedian (expanding ventrally) and pos- teromedian pleural sulci, posteromedian ones ex- tending dorsally to a point located at least at 0.25 of the height of the somite from the dorsal midline. Fourth somite bearing anterior and posterior ter- gal sulci, posterior one merging with deep, long posteromedian sulcus. Fifth somite marked with anterior tergal sulcus and united posterior tergal-posteromedian pleural sulci. Sixth somite with short anterior tergal sulcus, strongly arched posterior pleural one, and setose, longitudinal de- pression delimited dorsally by rib and ventrally by usually strong cicatrix. Telson with pair of small but well-developed fixed spines. Rami of uropod subequal in length, reaching or slightly overreaching apex of telson. Petasma (Fig. 41) with rigid distal projection of dorsolateral lobule curved mesially, raised prox- imodorsally in rounded prominence, and com- pressed distally; its truncate tip with ventral ex- tremity rounded and dorsal extremity sharply produced in minute spine. Fleshy distal projection of ventrolateral lobule with firm, terminal part directed laterally almost at right angle, dorsally bulbous, ventrally flat, and tapering to pointed, ventrally inclined apex. Petasmal endopods coupled in males 8.2 mm cl, 49 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 41. — Sicyonta aliaffinis ,6 16 mm cl, west of Puerto Madero, Golfo de Tehuantepec, Mexico. A , Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. about 27 mm tl, but may not be joined in individu- als as much as 11 mm cl, about 36 mm tl. Appendix masculina as illustrated in Figure 41C. Thelycum (Fig. 42) with plate of sternite XIV, delimited anterolaterally by strongly convex mar- gins, flat or very slightly raised in paired ill- defined bulges flanking depressed median portion. Median plate of sternite XIII flask-shaped in out- line, tapering gradually into long, slender spine reaching as far as distal margin of coxae of an- teriorly extended second pereopods; posterior component of plate with posterolateral margins strongly arched and separated by median emargi- nation variable in width. Sternite XI armed pos- teriorly with paired short spine. Posterior thoracic ridge with weakly concave or virtually straight anteromedian portion slightly elevated, but areas lateral to it merging indistinctly with plate of sternite XIV. The smallest impregnated female encountered has a carapace of 5 mm, about 23 mm tl. Color. — Specimens from Peruvian waters were described by Arana Espina and Mendez G. (1978) as follows: dorsum dark, petroleum green; carapace lighter laterally, exhibiting various shades of gray, green, or pink, and bearing striking dark mark resembling longitudinally disposed "9" on branchial region. Antennae with light and dark Figure 42. — Sicyonia aliaffinis, * 25 mm cl, west of Puerto Madero, Golfo de Tehuantepec, Mexico. Thelycum. Scale = 2 50 PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA bands. Pereopods and pleopods pink. In addition, spot — dark proximally, intense blue distally — present on lateral ramus of uropods. Sosa Hernan- dez et al. (1980) also presented color notes on specimens from the Golfo de Tehuantepec: body cream, suffused with reddish brown; carapace bearing hook-shaped brown mark on each side; lateral ramus of uropod with violet ventral spot; antennae banded with violet with cream. In Anonymous' (1980) work on the crustacean dec- apods of the Gulf of California, the "9" is described as purplish brown. Maximum size. — Male, 22.0 mm cl, 86.6 mm tl; female, 28.5 mm cl, 100.7 mm tl (both recorded by Arana Espina and Mendez G. 1978). Largest indi- viduals examined by me: male, 21.4 mm cl, about 71 mm tl; female, 28 mm cl, about 89 mm tl. Geographic and bathymetric ranges. — Isla Santa Margarita (24°20'00"N, 111°45'30"W - 24°20' 10"N, 111°46'40"W), Baja California Sur, Mexico, to Cabo San Lucas, in the southern part of the Gulf of California along both the east and west coasts, and southward to Bahia Chamela (19°34'00"N, 105°07'24"W), Jalisco. Also from off Salina Cruz (16°10'00"N, 95°00'00"W), Oaxaca, Mexico, to Santa Maria (12°24'S), Peru, except off middle Central America, Colombia, and most of Ecuador. In the waters of Ecuador, it has been recorded from the Golfo de Guayaquil and Islas Galapagos (Fig. 38). This species has been found at depths between 4-9 and 242 m, mostly at <85 m, on substrates of sand and mud. The report of the occurrence of this shrimp in Santa Maria, Peru, (Velez J., J. Zeballos, and M. Mendez G., in press) is the first from waters south of Bahia Sechura (5°43.1'S, 81°05.0'W), the southernmost record cited by Arana Espina and Mendez G. (1978). These specimens from Santa Maria were collected at a depth of 10.5 m by A. Robles on 28 June 1983. Discussion. — The closest relative of S. aliaffinis is iS. affinis. The former, however, reaches a larger size (about 29 mm cl) than S. affinis, the largest known specimen of which has only a 17 mm cl. Sicyonia aliaffinis also differs from S. affinis in having a strongly buttressed antennal spine and in the position of the first dorsal rostral tooth, which is situated opposite or barely anterior to the orbital margin instead of at about the anterior end of the basal third of the rostrum. Differences in the abdominal characters are even more striking. In S. aliaffinis the abdomen is granulose, heavily so on the first three somites, and the transverse sulci are deeply incised whereas in S. affinis it is glabrous and bears weak sulci, some of which are incomplete, adding to the smooth appearance of the abdomen. In S. aliaffinis the anteromedian sulcus of the first somite is long, and although it becomes shallow ventral ly, it ex- tends to near the ventral margin of the pleuron; in S. affinis, in contrast, it is short, ending consider- ably above the ventral margin of the pleuron. The anteromedian sulci of the second and third somites in iS. aliaffinis are long instead of short, recogniz- able only on the ventral half of the somites; the posteromedian sulci of these somites in S. aliaffinis extend dorsally to a point at least at 0.25 of the height of the somite from the dorsal midline and do not turn anteriorly, whereas in S. affinis they extend only to about 0.35 from the dorsal midline and turn anteriorly, delimiting ventrally a weak longitudinal ridge which is absent in S. aliaffinis. Also in iS. aliaffinis, the anterior tooth of the first somite is acute but not curved at the tip as it is in the other species, and the dorsal carina of the fifth somite ends in a sharp angle or more often in a tooth, whereas in S. affinis it slopes gradually to the base of the caudal cleft. The anteroventral extremities of the first through fourth pleura in S. aliaffinis bear a small spine rather than being unarmed. Features of the external genitalia also allow a ready separation of these two species. In S. aliaffinis, the distal projection of the ventrolateral lobule of the petasma is bulbous dorsally; in con- trast, that of S. affinis is comparatively thin. The thelycal plate of sternite XIV is flat or very faintly raised laterally in ill-defined elevations in S. aliaffinis, whereas in S. affinis it bears a pair of low but well-marked ovoid or subellipticai bulges. In addition to the morphological characters dis- cussed above, S. aliaffinis exhibits a striking 9-shaped color pattern on the branchial region which distinguishes it from all of its congeners occurring in the American Pacific. Discussing the diagnostic characters of S. aliaffinis, Burkenroad (1934a) stated that "The carina of the second somite is, although not notched above the juncture of the tergal sulci, shallowly emarginate at this point." I have ob- served that this carina may be entire or slightly depressed either at the point where Burkenroad noted it or more posteriorly; consequently, in this shrimp the contour of the carina is insignificant. 51 FISHERY BULLETIN: VOL. 83, NO. 1 Remarks. — Arana Espina and Mendez G. (1978) graphed the size distribution of each sex in sam- ples of this shrimp from the Golfo de Guayaquil. They included correlations between carapace length and total length, total weight, and abdomi- nal weight. They determined that the relative growth rate in males is higher than that in females, and that within the size range of the shrimp studied, eight molts occurred with an in- crease of 7.25% at each molt. Although Castro (1966) did not cite S. aliaffinis by name, he stated that among the specimens of "Eusicyonia" collected off Puerto Pehasco and near Isla de San Jorge, Sonora, Mexico, there were some bearing a 9-like shaped spot, which undoubtedly indicates that they belonged to this species. Commercial im,portance . — Sicyonia aliaffinis, like the other six relatively large species of the genus occurring in the American Pacific, is fre- quently taken together with other penaeoids of greater economic value. In the Gulf of California it is present in the commercial catches made on the eastern side. In some other areas along its range, e.g., the Golfo de Guayaquil (Arana Espina and Mendez G. 1978), it is found in quantities that might support development of a fishery. Material. — 251 specimens from 27 lots. Mexico— Baja California Sur: 2c5 19, SIO, SW of Isla Santa Margarita, 29-40 m, 13 November 1964, Black Douglas . 16 , SIO, NW of Todos Santos, 38 m, 9 November 1964, Black Douglas. IS, YPM, Bahia San Lucas, 24 m, 7 May 1936, Zaca stn 135D-26. Id 19, USNM, W of Estero de los Al- godones, 47 m, 3 April 1978, Toral Garcia. Id, YPM, Bahia Santa Ines, 37 m, 10 April 1936, Zaca stn 141-D4. Nayarit: 57d 389 , SIO, NE of Isla Maria Madre, 51 m, 31 March 1973, Agassiz. 15d 89, SIO, NE of Isla Maria Madre, 55 m, 31 March 1973, Agassiz. Jalisco: 19, USNM, Puerto Vallarta, 13 April 1937. 29 , SIO, N part of Bahia Chamela, 15-18 m, 2 April 1973, Agassiz. Oaxa- ca: 13d 139, USNM, E of SaUna Cruz, Golfo de Tehuantepec, 18 m, 10 July 1963, 1. Mayes A. 2d 19, SIO, Golfo de Tehuantepec, 55 m, 6 June 1965, T. Matsui. 19, USNM, Laguna Lagartero, Ixhuatan, 25 July 1963, G. Solorzano. Chiapas: 2d 19, USNM, Puerto Arista, 14 January 1964, I. Mayes A. 8d 89 , SIO, Golfo de Tehuantepec, 46-48 m, 10 April 1973, Agas- siz. 3d 29 , SIO, Golfo de Tehuantepec, 73 m, 10-11 July 1963, D. Dockins. Id, holotype, YPM, off "southern Mexico" [NW of Puerto Madero], 35-55 m, 9 April 1926, Pawnee. 17d 389, SIO, W of Puerto Madero, Golfo de Tehuantepec, 55 m, 10 April 1973, Agassiz. Guatemala — Id, AHF, off San Jose light, 42 m, 23 March 1939. Costa Rica— Id , USNM, near Quepos, 242 m, 26 April 1973, Enriqueta. Panama— Id, AHF, Isla Taboga, 4-9 m, 2 May 1939. Id, USNM, Bahia Santelino, 1.6 km N of Punta de Cocos, Archipielago de las Perlas, 9 Feb- ruary 1939. 19 , USNM, S of Isla del Rey 44-42 m, 7 May 1967, Pillshury stn 551. 19 ,USNM, SW of Bahia San Miguel, 55 m, 7 May 1967, Pillsbury stn 549. 2d 19, USNM, 12 km NW of Punta Caracoles, Darien, 84 m, L. G. Abele. Ecuador— 19, USNM, S of Isla Seymour, Galapagos, 7-13 m, 9 March 1938, F E. Lewis. 19 , USNM, off Playas, Golfo de Guayaquil, 16 m, 1976, P Arana Espina. Peru— Id, USNM, off Caleta Cruz, 10-14 m, 1970, E. Valdivia. Sicyonia martini Perez Farfante and Boothe 1981 Figures 38, 43-46 Eusicyonia species, Burkenroad 1938:81, fig. 26, 28-30. Sicyonia martini Perez Farfante and Boothe 1981:424, fig. 1-4 [holotype 9, USNM 180235; type-locality: SW of Punta Ana Maria, Golfo de Panama, 7°50'30"N, 78°49'00"W, 58 m, Pillsbury stn 556]. Vernacular names: rock shrimp (United States); camaron de piedra, camaron de roca (Mexico); camaron conchiduro (Mexico, Panama). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral ca- rina bearing one tooth posterior to level of hepat- ic spine and raised in high crest behind posterior tooth. Rostrum long, conspicuously overreaching distal margin of eye. Abdomen tuberculate; second and third somites bearing unusual inverted V-shaped ridges laterally; fifth somite with dor- somedian carina sharply truncate posteriorly. Petasma with distal projection of dorsolateral lobule tapering distally to minutely bifurcate tip, arms sharp. Thelycum with plate of sternite XIV bearing anteromedian tubercle. Branchiostegite without 9-shaped color pattern. 52 PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Figure 43.—Sicyonia martini Perez Farfante and Boothe (1981), paratypec? 19 mm cl, off Puerto Escondido, Golfo de Panama. Lateral view. Scale = 5 mm. Description. — Body relatively slender (Fig. 43). Carapace with sparse long setae intermingled with elongate patches of shorter ones situated on dorsum, ventral to hepatic sulcus, posterior to pterygostomian region, and on posterodorsal part of branchiostegite; patches also present on abdom- inal terga. Abdomen tuberculate, tubercles numerous on first five somites, few on sixth. Rostrum comparatively long, conspicuously surpassing eye, reaching as far as distal 0.33 of second antennular article, its length, 0.40-0.54 cl, increasing linearly with carapace length (Fig. 44); armed with two or three dorsal teeth and cluster of apical teeth, both groups varying in disposition and number in males and females. In males, ros- trum horizontal or directed upward at slight angle of no more than 10°, but weakly decurved at tip, with three dorsal teeth evenly spaced; first rostral tooth situated immediately anterior to orbital margin, last usually separated from upper apical tooth by interval (about 0.33 rl) slightly greater than that between dorsal teeth; apical cluster con- sisting of three or four teeth (76% and 24% , respec- tively), with subterminal ventral tooth situated not far from adjacent apical tooth. In females, ros- trum strongly elevated at angle of 40°-50°, its ven- tral margin straight or, more often, strongly con- vex along midlength, and with two dorsal teeth; first rostral tooth placed distinctly anterior to orbi- 10 E E 6 • c _« E 3 4 10 15 20 carapace length (mm) 25 Figure 44. — Sicyonia martini. Relationship between rostrum length and carapace length (regression equation, j = 0.76705 + 0.39436X). tal margin; interval between second tooth and upper apical tooth slightly shorter than that be- tween first and second; apical cluster consisting of three, four, or five teeth (1, 87, and 12%, respec- tively), ventral one subterminal, distinctly re- moved from adjacent apical tooth; teeth of apical 53 FISHERY BULLETIN: VOL. 83, NO. 1 cluster either turned ventrally, or less frequently directed forward. Adrostral carina strong, some- times sharp, reaching between level of last apical tooth and base of ventral apical one. Carapace with postrostral carina well-marked, bearing two teeth: 1) epigastric tooth, small, sub- equal to first rostral tooth, situated opposite or only slightly anterior to hepatic spine; and 2) pos- terior tooth, placed between 0.63 and 0.73 (mean 0.69) cl from orbital margin, large, about three times as high as epigastric tooth, hooklike at tip; postrostral carina low anteriorly, slightly elevated below apex of posterior tooth, and forming crest from latter descending gently toward ridge on posterior margin of carapace; each tooth preceded by tuft of long setae. Antennal spine small, project- ing from rather long buttress; hepatic spine acutely pointed, much longer than antennal, aris- ing from moderately raised area, and situated be- tween 0.14 and 0.20 (mean 0.18) cl from orbital margin. Postocular sulcus short but deep an- teriorly, continuing posteriorly as weak groove; hepatic sulcus deep, subhorizontal, accompanying inconspicuous carina. Branchiocardiac carina low but clearly distinct, long, extending from base of hepatic region to near posterior margin of carapace, then curving dorsal ly toward base of posterior tooth. Antennular peduncle with stylocerite produced in spine distally, extending about 0.8 distance be- tween lateral base of first antennular article and mesial base of distolateral spine; distolateral spine strong, reaching as far as midlength of sec- ond article; antennular flagella short, mesial one slightly shorter than lateral. Scaphocerite almost reaching (occasionally overreaching) distal end of antennular pedun- cle; length of antennal flagellum as much as 2.5 times cl. Third maxilliped slightly stouter than pereopods. Basis and ischium of first pereopod un- armed. Abdomen with high dorsomedian carina ex- tending from first through sixth somites; carina on first somite produced in large, apically hooked, triangular anterior tooth, more elevated than posterior tooth on carapace; carina of fifth somite abruptly truncate posteriorly; and that of sixth produced in large, acute posterior tooth. Anteroventral margin of pleuron of first abdom- inal somite barely to distinctly concave; antero- ventral angle 90°-100°, that of third and fourth 90° or less, with vertex slightly produced anteroven- trally; pleuron of fifth roughly pentagonal, an- 54 teroventral and posteroventral angles with ver- tices slightly produced, posteroventral one often armed with small spine; posteroventral angle of pleuron of fifth and sixth somites armed with spine, that of fifth larger. First somite with anteromedian sulcus well de- fined only dorsally but continuing ventrally as shallow depression joining deep posterior tergal- posteromedian pleural sulcus, ridge often extend- ing posteriorly from ventral portion of anterome- dian pleural sulcus to fused posterior sulci. Second and third somites with relatively short, anterior and posterior tergal sulci; short anteromedian pleural sulcus merging ventrally with conspicu- ous broad depression, latter terminating near an- teroventral margin of corresponding pleuron; posteromedian pleural sulcus extending dorsally to about 0.3 height of somite measuring from mid- dorsal line, there curving anteriorly; special in- verted V-shaped ridge Ijdng between tergal and pleural sulci. Fourth somite with anterior tergal and long, united posterior tergal-posteromedian pleural sulci; anteroventral part of latter curving dorsally; often short longitudinal ridge present at about 0.3 height of somite from middorsal line. Fifth somite with anterior tergal sulcus continu- ous with united posterior tergal-posteromedian pleural sulci, anteroventral portion of latter fad- ing as shallow depression; cicatrix extending post- eriorly from ventral end of anterior tergal sulcus. Sixth somite with arched posterior pleural sulcus and with shallow setose depression situated dorsal to long but interrupted strong cicatrix. Telson with pair of small, fixed, subterminal spines. Both rami of uropod reaching, or almost reaching apex of telson. Petasma (Fig. 45 A, B) with rigid distal projec- tion of dorsolateral lobule strongly curved me- sially, raised proximodorsally in subhemispheric prominence, and ending in bifurcate apex, both tips sharp. Fleshy distal projection of ventrolat- eral lobule falling short of adjacent one, and with terminal part truncate and curved dorsally. Petasmal endopods coupled in males as small as 5.8 mm cl, about 23 mm tl, but may not be joined in individuals as large as 9 mm cl, about 32 mm tl. Appendix masculina as illustrated in Figure 45C. Thelycum (Fig. 46A, B) with plate of sternite XIV forming slightly to broadly rounded lateral flanges partly surrounding and merging with roughly semicircular, low mesial bulges; latter separated by median depression bearing oval or, occasionally, subhemispheric anterior tubercle (if P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Figure 45. — Sicyonia martini, paratyped 13.7 mm cl, south of Archipielago de las Perlas, Golfo de Panama. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. Figure 46. — Sicyonia martini. A , holotype? 21 mm cl, southwest of Punta Ana Mari a, Golfo de Panama; B ,9 16.5 mm cl, Banco Gorda de Afuera, Baja California Sur Thelyca. Scales = 1 mm. 55 FISHERY BULLETIN: VOL. 83, NO. 1 oval, long axis disposed either longitudinally or transversely). Median plate of sternite XIII flaskshaped in outline, tapering into long, slender spine reaching between anterior and posterior ex- tremities of coxa of anteriorly extended second pereopods; plate constricted, its ventral surface strongly excavate at level of coxae of fourth pereopods; posterior component of plate, with rounded posterolateral margins and broad shal- low, median emargination. Paired, broad based spines projecting anteriorly from posterior margin of sternite XI. Posterior thoracic ridge narrow, with concave, sharp anteromedian margin but merging laterally with preceeding plate. The smallest impregnated females encountered have a carapace of 8 mm, about 31 mm tl. Color. — Specimens preserved in Formalin'* buff with purplish blue markings: antenna, lateral ridge of scaphocerite, postrostral and abdominal carina, and dorsal ribs of telson transversely banded; anterior margin and posterior ridge of carapace, anterior margin of pleuron of first ab- dominal somite, and posterior margin of all ab- dominal somites with series of small spots; tip of teeth on rostrum, carapace, and first abdominal somite also purplish blue; lateral ramus of uropod with subterminal spot on lateral ridge and large mesial blotch at same level. Maximum size. — Males 15.6 mm cl, 60.5 mm tl; females 22.5 mm cl, 87.2 mm tl. Geographic and bathymetric ranges. — From southwest of Isla Santa Margarita (24°19'48" N, lir47'06"W - 24°19'36"N, lir47'06"W), Baja California Sur, Mexico, to southern tip of Baja California Sur and throughout the Gulf of California southward to off Punta Lizardo (18°06'00"N, 102°5718"W), Michoacan; also from E of Puerto Angel (15°41'00"N, 96°07'30"W), Oaxaca, Mexico, to SW of Punta Ana Maria ( 7°50'30"N, 78°49'00"W - 7°50'48"N, 78°48'00"W), Panama (Fig. 38). It has been found at depths between 9 and 242 m, on substrates of sand, rock, mud, and coralline debris. D/scuss/on. — Although closely allied to S. affinis and S. aliaffinis, S. martini can be distinguished readily from both of them by the length, shape, and armature of the rostrum; the shape of the tooth on Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. 56 the first abdominal somite; the sculpture of the ab- domen; and features of the petasma and thelycum. In S. martini the rostrum is quite long, surpass- es the eyes, and almost reaches the distal margin of the second antennular article. In males, the rostrum is straight or upturned at an angle of no more than 10°, and armed with three dorsal teeth and three or four apical teeth, the ventral one of which is occasionally subterminal. In females, the rostrum is strongly elevated (40°-50°), with its ventral margin usually markedly convex in the middle and concave posterior to the base of the subterminal tooth, and bears two dorsal teeth and three to five apical teeth, the ventral one of which lies distinctly posterior to the adjacent tooth. In S. affinis and S. aliaffinis the rostrum is shorter than in S. martini, reaching at most the distal margin of the eye; in both males and females it is upturned at an angle of about 30°, thus more elevated than in males of S. martini but less so than in females, and its ventral margin is usually straight or, occa- sionally, slightly convex basally. Also, in these two species the rostrum is armed with only two dorsal teeth, and the ventral of the two or three apical teeth (four or five have not been observed) is ter- minal, instead of subterminal as it is in all females and some males of S. martini. The tooth on the first abdominal somite is pro- portionately higher in S. martini than in the other two species; its dorsal margin is sigmoid and it ends in a strong, recurved, hooklike tip. In S. affinis and S. aliaffinis the dorsal margin of the tooth is gently curved in an arc, and the tooth is inclined more anteriorly than in S. martini; in S. affinis it ends in a slightly curved tip, and in S. aliaffinis the tip is triangular rather than hook- like. Also, the abdominal sculpture of S. martini is much stronger than that of its two closest congen- ers, and exhibits unusual, longitudinally disposed, inverted V-shaped ridges at the ventral end of the dorsal third of the second and third somites, which are absent in the other two species. In S. martini, the projection of the dorsolateral lobule of the petasma is bifurcate apically, the tips sharp. In S. affinis and S. aliaffinis, the projection is compressed distally with the ventral extremity rounded, the dorsal extremity sharply produced in a simple spine, and the distal margin (im- mediately ventral to the spine) truncate or slightly emarginate. Furthermore, the projection of the ventrolateral lobule of the petasma of S. martini, like that of S. affinis but in contrast to that of S. aliaffinis, is flattened distally rather than thickened (dorsally) into a subovoid pro- PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA tuberance. The females of S. martini differ from those of all the other Sicyonia occurring in the region by possessing a conspicuous tubercle on the anteromedian extremity of the plate of sternite XIV. In addition to the distinguishing characters cited above, S. martini exhibits other features that differ from those of S. affinis. The dorsomedian carina of the fifth abdominal somite is abruptly truncate; the abdomen is coarsely tuberculate and the tergal sulci on the third through fifth somites are deep. Sicyonia martini also differs from S. affinis in the barely to distinctly concave (instead of convex) anterior margin of the pleuron of the first somite; the anteroventral extremity of the four anterior pleura which are markedly angular, forming angles of about 90° or less, rather than being rounded or broadly angular; and the outline of the fourth abdominal pleuron which is subpen- tagonal and often bears a spine on the posteroven- tral angle, whereas in S. affinis it is subcircular and always unarmed. Sicyonia martini differs further from S. aliaffinis by the presence on the first abdominal somite of a short but conspicuous longitudinal ridge extending posteriorly from the ventral end of the anteromedian pleural sulcus. The anterome- dian and posteromedian pleural sulci of the second and third abdominal somites are shorter than in S. aliaffinis, extending dorsally only to about a third of the height of the somite from the middor- sal line rather than to a fourth, and the pos- teromedian ones are curved anteriorly at their dorsal extremities. Finally, S. martini lacks the conspicuous purplish-brown mark (resembling a longitudinally disposed "9" located posterior to the hepatic sulcus and just ventral to the branchiocar- diac carina) present in S. aliaffinis. Material. — 193 specimens from 41 lots. For list of records see Perez Farfante and Boothe 1981. Sicyonia picta Faxon 1893 Figures 47-52 Sicyonia picta Faxon 1893:210 [syntypes: 4c? 29, MCZ 4639, and 25 29, USNM 21172, off Golfo de Panama (7°40'00"N, 79°17'50"W), 127 fm (232 m), 8 March 1891, Albatross stn 3387; 16 , USNM 21171, off Punta Mariato (7°12'20"N, 80°55'00"W), Panama, 182 fm (333 m), 23 Feb- ruary 1891, Albatross stn 3355]. Faxon 1895:180, pi. 46, fig. 2, 2a-c. H. Milne Edwards and Bouvier 1909:244. De Man 1911: 112. Bayer et al. 1970:A97. Arana Espina and Mendez G. 1978:27, fig. 10-13. Brusca 1980:256. Mendez G. 1981:47, pi. 10, Fig. 83-86. Perez Farfante 1982:372. Eusicyonia picta. Burkenroad 1934a:95, fig. 35, 1934b: 126, 1938:87. Anderson and Lindner 1945:318. Vernacular names: rock shrimp, target shrimp, Japanese shrimp (United States); cacahuete, camaron de piedra, camaron de roca, camaron japones (Mexico). FAQ names: peanut rock shrimp (English), camaron cacahuete (Spanish), boucot cacahouette (French). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing one tooth posterior to level of he- patic spine and raised in high, arched crest behind posterior tooth. Abdomen with tooth on dorsome- dian carina of first somite conspicuously larger than posterior tooth on carapace. Petasma with distal projection of dorsolateral lobule slightly curved mesially, its compressed tip produced dor- sally in strong, hooklike spine. Thelycum with plate of sternite XIV flat or slightly elevated lat- erally; posterior component of median plate flat or slightly raised laterally. Branchiostegite with ocellus consisting of red center svirrounded by yel- low ring. Description. — Body relatively slender (Fig. 47). Carapace sparsely studded with long setae and bearing patches of shorter setae on dorsum; patch also present anteroventral to hepatic sulcus, another elongate obliquely disposed on bran- chiostegite, and others on lateral depression and anteroventral part of sixth abdominal somite. Ab- domen with few small tubercles on first three so- mites, most on row behind posterior sulci. Rostrum short, usually not overreaching distal margin of eye, its length increasing linearly with carapace length (Fig. 48), but proportionately longer in young (0.40-0.25 cl); in males (Fig. 49fi), weakly arched, subhorizontal or upturned, usu- ally not more than 25° but occasionally 30°, deep basally, gently narrowing to slender, short tip (Fig. 49A); in females, nearly straight, raised 25°-40°, deep along almost entire length, slightly narrower and truncate apically; in both sexes armed with two to four dorsal teeth and two or three apical 57 FISHERY BULLETIN: VOL, 83, NO. 1 Figure 47. —Sicyonia picta Faxon 1893. Syntype 5 19 mm cl, off Golfo de Panama. Lateral view. Scale = 5 mm. ones (3 + 3, bmc; 4+3, 19%; 4+2, 22%; 3 + 2, 0.5%; 2 + 3, 0.5% ), dorsal teeth in females often crowded anteriorly with apical ones, about evenly spaced along margin in males. Adrostral carina, subpar- allel and close to ventral margin, extending to base of apical teeth. Carapace with well-marked postrostral carina bearing two teeth: 1) epigastric tooth small, sub- equal to or only slightly larger than first rostral tooth, situated distinctly anterior to hepatic spine, between 0.13 and 0.16 (mean 0.15) cl from orbital margin; and 2) posterior tooth, much larger, four or five times higher, than epigastric, hooklike, its apical portion acutely pointed and strongly curved anteroventrally, situated far posterior to hepatic spine but well in advance of posterior margin of carapace, between 0.60 and 0.68 (mean 0.64) cl from orbital margin. Postrostral carina low an- teriorly, slightly elevated just in front of posterior tooth, and forming high crest descending gently from latter to posterior margin of carapace. Tuft of setae present at anterior base of each tooth. An- tennal spine sharp, projecting from short, low buttress; hepatic spine considerably larger than antennal, arising from moderately raised area, and situated between 0.18 and 0.24 (mean 0.22) cl from orbital margin. Postocular sulcus deep an- teriorly, continuing posteriorly as low groove; 58 B • ■ ?* • • ^ X 1- • ^ • ^ 0» c y^ • £. 4 • V^ • E •^ • • 3 A^ • 1 S^ • M • J • • 0 V* w ^ kr? 2 V 10 15 20 carapace length (mm) 25 Figure A&.— Sicyonia picta. Relationship between rostrum length and carapace length (regression equation, y = 0.51173 + 0.26668X). hepatic sulcus deep; hepatic carina indistinct; branchiocardiac carina recognizable only pos- teriorly, extending for short distance subparallel to slope of posterior tooth, then curving dorsally to posterior margin of carapace. Antennular peduncle with stylocerite produced in long spine, its length 0.75-0.85 distance be- tween lateral base of first antennular article and I PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA mesial base of distolateral spine; latter, slender and sharp, extending to about midlength of second antennular article; antennular fiagella short, me- sial one slightly shorter, 0.20-0.30 cl, than lateral, 0.25-0.35 cl. Scaphocerite extending to midlength of third antennular article or slightly overreaching it. An- tennal flagellum about twice as much as twice length of carapace. Third maxilliped slightly stouter than pereopods. Basis and ischium of first pereopod un- armed. Abdomen with dorsomedian carina extending from first through sixth somites, carina on first produced in strong, anterodorsally directed tooth tapering to sharp apex and considerably larger than posterior tooth on carapace; carina on fifth produced in conspicuous sharp tooth and that on sixth terminating in strong, acute one. Anteroventral margin of pleuron of first abdom- inal somite concave; posteroventral margin of first through fourth somites rounded; anteroventral extremity of pleuron of first through fourth so- mites ending in spine, that of first directed ven- trolaterally, those of second through fourth curved posterolate rally. Pleuron of fifth and sixth somites bearing posteroventral, caudally directed, rela- tively small spine, that of fifth slightly larger than that on sixth. First abdominal somite traversed by anterome- dian pleural sulcus, deep dorsally and disappear- ing at about 0.30 height of somite before reappear- ing ventrally as broad shallow depression merging with united posterior tergal-posteromedian pleural sulci. Second and third somites with an- terior tergal and posterior tergal sulci joining an- terodorsally; anteromedian pleural sulcus short (not reaching dorsally posteromedian pleural) but deep and merging with shallow anteroventral depression, latter setting off subelliptical promi- nence dorsally and low ridge posteriorly; pos- teromedian pleural sulcus long, extending anterodorsally ventral to (not joining) posterior tergal. Fourth and fifth somites with anterior ter- gal sulcus and curved, united posterior tergal- posteromedian pleural sulci merging. Sixth so- mite with arched posteromedian pleural sulcus and longitudinal rib situated along base of mid- dorsal carina and delimited ventrally by deep de- pression lying just dorsal to weak cicatrix. Telson with median sulcus deep anteriorly, in- creasingly shallow posteriorly, its terminal por- tion elongate and sharp, and bearing pair of small fixed subterminal spines. Rami of uropod sub- equal in length, falling slightly short of or barely overreaching apex of telson. Petasma (Fig. 50A, B) with rigid distal projec- tion of dorsolateral lobule only slightly curved me- sially, raised proximodorsally in rounded promi- nence and compressed distally, its tip with ventral extremity rounded and dorsal extremity produced in strong, sharp spine directed dorsally. Fleshy distal projection of ventrolateral lobule curving laterally, roughly sickle shaped in outline, with apex directed proximoventrally. Petasmal endopods coupled in males as small as 6.7 mm cl, about 27 mm tl, but may not be joined in individuals as much as 9 mm cl, about 34 mm tl. Figure 49.— Skyonia picta. A, c? 16 mm cl, NW of Isla Monserrate, Baja California Sur, Mexico. Lateral view of carapace. Scale = 5 mm. B, syntjrpe 6 15.5 mm cl. Golfo de Panama. Lateral view of dorsal part of cara- pace. Scale = 2 mm. 59 FISHERY BULLETIN: VOL. 83, NO. 1 \gP Figure 50. — Sicyonia picta, syntype 6 15.5 mm cl, off Golfo de Panama. A, Petasma dorsal view; fi, ventral view of same; C, right appendix masculina, dorsolateral view. Scales = 1 mm. Appendix masculina as illustrated in Figure 50C. Thelycum (Fig. 51) with plate of sternite XIV flat or slightly elevated laterally, inclined toward broad median depression, and bordered anteriorly and laterally by narrow, sometimes thickened, flange. Median plate of sternite XIII flask-shaped in outline or subtriangular, tapering anteriorly into long, slender spine reaching between proxi- mal end and midlength of basis of anteriorly ex- tended second pereopods; plate at level of fourth pereopods excavate and constricted by pair of shal- low, widely separated lateral incisions; posterior component of median plate, often convex laterally, with rather deep median emargination. Paired short spines projecting from posterior margin of sternite XI. Posterior thoracic ridge narrow, with well-marked anteromedian margin but flush with lateral parts of plate of sternite XIV. The smallest impregnated females encountered have a carapace of 7 mm, about 28 mm tl. Color. — Mendez G. (1981) described recently caught specimens as follows: body light red or orange red, with white areas on ventral part of abdominal somites; carapace marked by conspicu- ous ocellus consisting of red center surrounded by yellow ring. Diffuse dark spot on lateral ramus of 60 Figure 51.— Sicyonia picta, syntype 9 28 mm cl, off Golfo de Panama. Thelycum. Scale = 1 mm. PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA uropod situated distomesially according to her figure 84. Antennae with alternating light and dark red bands. In Formalin, body turns darker red with ring around ocellus, garnet. Faxon (1893, 1895) noted that in specimens preserved in alcohol there is a dark ring on the posterior part of the branchial region and traces of color are present on margins of rostrum, dorsal carinae, and append- ages. In most specimens preserved in either agent examined by me, the color pattern described above is still distinguishable. Maximum size. — Males 17.5 mm cl, about 70 mm tl; females 24 mm cl, 87.9 mm tl (Faxon 1893 and Arana Espina and Mendez G. 1978; corroborated by me). Geographic and bathymetric ranges. — Bahia Magdalena (24°33'00"N, 112°00'30"W) to south- ern tip of Baja California Sur, Mexico, throughout the Gulf of California and southward to northeast of Isla Maria Madre (22°00'N, 106°16'W), Nayarit, Mexico; also from Champerico (13°55'36"N, 92°02'30"W), Guatemala, to Islas Lobos de Afuera (06°45'S, 80°45'W), Peru (Fig. 52). It occurs at depths between 16 and 400 m (shallowest cited by Arana Espina and Mendez G. 1978), but most of the recorded depths are <150 m. It occupies a large variety of bottom types: sand, shell, sand and shell, sand and mud, shell and mud, rock and mud, green, grey and brown mud, broken gravel and shells, and a mixture of mud, rocks, and coralline detritus. Discussion. — Sicyonia picta is most similar to S. disdorsalis; both are of moderate size and in addi- tion bear a small epigastric tooth, a large posterior tooth on the postrostral carina, and a strongly developed one on the first abdominal somite. These shrimps can be readily separated by their color pattern and a number of morphological characters. In S. disdorsalis an ocellus is lacking on the pos- terior part of the branchiostegite, the rostrum is slender throughout its entire length, and less ele- vated than in S. picta, its inclination not exceed- ing 20°; the epigastric and posterior teeth on the postrostral carina are situated closer to the orbital margin, between 0.06 and 0.12 (mean 0.10) cl and 0.55 and 0.65 (mean 0.60) cl, respectively; and the posterior tooth rises from a uniformly low post- rostral carina. The two species also differ in sculpture of the abdomen. In S. disdorsalis the first abdominal somite is traversed by a short anteromedian sul- cus which is not represented ventrally by a depres- sion; the posterior tergal and posteromedian pleural sulci of the second and third somites are coalescent; the anteroventral extremities of the second through fourth are unarmed or are pro- duced in a small, ventrally projecting spine; and the posteroventral extremities of first through fourth somites are angular, that of the fourth bear- ing a well-developed spine, and that of the fifth, an extremely long one (instead of small as in S. picta) in adults. In both species the petasma and the thelycum also exhibit distinctive features. In S. disdorsalis the distal projection of the distolateral lobule of the petasma terminates in an acute tip rather than being compressed laterally and produced in a • S. picta A S. disdorsalis * S. in gent is FIGURE 52.— Geographic distribution of Sicyonia picta, S. dis- dorsalis, andS. ingentis. 61 FISHERY BULLETIN: VOL. 83, NO. 1 dorsally directed salient as it is in S. picta. Fur- thermore, the distal projection of the ventrolateral lobule in S. disdorsalis is laminar, bifurcate later- ally, and bears a proximal plate bordered by a transverse rib; in S. picta, it is roughly sickle shaped in outline, tapering laterally to a sharp proximoventrally directed apex. Finally, the thelycal plate of sternite XIV in S. disdorsalis is raised in a low but well-defined pair of lateral protuberances instead of being flat or slightly raised laterally as it is in S. picta. In the field, S. picta may be distinguished from its sympatric congeners by the striking yellow ocellus with a red center located on the bran- chiostegite. Burkenroad (1938) discussed in detail the fea- tures that distinguish S. picta from the western Atlantic S. stimpsoni Bouvier 1905. Actually, fea- tures cited by him for S. stimpsoni also apply to S. burkenroadi, another western Atlantic species which was not recognized until described by Cobb in 1971. S icy onia picta differs from S. stimpsoni, but resembles S. burkenroadi, in bearing strongly curved spines on the anteroventral angles of the second through fourth abdominal pleura. It, in turn, can be separated from .S. burkenroadi, but resembles S. stimpsoni, in lacking a posterior tooth on the tergal carina of the fourth abdominal somite. Remarks. — Arana Espina and Mendez G. (1978) presented an illustration (Fig. 11) in which the posterior tergal and posteromedian sulci of the second and third abdominal somites appear co- alescent. The disposition of the posteromedian sul- cus seems to be in error because, as stated above, the two sulci in this shrimp do not merge; instead the posteromedian one extends anterodorsally, ventral to the posterior tergal sulcus. In addition to citing many new localities, this paper contains the first records of the species from the ocean side of Baja California Sur, Mexico, as far north as Bahia Magdalena. Material.— 602 specimens from 61 lots. Mexico — Baja California Sur: 19, AHF, 5.5 km W of mouth of Bahia Magdalena, 64 m, 8 March 1949, Velero IV. U, USNM, off Isla Santa Mar- garita, 86 m, 8 April 1889, Albatross stn 3039. 16 29, SIO, off Punta Marquez, 64 m, 9 November 1964. 29, SIO, Bahia de la Paz, 82-119 m, 12 January 1968, Thomas Washington . 26 69 , SIO, Bahia de la Paz, 119-128 m, 24 July 1965, R. Rosenblatt. 1<5, AHF, entrance to Bahia Agua 62 Verde, 42-48 m, 17 March 1949, Velero IV. 66 119, SIO, NW of Isla Monserrate, 170-192 m, 11 July 1965, Horizon. 29, AHF, NW of Isla Danzante Primero, 44-73 m, 18 March 1949, Velero IV. 16 , YPM, Bahia de Santa Ines, 101 m, 17 March 1926, Pawnee. 49, SIO, off Santa Rosalia, 35-26 m, 25 March 1960, R. Parker Baja California Norte (all in Gulf of California): 19, YPM, Bahia de los Angeles, 31-42 m, 13 May 1926, Pawnee. 3d 19, AHF, Puerto Refugio, Isla Angel de la Guarda, 143-165 m, 28 January 1940. 39 , AHF, N of Isla Angel de la Guarda, 104 m, 28 January 1940. 66 119, SIO, SE of San Felipe, 75-86 m, 19 January 1968, Washington. 16, YPM, Bahia San Luis Gonzaga, 17 May 1916, Pawnee. 16 19, USNM, off Punta San Fermin, 55 m, 27 March 1889, Alba- tross stn 3035. 19 , SIO, SE of San Felipe, 120 m, 19 January 1968, Washington. Sonora: 26 29, USNM, off Cabo Tepoca, 65 m, 24 March 1889, Albatross stn 3018. 266 299, USNM, SW of Cabo de Lobos, 139 m, 24 March 1889, Albatross stn 3016. 16 , USNM, NW of Isla Tiburon, 265 m, 24 March 1889, Albatross stn 3015. 4(5 19, AHF, 3 km W of Tastiota, 60 m, 21 December 1978, A. Kerotitch. Sinaloa: 3d 29 , USNM, off Punta Santo Domingo, 135 m, 10 April 1889, Albatross stn 3043. 19 , USNM, off Peninsula de Quevedo, 67 m, 30 March 1978, Toral Garcia. Nayarit: 26 49, SIO, NE of Isla Maria Madre, Islas Tres Marias, 82-88 m, 30 March 1973, Agassiz. Guatemala— 19, SIO, Champerico, 91-104 m, 13 April 1973, Agassiz. Nicaragua— 29, SIO, off N of Nicaragua, 53-59 m, 17 April 1973, C. Hubbs and S. Luke. Costa Rica— Id 19 , USNM, Golfo del Papagallo, 2 April 1978, D. Hedgecock. Id, SIO, Punta Guiones, 104 m, 19 April 1973, C. Hubbs and S. Luke. 1505 1509 , SIO, Golfo de Nicoya, 86 m, 22 April 1973, A^ass/e. 6d 19, SIO, off Cabo Blanco, 60 m, 18 April 1973, Agassiz. 19, UCR, off Cabo Blanco, 249 m, 28 April 1973, Enriqueta. 16 149, SIO, off Cabo Blanco, 137-144 m, 19 April 1973, C. Hubbs and S. Luke. 19 , USNM, off Cabo Blanco, 247 m, 27 April 1973, Enriqueta. Panama— Id 19, AHF, Islas Secas, 46-48 m, 27 March 1939. 2d 69 , AHF, off Isla Medidor, 55-64 m, 28 March 1939, Velero III. 26 89, UR 25 km S of Isla Cebaco, 256 m, 8 August 1972, Canopus. 16, syntype, USNM, off Punta Mariato, 333 m, 23 February 1891, Albatross stn 3355. Id 19 , USNM, E of Isla Iguana, 79-77 m, 2 May 1967, Pillsbury stn 502. 39, USNM, NE of Isla Iguana, 79-77 m, 4 May 1967, Pillsbury stn 515. 4d 29, MCZ, and 2d 29, USNM, syntypes. PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA off Golfo de Panama, 232 m, 8 March 1891, Alba- tross stn 3387. IS 29 , USNM, S of Isla San Jose, 84 m, 6 May 1967, Pillsbury stn 529. bS 69, USNM, S of Isla San Jose, 99 m, 7 May 1967, Pillsbury stn 553. 19 , USNM, SE of Isla San Jose, 68 m, 7 May 1967, Pillsbury stn 555. 26 39, USNM, SE of Isla San Jose, 60 m, 5 March 1888, Albatross stn 2797. 26 , USNM, S of Isla del Rey, 44-47 m, 7 May 1967, Pillsbury stn 551. 16 19, USNM, S of Isla del Rey, 59 m, 8 May 1967, Pillsbury stn 556. 16 59 , USNM, SW of Golfo de San Miguel, 64-60 m, 7 May 1967, Pillsbury stn 550. Colombia— Call: 2c?, USNM, off Bahia de Buenaventura, 80 m, 16 September 1966, Anton Bruun, 18B, stn 783. Ecuador — Manabi : many 6 and 9 , USNM, off Cabo Pasado, 93 m, 12 September 1966, Anton Bruun, 18B, stn 778. 16 39, USNM, off Bahia de Manta, 120-150 m, 12 September 1966, Anton Bruun, 18B stn 776. 36 39, USNM, off Cabo San Lorenzo, 185 m, 12 September 1966, Anton Bruun, 18B, stn 775. 16 19 , AHF, off Isla La Plata, 82-101 m, 10 February 1934. El Oro: 26 , USNM, SW of Isla Santa Clara, depth unrecorded, 10 Sep- tember 1966, Anton Bruun, 18B, stn 769-D. 27d 679, USNM, SW of Puerto Bolivar, 80 m, 10 Sep- tember 1966, Anton Bruun, 18B, stn 769. Peru— Tumbes: lid 89, USNM, off Casitas, 90 m, 8 September 1966, Anton Bruun, 18B, stn 764. Piura: 46 49, USNM, Bahia de Paita, 70-69 m, 8 September 1966, Anton Bruun, 18B, stn 762-A. 26 49 , USNM, Bahia de Paita, 118-133 m, 2 June 1966, Anton Bruun, 16, stn 625-A. 16, USNM, SW of Isla Foca, 120 m, 7 September 1966, Anton Bruun, 18B, stn 761. 46 109, USNM, off Punta Negra, 100 m, 4 June 1966, Anton Bruun, 16, stn 631- A. Lambayeque: Id 19 , IMARPE, Islas Lobos de Afuera, 360-400 m, 1977, R. Mar- quina. Sicyonia disdorsalis (Burkenroad 1934) Figures 52-56 Eusicyonia disdorsalis Burkenroad 1934a: 96, fig. 25, 36 [syntypes: Id 19, YPM 4391, 5d 69 (not 4d 79 as originally cited), YPM 5075, and Id 19, YPM 4391, Pearl Islands (Archipielago de las Perlas), 8°29'40"N, 78°52'30"W, Golfo de Panama, 19-24 fm (35-44 m), 31 March 1926, Pawnee; 56 59 (not 6d 49), YPM 5079, and 19, YPM 5078, Golfo de Panama, 1868, F H. Brad- ley Id 39, YPM 5076, and Id, YPM 5077, west coast of Central America, 1872, Capt. Dow]. Burkenroad 1938:87. Anderson and Lindner 1945:318. Sicyonia disdorsalis. Chirichigno Fonseca 1970:7, fig. 4. Bayer et al. 1970:A97. Del Solar et al. 1970:18. Rosales Juarez 1976:41, pi. 1, fig. 3. Rodriguez de la Cruz 1977:11. Arana Espina and Mendez G. 1978:29, fig. 14-17. Brusca 1980:256. Paul and Hendrickx 1980: 110. Sosa Hernandez et al. 1980:14. Mendez G. 1981:48, pi. 10, fig. 87-90. Per- ez Farfante 1982:370. Vernacular names: rock shrimp (United States); camaron conchiduro (Mexico, Panama); cama- ron duro (Ecuador, Peru); langostino cascara dura, camaron cascara dura (Peru). FAO names: keeled rock shrimp (English), camaron carenado (Spanish), boucot carene (French). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing one tooth posterior to level of he- patic spine and low throughout its entire length, not raised in crest behind posterior tooth. Abdo- men with tooth on dorsomedian carina of first so- mite considerably larger than posterior tooth on carapace. Petasma with distal projection of dor- solateral lobule curved mesially, tapering to apex, and lacking filament. Thelycum with plate of sternite XIV raised in pair of lateral bulges; pos- terior component of median plate flat or slightly raised laterally. Branchiostegite lacking large spot or ocellus. Description. — Body relatively slender (Fig. 53). Carapace bearing patches of short setae on dor- sum, antero ventral to hepatic spine, and ventral to hepatic sulcus; patches of setae also present on dorsal extremity of abdominal sulci and in depres- sion of sixth abdominal somite. Abdomen with numerous tubercles on first three somites. Rostrum relatively short, rarely overreaching eye, its length increasing linearly with carapace length (Fig. 54) to about 18 mm cl, then increasing little, not surpassing 6.2 mm ^proportional length decreasing with increasing size from as much as 0.36 to as little as 0.21 cl); slender but occasionally moderately deep; in males subhorizontal with tip strongly decurved, in females upturned as much as 20° with tip slightly decurved; armed with three, occasionally two, dorsal teeth and two or three apical teeth (2+2, 2%, 3+2, 92%, 3+3, 6%); 63 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 53 . —Sicyoma disdorsalis (Burkenroad), ? 25 mm cl, Golfo de Fonseca, El Salvador. Lateral view. Scale = 10 mm. ventral apical tooth smaller than dorsal and placed almost at same level or, more often, pos- terior to it. First rostral tooth located well in ad- vance of orbital margin, between 0.18 and 0.33 (mean 0.25) rl; second tooth from 0.45 to 0.70 (mean 0.58) rl; and third from 0.75 to 0.96 (mean 0.81) rl. Adrostral carina extending to near tip, relatively far from ventral margin, often slightly arched either along middle or less often anteriorly, and occasionally directed anterodorsally. Carapace with postrostral carina well marked but low throughout its entire length, bearing two teeth: 1) epigastric tooth small, subequal to or slightly larger than first rostral tooth and situated well in advance of hepatic spine, between 0.06 and 0.12 (mean 0.10) cl from orbital margin; and 2) posterior tooth, as large as or larger, sometimes as much as three times higher, than epigastric, acutely pointed, strongly inclined anteriorly, and placed considerably in advance of posterior mar- gin of carapace, between 0.55 and 0.65 (mean 0.60) cl from orbital margin (both teeth farther anterior in large individuals than in young). Tuft of setae present at anterior base of each tooth. Antennal spine moderately long, sharp, buttressed; hepatic spine long, conspicuously larger than antennal, projecting from raised area, and situated between 0.19 and 0.24 (mean 0.22) cl from orbital margin. 64 12 18 carapace length (mm) 24 30 FIGURE 54.— Sicyonia disdorsalis. Relationship between ros- trum length and carapace length (regression equation for speci- mens with about 18 mm cl or less, y = -0.03933 + 0.30998x; regression equation for those larger, > = 2.33498 + 0.14502x). Postocular sulcus deep anteriorly, continuing posteriorly as very shallow arched groove; hepatic sulcus subhorizontal; hepatic carina indistinct; branchiocardiac carina weak. Antennular peduncle with stylocerite produced in long, sharp spine, its length 0.75-0.85 distance between lateral base of first antennular article and ,1 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA mesial base of distolateral spine; latter long, reaching as far as distal 0.25 of second antennular article; antennular flagella short, mesial one, more slender and shorter than lateral, 0.20-0.30 as long as carapace; lateral flagellum 0.25-0.35 as long. Scaphocerite surpassing antennular peduncle by no more than 0.1 its own length; lateral rib ending distally in long, acute spine conspicuously overreaching margin of lamella. Antennal flagel- lum about twice as long as carapace. Third maxilliped stouter than pereopods. Basis and ischium of first pereopod unarmed. Abdomen with dorsomedian carina extending from first through sixth somites, carina on first produced in strong, anteriorly inclined tooth, ta- pering suddenly near end to minute, sharp apex (usually broken and thus appearing blunt); tooth considerably larger than posterior one on car- apace; carina on fifth somite abruptly truncate or produced in sharp tooth posteriorly, and that on sixth terminating in strong, acute posterior tooth. Antero ventral margin of ple\iron on first abdom- inal somite concave in adults, sometimes nearly straight in juveniles, its extremity ending in strong spine, that of second through fourth (in adults) often projecting slightly or forming ven- trally pointed spine; posteroventral extremity of third through sixth somites sharply angular, that of third often, but in last three always bearing caudally directed spine, that of fifth and sixth largest and smallest, respectively; additional spine occasionally present immediately dorsal to posteroventral one on fourth somite. First abdominal somite marked with short, deep anteromedian pleural sulcus and long, united posterior tergal-posteromedian pleural sulci; short, longitudinal ridge extending between an- teromedian pleural and posterior sulcus. Second and third somites with anterior tergal sulcus join- ing united posterior tergal-posteromedian pleural sulci dorsally, and with anteromedian pleural sul- cus represented by shallow depression setting off elevation at dorsal extremity. Fourth and fifth so- mites bearing curved, united posterior tergal- posteromedian pleural sulci; sometimes fourth also with faint anterior tergal sulcus. Sixth somite often marked by weak, sometimes indistinct, arched, posteromedian sulcus and bearing con- spicuous cicatrix frequently divided in two. Telson with median sulcus well defined only along anterior 0.65 of its length and armed with pair of minute, fixed, subterminal spines; latter clearly developed in juveniles but vestigial or lack- ing in adults. Rami of uropod subequal in length, falling slightly short of or barely overreaching apex of telson. Posterior spine on first abdominal sternite with wide base and usually concave but sometimes straight lateral margins. Petasma (Fig. 55A, B) with cornified distal pro- jection of dorsolateral lobule raised in prox- FIGURE 55. Sicyonia disdorsalis , 6 17 mm cl, off Balboa, Panama. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scales = 1 mm. 65 FISHERY BULLETIN: VOL. 83, NO. 1 imodorsal rounded prominence and ending in acute tip curved mesially. Distal projection of ven- trolateral lobule, falling short of projection of dor- solateral lobule, laminar, bifurcate laterally, and thickened proximally forming plate bordered dis- tally by transverse rib; latter supporting long proximal salient of bifurcation; terminal part of projection truncate and curved dorsally. Petasmal endopods joined in males as small as 3 mm cl, about 13 mm tl, but may be unjoined in individuals with as much as 8.4 mm cl, about 21 mm tl. Appendix masculina as illustrated in Figure 55C. Thelycum (Fig. 56) with plate of sternite XIV bearing pair of low protuberances bordered later- ally (in adults) by narrow flanges, and separated by moderately deep median depression. Median plate of sternite XIII flask-shaped in outline, ta- pering into long, slender spine reaching between base and about midlength of basis of extended second pereopod; plate set off from posterior com- ponent by shallow incisions, flat or concave pos- terolaterally, and with broad median depres- sion (broader than areas and incisions flanking it); posterior component of median plate flat or slightly raised laterally, with posteromedian mar- gin straight or convex. Paired short spines project- ing anteroventrally from posterior margin of ster- nite XI, spines broad basally, sharp and sometimes produced in fine needle apically. Posterior thoracic ridge narrow, with well-marked anteromedian margin. The smallest impregnated females encountered have a carapace of 5.3 mm, about 25 mm tl. Color. — Arana Espina and Mendez G. (1978) de- scribed specimens from the waters of Peru as fol- lows: Dorsum greenish gray, lighter-gray to pink laterally Antennae red. Pereopods and pleopods pink; merus of third maxilliped and pereopods with red and yellow bands (toward distal end ac- cording to their figure 15). Subdistal striking or- namentation on lateral ramus of uropod consisting of oval deep blue blotch bordered in yellow. In contrast, Sosa Hernandes et al. (1980) found that the specimens from southeast of Salina Cruz, Golfo de Tehuantepec, Mexico, were cream with orange hues. My observations, based on a large number of live specimens taken off Panama Viejo, Panama, indi- cate a color pattern much, but not exactly, like that noted by Arana Espina and Mendez G. Dorsum of carapace gray with broad transverse dark 66 Figure 56. — Sicyonia disdorsalis, 9 16 mm cl, off Punta Calabazo, Panama. Thelycum. Scale = 2 mm. purplish band extending across epigastric tooth and produced anteriorly in roughly pentagonal spot reaching base of first rostral tooth. Bran- chiostegite varying from dark purplish pink (in most specimens) to milky white, sometimes with grayish white longitudinal stripe along junction of branchiostegite and dorsum. Abdomen gray with middorsal carina banded: transverse dark gray ones alternating with buff bands; large tooth pro- jecting from carina on first somite purplish pink. Anterior margin of pleuron of first three somites bordered by white vertical stripe and posteroven- tral areas of third and fourth somites also white. Tergum of fifth somite bearing purplish pink V-shaped (vertex anterior) marking posteriorly. Telson and uropod gray with densely set yellow chromatophores. Lateral ramus of uropod bearing large garnet marking subdistally preceded by yel- low patch, and contiguous to buff dot lying against its mesial extremity. Antenna dark garnet. Third maxilliped and pereopods pink, former with two contiguous transverse bands, orange red one on PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA distal third of merus, followed by bright yellow band crossing merus and proximal part of carpus; fourth and fifth pereopods marked with orange red band flanked by yellow ones extending across merus and proximal part of carpus. Pleopods mostly pink, lateroventral part of basis white. Maximum size. — Males 21 mm cl, 88.8 mm tl; females 28 mm cl, 98.7 mm tl (Arana Espina and Mendez G. 1978). Largest specimens examined by me: males 20 mm cl, females 25.7 mm cl. Geographic and bathymetric ranges. — Bahia Santa Maria (24°42'48"N, 112°13'54"W), to the tip of Baja California Sur, Mexico, and from southern Sonora, Gulf of California, southward to Pisco (13°55'S), Peru(Fig. 52). This species has been found between 5 and 139-93 m, but rarely in water deeper than 65 m; contrary to most of its congeners from the Ameri- can Pacific, it does not seem to range beyond the continental shelf. It occurs on bottoms of mud, detritus, fine sand, and rock and coral, seemingly with no preference. Abundant in the southeastern part of the Gulf of California (Paul and Hendrickx 1980), it has not been reported off northern Sonora or along the east coast of Baja California; the present records from these latter areas are the first for the species north of Bahia San Lucas, a locality cited by Bur- kenroad (1938). The southern limit, Pisco, Peru, given here is also the first report of the presence of this shrimp in waters south of San Lorenzo (12°0.5'S), Peru, the southernmost locality cited by Arana Espina and Mendez G. (1978). The record from Pisco is based on specimens collected by M. Mendez G. and J. Zeballos, at 5.5-13 m, on 5 November 1983 (Matilde Mendez G. footnote 3). Discussion. — Burkenroad (1938) was the first to point out some of the differences that separate this species from its closest relative S. ingentis. Sicyonia disdorsalis can be distinguished from the latter by having 1) a sparsely setose carapace, 2) a less elevated postrostral carina, 3) a weak, al- most indistinct branchiocardiac carina, 4) a pos- teriorly truncate carina on the fifth abdominal somite that is sometimes produced in a spine, 5) an anteromedian pleural sulcus on the first abdomi- nal somite which ends abruptly far from the ven- tral margin, and does not continue ventrally as a shallow depression, 6) a strong spine on the an- teroventral extremity of the pleuron of the first abdominal somite, 7) angular posteroventral pleural margins on the first two abdominal so- mites, 8) minute telsonic spines in juveniles and vestigial or indistinct ones in adults, and 9) short uropodal rami that fall short of or barely surpass the apex of the telson. Various features of the genitalia also allow the separation of iS. disdorsalis from S. ingentis. The distal projection of the dorsolateral lobule of the petasma is curved distomesially instead of extend- ing distolaterally, and is not produced in a short, apical filament; and the distal projection of the ventrolateral lobule does not extend so far distally as the projection of the dorsolateral lobule, is bifurcate laterally, and bears a conspicuous trans- verse rib. In the thelycum, the breadth of the fiat or concave posterolateral areas of the median plate, as well as the depth of the delimiting inci- sions are much less than the depressed area be- tween them. Furthermore, the posterior compo- nent of the median plate is sometimes slightly raised posterolaterally but not forming well- defined lateral bosses traversed by a suture as in S. ingentis. The differences between S. disdorsalis and the geminate western Atlantic S. dorsalis Kingsley 1878, were discussed in detail by Burkenroad (1934a). Commercial importance. — Throughout its range, S. disdorsalis is present in the commercial catches of other penaeoid shrimps. It was recorded by Rosales Juarez (1976) in those off the coast of Sinaloa, in the Gulf of California, and more re- cently, on the basis of its abundance in the shrimp bycatch from the waters off Sinaloa and Nayarit, Paul and Hendrickx (1980) suggested that this shrimp has a possible commercial value in that area. Arana Espina and Mendez G. (1978) recorded that in 1977 it made up to 5.8% of the total catches made in northern Peru, a notable increase from that of previous years in which it constituted <0.5%. This species is considered by them to have a significant economic potential. Material. — 1054 specimens from 85 lots. Mexico— Baja California Sur: 16 29, SIO, Bahia Santa Maria, 0-37 m, 8 December 1962, H. C. Perkins. 16, SIO, NW of Punta Marquez, 37 m, 4 December 1962, H. C. Perkins and R. Wisner 16 , SIO, NW of Punta Marquez, 18 m, 4 December 1962, H. C. Perkins. 16 29, YPM, Bahia San Lucas, 5.5 m, 7 May 1936, Zaca stn 135D-20. 26, YPM, Bahia San Lucas, 5-17 m, 7 May 1936, Zaca stn 135D-18-D19. Sonora: 7c? 29, USNM, Bahia de Lobos (boca sur), 30 m, 18 67 FISHERY BULLETIN; VOL. 83, NO. 1 July 1979, F. Paredes M. Sinaloa: 216 169, SIO, Isla Altamura, 21-31 m, 26 May 1965, £/ Golfo II stn 50-6. 29, USNM, off San Ignacio, 25 May 1962, R. Bush M. 19 , USNM, N of Mazatlan, 3.5 km off Marmol, 12 January 1964, A. Villania and E. Chavez. 3d 89, SIO, off Boca Teacapan, 55 m, 25 August 1961, F. H. Berry Nayarit: 89, SIO, W of Laguna de Agua Brava, 20 m, 24 August 1961, H. DeWitt. 26 79, SIO, SW of Laguna de Agua Brava, 15 m, 24 August 1961, H. DeWitt and H. C. Perkins. Id 159, SIO, NE of Isla Maria Madre, 51 m, 31 March 1973, Agassiz. 29, SIO, NW of mouth of Rio Grande de Soledad, 38-39 m, 24 August 1961, H. C. Perkins and H. De- Witt. 30d 309 , SIO, Bahia de Banderas, 28-33 m, 2 June 1965, El Golfo II stn BT-150. 12d 149 , SIO, Bahia de Banderas, 46-55 m, 21 August 1961, F. H. Berry Id 19, SIO, Bahia de Banderas, 5-9 m, 19 August 1961, F H. Berry Jalisco: 29 , USNM, Puerto Vallarta, 13 April 1937. lid 149, SIO, Bahia Chamela, 15-18 m, 2 April 1973, Agas- siz. 66 159, SIO, Bahia Chamela, 27-18 m, 2 April 1973, Agassiz. 16 39, AHF, Bahia Tenaca- tita, 4-15 m, 8 May 1939. Michoacan: 19, CAS, 14.5 km SE of Punta San Telmo (off Ma- ruata), 17 July 1932, Zaca. 16 19, SIO, Punta Lizardo, 22-24 m, 4 April 1973, Agassiz. 21d 25j, SIO, Punta Lizardo, 37-38 m, 4 April 1973, Agassiz. Guerrero: 39, CAS, 6.5 km SE of entrance of Bahia de Acapulco, 27 m, 5 April 1932, Zaca. Oaxaca: Id 19, USNM, 24 km off Puerto Angel, 84-57 m, 13 July 1963, I. Mayes A. 8d 49, SIO, Golfo de Tehuantepec, 55 m, 6 June 1965, El Golfo II, stn BT-162. 5d 49, SIO, SW of Santiago Astata, 54 m, 6 June 1965, El Golfo II, stn BT-162. Id 39, USNM, 16 km W of Ayutla lighthouse, 54 m, 15 June 1963, I. Mayes A. 7d 30t, SIO, off Salina Cruz, 44 m, 7 June 1965, El Golfo II. 59 , USNM, Salina Cruz, 64 m, 23 August 1963, I. Mayes A. 4d 29, SIO, off Salina Cruz, 31-35 m, 8 July 1963, D. Dock- ins. 19, SIO, off Salina Cruz, 49-73 m, 8 July 1963, D. Dockins. 44d 509, SIO, Golfo de Tehuantepec, 22 m, 10 April 1973, Agassiz. 56 59, INP, off Tangola, 68 m, 10 July 1963, 1. Mayes A. 5d , USNM, off Tangola, 139-93 m, 13 July 1963, 1. Mayes A. Guatemala— 1j, AHF, San Jose, 4-9 m, 23 March 1939. El Salvador— 3d 89, SIO, Golfo de Fonseca, 18 m, 17 April 1973, C. Hubbs and S. Luke. 6d 49, SIO, Golfo de Fonseca, 18 m, 17 April 1973, C. Hubbs and S. Luke. 5d 149, SIO, Golfo de Fon- seca, 24-29 m, 17 April 1973, Agassiz. 68 Nicaragua — Id 59, USNM, off northern Nica- ragua, 53-59 m, 17 April 1973, C. Hubbs and S. Luke. Costa Rica— 39, USNM, 3 km off Rio Savegre, Puntarenas, 24 m, 2 December 1981, M. Hat- ziolos. lOd 149 , SIO, Cabo Blanco, 60 m, 18 April 1973, Agassiz. 56 159, SIO, Cabo Blanco, 60 m, 18 April 1973, Agassiz. 16 29, AHF, Golfo de Nicoya, about 1 km of east end of Islas Negritos Afuera, 64 m, 29 June 1973, Velero stn 19132. 30d 309 , USNM, Golfo de Nicoya, 31 m, 22 April 1973, C. Hubbs and S. Luke. 40d 409, SIO, Golfo de Nicoya, 31 m, 22 April 1973, Agassiz. Panama— 19, USNM, off Bocas del Toro, 91-97 m, 26 January 1971, Pillsbury stn 1313. 5d 39, SIO, Isla Cavada, Islas Secas, 40 m, 23 September 1970, W Newman, T Dana, S. Luke. 4d 49, USNM, S of Rio Hato, 17 m, 1/2 May 1967, Pillsbury stn 488. 19, USNM, Bahia de Parita, 22-18 m, 2 May 1967, Pillsbury stn 490." 19, USNM, E of Chitre, 20 m, 2 May 1967, Pillsbury stn 491. 2d 49, USNM, E of Chitre, 18-16 m, 2 May 1967, Pillsbury stn 492. 6d 69, USNM, N of Isla Iguana, 37-33 m, 2 May 1967, Pillsbury stn 493. Id , USNM, southern end of Bahia Limon, 3 m, 23 July 1966, Pillsbury stn 449. 5d 59, USNM, off Punta Calabazo, 20 m, 1 May 1967, Pillsbury stn 486.. 2d , USNM, off Rio Hato, 15 m, 1 May 1967, Pillsbury stn 485. Id, USNM, En- senada de Chame, Shimada stn 48. 2d 29, USNM, S of Isla Bona, 31-26 m, 1 May 1967, Pillsbury stn 484. 18d 149, USNM, E of Punta Chame, 22 m, 1 May 1967, Pillsbury stn 483. 2d 19, USNM, off Balboa, surface, 9/10 May 1967, Pillsbury stn 564. 20d 209, USNM, off Panama Viejo, 4 m, 23 February 1973, I. Perez Far- fante. 30d , USNM, off Juan Diaz, 12-22 m, 15 February 1973, Patricia. Id 79, USNM, Juan Diaz, 5 m, 4 February 1969, L. G. Abele. Id 19, USNM, W of Punta Brujas, 18 m, 6 May 1967, Pillsbury stn 536. Id 19, syntypes, YPM, Ar- chipielago de las Perlas, 35-44 m, 31 March 1926, Pawnee. 56 69, syntypes, YPM, Archipielago de las Perlas, 35-44 m, 31 March 1926, Pawnee. 56 59, syntypes, YPM, Golfo de Panama, 1868, F H. Bradley. 19, syntype, YPM, Golfo de Panama, 1868, F H. Bradley Id 39, syntypes, YPM, W coast of Central America, 1872, Capt. Dow. Id 39, syntypes, YPM, W coast of Central America, 1872, Capt. Dow. Colombia— 19, USNM, Bahia Humboldt, 20 April 1967, Shimada stn 76B, haul 1. 29, USNM, Bahia Humboldt, 20 April 1967, Shimada stn 76B haul 2. 69, USNM, off Timbiqui, Cauca, PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA 38-35 m, 16 September 1966, Anton Bruun 18B stn 785. Ecuador— 19, USNM, Golfo de Guayaquil, 32 m, 11 September 1966, Anton Bruun 18B stn 772. Peru— 13d 229, USNM, off Tumbes, 13 m, 10 September 1966, Anton Bruun 18B stn 768. 169, USNM, off Caleta Cruz, Tumbes, November 1979, Promaresa. 16d, USNM, off Caleta Cruz, Tumbes, November 1979, Promaresa. 16d, USNM, off Caleta Cruz, Tumbes, November 1979, Promaresa. 169, USNM, off Caleta Cruz, Tumbes, November 1979, Promaresa. 69, USNM, off Negritos, 16 m, 2 June 1966, Anton Bruun 16 stn 624-B. 26 19, USNM, Paita, 1969, J. Sanchez and E. Valdivia. 29, USNM, NW of Paita, 40 m, 1977. Sicyonia ingentis (Burkenroad 1938) Figures 52, 57-60 Eusicyonia ingentis Burkenroad 1938:88, fig. 31-34 [holotype: 6, AMNH 12388; type-locality: off east coast of Cedros Island (Isla Cedros), 28°05'N, 115°09'W, Baja California, Mexico, 38 fm (69 m), 27 March 1936, Zaca stn 127D- 1]. Anderson and Lindner 1945:318. Feinberg 1971:6. Frey 1971:16. Sicyonia ingentis. Parker 1964:162. Carlisle 1969:239. Longhurst 1970:272. Word and Charwat 1976:19, 3 fig. Holthuis 1980: 61. Wicksten 1980:360. Perez Farfante and Boothe 1981:424. Perez Farfante 1982:371. "?Sicyonia sp.", Mathews and Gonzalez, 1975:51. Sicyonia ringens. Mathews, 1981:329. Vernacular names: ridgeback prawn, rock shrimp, Japanese shrimp (United States); camaron de piedra, cacahuete (Mexico). FAO names: Pacific rock shrimp (English); camaron de piedra del Pacifico (Spanish); boucot du Pa- cifique (French). Diagnosis. — Antennal spine well developed and buttressed. Second abdominal somite with dor- somedian carina lacking incision. First pereopod with basis and ischium unarmed. Postrostral carina bearing one tooth posterior to level of he- patic spine and low throughout entire length, not raised in crest behind posterior tooth. Abdomen with tooth on dorsomedian carina of first somite smaller or only slightly larger than posterior tooth on carapace. Petasma with distal projections of dorsolateral lobules divergent and bearing short terminal filament. Thelycum with plate of ster- nite XIV raised in pair of lateral bulges; posterior component of median plate bearing pair of lateral bosses cut by transverse suture. Branchiostegite lacking large mark. Description. — Body slender (Fig. 57) and lacking Figure 57.— Sicyonia ingentis (Burkenroad 1938),? 36 mm cl, southeast of Punta Abreojos, Baja Califor- nia Sur, Mexico. Lateral view. Scale = 10 mm. [ FISHERY BULLETIN: VOL. 83, NO. 1 tubercles or occasionally with few on first abdomi- nal somite. Carapace bearing patches of short setae on dorsum, on pterygostomian and anterior part of branchial regions, bordering branchiocar- diac carina, and cluster immediately anteroven- tral to hepatic spine; patches also present on abdomen flanking dorsomedian carina and on anteroventral part of pleuron of first three and sixth somites. Abdomen lacking tubercles or with few on first somite. Rostrum slender, usually somewhat decurved but sometimes subhorizontal, occasionally with anterior extremity upturned; moderately long, overreaching distal margin of eye (extending as far as basal 0.2 of second antennular article), its length 0.30-0.43 cl, increasing linearly with carapace length (Fig. 58); armed with three dorsal teeth and two (rarely three) apical teeth, ventral one considerably smaller than dorsal and usually placed posterior to it but occasionally at same level or even more anteriorly; first rostral tooth sub- equal to, or slightly smaller than epigastric and located opposite and anterior to level of orbital margin, second tooth situated between 0.32 and 0.44 (mean 0.37) rl from orbital margin; and third tooth betwen 0.58 and 0.80 (mean 0.66) rl. Strong adrostral carina, parallel to and rather near ven- tral margin, extending along entire length of ros- trum. Carapace with postrostral carina low but robust throughout its entire length in adults, weak in juveniles, and bearing two teeth: 1) epigastric tooth small, subequal to or barely larger than first rostral tooth, situated anterior to but relatively near level of hepatic spine, between 0.11 and 0.17 (mean 0.16) cl from orbital margin; and 2) pos- terior tooth usually slightly, sometimes conspicu- ously, larger than epigastric and placed well in advance of posterior margin of carapace, between 0.57 and 0.65 (mean 0.63) cl from orbital margin. Tuft of setae present immediately anterior to base of each tooth. Antennal spine moderately long, projecting from sharp, elongate buttress; hepatic spine long, acutely pointed, arising from raised area, and placed between 0.20 and 0.25 (mean 0.22) cl from orbital margin. Postocular sulcus deep anteriorly, continuing posteriorly as long, well-marked arched groove; hepatic sulcus sub- horizontal; hepatic carina indistinct; branchio- cardiac carina strong, longitudinally disposed but curving dorsally near posterior margin of carapace where also often sending short branch ventrally. First article of antennular peduncle and gna- 10 20 30 40 carapace length (mm) 50 Figure 58. — Sicyonia ingentis. Relationship between rostrum length and carapace length (regression equation, >- = 0.75763 + 0.33933X). thai appendages, except third maxilliped, illus- trated in Figure 4. Antennular peduncle with stylocerite produced in long, sharp spine, its length 0.85-0.95 distance between lateral base of first antennular article and mesial base of distolateral spine; latter reaching as far as midlength of second antennular article; antennular flagella relatively long, mesial one, about 0.4 as long as carapace, longer and more slender than lateral; latter about 0.30 as long as carapace. Scaphocerite overreaching antennular peduncle by as much as 0.2 of its own length; lateral rib produced distally in long, acute spine considerably surpassing margin of lamella. Antennal flagellum about 2 times as long as carapace. Third maxilliped about as slender as pereopods. Basis and ischium of first pereopod unarmed. Abdomen with dorsomedian carina extending from first through sixth somites, carina on first very low and produced in small, stubby, anterior tooth, smaller or only slightly larger than pos- terior tooth on carapace; carina on first five somites sloping posteriorly, on sixth produced in large acute posterior tooth. Anteroventral margin of pleuron of first abdom- inal somite slightly convex, sometimes straight in juveniles; posteroventral margin, similar to that of second and usually third somites, gently curved. Anteroventral extremity of pleuron of first four somites lacking spine, although that of second and 70 PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA third occasionally projecting slightly or forming small spine. Posteroventral extremity of pleuron of first and second somites rounded, that of third variably angular (sometimes bearing spine), and that of fourth through sixth produced in caudally directed spine. First somite marked with short anteromedian pleural sulcus continuing as shallow depression to near ventral margin of pleuron, and long, united, posterior tergal-posteromedian pleural sulci. Sec- ond through fourth bearing anterior tergal sulcus (that of fourth weak or sometimes lacking), and united posterior tergat-posteromedian pleural sulci; second and third also bearing shallow depression representing anteromedian pleural sulcus. Fifth somite with almost indistinct an- terior tergal and strongly arched, united posterior tergal-posteromedian pleural sulci. Sixth somite with arched posteromedian pleural sulcus barely, if at all, distinct, and bearing well-marked, long cicatrix. Telson with median sulcus deep basally, fading posteriorly, and armed with small but well- defined, fixed subterminal spines. Rami of uropod subequal in length, exceeding apex of telson by as much as 0.25 of their length. Posterior spine on first abdominal sternite broadly subtriangular with blunt apex and straight or usually convex, instead of concave, lat- eral margins. Petasma (Fig. 59A,B) with cornified distal pro- jection of dorsolateral lobule directed distolater- ally, acutely pointed, ending in short filament, and raised in proximodorsal, subhemispheric promi- nence. Distal projection of ventrolateral lobule reaching as far as projection of dorsolateral lobule, mostly fieshy, blunt, and produced in small lateral tooth just proximal to midlength. Petasmal endopods joined in males 10.5 mm cl, about 41 mm tl, but in individuals with carapace of as much as 19 mm, about 70 mm tl, they may not be joined. Appendix masculina as illustrated in Figure 59C. Thelycum (Fig. 60) with plate of sternite XIV bearing paired strong protuberances bordered lat- erally by narrow flanges and separated by deep median depression sharply delimiting their pos- teromesial margins. Median plate of sternite XIII flask-shaped in outline, tapering gradually into long, slender spine reaching between base and midlength of basis of extended second pereopods; plate set off from posterior component by deep incisions and usually raised posterolaterally in paired rounded prominences flanking narrow de- pression (narrower than prominences); posterior component of median plate bearing paired strong, short bosses separated by deep median depression, each boss cut by transverse suture. Paired short spines projecting anteroventrally from posterior Figure 59. — Sicyonia ingentis, i 16 mm cl, southeast of Punta Tasco, Isla Santa Margarita, Baja California Sur, Mexico. A, Petasma, dorsal view; B, ventral view of same; C, right appendix masculina, dorsolateral view. Scale = 1 mm. 71 FISHERY BULLETIN: VOL. 83, NO. 1 Figure 60. — Sicyonia ingentis, -. 23 mm cl, southeast of Punta Tasco, Isla Santa Margarita, Baja California Sur, Mexico. Thelycum. Scale = 1 mm. margin of sternite XI, spines broad basally and sharp or sometimes needlelike apically. The smallest impregnated females encountered had a carapace of 14 mm, about 57 mm tl. Maximum size. — Males 157 mm tl; females 180 mm tl, measured from "telson to base of antenna" (Herkelrath 1977). In my sample: males 31 mm cl, about 112 mm tl; females 40.2 mm cl, about 133 mm tl. Geographic and bathymetric ranges. — Monterey Bay, 36^50 'N, 121°50'W (Perez Farfante and Boothe 1981), California, southward to Isla Maria Madre, 22°00'N, 106°16'W, Nayarit, Mexico; in the Gulf of California (Fig. 52) along the central part of the eastern coast; and in the southern part along both coasts. This species has been found between 5 and 293-307 m and is most abundant from 55 to 82 m, at which depth it is commercially fished off Ventura, Calif. (Frey 1971); also most of the many specimens examined by me were taken within this range. According to Carlisle (1969) it is common at depths between 61 and 183 m. Sicyonia ingentis occurs on substrates of sand, shell, and green mud, but seems to prefer sandy bottoms on which com- mercial concentrations are located. This species is the only member of Sicyonia that has been recorded along the west coast of the United States and north of Punta Canoas, Baja California Sur — about 1,000 km south of Mon- terey Bay, the northern limit of its range. The record from Isla Maria Madre, Nayarit, is the first from waters south of the Gulf of California. Discussion. — Sicyonia ingentis, the largest east- ern Pacific species in the genus, has its closest affinities with the much smaller, sympatric S. disdorsalis . It differs from the latter in possess- ing 1) a carapace bearing, not lacking, patches of long setae, 2) a robust postrostral carina rather than a slender one, 3) a strong branchiocardiac carina instead of an almost indistinct one, 4) a carina on the fifth abdominal somite which slopes gently to near the posterior cleft rather than being truncate or produced in a spine, 5) an anterome- dian pleural sulcus on the first abdominal somite which continues as a shallow depression almost to the ventral margin of the pleuron instead of end- ing abruptly and well above it, 6) an unarmed antero ventral extremity on the pleuron of the first abdominal somite rather than one armed with a strong spine, 7) curved, instead of angular, pos- teroventral pleural margins on the first two ab- dominal somites, 8) well-developed telsonic spines instead of minute or indistinct ones, and 9) long uropodal rami that considerably surpass the apex of the telson instead of falling short of or barely overreaching it. Sicyonia ingentis also differs from S. disdorsalis in characters of the genitalia. The distal projection of the dorsolateral lobule of the petasma is di- rected distolaterally instead of curving distome- sially and is produced in a short apical filament which is lacking in S. disdorsalis. The distal pro- jection of the ventrolateral lobule reaches, instead of falls short of, the terminal margin of the dor- solateral lobule; furthermore, it is neither bifur- cate laterally nor does it bear a transverse rib. In the thelycum, the median depression on the pos- terior part of the median plate of sternite XIII is narrower than the usually rounded protuberances flanking it, whereas in S. disdorsalis the depres- sion is much broader than the flat or concave areas which occupy the position of the two protuber- 72 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA I ances. Finally, the paired lateral bosses, repre- senting the posterior component of the median plate and each cut by a transverse suture, are found only in S. ingentis. The "pencil of hairs" that Burkenroad (1938) stated to be located on the dorsal surface of the distal part of the ocular peduncle, near its disto- lateral margin, actually is placed on the distome- sial margin. He distinguished S. ingentis from S. disdorsalis by, among other characters, the rela- tive length of that tuft of setae, stating that in large adults of the former species it overreaches the eye whereas in S. disdorsalis it spans no more than half the cornea. This character does not seem to be a reliable one for, except in occasional speci- mens of S. ingentis, in neither species does the tuft surpass the eye. Burkenroad also considered the disposition of the adrostral rib as a diagnostic fea- ture that would serve to separate the two species. Although in S. ingentis the rib lies parallel to the ventral margin of the rostrum, in S. disdorsalis its course varies: sometimes it is slightly arched near the anterior end, as Burkenroad described it, but often it is curved along the middle and occasion- ally is slightly turned anterodorsally. The distance of the rib from the ventral margin in the two shrimps, however, exhibits a slight difference — in S. ingentis it extends close to the margin whereas in S. disdorsalis it lies more dorsally. The characters exhibited by S. ingentis, S. dis- dorsalis, and S. picta suggest that they must have diverged quite early from a common ancestor in the group of species that share two teeth on the postrostral carina. Notes on biology. — Herkelrath (1977) investigated the temperature tolerance and age- growth and length-weight relationships in this shrimp. He found that within a salinity range of 33-35%o it exhibits a wide range of tolerance to temperature (4°-30°C). At a stressed salinity (261) this toler- ance was considerably reduced (7°-25°), and mor- tality increased proportionately with the duration of exposure, regardless of temperature. His studies indicated that shrimp with a total length of 50-90 mm increased 10 mm per month and also that there is no difference in length-weight ratio between sexes. He also stated that among shrimp "averaging 70 mm or greater in total length, the average length of females was greater than that of males." Anderson (1983) studied growth rates, molting, and certain aspects of reproduction in a population of S. ingentis occurring off Santa Barbara, CA. She found that spawning takes place far offshore in deep water, about 145 m, and lasts from May through October with the peak during the late summer. She also observed that molt frequency is highest in the winter and spring, that females do not molt during the summer (the reproductive period), and that males exhibit a similar pattern. Size-frequency analyses based on monthly off- shore and nearshore sampling indicated that ju- veniles increased at a monthly rate of about 1-2 mo. Commercial importance. — There is a fishery for this shrimp between Santa Barbara and Ventura, Calif According to the California Department of Fish and Game, landings in 1982 amounted to 127,000,956 lb with a value of $156,000,385. Mathews (1981) stated that "Sicyonia ringens" is occasionally fished in Magdalena Bay, which is located on the ocean side of Baja California Sur. I have little doubt that his remark applies to S. ingentis and that "ringens" is an erroneous spell- ing. Moreover, it seems to me almost certain that the study of "Sicyonia sp." (distribution, abun- dance, rate of growth, ratio total weight/total length) in Magdalena Bay by Mathew and Gon- zalez (1975), was based on a population of this species, apparently the only abundant rock shrimp in the area. However, because Magdalena Bay is within the range of S. penicillata , another species reaching sizes reported by the authors, it is not possible to be certain of the identity of the shrimp studied by them. It is indeed unfortunate that the valuable information presented cannot be definitely associated with a specific shrimp, par- ticularly in view of the fact that so little is known of the biology of any of the eastern Pacific rock shrimps. Although S. ingentis is present in the Gulf of California, it is not commercially exploited there. Material. — 946 specimens from 52 lots. United States— California: 1?, CAS, 2 km W of Moss Landing, Monterey Bay, 50 m, 23 Sep- tember 1978, D. D. Chivers. IS 3$ , AHF, 5 km off Point Mugu, 40-59 m, 25 April 1976, Velero IV stn 24833. Id , AHF, 8.4 km W of Venice, 70-73 m, 22 July 1958, J. L. Baxter. 1$, SIO, SW of Santa Monica Bay, 22 March 1962, F H. Berry and H. C. Perkins. 39 , SIO, San Pedro Bay 27 m, 20 March 1964, U.S. Fish and Wildlife Service staff. 16 46 , SIO, N of Dana Point, 53-48 m, 29 March 1974, Agassiz. 316 20V, JlO, off San Onofre, 54 m, 29 March 197 4, Agassiz. 26 29 , SIO, off San Onofre, 73 FISHERY BULLETIN: VOL. 83, NO. 1 91 m, 29 March 1974, Agassiz. 116 159, SIO, off Encinitas, 51 m, 29 March 1974, Agassiz. Mexico— Baja California Norte: 76 59, SIO, off Bahia de San Quintin, 57 m, 1 April 1962, Fish and Wildlife Service staff. 15c? 119, SIO, off Bahia de San Quintin, 73 m, 4 December 1960, C. Boyd and D. Dockins. 126 109 , SIO, off Bahia de San Quintin, 74-77 m, H. C. Perkins. 66 69 , SIO, off Bahia de San Quintin, 143-148 m, 1 April 1962, H. C. Perkins. 26 59, SIO, SE of San Felipe, 120 m, 19 January 1968, Thomas Washington. 19, SIO, W of Punta Prieta, 23 March 1960, H. C. Perkins. 66 79, SIO, Bahia Sebastian Vizcaino, 88 m, 19 August 1960, W. D. Clarke. 19, YPM, E of Isla Cedros, 110 m, 22 May 1936, Zaca stn 126D-10. 46 19, SIO, between San Benito and Isla Cedros, 247-265 m, 27 May 1971, C. Hubbs and S. Luke. 4(5, YPM, E of Isla Cedros, 80 m, 27 March 1936, Zaca stn 125D-1. 246 219 , USNM, E of Isla Cedros, 80 m, 5 May 1888, Albatross stn 2838. 126 199, SIO, WSW of Red Rock, Bahia Sebastian Vizcaino, 113-119 m, 25 November 1961, E H. Berry Id , YPM, E of Isla Cedros, 73 m, 22 May 1936, Zaca stn 126D-4. 46 69, YPM, E of Isla Cedros, 69 m, 27 March 1936, Zaca stn 126D-2. 6 holotype, AMNH, off east coast of Isla Cedros, 69 m, 27 March 1936, Zaca stn 127D-1. 36 19 and 26 29 paratypes, AMNH and YPM, respectively, collected with holotype. 36 li, YPM, E of Isla Cedros, 70-110 m, 27 March 1936, Zaca stn 125D-1. Baja California Sur: 36 19, SIO, Bahia Sebastian Vizcaino, 55 m, 11 August 1952, K. S. Norris. 26 29, SIO, Bahia de San Cristobal, 83-87 m, 2 December 1961, F H. Berry 196 179, SIO, Bahia Asuncion, 68-64 m, 17 November 1964, Black Douglas. 206 209, SIO, SW of Punta San Hipolito, 6 March 1954, "J.M. and W.H." 13d 209, SIO, SE of Punta Abre- ojos, 55-59 m, 17 November 1964, Black Doug- las. 46 179, SIO, SE of Punta Abreojos, 73-79 m, 17 November 1964, Black Douglas. 216 219, SIO, SE of Punta Abreojos, 91 m, 2 December 1960, C. Boyd and D. Dockins. 20d 209, SIO, WSW of Punta Pequeha, 68-73 m, 16 November 1964, Black Douglas. 26 39, SIO, 15 km WSW of Boca de las Animas, 55-57 m, 16 November 1964, Black Douglas. 146 69, SIO, SW of Santo Domingo del Pacifico, 100 m, 20 April 1969. 19d lb , SIO, 16 km NW of Isla Magdalena, 99-102 m, 16 November 1964, Black Douglas. 236 229, SIO, off Bahia Magdalena, 88 m, 3 February 1964, C. Hubbs. 30d 30.-, SIO, SW of Isla San- ta Margarita, 75-81 m, 13 November 1964, Black Douglas. 446 269, SIT, SE of Punta Tasco, Isla 74 Santa Margarita, 102-106 m, 27 June 1965, Hori- zon. 256 259, SIO, W of Inocentes, 91-93 m, 10 November 1964, Black Douglas. 29, SIO, WNW of Punta Lobos, 183-201 m, 9 November 1964, Black Douglas. 16, SIO, Bahia de la Paz, 82- 119 m, 12 January 1968, Thomas Washington. Sonora: Id 19, SIO, off Hermosillo coast, 289- 304 m, 25 March 1960, Curray and R. H. Parker. 19, AHF, S of Isla Tiburon, 4-29 m, 25 January 1940. 15d 119, SIO, off Santa Rosalia, 64-48 m, 25 March 1960, R. H. Parker. 25d 259, SIO, off Isla San Pedro Martir, 293-307 m, 21 January 1968, Thomas Washington. 19, AHF, Bahia de Guaymas, 5 m, 23 March 1949. Sinaloa: 2d 19, USNM, Puerto de la Punta Altata, 9 May 1962, R. E. Bush. 19, USNM, Los Cocos, 42 m, 18 May 1962, R. E. Bush. Nayarit: 24d 219, SIO, NE of Isla Maria Madre, Islas Tres Marias, 82-88 m, 30 March 1973, Agassiz. ACKNOWLEDGMENTS Many persons have assisted me throughout the course of this study. Because of the large area covered I was dependent upon many sources for specimens. I am deeply indebted to the following who have lent collections to me, many of which included materials representing new locality records and range extensions: Dustin D. Chivers, California Academy of Sciences; Enrique M. Del Solar, Instituto del Mar del Peru; Harold S. Feinberg, American Museum of Natural History; Jacques Forest, Museum National d'Histoire Naturelle, Paris; H. E. Gruner, Zoologisches Museum, Humboldt- Universitat, Berlin; Janet Haig, Allan Hancock Foundation, University of Southern California; Willard D. Hartman, Pea- body Museum of Natural History, Yale University; Leslie W. Knapp, Smithsonian Oceanographic Sorting Center, Smithsonian Institution; Herbert V. Levi and Catharine G. Sibble, Museum of Com- parative Zoology, Harvard University; Spencer R. Luke, Scripps Institution of Oceanography; Juan B. del Rosario, Universidad de Panama; and Mary K. Wicksten, Texas A & M University, formerly of Allan Hancock Foundation. I have the privilege of drawing on the consider- able knowledge of decapod crustaceans system- atics of Horton H. Hobbs, Jr., of the Smithsonian Institution, who has given me numerous sugges- tions during the investigation and has criticized the manuscript. Fenner A. Chace, Jr., of the Smithsonian Institution, and Bruce B. Collette and Austin B. Williams, both of the Systematics P6REZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Laboratory, National Marine Fisheries Service, also have read the manuscript and made helpful comments. I acknowledge their assistance with deep appreciation. I am deeply grateful to Matilde Mendez G., Instituto del Mar del Peru, and to Con- cepcion Rodriguez de la Cruz, Instituto de Pesca, Mexico, for the loan of specimens and for invalu- able information concerning materials deposited in their respective institutions. Thanks are also extended to Meredith L. Jones, Smithsonian In- stitution, whose interest and efforts facilitated my field work in Panama; and for making available critical bibliographic references to Lipke B. Holthuis, Rijksmuseum van Natuurlijke Historie, Leiden; Alexander Dragovich, Southeast Fisheries Center, National Marine Fisheries Ser- vice; and to Michel E. Hendrickx, Instituto de Ciencias del Mar y Limnologia, Estacion Maza- tlan, Mexico. I wish to thank Billy B. Boothe, Jr., who sorted and identified the Sicynnia that had been collected among large quantities of penaeoids from the re- gion under investigation and assisted me with the morphometric work. Joseph L. Russo, formerly of the NMFS Systematics Laboratory, helped with statistical analyses and computer work, and Ruth Gibbons also assisted in the analyses and prepared the graphs and distribution maps. Maria M. Dieguez applied her artistic talent to the prepara- tion of all of the illustrations except Figures 3, 4, and 7 which were ably rendered by Keiko H. Moore. Finally, I also acknowledge with gratitude the assistance of Arleen S. McClain and Virginia R. Thomas, who patiently typed several drafts of the manuscript. LITERATURE CITED ALLEN, D. M. 1973. Rock shrimp (Sicyonia brevirostris) production - South Atlantic coast of the United States and Gulf of Mexico. Summary Report. Natl. Mar Fish. Serv, NOAA, Southeast Fish. Cent., Miami Lab., 15 p. ANDERSON, S. L. 1983. Molt and reproduction of the ridgeback prawn, Sicyonia ingentis (Penaeidae). Ph.D. Thesis, Univ. California, Davis, 115 p. ANDERSON, W. W. 1956. January to April distribution of the common shrimp on the South Atlantic Continental Shelf. U.S. Fish Wildl. Serv, Spec. Sci. Rep. Fish. 171, 14 p. 1962. Recognizing important shrimps of the South. U.S. Fish Wildl. Serv, Fish. Leafl. 536, 5 p. (Supersedes Fish. Leaf!. 366.) ANDERSON, W. W., AND M. J. LINDNER. 1945. A provisional key to the shrimps of the family Penaeidae with especial reference to American forms. Trans. Am. Fish. Soc. 73:284-319. ANDERSON, W. W., M. J. LINDNER, AND J. E. KING. 1949. The shrimp fishery of the southern United States. Commer. Fish. Rev 11(2):1-17. ANONYMOUS. 1980. Crustaceos colectados con red de arrastre en el Golfo de California. Invest. Oceanogr. B-80-04, Seer. Mar., Mex., 25 p. ARANA ESPINA, R, AND M. MENDEZ G. 1978. El genero Sicyonia H. Milne Edwards, 1830 en el Pacifico Sur Oriental, con observaciones biologicas sobre Sicyonia alliaffinis Burkenroad 1934 (Crustacea: De- capoda: Penaeidae). Rev Com. Perm. Pac. Sur. 9:19-40. ARREGLfiN SANCHEZ, F 1981. Diagnosis de la pesqueria de camaron de roca {Sicyonia brevirostris Stimpson, 1871) de Contoy, Q. Roo, Mexico. Cienc. Pesq., Inst. Nac. Pesca, Dep. Pesca, Mex. 1(2):21-4L BALSS, H. 1914. Ostasiatische Decapoden II. Die Natantia und Rep- tantia. Beitrage zur Naturgeschichte Ostasiens. Herausgegeben von Dr. F. Doflein. Abh. Bayer Akad. Wiss. Math-Phys. Kl., Suppl. 2, Pt. 10, p. 1-101. 1925. Macrura der Deutschen Tiefsee-Expedition. 2. Natantia, Teil A. Wiss. Ergeb. "Valdivia" Exped. 20, p. 217-315. BARNARD, K. H. 1950. Descriptive catalogue of South African decapod Crustacea. Ann. S. Afr Mus. 38:1-837. BATE, C. S. 1888. Report on the Crustacea Macrura collected by H.M.S. Challenger during the years 1873-76. Rep. Sci. Results Voyage H.M.S. Challenger. Zool. 24, 942 p. BAYER, F M., G. L. VOSS, AND C. R. ROBINS. 1970. Bioenvironmental and radiological safety feasibility studies Atlantic-Pacific interoceanic canal. Report on the marine fauna and benthic shelf-slope communities of the Isthmian region. Rosenstiel School Mar. Atmos. Sci., Univ Miami, 311 p. BOECK, A. 1864. Beskrivelse og fremlagde Tegninger af 4 norske De- capoder, underaOgte af Overlaege Danielssen og ham. Forh. Vidensk. Selsk., Christ. (1863):189-190. BOONE, L. 1930. Crustacea: Anomura, Macrura, Schizopoda, Isopoda, Amphipoda, Mysidacea, Cirripedia, and Copepoda. Scien- tific results of the cruises of the yachts "Eagle" and "Ara", 1921-1928, William K. Vanderbilt, commanding. Bull. Vanderbilt Mar. Mus. Vol. 3, 221 p. BOUVIER, E. L. 1905. Sur les macroures nageurs (abstraction faite des carides), recueillis par les expeditions americaines du Hassler et du Blake. C. R. Acad. Sci., Paris 140:746-749. 1918. Sur une petite collection de crustaces de Cuba offerte au Museum par M. de Boury Bull. Mus. Hist. Nat. Paris 24:6-15. BRUNNICH, M. T 1768. Ichthyologia Massiliensis, sistens piscium descrip- tiones eorumque apud incolas nomina. Accedunt spolia maris Adriatici, Hafniae at Lipsiae, apud Rothii viduam et Proft, 110 p. BRUSCA, R. C. 1973. A handbook to the common intertidal invertebrates of the Gulf of California. 1st ed. Univ Ariz. Press, Tuc- son, 427 p. 75 FISHERY BULLETIN: VOL. 83, NO. 1 1980. Common intertidal invertebrates of the Gulf of California. Univ. Ariz. Press, Tucson, 513 p. BRUSHER, H. A., AND L. H. OGREN. 1976. Distribution, abundance, and size of penaeid shrimps in the St. Andrew Bay system, Florida. Fish. Bull., U.S. 74:158-166. BRUSHER, H. A., W. C. RENFRO, AND R. A. NEAL. 1972. Notes on distribution, size, and ovarian development of some penaeid shrimps in the northwestern Gulf of Mexico, 1961-1962. Contrib. Mar Sci. 16:75-87. BRYAN, C. E., AND T. J. CODY. 1975. Abundance of rock shrimp, Sicyonia brevirostris (Stimpson), in Texas. In A study of commercial shrimp, rock shrimp, and potentially commercial finfish 1973- 1975, Part II, p. 1-11. Texas Parks Wildl. Dep., Austin, Texas. BURKENROAD, M. D. 1934a. Littoral Penaeidea chiefly from the Bingham Oceanographic Collection with a revision of Penaeopsis and descriptions of two new genera and eleven new Ameri- can species. Bull. Bingham Oceanogr. Collect., Yale Univ 4(7):1-109. 1934b. The Penaeidea of Louisiana with a discussion of their world relationships. Bull. Am. Mus. Nat. Hist. 68:61-143. 1938. The Templeton Crocker Expedition. XIII. Penaeidae from the region of Lower California and Clarion Island, with descriptions of four new species. Zoologica (N.Y.) 23:55-91. 1939. Further observations on Penaeidae of the northern Gulf of Mexico. Bull. Bingham Oceanogr. Collect., Yale Univ 6(6):l-62. 1946. Status of the name Sicyonia H.M.E., with a note on S. typica (Boeck) and descriptions of two new species. Ark. Zool. 37A(9):1-10. Cain, T. D. 1972. Additional epifauna of a reef off North Carolina. J. Elisha Mitchell Sci. Soc. 88:79-82. Camp, d. K., n. h. whiting, and r. E. Martin. 1977. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974. V. Arthropods. Fla. Mar. Res. Publ. 25, 63 p. Carlisle, J. G., Jr. 1969. Invertebrates taken in six year trawl study in Santa Monica Bay Veliger 11:237-242. Castille, F L., Jr., and A. L. Lawrence. 1981. The effect of salinity on the osmotic, sodium and chloride concentrations in the hemolymph of the rock shrimps, Sicyonia brevirostris and Sicyonia dor- salis. Comp. Biochem. Physiol. 70A:519-523. Castro, p 1966 The shrimps (Crustacea Decapoda Natantia) of the Sonoran coast of the Gulf of California. Univ Ariz. Biol. Stud. Gulf Calif 4(2):l-33. CERAME-VIVAS, M. J., AND I. E. GRAY. 1966. The distributional pattern of benthic invertebrates of the continental shelf off North Carolina. Ecology 47:260-270. CHACE. f a., Jr. 1972. The shrimps of the Smithsonian-Bredin Caribbean expeditions with a summeu-y of the West Indian shallow- water species (Crustacea: Decapoda: Natantia). Smith- son, Contrib. Zool. 98. 179 p. CHAPA SALDANA, H. 1%4. Lista preliminar de los crustaceos existentes en el Laboratorio Central del I.N.I. B. P. Inst. Nac. Invest. 76 Biol.-Pesq. Ser. Trab. Divulg. 9(87), 40 p. CHIRICHIGNO FONSECA, N. 1970. Lista de crustaceos del Peru (Decapoda y Stomatopoda) con dates de su distribucion geografi- ca. Inst. Mar. Peru (Callao), Inf. 35, 95 p. COBB, S. P 1971. A new species of Sicyonia (Decapoda, Penaeidae) from the western Atlantic with notes on S. stimpsoni Bouvier Crustaceana 20:104-112. COBB, s. P, c. R. futch, and d. k. Camp. 1973. Memoirs of the Hourglass Cruises: the rock shrimp, Sicyonia brevirostris Stimpson, 1871 (Decapoda, Penaeidae). Fla. Dep. Nat. Res. Mar. Res. Lab. 3 (Part 1), 38 p. coen, l. d., and k. l. heck, Jr. 1983. Notes on the biology of some seagrass-dwelling crus- taceans (Stomatopoda and Decapoda) from Caribbean Panama. Proc. Biol. Soc. Wash. 96:202-224. DAY, J. W, JR., W. G. SMITH, P R. WAGNER, AND W. C. STOWE. 1973. Community structure and carbon budget of a salt marsh and shallow bay estuarine system in Louisi- ana. Cent. Wetland Resour, La. State Univ. Publ. LSU- SG-72-04, 79 p. DE HAAN, W. 1833-1850. Crustacea. In P. F. von Siebold, Fauna Japonica sive descriptio animalium, quae in itinere per Japoniam, jussu et auspiciis superiorum, qui summum in India Batava Imperium tenent, suscepto, annis 1823-1830 collegit, notis, observationibus et adumbrationibus illustravit. Lugduni-Batavorum [Leiden], 243 p. DE MAN, J. G. 1907. On a collection of Crustacea, Decapoda and Stomatopoda, chiefly from the Island Sea of Japan; with descriptions of new species. Trans. Linn. Soc. Lond. Zool., Ser. 2, 9:387-454. 1911. The Decapoda of the Siboga Expedition. Part I, Fam- ily Penaeidae. Siboga-Exped. Monogr. 39a, 131 p. (Suppl. to Part 1, 1913.) DE Natale, G. 1850. Descrizione zoologica d' una nuova specie di Plojaria e di alcuni crostacei del porto di Messina con poche con- siderazioni generali sulla natura delle appendici aculeiformi delle pinati e degli animali. Giuseppe Fiumara, Messina, 31 p. app. DE SAUSSURE, H. 1857. Diagnoses de quelques crustaces nouveaux des An- tilles et du Mexique. Rev Mag. Zool., Ser. 2, 9:304-308. DEL SOLAR C.,E. M. 1972. Addenda al Catalogo de crustaceos del Peru. Inst. Mar. Peru (Callao), Inf 38, 21 p. DEL SOLAR C, E. M., F BLANCAS S., AND R. MAYTA L. 1970. Catalogo de crustaceos del Peru. D. Miranda, Lima, Peru, 53 p. Dunn, D. f 1982. Announcement. J. Crustacean Biol. 2:153. ELDRED, B. 1959. A ref)ort on the shrimps (Penaeidae) collected from the Tbrtugas controlled area. Fla. Board Conserv., Mar Lab., Spec. Sci. Rep. 2, 6 p. ELDRED, B., J. WILLIAMS, G, T MARTIN, AND E. A. JOYCE, JR. 1965. Seasonal distribution of penaeid larvae and postlar- vae of the Tampa Bay area, Florida. Fla. Board Conserv. Mar. Lab., Tech. Ser. 44, 47 p. FAUSTO FILHO, J. 1966a. Primeira contribuicao ao inventario dos crustaceos decapodos marinhos do nordeste brasileiro. Arq. Estac. PfeREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Biol. Mar. Univ. Fed. Ceara 6:31-37. 1966b. Sbbre os peneideos do nordeste brasileiro. Arq. Estac. Biol. Mar. Univ. Fed. Ceara 6:47-50. 1968. General considerations on the peneids of north and northeast of Brazil. Arq. Estac. Biol. Mar. Univ. Fed. Ceara 8:69-73. Faxon, W. 1893. VI. Preliminary descriptions of new species of Crus- tacea. Reports on the dredging operations off the west coast of Central America to the Galapagos, to the west coast of Mexico, and in the Gulf of California, in charge of Alexander Agassiz, carried on by the U.S. Fish Commis- sion steamer "Albatross," during 1891, Lieut. Commander Z. L. Tanner, U.S.N., commanding. Bull. Mus. Comp. Zool, Harv Coll. 24:149-220. 1895. XV. The stalk-eyed Crustacea. Reports on an ex- ploration off the west coast of Mexico, Central and South America, and off the Galapagos Islands, in charge of Alexander Agassiz, by the U.S. Fish Commission steamer "Albatross," during 1891, Lieut. Commander Z. L. Tanner, U.S.N. , commanding. Mem. Mus. Comp. Zool., Harv. Coll. 18:1-292. 1896. XXXVII. Supplementary notes on the Crustacea. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico and the Carib- bean Sea, and on the east coast of the United States, 1877 to 1880, by the U.S. Coast Survey steamer "Blake," Lieut. -Commander C. D. Sigsbee, U.S.N., and Com- mander J. R. Bartlett, U.S.N. , commanding. Bull. Mus. Comp. Zool., Harv Coll. 30:153-166. FEINBERG, H. S. 1971. A catalogue of type specimens in the Department of Living Invertebrates. The American Museum of Natural History. New York, New York, U.S.A. Phylum Arthropoda; Class Crustacea. Part one: Order Decapoda. Am. Mus. Nat. Hist., N.Y, 75 p. FRANKS, J. S., J. Y. CHRISTMAS, W. L. SILER, R. COMBS, R. WALLER, AND C. BURNS. 1972. A study of nektonic and benthic faunas of the shal- low Gulf of Mexico off the state of Mississippi as related to some physical, chemical and geological factors. Gulf Res. Rep. 4:1-148. FREY, H. W. 1971. California's living marine resources and their utili- zation. Calif. Dep. Fish Game, Resour Agency, 148 p. Garcia Pinto, l. 1971. Identificacion de las postlarvas del camaron (genero Penaeus) en el occidente de Venezuela y observaciones sobre su crecimiento en el laboratorio. Proyecto Invest. DesarroUo Pesq. MAC-PNUD-FAO Inf. Tec. 39, 23 p. Greening, H. S., and r. j. Livingston. 1982. Diel variation in the structure of seagrass-associated epibenthic macroinvertebrate communities. Mar. Ecol. Prog. Ser. 7:147-156. GUNTER, G. and D. MCCAUGHAN. 1959. Catalepsy in two common marine animals. Science 130:1194-1195. HAAN, W. de. See De Haan, W. HALL, D. N. F 1956. The Malayan Penaeidae (Crustacea, Decapoda). Part I. Introductory notes on the species of the general Solenocera, Penaeus, and Metapenaeus. Bull. Raffles Mus. 27:68-90. HAY, W P, AND C. A. SHORE. 1918. The decapod crustaceans of Beaufort, N.C., and the surrounding region. Bull. U.S. Bur. Fish. 35:369-475. HEDGPETH, J. W 1953. An introduction to the zoogeography of the north- western Gulf of Mexico with reference to the inver- tebrate fauna. Publ. Inst. Mar. Sci., Univ. Tex. 3:111- 224. HELLER, C. 1863. Die Crustaceen des siidlichen Europa. Crustacea Podophthalmia. Mit einer iibersicht iiber die horizontale Verbreitung sammtlicher europaischer Arten, Vien- na. Wien, W. Brawmiiller, 336 p. HEMMING, F. (editor). 1958. Official list of generic names in zoology. First in- stallment: Names 1-1274. Int. Trust. Zool. Nomencl., Lend., 200 p. HERKELRATH, J. M. 1977. Some aspects of the growth and temperature toler- ance of the ridgeback prawn, Sicyonia ingentis Burken- road in southern California waters. M.S. Thesis, Whit- tier College, Whittier, CA, 67 p. HILDEBRAND, H. H. 1954. A study of the fauna of the brown shrimp {Pe- naeus aztecus Ives) grounds in the western Gulf of Mexico. Publ. Inst. Mar. Sci., Univ. Tex. 3:233- 366. 1955. A study of the fauna of the pink shrimp (Penaeus duorarum Burkenroad) grounds in the Gulf of Caim- peche. Publ. Inst. Mar. Sci., Univ Tex. 4:169-232. HOLTHUIS, L. B. 1952. Proposed use of the Plenary Powers to validate the generic name "Sicyonia" H. Milne Edwards, 1830 (Class Crustacea, Order Decapoda). Bull. Zool. Nomencl. 6:339-342. 1980. FAO species catalogue. Vol. 1. Shrimps and prawns of the world. An annotated catalogue of species of interest to fisheries. FAO Fish. Synop. 125, 261 p. HOOKS, T A., K. L. HECK, JR., AND R. J. LIVINGSTON. 1976. An inshore marine invertebrate community: struc- ture and habitat associations in the northeastern Gulf of Mexico. Bull. Mar. Sci. 26:99-109. HUFF, J. A., AND S. P COBB. 1979. Memoirs of the Hourglass Cruises: penaeoid and sergestoid shrimps (Crustacea: Decapoda). Fla. Dep. Nat. Resour. Mar. Res. Lab. 5(part 4), 102 p. HUTTON, R. E, F. SOGANDARES-BERNAL, B. ELDRED, R. M. INGLE, AND K. D. WOODBURN. 1959. Investigations on the parasites and diseases of salt- water shrimps (Penaeidae) of sports and conamercial im- portance to Florida. Fla. Board Conserv. Tech. Sen 26, 36 p. INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLA- TURE. 1956. Opinion 382. Validation under the Plenary Powers of the generic name Sicyonia Milne Edwards (H.), 1830 (Class Crustacea, Order Decapoda) and action consequen- tial thereon. In Opinions and declarations rendered by the International Commission on Zoological Nomencla- ture, Vol. 12, p. 43-58. JOYCE, E. A., JR. 1965. The commercial shrimps of the northeast coast of Florida. Fla. Board Conserv. Mar. Res. Lab. Prof Pap. Ser. 6, 224 p. 1968. Project Hourglass explores the continental shelf. Sea Front. 14:352-359. JOYCE, E. A., Jr., AND B. ELDRED. 1966. The Florida shrimping industry Fla. Board Con- serv. Mar Res. Lab. Educ. Ser. 15, 47 p. 77 FISHERY BULLETIN; VOL. 83, NO. 1 KENNEDY, F. S., J. J. CRANE, R. A. SCHLIEDER, AND D. G. BARBER. 1977. Studies of the rock shrimp, Sicyonia brevirostris , a new fishery resource on Florida's Atlantic shelf. Fla. Mar Res. Publ. 27, 69 p. KINGSLEY, J. S. 1878. List of the North American Crustacea belonging to the Suborder Caridea. Bull. Essex Inst. 10:53-71. 1880. On a collection of Crustacea from Virginia, North Carolina, and Florida, with a revision of the genera of Crangonidae and Palaemonidae. Proc. Acad. Nat. Sci., Phila. 31:383-427. KUBO, I. 1949. Studies on penaeids of Japanese and its adjacent waters. J. Tokyo Coll. Fish. 36(l):l-467. KUTKUHN, J. H. 1962. Conversion of "whole" and "headless" weights in commercial Gulf of Mexico shrimps. U.S. Fish Wildl. Serv, Spec. Sci. Rep. Fish. 409, 7 p. LOCKINGTON, W. N. 1879. Notes on Pacific coast Crustacea. Bull. Essex Inst. 10:159-165. LONGHURST, A. R. 1970. Crustacean resources. In J. A. Gulland (editor), The fish resources of the ocean, p. 252-305. FAO Fish. Biol. Tech. Pap. 97. LUNZ, G. R. 1945. Carolina shrimps of the genus £i/sJC3'o«ia. Char- leston Mus. Leafl. 20, 12 p. 1957. Notes on rock shrimp Sicyonia brevirostris (Stimpson) from exploratory trawling off the South Carolina coast. Contrib. Bears Bluff Lab. 25, 10 p. Lyons, w. G., S. P Cobb, d. k. Camp, J. A. Mountain, T Savage, L. Lyons, and E. a. Joyce, jr. 1971. Preliminary inventory of marine invertebrates col- lected near the electrical generating plant, Crystal River, Florida, in 1969. Fla. Dep. Nat. Resour Mar Res. Lab. Prof Pap. Sen 14, 45 p. MaN,J. G., DE. See De Man, J. G. Mathews, C. R 1981. A review of the North American penaeid fisheries with particular reference to Mexico. Kuwait Bull. Mar. Sci. 2:325-409. Mathews, C. P, and J. A. Gonzalez. 1975. Potencial pesquero y estudios ecologicos de Bahia Magdalena I. Las existencias de camaron con referenda especial al camaron de roca o camaron japones. Cienc. Pesq., Inst. Nac. Pesca, Mex. 2(l):51-57. MCCONNELL, R. H. 1960. Part II. The interpretation ofthe catch records. The trawl survey carried out by the RA^ "Cape St. Mary" off British Guiana 1957-59. Br Guiana Fish. Bull. 2:25-53. MENDEZ G., M. 1981. Claves de identificacion y distribucion de los langos- tinosycamarones (Crustacea: Decapoda)del mary riosde la costa del Peru. Bol. Inst. Mar Peru 5:1-170. MENZEL, R. W. 1956. Annotated check-list ofthe marine fauna and flora of the St. George's Sound -Apalachee Bay region, Florida Gulf coast. Oceanogr. Inst., Fla. State Univ., Contrib. 61, 78 p. MIERS, E. J. 1881. On a collection of Crustacea made by Baron Hermann-Maltzam at Goree Island, Senegambia. Ann. Mag. Nat. Hist., Ser 5, 8:204-220, 259-281, 364-377. MILNE Edwards, a., and E. L. bouvier. 1909. XLIV. Les Peneides et Stenopides. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78), in the Caribbean Sea (1878-79), and along the Atlantic coast ofthe United States (1880), by the U.S. Coast Survey steamer "Blake," Lieut. -Com. C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., commanding. Mem. Mus. Comp. Zool., Harv. Coll. 27:177-274. MILNE Edwards, H. 1830. Description des genres Glaucothoe, Sicyonie, Sergeste et Acete, de I'ordre des Crustaces Decapodes. Ann. Sci. Nat., Sen 1, 19:333-352. 1837. Histoire naturelle des crustaces, comprenant I'anatomie, la physiologie et la classification de ces animaux. [Atlas] Roret, Paris, 2, 532 p. MONOD, T, AND P CALS. 1970. Sur une espece nouvelle de crevette cavernicole: Typhlatya galapagensis (Decapoda Natantia: Aty- idae). Mission zool. belg iles Galapagos Ecuador 2:57- 103. NatALE, G. De. See De Natale, G. Office of Geography, department of the Interior. 1956. NIS Gazetteer Mexico. U.S. Printing Office, Wash- ington, D.C., 750 p. PARKER, R. H. 1964. Zoogeography and ecology of some macro- invertebrates, particularly mollusks, in the Gulf of California and the continental slope off Mexico. Vid- ensk. Med. Dan. Naturh. Foren. 126:1-178. PAUL, R. K. G., AND M. E. HENDRICKX. 1980. Crustaceans in the shrimp by-catch from off the coasts of Sinaloa and Nayarit, Mexico. Bull. South. Calif Acad. Sci. 79:109-111. PEARSE, A. S., AND L. G. WlLLL^MS. 1951. The biota of the reefs off the Carolinas. J. Elisha Mitchell Sci. Soc. 67:133-161. PEREZ FARFANTE, I. 1978. Families Hippolytidae, Palaemonidae (Caridea), and Penaeidae, Sicyoniidae and Solenoceridae (Penaeoidea). In Fischer, W. (editor), FAO species identification sheets for fishery purposes, western central Atlantic (fishing area 31), Vol. VI (unpaginated). FAO, Rome. 1980. A new species of rock shrimp of the genus Sicyonia (Penaeoidea), with a key to the western Atlantic spe- cies. Proc. Biol. Soc. Wash. 93:771-780. 1982. Camarones. In N. Chirichigno, W. Fischer, and C. E. Nauen (compositors), INFOPESCA. Catalogo de es- pecies marinas de interes economico actual o potencial para America Latina. Parte 2. - Pacifico centro y sur- oriental, p. 363-381. Roma, FAO/PNUD, SIC/82/2. P6REZ FARFANTE, I., AND B. B. BOOTHE, JR. 1981. Sicyonia martini, a new rock shrimp (Decapoda: Penaeoidea) from the American Pacific. J. Crustacean Biol. 1:424-432. PESTA, O. 1915. Die Penaeidea des Wiener naturhistorischen Hof- museums. Arch. Naturgesch. 81, Abt. A, Hft. 1:99-122. RATHBUN.M. J. 1901. The Brachyura and Macrura of Porto Rico. [Pre- print from U.S. Fish Comm. Bull, for 1900 [1902] 20(2): 1-127, 129-137 (preprint index)]. 1920. Stalk-eyed crustaceans ofthe Dutch West Indies. In J. Boeke, Rapport betreffende een voorloopig onderzoek naar den toestand van de Visscherij en de Industrie van 78 PEREZ FARFANTE: ROCK SHRIMP GENUS SICYONIA Zeeproducten in de Kolonie Curapao, ingevolge het Minis- terieel Besluit van 22 November 1904, Vol. 2, p. 317- 348. F. J. Belinfante, 'S-Gravenhage. RODRIGUEZ, G. 1980. Los crustaceos decapodos de Venezuela. Inst. Venez. Invest. Cient., Caracas, 494 p. RODRIGUEZ DE LA CRUZ, M. C. 1977. Decapodos marines del Golfo de California. Tesis Maestro en Ciencias Marinas, Escuela de Ciencias MaritimasyAlimentariasdell.T.Y.E.S.M. Inst. Tec. Es- tud. Super. Monterrey, Guaymas, Mex., 243 p. 1981. Aspectos pesqueros del camaron de alta mar en el Pacifico mexicano. Cienc. Pesq., Inst. Nac. Pesca, Mex. 1(2):1-19. ROSALES JUAREZ, E J. 1976. Contribucion al concimiento de la fauna de acom- panamiento del camaron de alta mar, frente a la costa de Sinaloa, Mexico. Mem. Reun. Recursos Pesca Costera Mex., Veracruz, Ver, 23-25 Nov. 1976, p. 25-88. ROUSE, W. L. 1969. Littoral Crustacea from southwest Florida. Q. J. Fla. Acad. Sci. 32:127-152. SAUSSURE, H., De. See De Saussure, H. SCHMITT, W. L. 1924. Crustacea (Macrura and Anomura). Expedition of the California Academy of Sciences to the Gulf of Califor- nia in 1921. XXIV Proc. Calif. Acad. Sci., Ser. 4, 13:381- 388. 1935. Crustacea Macrura and Anomura of Porto Rico and the Virgin Islands. In Scientific Survey of Porto Rico and the Virgin Islands, p. 125-227, 255-262. N.Y. Acad. Sci. 15, Part 2. SOTO, L. A. 1980a. Decapod crustacean shelf-fauna of the Campeche Bank: fishery aspects and ecology. Proc. Gulf Caribb. Fish. Inst. 32d Annu. Sess., p. 66-81. 1980b. Decapod Crustacea shelf-fauna of the northeastern Gulf of Mexico. Ann. Cent. Cienc. Mar Limnol., Univ. Nac. Auton. Mex. 7(2):79-110. SOSA HERNANDEZ, R, J. L. HERNANDEZ AGUILERA, AND J.L. VILLALOBOS HIRIART. 1980. Estudio prospective de los crustaceos (Decapoda y Stomatopoda) del Golfo de Tehuantepec, Mexico. Invest. Oceanogr. B-80-10, Seer Mar, Mex., 50 p. STEBBING, T. R. R. 1914. South African Crustacea (Part VII. of S. A. Crus- tacea, for the marine investigations in South Afri- ca). Ann. S. Afr Mus. 15:1-55. STEINBECK, J., AND E. RICKETTS. 1941. Sea of Cortez. The Viking Press, N.Y, 598 p. STIMPSON, W. 1871. Notes on North American Crustacea, in the museum of the Smithsonian Institution. No. III. Ann. Lyceum Nat. Hist. N.Y, 10:92-136. [Also reprinted with Notes. . . I and II as a separate publication with continuous pagination; the pagination of part III being: 119-163.] V6LEZ, J., J. ZEBALLOS, AND M. MENDEZ G. In press. Changes in the distribution pattern and abun- dance of some fishes and crustaceans in the central coast of Peru during the 1982-83 El Niiio. Tropical Ocean- Atmosphere Newsletter. WASS, M. L. 1955. The decapod crustaceans of Alligator Harbor and adjacent inshore areas of northwestern Florida. Q. J. Fla. Acad. Sci. 18:129-176. WELLS, H. W. 1961. The fauna of oyster beds, with special reference to the salinity factor Ecol. Monogr. 31:239-266. WENNER, E. L., AND D. F BOESCH. 1979. Distribution patterns of epibenthic decapod Crus- tacea along the shelf-slope coenocline. Middle Atlantic Bight, U.S.A. Bull. Biol. Soc. Wash. 3:106-133. WENNER, E. L., AND T. H. READ. 1981. Distribution and assemblages of decapod crusta- ceans from the continental shelf of the South Atlantic Bight: 1977-1979 MARMAP Investigations. Mar Resour Res. Inst., S.C. Wildl. Mar Resour. Dep., Charleston, Tech. Rep. 49, 41 p. 1982. Seasonal composition and abundance of decapod crustacean assemblages from the South Atlantic Bight, USA. Bull. Mar. Sci. 32:181-206. WICKSTEN, M. E. 1980. Mainland and insular assemblages of benthic decapod crustaceans of southern California. In D. M. Power (editor). The California islands: Proceedings of a multidisciplinary symposium, p. 357-367. Santa Bar- bara Mus. Nat. Hist., Santa Barbara, Calif WILLIAMS, A. B. 1965. Marine decapod crustaceans of the Carolinas. U.S. Fish Wildl. Serv, Fish. Bull. 65:1-298. 1984. Shrimps, lobsters, and crabs of the Atlantic coast of the eastern United States, Maine to Florida. Smithson. Inst., 550 p. WORD, J. Q., AND D. K. CHARWAT. 1976. Invertebrates of southern California coastal waters. II. Natantia. S. Calif. Coastal Water Res. Proj., 238 p. YOUNG, J. H. 1959. Morphology of the white shrimp Penaeus setiferus (Linnaeus 1758). U.S. Fish Wildl. Serv., Fish. Bull. 59:1-168. ZARIQUIEY ALVAREZ, R. 1968. Crustaceos decapodos ibericos. Cons. Superior In- vest. Cient., Patronato Juan de la Cierva, Invest. Pesq. 32, 510 p. ZYZNAR, E. S. 1970. The eyes of white shrimp, Penaeus setiferus (Lin- naeus) with a note on the rock shrimp, Sicyonia breviros- tris Stimpson. Contrib. Mar. Sci. 15:87-102. 1 79 VARIABILITY IN DIMENSIONS OF SALMONID OTOLITH NUCLEI: IMPLICATIONS FOR STOCK IDENTIFICATION AND MICROSTRUCTURE INTERPRETATION John D. Neilson.i Glen H. Geen ^ and Brian Chan^ ABSTRACT Sagittal otoliths in rainbow trout, Salmo gairdneri, and chinook salmon, Oncorhynchus tshawytscha, arise by fusion of otolith precursors (primordia) before hatching. Size of the otolith nucleus exhibited considerable variability even in the progeny of a single female. Otolith nucleus length was directly related to the number and position of the primordia and water temperature at which the eggs were incubated. This variability limits the utility of nucleus dimensions as criteria for separating sympatric populations of juvenile steelhead and rainbow trout. Variability in otolith nucleus dimensions also accounted for a significant error in otolith size-fish size relationships in recently hatched alevins. The early development of otoliths is poorly un- derstood considering their potential use in stock identification (Postuma 1974; Rybock et al. 1975) and in the provision of data on fish age and growth to the daily level of precision (Pannella 1971; Wilson and Larkin 1982). Variability of otolith nucleus size and shape is of particular concern in stock identification studies since nucleus dimen- sions may be racial characteristics. Rybock et al. (1975) have suggested a positive correlation of the rainbow trout, Salmo gairdneri, otolith nucleus size and the mean egg size of the female which, in turn, is positively correlated to the size of the female. Their data on Deschutes River steelhead trout (the sea-run form of S. gairdneri) females, which were larger, on average, than females of the sympatric population of freshwater resident rain- bow trout, led to the suggestion that otolith nu- cleus dimensions would differ significantly and provide a basis for racial identification of juveniles. This hypothesis was of particular significance since no other meristic or morphometric trait is known which permits identification of juvenile sea-run and freshwater resident S. gairdneri. Nucleus dimensions might affect the widths of concentrically formed daily growth increments deposited around the otolith nucleus. Bipartite daily growth increments consist of alternating ^Department of Biological Sciences, Simon Fraser University, Bumaby, B.C. V5A 1S6; present address: Marine Fish Division, Canada Department of Fisheries and Oceans, Biological Station, St. Andrews, N.B., Canada EOG 2X0. ^Department of Biological Sciences, Simon Fraser University, Bumaby, B.C., Canada V5A 1S6. ^British Columbia Fish and Wildlife Branch, Ministry of Envi- ronment, Kamloops, B.C., Canada V2C 5Z5. protein and calcium rich zones (Brothers 1981), and their widths are proportional to fish growth during the period of increment formation (Wilson and Larkin 1982). If increment width and number vary as a function of nucleus size and shape, then a source of the 15% error described by Wilson and Larkin in the estimation of fish growth from otolith growth could be identified. In this paper, we describe development of sagit- tal otoliths of S. gairdneri (sea-run and fresh- water resident) and chinook salmon, Oncorhyn- chus tshawytscha, and examine the effect of water temperature on otolith nucleus dimensions. These data permit a reexamination of the hypothesis of Rybock et al. (1975). Finally, the implications of variability in otolith nucleus size on otolith microstructure and its interpretation are con- sidered. METHODS To study otolith nucleus development in S. gairdneri, we obtained eggs from steelhead trout in the Deadman River, British Columbia (B.C.), in 1981 and from the Nicola and Deadman Rivers in 1982 (Thompson River tributaries). Rainbow trout eggs were taken from the Deadman River in 1981, and from stocks in Mission Creek and Pennask Lake in south-central B.C. in 1982. Prior to fertili- zation, samples of eggs (n = 20) were taken for dry weight determination (17 of 18 fish collected in 1982). In all cases, eggs were fertilized with pooled sperm from 2 to 3 males of similar size and origin as the female. In total, eggs from 10 steelhead and 11 rainbow trout were used in this study. Manuscript accepted April 1984. FISHERY BULLETIN: VOL. 83, NO. 1, 1985. 81 FISHERY BULLETIN: VOL. 83, NO. 1 The fertilized eggs of each female were incu- bated in separate compartments in Heath Trays at Abbotsford and Loon Lake trout hatcheries. In 1981, fertilized eggs from two female steelhead and one female rainbow trout were subdivided into three lots and held at 6.5° 9.5° and 15.0°C until yolk-sac absorption. In 1982, all fish were held at 11°C. An approximate 12:12 LD photoperiod was maintained through incubation and rearing. Samples of steelhead and rainbow trout eggs or alevins were taken at biweekly intervals in 1981. Alevins only were sampled in 1982. Oncorhynchus tshawytscha eggs were taken from the 1981 Capilano River stock and were incu- bated at 6°C under an approximate 12:12 LD photoperiod. Hatchery practice did not allow sepa- rate rearing of groups of eggs from individual females. Otolith development in S. gairdneri embryos was studied by dissecting the embryo from the egg, clearing it with carbol xylol, and then squashing the embryo between two microscope slides. This treatment, which made noncalcified tissue trans- parent and amorphous compared with otoliths and other hard parts, permitted otolith examination with a transmitted light microscope at 400 x. While we also examined embryos with X-ray and xeroradiographic techniques, satisfactory results were obtained more simply with the carbol xylol treatment. Examination of the nuclei of otoliths from ale- vins required that otoliths be ground and polished following the method of Neilson and Geen (1981). The extent of the otolith nucleus in both embryos and alevins was delimited by the first growth in- crement encircling all central otolith precursors or primordia (Fig. 1). The first growth increment en- circling the central primordia generally appeared dark when viewed with a transmitted light micro- scope. The only primordium outside the nucleus was in the anterior-ventral quadrant and was as- sociated with the formation of the rostrum, the pointed anterior extremity of the otolith showti in Figure 1. To avoid bias, otolith nucleus length was mea- sured from coded preparations with an ocular Nucleus length Primordium Rostral primordium *' I lOOiJm Figure l.— SagitUl otolith from a Capilano River chinook salmon, Oncorhynchus tshawytscha, alevin showing the otolith nucleus, primordia, and rostral primordium. 82 NEILSON ET AL.: DIMENSIONS OF SALMONID OTOLITH NUCLEI micrometer along the longest axis through the nuclear zone. The area of the otolith nucleus was measured from photographic enlargements with a polar planimeter. Increment widths were mea- sured with a vernier caliper from photographic enlargements (final magnification 9700 x). The frequency of increment formation was determined from slopes of regressions of increment counts from otoliths offish of known age. Nucleus measurements and primordia counts are only reported for otoliths removed from the fishes' left side as nucleus lengths were signifi- cantly greater in left-side than right-side sagittae, albeit at a low level of significance {P < 0.10, Wilcoxon Paired Sample Test). During the course of this study, otoliths from 257 rainbow trout, 187 steelhead trout, and 50 O. tshawytscha were examined. RESULTS To examine the hypothesis that egg size (a func- tion of female fork length) influences otolith nucleus length in progeny, we examined the rela- tionship of female fork length to egg dry weight and nucleus length in S. gairdneri. The dry weight of steelhead and rainbow trout eggs was positively correlated with the size of the female from which the eggs originated (r^ = 0.54, P < 0.001, Fig. 2). The slope of the geometric mean regression shown 0.042 -1 0.040- 0.038- 0.036 D) 0.034- s: ^ 0.032 >. d Q) 0.030 o> LlJ c g 0.028 0.026- 0.024 0.022- 0,020 y=0.2204x10-^(x)+0.0198 200 300 400 500 600 700 800 900 Fork Length (mm) Figure 2. — Geometric mean regression of mean unfertilized egg dry weight on fork length of female Salmo gairdneri from which eggs were obtained. Each point is the mean of 20 eggs from each female. Fish in the 300-400 mm size interval were rainbow trout from Pennask Lake, those 500-600 mm were rainbow trout from Mission Creek, and those >700 mm were Deadman or Nicola River steelhead. 83 FISHERY BULLETIN: VOL. 83, NO. 1 in Figure 2 differed significantly from zero (t-test, P < 0.001). However, there was no significant rela- tionship between otolith nucleus length and fe- male fork length {t-test, P > 0.05, Fig. 3), or egg dry weight {t-test, P > 0.10). We also investigated the utility of otolith nucleus lengths as a racial characteristic by calculating D^, a part of a dis- criminant function analysis. In this instance, D^ is a measure of the power of discrimination of nucleus length in separating juvenile sea-run and freshwater S. gairdneri. D^ was 0.063 and was not significant {P > 0.1). A major source of the variability in the otolith nucleus length-female parent length relationship (Fig. 3) was apparently related to the ontogeny of otolith nuclei in the salmonid embryos. Otolith nuclei result from the fusion of primordia. Pri- mordia, the first calcified structures to arise in S. gairdneri during embryonic development, ap- peared at 115-214 Centigrade degree-days. Indi- vidual primordia increase in size by concentric accretions, ultimately fusing with neighboring 0.24 n 0.22 ^ ^ 0.20 - Z UJ o o 0.18- O X I— o z UJ _J Ul o § 0.16- 0.14- 0.12 0.10- 0.08- 0.06 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •>• • • • •> • • • • • • • • • • • • •• • • • • • • • !• • • • • • • •>• • • • '• • • • • • • • •>• • • • •> • • • • • • • • • • • • • • • • • I I r -1 1 1 1 300 400 500 600 700 800 900 FEMALE PARENT FORK LENGTH (mm) Figure 3. — Scatter plot of Salmo gairdneri female parent size on otolith nucleus length of progeny. The origin of the adults is given in the caption of Figure 2. 84 primordia to form the nucleus of the otolith at 226-241 degree-days (Fig. 4). Hatching occurred at about 320 degree-days. The pattern of nucleus de- velopment was similar in both rainbow and steelhead trout. Although we did not follow otolith development in O. tshawytscha, examination of their nuclei suggested that they also arose from fusion of multiple primordia. Deposition of growth increments commenced immediately after fusion. The number of primordia fusing to form the otolith nucleus in the salmonid species we examined was variable, even within the progeny of a single female. In rainbow trout, there was an average of 8. 2 ± 2.7 primordia (±1 standard devia- tion indicated). In steelhead trout and O. tshawytscha numbers of primordia averaged 10.7 ± 2.4 and 10.1 ± 2.7, respectively. There were no significant differences in mean primordia counts among the three stocks of rainbow trout or the two stocks of steelhead trout examined (analysts of variance, P > 0.05). Figure 5 shows the relation- ship between the number of primordia deposited and otolith nucleus length. The variable location of primordia within the nucleus also affects nuclear dimensions and further increases variability. In some instances (<5%), primordia were formed at the periphery of the nucleus, resulting in a local distortion of otherwise regular growth increments (Fig. 6). Otolith nucleus length (mm) ±1 SE in S. gairdneri from the Deadman River was also af- fected by incubation temperature as shown below: Water temperature 6.5°C 9.5°C 15.0°C Mean nucleus length (mm) Rainbow trout 0.142 ± 0.009 0.174 ± 0.009 0.172 ± 0.008 Steelhead trout 0.154 ± 0.004 0.197 ± 0.008 0.191 ± 0.005 One-way analysis of variance and the Student- Newman-Keuls test indicated that the mean otolith nucleus length in rainbow or steelhead trout reared at 6.5°C was significantly less (P < 0.01) than at 9.5° or 15.0°C, although no significant differences in otolith nucleus length (P > 0.05) existed in fish reared at the two higher tempera- tures. The number of primordia formed in both Deadman River steelhead and rainbow trout was independent of the water temperature at which the eggs and alevins were incubated (analysis of variance, P > 0.05). NEILSON ET AL.: DIMENSIONS OF SALMONID OTOLITH NUCLEI / Figure 4. — Deadman River steelhead trout sagittal primordia before fusion (right, 214 degree-days) and after fusion (left, 331 degree-days). Bar = 10 /xm. We determined the effect of nucleus size varia- tion on otolith size by examining correlations be- tween nucleus area and otolith area at several stages of development of steelhead trout and O. tshawytscha of similar size. We chose to report nucleus area in this case, as it reflects nucleus dimension more precisely than one-dimensional measurements such as nucleus length. While nucleus area and length are significantly corre- lated (P < 0.001), nucleus length accounted for only 47 and 52% of the variability in nucleus area in steelhead trout and O. tshawytscha, respec- tively. The best correlations between nucleus area and subsequent otolith area were noted in rela- tively small otoliths of recently hatched alevins. The greatest degree of variability in otolith area occurred up to 15 d after nucleus formation (Table 1). Table l. — Coefficients of variability in otolith area at several stages of development, and coefficients of determination for re- gressions of otolith area at several stages of development. N = 15 for both steelhead trout and Oncorhynchus tshawytscha. The steelhead trout were 29-30 mm FL, and O. tshawytscha 30-31 mm. Trout were reared at 9.5°C and O. tshawytscha at 6°C. Steelhead trout 0. tshawytsctia Coefficient of Coefficient of Coefficient determination Coefficient determination ( Df variation (r^) when of variation (r^) when stage of otolith in otolith regressed on in otolith regressed on development area (%) nucleus area area (%) nucleus area Otolith area at nucleus formation 33 n/a 23 n/a Otolith area 15 d after nucleus formation 15 0.41" 14 0.62" Otolith area 35 d after nucleus formation 6 0.21 NS 10 0.21 NS Otolith area 50 d after nucleus formation 7 0.16 NS 11 0.15 NS " = P sO.01. NS = not significant (P > 0.05). 85 FISHERY BULLETIN: VOL. 83, NO. 1 .275 .250 - .225 - .200 - .175 - .150 - .125 .100 y=0.006(x) + 0.104 r2 = 0.34 •(2) J L £ .225 I- .200 O -z. UJ .175 CO 3 .150 LU d .1251- 3 Z .100 .225 .200 .175 .150 .125 .100 y=0.006{x)+0.105 r^ = 0.32 y=0.006(x)+ 0.114 r^ = 0.26 •12) • (2) • (2) 4 8 12 16 NUMBER OF PRIMORDIA PER SAGITTAL OTOLITH Figure 5. — Geometric mean regressions of number of primordia per sagittal otolith on otolith nucleus length for steelhead trout (top), rainbow trout (middle), and Capilano River Oncorhynchus tshawytscha (bottom). Trout were incubated at 9.5°C and salmon at 6°C. 86 NEILSON ET AL.: DIMENSIONS OF SALMONID OTOLITH NUCLEI Figure 6. — Development of a steelhead trout otolith nucleus resulting from a peripheral primordium (top) and the typical pattern of nucleus development (bottom). Note compression of otolith growth increments in the postrostral quadrant. Otoliths were from progeny of the same female parent. We did not find any correlation between mean increment width through the various stages of development and nucleus area in either species «-test, P > 0.05). In addition, examination of re- gressions of increment counts on nucleus area indicated that the frequency of increment forma- tion did not vary as a function of nucleus dimen- sion (P > 0.10 for both S. gairdneri and O. tsha- wytscha). DISCUSSION Sagittal otoliths in S. gairdneri embryos arise by fusion of primordia, the first calcified struc- tures to appear during development (McKern et al. 1974). Radtke and Dean (1982) reported similar results for mummichogs, Fundulus heteroclitus, and also noted that the otolith nucleus was first apparent as an amorphous gel-like mass in the area of the labyrinth in the developing larvae. Calcified primordia appeared later although Radtke and Dean did not describe any variability in their number or position. The number and position of the primordia were variable, even within the progeny of a single female. This variation affected the extent of the otolith nucleus. In addition, we observed that water temperature influenced nucleus size. The observed variation in nucleus size limits the util- ity of this feature as a criterion for stock identifica- tion. However, differences in nucleus size did not affect the number of growth increments sub- sequently formed and had no significant influence on their width. In our studies eggs were fertilized with the pooled sperm of several males. It is possible that the observed variability in otolith nucleus size was related to the differences between the male par- ents. There was little difference in the size of the males used, either within the group or relative to the females. We cannot rule out genetic differ- ences between males as a factor affecting variabil- ity in nucleus size. However, any genetic effects influencing our results would be no greater than would be expected in natural populations. The numbers of males from which sperm was pooled was usually three, a number frequently involved in fertilization of eggs of a single female in nature (Schroeder 1982; Gross in press). In developing a hypothesis to explain the basis for use of otolith nucleus length as a means of distinguishing races, Rybock et al. (1975) suggested that nucleus length was related to egg size, although no data were presented. While we found that greater nucleus lengths were as- sociated with larger eggs on average, and larger eggs originated from larger female parents, the slope of the regression of nucleus length on egg weight was not significant (Fig. 3). Furthermore, the variability of otolith nucleus dimensions in rainbow and steelhead trout from south-central B.C. made their measurement much less useful for stock identification that has been suggested for S. gairdneri from the Deschutes River, Oreg. (Rybock 87 FISHERY BULLETIN: VOL. 83, NO. 1 et al. 1975). However, otolith nucleus dimensions did serve to separate summer and winter races of steelhead trout (McKern et al. 1974). Workers proposing to use otolith nucleus dimensions as stock identification criteria should consider rear- ing fish under controlled conditions to establish the extent of nucleus size variability in the stocks in question. Otolith nucleus length is also influenced by water temperature during embryonic develop- ment. Our data showed an increase of about 25% in length in fish reared at 9.5° or 15°C relative to that observed in fish incubated at 6.5°C. The sen- sitivity of otolith nucleus length to water tempera- ture may allow separation of selected fish stocks whose eggs are incubated at different water tem- peratures. For example, O. tshawytscha juveniles originating from Campbell River stock reared in the Canada Department of Fisheries and Oceans Quinsam Hatchery on Campbell River had sig- nificantly greater otolith nucleus lengths (P < 0.01) than wild Campbell River O. tshawytscha incubated in cooler waters (M. Bradford pers. commun.^). Increased water temperature may in- fluence nucleus length through a greater rate of accretion of the calcium/protein matrix around primordia, reflecting a faster rate of embryonic development. The definition of otolith nucleus suggested here can be consistently applied. With relatively simple preparation techniques, otolith nucleus dimen- sions can be measured from micrographs or by using a light microscope equipped with an ocular micrometer Previous workers have delimited the otolith nucleus in relation to metamorphic or nuclear checks. Such terms are ill-defined and should be avoided since they imply that otolith checks result from important developmental events. While it seems likely that such events may result in growth interruptions or checks, causal links have not yet been demonstrated. The imprecise definition of the periphery of the otolith nucleus may reduce the comparability of measured dimensions derived in various studies. While we have defined the nucleus as lying within the first increment surrounding the primordia, several checks occur during early otolith develop- ment. Use of one of these checks to define the periphery of the nucleus would result in inconsis- tency between various investigations. For exam- ple, nucleus lengths of steelhead trout used in this *M. Bradford, Department of Biological Sciences, Simon Fraser University, Burnaby, B.C., Canada V5A 1S6, pers. com- mun. November 1983. 88 study were generally <0.2 mm (Fig. 3). The mean diameter of the otolith nucleus of summer and winter steelhead reported by McKern et al. (1974) were 0.348 and 0.436 mm, respectively. Differ- ences between studies of this magnitude may be racial in nature or may reflect differences in defi- nition of the extent of the nucleus. Data on variation in primordia number and lo- cation have not been reported previously although the existence of primordia was described by Radtke and Dean (1982) in mummichogs. McKern et al. (1974) did not describe primordia in their work involving the otolith nucleus in steelhead trout. Their results were based on the use of X-ray techniques. We were not able to detect primordia using this method. It is likely that the otoliths of many fish species are formed by fusion of multiple primordia. From our observations, this is apparently the case in all five species of Pacific salmon and the Pacific her- ring, Clupea harengus pallasi. Radtke and Dean (1982) noted multiple primordia in masou salmon, O. masou; Arctic char, Salvelinus alpinus; brook trout, S. fontinalis; and the sculpin, Cottus nozawa. While both steelhead trout and O. tshawytscha otolith nucleus areas were variable, otolith areas in older fish (longer than 15 d after primordia fusion) were less so as indicated by the decreasing coefficient of variation of otolith area with increas- ing age (Table 1). The decreased variation proba- bly reflects the development of otoliths from an indeterminant array of primordia to the otoliths of adult fish, the latter considered a species-specific characteristic (Fitch 1968; Morrow 1979). How- ever, variation in otolith development in the juvenile salmonids studied here do not present difficulties for the interpretation of microstruc- ture as neither the number nor width of growth increments is significantly affected by nucleus size variation. ACKNOWLEDGMENTS The cooperation of the staff at British Columbia Department of the Environment fish hatcheries at Abbotsford, Loon Lake, and Summerland is grate- fully acknowledged. In particular, we wish to thank John Cartwright, Dennis Graf, Chris Hous- ton, Bob Land, Don Peterson, and Hugh Sparrow for their exceptional support. Eldon Stone of the Canada Department of Fisheries and Oceans Capilano Hatchery provided the chinook salmon used in this study. Mike Bradford suggested calcu- NEILSON ET AL.: DIMENSIONS OF SALMONID OTOLITH NUCLEI lation of the D^ statistic. Barbara Puselja and Tracey Crawford assisted with otolith preparation and examination. Ron Long, Department of Biological Sciences, Simon Fraser University, pre- pared the photographs. This work was supported by a Graduate Research Engineering and Technology Award to G. H. Geen from the British Columbia Secretariat of Science, Research and Development, and a Canada Department of Fisheries and Oceans Subvention Grant. LITERATURE CITED Brothers, E. B. 1981. What can otolith microstructure tell us about daily and subdaily events in the early life history of fish? Rapp. P.-v. Reun. Cons. int. Explor Mer 178:393- 394. Fitch, J. E. 1968. Fish otoliths in cetacean stomachs and their impor- tance in interpreting feeding habits. J. Fish. Res. Board Can. 25:2561-2574. Gross, m. r. In Press. Sunfish, salmon and the evolution of alternative reproductive strategies and tactics in fishes. In R. J. Wooton and G. Potts (editors). Fish reproduction: strategies and tactics. Acad. Press, N.Y. MCKERN, J. L., H. F HORTON, AND K. V. KOSKI. 1974. Development of steelhead trout (Salmo gairdneri) otoliths and their use for age analysis and for separating summer from winter races and wild from hatchery stocks. J. Fish. Res. Board Can. 31:1420-1426. MORROW, J. E. 1979. Preliminary keys to otoliths of some adult fishes of the Gulf of Alaska, Bering Sea, and Beaufort Sea. U.S. Dep. Commer., NOAA Tech. Rep. NMFS CIRC 420, 32 p. NEILSON, J. D., AND G. H. GEEN. 1981. Method of preparing otoliths for microstructure examination. Prog. Fish-Cult. 43:90-91. PANNELLA, G. 1971. Fish otoliths: daily growth layers and periodical pat- terns. Science (Wash., D.C.) 173:1124-1127 POSTUMA, K. H. 1974. The nucleus of the herring otoliths as a racial character. J. Cons. Int. Explor. Mer 35:121-129. RADTKE, R. L., AND J. M. DEAN. 1982. Increment formation in the otoliths of embryos, lar- vae, and juveniles of the mummichog, Fundulus hetero- clitus. Fish. Bull., U.S. 80:201-215. RYBOCK, J. T, H. F. HORTON, AND J. L. FESSLER. 1975. Use of otoliths to separate juvenile steelhead trout from juvenile rainbow trout. Fish. Bull., U.S. 73:654- 659. SCHROEDER, S. L. 1982. The influence of intrasexual competition on the dis- tribution of chum salmon in an experimental stream. In E. L. Brannon and E. O. Salo (editors), Salmon and trout migratory symposiiun, p. 275-285. School Fish. Univ. Wash. Wilson, k. h., and p a. larkin. 1982. Relationship between thickness of daily grovrth in- crements in sagittae and change in body weight of sockeye salmon (Oncorhynchus nerka) fry. Can. J. Fish. Aquat. Sci. 39:1335-1339. k 89 EFFECTS OF FEEDING REGIMES AND DIEL TEMPERATURE CYCLES ON OTOLITH INCREMENT FORMATION IN JUVENILE CHINOOK SALMON, ONCORHYNCHUS TSHAWYTSCHA John D. Neilson' and Glen H. Geen'^ ABSTRACT The effects of constant and diel cyclic water temperature regimes, feeding frequency, fish activity, and ration level on growth increment formation in juvenile chinook salmon, Oncorhynchus tshawytscha, are described. Of the variables examined, any event which recurred more than once every 24 hours increased the rate of increment production above 1 increment per 24 hours. The results were consistent with the hypothesis that environmental variables modify the rate of increment formation by altering the periodicity of fish activity. Both water temperature and ration level interacting with water temperature affected otolith increment width, a measure offish growth, although ration level did not. To realize the potential of otolith microstructure in detailed age and growth studies of fishes, knowledge of factors influencing otolith growth is required. The principle features of otoliths likely to be used in such studies are the growth incre- ments which are deposited in a concentric fashion around the otolith nucleus. The frequency of depo- sition (often 1 increment/24 h) and the width of the increments are both affected by environmental conditions (Neilson and Geen 1982). The growth increments result from accretion of CaCO^ and to a lesser extent, protein (Simkiss 1974). The daily nature of their deposition ob- served by many workers (Pannella 1971; Brothers et al. 1976; Wilson and Larkin 1982; and others) appears related to a daily rhythm in the relative rates of calcium carbonate and protein deposition (Mugiya et al. 1981). The cyclic deposition of cal- cium and protein over a 24-h period results in the formation of the bipartite features now referred to as daily growth increments. The effects of environmental variables on otolith increment formation have been the subject of some controversy. Taubert and Coble (1977) con- cluded that a 12:12 LD photoperiod was responsi- ble for entraining diel rhythms in the growth of juvenile Lepomis and Tilapia sp. otoliths. How- ever, juvenile starry flounder, Platichthys stel- ^ Department of Biological Sciences, Simon Fraser University, Bumaby, B.C., Canada V5A 1S6; present address: Marine Fish Division, Canada Department of Fisheries and Oceans, Biologi- cal Station, St. Andrews, N.B., Canada EOG 2X0. ^Department of Biological Sciences, Simon Fraser University, Bumaby B.C., Canada V5A 1S6. Manuscript accepted February 1984. FISHERY BULLETIN: VOL. 83. NO. 1, 1985. latus, and chinook salmon, Oncorhynchus tshawytscha , continued to produce daily growth increments when exposed to constant light (Cam- pana and Neilson 1982; Neilson and Geen 1982). The latter authors presented evidence that feed- ing frequency affected both increment number and width in O. tshawytscha . They also suggested that feeding frequency (or any other environmen- tal variable) was probably not the ultimate factor determining the frequency of otolith increment production. Environmental modulation of endo- crine rhythms (Menaker and Binkley 1981) may ultimately control otolith increment periodicity. Diel cycles in water temperature have received little consideration as an environmental variable potentially affecting increment formation. Brothers (1978) suggested that diel temperature variations were responsible for otolith increment formation in temperate stream-dwelling fish al- though no data were presented. This gap in our understanding of factors influencing otolith in- crement production is significant since diel changes in water temperature are a common fea- ture of aquatic environments. In this study we examined the effects of diel water-temperature regimes on formation of otolith growth increments in O. tshawytscha alevins and fry. We also present data on the effects of interactions of water- temperature regimes, feeding frequency, and ra- tion level on otolith increment formation in O. tshawytscha fry. Finally, we tested the suggestion made earlier (Neilson and Geen 1982) that feeding periodicity (or any other periodic event affecting fish activity) modifies the rate of otolith increment production through changes in fish activity. 91 FISHERY BULLETIN: VOL. 83, NO. 1 METHODS Alevins Fish used in the experiments described below originated from the 1981 brood of the Canada De- partment of Fisheries and Oceans Capilano Hatchery. Eggs were transferred to incubation facilities at Simon Fraser University at the "eyed" stage of development, corresponding to 347 Cel- sius degree-days. Prior to transfer, the eggs were held under a 12:12 LD photoperiod and at a con- stant 8°C water temperature. The eggs were held for 5 d in our laboratory at 8.5°C before exposure to diel water-temperature regimes. Two lots of 100 fish were exposed as eggs and later, as alevins, over a 69-d period to a water- temperature regime whose diel amplitude aver- aged 2" and 4°C (range 1.8°-2.4° and 3.0°-4.5°C) above a daily average minimum temperature of 8.5°C. These temperatures were similar to those observed in May-June 1981 in the Deadman River, B.C., a stream supporting an O. tshawytscha population. All eggs hatched by day 29. Eggs or alevins in = 10) were sampled at days 19, 40, 55, and 69. On day 39, 20 alevins were transferred from a temperature regime with a 4°C amplitude and 24-h period to a regime with the same temper- ature amplitude but a 12-h period. A fourth group was held at a constant 8.5°C. The constant water temperature corresponded to that of the cool period of the diel water-temperature regimes. Sagittal otoliths were removed from preserved fish and prepared following the methods of Neilson and Geen (1981). Otolith sections were examined using a light microscope or scanning electron mi- croscope (SEM) as described in Neilson and Geen (1982). Fry Fry used in these experiments were about 90-d- old posthatch and originated from Capilano River hatchery stock. Prior to transfer to 25 1 aquaria at Simon Fraser University, fish were held under natural light at a constant 8°C and fed once every 24 h. After transfer to our laboratory, fry were held for 2 wk in flow-through aquaria supplied with aerated and dechlorinated water at 6°C before ex- periments commenced. During this period the 50 fish in each aquarivun were fed to satiation with Oregon Moist Pellets once per 24 h and exposed to a 12:12 LD photoperiod. Experimental feeding and temperature regimes to which fry were exposed are summarized in Table 1. Amplitude of daily temperature fluctua- tions was 4°C (range 3.6°-4.4°C) above the average minimum of 6°C. The diel temperature cycle in relation to photoperiod and feeding events is shown in Figure 1. The activity of one group offish was artificially increased to examine the effects of activity on otolith increment formation. These fish were forced to evade a slowly moving aquarium net for 10-min beginning at 1900 h daily. The in- duced activity level appeared similar to that as- sociated with feeding. Ration provided to experi- mental lots of fish was maintained as a constant proportion (4% or 8% ) of average fish dry weight by 105- 9.5- O 0) 2 « a. E ® 0) IS $ w 8 5" 7.5- 6.5- 5.5 0100 0500 f 1000 1500 f 2000 0100 0600 f 1 1 00 1 600 t 2100 2400 Time (24-h clock) Figure \. — Did water-temperature cycle in relation to photoperiod and feeding events (f) for Oncorhynchus tshawytscha fry. Light and dark periods are indicated by the open and solid bars respectively. 92 NEILSON and GEEN: CHINOOK SALMON OTOLITH INCREMENT FORMATION Table L — List of abbreviations denoting experimental regimes to which On- corhynchus tshawytscha fry were exposed in 1982. Percent ration (% of body weight offered every 24 h ) is given and the water temperature at time of feeding dur- ing the diel cycle, if applicable, is indi- cated in parentheses. Refer to Figure 1 for details of feeding, temperature, and photoperiod regimes. Time of feeding Treatment (h) 8% (warm) 0700 8% (cool) 1900 8% (constant)' 0700 4% (warm) 0700 4% (cool) 1900 4% (constant)' 0700 2 X 4%2 0700 and 1900 2 X 2% 2 0700 and 1900 4% + activity^ 0700 Starvation n/a ' Fish in these treatments were held at constant temperature. ^Fish in these treatments were fed 2 times per 24 h. ^Fish in this treatment were held at a constant temperature and exposed to a 10-min bout of forced activity at 1900 h every day. adjusting total food offered as fish grew or were sampled. Every third day, excess food was removed from the aquaria within 30 min of offering, weighed, and consumption estimated. On day 26, we exposed fry for 30 min to a hyper- tonic solution of 1 g/1 sodium chloride and 40 mg/1 oxytetracycline hydrochloride. The tetracycline was incorporated into the otolith and provided a time marker which exhibited fluorescence when viewed with ultraviolet illumination. All fry were successfully marked by this method. Originally, we had intended to sample 15 fish at days 10, 20, and 40. However, an accidental inter- ruption of the dechlorinated water supply on day 19 resulted in the mortality of some fish in treat- ments 4% (cool), 4% (constant), 2 x 4%, and 4% + activity. Complete mortality of starved fish oc- curred at that time. To ensure an adequate (N ^ 10) sample on experiment completion, no samples were taken at day 20 for the above four treat- ments. Even so, only five fish remained by day 40 in the 4% (cool) treatment. Fork lengths were determined immediately after the fish were sacrificed. Fish were then dried to a constant weight (60°C for 48 h) in individual labeled containers, and weighed. Sagittal otoliths were then removed, weighed with an electrobal- ance, and prepared for examination with the SEM or a light microscope. Increment counts were conducted as described by Neilson and Geen (1982). No attempt was made to distinguish between the daily and subdaily in- crements as did Brothers (1978) and Campana (1983). Such distinctions are often based on subjec- tive appraisals of increment continuity and ap- pearance when viewed with a light microscope. We did not observe any such differences in growth increments of O. tshawytscha. Moreover, as the purpose of this study was to determine the periodicity of increment formation as a basis for detailed study of fish growth, the classification of increments as daily or subdaily was not neces- sary. RESULTS Eggs and Alevins The formation of growth increments com- menced before hatching under all experimental regimes. One increment/24 h was formed on aver- age under all temperature regimes (Table 2). No significant departure from unity was noted (anal- ysis of variance, P > 0.05). However, the appear- ance of the daily growth increments differed be- tween treatments. Otoliths offish subject to a cycle of temperature were characterized by more regu- lar and easily observed growth increments than those held under constant temperatures (Fig. 2). Examination with a SEM at 1,000 x revealed that the bipartite nature of otolith growth incre- ments differed between the temperature regimes. After etching with a weak acid (Neilson and Geen 1982), the relatively deeply etched portion of the bipartite growth increment (corresponding to the opaque portion of the bipartite structures when viewed with a transmitted light microscope) com- prised a larger average fraction of the growth in- crements (P < 0.01) in otoliths offish subjected to a diel cycle in temperature than those offish held at constant water temperatures. The lightly etched portion of daily growth increments did not differ significantly between fish held in diel temperature regimes with 2°C and 4°C amplitude (analysis of Table 2. — Summary of Oncorhynchus tshawytscha otolith in- crement counts for alevins held under various temperature re- gimes. Experiment 1 Increment count' day Constant temperature 2°C amplitude 4°C amplitude 19 17.8±2.6 18.5± 1.2 17.4± 1.0 40 — 38.0 ±2.4 39.5±2.1 55 51.82:2.8 54.1 ± 1.9 53.3 ±3.0 69 68.4 ±5.6 68.4±4.1 70.2 ±4.6 '± 1 standard deviation indicated, n = 1 0. 93 V* \ i -»k'. E :t o in FISHERY BULLETIN; VOL. 83, NO. 1 variance and the Student-Newman-Keuls test, P > 0.05). Oncorhynchus tshawytscha transferred from a 4°C diel temperature regime (24-h period) to a regime with a 12-h period and similar amplitude produced an average of 1.56 increments/24 h. The slope of the regression of mean increment count on experiment day differed significantly from unity (P < 0.01). An example of an otolith from a fish exposed to the 12-h period, cyclic temperature re- gime is shown in Figure 3 and illustrates the nar- rower increments associated with the 12-h cycle. y u o CM SO - 0) O e 6 1 « ^^ CC -4-* 2 I Fry Otolith growth increments were formed at the rate of one every 24 h in fish fed once per day. No significant departures were noted (^-tests, n ^ 20, P > 0.05). Fish which received 2 feedings/24 h or 1 feeding and a 10-min bout of activity deposited significantly >1 increment/24 h (i-tests, P < 0.01). Arithmetic mean regressions of increment counts on experiment day for the latter treatments are given below: Treatment Regression equation r^ 8% B.W. ration fed y= 1.45U)+1.58 0.91 2 times/24 h 4% B.W. ration fed J = 1.76 (x)- 1.40 0.98 2 times/24 h 4% B.W. ration and y = 1.50(x)-0.80 0.93 forced activity Slopes of regressions in groups of fish producing >1 increment/d did not differ significantly from each other (analysis of covariance, P > 0.10). The distributions of increment widths in fed groups of fish are presented in Figure 4. A sum- mary of the comparisons of increment width data among treatments is provided in Figure 5. One- way analysis of variance and the Student- Newman-Keuls test indicated that mean incre- ment widths in otoliths offish receiving a ration of 8% B.W./24 h in one feeding differed significantly between groups (P < 0.05). Mean increment widths in otoliths of fish receiving a ration of 4% B.W./24 h in one feeding did not differ signifi- cantly in fish receiving the ration either during the cool or warm portion of the diel temperature cycle (Student-Newman-Keuls test, P > 0.05). However, fish receiving 4% B.W./24 h under con- stant water temperature produced growth incre- ments whose mean width was significantly less than those of fish held in the diel water- 94 NEILSON and GEEN: CHINOOK SALMON OTOUTH INCREMENT FORMATION Figure 3. — Change in otolith microstructure in a Oncorhynchus tshawytscha alevin transferred from a 24-h period temperature cycle (4°C amplitude) to a 12-h period tempera- ture cycle (4°C amplitude). 50 ^tm temperature regimes (P < 0.05, Student- Newman-Keuls test). The top two horizontal strata of Figure 4 consti- tute a 3 X 2 factorial design and were examined with a two-way analysis of variance. The effects of time of feeding with respect to the diel tempera- ture cycle, ration level, and their interaction were examined in relation to mean otolith increment width. The effect of time of offering with respect to the diel temperature cycle on mean increment width was significant (P < 0.001), whereas ration level was not (P > 0.05). The interaction of time of offering in relation to the diel temperature cycle and ration level on otolith increment width was also significant (P < 0.001). Fish in treatments receiving 2 feedings/24 h or fed once per 24 h and exposed to a 10-min bout of activity produced growth increments whose aver- age widths were significantly less than those of fish in treatments fed the same ration once per 24 h. Treatments in which fish received either rations of 8% or 4% with 1 or 2 feedings/24 h comprise a 2x2 factorial design, and were analyzed with a two-way analysis of variance. Increased feeding frequency significantly reduced mean increment width (P < 0.001), although ration level did not (P > 0.1). The interaction of feeding frequency and ration level was not significant (P > 0.1). Widths of otolith increments formed when fish were fed 4% B.W./24 h and subjected to a 10-min bout of activity were not significantly different from widths of increments in fish which received two feedings equivalent to the 4% B.W./24 h ration level ( ^-test, P > 0.05). However, fish fed a ration of 8% B.W./24 h with two feedings produced incre- ments whose average width was significantly greater than the latter two treatments (analysis of variance and the Student-Newman-Keuls test, P < 0.01). Mean increment widths in fish fed 4% B.W./24 h and exposed to a constant water- temperature regime were compared with incre- ment widths in fish receiving the same ration plus 95 FISHERY BULLETIN: VOL. 83, NO. 1 (0 o c (1) k_ o o O > DC 50n 8% (warm mean width = 3. 77 no. of increments=386 (099) 8% (cool) 8%(constant) mean wiclth= 2.78 no. of increments =384 (0.96) mean width = 2.29 no. of increments = 36 3 (0.94) J 50-, 25- 4% (warm) mean width = 3.06 no. of increments = 37 4 (0.98) rw 0 50n 4%(cool) I2=i mean width = 3.21 no of increments = 1 80 (0.99) 4% (constant) mean width = 2.23 no. of increments = 393 (1.02) i=. 25- 0- 2 X 4% mean width = 1 66 no of increments=588 (1.45) J^ 5 10 2x2% mean width = 1 .45 no of increments = 733 (1-76) i. — r- 5 10 4% + activity mean width = 1 .34 no of increments =556 (1 50) 10 Increment Width i/jjm) Figure 4. — Distribution of otolith increment widths under the experimental regimes. Treatments are identified by numbers in the top-left comers of histograms and correspond to treatments listed in Table 1. The average rate of increment formation every 24 h is shown in brackets. 8%(warm) ». 8%(cool) 8%(constant)- 4%(warm Figure 5. — Summary of Student-Newman-Keuls or <-test (a = 0.05) comparisons of mean increment widths in Oncorhynchus tshawytscha fry held under the various experimental regimes. Arrow heads pointing left or right signify "less than" and "greater than" respectively. 96 a 10-min period of enforced activity. The mean increment width associated with the latter treat- ment was significantly less (^-test, P < 0.01). Production of the narrower growth increments associated with 2 feedings/24 h or 1 feeding and induced activity did not occur immediately upon commencement of the experimental regimes. A period of transition in otolith microstructure was evident. Figure 6 shows the decrease in increment widths with time in fish previously provided a ration of 8% B.W./24 h in one feeding and then offered the same total ration in 2 feedings/d. For comparison, data on increment widths in fish fed 8% B.W./24 h are provided (Fig. 6). The slope of the regression of increment width on date in the latter NEILSON and GEEN: CfflNOOK SALMON OTOLITH INCREMENT FORMATION 3.5n E 3- •g 5 2.5- c (0 0) 2- c 0 E b c ii.5- o O 1- 0.5- ^- '^o ■ "o ■ o 0 I 10 ~^ — I — 30 — I — 40 Experiment Day 50 Figure 6. — Mean otolith increment widths (0) for Oncorhyn- chus tshawytscha fry from the 2 x 4% experimental feeding regime over days 1-40. Prior to day 1, fish were fed once every 24 h. Also shown are mean increment widths ( ■ ) offish from the 8% (constant) feeding regime, where fish received one feeding only every 24 h. treatment did not significantly differ from zero (P > 0.10), whereas the former did (f-test, P < 0.01). Starved fry continued to produce one otolith in- crement every 24 h. However, the growth incre- ments were faint when observed with a transmit- ted light microscope. That portion of otolith growth formed under starvation conditions was more transparent than the portion of otolith growth produced when fish were fed. Growth in- crement diel periodicity was also more pronounced during the portion of otolith growth corresponding to that period when fish were fed (Fig. 7). To confirm that increment widths were propor- tional to fish growth, we plotted instantaneous growth in dry weight against average increment width for all treatments except the starved group (Fig. 8). The coefficient of determination (r^) as- sociated with those treatments in which fish formed 1 growth increment/24 h was 0.735 and the slope of the regression was significantly different from zero (P < 0.01). Note that points associated with treatments in which fish formed more than 1 increment/24 h lie considerably above the regres- sion. The regression of these data differs signifi- Fed 1x/24h % SOyum Figure 7. — Example of otolith microstructure from a starved Oncorhynchus tshawytscha salmon fry when viewed with transmitted light microscopy. The relatively transparent region near the otolith periphery corresponds to the starvation period. 97 FISHERY BULLETIN: VOL. 83, NO. 1 UJ o or o (/) Z) o to 3.0 ^ 2.5-1 2.0- 1.5- 1.0 - 0.5 0.0 8% (warm) ,4% (cool) 4%X2 >^ 4% (warm) D 4% (constant) 8% (cool) 2% X 2 4% + activity o 8% (constant) — 1 1 1 1 AC o Q ^ '^ MEAN INCREMENT WIDTH (urn) 3.5 4 Figure 8. — Regressions of mean otolith increment width versus instantaneous growth rate (dry weight) for the various experimental regimes. Solid line represents groups where fish produced one increment every 24 h on average; dashed line represents treatments where fish produced significantly more than one increment every 24 h. cantly in both slope and y-intercept (analysis of covariance and Mest, P < 0.01) from that of fish fed once per day. Slopes of arithmetic mean linear regressions of fish dry weight on experiment day indicated that the average rate of growth offish fed 8% B.W./24 h at the beginning of the warm portion of the diel temperature cycle was significantly greater than that offish fed at the beginning of the cool period of the diel temperature regime or at the constant water temperature (6°C) (analysis of covariance and the Student-Newman-Keuls test, P < 0.01). Similar analyses among treatments in which fish were fed 4% B.W./24 h [4% (warm), 4% (cool), 4% (constant)] or among fish that received two feed- ings or one feeding coupled with an additional bout of activity (2 x 4%, 2 x 2%, 4^^ + activity) indicated no significant differences in growth rate (P> 0.05). To determine whether otolith growth-fish growth relationships were similar among treat- ments, we calculated otolith weight-fish dry weight regressions for data from all experimental regimes. Analysis of covariance indicated that the 98 slopes of the predictive regressions among groups of fish fed 8% B.W./24 h and exposed to different temperature regimes did not significantly differ from each other (P > 0.1). Nor were there signifi- cant differences among treatments in which fish were fed twice/24 h or fed once/24 h and exposed to an enforced 10-min bout of activity. The slope of the regression representing the otolith weight-fish weight relationship for those fish receiving a ra- tion of 4% B.W./24 h on the cool portion of the diel cycle was significantly greater than the slopes of regressions representing fish fed 4% B.W./24 h (warm or constant) (analysis of covariance and the Student-Newman-Keuls test, P < 0.05). However, as mentioned earlier, the treatment where fish received a ration of 4*^ B.W./24 h on the cool por- tion of the diel temperature cycle was affected by an interruption in water supply. Only five fish survived to day 40 and may not have been repre- sentative offish held under those conditions. DISCUSSION Under most environmental conditions consid- NEILSON and GEEN: CHINOOK SALMON OTOUTH INCREMENT FORMATION ered during this study and reported by Neilson and G«en (1982), one otolith growth increment was formed each day. This supports the hypothesis that an endogenous rhythm influences growth in- crement formation. Earlier, Neilson and Geen (1982) reported that multiple feedings within a 24-h period resulted in the formation of >1 increment/24 h. We suggested that this resulted from the interaction of an endogenous diel rhythm of increment production and some regularly re- curring environmental event. Data presented here are consistent with that view, as increased feeding frequency, exposure to a warm/cool tem- perature cycle twice in 24 h and an enforced in- crease in fish activity were all associated with an increased rate of increment formation. The effects of at least some of these environmental events on otolith microstructure may be mediated through activity-induced modification of fish metabolism, which often follows a circadian rhythm (Matty 1978). If otolith growth increment production fol- lows a circadian rhythm that is sometimes over- lain by environmental events, it seems reasonable to assume that fish may produce one or more growth increments but not less than one every 24 h. In our studies, O. tshawytscha alevins and fry produced one or more growth increments every 24 h, a result consistent with most earlier studies. Even when fish were exposed to light and tempera- ture stimuli with periods >24 h, Campana and Neilson (1982) reported that only one increment was formed every 24 h. Diel water-temperature fluctuations were not required for otolith increment production in O. tshawytscha. However, cyclic changes in tempera- ture with a 24-h periodicity apparently result in differences in the appearance of otolith growth increments (Fig. 2). The deeply etched portion of the increments is significantly wider in otoliths of fish taken from a diel water-temperature regime than those from fish held in water of constant temperature. Mugiya et al. (1981) concluded that the deeply etched portions of goldfish, Carassius auratus, otoliths have a relatively high concentra- tion of protein relative to calcium carbonate. Degens et al. (1969) suggested that the deposition of the organic matrix is not readily modified by environmental events. If these results are appli- cable to salmonids, the greater contrast in otoliths of fish reared under a diel temperature regime may result from changes in the rate of calcium carbonate deposition. However, the presumed change in composition and structure of daily growth increments produced under various en- vironmental conditions does not affect the incre- ment width-fish growth relationship illustrated in Figure 8. Interactions between ration level and time of feeding with respect to the 24-h temperature cycle affected mean increment width. Ration level as a single factor influencing increment width was not significant. However, the interaction between temperature and ration on increment width was not significant suggesting higher calcium car- bonate deposition on the otolith when temper- atures were elevated at time of feeding. In a two-way comparison with ration level and feed- ing frequency (water temperature was constant), increment width was affected by feeding fre- quency but not by ration level. This agrees with the results of Neilson and Geen (1982) who showed that the rate of increment production is affected by feeding frequency. Mean increment width reflected fish growth under a variety of water-temperature and ration regimes (Fig. 8). However, different equations de- scribed increment width-growth relationships under conditions that produced 1 increment/d or >1 increment/d (Fig. 8). The extent to which in- crement width data can be used to predict instan- taneous growth rates in natural populations re- mains to be examined. Increment widths can provide an indicator of environmental changes and consequent alteration of growth rates. However, such changes, at least under laboratory conditions, did not occur rapidly (Fig. 6). These data suggest that at least 3 wk would be required before the change in increment width would be statistically detectable. Our data indicate that otolith weight-fish weight regressions are similar under a range of experimental conditions suggesting that otolith growth in salmon fry is closely coupled to fish growth. Marshall and Parker (1982) also reported that differences in ration and water temperature did not significantly affect slopes of otolith size- fish size regressions among fed sockeye salmon, O. nerka, fry. Exceptions to the isometric growth re- lation between fish size and otolith size have only been observed in recently hatched salmonid ale- vins (Neilson unpubl. data) and in starved O. tshawytscha fry. Fry deprived of food for 19 d con- tinued to form daily growth increments. Assum- ing fish dry weight did not increase over this period, then the slope of the otolith weight-fish weight regression would probably be greater than for fed fish. Marshall and Parker (1982) also re- ported continued otolith growth in O. nerka fry 99 FISHERY BULLETIN: VOL. 83, NO. 1 over a 2-wk starvation period. Evidently con- tinued otolith growth in starved fish resulted from the metabolism of stored energy reserves. Estimates of food consumption indicated that fish held under diel cyclic temperatures and fed 89c B.W./24 h consumed significantly more food per gram of fish when the food was offered during the warm period (^-test, P < 0.01). No differences in food consumption were noted in fish receiving a 4% B.W./24 h ration on either the warm or cool portion of the diel water-temperature regime (^-test, P > 0.05). It is likely that fish were not satiated at this ration under either water- temperature regime. Under the high ration, fish were satiated even when the food was offered during the cool period of the water-temperature cycle. Additional con- sumption occurred only if food was offered during the warm portion of the diel temperature cycle. The additional food consumption was associated with increased growth rates. It is not clear whether the increased growth was simply a re- sponse to differences in food consumption or also reflected enhanced efficiency of food utilization in fish exposed to cyclic temperatures similar to that described by Brett (1979) and Biette and Geen (1980). Differences in growth rate of fish fed 4% B.W./24 h strongly suggest more efficient food utilization in fish exposed to a cyclic temperature regime. Food consumption did not differ although growth rates (and increment widths) are signifi- cantly greater Given that water temperature and food con- sumption are considered the most important fea- tures of fishes' environment affecting their growth (Paloheimo and Dickie 1966), it is not surprising that water-temperature regimes and ration levels influence otolith growth increment production. Our findings and those of English (1981) suggest that interpretation of prey abundance and feeding success from otolith microstructure data may be masked by relatively small changes in water temperature. Workers attempting to quantify fish growth with respect to ration size through exami- nation of otolith microstructure should be aware of the effects of water temperature documented here and design studies accordingly. ACKNOWLEDGMENTS We thank Eldon Stone of the Canada Depart- ment of Fisheries and Oceans Capilano Hatchery for providing the fish used in this study Tracey Crawford and Jeff Johansen assisted with otolith preparation. Ron Long, Department of Biological 100 Sciences, Simon Eraser University, prepared the photographs. The critical review of an earlier draft of this manuscript by S. Campana and M. Healey is particularly appreciated. This work was sup- ported by grants from Canada Department of Fisheries and Oceans and the Natural Sciences and Engineering Research Council to Glen H. Geen. LITERATURE CITED Biette, R. M., and G. H. Geen. 1980. Growth of underyearling sockeye salmon (On- corhynchus nerka) under constant and cyclic tempera- tures in relation to live zooplankton ration size. Can. J. Fish. Aquat. Sci. 37:203-210. Brett, j. r. 1979. Environmental factors and growth. In W. S. Hoar, D. J. Randall, and J. R. Brett (editors), Fish physiology, Vol. 8, p. 599-675. Acad. Press, N.Y. BROTHERS, E. B. 1978. Exogenous factors and the formation of daily and subdaily g^rowth increments in fish otoliths. Am. Zool. 18:631. BROTHERS, E. B., C. R Mathews, and R. Lasker. 1976. Daily growth increments in otoliths from larval and adult fishes. Fish. Bull., U.S. 74:1-8. Campana, S. E. 1983. Factors affecting the production of daily growth in- crements in the otoliths of fishes. Ph.D. Thesis, Univ. British Columbia, Vancouver, B.C., 146 p. Campana, S. E., and J. D. Neilson. 1982. Daily growth increments in otoliths of starry floun- der (Platichthys stellatus ) and the influence of some en- vironmental variables in their production. Can. J. Fish. Aquat. Sci. 39:937-942. Degens, E. T, w. g. deuser, and R. L. HAEDRICH. 1969. Molecular structure and composition of fish oto- liths. Mar. Biol. (Berl.) 2:105-113. English, k. k. 1981. Growth and feeding of juvenile chinook salmon, On- corhynchus tshaivytscha , in in situ enclosures. M.S. Thesis, Univ. British Columbia, Vancouver, B.C., 86 p. Marshall, S. L., and S. S. Parker. 1982. Pattern identification in the microstructure of sock- eye salmon (Oncorhynchus nerka ) otoliths. Can. J. Fish. Aquat. Sci. 39:542-547. Matty, a. J. 1978. Pineal and some pituitary hormone rhythms in fish. In J. E. Thorpe (editor). Rhythmic activity of fishes, p. 21-30. Acad. Press, N.Y. Menaker, m., and S. BINKLEY. 1981. Neural and endocrine control of circadian rhythms in the vertebrates. In J. Aschoff (editor). Handbook of be- havioral neurobiology, p. 234-256. Plenum Press, N.Y. MUGIYA, Y, N. WATABE, J. YAMADA, J. M. DEAN, D. G. DUN- kelberger, and M. SHIMUZU. 1981. Diurnal rhythm in otolith formation in the goldfish, Carassius auratus. Comp. Biochem. Physiol. 68A:659- 662. NEILSON, J. D., AND G. H. GEEN. 1981. Method for preparing otoliths for microstructure examination. Prog. Fish-Cult. 43:90-91. NEILSON and GEEN: CHINOOK SALMON OTOLITH INCREMENT FORMATION 1982. Otoliths of chinook salmon (Oncorhynchus tshawytscha): daily growth increments and factors in- fluencing their production. Can. J. Fish. Aquat. Sci. 39:1340-1347. PALOHEIMO, J. E., AND L. M. DICKIE. 1966. Food and growth of fishes. II. Effects of food and temperature on the relation between metabolism and body weight. J. Fish. Res. Board Can. 23:869-908. Pannella, G. 1971. Fish otoliths: daily growth layers and periodical pat- terns. Science (Wash., D.C.) 173:1124-1127. SIMKISS, K. 1974. Calcium metabolism of fish in relation to age- ing. In T. B. Bagenal (editor), Ageing of fish, p. 1-12. Unwin Bros., Ltd., Old Woking Surrey Taubert, B. d., and D. W Coble. 1977. Daily rings in otoliths of three species of Lepomis and Tilapia mossambica. J. Fish. Res. Board Can. 34:332-340. Wilson, k. h., and p a. larkin. 1982. Relationship between thickness of daily growth in- crements in sagittae and change in body weight of sockeye salmon (Oncorhynchus nerka) fry. Can. J. Fish. Aquat. Sci. 39:1335-1339. 101 INFORMATION FOR CONTRIBUTORS TO THE FISHERY BULLETIN Manuscripts submitted to the Fishery Bulletin will reach print faster if they conform to the following instructions. These are not absolute requirements, of course, but desiderata. CONTENT OF MANUSCRIPT The title page should give only the title of the paper, the author's name, his affiliation, and mailing address, including ZIP code. The abstract should not exceed one double-spaced page. In the text, Fishery Bulletin style, for the most part, follows that of the U.S. Government Printing Office Style Manual. Fish names follow the style of the Amer- ican Fisheries Society Special Publication No. 12, A List of Common and Scientific Names of Fishes from the United States and Canada, Fourth Edition, 1980. Text footnotes should be typed separately from the text. Figures and tables, with their legends and headings, should be self-explanatory, not requiring reference to the text. Their placement should be indicated in the right-hand margin of the manuscript. Preferably figures should be reduced by photography to 5% inches (for single-column figures, allowing for 50% reduction in printing), or to 12 inches (four double- column figures). The maximum height, for either width, is 14 inches. Photographs should be printed on high quality or glossy paper. Do not send original drawings to the Scientific Editor; if they, rather than the photographic reductions, are needed by the printer, the Scientific Publications Office will request them. Each table should start on a separate page. Consis- tency in headings and format is desirable. Vertical rules should be avoided, as they make the tables more expensive to print. Footnotes in tables should be num- bered sequentially in arable numerals. To avoid confu- sion with powers, they should be placed to the left of numerals. Acknowledgments, if included, are placed at the end of the text. Literature is cited in the text as: Lynn and Reid (1968) or (Lynn and Reid 1968). All papers referred to in the text should be listed alphabetically by the senior author's surname under the heading "Literature Cited." Only the author's surname and initials are required in the literature cited. The accuracy of the literature cited is the responsibility of the author. Abbreviations of names of periodicals and serials should conform to Serial Sources for the BIOSIS Data Base. (Chemical Abstracts also uses this system, which was developed by the American Standards Association.) Common abbreviations and symbols, such as mm, m, g, ml, mg, °C (for Celsius), %,7«, , and so forth, should be used. Abbreviate units of measure only when used with numerals. Periods are only rarely used with abbreviations. We prefer that measurements be given in met- ric units; other equivalent units may be given in parentheses. FORM OF THE MANUSCRIPT The original of the manuscript should be tjqjed, double-spaced, on white bond paper. Please triple space above headings. We would rather receive good dupli- cated copies of manuscripts than carbon copies. The sequence of the material should be TITLE PAGE ABSTRACT TEXT LITERATURE CITED TEXT FOOTNOTES APPENDIX TABLES (Each table should be numbered with an arable numeral and heading provided) LIST OF FIGURES (Entire figure legends) FIGURES (Each figure should be numbered with an arable numeral; legends are desired) ADDITIONAL INFORMATION Send the ribbon copy and two duplicated or carbon copies of the manuscript to: Dr. William J. Richards, Scientific Editor Fishery Bulletin Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA 75 Virginia Beach Drive Miami, FL 33149-1099 Fifty separates will be supplied to an author free of charge and 50 supplied to his organization. No covers will be supplied. • GPO 693-007 OF, Fishery Bulletin Vol. 83, No. 2 April 1985 BOEHLERT, GEORGE W. Using objective criteria and multiple regression models for age determination in fishes 103 HUNTER, J. ROE, and BEVERLY J. MACEWICZ. Rates of atresia in the ovary of captive and wild northern anchovy, Engraulis mordax 119 LO, NANCY C. H. Egg production of the central stock of northern anchovy, Engraulis mordax, 1951-82 137 MULLIN, M. M., E. R. BROOKS, R M. H. REID, J. NAPP, and E. E STEWART. Ver- tical structure of nearshore plankton off southern California: a storm and a larval fish food web 151 DEMARTINI, EDWARD E., LARRY G. ALLEN, ROBERT K. FOUNTAIN, and DALE ROBERTS. Diel and depth variations in the sex-specific abundance, size composition, and food habits of queenfish, Seriphus politus (Sciaenidae) 171 HEWITT, ROGER R Reaction of dolphins to a survey vessel: effects on census data 187 CROSS, JEFFREY N. Fin erosion among fishes collected near a southern Cali- fornia municipal wastewater outfall (1971-82) 195 DEC 6 1985 Woods Ho!e, Um\ Notice NOAA Technical Reports NMFS published during last 6 months of 1984 V Seattle, Washington U.S. DEPARTMENT OF COMMERCE Malcolm Baldrige, Secretary NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION John V. Byrne, Administrator NATIONAL MARINE FISHERIES SERVICE William G. Gordon, Assistant Administrator 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. Separates 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. Government Printing Office, Washington, DC 20402. It is also available free in limited numbers to libraries, research institutions. State and Federal agencies, and in exchange for other scientific publications. SCIENTIFIC EDITOR, Fishery Bulletin Dr. William J. Richards Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA Miami, FL 33149-1099 Editorial Committee Dr. Bruce B. Collette Dr. Donald C. Malins National Marine Fisheries Service National Marine Fisheries Service Dr. Edward D. Houde Dr. Jerome J. Pella Chesapeake Biological Laboratory National Marine Fisheries Service Dr. Merton C. Ingham Dr. Jay C. Quast National Marine Fisheries Service National Marine Fisheries Service Dr. Reuben Lasker Dr. Carl J. Sindermann National Marine Fisheries Service National Marine Fisheries Service Mary S. Fukuyama, Managing Editor The Fishery Bulletin (ISSN 0090-06561 is published quarterly by the Scientific Publications Office, National Marine Fisheries Service, NOAA, 7600 Sand Point Way NE, BIN C1570O, Seattle, WA 98115. Second class postage is paid at Seattle, Wash., and additional offices. POSTMASTER send address changes for subscriptions to Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. Although the contents have not been copyrighted and may be reprinted entirely, reference to source is appreciated. 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 1 April 1985. Fishery Bulletin CONTENTS Vol. 83, No. 2 April 1985 BOEHLERT, GEORGE W. Using objective criteria and multiple regression models for age determination in fishes 103 HUNTER, J. ROE, and BEVERLY J. MACEWICZ. Rates of atresia in the ovary of captive and wild northern anchovy, Engraulis mordax 119 LO, NANCY C. H. Egg production of the central stock of northern anchovy, Engraulis mordax, 1951-82 137 MULLIN, M. M., E. R. BROOKS, F M. H. REID, J. NAPP, and E. R STEWART. Ver- tical structure of nearshore plankton off southern California: a storm and a larval fish food web 151 DEMARTINI, EDWARD E., LARRY G. ALLEN, ROBERT K. FOUNTAIN, and DALE ROBERTS. Diel and depth variations in the sex-specific abundance, size composition, and food habits of queenfish, Seriphus politus (Sciaenidae) 171 HEWITT, ROGER P Reaction of dolphins to a survey vessel: effects on census data 187 CROSS, JEFFREY N. Fin erosion among fishes collected near a southern Cali- fornia municipal wastewater outfall (1971-82) 195 Notice NOAA Technical Reports NMFS published during last 6 months of 1984 L5 T^ ;r>, .': c-j DEC 6 1985 Woods Hole, Mass. Seattle, Washington 1985 I For sale by the Superintendent of Documents, U.S. Government Printing Office, Washing- ton DC 20402 — Subscription price per year: $21.00 domestic and $26.25 foreign. Cost per single issue: $7.00 domestic and $8.75 foreign. The National Marine Fisheries Service (NMFS) does not approve, recommend or endorse any proprietary product or proprietary material mentioned in this pub- lication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales promotion which would indicate or imply that NMFS approves, recommends or endorses any proprietary product or proprietary material mentioned herein, or which has as its purpose an intent to cause directly or indirectly the advertised product to be used or purchased because of this NMFS publication. USING OBJECTIVE CRITERIA AND MULTIPLE REGRESSION MODELS FOR AGE DETERMINATION IN FISHES George W. Boehlert^ ABSTRACT Analysis of the age structure of exploited fish papulations is necessary for models upon which manage- ment decisions are made, but existing aging methodology for many species is hindered by subjective criteria used in age determination. A new technique is described in which age is estimated using multiple regression models based upon the measurable parameters otolith weight, otolith length, and otolith width in the splitnose rockfish, Sebastes diploproa , and the canary rockfish, S. pinniger. Models were calibrated using ages determined by interpretation of both whole otoliths and otolith sections which differ within these species, particularly at greater lengths. The models typically explained from 70 to 92% of the variability in age depending upon species, sex, and method of age analysis. In another sample used to verify the precision of the models, variability associated with model-estimated ages was generally less than that induced by variability in ages between different agencies. Based upon the pattern of otolith growth in length, width, and weight in these and other species, it is suggested that these methods would be applicable to a wide variety of fishes. Implementation of this type of age determination methodology could result in savings in time and cost for fisheries management agencies while decreasing variability among age estimates between different laboratories. Virtually all methods of age determination in fishes involve a certain degree of subjectivity. De- ciding whether a mark on an otolith or scale con- stitutes 1 year's growth is difficult; precision in fish aging improves only with experience. Even so, var- iability between experienced readers may be great. Sandeman (1969), for example, observed only 9% agreement between readers for a wide age range of otoliths of Sebastes marinus and S. men- tella, and noted greater variability with increas- ing age of the fish. Kimura et al. (1979) suggested that bias between readers within a given agency is likely to be much less than among different agen- cies. In a situation such as exists on the Pacific coast, where several management agencies may routinely determine ages for the same species, interagency calibrations are necessary but are rarely achieved. Williams and Bedford (1974) suggested ". . . that otolith reading remains, for the present at least, as much an art as a science, and that proficiency cannot easily be achieved without examination of very large numbers of otoliths." Clearly, objective, repeatable age determination methodology which will minimize variability is desirable. Traditional methodology for age determination ^Oregon State University, College of Oceanography, Marine Science Center, Newport, Oreg.; present address: Southwest Fisheries Center Honolulu Laboratory, National Marine Fisheries Service, NOAA, PO. Box 3830, Honolulu, HI 96812. in fishes generally involves some calcified struc- ture; in Sebastes, Six and Horton (1977) tested 25 different structures. By far the most commonly used structures, however, are the otolith and scales. Scales are often best for short-lived, fast- growing species because annuli become indistinct near the margin in long-lived, slower growing species (Power 1978; Maraldo and MacCrimmon 1979). When this is the case, the otolith becomes the superior structure for age determination; even in the otolith, however, annuli may become indis- tinct on the margin as otoliths thicken and become opaque with age. For this reason several inves- tigators have used broken or sectioned otoliths to determine age from internal banding patterns. While some studies using otolith sections have provided clear continuation of growth patterns obvious on whole otoliths from younger specimens, others have suggested maximum ages which are double or triple those estimated from whole otoliths. Power (1978), for example, suggested ages of >50 yr in Salvelinus namaycush and Coregonus clupeaformis and provided confirming evidence based upon population structure. In the redfish, Sebastes marinus, Sandeman (1961) suggested that specimens exceeding 50 yr of age were pres- ent in the population; ages up to 80 yr have since been estimated (Sandeman^). Similarly, Beamish Manuscript accepted April 1984 FISHERY BULLETIN; VOL. 83, NO. 2, 1985. ^E. J. Sandeman, Biological Station, St. John's, Newfound- land, Canada, pers. commun. July 1978. 103 (1979b) estimated ages approaching 90 yr in cer- tain Pacific species of Sebastes, including S. alutus. In the genus Sebastes, these estimates of extended longevity have recently been confirmed by Bennett et al. (1982), who used geochronologi- cal methods to confirm age in S. diploproa. Under- standing population structure for such long-lived species will require a large number of age esti- mates using otolith sections. Routine sectioning and interpretation of otoliths, however, is a time- consuming process, and age structure would need to be determined frequently for management of an active fishery. In this paper I suggest a possible alternative method for age determination. Otolith growth begins with the initial "focus" and thereafter by incremental concretions of cal- cium carbonate in the form of aragonite. Otolith size increases with increasing size and age of the fish. Differential addition of crystalline material to the otolith, however, results in a species-specific shape (Bingel 1981). In flatfish and certain other species, Williams and Bedford (1974) observed con- tinued linear growth of the otolith with growth of the fish only until maximum size was achieved; beyond this time, the otolith began to thicken. This has been observed in several other species (Blacker 1974a). Linear measurements of the otolith (i.e., length and width) are directly related to fish length and show little variability, but otolith thickness and weight are highly variable in larger fish (Templeman and Squire 1956; Beamish 1979a, b). Templeman and Squire (1956) observed that length and width of otoliths from slow- and fast- growing populations of haddock did not differ at the same fish length, whereas otolith weight was consistently greater in the slower growing (and therefore older) populations at a given length. The same trend appears to exist in some members of the genus Sebates (G. W. Boehlert unpubl. data). Beamish (1979a) observed an increase in thick- ness of the hake otolith with increasing otolith section age and a nearly linear relationship of otolith thickness and otolith weight. If otolith thickness, and therefore weight, is a function of fish age, then if fish length (or otolith length, since the two are related) is known, one should be able to estimate fish age. This was suggested by Brander (1974) with Irish Sea cod. The objective of this study is to determine the trends of otolith growth in terms of thickness, length, width, and weight, and to determine the potential of these criteria for estimation of age in splitnose rockfish, S. diplop- roa, and canary rockfish, S. pinniger. FISHERY BULLETIN: VOL. 83, NO. 2 MATERIALS AND METHODS Otolith Collection Otoliths of S. pinniger and S. diploproa were collected during the 1980 West Coast Survey con- ducted by the Northwest and Alaska Fisheries Center on the FV Pat San Marie and the FV Mary Lou. Gear and sampling strategy were similar to that described in Gunderson and Sample (1980). Otoliths were collected from fish captured in all hauls until desired numbers of specimens in specified length categories were obtained. Both otoliths from each specimen were removed, cleaned, and stored in individual, labeled vials containing 50% ethanol. Data taken with each specimen included vessel, haul (with latitude, lon- gitude, and bottom depth), sex, and fork length (to the nearest 0.1 cm). After returning to the laboratory, otoliths were thoroughly cleaned and the preservative renewed. Age Determination General information on otolith morphology and whole otolith aging methodology in Sebastes is described in detail by Kimura et al. (1979). Age determined from whole otoliths followed the aging methodology of Boehlert (1980) for S. diploproa and that of Six and Horton (1977) for S. pinniger. Ages determined in this manner are referred to as whole otolith ages. Otolith sections were prepared for selected specimens using the left otolith after the methodology of Nichy^ with several modifications. Specimens were affixed to heavy-duty cardboard tags with double-faced tape and embedded in polyester casting resin in preparation for section- ing. Specimens were mounted in a chuck specifi- cally designed to accommodate the cardboard tags and fed onto a pair of thin diamond blades sepa- rated by acetate spacers on a Buehler** low-speed Isomet saw. Dorsal-ventral sections through the focus and perpendicular to the sulcus, about 0.4 mm thick, were removed from the center of the otolith. Sections were removed from the tag and attached to labeled microscope slides with his- tological mounting medium. They were sub- sequently ground to eliminate surface artifacts, 'F. Nichy, Northeast Fisheries Center Woods Hole Laboratory, National Marine Fisheries Service, NOAA, Woods Hole, MA 02543. ■* Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. 104 BOEHLERT: AGE DETERMINATION IN FISHES first with 400-grit carborundum paper and then polished with 3 ^im jeweler's rouge. To compare internal otolith section annuli with surface annuli, 25 whole left otoliths from S. pin- niger and 50 from S. diploproa were selected. Sample size was chosen to represent the range of ages estimated from whole otoliths. I determined the distance from focus to each annulus on the whole otolith along the dorsal- ventral axis from focus to dorsal edge of the otolith using an ocular micrometer on a dissecting microscope. These measurements were used to identify the first sev- eral annuli on corresponding sections. By follow- ing these identified annuli around to the internal dorsal surface it was determined that each small ring in the direction of counting (from focus to dorsal, interior surface) corresponded to a single year of growth (Fig. 1). Sections were initially examined under a dis- secting microscope at 30 x magnification with either reflected light and a black background or transmitted light, depending upon the clarity of the annuli. Discerning and counting the narrow zones in otoliths from older fish was facilitated by the use of a compound microscope interfaced with a video camera and television screen. A more ac- curate estimate of age was made possible by the increased magnification and enhanced contrast of the compound microscope, coupled with the ease of viewing annuli on an enlarged screen. Sections were aged by identifying the first translucent annulus (winter growth zone) and counting sequential growth zones from the center to the dorsal edge. Subsequent annuli were fol- lowed from the dorsal edge to the interior dorsal quadrant (after Beamish 1979b), and counted to the internal surface. In this paper, ages deter- mined by different methods and sources will be discussed; none of these ages is known with cer- tainty. For this reason, given ages will be defined as "standard ages" only for purposes of compari- son. Calibration Subsample To establish models of age based upon otolith dimension and weight criteria, otoliths from the entire collection were subsampled. Every fourth otolith pair of S. diploproa and every third of S. pinniger were selected to provide roughly equal sample sizes representative of all sizes and collec- tion (latitudinal) areas. These subsampled otoliths were used to develop the multiple regres- sion models (see section on Data Analysis) and were treated as described below. Whole otolith ages were determined by an ex- perienced otolith reader to whom fish length re- mained unknown. This practice has been recom- mended by Williams and Bedford (1974), among others, to minimize bias in otolith reading. Otoliths were then dried to a constant weight at 58°C and placed in a dessicator for 8 h. Intact left otoliths were weighed to the nearest milligram. Otoliths were measured with dial calipers in the anteroposterior dimension (length) to the nearest 0.02 mm and in the maximum dorsoventral di- FlGURE 1. — Dorsal-ventral section of the left otolith of a 305 mm FL female Sebastes diploproa. Whole otolith ages are generally determined from the focus (F) to the dorsal edge (A), but often extend to the posterior margin (not shown) which may include additional annuli extending to greater ages (A to B). Section ages are determined from the focus (F) to the internal dorsal surface (C). Note the additional growth zones on axis F-C which have been deposited after the latest visible zones on axis F-A. The otolith section age of this specimen is 40 yr. 105 FISHERY BULLETIN; VOL. 83, NO. 2 mension (width) to the nearest 0.05 mm. When the left otolith was chipped or broken, the right one was substituted for measurements, since no sys- tematic differences between left and right otolith measurements were apparent for either species. The left otolith was subsequently sectioned and age determined by the same otolith reader. Otolith thickness, which is too variable to measure on the whole otolith, was measured on the section from internal to external surface just dorsal to the sul- cus (Fig. 1). Confirmation Subsample In order to test the precision of the model, sub- samples of 50 otoliths by sex and species were drawn randomly from samples not used in the calibration subsample. These samples were han- dled in the following way: A second whole otolith age was determined by reader A to determine within-reader variability for S. diploproa and between-reader variability for S. pinniger (reader B had left this laboratory). The otoliths were sent to the Northwest and Alaska Fisheries Center (Seattle, Wash. ) for an additional whole otolith age to determine between-agency variability. The otolith was dried, weighed, measured, and sec- tioned as described above; a single otolith section age for each specimen was determined by reader A for both species. Model-estimated ages were de- termined by use of the multiple regression models described below. Data Analysis Generally, data were recorded in a standard format and stored on the Oregon State University Cyber 70 computer. Data management and analy- sis were assisted by use of the Statistical Package for the Social Sciences (SPSS) (Nie et al. 1975). From the calibration subsample of otoliths, pre- dictive regression equations were developed to es- timate age from otolith morphometries. Multiple regression models were fitted in the following form: Age = 6iXi -I- 62^2 + 63X3 -I- b^X^ + c where age (years) is determined by conventional methods, bn's = regression coefficients, X 's = in- dependent variables, and c = constant. Models were developed for males and females separately within each species with both otolith section ages and whole otolith ages as dependent variables. Independent variables included otolith weight, otolith length, otolith width, the respective square and cubic terms of each, and the interaction vari- ables (otolith weight/otolith length and otolith length/otolith width). With the exception of otolith weight, where both weight and the cube of weight were used as independent variables, square or cubic terms were not used if the raw values were entered. This decreased problems of multicollinearity. Models were fitted in a forward stepwise manner (Nie et al. 1975) with the inclu- sion level for independent variables set at P = 0.10. The 1980 confirmation subsample was used to verify the models. Direct comparisons between ages determined for the same otoliths but dif- ferent reading methods were accomplished by paired ^-tests. Since age is not known with cer- tainty for any otolith, the ages determined by reader A for S. diploproa and by reader B for S. pinniger, which were used to calibrate the models in the calibration subsample, were considered as "standard age". To conduct multiple comparisons of variability, deviations from standard age were defined as follows: "model-induced variation" is the difference between the standard age and the model-predicted age; "within-agency variation" is the difference between ages determined by reader A for S. diploproa and between readers A and B for S. pinniger; "between-agency variation" is the dif- ference between the standard age and the age de- termined by the National Marine Fisheries Ser- vice (NMFS). A one-way analysis of variance (ANOVA) was used to compare these deviations. Multiple range testing was conducted using the least significance difference method with a = 0.05. This analysis was conducted only for whole otoliths since only a single section age was determined on the 1980 confirmation subsample. RESULTS Sehastes diploproa Locations of the collections of S. diploproa are shown in Figure 2; this species was taken from lat. 36°49' to 48°47'N and over a depth range of 62 to 338 m. The distribution was similar to that noted in 1977 (Boehlert 1980). A total of 975 male and 1,145 female specimens were taken during the survey. The length frequencies show a mode near 23 cm for males and 24 cm for females with sec- ondary modes at 26 and 27 cm, respectively. Cor- responding age frequencies (based upon whole 106 BOEHLERT: AGE DETERMINATION IN FISHES otoliths) show a clear mode at 7 yr for both males and females, with whole otolith age ranges from 1 to 46 for males and 0 to 55 for females. Mean lengths-at-age for males and females are similar until age 8, after which females grow more rapidly (Boehlert 1980; Boehlert and Kappenman 1980). Subsampling every fourth pair of otoliths from all collections of S. diploproa resulted in 290 female and 246 male specimens. The subsample was representative of the latitudinal distribution, age range, and length range of the whole collec- tion. Capture, otolith, and age data from these samples are summarized in Table 1. Otolith sec- tion ages, as expected, were typically greater than whole otolith ages (Table 1); this was particularly true at greater lengths. Correlation matrices of pertinent otolith and age data (Table 2) show that otolith weight has the strongest linear association with otolith section age; both otolith weight and age are exponential functions offish length. Plot- ting otolith length, fish length, and otolith weight against otolith section age demonstrates the pat- tern of otolith growth (Fig. 3). Past an age of about 25 yr, both otolith length and fork length reach approximate asymptotes, whereas otolith weight continues to increase. The wide fluctuations in otolith weight apparent at older ages correlate closely with changes in fork length (Fig. 3 ); for this reason, otolith weight alone is a relatively poor predictor of fish age at greater ages where fork length is highly variable. Addition of otolith FIGURE 2. — Locations of 1980 West Coast Survey collections from which otoliths of Sebastes diploproa were taken. 10 20 30 40 50 60 70 OTOLITH SECTION AGE (YEARS) Figure 3. — Otolith characteristics of male Sebastes diploproa from the calibration subsample as related to fish length and age. A'' = 246. Note the covariation among the three curves, particularly at older ages. Table l. — Summary of biological and otolith data from the subsampled groups of Sebastes diploproa used in developing the age models. Females (N = 290) Males (W = 246) Variable Minimum Maximum Mean SD Minimum Maximum Mean SD Depth of capture (fathoms) 34 185 137 29.36 53 185 136 28.45 Fork length (mm) 130 378 264 56.16 94 364 246 48.19 Otolith length (mm) 7.71 18.02 12.49 2.35 5.47 17.03 11.82 2.14 Otolith width (mm) 5.08 11.25 7.97 1.31 3.59 10.32 7.57 1.14 Otolith thickness (mm) 0.83 2.97 1.41 0.44 0.73 2.84 1.35 0.39 Otolith dry weight (mg) 59 724 244.6 150.4 25 659 208 117.4 Whole otolith age (yr) 1 56 15.2 11.97 1 40 13.5 9.78 Otolith section age (yr) 2 66 17.2 15.68 1 74 16.9 16.41 107 FISHERY BULLETIN: VOL. 83, NO. 2 Table 2. — Correlation matrix for selected otolith morphometric, weight, and age data for the calibration subsample of Sebastes diploproa. Whole Ofclith Otolith Otolith Otolith Otolith otolith section weight length width thickness age age Females (A/ = 290) Fork length 0.912 0.969 0.956 0.766 0.862 0.819 Otolith section age 0.947 0.859 0.788 0.938 0.917 Whole otolith age 0.925 0.893 0.837 0.901 Otolith thickness 0.930 0.843 0.778 Otolith width 0.893 0.948 Otolith length 0.940 Males (W =246) Fork length 0.895 0.971 0.959 0.815 0.835 0.769 Otolith section age 0.938 0.807 0.710 0.905 0.907 Whole otolith age 0.923 0.885 0.778 0.846 Otolith thickness 0.903 0.778 0.725 Otolith width 0.857 0.778 Otolith length 0.922 length and the interaction variables compensate for these changes in the pattern of otolith weight in the multiple regression models offish age. The multiple regression models relating fish age with otolith data were fitted with both whole otolith age and otolith section age as dependent variables. Independent variables included in the whole otolith age models, their coefficients, and significance levels are presented in Table 3. All coefficients were highly significant and the models explain 88.1% of the variation in age for females and 92.0% for males, as measured by the coeffi- cient of determination, R"^. Residuals from the models by age category show no trend up to age 35 for females and age 30 for males, after which there is a trend of increasing positive deviation with increasing age. The ages included in this part of the model, however, represented only 7.7% of female and 8.6% of male S. diploproa and are therefore not of great concern. These deviations are positive, however, suggesting that the model predictions may relate to otolith growth patterns which are more indicative of otolith section ages. Variables included in the otolith section age models, their coefficients, standard errors, and significance levels are presented in Table 4. Again, all coefficients are highly significant, but the co- efficients of determination are slightly less, ex- plaining 86.1% of the variation in age for females and 85.0% for males. Mean residuals for the dif- ferent age categories show no significant trend with age. The model based upon whole otolith ages suffers from inaccuracies in the older ages, where otolith section ages are much greater than whole otolith ages. This is demonstrated in the trend of increas- ing residuals with increasing age. The model based upon otolith section age, however, is charac- 108 Table 3. — Regression coefficients and associated statistics on the multiple regression models of whole otolith age for Sebastes diploproa . Variable Coefficient SE P Females (N = 290) Otolith weight 0.1343 0.0091 <0.001 (Otolith weight)^ -0.107 X 10-6 0.14 X 10"' <0.001 Otolith width -2.558 0.571 <0.001 Constant (a) 6.4303 3.004 0.033 SD = 4.15 Multiple correlation, R = 0.939 Males (W = 246) Otolith weight 0.2179 0.0145 <0.001 (Otolith weight) 3 -0.1945 X 10-6 0.14 X 10-' <0.001 Otolith width -3.4542 0.3942 <0.001 Otolith weight/length -1.0997 0.2402 <0.001 Constant (a) 16.2572 2.2186 <0.001 SD = 2.797 Multiple correlation, R = 0.959 Table 4. — Regression coefficients and associated statis- tics on the multiple regression models of otolith section age for Sebastes diploproa . Variable Coefficient SE Females (A/ = 290) Otolith weight (Otolith width) 2 (Otolith weight)^ (Otolith length)^ Constant (a) SD = 4.232 Multiple correlation, R Males (W = 246) Otolith weight Otolith width^ (Otolith weight)^ (Otolith length)^ Constant (a) SD = 4.620 Multiple correlation, R 0.2270 -0.3288 -0.1134 X 10-' -0.1114 5.0243 = 0.928 0.2496 -5.7233 -0.1315 X 10-' -0.0882 23.540 0.922 0.0137 0.0377 0.155 X 10- 0.0205 1 .2982 00158 0.6949 0.266 X 10- 0.0256 3.3823 <0.001 <0.001 <0.001 <0.001 <0.001 0.001 0.001 0.001 0.001 <0.001 terized by slightly lower multiple correlation co- efficients (Table 4). This may be a result of inac- curacies in estimates of otolith section age of younger fish, where greater difficulty in age de- BOEHLERT: AGE DETERMINATION IN FISHES termination exists with sections. For this reason, I also constructed a hybrid multiple regression model based upon a combination of otolith section and whole otolith ages. The decision on which age to use was arbitrary in the following way: If the difference (otolith section age minus whole otolith age) was "^ 5 yr, whole otolith age was chosen; if the difference was >5 yr, otolith section age was cho- sen. The resulting models are described in Table 5. Independent variables similar to those in the other two models were chosen, and the multiple correlation coefficients were greater in each case. To analyze the precision of the models, subsam- ples of 50 male and 50 female S. diploproa were taken from the remaining samples not used in the calibration subsample. Lengths and ages were representative of the respective ranges in the overall collection. Ranges of whole otolith age, NMFS age (that from the other agency), and otolith section age in these samples were 2-50, 3-49, and 2-75 for females and 3-34, 4-25, and 3-84 for males, respectively. Whole otolith age was predicted based upon the appropriate whole otolith age models. Values of estimated age, whole otolith age, and NMFS age as a function of length are plotted in Figure 4. The deviation of NMFS age from whole otolith age increases with increasing length for both males and females. Deviations from the first whole otolith age are presented in Figure 5. Model- induced variability is the difference between es- timated whole otolith age and whole otolith age; between-agency variability is whole otolith age minus NMFS age; within-agency variability is the difference of two successive age determinations by Table 5. — Regression coefficients and associated statis- tics on the multiple regression models of age in Sebastes diploproa . The ages used for the calibration of these mod- els are based upon either whole otoliths or otolith sec- tions as described in the text. Variable Coefficient SE Females (A/ = 290) Otolith weight 0.2233 (Otolith width) 2 -0.2983 (Otolith weight)^ -0.1244 x 10" Otohth length -2,495 Constant (a) 17.7993 SD = 4.3967 Multiple correlation, R = 0.962 Males {N = 246) Otolith weight 0.2504 (Otolith width)2 -0.3598 (Otolith weight) 3 -0.1272 x IQ- Otolith length -2.4123 Constant (a) 16.6069 SD = 4.7479 Multiple correlation, R = 0.958 0.0135 0.0403 0,1685 X 10- 0.5084 3,7339 0,0157 0,0549 0-2800 . 10-' 0,6071 3,9145 0001 0,001 ;0.001 -0.001 0,001 ; 0.001 0,001 0,001 ;0.001 tO.001 the same reader Mean values of these sources of variation are presented in Table 6 for females and Table 7 for males. In both cases, the mean between-agency variability is greater than either model-induced or within-agency variability. One-way ANOVA demonstrates a significant dif- ference among the three sources (Tables 6, 7). Mul- tiple range testing (least significant difference, a = 0.05), moreover, demonstrates that the means are significantly different for both females and males; the range tests suggest that within-agency and model-induced variability are equal and are both significantly less than the between-agency variability. Only a single otolith section age was determined for specimens from the 1980 confirmation subsam- ple. Ages were estimated from the multiple re- gression model of section age (Table 4) and com- pared with conventionally determined section age 35rT 14 16 18 20 22 24 26 28 30 32 34 36 38 40 FORK LENGTH (cm) FIGURE 4. — Comparisons of mean whole otolith ages at length for the confirmation subsample of Sebastes diploproa . Trian- gles represent age from reader A, circles the age estimated by the model, and squares the age determined by another laboratory. 109 FISHERY BULLETIN: VOL. 83, NO. 2 Table 6. — Results of one-way analysis of variance and multiple range tests comparing deviations of age from the standard age in Sebastes diploproa females. Group 1 = between-agency variabil- ity; group 2 = model-induced variability: group 3 = within- agency, within reader variability. Source df Sum of squares Mean squares Analysis of variance Table 7. — Results of one-way analysis of variance and multiple range tests comparing deviations of age from the standard age in Sebastes diploproa males. Group 1 = between-agency variabil- ity; group 2 = model-induced variability; group 3 = within- agency, within reader variability. Source df Sum of squares IVIean squares Analysis of variance Between groups 2 207.30 103.65 Within groups 147 1,118.30 7.61 Total 149 1.325,60 Group n Mean SD 1 50 2.360 3.306 2 50 0.108 2 294 3 50 -0320 2.575 13.62 <0.001 Multiple range lest (least significant difference, a = 0.05) Group 3 ^ Group 2 ■ Group 1 (Fig. 6). Ages were close to those predicted from the model with the notable exception of the maximum age for both males and females. In each instance, the maximum ages were greater than the maximum otolith section age in the calibra- tion subsample; the estimated section age is there- fore an extrapolation from the model. For the overall subsample, however, the estimated section ages were not significantly different from those determined by conventional methods (paired ^test, a = 0.05). The observed and predicted ages comparing the confirmation subsample with the predicted ages from the hybrid model are not pre- sented graphically, but the form of the curves for both males and females is virtually identical to that for the section age model (Fig. 6). Sebastes p/tjurger Sebastes pinniger were collected from lat. 43°ir to 49°26'N at depths from 58 to 375 m (Fig. 7). 110 to < Between groups Within groups Total 2 147 149 707,77 2.247.93 2.955.70 353.89 15.29 23.14 •0.001 o > UJ Group n Mean SD 1 2 3 50 50 50 4.000 -0.51 -0.700 4.686 4.134 2.613 Multiple range test (least significant Group 3 = Group 2 - Group 1 difference, « = 0.05) MALES 12 14 16 18 20 22 WHOLE OTOLITH AGE (YEARS) Figure 5. — Mean deviations of whole otolith ages from the confirmation subsample of Sebastes diploproa. Triangles rep- resent model-induced variability, circles within-agency variabil- ity, and squares between-agency variability. Pairs of otoliths from a total of 519 male and 369 female specimens were taken from the survey. Length frequencies for S. pinniger show a mode at 50 cm for males and 52 cm for females. Age fre- quencies of the entire sample (based upon whole otoliths) demonstrate a mode for both males and females at 12 to 13 yr Whole otolith ages from the collections ranged from 2 to 25 for males and 2 to 22 for females. Subsampling every third pair of otoliths from the whole collection resulted in 171 male and 121 female specimens of S. pinniger. Again, this sub- sample was representative of the latitudinal dis- tribution, age range, and length range of the whole sample. Capture, otolith, and age data from these specimens are summarized in Table 8. Otolith section ages in larger fish are generally greater than whole otolith ages, but not to the BOEHLERT: AGE DETERMINATION IN FISHES 90 80 70- — 60 (/> < uj 50 <0 40 < z < 30 20- 10 T I I MALES ' T I I 1 - - J \\ - _ I \ _ - / 9 - - A' - 1 1 1 1 1 1 1 1 - 14 16 18 20 22 24 26 28 30 32 34 36 38 40 FORK LENGTH (cm) Figure 6. — Comparisons of mean otolith section ages at length from the confirmation subsample of Sebastes diploproa. Triangles represent otolith section age and circles the model estimated section age. Figure 7. — Locations of 1980 West Coast Survey collections from which otoliths of Sebastes pinniger were taken for the current study. Samples from the FV Pat San Marie and the FV Mary Lou are included. Table 8. — Summary of biological and otolith data from the subsampled groups of Sebastes pinniger used in developing the age models. Females (A/ = 121) Males (N = 171) Variable Minimum Maximum Mean SD Minimum Maximum Mean SD Depth of capture (fathoms) 32 100 69.8 12.66 37 103 73.3 14.39 Fork length (mm) 152 610 497.8 69.25 170 579 481.64 64.20 Otolith length (mm) 8.00 23.40 19.62 2.27 8.59 22.89 19.56 2.31 Otolith width (mm) 4.45 12.02 9.60 1.06 4.69 11.07 9.46 1.01 Otolith thickness (mm) 0.83 2.01 1.54 0.19 0.79 2.41 1.64 0.29 Otolith dry weight (mg) 53 821 486.7 135.5 58 867 517.0 160.69 Whole otolith age (yr) 2 19 12.4 3.16 2 25 13.2 3.79 Otolith section age (yr) 2 33 14.83 5.09 2 54 20.02 9.77 extent seen for S. diploproa. Otolith weight is again an exponential function of length, particu- larly for males. For females, however, this rela- tionship was nearly linear. Of the ages determined in the calibration subsample, otolith weight has the strongest linear association with whole otolith age for females and whole otolith age and section age for males (Table 9). The multiple regression models constructed to predict whole otolith age were based upon fewer variables than for S. diploproa , but included vari- ables were highly significant (Table 10). The coeffi- 111 FISHERY BULLETIN: VOL. 83. NO. 2 Table 9. — Correlation matrix for selected otolith morphometric, weight, and age data for the calibration subsample of Sebastes pinniger. Whole Otolith Otolith Otolith Otolith Otolith otolith section weight length width thickness age age Females (W = 121) Fork length 0.915 0.948 0.923 0.779 0.895 0.755 Otolith section age 0.825 0.735 0.757 0.718 0.795 Whole otolith age 0.890 0.887 0.851 0.756 Otolith thickness 0.826 0.765 0.756 Otolith width 0.920 0.902 Otolith length 0.917 Males (W = 171) Fork length 0.844 0.940 0.909 0.754 0.847 0 682 Otolith section age 0 898 0.694 0.696 0.883 0,809 Whole otolith age 0.892 0.837 0.815 0.830 Otolith thickness 0.910 0 769 0.750 Otolith width 0.869 0.901 Otolith length 0.879 Table 10. — Regression coefficients and associated statistics on the multiple regression models of whole otolith age for Sebastes pinniger. Variable Coefficient SE P Females (N = 121) (Otolith length)^ 0.00095 (Otolith width)2 0.0448 0.00011 0.0126 ■0.001 0.001 SD = 1.30 Multiple correlation. R = 0 913 Males (W = 171) Otolith weight 0.0280 (Otolith weight)^ -0.845 x lO-« 0.00214 0.241 X 10-8 <0.001 0.001 SD = 1.665 Multiple correlation, R = 0,900 cient of determination (R ) suggests that the mod- els of whole otolith age explain 83.4% of the varia- tion in age for females and 81.0% for males. For both males and females, the constant in the re- gression was not significantly different from zero and was not included in the models. The residuals from the models show no distinct trend with the exception of a slight increase at ages >17 yr for males; this included 11.1% of the sample. The variables included in the otolith section age models, their coefficients, standard errors, and significance levels are presented in Table 11. As in the whole otolith age models, there are fewer vari- ables included than for S. diploproa; for the male section age model, for example, there is only one variable and the constant included for prediction of age. All variables are highly significant and the coefficients of determination suggest that the otolith section models explain 70.2% of the varia- tion in age for females and 84.6% for males. Mean residuals show a strong trend of increase at ages past 26 yr for male otolith section age models; this represented 23% of the sample. A model incorporating both otolith section age and whole otolith age was developed using the same criteria for age as in S. diploproa. These models were based upon more independent vari- ables but were not significantly better (as based upon the coefficient of determination) than the otolith section models (Table 12). Based upon the multiple correlation coefficients, the best models for S. pinniger would be the hybrid model for males and the whole otolith model for females. For analyzing the precision of the models, sub- Table 11. — Regression coefficients and associated statistics on the multiple regression models of otolith section age for Sebastes pinniger. Variable Coefficient SE P Females (N = 121) (Otolith weight)^ 0.272x10-' 0.382 X 10-5 <0.001 Otolith width 0.8368 0.4586 0.071 SD = 2.80 Multiple correlation, R = 0.838 Males (A/ =171) (Otolith weight)^ 0.546x10-" 0.179 X 10-5 < 0.001 Constant (a) 4.0297 0.6022 • 0.001 SD = 3.85 Multiple correlation, R = 0.920 Table 12. — Regression coefficients and associated statistics on the multiple regression models of age in Sebastes pinniger. The ages used for the calibration of these models are based upon either whole otoliths or otolith sections as described in the text. Variable Coefficient SE P Females (N = 121) (Otolith weight)^ 0.2621 X 10-" 0.4518 X 10-5 0.001 (Otolith width) 3 0.4038 X 10-2 0.2186 X 10-2 0.067 Constant (a ) 3.2137 1.1296 0.005 SD = 2.8239 Multiple correlation, R = 0.840 Males (W = 171) (Otolith weight)^ 0.1306 X 10-3 0.2359 X 10-" <0.001 (Otolith length)^ -0.2044 X 10-2 0.5456 X 10-3 <0.001 (Otolith weight) 3 -0.6026 X 10-' 0.2197 X 10-' 0.007 Otolith length/width 9.7349 4.1381 0.020 Constant (a ) -12.8239 7.4064 0.085 SD = 3.9989 Multiple correlation, R = 0.924 112 BOEHLERT: AGE DETERMINATION IN FISHES samples of 50 male and 50 female S. pinniger were taken from the remaining 1980 samples not used in the calibration subsample. These subsamples were representative of the length and age ranges in the overall collection. Ranges of whole otolith age, NMFS age, and otolith section age in these subsamples were 4-26, 4-25, and 4-29 for females and 7-35, 7-32, and 8-45 for males, respectively. Whole otolith age was estimated from the ap- propriate whole otolith age model for males and females. Values of model estimated age, whole otolith age, and NMFS age as a function of length are plotted in Figure 8. Female S. pinniger ages are similar for all three age determination methods. For males, model-estimated age is simi- lar to the whole otolith age but both are less than the NMFS age (Fig. 8). Deviations from the whole otolith age by the otolith reader whose ages were used to calibrate the model are shown in Figure 9. 26 24 22 20 18 16 14 12 10 8 6- D— Q MALES 10 8 6 4 « a: < 2 Ul >- 0 7 o 1- -2 < ■> III -4 o V) < UJ -6 -8 -lOf- "I 1 1 1 1 r MALES 9 ° > Q -2 -3 -4 -5 6 8 10 12 1 1 1 1 r FEMALES 16 18 20 22 24 1 1 ' r ntxv. ■A ^/ J \ L _L _!_ J_ 6 8 10 12 14 16 18 20 22 24 WHOLE OTOLITH AGE (YEARS) Figure 9. — Mean deviations of whole otolith ages from the confirmation subsample oi Sebastes pinniger. Triangles repre- sent model-induced variability, circles within-agency variabil- ity, and squares between-agency variability. 0 22 J I I I I I J I L 36 38 40 42 44 46 48 50 52 54 56 in < UJ > UJ < z < UJ S oL T r 25 30 35 40 45 50 FORK LENGTH (cm) 55 J 60 Figure 8. — Comparisons of mean whole otolith ages at length from the confirmation subsample of Sehastes pinniger. Trian- gles represent age from reader B, circles the age estimated by the model, squares the age determined by another laboratory. The explanation of these deviations is the same as for S. diploproa with the exception that the within- agency variability is a between-reader rather than a within-reader variability. One-way ANOVA within these deviations shows significant differences among the groups for both females (Table 13) and males (Table 14). Multiple range testing (least significant difference, a = 0.05) demonstrates that for females, mean between-agency variability and model-induced variability are equal but are both less than within- agency variability (for S. pinniger this was based upon two different readers). For males, between- agency variability is less than model-induced vari- ability which is less than within-agency variability. For the purposes of this comparison, however, the model-induced variability is significantly closer to zero than either of the other sources of variability (Table 14). In the confirmation subsample, section ages estimated from the multiple regression model are 113 FISHERY BULLETIN: VOL. 83, NO. 2 Table 13. — Results of one-wa.v analysis of variance and multi- ple range tests comparing deviations of age from the standard age in Sebastes pinnlger females. Group 1 = between-agency variability; group 2 = model-induced variability; group 3 = within-agency, between reader variability. Source df Sum of squares Mean squares Analysis of vanance Between groups 2 88.69 Within groups 147 751.84 Total 149 840 53 Group n Mean SD 44.34 5.11 8.67 0,001 50 50 50 -0.320 -0.021 1.44 2.817 1.516 2 260 Multiple range test (least significant difference, a = 0.05) Group 1 = Group 2 Group 3 Table 14. — Results of one-way analysis of variance and multi- ple range tests comparing deviations of age from the standard age in Sebastes pinniger males. Group 1 = between-agency var- iability; group 2 = model-induced variability; group 3 = within-agency, between reader variability. Source df Sum of squares Mean squares F Analysis of variance Between groups 2 1,840.42 920.21 Witfiin groups 147 2.006.21 13.65 Total 149 3.846.63 Group n Mean SD 1 50 -4.280 4.427 2 50 -0 111 2 107 3 50 4 300 4 112 67.43 -0.001 Multiple range test (least significant difference, a = 0.05) Group 1 Group 2 • Group 3 compared with conventional section ages in Fig- ure 10. The two ages are similar and as a whole are not significantly different for females but are sig- nificantly different for males (paired ^-test, a = 0.05). This is presumably a result of the con- sistently overestimated otolith section age for S. pinniger males. The ages estimated from the hy- brid model (Fig. 11) are not significantly different from those determined by the appropriate conven- tional age (paired /-test, a = 0.05). DISCUSSION The results of this research demonstrate the potential for using objective criteria and mul- tivariate models to determine age in fast- and slow-growing members of the genus Sebastes . Past studies have used weight of the eye lens for esti- mates of age in fishes, amphibians, and certain mammals (Crivelli 1980; Malcolm and Brooks 1981). In fishes, however, this technique is only 114 good for fast-growing species and provides poor estimates of age after several years when length at age becomes highly variable (Crivelli 1980); the same problems exist in estimating age from modal lengths. Growi;h of most body parts, including the eye lens, is allometric with length rather than age. Growth of the otolith, however, as described above, is a complex function of age as well as length. After a certain size is reached, the fish otolith does not increase in length or width, but continues to in- crease in thickness, and therefore weight, with age (Fig. 3). The increasing thickness is a function of addition of aragonite crystals only on the inter- nal surface of the otolith (Fig. 1). Similar patterns of otolith growth in length, width, thickness, and weight have been observed 1 1 ; 1 I 1 i 1 1 1 32 MALES A - 28 A A jj ^^^? v> jp rr Cfi^ - / \ 1 1 tij <•> 20 m < z < / '' UJ 16 ~ 0/ - z •^, / ■ 12 L ^ " B 7 • 38 40 42 44 46 48 50 52 54 56 30 35 40 45 FORK LENGTH (cm) 50 55 60 Figure lO. — Comparisons of mean otol ith section ages at length from the confirmation subsample of Sebastes pinniger. Trian- gles represent otolith section age and circles the model esti- mated section age. BOEHLERT: AGE DETERMINATION IN FISHES in other species offish, but the information has not been applied to the estimation of age, with the exception of preliminary tests using discriminant techniques by Brander (1974). Templeman and Squire (1956), however, noted the importance of this information: "In many fishes, in which accu- rate age reading is doubtful, otolith weights, which are more factual, may offer a better separa- tion of fish populations than growth rates which are dependent on the judgement of the scale- or otolith-reader." Weight and otolith measurements are valid criteria for age determination based upon the models (Tables 3-5, 10-12) and provide good estimates of age compared with other reading methods (Tables 6, 7, 13, 14; Figs. 4-6, 8-11). Based upon published patterns of otolith growth, these techniques should work for other species of Sebastes (Sandeman 1961; Beamish 1979b), Pacific hake (Beamish 1979a), haddock (Templeman and Squire 1956), plaice, sole, turbot, and horse mackerel (Blacker 1974a), and cod (Trout 1954; Blacker 1974a), among others. This technique may therefore be amenable to a wide variety of species of fishes. Ages determined by scale or otolith readers are generally based on subjective decisions by the age reader, who reads annuli but must distinguish from "false checks", "metamorphic checks", and "spawning checks" (Trout 1961; Bailey et al. 1977). FORK LENGTH (cm) Figure ll. — Comparison of ages determined from otoliths and those predicted by the hybrid regression model for Sebastes pin- niger males. Otolith ages were based upon whole otolith ages if the difference between section and whole otolith ages were s5; otherwise, otolith sections were used. Triangles represent whole otolith or section age and squares the model estimated age. With experience comes reduced individual vari- ability, but aging variability among different otolith readers and especially among different agencies is great; such variability can have impor- tant effects upon the estimates of growth parameters important for fisheries management (Sandeman 1961; Brander 1974; Hirschhorn 1974; Kimura et al. 1979). While otolith or scale ex- changes are occasionally made between agencies for calibration purposes, this represents addi- tional time spent for gaining greater consistency in ages (Westrheim and Harling 1973; Blacker 1974b), and difficulties may remain if disagree- ment in aging techniques cannot be resolved. Blacker (1974a) noted that "Recent progress in the use of otoliths for age determination has been lim- ited mainly to the development of new techniques for preparing otoliths for reading and for photog- raphy so that aging methods can be readily com- pared." The techniques described in the present study represent a new approach to the systematic and repeated age determination in species for which continued age determination is necessary; once calibrated and implemented, the models would reduce between-reader and between-agency variability in age determination. Further re- search, however, should be conducted on varia- tions in thg models over seasons, regions, and dif- ferent years to determine to what extent repeated calibration is necessary. Ancillary benefits of the proposed methodology include its simplicity. Reliable, repeatable esti- mates of age require a great deal of experience on the part of an otolith or scale reader using conven- tional aging methodology (Blacker 1974a). It is often difficult to maintain a staff of trained otolith readers and retraining may require a large time commitment. The techniques described here re- quire no special training, since the criteria (otolith length, otolith width, and otolith dry weight) are objective and can be measured with simple dial calipers and balance. Time expended for age de- termination by different methods is as follows: An experienced otolith reader averages about 17 ages/h on whole otoliths, but only 6 to 8 ages/h when otolith sections are used due to the addi- tional preparation necessary. An untrained tech- nician, however, can determine the measurements necessary for the model-based age estimates at a rate of about 40 otoliths/h on a long-term basis. Since the criteria for age are measurable, the techniques will be amenable to automation. Sev- eral attempts have been made in the past to auto- mate or semiautomate age analysis using imaging 115 FISHERY BULLETIN: VOL. 83, NO. 2 systems based upon differential light transmis- sion (Fawell 1974; Mason 1974). These techniques have generally not been implemented, however, due to the subjective and variable nature of the criteria. Implementation of these techniques with automated systems could result in even further savings of time. Since estimating the age distribution of exploited fish populations remains an important part of fishery biology, new and improved techniques of age determination are desirable. For shorter lived species, length-based methods are proving important (Pauly and David 1981). Age- length keys are also used quite extensively. Sam- ple sizes necessary for accurate age-length keys, however, must be quite large, particularly for long-lived species such as Sebastes. In my rela- tively small calibration subsamples, for example, there are up to 15 age classes in a single 1 cm length interval (Table 15). Considering the Table 15. — Number of age classes within single 1 cm length intervals from the calibration subsample. A'^ = number of Sebastes specimens in the subsample. Whole Otolith otolith section Species Sex N age age S diploproa Female 290 14 14 Male 246 12 14 S. pinniger Female 121 6 11 Male 171 9 15 maximum age of S. diploproa (Bennett et al. 1982), there could potentially be up to 50 age classes in a single length interval if a sufficient sample size were taken. For such species, age-length keys will be difficult to extrapolate meaningfully to the en- tire population without very large sample sizes, which must accordingly be aged. Similar, but more severe, problems will apply to techniques which attempt to extract growth parameters from length-frequency data for such long-lived species. The techniques developed by Pauly and David (1981) for faster growing species would be com- plemented by the current technique for slow- growing, difficult-to-age species. Otoliths could be collected by station, sex, and species without re- gard, to size. From each otolith, after calibration of an age model, the available information could in- clude both fish length and age. This approach to length data collection is not new and has been used by the International Pacific Halibut Commission for several years to estimate length (Southward 1962; Quinn et al. 1983). These techniques could therefore streamline not only the collection of otoliths at sea but also the analysis of age in the laboratory. The difficulty in age determination described above and the resulting variability between laboratories may have a negative impact upon ac- curacy of fishery models, particularly those using cohort or virtual population analysis (Brander 1974; Alverson and Carney 1975). The new methodology can provide significant time and cost savings over conventional methods and also de- crease variability in age estimates. Implementa- tion of these aging techniques, however, will re- quire careful calibration with ages determined by a consensus of expert otolith readers from all management agencies with an interest in each species for which a model is developed. ACKNOWLEDGMENTS This research was supported by Cooperative Agreement No. 80-ABH-00049 from the North- west and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, Seattle, Wash. I thank Tom Dark, Mark Wilkins, and other partic- ipants in the 1980 West Coast Survey for assis- tance in specimen collection; particular thanks are extended to Captains Bernie and Tom Hansen of the FV Pat San Marie and MV Mary Lou, re- spectively, and their capable crews. Technical as- sistance and otolith reading were provided by Mary Yoklavich, Dena Gadomski, and Robert McClure. I thank Jack Lalanne for providing the NMFS age estimates. Finally I thank W H. Lenarz and D. R. Gunderson for critically review- ing the manuscript. LITERATURE CITED Alverson, D. L., and M. J. Carney. 1975. A graphic review of the growth and decay of popula- tion cohorts. J. Cons. Int. Explor. Mer 36:133-143. BAILEY, R. F. J., K. W. ABLE, AND W. C. LEGGETT. 1977. Evidence for the presence of a metamorphic check in capelin {Mallotus villosus) otoliths and implications for age determination. J. Fish. Res. Board Can. 34:2008- 2014. BEAMISH, R. J. 1979a. Differences in the age of Pacific hake (Merluccius productus) using whole otoliths and sections of otoliths. J. Fish. Res. Board Can. 36:141-151. 1979b. New information on the longevity of the Pacific ocean perch (Sebastes alutus). J. Fish. Res. Board Can. 36:1395-1400. BENNETT, J. T, G. W. BOEHLERT, AND K. K. TUREKIAN. 1982. Confirmation of longevity in Sebastes diploproa (Pisces: Scorpaenidae) using ^"*Pb/"*Ra measurements in otoliths. Mar. Biol. (Berl.) 71:209-215. 116 BOEHLERT: AGE DETERMINATION IN FISHES BINGEL, F. 1981. Growth of a cod otolith on the crystal level. Meeres- forschung 28:212-215. BLACKER, R. W. 1974a. Recent advances in otolith studies. In F. R. Har- den Jones (editor), Sea fisheries research, p. 67-90. John Wiley and Sons, N.Y. 1974b. The ICNAF cod otolith photograph exchange scheme. In T. B. Bagenal (editor). The ageing of fish, p. 108-113. Unwin Brothers, Ltd., Surrey BOEHLERT, G. W. 1980. Size composition, age composition, and growth of the canary rockfish, Sebastes pinniger, and the splitnose rockfish, S. diploproa, from the 1977 Rockfish survey Mar. Fish. Rev. 42(3-4):57-63. BOEHLERT, G. W, AND R. F KAPPENMAN. 1980. Variation of growth with latitude in two species of rockfish {Sebastes pinniger and S. diploproa) from the northeast Pacific Ocean. Mar. Ecol. Prog. Ser. 3:1-10. BRANDER, K. 1974. The effects of age-reading errors on the statistical reliability of marine fishery modelling. In T. B. Bagenal (editor), The ageing of fish, p. 181-191. Unwin Brothers, Ltd., Surrey. Crivelli, a. 1980. The eye lens weight and age in the common carp, Cyprinus carpio L. J. Fish Biol. 16:469-473. Fawell, J. K. 1974. The use of image analysis in the ageing of fish. InT. B. Bagenal (editor), The ageing offish, p. 103-107. Unwin Brothers, Ltd., Surrey. GUNDERSON, D. R., AND T. M. SAMPLE. 1980. Distribution and abundance of rockfish off Washing- ton, Oregon, and California during 1977. Mar. Fish. Rev. 42(3-4):2-16. HIRSCHHORN, G. 1974. The effect of different age ranges on estimated Ber- talanffy growth parameters in three fishes and one mol- lusk of the northeastern Pacific Ocean. In T. B. Bagenal (editor), The ageing of fish, p. 192-199. Unwdn Brothers, Ltd., Surrey. KIMURA, D. K., R. R. MANDAPAT, AND S. L. OXFORD. 1979. Method, validity, and variability in the age determi- nation of yellowtail rockfish (Sebastes flavidus), using otoliths. J. Fish. Res. Board Can. 36:377-383. Malcolm, J. R. and R. J. Brooks. 1981. Eye lens weight and body size as criteria of age in beaver {Castor canadensis). Can. J. Zool. 59:1189-1192. Maraldo, D. C, and H. R. MACCRIMMON. 1979. Comparison of ageing methods and growth rates for largemouth bass, Micropterus salmoides Lacepede, from northern latitudes. Environ. Biol. Fish. 4:263-271. Mason, J. E. 1974 . A semi-automatic machine for counting and measur- ing circuli on fish scale. In T. B. Bagenal (editor). The ageing offish, p. 87-102. Unwin Brothers, Ltd., Surrey. NIE, N. H., C. H. HULL, J. G. JENKINS, K. STEINBRENNER, AND D. H. BENT. 1975. Statistical package for the social sciences. 2d ed. McGraw-Hill, N.Y, 675 p. Pauly, D., and N. David. 1981. ELEFAN 1, a basic program for the objective extrac- tion of growth parameters from length frequency data. Meeresforschung 28:205-211. POWER, G. 1978. Fish population structure in Arctic lakes. J. Fish. Res. Board Can. 35:53-59. Quinn, T J., n, E. A. Best, L. Bijsterveld, and L R. McGregor. 1983. Sampling Pacific halibut {Hippoglossus stenolepis) landings for age composition: History, evaluation, and estimation. Int. Pac. Halibut Comm. Sci. Rep. 68, 56 p. Sandeman, E. J. 1961. A contribution to the problem of age determination and growth-rate in Sebastes. Rapp. P-V Reun. Cons. Perm. Int. Explor. Mer 150:276-284. 1969. Age determination and growth rate in redfish, Sebastes sp., from selected areas around Newfound- land. Int. Comm. Northwest Atl. Fish. Res. Bull. 6:79- 106. Six, L. D., and H. F Horton. 1977. Analysis of age determination methods for yellow- tail rockfish, canary rockfish, and black rockfish off Ore- gon. Fish. Bull., U.S. 75:405-414. SOUTHWARD, G. M. 1962. A method of calculating body lengths from otolith measurements for Pacific halibut and its application to portFock-albatross grounds data between 1935 and 1957. J. Fish. Res. Board Can. 19:339-362. Templeman, W, and H. J. Squire. 1956. Relationship of otolith lengths and weights in the haddock Melanogrammus aeglefinis (L.) to the rate of growth of the fish. J. Fish. Res. Board Can. 13:467-487. TROUT, G. C. 1954. Otolith growth of the Bamets Sea cod. Rapp. R-V. Reun. Cons. Perm. Int. Explor. Mer 136:89-102. 1961. The otolith of group-O Sebastes mentella Tra- vin. Rapp. P-V. Reun. Cons. Perm. Int. Explor. Mer 150:297-299. WESTRHEIM, S. J., AND W R. HARLING. 1973 . Report on the 1972 comparison of Pacific ocean perch otolith and scale interpretations. Fish. Res. Board Can. Manuscr. Rep. Ser. 1259, 24 p. Williams, T, and B. C. Bedford. 1974. The use of otoliths for age determination. In T. B. Bagenal (editor), The ageing offish, p. 114-123. Unwin Brothers, Ltd., Surrey. 117 RATES OF ATRESIA IN THE OVARY OF CAPTIVE AND WILD NORTHERN ANCHOVY, ENGRAULIS MORDAX J. Roe Hunter and Beverly J. Macewicz ' ABSTRACT The process of ovarian atresia was described for northern anchovy using a laboratory group in which atresia was induced by starvation. Atretic characteristics of the ovary were described and illustrated, atretic ovarian states defined, and the rate that the ovary passed from one atretic stage to another was measured. The ovaries of starved females regressed rapidly; 3 days after the onset of starvation the ovaries of about half of the females contained yolked oocytes undergoing resorption of yolk (alpha stage of oocyte atresia) and by 23 days after the onset of starvation no yolk remained in the ovaries of any of the females. Gamma+delta stages of atretic follicles persisted in the ovary for over a month, but their decline in abundance indicated that eventually all signs of past reproductive activity would be lost in regressed ovaries. In the natural population, rates of ovarian atresia increased seasonally from only a few percent of the females showing some atresia in peak spawning months to over 50% near the end of the season. Females with low levels of alpha stage atresia (<50% yolked oocytes affected) spawned about half as frequently as did those with no alpha stage atresia. Spavraing was rare (1% of the females) or absent in females with high levels of alpha stage atresia (>50% yolked oocytes affected). Late in the spawning season, it may be possible to forecast the end of spawning in the populations using the frequency of females in the populations with high levels of alpha stage oocyte atresia. Throughout the spawning season atretic rates were higher in small females (standard length <10 cm) than in larger ones indicating that 1-year-old females spawning for the first time have a much shorter spawning season than do older females. Four approaches commonly used to determine the reproductive state of female fishes are 1) staging of ovaries using gross anatomical criteria such as the international Hjort scale (Bowers and Holli- day 1961); 2) calculation of the gonosomatic index (GSI), i.e., gonad weight divided by female weight or the equivalent (de Vlaming et al. 1982); 3) es- timating the mean diameter of the oocytes in the most advanced mode of oocytes (Hunter and Goldberg 1980; Hunter and Leong 1981); and 4) classifying ovaries histologically. Histological classification is superior to all other methods. Two of its great strengths are that the frequency of spawning of multiple spawning fish populations can be accurately estimated using the presence of postovulatory follicles (Hunter and Goldberg 1980) and that regressing ovaries can be distin- guished from immature and from postovulatory ovaries. The histological criteria used to identify regressing ovaries is the presence of many oocytes and follicles undergoing resorption, a process known as atresia. 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. Manuscript accepted April 1984. FISHERY BULLETIN: VOL. 83, NO. 2, 1985. The interpretive power of histological analysis could be enhanced if the process of ovarian atresia were better documented. Specifically, ovarian atretic stages need to be defined, rates of atresia and duration of stages estimated, and the relation between ovarian atretic state and the probability of spawning determined. Such information would facilitate process oriented field studies on re- productive biology, and increase the accuracy of estimates of size at first maturity and size- or age-specific duration of the annual spawning sea- son. This study provides the laboratory and field calibration necessary for the assessment of the reproductive state of northern anchovy, Engraulis mordax, using the atretic condition of the ovary. We identify a range of ovarian atretic characteris- tics that define the atretic condition of the ovary, estimate rates of atresia, and estimate the dura- tion that atretic characters persist in the ovaries of starving females in the laboratory. We use this information to classify ovaries of sea-caught females and estimate the probability of spawn- ing for females with various levels of ovarian atre- sia. We know of no similar work. A large descriptive 119 FISHERY BULLETIN: VOL. 83, NO. 2 literature exists on follicular atresia in fishes (re- viewed by Saidapur 1978) and stages of atretic oocytes and follicles have been defined (Bretschneider and Duyvene de Wit 1947; Lambert 1970a), but only the atretic rates in the guppy have been measured (Lambert 1970a). Considerable at- tention has been devoted to stages of atretic folli- cles (corpora atretica or "preovulatory corpora lutea") because of a presumed endocrine function (see reviews by Hoar 1965; Byskov 1978). The sea- sonal occurrence of atretic oocytes and follicles is often discussed as part of a general description of seasonal changes in the ovary of marine fish; see, for example, cycles described for the gobiid, Gil- lichthys mirabilis (de Vlaming 1972); plaice, Pleuronectes platessa (Barr 1963); Paracentropris- tis cabrilla (Zanuy 1977); and three species of Epinephelus (Bouain and Siau 1983). The propor- tion of females with atretic ovaries or the numbers of atretic oocytes within the ovary is given less often, but a few reports exist. For example, atresia ranged from 0 to 69c of the oocytes in female had- dock, Melanogrammus aeglefinus (L.) (Robb 1982); corpora atretica increased to about 39c of the oocytes during the postspawning period of the dab, Limanda limanda (L.) (Htun-Han 1978); and atre- tic oocytes varied from 13% of yolked oocytes dur- ing the prespawning period to 100% during the postspawning period of the snapper, Chrysophrys auratus (F.) (Crossland 1977). Some attention has been given to the issue of whether or not atretic rates can account for differences in fecundity among females fed high and low rations. It ap- pears that ration-related differences in fecundity are more closely tied to production rates of oocytes rather than atretic rates (Tyler and Dunn 1976; Wootton 1979). In summary, our literature review indicates that ovarian atresia has yet to be used for quantitative estimation of any reproductive pro- cesses in marine fish populations, although it has been used in general descriptions of the seasonal- ity of reproduction for many years. METHODS Laboratory Experiment Adult northern anchovy captured by commer- cial bait fishermen on 23 February 1982 were kept in a live car in San Diego Bay. Three days later about 1,000 fish averaging 104 mm SL (9.50 g) were taken to the laboratory and held in a 4.6 m diameter pool (1 m deep ) at which time the first fish sample was taken. Over the first 34 d in captivity, 120 samples of 18-24 females were taken at 3-4 d inter- vals with the final sample taken after 62 d in captivity. The temperature of the seawater ranged from 15.5° to 16.5°C. The fish were not fed during the first 27 d in captivity because starvation was used to trigger the resorption of the ovary; thereafter they were fed daily. On the 27th day of starvation the ovaries had regressed from A9c of female body weight to 0.8% and feeding was resumed because we wished to learn how long the atretic characters would last once the fish began to feed. In our calculations of atretic rates of laboratory females, we assumed that all the females at the time of capture had active ovaries without atresia, although no samples were taken until 3 d after the fish were captured. Only 3% of the 1,680 females taken in a survey conducted at the same time (28 January-8 March 1982) had atretic ovaries, and it was prominent in only 0.1% of the females (50% or more of yolked oocytes were affected). Ninety-six percent of the females in our first sample (taken 3 d after capture) had yolked eggs, and half of them had no atresia. All females sampled during the course of the laboratory experiment were weighed and mea- sured, and the ovary removed, weighed, and a sec- tion removed for histological analysis. Ovaries were fixed in 10% neutral buffered Formalin^ and embedded in Paraplast. Histological sections were cut at 6 /um and stained with Harris hematoxylin followed by eosin counterstain. Sea Data The ovaries of northern anchovy taken in trawl surveys used for biomass estimation (Stauffer and Picquelle^) and various other collections from commercial seiners and midwater trawls were his- tologically examined. The number of females examined per catch (trawl, purse seine, or lam- para net) has varied from 10 to 20. Some collec- tions were quite small, especially those taken out- side the main spawning season in the Southern California Bight; these small collections may con- sist of only two catches, whereas those taken dur- ing the main spawning months (February-March) ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. ■■'Stauffer, G., and S. Picquelle. The 1980 and 1981 egg produc- tion estimates of anchovy spawning biomass. Unpubl. manu- scr. Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY were large, often consisting of 100 or more catches of 10-20 females each. In all collections ovaries were classed according to atretic characteristics as well as on the basis of the presence of postovula- tory follicles (age 0 and age 1 d) and hydrated eggs according to the method described by Hunter and Goldberg (1980). All but two of the collections were from the Southern California Bight or northern Baja California, the region where the Central subpopulation of the northern anchovy is concen- trated (Vrooman et al. 1981). Two collections were from the vicinity of Monterey and San Francisco Bays. Fish from these areas appear to have a dif- ferent spawning season from those of fish to the south so they are listed separately in our seasonal tabulations. All collections were classified using histological criteria to determine the incidence of ovarian atretic states as a function of female size, season, and reproductive state. Histological Characteristics We describe below the histological characteris- tics of four oocyte classes and four stages of atresia in the northern anchovy. These stages and classes are subsequently used to define various ovarian atretic states in laboratory and sea-caught female anchovy. Oocyte Classes The northern anchovy is a multiple spawning fish (Hunter and Goldberg 1980) with asynchro- nous oocyte development (oocytes in many stages of development occurring simultaneously in re- productively active ovaries; Wallace and Selman 1981). During the spawning season oocyte devel- opment is a continuous process involving all stages with a new spawning batch maturing every week to. 10 d (Hunter and Leong 1981). Oocyte development and maturation in teleosts, reviewed recently by Wallace and Selman (1981), has fre- quently been subdivided into many stages (An- drews 1931^ Yamamoto 1956; Lambert 1970b), but our work required a simpler histological classifi- cation system. We have combined the stages of past authors into four oocyte classes (unyolked oocytes, partially yolked oocytes, yolked oocytes, and hydrated oocytes), and we describe the his- tological characteristics of each class below. "Andrews, C. B. 1931. The development of the ova of the California sardine (Sardina caerulea). Unpubl. manuscr., 88 p. Stanford Univ., Stanford, CA 94305. 1) Unyolked Oocytes — This class includes all oocytes without yolk that are about 0.04 mm or larger and range upward in size to about 0.35 mm (U, Fig. la, b). Oocytes <0.04 mm are excluded because they consist mostly of "oogonium nests", do not have a true follicle layer, and do not seem to undergo degeneration (o, Fig. lb). The smaller oocytes within this class (0.04-0.15 mm) are spher- ical, have a large nucleus with a narrow homoge- nous very densely staining cytoplasm (Fig. lb). A very thin single layer of elongated, spindlelike cells (the beginning of the granulosa layer) sur- rounds these small oocytes. The large oocytes in this class are oval, the C5rtoplasm stains faintly with hematoxylin and has a cloudy, mottled ap- pearance (Fig. Id). The oval nucleus of these oo- cytes contains several nucleoli and is surrounded by a granular perinuclear zone. In these larger oocytes a thin, definite, faintly eosinophilic stain- ing, hyaline membrane (precursor of the zona radiata) appears between the oocyte and the grow- ing follicle. The follicle consists of a narrow single inner layer of cuboidal granulosa cells and a single outer layer of flat elongated thecal cells with some blood capillaries. The larger oocytes also may have some small vesicles in the periphery of the cyto- plasm. These vesicles are at times difficult to dis- tinguish and they seem to disappear in yolked oocytes. No oil vacuoles exist as northern anchovy eggs do not contain oil droplets. 2) Partially Yolked Oocytes — Oocytes in this class are in the early stages of yolk deposition (vitellogenesis) and range in size from 0.3 to 0.5 mm (major axis) (P, Fig. Id, g). The class includes oocytes in the initial stage of yolk deposition up to and including those in which yolk granules or spherules extend three-fourths of the distance from the periphery to the perinuclear zone. Yolk deposition starts at the periphery of the oocyte cytoplasm as small eosinophilic staining granules and then subsequently spreads internally until they nearly reach the finely granular perinuclear zone. Usually by this time the granules have be- come small spherules. The oval-shaped nucleus of oocytes in this class contains several nucleoli. Delicate striations appear on the hyaline mem- brane between the oocyte and follicle layer at the time yolk appears in the oocyte. As maturation proceeds, the follicle layer becomes wider due to an increase in the width and proliferation of granulosa cells. The thecal cells do not increase in size but remain elongated, flat cells with occa- sional blood capillaries and form a thin outer cov- 121 FISHERY BULLETIN; VOL. 83, NO. 2 Figure l. — Development of northern anchovy ovary at various magnifications (stain = H & E, bar = 0.1 mm), a) Immature ovary consisting of unyolked oocytes and no atresia, b i Enlargement of (a I showing small spherical unyolked oocytes (U) with a large central nucleus and "oogonium nests" (o). c) Normal mature ovary with many fully yolked oocytes (Y). d) All stages of oocytes: unyolked (U), partial yolked (Pi, and yolked (Y), are present in normal mature ovaries, (g = granulosa cell layer, z = zona radiata, n = nucleus, y = yolked globules.) e) Prespawning ovary showing migration of nucleus to the animal pole, f) Enlargement of a migratory nucleus oocyte (M). (n = nucleus, y = yolk globules.) g) Imminent (• 12 h) spawning ovary with hydrated oocytes (H) still within the follicle layer (U = unyolked, P = partial yolked. ) h ) Enlargement of a hydrated oocyte. Note that the yolk globules have fused into yolk plates (yp) and there is no prominent nucleus due to disintegration of the nuclear membrane. 122 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY ering to the follicle. The thecal cells do not change until hydration when they become even flatter and have a stringy appearance. 3) Yolked Oocytes — Oocytes in this class range from 0.45 to 0.80 mm (major axis), and all contain yolk spherules or globules throughout the region between the periphery of the oocyte and the perinuclear zone (Y, Fig. Ic, d). As vitellogenesis continues, the yolk varies from spherules in the smaller oocytes to large globules in the larger ones. Just prior to spawning (<24 h) the globules fuse to form yolk plates (Fig. Ih). Such oocytes are excluded from this oocyte class, this characteristic being diagnostic of the last class (hydrated oo- cytes). The nucleus of oocytes in the yolked oocyte class is oval with numerous nucleoli. The granulosa cells have a wide rectangular shape in cross section and a large oval nucleus; their walls are clearly evident in sagittal section where they form polyhedrons. The zona radiata is a wide, striated, eosinophilic band until hydration when it stretches thin and the striations disappear. 4) Hydrated Oocytes — These oocytes range in size from 0.75 to 1.2 mm (major axis) (H, Fig. Ig, h). Hydration (rapid uptake of fluid by the follicle, Fulton 1898) begins when the nucleus has mi- grated to the animal pole (M, Fig. le, f) and yolk globules first fuse to form yolk plates, and it ends when the hydrated oocyte is ovulated. The nucleus of hydrated oocytes is not visible except in the earliest phase because after the nucleus migrates, the nuclear membrane disintegrates dispersing its contents into the cytoplasm. During hydration all yolk globules fuse into plates and the oocyte expands greatly, stretching the granulosa and thecal cell layers. At this time, the granulosa cells in cross section appear as long, thin rectangles, the thecal cells are extremely flat and have a stringlike appearance, and the zona radiata is very thin and lacks striations. Hydrated oocytes are the most ephemeral of all oocyte classes since this stage lasts for less than a day, whereas the other stages are always present in reproductively active anchovy ovaries. Migratory nuclei may be seen as early as 24 h before ovulation, but hy- drated oocytes in which all globulues are fused to form yolk plates do not occur earlier than 12 h before spawning. We have never observed atre- sia in hydrated oocytes; apparently, in northern anchovy, nearly all hydrated oocytes are ovu- lated. Atretic Stages The nomenclature and general characteristics used for the four atretic stages given below follow those of Bretschneider and Duyvene de Wit (1947) and Lambert (1970a). In the initial stage of the atretic process (alpha (a)), the entire oocyte is resorbed including the yolk, if present, by the hypertrophying granulosa cells of the follicle. In the next stage (beta (^)), the major degeneration and resorption of the follicle (granulosa and thecal cells) occurs. In the third (gamma (y) ) and fourth (delta (8)) atretic stages, regression of the theca and granulosa cells continues, greatly reducing the size of the follicle, and a yellow-brown pigment appears. The histological characteristics used to identify these stages are outlined below. 1) Alpha (a) Stage Atresia — In the alpha stage of atresia the oocyte is resorbing leaving only the follicular layers. The early phase of alpha stage atresia is characterized by the disintegration of the nucleus, evident by an irregular shape, and a granular, dark basophilic staining, and the disin- tegration of some of the yolk globules, indicated by less refractive globules, fused globules, or globules expanded and of less regular shape (Fig. 2a, b, c). The zona radiata slowly dissolves as indicated by the loss of striations and uneven diameter (Fig. 2b). In subsequent phases of alpha atresia, granulosa cells enlarge and, upon rupture of the zona radiata, invade the degenerating oocyte (Fig. 2d). Yolk adjacent to the invading granulosa cells liquifies (loses all structural integrity and appears as a homogeneous eosinophilic area) and becomes phagocytized by the granulosa cells as indicated by the presence of yolk in the vacuoles of these cells. The basophilic staining cytoplasm is also resorbed by the granulosa cells. In the alpha stage of atresia, blood capillaries and vessels are numer- ous in the thecal connective layer which does not proliferate or invade the oocyte but remains as a thin layer covering the granulosa cells. The alpha stage ends when resorption of the oocyte is com- plete (all cytoplasm and yolk are gone). The result- ing structure (beta stage) is usually much smaller than the original oocyte. The subsequent atretic stages (beta-delta) are steps in the resorption of the remaining follicle and the structure at this point is called an atretic follicle, the term atretic oocyte being reserved for only the alpha stage of atresia. In unyolked oocytes the alpha stage process is similar but without yolk (Fig. 2e, f). The nucleus 123 FISHERY BULLETIN: VOL. 83, NO. 2 Figure 2. — Alpha (a) stage atresia in yolked ( Y) and unyolked (U) oocytes (bar = 0.1 mm), a) and b) Yolked oocyte undergoing alpha atresia (Ya). Notice dark irregular nucleus (n), uneven dissolving zona radiata (z), and hypertrophic granulosa cells (g); iUa = alpha atresia of a large unyolked oocyte), c) and d) Only remnants of yolk (y) remain among the invasive phagocytizing granulosa cells in this late phase of alpha atresia (!« ). Note also the thecal layer (t) and the closely associated red blood cells ( b). e) and f) Unyolked oocytes in the alpha stage of atresia (Ua), note enlargement of granulosa (g) and disintegration of nucleus (n). (Ya = alpha yolked atretic oocyte, /3 = beta atresia.) 124 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY disintegrates, the thin prezona radiata (if present) dissolves and the granulosa cells enlarge, and, with only a slight proliferation, phagocytize the unyolked oocyte. When resorption is complete, all that remains is the follicle. 2) Beta ((3) Stage Atresia — Initially the beta stage atretic follicle is a compact structure com- posed of numerous disorganized granulosa cells surrounded by a thin thecal and blood vessel layer. The nucleus of some of the granulosa cells is pyc- notic and many of the cells contain a large in- tracellular vacuole that may be empty or contain amorphous particles. Occasionally one or more large intercellular cavities may exist among the granulosa cells (Fig. 3b, d). Preovulatory beta stage atretic follicles containing such cavities may easily be confused with postovulatory follicles (older than 48 h) and, as a consequence, we do not age postovulatory follicles older than 48 h (Hunter and Goldberg 1980). In addition, small (older) beta stage atretic follicles from yolked oocytes (Fig. 3c, d) are indistinguishable from beta stage atretic follicles from unyolked oocytes. Thus, we do not identify the original oocyte type undergoing at- resia in beta or subsequent atretic stages; such distinctions are made only for alpha stage atretic oocytes. Three different patterns of atresia may occur at the conclusion of the beta stage: 1) The follicle may follow the classic pattern outlined by Bretschneider and Duyvene de Wit (1947) and pass through subsequent gamma and delta stages (both characterized by increased pigmentation, see be- low); 2) the follicle may be completely resorbed during the beta stage leaving no histological characteristics that can be identified; and 3) the follicle may pass directly from a beta stage struc- ture to a delta stage structure without passing through the intervening gamma stage. In north- ern anchovy, either the duration of the gamma stage is very short or few follicles pass through the gamma stage into the delta stage, because in re- gressing ovaries the incidence of gamma stages is very low compared with those of either beta or delta stages. 3) Gamma (y) Stage Atresia — The gamma stage atretic follicle is usually much smaller than the typical beta stage follicle (Fig. 3e). The granulosa cells contain flocculent material of light-yellow hue and have nuclei of very irregular shape. The granulosa cells are surrounded by many fewer thecal cells and blood vessels than occur in the beta stage atretic follicles. Occasion- ally we see an atretic follicle of quite different appearance in anchovy ovaries which we classify as a gamma stage atretic follicle; they are included in the gamma stage because they also contain flocculent material of light-yellow hue. In this case, the flocculent yellow material is extracellu- lar rather than intracellular, and the material is encapsulated by a layer of granulosa and thecal cells. It is possible that the extracellular flocculent material is produced by the disintegration of granulosa cells. 4) Delta (6) Stage Atresia — The diagnostic characteristic of this stage is the presence of a dark yellow-brown, finely granular pigment in the granulosa cells (Fig. 3f). The delta stage atretic follicles are normally very small structures typi- cally composed usually of 2-20 granulosa cells in the ovarian connective tissue stroma. Thecal cells and blood vessels no longer encompass the granulosa cells. In our laboratory work 3-4 levels of abundance were recorded for each of three atretic classes seen in anchovy ovaries (alpha, beta, and gamma + delta stages). The gamma and delta stages were combined since gamma stages were rare. In addi- tion, the alpha stage atretic class was further sub- divided into three groups depending on the type of oocyte undergoing atresia (unyolked, partially yolked, and yolked oocytes). In the discussion that follows we have combined some of the abundance levels and have considered only what we believe to be the most diagnostic atretic characteristics, al- though all atretic characteristics as originally tabulated are given in Tables 1 and 2. The system of atretic classifications was further simplified in our presentation of the analysis of sea-caught specimens, but that will be discussed sub- sequently. RESULTS Rates of Atresia in the Laboratory The speed at which yolked oocytes were resorbed was striking. In the first sample (elapsed time from onset of starvation = 3 d) the ovaries of 11 of the 24 females (46%) had yolked ooc5d;es in the alpha stage of atresia (Table 1). By the 13th day, half of the females no longer had yolked oocj^s, and in the rest of the females 50% or more of their yolked oocytes were in the alpha stage of oocyte 125 FISHERY BULLETIN: VOL. 83, NO. 2 lt3k Figure 3. — Stages of atresia following after complete yolk absorption (bar = 0.1 mm), a) and b) Typical beta (/i) stage atresia. Note the disorganized granulosa cells with some pycnotic nuclei (p) or intracellular vacuoles (v) (t = outer layer of thecal cells, c = intercellular cavities), c) and d) Disintegration of granulosa continues in these older beta (fi) stage atresia. Note the large intercellular cavity (c) and the prominent, contracted thecal cell layer (t). Also present is an unyolked oocyte in early alpha (Ua) stage and several delta (8) stage atresia cells, ei Two types of gamma (y) atresia seen in northern anchovy ovaries. Note flocculent material (f) and the thecal layer (t). f) Delta (6) stage atresia characterized by dark yellow fine granular pigment and an irregular nucleus (n). 126 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY Table l. — Percentage of northern anchovy females with ovaries containing various levels of alpha stage atresia during starvation and after the resumption of feeding.' Feedi ing Percentage of ovaries with levels of alpfia stage atresia Partially yolked Elapsed time condition Unyoll5 None N -=5 N >5 None F ■50% F = 50-90% F ^91% 3 X 24 79 13 8 46 29 25 50 33 0 13 6 X 21 19 14 67 14 14 62 5 10 19 48 9 X 24 12 17 71 0 58 34 0 8 17 42 13 X 20 10 15 75 20 15 20 0 0 5 40 16 X 24 0 12 88 4 17 17 0 0 0 8 20 X 22 0 36 64 0 0 4 0 0 0 4 23 X 23 9 39 52 0 4 13 0 0 0 0 27 X 23 4 57 39 0 0 0 0 0 0 0 34 X 23 70 17 13 4 0 0 0 0 0 0 41 X 18 83 17 0 17 0 0 17 0 0 0 62 X 22 90 5 5 64 9 9 36 23 0 0 ' Feeding begins on the 28th day. ^N = mean number of atretic oocytes per 6 fim section. ^F = mean percentage of atretic oocytes per 6 /um section. Table 2. — Percentage of northern anchovy females with ovaries containing various levels of beta and gamma+delta stage atresia and yolked oocytes during starvation and after the resumption of feeding.' Feeding condition N Percentagi 3 of ovaries with levels of atresia Go Beta stage atresia^ Beta stage atresia with no yolked oocytes^ Gamma -rdelta stage atresia^ icyte types Elapsed time from capture (d) Yolked oocytes present Only partial Starved Fed None N -=5 W >5 None A/ 5=5 N >5 None N s5 N >5 and unyolked oocytes present 3 X 24 71 21 8 0 4 0 92 4 4 96 4 6 X 21 24 24 52 0 5 14 71 24 5 81 19 9 X 24 4 8 88 0 0 33 88 8 4 67 33 13 X 20 0 15 85 0 10 45 60 25 15 45 55 16 X 24 0 12 88 0 12 79 16 46 38 8 92 20 X 22 0 9 91 0 9 86 9 36 55 4 96 23 X 23 0 17 83 0 17 83 13 26 61 0 100 27 X 23 0 44 56 0 44 56 26 35 39 0 100 34 X 23 35 48 17 35 48 17 4 13 83 0 100 41 X 18 88 6 6 78 0 6 0 28 72 17 83 62 X 22 82 18 0 35 5 0 0 50 50 59 41 ' Feeding begins on the 28th day. ^N = mean number of atretic follicles per 6 /xm section. resorption (Fig. 4). None of the females sampled on the 23d day had yolked oocytes, indicating that all yolked oocytes had passed through the alpha stage of atresia by this time. The resorption of unyolked and partially yolked oocytes began just as rapidly as did the resorption of yolked oocytes. The percentage of females with atretic unyolked oocytes in the alpha stage in- creased sharply from 21% on the 3d day of starva- tion to 90% on the 13th day. Throughout the rest of the 27-d starvation period nearly all of the females (90-100% ) had some unyolked oocytes in the alpha stage of atresia, indicating a continual recruit- ment of atretic follicles from the unyolked and partially yolked oocyte classes. Thus, alpha stage unyolked and partially yolked oocytes are present in regressing ovaries for a much longer period than is the alpha stage of yolked oocytes. This difference probably is due to the greater number of unyolked and partially yolked oocytes in mature ovaries. Yolked oocytes constitute <1% of the total number of oocytes present in mature ovaries. The incidence of beta stage atretic follicles also increased sharply over the first 9 d of the starva- tion period and followed a pattern similar to that described for the incidence of alpha atresia from unyolked eggs (Fig. 4). After attaining a high value on the ninth day the incidence of beta atresia remained high until the end of the starva- tion period as atretic follicles from yolked and un- yolked oocytes degraded from the alpha to the beta stage of atresia. Incidence of gamma -I- delta stages (the third and fourth stages of follicle degenera- tion) increased later than did alpha and beta stages and remained high after the onset of feeding. Once feeding resumed (day 28), rapid resorption of yolked and unyolked follicles ceased and the 127 FISHERY BULLETIN: VOL. 83, NO. 2 STARVED FED lOOr YOLKED EGGS PRESENT 50% OR MORE YOLKED OOCYTES a ATRESIA • p ATRESIA PRESENT A a ATRESIA OF UNYOLKED OOCYTES PRESENT 30 40 ELAPSED TIME (days) FIGURE 4. — Percentages of captive female northern anchovy with ovaries having various atretic characteristics during a 27-d starvation period and after the onset of feeding. Each percentage is calculated from a sample of 18-24 females (see Tables 1 and 2); alpha, beta, and gamma through delta stages of atretic follicles are those defined by Bretschneider and Duyvene de Wit (1941). dominant process became maturation rather than resorption. This was indicated by sharp declines in the percentages of females with alpha stage atre- sia of unyolked oocj^es and beta stage atretic folli- cles, and the reappearance of yolked oocytes (day 41). After only 1 wk of feeding the percentage of females with alpha stage atresia from partially unyolked oocytes dropped from 96 to 30%. Some inferences can be drawn from these data regarding the duration of atretic stages. The sharp and simultaneous decline in beta stage atretic fol- licles and alpha stage atresia of unyolked oocytes (following the onset of feeding) indicates that alpha and beta stages must have a short and simi- lar duration. The duration of alpha and beta atre- sia probably is <2 wk, since the incidence of these two stages dropped to very low levels 2 wk after the onset of feeding; a lag of about 1 wk existed between the first high incidence of females with beta atresia (9 d) and that for gamma -H delta (16 d), indicating that the duration of the beta stage may be about 1 wk. The continued high inci- dence of gamma -I- delta stages of atretic follicles long after the onset of feeding indicates that these late atretic stages must persist in the ovary for much longer periods than alpha or beta stages. Although gamma -I- delta stages were present in all ovaries on the last day of the experiment their abundance within an ovary had decreased indicat- ing that even the delta stage would eventually disappear, eliminating the last histological sign of past reproductive activity. We conclude from these inferences that the alpha and beta stages persist in the ovary for 1 wk or less whereas gamma -i- delta stages persist for over a month, but eventu- ally all signs of past reproductive activity are lost. The occurrence of alpha stage atresia of yolked oocytes is the best characteristic to use to back- calculate the time of past reproductive activity in field-caught specimens because the stage is of relatively short duration and the time required to resorb all yolked oocytes is relatively short. On the other hand, alpha stage atresia of unyolked oo- cytes, and beta and gamma -^ delta stages are less useful for back-calculations because these stages may occur in an ovary for extended periods while atretic oocytes are recruited from the large reser- voir of unyolked oocytes in the ovary. In addition, estimates of the time since the onset of atresia in ovaries without yolked oocytes (using the inci- dence of beta or gamma + delta atretic stages) will always be uncertain because atresia of unyolked oocytes may occur at low levels in immature or developing ovaries as well as in regressing ovaries. For the laboratory specimens, we calculated the average elapsed time from the onset of ovary re- sorption using various classes of alpha stage atre- sia of yolked oocytes and beta atresia in ovaries without yolked oocytes (Table 3). We prefer the criteria of 50% or more of the yolked oocytes with alpha stage atresia because it is likely that no spawning will occur in such females. The average duration of this stage (alpha, yolked, ^ 50% ) in the starving laboratory females was about 9 d and ranged from <3 to 20 d from the onset of starva- tion. Starvation may have induced a higher rate of 128 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY TABLE 3. — Mean and maximum duration of various atretic characteristics of the ovaries of starved northern anchovy. Mean Metximum duration duration Atretic characteristics (d) (d) Alpha atresia of yolked oocytes present 8.0 20 Alpha stage atresia in: <50% of yolked oocytes 4.5 9 50-90% of yolked oocytes 8.1 13 91% or more of yolked oocytes 9.3 20 50% or more of yolked oocytes 9.0 20 No yolked oocytes present and beta atresia present >16 >27 oocyte resorption than usually occurs under natural conditions. Variation in the female nutri- tional state, food ration, water temperature, day length, and a host of other variables may affect rates of atresia. In addition, field data indicate (see next section) that some spavming may occur in females with low to moderate levels of alpha (yolked) atresia, indicating that such stages may persist under natural conditions for extended periods. Despite these uncertainties we believe that our laboratory estimates of atretic rates are useful for making a rough estimate of the mini- mum time elapsed since the end of the spawning season in sea-caught females. Natural Rates of Atresia In this section we analyze sea data taken since 1977 for the occurrence of four ovarian atretic states in a northern anchovy population: Atretic state 0 — no alpha atresia of yolked oo- cytes (yolked oocytes present). Atretic state 1 — alpha atresia of yolked oo- cytes where <50% of the yolked oocytes are af- fected. Atretic state 2 — alpha atresia of yolked oocytes where 50% or more of the yolked ooctyes are af- fected (Fig. 5a, b). Figure 5. — Northern anchovy ovaries wath increasing atresia states (bar = 0.1 mm), a) 50% of all yolkfed oocytes (Y) are in an alpha (a) stage of atresia (both early and late are counted). This is the division point between atretic state 1 and atretic state 2. b) 100% (all) yolked oocytes are in an alpha stage of atresia (Ya). Also present are a few unyolked alpha atretic oocytes and several beta (j8) stage atretic follicles. This is still in atretic state 2. c) All yolk has been resorbed leaving only unyolked oocytes (U) and many beta (/3) stage atretic follicles. This is atretic state 3. Atretic state 3 — ovaries with no yolked oocytes present and beta stage atresia present (Fig. 5c). In addition to the atretic condition of the ovary, we also include histological evidence of recent or imminent spawning using the system of Hunter '>S /3 129 FISHERY BULLETIN: VOL. 83, NO. 2 and Goldberg (1980), i.e., presence of hydrated eggs (imminent spawning), day 0 or new post- ovulatory follicles (spawning on the night of cap- ture), and 1-d-old postovulatory follicles (spawn- ing on the night before capture). We also include the number of females judged to have inactive or immature ovaries with no evidence of atresia. All data on the incidence of reproductive states are given in Table 4. In the discussion that follows we select and regroup these data in various fashions to test hypotheses and document trends. Incidence of Spawning in Atretic Females An important assumption underlying interpre- tation of ovarian atresia is that the spawning sea- son has or is going to cease, in other words, the probability of spawning in females with atretic ovaries would be expected to be low. To test this assumption we selected from Table 4 the females which had alpha stage atresia of yolked oocytes (atretic states 1 and 2) or yolked oocytes without alpha atresia (atresia state 0) and calculated the percentage of these females that had hydrated oocytes, new (day 0) postovulatory follicles, and 1-d-old postovulatory follicles. Of the females classed in atretic state 1 (females with <50% of the yolked oocytes in alpha stage of atresia), 14% showed evidence of recent or imminent spawning (postovulatory follicles or hydrated oocytes); 29% of the females without atresia showed evidence of spawning (Table 5). Only 1% of those in atretic state 2 (females 50% or more atretic yolked oo- cytes) had recently been reproductively active. That 1.8% of females in atretic state 1 had hy- drated eggs and 3.7% had age 0 d postovulatory TABLE 4. — Numbers of female northern anchovy in various atretic and reproductive states northern California Postovulatory Collection dates number of mature Atretic Hydrated follicles Yolked No yolked Immature no females/collection state' oocytes 0 day2 1 day3 oocytes oocytes histology" Total 1977 0 1 2 13 4 20 09/09-09/10 1 2 0 0 (10) 3 3 3 X 1 2 0 13 7 0 23 1978 0 1 1 05/07-05/1 1 1 2 1 5 4 6 4 (10) 3 6 6 X 0 0 1 10 6 0 17 1979 0 39 44 52 279 89 110 613 01/26-02/16 1 2 1 1 1 1 (10) 3 2 2 X 39 44 53 280 91 110 617 1979 0 16 51 45 284 27 18 441 03/22-04/14 1 2 1 5 36 6 42 6 (10) 3 16 16 X 16 52 50 326 43 18 505 1979 0 1 8 4 13 06/09-06/19 1 2 3 1 3 1 (12) 3 16 16 X 0 0 1 12 20 0 33 1979 0 9 31 40 09/19-09/23 1 2 5 0 5 (10) 3 25 25 X 0 0 0 14 56 0 70 1980 0 25 72 52 241 390 03/20-04/10 1 2 9 1 9 1 (20) 3 2 2 X 25 72 52 251 2 0 402 1980 0 4 9 63 B 6 90 04/24-04/27 1 2 11 14 11 14 (20) 3 63 63 X 0 4 9 88 71 6 178 'Atretic state 0 = no alpha stage atresia of yolked oocytes. state 1 = alpha stage atresia of yolked oocytes present but • 50% oocytes affected, state 2 = alpha stage atresia of yolked oocytes present, 50% or more oocytes affected state 3 = no yolked oocytes present and beta stage atresia present. 130 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY follicles indicate that some of the females in atretic state 1 spawn despite the atretic condition of their ovaries. On the other hand, only two females classed in atretic state 2 had a recent history of spawning. These two females had few yolked oocytes remaining, all but one was in alpha atresia. This atresia may have increased or even started during the time elapsed between hydra- tion and capture. In short, the females in atretic state 2 probably did not spawn in the highly atretic state in which they were captured. No doubt exists that females with moderate levels of ovarian atresia are capable of spawning because they often are hormonally induced to do so in the laboratory. In summary these data indicate that significant numbers of females in atretic state 1 may continue to spawn under natural conditions, although the probability of spawning appears to be about half that offish without ovarian atresia. Few or none of the females in atretic state 2 con- tinue to spawn indicating that this stage is proba- bly the best one to use to calculate cessation of spawning in the population. Forcasting the end of Spawning Using Ovarian Atretic States In our laboratory study atretic state 1 had an average duration of about 5 d and atretic state 2, 9 d; state 3 was in excess of 16 d and probably per- sists for 30 or more days (Table 3). In the sea, linear projections of the end of the spawning sea- son from early incidence of atresia is not realistic since the numbers of females with regressing ovaries would be expected to increase sharply at the end of the season. listed in order of collection dates for southern and Baja California (1977-82) and for (1979, 1982). Rostov /ulatory Collection dates number of mature Atretic Hydrated folli icles Yolked No yolked Immature no females/collection state' oocytes Oday^ 1 day 3 oocytes oocytes histology" Total 1980 0 3 32 11 46 05/15-05/28 1 2 1 5 2 6 2 (20) 3 16 16 X 0 1 3 39 27 0 70 1981 0 119 122 148 862 58 1,309 02/05-03/06 . 1 2 2 1 19 3 22 3 (15) 3 10 10 X 119 124 149 884 68 0 1,344 1981 0 77 96 113 559 7 852 04/01-04/19 1 2 3 1 2 57 45 62 46 (15) 3 19 19 X 77 100 115 661 26 0 979 1981 0 2 1 7 10 04/15-04/30 1 2 5 8 5 8 (15) 3 7 7 X 0 2 1 20 7 0 30 1982 0 104 101 189 1,172 52 8 1,626 01/28-03/08 1 2 2 2 10 32 2 46 2 (15) 3 6 6 X 106 103 199 1,206 58 8 1,680 1979^ 0 1 42 43 03/20-03/22 1 2 1 40 41 41 41 (30) 3 25 25 X 0 2 0 123 25 0 150 19825 0 2 27 2 11 42 01/22-01/25 1 2 3 2 1 14 5 19 6 (15) 3 13 13 X 5 0 3 45 16 11 80 ^New postovulatory follicles <24 h old, ^Postovulatory follicles about 24 h old "Female not examined histologically, ovary ^1% of body weight. ^Female northern anchovy from northern California; rest of females were from southern and Baja California. 131 FISHERY BULLETIN: VOL. 83, NO. 2 TABLE 5. — Percentage of northern anchovy females taken from 1977 to 1982' that were classed in three atretic states that occurred in each of five reproductive classes. Reproductive state Atretic state Percent yoltced oocytes with alpha stage atresia Postovulatory follicles Hydrated oocytes (%) Oday (%) 1 day (%) Spawning No evidence recent or of recent Total imminent^ spawning^ number (%) (%) of females 0 0 7.5 9.7 12.0 29.3 70.7 5.090 1 <50 1.8 3.7 8.1 13.6 86.4 273 2 =50 0 0.7 0.7 1.4 98.6 140 ' Calculated from data given in Table 4; only females with yolked oocytes are considered. ^Females with either hydrated oocytes or postovulatory follicles ages 0 or 1 d (the sum of the first three columns). ^Females with yolked oocytes but without hydrated oocytes or postovulatory follicles. This nonlinearity becomes obvious when the end of the spawning season is extrapolated from numbers of females classed in atretic state 2. For example, of the 1,620 mature females taken dur- ing the peak of spawning (28 January-18 March 1982) in southern California (Table 4), only two were in atretic state 2 and 1,612 had yet to pass through state 2. Since laboratory data indicate that about 9 d are required to pass through atretic state 2, it would require (1,612/2) x 9, or over 7,000 d for the entire population to become atretic at the rates of atresia observed in February, which, of course, is nonsense. Projections of the end of the spawning season using higher rates of atresia taken in April in southern California (24-27 April 1980) give a more realistic projection ((87/ 14) X 9 = 56 d). Such an arithmetic projection may be inappropriate for collections which have a very high rate of atresia such as those taken in Monterey in March 1979 ((84/41) x 9 = 18 d), and a geometric model might be preferable. The point we wish to emphasize is that atretic rates are nonlinear over the season with the rate increasing markedly as the season closes. Thus only samples taken near the close of the spawning season are of value for forecasting the end of spawning for the population. Seasonal Changes in Atresia Among Females of Different Lengths To evaluate how atretic rates change among females of different lengths, we segregated our data into two length classes (females ^ 10 cm SL and those >10 cm SL) and calculated the percent- age of mature females that had atretic ovaries (atretic states 1-3 combined). Mature is defined here as all females except those which have yet to reach first maturity (small females with small immature non-atretic ovaries). We also calculated the fraction of females in each length class with 1-d-old postovulatory follicles, a measure of the percentage of females spawning daily (Hunter and Goldberg 1980). In every case, regardless of cruise or season, small females (^10 cm SL) consistently had a higher rate of ovarian atresia than did larger ones (>10 cm SL) (Table 6). This is a strong trend as the probability of such an event (9 pairs of the same sign) is (1/2)^. In addition, the difference between pairs was statistically significant (chi-square test) even when the levels of atresia were quite low. For example, in February-March 1981, only 4.1% of the small females and 1.9% of the large females were atretic, yet this difference was significant at P < 0.05 using the chi-square test. As would be ex- pected, the percentage of females with atretic ovaries increased in both length classes as the season progressed from January through June. The consistency of the differences in the inci- dence of atresia between large and small females indicates that the smaller ones must have a much shorter spawning season than larger ones. Females <10 cm long are typically about 1-yr-old and are in their first spawning season whereas those longer than 10 cm are predominantly 2-3 yr old and have spawned during the previous sea- sons. These data indicate that the first spawning season of females may be quite short with signifi- cant numbers of females leaving the spawning population in early April, while the older fish con- tinue to spawn. That the rates of atresia in young fish were always higher even in the peak months of spawning such as February and March indicates that a small percentage of small females may only spawn a few times during the season in contrast to the older females which appear to be spawning at about weekly intervals for months. The fraction of small females spawning per day would be expected to be less than larger females since the small females have a higher incidence of ovarian atresia. We calculated the fraction of females spawning per 132 HUNTER and MACEWICZ: ATRESIA IN NORTHERN ANCHOVY OVARY TABLE 6. — Percentage of mature northern anchovy females in two length classes with atretic ovaries. Females from north of Point Conception and groups with fewer than nine females per length class excluded. Percent of Fraction mature Number of mature females females spawning Cruise period mature females' with atretic ovaries^ per day percent^ From to year - 10 cm >10cm ^10 cm >10cm ■ilOcm >10cm 01/26 -»02/ 16 1979 121 297 1.7 0.7 8.0 15.9 01/28^03/18 1982 97 1,523 14.6 2.6 14.1 12.2 02/05 ->03/06 1981 462 824 4.1 1.9 10.2 13.7 03/20—04/10 1980 68 334 8.8 1.8 11.7 15.2 03/22—04/14 1979 30 430 23.3 13.3 3.8 11.8 04/01 —04/19 1981 102 870 39.2 10.0 10.5 12.9 04/24—04/27 1980 64 100 969 26.0 0 8.6 04/15—04/30 1981 10 20 80.0 60.0 0 5.3 05/15-05/28 1980 15 44 73.3 29.5 6.2 4.4 'All females given in Table 4 except those with immature ovaries and those not examined histoloqicallv. ^Ail females in atretic states ,1,2, and 3 combined. ^Fraction of females spawnii nq = F. where F = m^. 2m, m ■ = mature nonspawning females, and m = females with 1-d-old postovulatory follicles. day for the two length classes to test this assump- tion. We used the Stauffer and Picquelle (footnote 3) method for estimating spawning fraction as it corrects for biases in the numbers of females with hydrated eggs, i.e., M, F = 2M, + m li ni where F = fraction of females spawning per day, M^^ = number of females with 1-d-old postovula- tory follicles, and m = number of mature females with no recent spawning history (females with postovulatory follicles or hydrated eggs are excluded). Examination of Table 6 indicates that differences in spawning fraction between the two size classes of females were much less distinct than were the differences in ovarian atresia. Using only the 8 cruises in which the numbers of females in each of the two length classes exceeded 10, the mean difference in spawning fraction (frac- tion for large females — fraction for small females) for the set of 8 cruises was +3.76% with 95% C.I. ±3.50% indicating a small difference in spawning frequency between the two length classes that is just barely significant at the 5% level. We believe the reason that differences in atretic fraction be- tween large and small females are much more consistent than those in spawning fraction is that spawning fraction has a greater variability and a much more limited dynamic range than does the atretic fraction. Spawning fraction varies from 0 to about 16% and may be affected by time of day and schooling behavior (Hunter and Goldberg 1980). Atretic fraction varies from 0 to nearly 100%, is not linked to reproductive behavior, and conse- quently, is probably not affected by time of day or schooling. DISCUSSION Evaluation of Atretic Classification Our objective was to evaluate the use of ovarian atretic states to characterize the reproductive bi- ology of northern anchovy populations. We in- cluded in our analysis of laboratory data many atretic characteristics not used to construct the three atretic states utilized in the analysis of sea data. These additional characters could be used to create additional states or to more precisely de- limit the existing ones. Our selection of charac- teristics was based in part on ease of identification since for population work thousands of histologi- cal sections were examined. Other considerations include the fact that statistical analysis indicated that classifiers frequently confused beta stage atretic follicles in yolked ovaries with postovula- tory follicles older than 24 h, and, as a conse- quence, beta atresia was not used as a diagnostic character in ovaries with yolked oocytes. Alpha stage atresia was the most useful atretic stage because the type of oocyte (yolked) undergoing atresia is still discernible. In addition, alpha stage atretic oocytes can be easily distinguished from postovulatory follicles whereas this is not the case for later atretic stages. Three atretic states were defined and applied to sea data. The incidence of all three atretic states combined was a sensitive index of the reproductive state of the population over the spawning season. 133 FISHERY BULLETIN: VOL. 83, NO. 2 In fact, the atretic condition of the ovary was a more sensitive index of seasonal changes in the reproductive rate among size classes of females than was the incidence of spawning based on the presence of postovulatory follicles. Atretic state 1 (<50% of yolked oocytes in the alpha stage of atresia) was not useful for estimat- ing atretic rates in an absolute sense since this state appeared to persist in natural populations for extended and probably variable periods. Some spawning occurred among females classed in atretic state 1, although the frequency of spawning was less than half of that of females without ovarian atresia. Batch fecundity might also be reduced in females classed in atretic state 1, a speculation worth further study. Atretic state 1 was a useful index of atretic rates during peak spawning months. At such times it was the most common atretic condition and detection of differ- ences in atretic rates among length classes was largely a function of the number of females in this state. Atretic state 2 (5(y7f or more of yolked oocytes in alpha atresia) persisted for about 9 d in the laboratory, and judging by its low frequency in field collections this state may have a similarly short duration in natural populations. Females with ovaries in this state rarely or never spawn, as might be expected, since more than half of the yolked oocytes are not viable. In addition, a short duration of this state also might be expected on the grounds that it seems maladaptive to prolong such a threshold condition. For the above reasons atretic state 2 seems to be the best absolute mea- sure of the rates of ovary resorption in the popula- tion and the only state that might provide an accu- rate forecast of the end decline of reproduction in a population. Unfortunately, accurate forecasts of the end of spawning for a population can be made only near the end of the spawning season. Atretic state 3 (no yolked oocytes with beta at- resia present) identifies females in late post- spawning condition. Such females cannot be sepa- rated from immature females on the basis of gonad weight or using gross anatomical criteria. This state persisted for about 30 d in the laboratory, but it may last much longer under natural conditions while the numerous small oocytes are resorbed. The laboratory data indicate that the duration of this state could be increased if the definitions were changed to include gamma + delta stages of atresia which have a longer life in the ovary than the beta stage. The laboratory data also indicated that even gamma + delta stages of atresia would even- tually disappear from the ovary so that no signs of previous spawning activity would exist in a re- gressed ovary. It is doubtful that the duration of atretic state 3 or any late postspawning state will ever be accurately estimated because it is depen- dent on too many environmental circumstances. Nevertheless, this state is very useful in separat- ing females in postspawning condition from females with no previous reproductive history. This is an essential distinction for estimating spawning biomass (Stauffer and Picquelle foot- note 3) and for determining the size or age at first reproduction (Hunter and Macewicz 1980). Possibly the most important future application of atretic classification of ovaries is for process oriented sea work on the reproductive biology of multiple spawning fish such as the northern an- chovy. Such work does not require a large sample as do estimates of reproductive characteristics for an entire population. The reproductive state of an individual female can be accurately defined by the atretic criteria we have discussed, and the spawn- ing state criteria described by Hunter and Goldberg (1980). The reproductive characteristics of a female can be related to its physiological state (age, fat content, biochemical composition, and in- stantaneous growth rate from otoliths or RNA/ DNA ratios) and functional relationships estab- lished between reproduction and the environ- ment. In this way the factors controlling the duration of the spawning season, and the total fecundity during the season, can be identified under natural conditions. Biological Implications Several important biological conclusions can be drawn from this work. Only a few attempts have been made to estimate the time needed for a folli- cle to disappear by atresia in vertebrates and no information exists for fishes (Byskov 1978). Our focus was on atretic rates of all oocytes in the ovary and not on an individual follicle; neverthe- less, the striking speed with which all yolked oo- cytes passed through the initial stages of atresia indicate that the rate for individual follicles must be high. Similar rates were observed in the guppy by Lambert (1970a). In the guppy, alpha stage atresia of yolked oocytes appears about 1 d after parturition, and beta stage atresia appeared about 2 d after the first alpha stages were detected; beta stages persisted for only 11 d. In the anchovy, the average time for all yolked oocytes in the ovary to pass through alpha atresia was 8.0 d and the 134 HUNTER and MACEWiCZ: ATRESIA IN NORTHERN ANCHOVY OVARY maximum time was 29 d. Thus the effect of atresia on fecundity may be underestimated since the du- ration of atretic stages is short and a small stand- ing stock of atretic oocytes could be an indication of a high loss rate. On the other hand, laboratory studies seem to indicate that atretic rates are not sufficiently high to account for the differences in fecundity observed when fish are fed high and low rations (Tyler and Dunn 1976; Wootton 1979). The duration of the atretic stages in these studies was unknown, however. Additional evidence for the volatility of the re- productive state of anchovy is an important con- tribution of this study. Our laboratory data indi- cated that given a shortage of food the ovary can be rapidly resorbed leaving no trace of former repro- ductive activity in a few months or less, but when given sufficient food atresia stopped, maturation and vitellogenesis resumed, and a reproductively active ovary was rapidly reformed within 35 d. Clearly, in such multiple spawning fishes as the anchovy, more than one spawning season per year is possible given the appropriate environmental conditions. This may explain the occurrence of a second annual spawning period in the Peruvian anchoveta (Santander and Castillo 1976) and the occasional heavy fall spawning of the northern anchovy ( Smith 1972 ). That active ovaries are con- sistently produced from small, inactive ones in 30-60 d in the laboratory (Leong 1971; Hunter and Leong 1981) and that some reproductively active females are found the year around also supports this view. Food shortage does not always lead to regression of the ovary in anchovy or any other multiple spawning fishes. In addition to food ration, regres- sion of the ovary also depends upon the level of energy reserves, the timing of the reproductive cycle, and perhaps certain environmental condi- tions such as temperature and day length. For example, starvation of 40-80 d did not block the initial increase in the size of ovaries of the goby Gillichthys at the start of the reproductive cycle in July but only 23 d of starvation resulted in ovarian regression in January when active vitellogenesis was occurring (de Vlaming 1971). Similarly we noted in a preliminary experiment that starving anchovy of 25^f greater wet weight than those used in this study produced a slower regression of the ovary over a 36-d period than occurred in the present study. The present study is more represen- tative of natural conditions since the fish were taken in midspawning season when their ovaries were active whereas in the preliminary study the fish were taken out of season and fed heavily for 30 d to induce gonad maturation before the onset of the 36-d starvation period. Another important conclusion from this study was that young female anchovy spawning for the first time probably have a much shorter reproduc- tive season than do older females. Hunter and Leong (1981) estimated that the average female spawns about 20 times per year. Thus the older females must spawn considerably more often than 20 times per year, and probably contribute a much larger fraction of the reproductive output of the population than a proportionate share by weight. This indicates the importance of maintaining older fish in the population and that danger may exist if older fish are overharvested. ACKNOWLEDGMENTS We thank Roderick Leong (Southwest Fisheries Center (SWFO) for providing and maintaining the northern anchovy used in the laboratory study. We thank Kenneth Mais (California Department of Fish and Game) for providing some specimens. Pedro Paloma (SWFC) and Eric Lynn (SWFC) as- sisted in histological classifications. Carol Kim- brell (SWFC) and Susan Picquelle (SWFC) pro- vided valuable assistance in analyzing the data. LITERATURE CITED barr, w. a. 1963. The endocrine control of the sexual cycle in the plaice, Pleuronectes platessa iL.) I. Cyclical changes in the normal ovary. Gen. Comp. Endrocrinol. 3:197-204. BOUAIN, A., AND Y. SlAU. 1983. Observations of the female reproductive cycle and fecundity of three species of groupers iEpmephelus^ from the southeast Tunisian Seashores. Mar. Biol. (Berl.) 73:211-220. BOWERS, A. B., AND F. G. T. HOLLID.W. 1961. Histological changes in the gonad associated with the reproductive cycle of the herring iClupea harengus L.). Dep. Agric. Fish. Scotl.. Mar Res. 5:1-16. BRETSCHNEIDER, L. H., AND J. J. DUYVENE DE WIT. 1947, Sexual endocrinology of non-mammalian verte- brates. Monogr. Prog. Res., Vol. II, Elsevier, N.Y. BYSKOV, A. G. 1978. Follicular atresia. In R. E. Jones (editor), The ver- tebrate ovary: Comparative biology and evolution, p. 533-562. Plenum Press, N.Y. CROSSLAND, J. 1977. Seasonal reproductive cycle of snapper Chrysophrys auratus (Forsten in the Hauraki Gulf N.Z. J. Mar. Freshwater Res. 11:37-60. DE VLAMING, V. L. 1971. The effects of food deprivation and salinity changes on reproductive function in the estuarine gobiid fish, Gil- lichthys mirabilis. Biol. Bull. (Woods Hole 1 141:458-471. 135 FISHERY BULLETIN: VOL. 83, NO. 2 1972. Reproductive cycling in the estuarine gobiifish Gil- lichthys mirabihs. Copeia 1972:278-291. DE VLAMING, V., G. GROSSMAN, AND F. CHAPMAN. 1982. On the use of the gonosomatic index. Comp. Biochem. Physiol. 73A:31-39. FULTON, W. 1898. On the growth and maturation of the ovarian eggs of Teleostean fishes. Annu. Rep. Fish. Board Scotl. 16: 88-124. HOAR, W. S. 1965. Comparative physiology: hormones and reproduc- tion in fishes. Annu. Rev. Physiol. 27:51-70. HTUN-HAN, M. 1978. The reproductive biology of the dab Limanda limanda (L.) in the North Sea: seasonal changes in the ovary J. Fish Biol, 13:351-359. Hunter, J. R., and S. R. Goldberg. 1980. Spawning incidence and batch fecundity in north- em anchovy, Engraulis mordax. Fish. Bull., U.S. 77: 641-652. Hunter, J. R., and r. Leong. 1981. The spawning energetics of female northern an- chovy, Engraulis mordax. Fish. Bull., U.S. 79:215-230. Hunter, J. R., and b. J. Macewicz. 1980. Sexual maturity, batch fecundity, spawning fre- quency, and temporal pattern of spawning for the north- em anchovy, Engraulis mordax, during the 1979 spawn- ing season. Calif. Coop. Oceanic Fish. Invest. Rep. 21:139-149. Lambert, J. G. D. 1970a. The ovary of the guppy, Poecilia reticulata. The atretic follicle, a Corpus atreticum or a Corpus luteum praeovulationis. Z. Zellforsch 107:54-67. 1970b. The ovary of the guppy Poecilia reticulata . The granulosa cells as sites of steroid biosynthesis. Gen. Comp. Endocrinol. 15:464-476. Leong, r. 1971. Induced spawning of the northern anchovy, En- graulis mordax Girard. Fish. Bull., U.S. 69:357-360. ROBE, A. R 1982. Histological observations on the reproductive biol- ogy of the haddock, Melanogrammus aeglefinus (L.). J. Fish Biol. 20:397-408. SAIDAPUR, S. K. 1978. Follicular atresia in the ovaries of non-mammalian vertebrates. Int. Rev Cytol. 54:225-244. Santander, h., and O. S. de Castillo. 1979. El ictioplancton de la costa Peruana. Inst. Mar Peru Bol. 4:69-112. Smith, R E. 1972. The increase in spawning biomass of northern an- chovy, Engraulis mordax. Fish. Bull., U.S. 70:849-874. Tyler, a. v., and r. S. Dunn. 1976. Ration, growth, and measures of somatic and organ condition in relation to meal frequency in winter flounder, Pseudopleuronectes americanus, with hypotheses regard- ing population homeostasis. J. Fish. Res. Board Can. 33:63-75. VROOMAN, A. M., P A. PALOMA, AND J. R. ZWEIFEL. 1981. Electrophretic, morphometric, and merisitic studies of subpopulations of northern anchovy, Engraulis mor- dax. Calif Fish Game 67:39-51. WALLACE, R. A., AND K. SELMAN. 1981. Cellular and dynamic aspects of oocyte growth in teleosts. Am. Zool. 21:325-343. WOOTTON, R. J. 1979. Energy costs of egg production and environmental determinants of fecundity in teleost fishes. Symp, Zool. Soc. Lond. 44:133-159. YAMAMOTO, K. 1956. Studies on the formation of fish eggs. 1. Annual cycle in the development of ovarian eggs in the flounder, Liopsetta obscura. J. Fac. Sci. Hokkaido Univ., Sen 6, Zool. 12:362-373. ZANUY, S. 1977 . Induccion a la puesta y estudio de la ovogenesis en un teleosteo marine: Paracentropristis cabrilla L. Invest. Pesq. 41:337-384. 136 EGG PRODUCTION OF THE CENTRAL STOCK OF NORTHERN ANCHOVY, ENGRAULIS MORDAX, 1951-82 Nancy C. H. Lo' ABSTRACT A model was developed for estimating daily production of eggs of northern anchovy from counts of the total numbers of eggs and size-frequency distribution of larvae. Estimates of egg production using this model were compared with three estimates based on the mortality rates of staged (aged) eggs. The model was used to calculate daily egg production of anchovy for a 24-year time series (1951-82) (data were collected each year from 1951 to 1966 and 1979 to present and every 3 years from 1966 to 1979). Comparisons of this index of stock abundance with ones based on the standing stock of larvae indicate that the present model is a better index of spawning biomass. It was found from the 1979-81 data that the eggs and larvae (< 20 days) have different forms of instantaneous mortality rate (IMR): The larval IMR was age dependent, i.e., zit) = pit for tc < t whereas the egg IMR was constant 2(n = a for t < tc where tc is incubation time or yolk-sac absorption. Based upon this model, the daily egg production, and egg-larval mortality rates for larvae <20 days (<8 mm preserved length), were estimated for 1951-82 from data collected with 1 m ring nets and bongo nets. Egg production varies with stock size proportionally if the reproduction effort remains constant. The egg production is a better index of stock size than the larval abundance because the latter is subject to the inherent egg and larval mortality in addition to reproductive output. Ichthyoplankton data have been used extensively for estimating biomass (or spawning biomass) of marine fish stocks (Murphy 1966; Ahlstrom 1968; Smith 1972). One of the tacit assumptions under- lying most of the methods used for estimating biomass from ichthyoplankton data is that egg or larval mortality is constant among years. In recent years, however, it has become increasingly evident that egg and larval mortality is quite variable among years and among life stages (Ahl- strom 1954; Marr 1956; Colton 1959; Burd and Parnell 1972; Gushing 1973; Fager 1973; Harding and Talbot 1973). As a result, biomass indices based on standing stock of eggs or larvae are subject to a considerable bias if the interannual variability in mortality is not taken into account. In order to eliminate the bias, attempts were made to estimate the spawning biomass by using the egg production and reproduction parameters (Saville 1964; Beverton and Holt 1965; Ciechomski and Capezzani 1973). The basic model is Po = BaRiE/W) where Po = egg production at age zero, Ba = spawning biomass, (1) 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, PO. Box 271, La Jolla, CA 92038. R = E = W = proportion of spawning biomass be- ing female, average batch fecundity, average mature female weight. Equation (1) is adequate for species that spawn only once during a season. But for the multiple spawners, like northern anchovy, Engraulis mor- dax, one needs to include another adult parameter, the proportion of mature spawning female (F), in the equation (Parker 1980). Moreover, Parker chose to use egg production per day, as this could be easily estimated from a single cruise. Thus, the egg production model (EPM) for northern anchovy (or any multiple spawning stock) becomes Po = BaRF{E/W). (2) Manuscript accepted April 1984. FISHERY BULLETIN: VOL. 83, NO. 2, 1985. Staged eggs are used to estimate the daily egg production (number of eggs per day) of the popula- tion (Po) while adult fish are sampled to estimate the number of eggs produced per fish weight (E/W), sex ratio (i?), and proportion of mature spawning female iF). This method is, without doubt, the best of all ichthyoplankton biomass estimation techniques. It is, however, a data rich method requiring both ichthyoplankton and adult sampling plus staging of eggs and various labora- tory measurements which may not be available. In this report, I present an alternative method for 137 FISHERY BULLETIN: VOL. 83, NO. 2 estimating biomass using conventional ichthyo- plankton data rather than the extensive sets of specialized information required by the EPM method. This alternative method provides esti- mates of the daily egg production (Pq) and is referred to as the historical egg production (HEP) to distinguish it from the current EPM. This model for HEP requires only the standing stock of unstaged eggs, and the numbers of larvae in various length classes subsequently transformed into age classes using Gompertz growth curve (Zweifel and Hunter"; Methot and Hewitt^; Lo 1983). Daily egg production varies proportionally with the stock size if the reproduction effort of the population remains constant. The production of eggs by a stock is certainly a better index of stock size than the standing stock of larvae (Smith 1972 ) because both egg and larval mortality rates are considered in the former case. In addition to development of the model, I provide a time series of northern anchovy HEP for 1951-82. This historical record of daily egg produc- tion rather than the EPM i Equation (2) ) was used to estimate anchovy biomass for these past years because data were not available for all the female reproductive parameters until 1980 and none of the eggs have been staged. It would be unpractical and take years to do all the staging of eggs that would be required for all the years. The HEP is an unbiased index for the spawning biomass (Ba ) of the anchovy population for those years if the annual reproductive output per fish weight has remained constant. I do not have sufficient data to validate the assumption of constant reproduc- tive output although 1981-82 data do indicate so. peak spawning season of northern anchovy was usually February- April, daily egg production for the central stock northern anchovy was computed from egg and larval data (CalCOFI'^) collected in January-April within these eight regions. The CalCOFI survey was conducted each year until 1966 after which the survey was conducted every 3 yr. Owing to various improvements in the design of the plankton nets over the past 20 yr (Smith and Richardson 1977; Stauffer and Picquelle 1980*^), different calibration factors were necessary to standardize the catch of eggs and larvae taken in different nets: Aim ring net with 0.55 mm silk mesh was used until 1969 when it was replaced by aim ring net with 0.505 mm nylon mesh; this net was used until 1978 when it was replaced by the bongo net of 0.505 mm nylon mesh. Beginning in 1979, a vertical tow of the 0.333 mm mesh, 25.23 cm diameter CalVET net (CalCOFI vertical egg net) (Hewitt 1983) was used along with the 0.505 mm mesh bongo net to collect egg and larval samples in order to estimate the northern anchovy spawning biomass using the egg production meth- od (EPM) (Parker 1980). In addition to the bias in catch caused by the different mesh sizes, biases also existed due to avoidance of the net, water volume filtered through the net (measured by water flowmeter readings), growth rate of larvae, temperature dependent incubation time (in days), and proportion of larvae from each plankton sample sorted (Zweifel and Smith 1981; Lo 1983). All data (counts of eggs and larvae) were adjusted for the above biases, when it was appropriate, following the procedures outlined by Zweifel and Smith (1981). ASSEMBLY AND BIAS CORRECTION OF EGG AND LARVAL DATA The northern anchovy spawning area lies off central and southern California and Baja Califor- nia. The sampling area was divided into 23 regions covering 17.556 x 10" m" (Fig. 1). The central anchovy stock is enclosed by eight regions (4, 5, 7, 8, 9, 11, 13, and 14) with a total of 5.703 x 10" m^ (Duke 1976^ Huppert et al. 1980). Because the Egg Data The counts of unstaged eggs from each tow were adjusted to a standardized volume of water filtered per unit depth (0.05 m^/1 m depth = 0.05 m^ sea surface area = area sampled by the CalVET net). The adjusted egg counts per 0.05 m^ sea surface area were then stratified by CalCOFI regions. A weighted mean egg count per 0.05 m^ was com- puted as ^Zweifel, J. R., and J. R. Hunter. Unpubl-. manuscr. Tem- perature specific equations for growth and development of an- chovy, Ertf^rciiilis mordax, during embryonic and larval stages. ^Methot, R. D., and R. R Hewitt." 1980. A generalized growth curve for young anchovy larvae, derivation and tabular example. Natl. Mar Fish. Serv, Southwe.st Fish. Cent. Admin. Rep. LJ-80-17, 8 p. •'Duke, S. 1976. CalCOFI station and region specifica- tions. Natl. Mar. Fish. Serv., Southwest Fish. Cent. Admin. Rep, LJ-76-3, 37 p. ^CalCOFI. California Cooperative Ocean Fisheries Investi- gation, a program sponsored by the State of California. The cooperating agencies in the program are California Department of Fish and Game, National Marine Fisheries Service, and Scripps Institution of Oceanography, University of California. "Stauffer, G.D., and S.J. Picquelle. 1980. E.stimates of the 1980 spawning biomass of central subpopulation of northern anchovy. Natl. Mar. Fish, Serv, Southwest Fish. Cent. Admin. Rep. LJ-80-09, 138 LO: EGG PRODUCTION OF NORTHERN ANCHOVY 135° 130° 125"W. / V ^ / y 0 o o o o o o 0 o o oJo O OO f ~ ' ^ iff o o o o o o o o o o /o o o of 35° NORTHERN CALIFORNIA / /^ ,_. ^ O o o ooooooA" « / CAPE MENDOCINO 1 INSHORE 3 x^ ^J ^^ \J ^^ t^ • ' ' 2 C\ t\ f*t f\ C\ /\ 2 NEARSHORE . \ 3 OFFSHORE °°t 7 < 3 O O o o o o o o o o oy 3 O O o o o o o o o o oV - O O O o o o o o \ o o o \ A V /U SAN FRANCISCO o o o 0 o o o o o o o o / if CENTRAL CALIFORNIA { 4 INSHORE o o o 6 o o o o o o o o 5 o o o o o o o o o^ 4 / o o o o\ 5 NEARSHORE 30° ^ 6 OFFSHORE \ o o o o o o o o o o o o\ . o o o o o o o o o o o o^ ( POINT CONCEPTION o o o O o o o o o I o o o V^ \ ciV SOUTHERN CALIFORNIA o o o o o o o o o\ o ^ ^^0] 7 INSHORE o o o o o o o o o o o o o o o j "^ 8 NEARSHORE 10 9 8 '^ A 9 OFFSHORE O O 0 o o o o o o o o o o Q o odL 10 EXTENDED o o o o o o o o o o o\o o o o o o/ \^ / SAN DIEGO 25° o o o O O O 0 o o o o o ^ o o o ■v I o o o o o o o o o o o o o o o oaJ - BAJA CALIFORNIA o o o o o o o o o o o o o o o oJ < II INSHORE II A 12 BAY 13 NEARSHORE 14 OFFSHORE o o o 15 o o o o o o o o 14 o o o o o o o o o ©yo oo/ / ^-^^ oooo/oooo) / \ 15 EXTENDED o o o o o o o o ooooJooooV i 7 12 \ \ . o o o o o o o o O 0 3/^0 0000 0[ \ ^ o o o o o o o o o o D o 0*^*^ coo) L, • SOUTH BAJA o o o o o o o o o o.o o oo^\p^^^^ / \ (EUGENIA / 20° K 16 INSHORE o o 1 o o o o o o o o o\ o o oV. \ 17 NEARSHORE 19 18 17 \ 16 l^ ) 18 OFFSHORE o o o o o o o o o o o o\o o o r u 19 EXTENDED o o O O 0 o o O O 0 o o o\ O O O V i{\ o o O O O 0 o 000000 0)0 o/ / 1 / \r o o o o o o o o o o o o oo o^ ) • CAPE o o O O O O 0 o o o o o .o o^S > 20 INSHORE o o O 0 o o o o o o o o o o\o o o\ L- 21 NEARSHORE 23 22 2! \20V Al 22 OFFSHORE O 0 o o o o o oooooooo\oo\ I 15° 23 EXTENDED o o o o o oooo ooooloo <3L-*-<-^^ / o o o o o oooooooooXooo / / 40° N. 35° 30° 25° 120^ 1 15° 110° Figure l. — Sampling area for estimating northern anchovy spawning biomass with CalCOFI sampling stations denoted by the open circles, and CalCOFI regions denoted by numbers (from Duke text footnote 4). 139 FISHERY BULLETIN: VOL. 83, NO. 2 i. i where xi is the adjusted mean egg count for re- gion i and W, is the relative area weight for region i. Region nmi^ xlO"^ m' xlO-''' Wi 4 18 6.105 0.107 5 29 9.878 0.174 7 20 6.896 0.119 8 12 4.116 0.072 9 29 9.878 0.174 11 9 3.171 0.0538 13 21 7.122 0.126 14 29 9.866 0.174 Total 167 ^57.031 1.00 ^ Sum is not equal to the total due to rounding error. Zero catch was assumed for regions where no samples were taken because historical records show those regions usually had low densities of eggs and larvae. The weighted Xu's were also corrected for extrusion through the mesh by mul- tiplying the catch by the ratio of the catch in a 0.150 mm CalVET net to the catch in the net used in a particular survey (r): r = 3.6 for 0.55 mm mesh silk 1 m ring net 1 1951-68), r = 3.04 for 0.505 mm mesh Nitex' 1 m ring net (1969-76), r = 12.76 for 0.505 mm mesh Nitex bongo net (1978-present) (Lo 1983). The 0.505 mm mesh bongo net seems to catch 4 times that of a 1 m ring net. The reason is unknown. (A field experiment was conducted in April 1983 to reestimate the extrusion rate of anchovy eggs from 0.505 mm mesh bongo net. The data have not been analyzed at the time of writing. Although the egg samples from bongo nets were u.sed to compute the HEP, the bongo net is pri- marily used for catching anchovy larvae, whereas the CalVET net is the egg sampler The discrep- ancy between bongo and 1 m ring net is not of major concern for the current anchovy biomass estimation. ) The standing stock of eggs per 0.05 m^ is then and mt, = Xu- r var (mil ' = var ixw )r + xfr var(r) where mt is the standing stock ofeggs( and larvae) Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. 140 up to age t days from fertilization. Here ti is the duration of incubation. The size of standing stock of eggs depends on not only egg production rate and mortality rate but also the duration of incubation (or the incubation time), which is a function of sea temperature. The average temperature for all positive egg tows (tows which contain one or more anchovy eggs) over January-April in each year was used to estimate incubation time {ti) using the equation (Lo 1983) ti - (18.73 e"°^25^^"^P) where ti = incubation time in days, temp = temperature in degrees centigrade. Both the standing stock of eggs imii') and the incubation time (ti) are essential in computing the time series of daily egg production. The temperature in January- April ranges from 11° to 19° C. The long-term average temperature from January to April is 14.25° C, thus the average incubation time is 3.15 d. Larval Data The anchovy larvae from all years were mea- sured to the nearest 0.5 mm preserved length. For the purpose of estimating mortality rate, larval data were grouped into 2.5 mm, ranging 2-3.0 mm; 3.75 mm, 3.5-4.0 mm; 4.75 mm, 4.5-5.0 mm: ... for larvae < 30 mm. Each preserved length was first converted to a live standard length using a shrink- age formula based on the tow duration (Theilacker 1980), and then converted to age (,t days) using a two-cycle Gompertz growth curve. The first cycle is from hatching to yolk-sac absorption, a temper- ature-dependent growth curve, and the second cycle is from yolk-sac absorption to 22 mm larvae, a food-dependent growth curve (Zweifel and Hunt- er footnote 2; Methot and Hewitt footnote 3; Lo 1983). Larval abundance by length (age) group was estimated using a negative binomial weighted model (Bissell 1972; Zweifel and Smith 1981) which incorporates the "effective sampler size" (relative sampler bias). All larval abundance data were adjusted to conform to the following standard conditions: no extrusion, no day-night difference in avoidance, and a constant water volume filtered per unit depth. These data were converted to daily production (Pt) by dividing the total number of larvae in each length group by the duration (the number of days larvae remain within each length LO: EGG PRODUCTION OF NORTHERN ANCHOVY group). It was necessary to compute a weighted mean of larval production (wPt) because the number of net tows was not proportional to the area size: The daily larval production [Ptj , tj ) was estimated first for each of the three subareas {j - 1: inshore = regions 7 and 11; 7 = 2: nearshore = regions 4, 8, and 13; and j = 3: offshore = regions 5, 9. and 14) (Fig. 1). The data set (u-Pt , t ) was used for final fitting of the mortality curve where w Pt = S Ptj uj , and uj = 0.17, 0.31, and 0.52 for^ = 1, 2, and 3, the relative area sizes. The unweighted average age t over three areas was used because little variation exists among tj's (Fig. 2). DAILY LARVAL PRODUCTION PER 0. (t) 05 m2 (^Pj) AT AGE t, (wP,) 1979 PRESERVED LIVE SIZE AVE, AGE DAILY LARVAL SIZE (mm) (mm) (day) PROD 70.05 m2 2.50 3.26 4 91 0.518 3.75 4.57 8 60 0.121 4 75 5.69 1 1 58 0.0838 5.75 6.27 14 15 0 0665 6.75 7.83 16 4 1 0.0481 7.75 8.87 18.62 0.036 P,= 1.364 (t/3.16)-2-2i' - 0.45 0 30 O 3 a o cc a. -I < > < _i 0.15 - < 0 00 2.0 3.0 4.0 5.0 AGE/3.16 (t/th) 6.0 FIGURE 2.— Weighted daily larval production (wPt) and age in days (t) of northern anchovy and the fitted larval mortality curve based upon Equation (8B) for larvae < 20 d old, 1979. MODEL If a cohort of eggs (larvae) is followed and Nt is defined as the number of eggs (larvae) at age t (days), then the ratio Nt/No measures the survival probability at age t: Sit; zit)) = P(T > t; z (t)). The sample ratio m/no estimates the survival probability Sit) where zit), the in- stantaneous mortality rate (IMR), is defined as lim PU:Si:^l±Al^:>i) If the sample data At^O A^ {nt,t) are taken from a single cohort and the form of Sit) is known, both No and zit) can be esti- mated through nt = no Siit); zit)). Assuming that the standing stock of eggs and larvae repre- sents a single cohort (with stable age distribution) as it ages, then iNt , t ) can be estimated from the number of eggs and larvae in various stages (lengths) which are later converted to age in the sample. Hewitt (1982) conducted a simulation study to check for possible bias in larval mortality rate caused by seasonal changes in the intensity of spawning of northern anchovy which violates the assumption of a stable age distribution. He found that mortality was overestimated in the begin- ning (January-February) of a season when spawn- ing was increasing and underestimated at the end (May-July) when spawning was decreasing. When the larval numbers were accumulated over the entire season, these two biases tended to cancel out. Therefore, the stable age distribution is a reasonable assumption if the egg and larval sam- ple covers the entire season. To compute larval mortality for each year, I chose larval data from January to April to be consistent with the current sampling scheme. According to Hewitt's study, the larval mortality may be overestimated. How- ever, because only young larvae (<8 mm pre- served length) were considered in the model, the upward bias is slight. The number of eggs and larvae at various stages or length classes int, ), as mentioned in a previous section, was further adjusted for the duration in days that eggs ( larvae) remained in a particular stage or length class (d; ), i.e., Pt, = nt.ldi. The quantity Pt, is egg (larval) production per day per unit area (e.g., 0.05 m^ ) at age ti , the average age of eggs (larvae) in the iih stage (length) class (Farris 1960; Saville 1964; Harding and Talbot 1973; Ciechomski and Capez- zani 1973). (In later sections, the subscript / is dropped, thus iPt, t) is used in place of (P^, , ti).) The model is based on the form of the mortality curves of northern anchovy eggs and those for anchovy larvae, the form of the curve for eggs and larvae being distinctly different. The daily egg and larval production Pt is modeled by three survivorship functions Si, S2, and S.3: {PoSiit;ziit)) \ t^tc (3A) Pt = PoSit) = {Ptc S2 it; Z2 it)\ T > tc) 1 tc tk) tk < t (3C) with the IMR 141 FISHERY BULLETIN: VOL. 83, NO. 2 Zit) ={z2it) Zzit) t< tc tctc)-= PiT > t; zit) \T >tc), tc is the age when the form of IMR changes, tk is max ti for ti < 20 d where ti is the average age for the iih. length class. The quality of larval data for larvae older than 20 d is questionable because older larvae avoid the net (Hewitt 1982). The forms ofzz{t) and Szit) for ^ > 20 d are unknown at the present time. Therefore, only the mortality of eggs and larvae < 20 d old was assessed. The IMR, zU), relates to the survivorship func- tion S{t), by definition, in the form of -I z\(u)du (4) -Si(0 t^tc -Jo-(" )du rtc -I zi(u)du •'0 •'tc Z2(u)du tc tc) = S2it). The critical age tc was defined as the age before which z{t) = ziit), after which zit) = Z2it). Pt and t from larval data were used to estimate both Ptc and Z2(t) through Equation (3B) after Sit) is specified. Both larval production iPt^ ) and the standing stocks of eggs and larvae up to age tc imtf) were then used to estimate Po and ziit) through Equation (3A) as below: and mt,= r Pf dt = f^' Po Slit; Ziit)) dt (5A) *'0 •'0 Pt, = PoSiitc;zi{t)). (5B) Now I have two Equations (5 A) and (5B) to be solved simultaneously for the unknowns Po and the parameters in ziit). An iterative procedure was used to obtain estimates of Po and ziit). Clearly, the selection of the function forms of 21 it) and Z2it) are important in obtaining accurate estimates of Po and Ptc ■ Anchovy Mortality Curves and Estimation of Egg Production Daily egg and larval production per 0.05 m^ and their ages iPt , t) were estimated for 1979-81 to model the mortality curves Po Siit) and Pt^ S2it\T > tc) (Equation (3)). The egg data were collected in vertical net tows from 70 m with the Table L — Daily egg and larval production per 0.05 m^ (Pt) at various ages in days ( t ) sampled from Cal VET and bongo tows, and the estimates of five parameters: egg production at age zero (Po *. egg mortality id), larval mortality coefficient (/3 ), larval production at hatching (P/ ) and incubation time in days ( 0 for all t. The relationship between zit) and t determined the function form of 2(0 (Table 2). The zitYs were 19791 T9801 ^ 12.0 H 90 h O Q 6.0 |- O CC ^ 3.0 h < c o u 3 ■D O w Q. >• "(5 "O 0.0 ^ 12.0 9.0 6.0 3.0 Q I Q Q— I 0.0 J] 19811 o , o 12.0 9.0 6.0 3.0 g IQ Q I 0.0 DD . ° °.° ° 3.0 2.0 1.0 0.0 -1.0 0.0 5.0 10.0 15.0 20.0 0.0 5.0 10.0 15.0 20.0 0.0 5.0 10.0 15.0 20.0 AGE (in days) AGE (in days) AGE (in days) ■J - a £ -2.0 3.0 2.0 1.0 0.0 h -1.0 a -2.0 3.0 2.0 1.0 0.0 -1.0 ' o ' -2.0 ■o o o 0.0 5.0 10.0 15.0 20.0 0.0 5.0 10.0 15.0 20.0 0.0 5.0 10.0 15.0 20.0 3.0 2.0 1.0 0.0 -1.0 I- J -2.0 -2.0 -1.0 0.0 1.0 2.0 3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 In (age) In (age) In (age) Figure 3. — Daily egg and larval production of northern anchovy per 0.05 m* (Pt) by age in days (t) and their log transformations (ln(Pt ) ), 1979-81. A linear relationship between ln(Pt ) and t indicates a constant instantaneous mortality rate (IMR) and a curvilinear relationship between ln(Pt ) and t indicates an age dependent IMR. Squares are egg data and open circles are larval data. ^ AGE (in days) c 3.0 r o a o 2.0 - o 3 ■o D D D D o 1.0 - O Q. O >. 0.0 - (0 o 3 -1.0 o o c -2.0 o J 1 1 1 1 3.0 D AGE (in days) 2.0 1.0 - a D D a 3 O O 0.0 - 1.0 -?.o - J 1 1 o o o o 1 1 r- AGE (in days) o a o o o 1 .._I 1 ( 1 1 — 3 CD O 143 FISHERY BULLETIN: VOL. 83, NO. 2 TABLE 2.— The instantaneous mortality rates of anchovy eggs and larvae < 20 days (ziti ) ) by age in days (ti ) computed from the daily egg and larval production estimates (P<, ) and age (<, ), 1980. z(t) = 0.0060 + 1.63/t is the function fitted to the data in the last two columns for f > 4.5 d. Daily egg and larval production ti (d) Pti Pti^ ti = -Pti ti-ti-^ {ti+ti--[)i2 z(tjy 1 2 3 4 5 6 7 8 9 10 11 12 0.67 1.67 2.60 3.57 5.65 5.91 7.69 8.90 11.47 13.83 15.91 17.99 9.28 5.53 3.70 2.37 1.04 0.99 0.86 0.49 0.39 0.26 021 0.15 3.75 1.83 1.33 2.28 0.05 0.13 0.37 0.10 0,13 0.05 0.06 1.00 0.93 0.97 2.08 0.26 1.78 1.21 2.57 2.36 2.08 2.08 1.17 2.14 3.09 4.61 5.78 6.80 8.30 10.19 12.65 14.87 16.95 0.40 0.36 0.37 0.46 0.18 0.07 0.36 0.08 0.14 0.09 0.14 'Z(f,) = (Pti-^ - Pti)l(ti - ti-A)IPti. quite constant for egg and larvae <4.5 d old and decreased thereafter. For t values >4.5 d, the function z{t) = a + hit fit the data best. Based upon the function relationship z{t) = bit (the intercept a is not distinguishable from zero and thus w^as dropped), I have the IMR z{t): zit) = \a ^filt t ^ tc tc0 (9A) ^c ■ Po a = 0 Ptc =Poe ■ate (9B) where mtc is the standing stock of eggs and larvae up to age tc ■ Equation (9A) divided by Equation (9B) results in mf,/Ptc = (e"'"-l)/a = h(a) a>0 tc a = 0 (10) where tc = ti or tys and q is the ratio of standing stock of eggs and larvae up to age tc to the larval production Pt^- The estimated IMR, a, was ob- tained by an iterative procedure using Equation (10). The estimated egg production obtained by rearranging the terms in Equation (9B): Po ^ Ptc-e ate The approximate variance of a and j8 were com- puted in the appendix. TIME SERIES ESTIMATES OF HISTORICAL EGG PRODUCTION (HEP) The HEP per 0.05 m^ (Po ) and the egg IMR (a) for the central stock of northern anchovy in the first 4 mo of the year, 1951-82, were estimated based upon Equations (9B) and (10). For years after 1978, catch data were available for CalVET and bongo nets, but I chose to use samples from 144 LO: EGG PRODUCTION OF NORTHERN ANCHOVY bongo nets because only bongo or similar nets (1 m ring nets) were used for sampling eggs and larvae prior to 1978. Two series of HEP estimates were constructed. Series 1 assumed a constant IMR for the egg stage with tc = ti , whereas series 2 with tc = tys assumed a constant IMR throughout the embry- onic period (Table 3). Both Ptj and Pty, , the daily larval production at hatching and yolk-sac stage, were obtained from the fitted line of Equation (8B) with tc = ti . Under series 1, nearly half of the egg IMR (a) were negative (11 out of 24 yr). This was because the egg IMR depended on the value of q through Equation (10) where q = mtj/Ptj. However, judg- ing from Equation (10), q = ti for a = 0. Therefore for those years where q < ti , egg IMR would be less than 0. The small ^'s could result from the underestimated mt/ or overestimated Ptj or both. The poor results of IMR (a) were likely due to the underestimation of mr: -^-M 1980 Figure 4. — Estimates of historical egg production of the central stock of northern anchovy using the series 2 method (Po ) and the larval abundance (La) of the larval census estimates, 1951-82. Table 4. — Daily egg production per 0.05 m^ (Po), egg instan- taneous mortality (a), egg abimdance (mtj) of northern an- chovy, and number of tows in) in CalCOFI regions 4, 5, 7, 8, 9, 11, 13, and 14, January-April 1980-82. 1980 1981 1982 Po a ^0 a Pq a (SE) (SE) (SE) (SE) (SE) (SE) Historical egg production Series 1 2.33 0.37 3.95 0.36 1.94 0.15 (1.46) (0.21) (2.70) (0.24) (1.70) (0.20) Series 2 2.67 0.36 4.37 0.38 3.29 0.36 (1.46) (0.08) (2.70) (0.08) (1.70) (0.09) Current egg pro- 2.29 0.45 1.82 0.14 1.18 0.15 duction method' (0.51) (0.11) (0.31) (0.08) (0.32) (0.104) 1980 1981 1982 Egg abundance per (SE) (Sl) mti (S^) 0.05 m^imti) n n n CalVET (0.333 mm)^ 961 3.20 (0.52) 1,134 4.72 (0.72) 992 3.48 (0.62) Bongo (0.505 mm) 97 4.12 (1.46) 403 6.88 (2.70) 113 4.93 (1.70) ' Picquelle, see text footnote 8. 2 Mesh size. point estimates of Pq's from the current EPM were lower than those estimated from the two series. The reason for the lower values is unknown at the moment. This could be due to random fluctuation of the statistics. The current EPM estimates of Po were much more precise than those derived from the historical egg-larval mortality model, where- as the precision of egg mortality rate from both methods was similar. As to the estimates from the two series of HEP, the point estimates of Pq from series 2 were always higher than those estimated from series 1. Recall that the assumption of series 2 was that the egg through yolk-sac larval stage suffers a constant mortality rate. However, if in fact the yolk-sac larvae suffer a higher mortality rate than eggs, the mortality rate of eggs and larvae when com- bined (series 2) would overestimate egg mortal- ity as well as egg production (Po) (Equation (9B)). DISCUSSION Historical production (Po) and egg IMR (a) of the central stock of northern anchovy for the first 4 mo of the year from 1951 to 1982 were estimated based upon the information of total number of eggs and yolk-sac larvae per 0.05 m^ and the egg- larval mortality model. Two series of Po and a were produced. Series 1 assumed a constant IMR for only the egg stage whereas series 2 assumed a constant IMR for the entire embryonic period. Both series of Po showed the same trend (Table 3, Fig. 4) with a peak in 1975. The high daily egg 146 LO: EGG PRODUCTION OF NORTHERN ANCHOVY production estimate (Po ) in 1975 was caused by the high standing stock of eggs {mti = 30.06/0.05 m^ per m depth) which was more than 10 times that of other years, and the high egg IMR {a = 0.36) (Table 3). The high daily egg production in 1975 reflects either a high fecundity (high spawning frequency) or a high spawning biomass or some combination of these effects. The present level of egg production is the same as that in the middle 1960's. Both egg IMR (a) and larval IMR coeffi- cient )8, 2(0 = pit, vary from year to year (Fig. 5). In addition to providing a 24-yr time series of HEP for the northern anchovy, two important conclusions can be drawn from this analysis: 1. The form of IMR of eggs (and yolk-sac larvae) is different from that of older larvae (6-20 d). 2. Egg production is a better index of stock abundance than is the standing stock of larvae. Little doubt exists that mortality rates change sometime between the hatching of the eggs and the onset of feeding. Analysis of the daily egg and larval production by age for 1979-81 (Fig. 3) suggested a constant IMR for eggs (or eggs and yolk-sac larvae) and an age-dependent IMR of Pareto form for older larvae {z{t) = fB/t for tc < t < 20 d) (Table 2). The age tc in Equation (3) could be 2.5 r 2.0 - 1.5 I U A V 1 / I; H Larval IMR coefficient (/3) 1950 1960 1970 1980 YEAR Figure 5. — Estimated egg instantaneous mortality rate (EMR) (a) from series 2 method of estimating egg production and the larval mortality coefficient (y3) of the central stock of northern anchovy, 1951-82. considered to mark the end of the critical period after which mortality decreases (Ahlstrom 1954; Marr 1956; Farris 1960; Saville 1964). Series 1 assumed tc = incubation time and series 2 as- sumed tc - average age of yolk-sac larvae. From the existing data, I could not ascertain which assumption was the more likely, but it was evi- dent that larvae at hatching or near first-feeding (yolk absorption) suffer higher mortality than do older larvae. The HEP (Po) is certainly preferable to larval standing stock (larval census estimate - LCE) for use as an index of spawning biomass. Egg produc- tion is related to the spawning biomass through Equation (2), i.e., Po = Ba'C, where the propor- tionality C is the reproductive output (R-F-EIW). If the reproductive output remains constant be- tween years, as shown by 1980-82 anchovy data (Picquelle^), the HEP will be an unbiased index of the spawning biomass. The LEG assumes Ba - K-La where La is the larval abundance and K is a constant proportionality (Smith 1972; Stauffer and Charter 1982) (Table 3, Fig. 4). Thus to provide an unbiased index of biomass, the method requires that not only the reproductive output be constant from year to year but also the egg and larval mortality must remain constant as well. Using Equation (8), the larval abundance (age < 30 d old) can be written as 30 La=J Ptdt 'ti I PoSit;z(t))dt •>ti 'ti = Ba , - atl tl /3 -k w\ where g{a, /3, ti) —' for /3 7^ 1 .e-«'^(ln30- \nti) a is the egg IMR and /3 is the larval mortality coefficient. The larval abundance (La) is proportional to the spawning biomass (Ba) with constant proportionality only if the reproductive output *S. J. Picquelle, Statistician, Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Mont- lake Boulevard E, Seattle, WA 98112, pers. commun. July 1983. 147 FISHERY BULLETIN: VOL. 83, NO. 2 (R-F-E/W) and the egg and larval mortality rates through the function g {a, 13, tj) remain constant from year to year It is clear that from 1951 to 1982 time series (Table 3) that the assumption of constant egg and larval mortality has not been met by the central California anchovy population. The HEP requires constant reproductive output. The validity of this assumption can only be tested with future data. In addition to the ichthyoplankton data, several other indices of anchovy biomass exist: acoustic trawl surveys conducted by California Depart- ment of Fish and Game, aerial survey records from aircraft associated with the fishery, catch-effort analysis (CPUE), and cohort analysis from the catch of the United States and Mexican fishery. In a recent management plan, all of these indices except cohort analysis have been calculated and compared with the time series of egg production presented in this paper (MacCall et al.^). The estimates of egg production covaried with these other indices from year to year and appeared to be the most consistent index of spawning biomass among these indices (Table 5). Selection of the appropriate method for esti- mating biomass depends upon the data availabil- ity and knowledge of the growth of eggs and larvae. If nothing is known of the age of eggs and larvae and no information exists on reproductive parameters, the LCE is the only method available 'MacCall, A. D., R. D. Methot, D. D. Huppert, H. W. Frey, and O. Mathisen. 1983. Northern anchovy second draft revised fishery management plan incorporating DEIS/RIR. Manage. Counc. Pac. Fish. Table 5. — Correlations among various indices of an- chovy spawning biomass. Upper value is correlation coefficient, lower value is number of observations. (Repro- duced from table 4.3-2 of MacCall et al. (text footnote 9). ) Historical egg production 0.458 Larval (23) census 0.807 0.708 Acoustic (8) (7) survey 0.818 0.327 0.659 Aerial (9) (9) (9) index 0.791 0.004 0.512 0.379 Spring (4) (4) (10) (9) CPUE 0.395 0.865 0.290 0.655 0.256 Fall (4) (4) (10) (10) (9) CPUE Spawning biomass index Consistency' Rank Historical egg production 0.654 1 larval census 0.480 5 Acoustic survey 0.606 2 Aerial index 0.583 3 Spnng CPUE 0.388 6 Fall CPUE 0.517 4 'Consistency is average of correlation coefficients. although subject to major biases. If egg and larval age data exist but no data on reproductive param- eters are available, then the HEP is the preferable method. The EPM is the best method; it requires not only knowledge of egg mortality but accurate estimates of adult reproductive parameters as well. In many time series, both growth and abun- dance of eggs and larvae are available but repro- ductive parameters are not. In these cases, the HEP is probably the most accurate means of creating a historic time series of biomass. ACKNOWLEDGMENTS I thank John Hunter, Roger Hewitt, Paul Smith, Rick Methot, Alec MacCall, Jay Barlow, National Marine Fisheries Service; David Farris, San Diego State University; Grace Yang, Uni- versity of Maryland; two referees for reviewing the manuscript and making valuable suggestions. I also thank Barry Finzel, Cynthia Meyer, Carol Miller, and Richard Charter for compiling his- torical egg and larval data files, and Mary DeWitt and Debra BrowTi for typing the manuscript. LITERATURE CITED AHLSTROM, E. H. 1954. Distribution and abundance of egg and larval popu- lations of the Pacific sardine. U.S. Fish Wildl. Serv., Fish. Bull. 56:83-140. 1968. An evaluation of the fishery resources available to California fishermen. In D. Gilbert (editor), The future of the fishing industry of the United States. Univ. Wash. Publ. Fish. New Ser. 4:65-80. BEVERTON, R. J. H., AND S. J. HOLT. 1965. On the dynamics of exploited fish populations. Fish. Invest. Minist. Agric. Fish. Food (G.B.) Ser. II, 19: 1-533. BISSEL, A. F 1972. A negative binomial model with varying element sizes. Biometrika 59:435-441. BURD, A. C, AND W. G. PARNELL. 1972. The relationship between larval abundance and stock in the North Sea herring. Rapp. R-v. Reun. Cons. Perm. int. Explor Mer 164:30-36. CIECHOMSKI, J. D. DE, AND D. A. CAPEZZANI. 1973. Studies on the evaluation of the spawning stocks of the Argentinean anchovy, Engraulis anchoita, on the basis of egg surveys. Rapp. R-v. Reun. Cons. Perm. int. Explor. Mer 164:293-301. COLTON, J. B. 1959. A field observation of mortality of marine fish larvae due to warming. Limnol. Oceanogr. 4:219-222. CUSHING, D. H. 1973. Dependence of recruitment on parent stock. J. Fish. Res. Board Can. 30:1965-1976. Eager, E. W. 1973. Estimation of mortality coefficients from field sam- 148 LO: EGG PRODUCTION OF NORTHERN ANCHOVY pies of zooplankton. Limnol. Oceanogr. 18:297-301. KARRIS, D. A. 1960. The effect of three different types of growth curves on estimates of larval fish survival. J. Cons. Perm. Int. Explor Mer 25:294-306, HARDING, D., AND J. W. TALBOT. 1973. Recent studies on the eggs and larvae of the plaice (Pleuronectes Platressa L.) in the southern Bight. Rapp. P-v. Reun. Cons. Perm. int. Explor. Mer 164:261-269. HEWITT, R. P 1982. Spatial pattern and survival of anchovy larvae: implications of adult reproductive strategy. Ph.D. The- sis, Univ. California, San Diego, 187 p. 1983. Roll, heave and vertical ichthyoplankton tows. Ocean Sci. Eng. 8(1):41-51. HUPPERT, D. D., A. D. MACCALL, G. D. STAUFFER, K. R. PARKER, J. A. MCMILLAN, AND H. W. FREY. 1980. California's northern anchovy fishery: Biological and economic basis for fishery management. U.S. Dep. Commer,, NOAA Tech. Memo., SWFC-TM-NMFS-1, 234 p. JOHNSON, N. L., AND S. KOTZ. 1970. Distributions in statistics: continuous univariate distributions. Houghton Mifflin Co., Boston, 300 p. LO, N. C. H. 1983. Re-examination of three parameters associated with anchovy egg and larval abundance: Temperature depen- dent of incubation time, yolk-sac growth rate and egg larva retention in mesh nets. U.S. Dep. Commer, NOAA Tech. Memo., SWFC-TM-NMFS-31, 32 p. Marr, J. C. 1956. The critical period in the early life history of marine fish. J. Cons. Perm. Int. Explor. Mer 21:160-170. METHOT, R. D. 1981. Growth rates and age distributions of larval and juvenile northern anchovy, Engraulis mordax, with infer- ences on larval survival. Ph.D. Thesis, Univ. California, San Diego, 203 p. Murphy, G. I. 1966. Population biology of the Pacific sardine (Sardinops caerulea ). Proc. Calif Acad. Sci., Ser. 4(34):l-84. Parker, K. 1980. A direct method for estimating northern anchovy, Engraulis mordax, spawning biomass. Fish. Bull., U.S. 78:541-544. Saville, a. 1964. Estimation of the abundance of a fish stock from egg and larval surveys. Rapp. P-v. Reun. Cons. Perm. int. Explor. Mer 155:164-170. SEBER, G. a. F 1973. The estimation of animal abundance and related parameters. Hafner Press, N.Y., 506 p. SMITH, P E. 1972. The increase in spawning biomass of northern an- chovy, Engraulis mordax. Fish. Bull., U.S. 70:849-874. Smith, P E., and S. L. Richardson. 1977. Standard techniques for pelagic fish egg and larva surveys. FAO Fish. Tech. Pap. 174, FIR/T175, 100 p. STAUFFER, G. D., AND R. L. CHARTER. 1982. The northern anchovy spawning biomass for the 1981-82 California fishing season. CalCOFI Rep. 23: 15-19. THEILACKER, G. H. 1980. Changes in body measurements of larval northern anchovy, Engraulis mordax, and other fishes due to handling and preservation. Fish. Bull., U.S. 78:685-692. ZWEIFEL, J. R., AND P E. SMITH. 1981. Estimates of abundance and mortality of larval anchovies (1951-75): application of a new method. Rapp. P-v Reun. Cons. Perm. int. Explor. Mer 178:248-259. APPENDIX The approximate variances of 6c and Pq were derived from the delta method (Seber 1973): var[fixi,..., xi)]= X 1 = 1 _d_ dxi fixi,..., XI ) warixi ) + 2S -^fixi,...,xi)-^f{xi,...,xi)coyixi,xj) iC,Ji^, "■■^^ ■ i^^ Wind Direction N W— I— E Rom Winds at Lindberg Field, San Diego Winds at San Clemente Island [10 m/sec <_) 2 - In Port O o 0 17 16 15 h Temperature Difference, Onn — 30nn, Dana Point 1 1 1 1 1 1 r Mean Temperature, 0 and 5m, Dana Point x ■In Port Mean Water Temperature, Scripps Pier, La Jolla 2 - E I - 0 Swell Height at Scripps _| Pier, La Jollo I I I.I 2 e I 0 IIMIIIIMII Significant Wave Height, Oceanside lliiiiiiiilinilll 29 30 31 ' I 2 3 4 5 6 MARCH APRIL Figure l. — winds, water temperature, temperature gradient, and wave height at Southern California locations during this study. For temperature, dots are Dana Point, x's are La Jolla. Wind direction is the direction from which the wind is blowing. note 3), and these differences were each significant by rank sum test (P < 0.05). This is not what one would expect from simple mixing, in which the nearsurface isotherms should shoal and the deeper isotherms deepen. A. Zooplanktonic Taxa and Community Structure We examined statistically the data on zoo- plankton summarized in the Appendix for answers to several questions concerning temporal changes in the distributions, using the ANOVA or nonparametric tests summarized in Tables 1 and 2. Daytime vertical distributions of many of these taxa off Southern California in late spring and fall are given by Fiedler (1983). As noted in the tables, there are potential ambiguities in the interpreta- tion of even statistically significant results, such as the difficulty in distinguishing diel migration of a zooplanktonic taxon from a diel variation in its capability to avoid capture by the pump. More serious, and applicable to phytoplankton as well as zooplankton, is the impossibility of distinguish- ing between 1) biological changes caused directly by the storm (such as vertical redistribution, changes in behavior, or changes in the balance between birth and death of a taxon) and 2) storm- driven advection into the area of water with planktonic populations differing in abundances or behavior from those present prior to the storm, but neither the original nor the replacement popula- tions having themselves changed in these proper- ties. Advection undoubtedly occurred before, dur- ing, and after the storm; the issue is whether biologically caused changes associated with the storm occurred as well. L Diel Vertical Migrations Based on results from ANOVA classification 1 (Table 1) or nonparametric Test 1 (Table 2), the taxa migrating into the upper 50 m from deeper water at night were the copepodites and adults of Pleuromamma and Metridia. These are real mi- grations, since sampling the water column of the Southern California Bight to greater depths re- veals a change in depth of maximal abundance from below 100 m by day to within the upper 50 m at night (Esterly 1912; Enright 1977; Brooks and Mullin 1983). Euphausiid furcilia were also more abundant at night than by day. Of the remaining zooplanktonic taxa (which did not have significant diel variation in total abun- dance within the upper 50 m), the populations of female, CV, and CIII Calanus, euphausiid calyp- topes, and cyphonautes larvae were centered sig- nificantly higher at night than by day in the water column. Again, these results are consistent with results of sampling to greater depths in nearby waters (Esterly 1912; Enright and Honegger 1977; Mullin 1979; Brooks and Mullin 1983). Other taxa probably belonging to this category of behavior are adult Rhincalanus and Eucalanus (numbers too small for reliability). Curiously, when tem- poral changes are removed from the analysis (ANOVA classification 3 (Table 1) ), female and CV Calanus a-nd SiduXt Rhincalanus , Eucalanus, and Metridia tend to be uniformly distributed in the upper 50 m. 2. Changes Following the Storm Several taxa were significantly different in abundance following the passage of the storm [ANOVA classification 2 (Table 1) or nonparamet- ric Test 3 (Table 2)], and most of these were larval 155 FISHERY BULLETIN: VOL. 83, NO. 2 forms. Acartia, "Paracalanus" , and "other" nau- plii were significantly more abundant in 3-6 April samples than in the prestorm set (Fig. 2), while the abundance of larval fish in our samples de- creased, as did that of adult and copepodid Oithona. An increase in abundance of appen- dicularians was almost significant. From the point of view of a larval or young juvenile fish, there were more items of desirable food {copepod nau- plii and fewer siblings after the storm. No change was detected in those taxa (adult Labidocera and Corycaeus, chaetognaths) likely to be important zooplanktonic predators on larval anchovy. Some of these changes appear to be continua- tions of trends evident before the storm (Fig. 2). However, the fact that larval copepods of several types were more abundant following the storm suggests that the storm directly or indirectly stimulated reproductive activity, though stimula- tion of hatching of benthic eggs (cf. Uye and Fleminger 1976; Landry 1978) or advection of populations from an area of greater fecundity 250 A 200 - /\ 150 - /V Vv 100 ^ Adult and Copepodid 50 — "Paracalanus" 0 1 1 1 1 1 1 1 -| 300 200 100 Nauplior "Paracalanus ruu 600 - A 500 - rv \ 400 - / ^ 300 - 1^ \i 200 -^1 Other 100 - Nauplii n 1 1 1 1 1 1 1 UJ o < z 03 < 500 Ncupliar /\ 400 Acartia (\ \ \ 300 - \ \ 200 - / V 100 \ n 0 V^ 1 1 1 1 1 1 1 250 - \ 200 - \ l\ 150 - \ M 100 - f^ ^ 50 kJ 0 Appendicularians 1 1 1 III 1 2.0 1.5 1.0 Female - Calanus as - 100 1 1 — — n 1 r 29 30 31 3 4 5 6 March — I 1 n 1 29 30 31 3 4 March April Morch Figure 2. — Temporal change in total abundance (m ) of selected zooplanktonic taxa. The vertical line in each panel separates "prestorm" on the left from "poststorm" on the right. "Paracalanus" includes some Clausocalanus spp. 156 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA would give the same result. Given the naupliar life span of a few days, one would expect that, if repro- duction were responsible, the ratio of nauplii to adults would continue to increase for the 3 d fol- lowing the storm (though such a finding would not rule out advection). This is most readily tested for Acartia (nauplii/female) and "Paracalanus" (nauplii/adults and copepodites), since the older stages remained within the 0-50 m water column day and night (ANOVA classification 1 not sig- nificant). The ratio, nauplii/female, for Acartia increased dramatically; indeed, so much so that published values of maximal fecundity (50 eggs/ female per day, Landry 1978; Uye 1981) are barely sufficient over the 5-d period including the storm (1-5 April) to account for the observed ratio on 6 April, even if no death or metamorphosis of nauplii occurred. This is because female Acartia, though unusually abundant immediately after the storm on 3 April, declined dramatically from that time until 6 April (Fig. 2). For "Paracalanus" , which has approximately the same maximal fecundity as, Acartia (Checkley 1980a), the ratio of nauplii to copepodites plus adults was slightly greater on 6 April than during the preceding 2.5 d. Evidence that the capacity of the environment to sustain reproduction of "Paracalanus" increased after the storm is presented in Section C below. Hence, the results are qualitatively consistent with the hypothesis that the storm stimulated reproduction, especially in Acartia; but other explanations cannot be ruled out. It is noteworthy, however, that no "exotic" taxa appeared after the storm. Significance of ANOVA interaction 1x2 indi- cated that for two taxa (euphausiid furcilia and CV Calanus), the tendency to be more abundant in the upper 50 m at night than by day was more pronounced before the storm. Another migratory taxon — cyphonautes larvae — showed both a change in average depth distribution within the upper 50 m and a change in pattern of diel migra- tion (significance in ANOVA interactions 2x3 and 1x2x3). The nighttime distributions of cyphonautes larvae were similar before and after the storm, but the daytime distribution was shifted to shallower water after the storm; adult Corycaeus showed the same (but nonsignificant) tendency, though they did not have a significant diel migration over the whole period (cf., evidence for a reverse migration by this species off Southern California in Fiedler 1983). Pleuromamma was virtually absent from the upper 50 m during the day both before and after the storm, but at night tended to occur shallower within this layer after the storm. Diel migration was not detected in Calanus CII and cm copepodites (ANOVA interaction 1x3 not significant), nor did they change significantly in total abundance after the storm (ANOVA clas- sification 2 not significant). There was, however, a shoaling of the distributions of both stages both day and night after the storm (ANOVA interaction 2x3 significant). Larval fish were also shallower in our samples by day after the storm. The vertical distributions of these three taxa were still stratified after the storm but were shifted with respect to depth. Another potential effect of the storm, which could also result in sig- nificance of ANOVA interaction 2x3, is homogenization of strongly stratified distribu- tions into more nearly uniform ones. Reduction of the temperature gradient (Fig. 1) reinforces this possibility. The reverse process — an increase in stratification — is possible as a result of biological responses to the physical disturbance. Test 4 is a simple way to examine this question, though it is insufficient to detect some possible complex redis- tributions. The results of this test were contrary to expectation; only 2 taxa, Labidocera nauplii and copepodites, had greater ranges of abundance in the water column before the storm, while 13 taxa had greater ranges after the storm. Included in the latter group were Acartia and "other" nauplii, both of which increased in overall abundance after the storm, and all five juvenile copepodid stages of Calanus. Thus, as far as the zooplankton is con- cerned, poststorm stratification was generally more marked than that prestorm. It may be of significance, however, that the two taxa whose prestorm abundances were more strongly stratified than in the poststorm condition were taxa with strong neustonic (nearsurface) affinities (Barnett 1974; Appendix). Inspection of the data (see Appendix) revealed several other taxa which appeared to have dis- tributional changes of the sorts described above, though these were not significant by the criteria used in the statistical tests. This means that other sources of variability in abundance — notably, horizontal patchiness on the scales of a few kilo- meters, or vertical internal motions creating high variability at a fixed depth from day to day as water passed the sampling location — were more important than were those patterns of change the statistical tests were chosen to detect. Finally, we can examine the overall similarities in the zooplanktonic community of the upper 50 m 157 FISHERY BULLETIN: VOL. 83, NO. 2 (as abundances m^^), excluding those taxa which significantly changed in total abundance in this layer from day to night. The weighted-pair den- drogram of Spearman's rank difference correla- tion coefficients (Fig. 3A) shows an imperfect sep- aration into profiles taken before and after the storm, the first poststorm profile (#7) being more like those before the storm. This is evidence against the hypothesis that physical advection of new populations caused all the poststorm differ- ences, though it is also possible that advection caused by the storm affected our site only after a delay. The storm's apparent effect on the thermal gradient (Fig. 1) was also delayed for some time. Even with some of the migrating taxa excluded, there is a partial separation in the dendrogram of nocturnal from diurnal profiles. o UJ a: o (J ui o z < a: 1.0 .95 .9 0 .85 .80 .75 1.0 .9 5 ,90 8 5 .8 0 .75 .7 0 .65 .60 .55 DDNDD DNN NN DD BBBBA AAA BB AA 15 6 3 7 9 10 12 2 4 1113 N Doy or Night A Before or After storrr 8 Profile number A DNDND NDNDD DM 8B8BB AAAAA AA 12 3 4 5 8 9 12 II 13 7 10 N Day or Night B Before or After storm 6 Profile number B. FIGURE 3. — Dendrograms of faunal (A) and floral (B) similarities of the upper 50 m of water off Dana Point, Calif Faunal assemblages are based on 39 taxa, floral assemblages on 126 taxa (not just those listed in Appendix). "Floral" includes protozoans. All coefficients are significant at P ■: 0.001. B. Chlorophyll, Phytoplankton, and Protozoa Because of the mechanisms of feeding used to separate small particles of food from water, there are probably no strict herbivores among the zoo- plankton we studied, i.e., no animals which ingest living phytoplankton without also ingesting other particulate organic matter Nevertheless, we used the distribution of chlorophyll (see Appendix) as the measure of the distribution of food for particle-grazing species; in the euphotic zone of the Southern California Bight, the concentration of chlorophyll is closely correlated with that of particulate organic carbon, with particulate ATP, and (within any one season) with the chlorophyll in particles >5 ^tm (Mullin and Brooks 1976; Eppley et al. 1977; Mullin 1979). We had adequate data to answer Questions 1-4 from Table 2 for chlorophyll ( = "taxon"). We used the phytoplanktonic and protozoan abundances from the physically integrated samples for all 13 profiles (see Methods) to perform Tests 1, 3, and 5 concerning the whole 50 m water column. We re- stricted Questions 2 and 4 to the upper 40 m (since these taxa were rare below this depth) and used data from five diurnal and three nocturnal profiles in answering these questions, since only those profiles were suitable for counting (see Methods). Only one of the nocturnal profiles was poststorm. In order to obtain estimates of "within classifica- tion" variability and still maintain a balanced de- sign, we reduced the ANOVA to a two-way design, retaining "before vs. after storm" and "depth" as classifications. Thus, diurnal and nocturnal sam- ples were considered replicates (there was no evi- dence of diel migration in the phytoplanktonic taxa). We again restricted the analysis to the upper 40 m. Variances of log-transformed data for these taxa were all homogeneous in the four pro- file data set (profiles 5, 6, 9, and 10). Thus we applied the ANOVA to a subset of those profiles suitable for nonparametric tests. The concentration of chlorophyll per m^ did not change from day to night (Ho 1 accepted), nor did the vertical distribution of chlorophyll within the upper 50 m change from day to night (Ho 2 ac- cepted). The median chlorophyll concentration (m~^) was greater after the storm, but not sig- nificantly so by Test 3. Vertical profiles of in vivo fluorescence of chlorophyll and samples of phyto- plankton from the fluorescence maximum layer (cf. Kiefer and Lasker 1975; Cullen et al. 1982) were taken from the second ship working concur- rently at Dana Point. Comparison of the inte- grated fluorescence profiles indicated that this measure of chlorophyll increased significantly after the storm (P 0.01 by a variant of Test 3). Inspection of the data ( see Appendix ) indicated a shoaling of the chlorophyll maximum layer after the storm, and this was significant by a Mann- Whitney U test for differences in depth of occur- 158 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA rence of the median value before vs. after the storm. This tendency was also shown by isotherms (see above). The range of concentrations of chlorophyll in the water column tended to in- crease, meaning that chlorophyll maxima were accentuated after the storm, though hypothesis 4, based on the median profiles, was not rejected (0.01 < P < 0.025). No phytoplanktonic or protozoan taxa we examined migrated dielly into and out of the upper 50 m nor did any taxon migrate dielly within the upper 40 m. The two-way ANOVA of four profiles detected significant decreases in poststorm abun- dances of the diatoms Nitzschia spp., Bacterias- trum spp., Rh. alata, and Rh. fragilissima; all but the last of these decreases were also significant by nonparametric Test 3 applied to the full 13- member set of integrated profiles. This latter test also revealed a significant decrease in poststorm abundance of another diatom, S. costatum. Only the dinoflagellate, Prorocentrum, was more abun- dant after the storm by the ANOVA test. When the data set of 13 integrated profiles was examined by nonparametric Test 3, significant increases were also detected in the poststorm abundance of Lohmanniella (a potential larval fish food) and Ceratium spp. Neither Gymnodinium splendens nor Cochlodinium catenatum (two potential food items for larval anchovy) changed significantly in abundance in samples taken at the depth of the fluorescence maximum layer from the second ves- sel. The large diatom category, Chaetoceros spp., did not change in total abundance, but the species comprising this category changed at the time of the storm; in particular, Ch. constrictus was the dominant member of the genus after the storm, but was not encountered in the prestorm samples. From the ANOVA, no phytoplanktonic taxa had poststorm vertical distributions different from their prestorm ones, when the criterion of P "^ 0.01 was used for significance, and only Rh. alata and Prorocentrum had significant changes as defined by P < 0.05. Hence, as far as we could tell from the four profiles which were usable in the ANOVA, the storm had much less effect in changing the verti- cal distributions of specific phytoplankters (and protozoans) than it did for zooplankton. This con- clusion is, however, suspect (see below). Nonparametric Test 4, for which eight profiles were usable, indicated that the poststorm range of abundances in the upper 40 m was greater than the prestorm range for five of the dinoflagellate taxa, Mesodinium rubrum, and Lohmanniella, while Nitzschia, Rh. fragilissima , and S. costatum had significantly smaller poststorm ranges. These eight profiles strongly suggested poststorm shoal- ing of the vertical distributions of the potential food species, C. catenatum, G. splendens, and Laboea, but the data sets were too small to estab- lish statistical significance at P ^ 0.01. The general changes associated with the storm were therefore decreases in the abundances and in the degree of stratification of some diatoms, and increases in abundances and degree of stratifica- tion of some dinoflagellates and protozoans. How- ever, significant changes in the pattern of stratifi- cation with depth were more difficult to detect because of the reduced data sets, except for the shoaling of the distribution of chlorophyll. The floral composition of the profiles permitted a clear separation into prestorm and poststorm as- semblages, with the exception of the last prestorm profile, which was quite different from the others (Fig. 3B). This result was different from the analy- sis of zooplankton (Fig. 3A), where the first post- storm profile was unexpectedly grouped with pre- storm profiles. Both results, however, indicate that the compositional changes associated with the storm were gradual rather than abrupt. Unlike the faunal assemblages, the floral grouping showed no tendency to separate day from night. The difference in correlation coefficients between dendrograms A and B probably reflects the fact that quite different numbers of taxa were counted, and that samples were counted by different techniques, rather than any fundamental distinc- tion between phytoplanktonic and zooplanktonic assemblages. C Relations Between Zooplankton and Phytoplanktonic Biomass If positive correlations between the abundances of particle-grazing zooplanktonic taxa and chlorophyll existed before the storm, it is reason- able to hypothesize that such correlations would be weaker or nonexistent after the storm due to turbulent disruption of associations. We examined the following taxa of zooplankton in this regard, sometimes combining categories from the Appendix: Naupliar Acartia, naupliar Calanus, naupliar "Paracalanus" , copepodid and adult Acartia, CI-CIV Calanus, CV and female Calanus (nocturnal only), copepodid and adult "Paracalanus" , adult Metridia (nocturnal only), adult Pleuromamma (nocturnal only), and the appendicularians. We grouped data into four sets of profiles: three diurnal, prestorm; three noc- 159 FISHERY BULLETIN: VOL. 83, NO. 2 turnal, prestorm; three diurnal, poststorm (excluding the 6 April profile); and three noctur- nal, poststorm. In order to give each profile w^ithin a set equal weight and to restrict attention to vertical relations, we arranged data from each profile in order of increasing concentration of chlorophyll; next ranked the samples in order of increasing abundance of the taxon of interest; then calculated the Kendall's tau coefficient as a measure of correlation between that taxon and chlorophyll within each profile; and finally calcu- lated the coefficient of concordance between the rearranged ranks of the taxon in the three profiles of a set as a measure of agreement on a common tendency (see Mullin and Brooks 1972). We then defined a persistent relation between a taxon and chlorophyll in one full set of profiles as requiring a significant {P ^ 0.05) concordance between the individual profiles of the set, tau coefficients of all profiles of the same sign (positive or negative), and at least one of the tau coefficients significant (P ^ 0.05). No persistently negative relations were found between any taxon and chlorophyll in any set of profiles. In the diurnal, prestorm set, naupliar Acartia, naupliar Calanus, copepodid and adult Acartia, and appendicularians were all positively related to chlorophyll, and CI-CIV Calanus tended in this direction. These relations all van- ished at night by our criteria, though naupliar Calanus tended to retain a positive association. After the storm, the strength of the diurnal, posi- tive relations of naupliar Acartia, copepodid and adult Acartia, CI-CIV Calanus, and appendicula- rians increased, and naupliar "Paracalanus" also had a positive relation. At night after the storm, all taxa except naupliar "Paracalanus" , CV and adult Calanus, Metridia, and Pleuromamma had positive relations with chlorophyll. Thus, contrary to expectations, after the storm there were more positive relations between these particle-grazing taxa and the concentration of their food, measured as chlorophyll. We reached a similar conclusion for the ciliates, Laboea and Lohmanniella; neither were persis- tently related to the vertical distribution of chlorophyll before the storm, but both were posi- tively related after the storm by our criteria. Since fewer profiles for these protozoans were counted, we did not separate night from day in searching for the correlations. Such correlations can also show seasonal vari- ability; for example, Fiedler (1983) found strongly positive correlations between the vertical dis- tributions of chlorophyll, Paracalanus, and Penilia avirostris (a cladoceran) in October, but strongly negative correlations between these zoo- plankters and chlorophyll in May; Ctenocalanus vanus showed a seasonal reversal of its relation to chlorophyll in the opposite direction. In spite of the increased correlation after the storm between particle-grazers and their food, there is some evidence that the poststorm grazing pressure on phytoplankton was less than that pre- storm. The ratio of chlorophyll to phaeopigments in the water column is an indicator of the ratio of living phytoplankton to the fecal material of graz- ers, and hence is inversely related to the grazing pressure per unit phytoplanktonic crop (Lorenzen 1967). The chlorophyll/phaeopigment ratio was significantly greater (P < 0.05 by rank sum test) after the storm, indicating a reduction in grazing relative to the available crop. We derived a second indicator of the effect of the storm on relations between phytoplankton and zooplankton from a study of egg production of the copepod, Paracalanus parvus, and chlorophyll and particulate nitrogen in the Southern Califor- nia Bight (Checkley 1980b). Checkley found that the nitrogen in phytoplankton was the best mea- sure of fecundity-stimulating food, that about half the chlorophyll retained on a fiberglass filter was in particles >5 pim, and that the weight ratio of nitrogen in phytoplankton to chlorophyll was 12. From these relations, the egg production of Paracalanus is food-limited where the concentra- tion of total chlorophyll is below 1.3 /xg/1. By this standard, only 18% of the upper 50 m contained sufficient food for maximal egg production prior to the storm, while 34% of the water column met this criterion afterwards. This conclusion is likely to be qualitatively cor- rect unless the size distribution of phytoplankton was altered markedly by the storm, or the breadth of the copepods' diet with respect to nonphyto- plankton was changed. Neither of these sources of error is particularly likely, since the ratio of >5 /i,m to total chlorophyll agrees with earlier results in the Bight (Mullin and Brooks 1976) and since the range of the data from which Checkley de- duced the importance of chlorophyll in regulating egg production included all but one of the concen- trations of chlorophyll we measured. Further, the vertical distribution of adult and copepodid "Paracalanus" was positively corre- lated with that of chlorophyll after the storm and at night (see above). If this finding applies to female "Paracalanus" by themselves, a consider- 160 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA ably greater fraction of total reproduction oc- curred at maximal (i.e., nonfood-limited) rates after the storm. A similar quantitative example of augmenta- tion of zooplanktonic nutrition related to the storm can be calculated for CIV-adult Calanus, though the vertical distribution of these stages was not well correlated with that of chlorophyll. In June 1980, Cox et al. (1983) estimated the carbon budget of Calanus at various stations and depths in the Southern California Bight, and concluded that gain in biomass of these copepods was possi- ble where the concentration of chlorophyll ex- ceeded 0.9 Mg/1- By this standard, the fraction of the upper 50 m where some gi-owth was possible (nighttime only, because of diel migration) was 36% before and 58% after the storm. A third test of the significance of vertical dis- tributions and the effect of the storm on them was based on the plant pigments in the guts of the large copepods caught at various times and depths. The measurement of fluorescence of gut contents can be used as a quantitative estimate of the rate of ingestion of plant material if the break- down of pigment, the gut passage time, and the background fluorescence due to an animal's own pigmentation are known (Mackas and Bohrer 1976). We chose to ask two simpler questions based on changes in fluorescence: 1) Were the total gut pigments (chlorophyll -I- phaeopigments) of copepods caught at specific depths correlated with the concentration of chlorophyll measured at the same depths, before or after the storm or both? 2) Did the amount of gut fluorescence of a tax on, independent of specific depths, change coincident with the storm? The first question addresses the issue of whether the copepods can be shown to have fuller guts at depths where phytoplanktonic food (as measured by chlorophyll) is more concen- trated. If copepods move frequently from the depths at which they feed, such correlations would be difficult to establish (cf. Dagg and Wjmian 1983). The second question is the more general one of whether the copepods were better nourished after the storm. We tested data concerning female Acartia, female and CV Calanus, female Metridia, and female Pleuromamma in this regard, with 6-28 pre- or poststorm data points per taxon. Of these taxa, only Acartia's abundance was significantly positively associated with the vertical distribution of chlorophyll (see above). The gut pigment per Acartia showed no relation to the ambient concentration of chlorophyll, how- ever, while that of Pleuromamma was positively correlated with chlorophyll. In no case was the poststorm correlation (tau coefficient) between gut fluorescence and chlorophyll stronger than that prestorm. Hence, we could not show that for these taxa the distribution of degree of satiety became more strongly associated with the vertical dis- tribution of chlorophyll after the storm, even though the range of chlorophyll concentrations available in the upper 50 m had increased. Nor for any of these taxa was the poststorm amount of gut fluorescence significantly greater than that prestorm. Based on comparison between field-caught female Acartia and Calanus, and these same taxa fed to excess or starved in the laboratory, we conclude that both these popula- tions were well fed in general both before and after the storm, and animals had plant food in their guts at all depths sampled. Hence, we could not demon- strate a change in nutritional status of the taxa as a result of the storm, even though the overall con- centration of chlorophyll increased. All these taxa have been shown to feed on nauplii as well as phytoplankton (e.g., Haq 1967; Lonsdale et al. 1979; Landry 1981), but we could not test whether their nutrition from animal sources had improved coincident with the increase in abundance of nauplii following the storm. D. Abundance and Vertical Distribution of Food for Larval Fish Because larval fish are visual predators, it is the diurnal distributions of potential prey which are particularly relevant. Different species select (or are physically able to ingest) different prey, and of course different types of prey differ in their catch- ability, digestibility, and nutritive value. We will consider the distributions of food for two prototyp- ical larvae representing extremes in a continuum of actual types. One is a small-mouthed larva which we v^dll call "anchovy-like", based on Bemer (1959), Lasker et al. (1970), O'Connell and Raymond (1970), Arthur (1976), and Lasker and Zweifel (1978). For these larvae, "large" prey con- sists of all copepod nauplii and lamellibranch and cyphonautes larvae (Appendix); "small" prey con- sists of all ciliates and all nonthecate, large di- noflagellates. Laboratory studies suggest the crit- ical concentrations for both good survival and rapid grov^rth are ^ 5 x 10^ large or s: 5 x 10^ small prey 1~^ , or an equivalent combination. The other prototypical larva has a larger mouth and is more active; based on Arthur (1976), Hunter 161 FISHERY BULLETIN: VOL. 83, NO. 2 and Kimbrell (1980), Lipskaya (1982\ and De- vonald (1983), this larva is "mackerel-like" (though Scomber and Trachurus, especially the latter, tend to spawn farther offshore than our sampling location). This larva requires a much lower concentration of "large" prey, 50 1 \ and a large number of zooplanktonic taxa are potential food: all copepod nauplii; lamellibranch and cyphonautes larvae; Acartia, Labidocera, Met- ridia, and Pleuromamma immature copepodites; "Paracalanus" , Oithona, Euterpina, Corycaeus, Oncaea, and Microsetella copepodites and adults; euphausiid nauplii and calyptopes; CI, CII, and CIII Calanus; "other copepods"; cladocerans; and (see Lipskaya 1982) appendicularians. Nauplii and lamellibranch and cyphonautes larvae are considered small prey, the remainder being large. This spectrum of prey is also appropriate for young postlarval anchovy. Figure 4 shows the prestorm and poststorm diurnal vertical distributions of food for the two prototypical larval types, in terms of the equiva- lent "large prey" for each; the figure legend gives the conversion factors used. In no instance was the laboratory-determined critical concentration of prey exceeded. We do not believe that this conclu- sion is due to destruction of prey during preserva- tion. We tested hypotheses concerning the vertical stratification and the effect of the storm on dis- tributions of prey by two-way ANOVAs on log- transformed abundances from the diurnal profiles (3, 5, 7, and 9) similar to those used for phyto- planktonic taxa (Section B above), since variances were homogeneous by Barlett's test. We used our data on the diurnal abundances of total larval fish to examine correlations with the food of "anchovy-like" larvae by means of the tau coeffi- cient for these profiles. It is apparent from Figure 4A that "small prey" dominated the food supply for "anchovy-like" lar- vae, even when expressed as its equivalence in terms of large prey. Because this category had not increased significantly after the storm, neither had total prey for these larvae; however, large prey were both more abundant and more strongly stratified. The food supply of "mackerel-type" larvae was E I- CL UJ Q EQUIVALENT LARGE PREY, lOV liter 1 EQUIVALENT LARGE PREY/ liter 20 \ 30 I 40 I 50 I B 30 I 40 I Large prey Small prey 50 _i 60 I Large prey Small prey PRE -STORM POST- STORM Figure 4. — Median vertical, diurnal distributions of larval fish food, as "equivalent large prey", before and after the storm. Taxa comprising categories of prey are li.sted in text. A. Prey of "anchovy-like" larvae. Graphed concentrations of small prey are 0.1 x actual concentrations. B. Prey of "mackerel-like" larvae. Graphed concentrations of small prey are 0.2 x actual concentrations. 162 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA dominated by "large" rather than "small" prey. That these types of zooplankters are less abundant very near surface and below 30 m than at inter- mediate depths is apparently not unusual in spring ( Fiedler 1983:fig. 5). Both types of prey were more abundant after the storm. Total food for both types of larvae tended to be concentrated nearer the surface in the poststorm condition. Although the numbers of larval fish in our sam- ples are too small, especially after the storm, to provide a strong test of spatial correlation with their food supplies, the tau coefficients of correla- tion between total larval fish and their food by day were positive in all cases, but somewhat less so after the storm for the nonthecate dinoflagellates and ciliates which dominated the food supply of "anchovy-like" larvae. This was the case even though the poststorm distributions of both larval fish and food were concentrated nearer the surface than were the prestorm distributions. SUMMARY AND DISCUSSION We concentrated during this study on the con- sequences of the vertical distributions of plankton for the production of zooplanktonic food for larval fish, and on the differences in distributions of food experienced by larval fish at one coastal location before and after a small storm. Since advection surely occurred, we do not intend to imply that the same individual larvae experienced both sets of conditions. Conclusions we believe to be ecologically sig- nificant and statistically verified are as follows: 1. The biotic environment was vertically struc- tured. a. Of the 28 zooplanktonic taxa for which the ANOVA was appropriate, 22 had a consis- tently uneven (i.e., stratified) pattern of dis- tribution with depth in the upper 50 m. Of the remaining six taxa, three had regular temporal changes in vertical distribution. Hence, only 3 of the 28 taxa were uniformly distributed both vertically and dielly. b. Chlorophyll was stratified in the upper 50 m, and 9 of the 18 phytoplanktonic- protozoan taxa examined had stratified dis- tributions in the upper 40 m; the stratified taxa were notably dinoflagellates and the oligotrich ciliates rather than the dia- toms. c. Both small prey and total prey for "anchovy-type" larvae were vertically stratified, but prey for "mackerel-type" lar- vae was not. d. Though the abundance of Acartia was cor- related vertically with that of chlorophyll, its gut fullness was not. 2. Several features were different after the storm. a. Several zooplanktonic taxa — notably, vari- ous nauplii — were more abundant, while larval fish were less so. Ceratium, Prorocen- trum, and Lohmannlella had increased, while several diatoms had decreased. Evi- dence suggested a poststorm increase in chlorophyll, but contained ambiguities. b. Several zooplanktonic taxa — Pleuro- mamma at night, cyphonautes by day, Calanus CII and CIII, larval fish — tended to be concentrated in shallower depths after the storm, as did chlorophyll, but data were insufficient to show that the large-sized phytoplanktonic taxa we studied responded in this way. Food for both types of larval fish was concentrated in shallower water after the storm. The neustonic distribution of Labidocera nauplii and copepodites was less pronounced after the storm, but in gen- eral the poststorm vertical stratification was at least as great as that prestorm, even though the temperature gradient was les- sened. This general conclusion was also true for phji^oplankton (except for some diatoms which were less abundant after the storm) and for the sum of forms represent- ing "large food" for "anchovy-like" larvae and "small food" for "mackerel-like" larvae. 3. Relations between predators and prey were dif- ferent following the storm. a. Several taxa maintained or established abundant populations in those parts of the water column where food was most plenti- ful. However, this was not true for taxa with pronounced diel vertical migrations. The estimated poststorm reproduction of "Paracalanus" was less limited by food than was the prestorm reproduction, and Calanus could obtain sufficient food for growth in a greater fraction of the water column after the storm; but we could not demonstrate a poststorm increase in gut fullness of large herbivores. b. Larval fish, both those categorized as "anchovy-like" and "mackerel-like", also were exposed to augmented concentrations of their respective "large" food items, both immediately and perhaps as a result of en- 163 FISHERY BULLETIN: VOL. 83, NO. 2 hanced growth and reproduction of zoo- plankton and reduced competition from other larvae. However, the supply of food for larvae was less than that thought neces- sary for rapid growth and high survival, and the spatial association between total larvae and abundant, small food (di- noflagellates and protozoans) was slightly less strong after the storm; this category of food was not significantly more abundant after the storm. Lacking information on the planktonic stocks and their distribution, we might have hy- pothesized that the decrease in abundance of lar- val fish following the storm (Fig. 2) was due to starvation because the storm-induced turbulence homogenized the vertical distributions of food. The results shown in Figure 4 make this hypothesis untenable. Even though we did not find concentrations of food exceeding laboratory-determined thresholds for growth, certainly the most important conclu- sion with respect to the storm from the point of view of a larval fish is that there was as much food available after the storm and that copepod nauplii (which laboratory studies have shown to be desir- able prey) increased significantly. In view of this, we predict that the larvae present after the storm were growing faster (or starving more slowly), were in better condition, and were more likely to have food in their guts than those present before the storm, even though the latter were the more numerous. Also, since the available food increased at several depths in the water column, we predict that the occurrences of well-nourished anchovy larvae (if any were present) should be shallower after the storm and less strictly confined to one or two depth strata. A tendency for larvae to be less closely as- sociated after the storm with layers of abundant dinoflagellates and ciliates might negate this pre- diction; the nature of the vertical relations should now be examined using the more reliable distribu- tions of larvae determined by a towed opening- closing net. Another condition which would result in failure of our prediction is if the larvae actually rely for nutrition on micropatches of food, such as organic aggregates and an associated assemblage of phytoplankton and microzooplankton (e.g., Alldredge 1976; Silver et al. 1978). Devonald (1983) has suggested this for larvae of jack mack- erel, Trachurus symmetricus, farther offshore in the Southern California Bight. If this is true, sam- pling on the scale of hundreds of liters, as we did, would not detect the redistribution of food on the scale most important for larval survival and growth; storm-induced turbulence could have dis- rupted such micropatches, making the supply of food less rather than more favorable. A large amount of true microscale sampling, such as that done by Owen (1981), would then be required to predict correctly the effect of the storm on the larvae. ACKNOWLEDGMENTS We thank R. Lasker, G. Moser, and R. Owen of the National Marine Fisheries Service for collab- oration in this project. J. Star and P. Peterson assisted with sampling, as did D. Carlson, owner and operator of the Fisherette. D. Cayan and R. Seymour supplied some of the wind and wave data. This long after the fact, we thank Neptune for the storm. E. Venrick made helpful comments on the manuscript (especially in its statistical aspects), and D. Osborn typed it several times. Financial support was from the Department of Energy, DE- AT03-82-ER60031, and ship funds from the Marine Life Research Group. LITERATURE CITED ALLDREDGE, A. L. 1976. Discarded appendicularian houses as sources of food, surface habitats, and particulate organic matter in planktonic environments. Limnol. Oceanogr. 21:14-23. ARTHUR, D. K. 1976. Food and feeding of larvae of three fishes occurring in the California Current, Sardinops sagax, Engraulis mordax, and Trachurus aymmetricus. Fish. Bull., U.S. 74:517-530. BARNETT, A.M. 1974. The feeding ecology of an omnivorous neritic copepod, Lahidocera trispmosa Esterly. Ph.D. Thesis, Univ. California — San Diego, La Jolla, 215 p. BERNER, L., JR. 1959. The food of the larvae of the northern anchovy En- graulis mordax. Bull. Inter-Am. Trop. Tuna Comm. 4:1- 22. Brooks, e. r., and m. m. Mullin. 1983. Diel changes in the vertical distribution of biomass and species in the Southern California Bight. CalCOFI Rep. 24:210-215. CHECKLEY, D. M., JR. 1980a. The egg production of a marine planktonic copepod in relation to its food supply: Laboratory studies. Lim- nol. Oceanogr. 25:430-446. 1980b. Food limitation of egg production by a marine, planktonic copepod in the sea off Southern Califor- nia. Limnol. Oceanogr. 25:991-998. Cox, J. L., S. WILLASON, AND L. HARDING. 1983. Consequences of distributional heterogeneity of 164 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA Calanus pacificus grazing. Bull. Mar. Sci. 33:213-226. CULLEN, J. J.. AND R. W. EPPLEY. 1981. Chlorophyll maximum layers of the Southern California Bight and possible mechanisms of their forma- tion and maintenance. Oceanol. Acta 4:23-32. CULLEN, J. J., F. M. H. REID. AND E. STEWART. 1982. Phytoplankton in the surface and chlorophyll maximum off Southern California in August, 1978. J. Plank. Res. 4:665-694. DAGG, M. J., AND K. D. WYMAN. 1983. Natural ingestion rates of the copepods Neocalanus plumchrus and N. cristatus calculated from gut con- tents. Mar. Ecol. Prog. Ser. 13:37-46. DEVONALD, E K. 1983. Evaluation of the feeding success of jack mackerel larvae off Southern California and some contributing fac- tors. Ph.D. Thesis, Univ. California — San Diego, La Jolla, 209 p. DIXON, W. J., AND R J. MASSEY, JR. 1957. Introduction to statistical analysis. 2d ed. McGraw-Hill, N.Y., 488 p. DORMAN, C. E. 1982. Winds between San Diego and San Clemente Is- land. J. Geophys. Res. 8:9636-9646. ENRIGHT, J, T. 1977. Copepods in a hurry: Sustained high-speed upward migration. Limnol. Oceanogr. 22:118-125. ENRIGHT, J. T, AND H.-W. HONEGGER. 1977. Diurnal vertical migration: Adaptive significance and timing Part 2. Test of the model: Details of tim- ing. Limnol. Oceanogr. 22:873-886. EPPLEY, R. W., W. G. HARRISON, S. W. CHISHOLM. AND E. STEWART. 1977. Particulate organic matter in surface waters off Southern California and its relationship to phyto- plankton J. Mar. Res. 35:671-696. ESTERLY, CO. 1912. The occurrence and vertical distribution of the Copepoda of the San Diego Region with particular refer- ence to nineteen species. Univ. Calif Publ. Zool. 9:253- 340. Fiedler, R C. 1982. Zooplankton avoidance and reduced grazing re- sponses to Gymnodinium splendens (Dinophy- ceae). Limnol. Oceangr. 27:961-965. 1983. Fine-scale spatial patterns in coastal epiplankton off Southern California. J. Plank. Res. 5:865-877. Gruber, d., e. h. ahlstrom, and M. M. MULLIN. 1982. Distribution of ichthyoplankton in the Southern California Bight. CalCOFI Rep. 23:172-179. HAQ, S. M. 1967. Nutritional physiology of Metridia lucens and M. longa from the Gulf of Maine. Limnol. Oceanogr. 12:40- 5L Holm-Hansen, O., C. J. lorenzen, r. w. holmes, and J. D. H. Strickland. 1965. Fluorometric determination of chlorophyll. J. Cons. Perm. Int. Explor Mer 30:3-15. Hunter, j. r. 1972. Swimming and feeding behavior of larval anchovy Engraulis mordax. Fish. Bull., U.S. 70:821-838. 1976. Culture and growth of northern anchovy, Engraulis mordax, larvae. Fish. Bull., U.S. 74:81-88. Hunter, j. r., and C. A. Kimbrell. 1980. Early life history of Pacific mackerel. Scomber japonicus. Fish. Bull., U.S. 78:89-101. Huntley, m. e. 1982. Yellow water in La Jolla Bay, California, July 1980. II. Suppression of zooplankton grazing. J. Exp. Mar. Biol. Ecol. 63:81-91. KIEFER, D. A., and R. LASKER. 1975. Two blooms of Gymnodinium splendens, an unar- mored dinoflagellate. Fish. Bull., U.S. 73:675-678. LANDRY, M. R. 1978. Population dynamics and production of a planktonic marine copepod, Acartia clausii, in a small temperate lagoon on San Juan Island, Washington. Int. Rev. Gesamten Hydrobiol. 63:77-119. 1981. Switching between herbivory and carnivory by the planktonic marine copepod Calanus pacificus. Mar. Biol. (Berl.) 65:77-82. LASKER, R. 1975. Field criteria for survival of anchovy larvae: The relation between inshore chlorophyll maximum layers and successful first feeding. Fish. Bull., U.S. 73:453- 462, 1981. Factors contributing to variable recruitment of the northern anchovy (Engraulis mordax) in the California Current: Contrasting years 1975 through 1978. Rapp. P-V, Reun. Cons. Int. Explor. Mer 178:375-388. LASKER, R,, H. M. FEDER, G. H. THEILACKER, AND R. C. MAY. 1970. Feeding, growth, and survival of Engraulis mordax larvae reared in the laboratory. Mar Biol. (Berl.) 5:345- 353. LASKER, R., J. PELAEZ, AND R. M. LAURS. 1981. The use of satellite infrared imagery for describing ocean processes in relation to spawning of the northern anchovy {Engraulis mordax). Remote Sensing Environ. 11:439-453. LASKER, R., AND J. R. ZWEIFEL. 1978. Growth and survival of first-feeding northern an- chovy larvae (Engraulis mordax) in patches containing different proportions of large and small prey. In J. H. Steele (editor). Spatial pattern in plankton communities, p. 329-354. Plenum Press, N.Y. LIPSKAYA, N. YA. 1982. [The feeding of larvae of the chub mackerel, Scomber japonicus (Scombridae), from the southeastern Pacific] Vopr Ikhtiol. 22:97-104. LONSDALE, D. J., D. R. HEINLE, AND C. SIEGFRIED. 1979. Carnivorous feeding behavior of the adult calanoid copepod Acartia tonsa Dana. J. Exp. Mar. Biol. Ecol. 36:235-248. LORENZEN, C.J. 1967. Vertical distribution of chlorophyll and phaeopig- ments: Baja California. Deep-Sea Res. 14:735-745. MACKAS, D., AND R. BOHRER. 1976. Fluorescence analysis of zooplankton gut contents and an investigation of diel feeding patterns. J. Exp. Mar. Biol. Ecol. 25:77-85. METHOT, R. D., JR., AND D. KRAMER. 1979. Growth of northern anchovy, Engraulis mordax, lar- vae m the sea. Fish. Bull., U.S. 77:413-423. MULLIN, M. M. 1979. Longshore variation in the distribution of plankton in the Southern California Bight, CalCOFI Rep. 20:120-124. MULLIN, M. M., AND E. R. BROOKS. 1970. The effect of concentration of food on body weight, cumulative ingestion, and rate of growth of the marine copepod Calanus helgolandicus . Limnol. Oceanogr. 15:748-755. 165 FISHERY BULLETIN: VOL. 83, NO. 2 1972. The vertical distribution of juvenile Calanus (Copepoda) and phytoplankton within the upper 50m of water off La Jolla, California. In A. Y. Takenouti (editor). Biological oceanography of the northern North Pacific Ocean, p. 347-354. Idemitsu Shoten, Tokyo. 1976. Some consequences of distributional heterogeneity of phytoplankton and zooplankton. Limnol. Oceanogr. 21:784-796. O'CONNELL, C. P. 1980. Percentage of starving northern anchovy, Engraulis mordax, larvae in the sea as estimated by histological methods. Fish Bull., U.S. 78:475-489. O'CONNELL, C. P, AND L. P RAYMOND. 1970. The effect of food density on survival and growth of early post yolk-sac larvae of the northern anchoyy ^En- graulis mordax Girard) in the laboratory. J. Exp. Mar Biol. Ecol. 5:187-197. OWEN, R. W. 1981. Microscale plankton patchiness in the larval an- chovy environment. Rapp. R-V. Reun. Cons. Int. Explor. Mer 178:364-368. PaffenhOfer, G.-A. 1976. Feeding, growth, and food conversion of the marine planktonic copepod Calanus helgolandicus . Limnol. Oceanogr. 21:39-50. SILVER, M. W., A. L. SHANKS, AND J. D. TRENT. 1978. Marine snow: Microplankton habitat and source of small-scale patchiness in pelagic populations. Science (Wash., D.C.) 201:371-373. TATE, M. W, AND R. C. CLELLAND. 1957. Nonparametric and shortcut statistics in the social, biological, and medical sciences. Interstate Printers and Publishers, Inc., Danville. 111., 171 p. UYE, S. 1981. Fecundity studies of neritic-calanoid copepods Acar- tia clnusi Giesbrecht and A. sfeucri Smirnov: A simple empirical model of daily egg production. J. Exp. Man Biol. Ecol. 50:255-271. UYE, S., AND A. FLEMINGER. 1976. Effects of various environmental factors on egg de- velopment of several species of Acartia in Southern California. Mar. Biol. (Berl.) 38:253-262. 166 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA APPENDIX Vertical Distribution of Taxa APPENDIX TABLE 1. — Diurnal profiles before storm. * = variances heterogeneous; ANOVA not used. Depth (m): 0 5 10 15 20 25 30 35 40 45 50 ZOOPLANKTON ■-median number per cuoic meter- 5,074 1,345 448 Naupliar Acartia 75 407 786 1,533 78 0 39 49 Naupliar Labidocera 1 1 ,688 325 75 77 70 67 34 0 0 0 0 "Naupliar Paracalanus'"' 2,397 3,004 2,000 1,529 1,761 2,941 8,679 3,333 1.418 1,023 1,478 "Naupliar Calanus 693 956 597 536 515 1,070 299 107 75 79 33 Naupliar Rhincaianus 0 0 0 77 7 252 163 71 7 4 0 "Other nauplii 1,653 1,832 3,283 4,138 7,647 2,809 5,000 2,679 3.060 1.732 1,569 Female Acartia 4 22 23 19 21 8 4 0 0 0 0 Male Acaitia 4 11 19 23 11 0 0 0 0 0 0 Copepodite Acartia 12 18 302 284 345 147 0 0 0 0 0 "Adult Labidocera 0 0 0 0 0 0 0 0 0 0 0 "Copepodite Labidocera 1,571 234 4 0 0 0 0 0 0 0 0 Adult and copepodite "Paracalanus"' 3,117 2,711 2,239 1,456 1,029 840 3,051 4,783 3,694 2.598 1,814 "Appendiculanans 285 1,245 1,214 3,218 1,985 4.622 2,164 856 67 24 25 "Adult and copepodite Oithona 0 293 970 1,116 2,746 1,070 1,661 1,739 521 736 1,100 Adult and copepodite Euterpina 9 0 71 284 662 602 293 107 0 0 0 Euphausiid nauplii 0 0 11 0 4 0 3 11 16 4 13 'Euphausiid calypiopis 0 0 4 17 14 30 4 23 7 8 5 Euphausiid furcilia 0 0 4 0 4 20 7 31 8 12 14 Chaetognalhs 11 163 204 307 121 172 97 43 16 35 31 Female Calanus 0 0 0 0 0 0 0 0 0 0 0 Male Calanus 0 0 0 0 0 0 0 0 0 0 0 0 V Calanus 0 0 0 0 0 0 0 0 0 3 0 C IV Calanus 0 0 0 0 0 4 4 4 8 0 14 C III Calanus 0 0 0 4 0 3 8 14 15 4 5 C II Calanus 0 4 0 4 7 8 57 21 0 3 0 C 1 Calanus 0 8 4 0 13 13 57 18 4 0 0 "Adult Corycaeus 15 33 79 87 92 97 157 74 92 67 64 "Adult and copepodite Oncaea 83 73 38 230 70 168 2,463 2,536 1,679 1,299 1,225 Microsetella 0 4 4 843 1,513 1,271 305 286 65 173 196 Adull Melndia 0 0 0 0 0 0 0 0 0 0 0 "Adult Pleuromamma 0 0 0 0 0 0 0 0 0 0 0 "Copepodite Metndia and Pleuromamma 0 0 0 0 0 0 0 12 22 91 163 "Adult Rhincaianus 0 0 0 0 0 0 0 0 0 0 0 Copepodite Rhincaianus 0 0 0 0 0 0 34 18 15 17 15 "Copepodite Eucalarius 0 0 0 0 0 0 0 25 0 0 0 Other copepods 0 0 15 0 0 0 75 11 7 0 3 Fish eggs 43 29 18 8 7 3 0 0 0 0 0 Fish larvae 0 0 30 74 59 57 23 7 C 0 0 Cladocerans {Evadne) 50 121 86 12 0 4 0 0 0 0 0 Polychaete larvae 0 22 104 92 56 104 78 5^ 28 22 " 21 Lamellibranch larvae 0 0 0 0 C 0 0 0 0 104 0 Cyphonautes larvae 0 0 0 0 0 0 52 140 116 30 0 'CHI OROPHYi 1 — median micrograms per liter — *^ 1 IL_^^l tV^ 1 1 1 T k_ L_ 0.33 0.36 0.44 0.63 097 1.03 2.06 0.77 0.49 0.32 0.28 PHYTOPLANKTON -r CILIATES Nitzschia spp. S rYlCwHiar"! rMimKi^r i-i^^r inn ml ... 1,627 946 2,412 1,536 2,326 1 lUI 1 ii->c;i y^i 987 r \J\J INI 1,369 1,008 261 Bactenastrum spp. (chains) 56 86 84 78 78 95 66 54 16 Chaetoceros spp. 2,488 3,845 4,307 3,246 3,576 1,427 4.270 1.764 413 Rhizosolenia alata 112 116 90 86 80 35 18 4 2 Rhizosolenia fragillssima 98 120 156 120 268 53 6 0 8 Skeletonema costatum 78 56 18 96 76 24 3 88 40 Ceratium spp. 42 30 24 32 14 11 54 4 0 Gonyaulax polyedra Ipolygramma 26 14 16 20 12 15 14 0 0 Prcrocentrum sp. C 0 0 2 4 94 34 0 0 0 Piotoperidinium spp 26 30 26 20 18 7 22 2 4 Cochlodinium catenatum 50 56 290 714 1,272 4,065 1,182 148 68 Gymnodinium splendens 0 0 0 2 0 341 300 0 0 Torodinium robustum 30 26 20 10 0 16 12 0 0 Umbilicosphaera sibogae 50 52 42 34 44 0 18 4 0 'Emiliania huxleyi 1,213 1,821 1,822 1,298 1,883 1,928 927 2,489 1,152 Mesodinium rubrum 8 2 6 8 2 8 4 8 14 Laboea spp 30 146 110 74 82 104 124 72 36 Lohmaniella spp. 68 128 116 94 134 154 218 78 50 'Includes some Clausocalanus. 167 FISHERY BULLETIN: VOL. 83, NO. 2 APPENDIX TABLE 2.— Diurnal profiles after storm. Depth (m): 10 15 20 25 30 35 40 45 50 ZOOPLANKTON Naupliar Acartia Naupliar Labidocera Naupliar Paracalanus"^ Naupliar Calanus Naupliar Rhincalanus Other nauplll Female Acartia Male Acartia Copepodlte Acartia Adult Labidocera Copepodlte Labidocera Adult and copepodlte "Paracalanus"' Appendicularians Adult and copepodlte Oithona Adult and copepodlte Euterpina Euphausiid nauplii Euphausiid calyptopis Euphausiid furcilia Chaetognaths Female Calanus Male Calanus C V Calanus C IV Calanus cm Calanus C II Calanus C I Calanus Adult Corycaeus Adull and copepodlte Oncaea Microsetella Adult Metndia Adull Pleuromamma Copepodlte Metndia and Pleuromamma Adult Rhincaianus Copepodlte Rhincalanus Copepodlte Eucalanus Other copepods Fish eggs Fish larvae Cladocerans {Evadne ) Polychaete larvae Lamellibranch larvae Cyphonautes larvae CHLOROPHYLL PHYTOPLANKTON -r CILIATES Nitzscnia spp S Bacteriastrum spp. (chains) Chaetocoros spp Rhizosolenia alaia Rhizosolenia fragilissima Skeletorema costatum Ceratium sop Gonyaula\ pclyedra Ipclygramma Prorocentrum sp C Protcperidinium spp Cuchiodinium catenatum Gymnodinium splendens Torodinium robustum Umbiliccsphaora sibogae Emiliania huxleyi Mesodinium ruorurn Laboea spp Lohmanielia spp. 1,834 8,333 10,525 21,954 1 1'cuiai i 1 lu 6,583 1,867 197 153 60 0 57 1.600 4.679 0 0 0 0 0 0 0 0 0 2.644 10,114 3,669 5,576 7,219 7,810 5,015 1.651 972 888 1,115 950 810 424 2,180 1.833 837 456 391 193 128 196 0 631 0 95 134 9 204 228 0 36 124 2,740 14,601 15,091 18,210 14,232 8,216 5,327 3,990 3,414 2,570 1,927 0 5,666 69 9,736 6,617 3,994 0 1.907 0 0 0 0 5,645 27 14 6,617 7 0 0 0 0 0 36 643 1.758 1,756 1,807 288 0 1,905 1,903 0 0 0 0 0 0 0 0 0 0 0 0 0 153 79 7 0 0 0 0 0 0 0 0 5.646 6,200 1,726 1,796 2,279 4,808 6,434 3. 090 1 371 660 435 3,053 3,165 3.551 5,785 8,364 5,651 1,745 754 122 79 44 21 696 391 983 814 ^ei? 1.953 1.014 :'30 567 516 5 229 1,196 2,358 962 190 80 23 0 25 0 0 0 171 0 214 9 19 33 27 26 25 0 13 11 89 41 38 21 2C 15 9 14 0 0 12 15 23 36 27 28 IS 16 20 8 164 511 251 190 60 42 24 12 6 12 0 0 0 0 0 0 0 22 8 8 4 0 0 0 0 0 0 0 20 4 0 4 0 0 0 116 0 11 12 18 12 0 4 0 68 0 119 45 16 32 6 0 8 0 0 0 0 119 90 43 14 8 0 0 0 7 0 297 27 50 46 11 0 0 0 0 7 79 303 15 31 25 12 0 0 0 C 20 107 337 305 478 74 87 69 43 46 43 0 711 255 215 707 2,678 3.466 2.011 1,797 1,322 1,204 14 40 719 476 45 171 236 190 23 37 33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1,047 0 0 0 0 0 0 0 0 0 81 153 176 135 115 0 0 0 0 0 0 0 0 97 80 0 0 0 0 0 30 31 57 8 99 80 6 0 0 0 0 23 16 0 7 97 80 0 0 0 0 15 0 17 6 6 10 0 6 14 18 19 89 0 0 0 0 C 0 0 0 11 15 95 19 0 0 0 0 0 0 164 273 101 102 16 0 0 0 0 0 0 0 100 294 193 375 256 181 101 52 33 19 0 0 0 0 0 54 195 138 0 0 0 0 193 20 36 15 58 0 3 0 0 0 0.74 1.08 2.91 median micrograms per liter ■ 2.05 0.86 0.62 median number oer 100 ml ■ 0.52 0.36 67 48 0 0 0 0 0 130 0 0 20 24 8 4 4 a 12 0 7,957 3.186 265 1,416 1,099 696 87 565 174 28 24 8 16 8 0 0 0 0 44 16 36 8 4 0 0 0 0 0 0 0 0 0 0 0 0 24 64 80 72 104 56 0 0 4 0 16 56 88 32 16 0 0 0 0 20 4 681 784 296 4 0 4 0 48 64 72 40 16 4 0 0 C 52 363 1,837 1,672 1,720 680 56 52 4 0 0 672 583 320 4 C 0 0 24 24 26 28 8 0 0 0 0 0 104 40 56 32 4 0 0 0 1,681 1,504 3,009 1,858 609 1,130 696 1,478 973 16 24 32 IS 8 0 0 0 0 52 464 326 134 204 116 32 20 4 143 520 580 568 616 152 36 16 6 0.31 0.21 ' Includes some Clausocalanus . 168 MULLIN ET AL.: VERTICAL STRUCTURE OF PLANKTON OFF CALIFORNIA APPENDIX TABLE 3.— Nocturnal profiles before storm. Depth (m): 10 15 20 25 30 35 40 45 50 ZOOPLANKTON Naupllar Acartia Naupliar Labldocera Naupllar "Paracalanus"^ Naupliar Calanus Naupliar Rhincalanus Other nauplii Female Acartia Male Acartia Copepodite Acartia Adult Labidocera Copepodite Labidocera Adult and copepodite Paracalanus^^ Appendicularians Adult and copepodite Oithona Adult and copepodite Euterpina Euphausiid nauplii Euphausiid calyptopis Euphausiid furcilia Chaetognaths Female Calanus Male Calanus 0 V Calanus C IV Calanus cm Calanus C II Calanus C I Calanus Adult Corycaeus Adult and copepodite Oncaea Microsetella Adult Metndia Adult Pleuromamma Copepodite Metndia and Pleuromamma Adult Rhincalanus Copepodite Rhincalanus Copepodite Eucalanus Other copepods Fish eggs Fish larvae Cladocerans (Evadne) Polychaete larvae Lamellibranch larvae Cyphonautes larvae CHLOROPHYLL PHYTOPLANKTON ^ CILIATES Nitzschia spp. S Bactenastrum spp. (chains) Chaetoceros spp. Rhizosolenia alata Rhizosolenia fragilissima Skeletonema costatum Ceratium spp Gonyaulax polyedra polygramma Prorocentrum sp. C Protopendinium spp. Cochlodinium catenatum Gymnodinium splendens Torodinium robustum Umbilicosphaera sibogae Emiliania huxleyi Mesodinium rubrum Laboea spp Lohmaniella spp. 352 315 4,286 5.874 1 iicuiai 1 1 lu 6,723 Miuci per ui 1,992 0 43 0 0 0 1.636 870 75 0 0 0 0 0 0 0 0 5.282 4,229 2,491 2,379 2,500 8,352 2,215 1,277 1,051 1,041 748 691 441 1.008 744 2,033 579 340 81 81 41 71 0 72 0 8 8 97 23 0 0 0 0 2,636 5,507 7,143 1 1 ,822 12,602 9.650 2,764 3,333 2,358 2,073 1,594 14 8 25 59 52 16 0 0 0 0 0 21 0 25 16 21 4 0 0 0 0 0 45 47 517 1,784 2.546 83 72 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 532 98 75 80 0 0 0 0 0 0 0 3.040 2.047 2.264 2,066 2,033 6.364 3,574 3,191 2,412 1.165 435 2,535 2,907 3,218 4,164 6,134 1.736 488 255 45 37 7 282 157 1,345 2,231 3.821 1.992 1,824 1.404 545 805 725 67 36 130 2.320 1,736 413 130 0 0 0 0 0 0 4 33 4 17 0 8 4 0 0 14 13 4 40 46 21 0 4 0 4 7 8 4 4 24 25 19 47 26 33 73 7 45 141 189 252 142 93 85 41 28 22 11 0 0 0 4 16 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 17 15 21 4 0 4 0 0 0 14 14 17 19 10 4 4 0 4 0 4 0 0 4 0 16 50 10 0 0 0 0 0 0 0 12 21 161 20 0 0 0 0 0 4 13 4 49 119 9 0 0 0 0 14 16 91 123 142 194 163 119 70 61 36 240 157 0 0 325 3,636 3,453 1,707 1,284 1,487 1,143 141 394 613 1.440 1,220 579 488 71 19 33 36 0 0 0 0 0 4 0 20 33 15 7 0 0 0 4 0 11 23 45 28 26 18 4 0 4 45 29 153 293 203 167 134 112 0 0 0 0 0 0 0 0 0 0 0 0 0 8 7 10 23 51 33 4 0 0 0 0 0 0 0 12 8 8 0 0 0 4 4 0 0 0 0 0 8 12 23 24 4 0 0 0 3 0 0 0 0 0 0 21 7 67 59 28 11 8 0 0 0 0 120 101 13 4 0 4 0 0 0 0 0 14 22 50 96 366 257 169 80 12 31 32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20 8 4 0 0 0 0 0 0.24 0.29 0.49 0.68 median micrograms per liter- 0.85 1.82 1.42 median number per 100 ml ■ 0.97 0.51 1,036 2,181 791 372 1,593 162 841 241 217 36 90 44 78 14 168 108 50 18 5,974 1,855 3,628 2,008 1,814 3,585 2,035 1,428 435 88 122 96 72 34 28 6 2 4 108 160 114 70 68 11 8 2 4 56 74 70 8 10 18 96 52 116 18 24 38 32 18 28 4 2 0 10 2 54 26 18 2 0 0 0 2 0 58 76 50 6 0 0 0 14 16 12 22 23 32 10 0 0 46 96 456 1,092 1.869 2.064 522 68 0 0 0 2 12 241 21 0 0 0 28 30 12 12 7 0 0 0 2 16 0 16 86 0 30 2 2 0 858 1.518 1,498 706 1.372 2,173 3,009 1,580 957 2 34 6 4 0 11 4 6 4 102 74 134 122 217 125 44 24 14 210 84 152 174 118 123 30 20 12 0.32 0.26 includes some Clausocalanus 169 APPENDIX TABLE 4.— Nocturnal profiles after storm. FISHERY BULLETIN: VOL. 83, NO. 2 Depth (m): 10 15 20 25 30 35 40 45 50 ZOOPLANKTON Naupliar Acartia Naupliar Labidocera Naupliar "Paracalanus"^ Naupliar Calanus Naupliar Rhincalanus Other nauplii Female Acartia Male Acartia Copepodite Acartia Adult Labidocera Copepodite Labidocera Adult and copepodite "Paracalanus'^ Appendicularians Adult and copepodite Oitliona Adult and copepodite Euterpina Euphauslid nauplii Euphausiid calyptopis Euphausiid furcilia Chaetognaths Female Calanus Male Calanus C V Calanus C IV Calanus cm Calanus C II Calanus C I Calanus Adult Corycaeus Adult and copepodite Oncaea Microsetella Adult Metridia Adult Pleuromamma Copepodite Metridia and Pleuromamma Adult Rhincalanus Copepodite Rhincalanus Copepodite Eucalanus Other copepods Fish eggs Fish larvae Cladocerans (Evadne) Polychaete larvae Lamellibranch larvae Cyphonautes larvae CHLOROPHYLL PHYTOPLANKTON + CILIATES Nitzschia spp. S Bacteriastrum spp. (chains) Chaetoceros spp. Rhizosolenia alata Rhizosolenia fragilissima Skeletonema costatum Ceratium spp. Gonyaulax polyedra Ipolygramma Prorocentrum sp. C Protoperidinium spp. Cochlodinium catenatum Gymnodinium splendens Torodinium robustum Umbilicosphaera sibogae Emiliania huxleyi Mesodinium rubrum Laboea spp. Lohmaniella spp. median number per cubic meter 6,590 9,360 17.323 18,008 15,106 11.371 660 176 90 230 0 1,172 1.181 394 336 0 0 0 0 0 0 0 4,908 5,256 7,591 4.370 5,184 9,091 6,400 3,424 1,345 1,073 1,126 1,099 1.378 1,969 1,533 1,082 1,489 495 467 545 236 260 0 0 0 0 0 167 7 4 0 0 0 3.678 7,102 12.795 27,969 11,873 12,553 8,000 6,154 5.364 4.245 3,520 11 23 66 138 13 0 0 0 0 0 0 11 28 17 123 27 0 0 0 • 0 0 0 280 661 1,575 2,299 1,505 3,617 32 0 0 0 0 11 0 0 0 0 0 0 0 0 0 0 119 80 31 15 0 0 0 0 0 0 0 4,215 3,977 2,953 2,490 1.915 4,013 5,400 2,731 1.887 720 1.039 2,835 4,400 4,921 5,556 2,814 4,255 4,200 881 377 307 80 172 394 495 347 1,003 1,505 1,800 952 755 169 560 57 142 110 1.261 3,617 167 165 16 27 5 0 0 0 0 0 0 0 14 9 14 20 13 57 40 37 83 17 4 13 7 9 8 0 11 0 29 61 27 13 11 26 9 16 9 95 100 202 215 54 112 49 57 9 19 9 0 0 0 15 17 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 4 4 11 0 15 9 13 0 0 4 0 0 4 23 26 0 43 27 7 0 0 4 0 7 31 26 31 34 80 13 0 0 0 0 0 16 26 42 43 85 0 0 0 0 0 7 16 8 0 0 68 7 0 0 0 0 11 0 47 252 732 190 130 55 45 28 14 230 240 94 1,513 1,003 2,766 4.400 1.868 1,345 1,792 1,200 115 320 495 347 426 334 660 78 245 46 20 0 8 16 0 0 13 7 4 9 5 0 8 16 16 0 0 0 32 44 31 5 9 8 8 16 17 0 202 366 229 188 88 44 0 0 0 0 0 0 0 0 0 0 0 0 0 12 0 17 67 18 4 9 0 4 0 0 0 0 0 0 0 0 5 0 0 4 0 0 15 27 51 16 18 54 15 26 0 16 4 0 0 0 0 0 0 0 0 0 0 26 42 0 0 0 0 0 0 0 29 197 330 17 0 0 0 0 0 0 4 18 20 106 276 272 334 165 311 47 27 20 0 0 0 0 0 0 0 0 0 0 0 0 0 4 56 121 67 25 0 0 0 0 1.17 1.34 2.87 3.87 median micrograms per liter 4.15 3.11 1.01 median number per 100 ml ■ 0.61 0.46 88 0 0 88 0 0 0 43 130 194 9 8 0 0 0 12 12 0 5,221 9.469 3,363 442 4,071 4,690 783 174 217 56 19 32 19 8 0 0 4 0 83 0 0 111 0 0 4 20 0 0 0 0 0 0 0 40 0 8 111 46 80 167 32 24 4 0 2 0 28 40 37 41 0 0 0 0 556 806 1,360 1,704 290 88 4 0 0 56 102 32 0 16 16 12 0 2 1,833 1,519 1.488 3.259 3,343 3,424 516 56 10 306 352 464 370 4,102 1,016 32 0 0 56 37 64 74 16 40 4 0 0 28 9 24 74 24 16 4 0 0 1,062 1,327 1,858 1,416 2,035 2,035 1.652 1.043 478 389 176 72 241 14 16 12 0 0 472 222 344 389 97 400 72 36 8 528 435 512 1,185 207 552 180 24 4 0.38 0.35 'Includes some Clausocalanus. 170 DIEL AND DEPTH VARIATIONS IN THE SEX-SPECIFIC ABUNDANCE, SIZE COMPOSITION, AND FOOD HABITS OF QUEENFISH, SERIPHUS POLITUS (SCIAENIDAE) Edward E. DeMartini/ Larry G. Allen ,2 Robert K. Fountain,^ AND Dale Roberts* ABSTRACT Lampara seine-hauls were taken during day and night over 5-27 m bottom depths off the coast of northern San Diego County, California, from September 1979 to March 1981. These samples were used to characterize the temporal and spatial patterns of the abundances and size and sex compositions of queenfish, Seriphus politus, in an unprotected, coastal environment. Stomach contents of sample queenfish were examined to aid our interpretation of these patterns. Adult queenfish of both sexes made diel, onshore, and offshore migrations, but immature fish generally did not. Both immatures and adults occurred in epibenthic, resting schools in shallow areas ( -10 m or less depth, within -1.5 km of shore) during the day. At night, adult fish dispersed (to >3.5 km ) offshore. On average, a greater fraction of the adult males emigrated farther offshore at night than adult females. Immature fish remained inshore of 16 m bottom depths (within -2.5 km of shore) at night, with the majority staying inshore of -10 m depth. Regardless of maturity class, larger fish occurred farther offshore at night. Stomach contents data confirmed the primarily nocturnal feeding habits of both immature and adult queenfish. Immatures fed primarily on meroplankton and other nearshore prey; however, adults captured offshore had also eaten some nearshore prey. Thus, food habits explain much, but not all of the diel migratory pattern. Immature queenfish may also remain nearshore at night because migration is not worthwhile energetically and because of greater risk of predation offshore. Adults perhaps also migrate offshore at dusk to spawn. Numerous physical and biological factors influ- ence the spatial and temporal distribution pat- terns of fishes. In response to such factors, coastal marine fishes often undergo diel shifts in spatial distributions (reviewed by Woodhead 1966; Blax- ter 1970). Examples of horizontal (Hobson 1965, 1973; Hobson and Chess 1976; Quinn et al. 1980; Allen and DeMartini 1983) and vertical or water- column (Parrish et al. 1964; Woodhead 1964; Beamish 1966) diel migrations are recognized. Diel horizontal migrations may vary with life stage (e.g., see Hobson and Chess 1973). The type of diel vertical movement also may vary with sea- son and with age and spawning condition of fish (Hickling 1933; Lucas 1936; Brawn 1960; Blaxter and Parrish 1965; Beamish 1966). In other cases, relatively static differences between the depth dis- tributions of juvenile and adult life stages have ^Marine Science Institute, University of California, Santa Barbara, Calif; present address: Marine Review Committee Re- search Center, 531 Encinitas Boulevard, Suite 114, Encinitas, CA 92024. Marine Science Institute, University of California, Santa Barbara, Calif.; present address: Department of Biology, Califor- nia State University, Northridge, CA 91330. Manuscript accepted March 1984. FISHERY BULLETIN; VOL. 83, NO. 2, 1985. been documented (reviewed by Helfman 1978). Spatial segregation of adult males and females has been commonly observed only in tropical reef fishes (Moyer and Yogo 1982; Clavijo 1983; and others). This study describes the manner in which a complex interplay of the factors listed above can determine the temporal and spatial patterns of the distribution of a temperate marine fish. Specifi- cally, we report on diel shifts in the onshore, offshore distribution of queenfish, Seriphus politus, characterize the variation in these diel shifts for immature, adult male, and adult female fish, and relate these shifts to feeding, anti- predator, and breeding functions previously de- scribed. The queenfish is a small, schooling sciaenid whose center of geographic distribution lies in the Southern California Bight, south of Point Concep- tion (Miller and Lea 1972). The species contributes significantly to the sport fish catch on piers in southern California (Frey 1971) and provides for- age for several game fishes (Young 1963; Feder et al. 1974). Queenfish form inactive, epibenthic schools nearshore (at — 10 m or less bottom depth) 171 FISHERY BULLETIN: VOL. 83, NO. 2 during the day (Hobson and Chess 1976; Allen and DeMartini 1983). Queenfish are dispersed throughout the water column and also occur farther offshore (to 20-30 m depths) at night, where they feed (Hobson and Chess 1976; Hobson et al. 1981; Allen and DeMartini 1983) and perhaps spawn (DeMartini and Fountain 1981). This study represents part of an ongoing en- vironmental impact assessment of the fishes of coastal waters off San Onofre Nuclear Generating Station near Oceanside, Calif., using the queenfish as a target species. Recognition of poten- tially complex patterns of spatial and temporal distribution has general applicability for the de- sign and interpretation of analogous monitoring studies and for other assessments of nearshore fish stocks (June 1972). METHODS AND MATERIALS Sampling Design Catches made by lampara seines (a type of semipursing, roundhaul net, Scofield 1951) were used to characterize the distribution and abun- dance of queenfish in terms of catch per unit effort (CPUE), where a standard-area seine-haul was defined as the unit of effort (Allen and DeMartini 1983). All queenfish present in each seine-haul were counted aboard ship. Seines fished from sea surface to seabed over bottom depths from 5 to 27 m. For diel comparisons, a total of 14 pairs of "day" (1-6 h after sunrise) and "night" (1-6 h after sunset) cruises were made during the period from Sep- tember 1979 to March 1981, inclusive. On each cruise, 1 or 2 seine-hauls were made within ran- domly selected subareas within each of three depth blocks (shallow, 5-10 m, 0.5-1.5 km offshore; middepth, 11-16 m, 1.5-2.5 km offshore; deep, 18-27 m, 2.5-3.5 km offshore) at each of two longshore locations, about 5 and 22 km upcoast of Oceanside, Calif. Two replicate hauls were made at each longshore location in the shallow depth block (wherein catches were most variable) on day cruises, and the two catch values averaged. For a chart of the study area and further details of gear and sampling designs, see Allen and DeMartini (1983). CPUE and Size-Composition Data A maximum of two subsamples of —50 individu- als each of queenfish of all sizes were randomly selected from each seine-haul and placed on ice 172 aboard ship. In the laboratory, all fish in the sub- samples were sexed macroscopically (DeMartini and Fountain 1981) into immatures (of both sexes), adult males, adult females, and sex indetermin- able. (Fish of indeterminable sex comprised <5% of total catch.) Fish were measured to the nearest millimeter standard length (SL) and grouped into 5 mm length classes for analysis. For seine-hauls in which the total number of queenfish caught exceeded the total number measured, the numbers of fish of each maturity and sex category caught were estimated from the respective number mea- sured, standardized to the total number of queenfish caught. In these cases, the length fre- quencies of the fishes in each sex category mea- sured were then weighted by the estimated number of that category present in the haul. Queenfish length-frequency data were com- pared between diel periods and depth blocks by Kolmogorov-Smirnov Two-Sample test (Siegel 1956). A nonparametric 3-way ANOVA (Wilson 1956), available in the IMSL Library's^ statistical package, was used to simultaneously evaluate the effects of diel period, depth block, sampling date (cruise), and their potential interactions on the numerical CPUE of immature, adult male, and adult female fish. In all ANOVA analyses, catches made w.ithin the same depth block at the two long- shore locations on a given cruise were considered separate estimates, as differences between loca- tions were sometimes evident. Food Habits Additional subsamples of one queenfish per 10 mm SL length class were randomly selected from seine-hauls for analysis of food habits. Fish were examined from a larger series of 11 day and 23 night cruises (that included 8 of the aforemen- tioned 14 paired, day/night cruises) conducted during September 1979-October 1980. These sub- sampled fish were placed in lO^c Formalin'* immediately following capture, after their abdom- inal walls had been slit to accelerate preservation. Stomachs were dissected and placed in 70% ethyl alcohol after about 1 wk of fixation. Contents of stomachs were scored for state of digestion on a scale of 0 (undigested) to 10 (prey present but to- tally indistinguishable). All prey were identified to lowest taxonomic category, their numbers tal- 'IMSL Library, Sixth Floor, NBC Building, 7500 Bellaire Blvd., Houston, TX 77036. ■* Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. DeMARTINl ET AL: DISTRIBUTION PATTERNS OF QUEENFISH lied, and reconstructed wet weight biomass esti- mated (to the nearest milligram, based on a key of reference prey weights). Values were summed for the aggregate of each taxon in each stomach. A compound measure of numbers (N), weight (Wt), and frequency occurrence (FO) of prey (IRI = {(%N + 7r Wt)%FO}; Pinkasetal. 1971) was used to characterize temporal and spatial variations in the overall importance of various prey to the diet of immature and adult male and female queenfish. RESULTS Catch per Unit of Effort A large majority of the adults of both sexes, as well as immature queenfish, occurred at shallow depths (5-10 m) during daylight hours throughout most of the year (Fig. 1). A plurality of immature and adult fish of both sexes occurred at shallow depths at night as well; however, the distribution of numbers spread farther offshore at night, espe- cially for adult fish (Fig. 1). The nocturnal offshore distribution appears to have been especially marked for adult males (Fig. IB). This diel shift in the depth distribution of queenfish is charac- terized by the diel-by-depth interaction term in the ANOVA (Table 1). The (nearly) significant date effect for adults (and insignificant date effect for immatures) in the ANOVA (Table 1) reflects the general offshore emigration of adult, but not im- mature, queenfish during late fall and early Table l. — Results of Wilson's Three- Way Non- parametric ANOVA with equal replication (Wil- son 1956) for the effects of diel period (day, night), bottom depth (5-10, 11-16, 18-27 m), and date (cruise ) on the lampara seine CPUE of immature, adult male, and adult female queenfish. Data for 14 paired, day/night cruises made during the period from September 1979 to March 1981, inclu- sive. (* denotes significance at P s 0.05). Maturity' sex category Factor / df P Immatures Diel 27.5 1 - o UJ o UJ a: UJ (J X z 10- 5- 0-- 5- 10- 15- 10 20 30 40 rrrrp, 50 rrrrp-, 80 70 ""I" eo 90 '"I"" 100 no 120 130 IdO 150 STANDARD LENGTH (MM) Figure 2. — Relative (percenti length-frequency distributions of lAi immature, - (J 3 o LjJ 15- 10- 5- 0-- 5- 10- 15- 20- 50 ""1 1 1 1" BO 70 80 90 '"I I I I I 1 1 1 1 1 1 1 100 110 120 130 140 150 160 170 160 190 200 ADULT FEMALE QUEENFISH SIZE FREQUENCIES DEPTH=5-10 M DAY N=3292 NIGHT N=S42 Ld O Ld Q. < a X o z 20- 15- 10-1 0-- 10- 15- 60 c '"1 1" 80 90 ■"I ' I I I I I I I I ' I I 100 110 120 130 140 150 160 170 160 190 200 210 220 STANDARD LENGTH (MM) 175 FISHERY BULLETIN: VOL. 83, NO. 2 IMMATURE QUEENFISH SIZE FREQUENCIES NIGHT 5-lOM DEPTH N=2247 NIGHT 11-16M DEPTH N=725 n a. lit a I o liJ O z UJ a: Z) o o o u. o >- o z LJ o UJ 10- 5- 5- 10- 15- 30 40 ' ' ' ' I I ■ 50 80 70 80 90 "I I " 100 110 ADULT MALE QUEENFISH SIZE FREQUENCIES NIGHT 5-lOM DEPTH N=984 NIGHT 11-16M DEPTH N=1096 " 1 I ■ 120 130 z UJ o a: a- § a. UJ o I a 20- 15- 10- 5- 10- 15- B 'I I T"' I" 80 70 80 90 100 '"1"" 110 'I I' MO 150 '"I"" 180 ^V 120 130 MO 150 180 170 STANDARD LENGTH (MM) "■1 1 1 1 1"" 180 190 200 210 220 176 DeMARTINI ET AL: DISTRIBUTION PATTERNS OF QUEENFISH ADULT FEMALE QUEENFISH SIZE FREQUENCIES NIGHT 5-lOM DEPTH N=642 NIGHT 11-16M DEPTH N=350 LiJ O q: O (J o >- o UJ o ir UJ Q. o X en g 2 10- 0-- 5- 10- 15- c 5 1'" 0 60 70 r" 80 I'" 90 1"" 100 ""1 1 110 120 STANDARD '"1 1 1"" 130 140 150 LENGTH (MM) 160 — I — 170 180 190 ""] ! 200 Figure 3. — Relative length-frequency distributions of (A) immature, (B) adult male, and (C) adult female queenfish caught during the night in the shallow versus middepth blocks (see Figure 1 caption for details). (Data for the deep depth block were too few to evaluate independently. ) m depths nearshore during the day (also see Dis- cussion and Conclusions). In order to further aid our interpretation of the function of the observed nocturnal offshore move- ments, we subdivided our diel catch data into three periods of year: 1) February-July (the onshore, breeding season; see DeMartini and Fountain 1981); 2) August-October (the onshore, non- breeding season); and 3) November-January (the offshore, nonbreeding season). Analyses were restricted to size-frequency data for queenfish seined at shallow and middepths during the two periods of onshore distribution, as scant data on the size composition of adults were available for the offshore season. During both breeding and nonbreeding onshore periods, queenfish seined at shallow depths were larger during the day versus at night, and fish caught at night were consis- tently larger in middepth versus shallow collec- tions (Table 3). Thus the year-round patterns illustrated by Figures 2 and 3 also basically characterize both breeding and nonbreeding periods of onshore distribution. Food Habits The food habits of immature, adult male, and adult female queenfish were evaluated for day and night collections made in the shallow- and mid- depth blocks during the onshore, breeding and onshore, nonbreeding periods of year (Stomachs of fish from the deep depth block were not examined. ) The purpose of these comparisons was to help interpret the relative importance of the feeding and breeding functions of diel offshore movements. We hypothesized that immature fish might remain onshore at night to feed on meroplanktonic (noc- turnally active) demersal crustaceans and other prey more abundant at shallow depths. We further expected that adults emigrated offshore to spawn (DeMartini and Fountain 1981) and thereafter fed on relatimely larger prey that were more preva- lent farther offshore. In general, immature queenfish fed on smaller prey than adult males, and adult males, being smaller than adult females, fed on generally smaller prey than females (Table 4). Contrary to expectations, adult 177 FISHERY BULLETIN: VOL. 83, NO. 2 IMMATURE QUEENFISH SIZE FREQUENCIES DAY 5-lOM DEPTH N=9394 NIGHT 5-27M DEPTH N=3013 I X o z UJ O Z UJ q: a: o o o u. o >- o z llJ z> o UJ on u. UJ o UJ Q. i 10- 5- 0-- 5- 10- 15- I I I [ 1 1 1 1"! I I iiimiiiii»iniiiiMnn ii*! j I 0 10 20 30 40 50 80 70 80 90 100 110 120 130 140 150 ADULT MALE QUEENFISH SIZE FREQUENCIES DAY 5-lOM DEPTH N=3238 NIGHT 5-27M DEPTH N=3155 'I iijii mill n 1 1*1 II iiji ;iiiiiiii>pinniii[iniMiii[i ■iiliiii[iMi |ii |iiii [ III! imtiMi ji \t |llllllli« | 80 70 80 90 100 110 120 130 140 150 180 170 180 190 200 210 220 STANDARD LENGTH (MM) 178 DeMARTINI ET AL: DISTRIBUTION PATTERNS OF QUEENFISH Ld O cr o o o o >- o u o UJ ij- 1- ^- N? Dmax P D>N? Dmax P Day vs. night, shallow depths Immatures yes Adult males yes Adult females yes 0.37 0.06 0.28 <0.001 0.05 <0.001 yes yes yes 0.10 0.10 0.25 <0.001 0.05 >P> 0.01 <0.001 Mid > Shal? Dmax P Mid > Shal? Dmax P Shallow vs. mid- Immatures Adult males Adult females depths, at night yes yes yes 0.19 0.32 0.15 <0.001 <0.001 0.01 yes yes no 0.33 0.12 0.14 <0.001 0.05 >0.1 queenfish of both sexes, as well as immatures, fed to large extents on prey (Table 4) whose centers of abundance were shallow (see Discussion and Con- clusions). For immatures such major prey included holoplankton {Labidocera trispinosa) and mero- planktonic cumaceans (primarily Diastylopsis tenuis) (Table 4). Diastylopsis tenuis and other nearshore crustaceans also comprised nontrivial fractions of the diet of adult queenfish d\iring the periods of onshore depth distribution (Table 4). Diastylopsis tenuis, a night-active meroplankter characteristic of the shallow region (see below), also occurred in the stomachs of fish collected at shallow depths during the day and during the night at middepths. Hence we feel justified to characterize the queenfish diet using data for both diel periods and depth blocks pooled in Table 4. State of prey digestion, though, tended to be less at 179 FISHERY BULLETIN: VOL. 83, NO. 2 Table 4. — Prey taxa comprising ^1*^ of the diet (by IRI) of immature, adult male, and adult female queenfish during the two periods of onshore distribution combined. Sample fish collected during both day and night and at shallow (5-10 m) and middepths (11-16 m) are pooled (see text). Diet characterized by number (N), reconstructed wet weight (Wt), and frequency of occurrence (FO) of prey; overall contribution to diet evaluated by IRI (Pinkas et al. 1971). Prey ranked by IRI within queenfish categories. A^ = number of fish examined that had food in their stomachs. Mean (and range I of body lengths (SL, mm) of fish examined were immatures — 70 (34-100), adult males— 127 (76-210), and adult females — 146 (102- 214). % Mean Wt/ Type of prey IRI %A/ %wt %FO prey (mg) Immatures {n = 57) Labidocera thspinosa holoplanktonic copepod 35.8 26.7 3.2 56.1 <0.2 Acartia tonsa holoplanktonic copepod 20.9 40.2 2,8 22.8 <0.1 Metamysidopsis elongata meroplanktonic mysid 15.6 5.9 7.9 52.6 1.7 Engraulis mordax clupeold fish 9.0 1.1 11.6 33.3 13.7 Diastylopsis tenuis meroplanktonic cumacean 5.1 3.6 3.2 35.1 1.1 Ogyrides sp. ? meroplanktonic caridean shrimp 2.4 <0.1 31.9 3.5 472 Atylis tridens meroplanktonic amphipod 2.1 2.0 2.6 21.1 1.7 Acanthomysis macropsis meroplanktonic mysid 1.3 0.6 2.2 21.1 4.7 Blephahpoda occidentalis ? meroplanktonic megalops/juvs. 1.0 0.2 9.0 5.3 53.0 (anomuran) All other prey 6.8 19.6 25.6 — 1.7 Adult males (n = 228) Engraulis mordax clupeoid fish 74.6 6.3 89.0 47.8 385 Diastylopsis tenuis meroplanktonic cumacean 7.1 18.4 0.9 224 1.3 Metamysidopsis elongata meroplanktonic mysid 6.4 10.0 08 36.4 2.1 Labidocera trispinosa holoplanktonic copepod 6.4 14.9 0.1 25.9 <0.2 All other prey 5.5 50.4 92 — 5.0 Adult females (n = 236) Engraulis mordax clupeoid fish 78.9 5.7 90.8 52.1 783 Metamysidopsis elongata meroplanktonic mysid 6.5 10.8 0.5 36.4 2.3 Diastylopsis tenuis meroplanktonic cumacean 3.7 10.8 0.3 21.2 1.2 Labidocera trispinosa holoplanktonic copepod 3.1 10.2 <0.1 19.4 <0.2 Acanthomysis sculpta meroplanktonic mysid 1.4 6.1 0.5 13.6 3.6 Caridean shrimp ? meroplanktonic 1.0 3.4 0.6 15.7 8.7 All other prey 5.4 53.0 7.3 — 6.9 night for both immature and adult queenfish (Table 5), indicating that all sizes of fish fed primarily at night. DISCUSSION AND CONCLUSIONS Functions of Nocturnal Offshore Dispersal in Queenfish Diel migrations of queenfish have been previ- ously reported. Queenfish have been directly ob- served emigrating offshore at dusk from inactive daytime schools nearshore at Santa Catalina Is- land, one of the Channel Islands offshore of the southern California mainland (Hobson and Chess 1976; Hobson et al. 1981). Similar behavior has been noted by Hobson^ in mainland waters off La Jolla, near San Diego. Allen and DeMartini (1983) have characterized the general pattern of noctur- nal offshore dispersal of queenfish near San Diego. Direct observation (Hobson and Chess 1976; Hob- son et al. 1981) and examination of stomachs offish collected during the day and at night (Hobson and Chess 1976; Hobson et al. 1981; Allen and DeMar- tini 1983; this study) confirm the primarily noc- turnal feeding habits of queenfish. A spawning function has also been implicated for the offshore movements of adult queenfish at dusk (DeMartini 'Edmund S. Hobson, Southwest Fisheries Center Tiburon Laboratory, National Marine Fisheries Service, NOAA, 3150 Paradise Drive, Tiburon, CA 94920, pers. commun. May 1978. 180 DeMARTINI ET AL: DISTRIBUTION PATTERNS OF QUEENFISH Table 5. — Results of Spearman's rank correlations (Siegel 1956) be- tween index of state of digestion of stomach contents and time of collec- tion for immature, adult male, and adult female queenfish. All samples collected prior to midnight. Sample fish collected during day, night, and at shallow (5-10 m) and middepths (1 1-16 m) are pooled over both onshore periods of distribution (see text). Digestion versus time rho N Immatures Adult males Adult females -0.35 -0.20 -0.22 57 228 236 0.007 0.003 ;0.001 and Fountain 1981). During the February-July/ August spawning season, ripe females with ovaries in hydrated (ready-to-spawn) condition can be collected throughout the daylight period beginning 1 h after sunrise, while females col- lected as soon as 1 h after sunset are either ripen- ing (but nonhydrated) or are recently spent (De- Martini and Fountain 1981). The diel distributional (CPUE) data of this study (Fig. 1) clearly illustrate the differences in diel migration made by immature, adult male, and adult female queenfish. Size-composition data (Figs. 2-4) further characterize the diel migrations as related to size of fish, regardless of maturity state or whether adults in the populations were reproductively active. Certain aspects of the diel CPUE data suggest a breeding function for offshore dispersal at night, since only the distribution of immatures remained centered onshore at night. Also, a dispropor- tionately greater number of adult males versus females emigrated offshore at night (Fig. IB, C). This is consistent with an offshore migration by females for spawning that occurs on a less fre- quent than daily basis, since individual female queenfish ripen and spawn batches of eggs on av- erage only once a week (DeMartini and Fountain 1981). The male-biased, daytime aggregations of ready-to-spawn queenfish (DeMartini and Foun- tain 1981) suggest that individual males spawn at more frequent than weekly intervals. Also, pre- flexion stages of queenfish larvae are most abun- dant in midwaters over 12-45 m bottom depths from 1.9 to 5.4 km offshore in the region of San Onofre-Oceanside (Barnett et al.^), which strongly suggests that most spawning occurs in outer near- shore regions. Other distributional data, however, indicate a primarily feeding function for offshore dispersal at night. The nighttime, offshore shift in the dis- tribution of adults of both sexes, for example, oc- curred during both the nonbreeding and breeding seasons. In addition, relatively more of the larger individuals among the immatures (as well as more of the larger adults of both sexes) moved offshore at night (Fig. 3) from the shallow region wherein queenfish of all sizes co-occurred during the day (Figs. 1, 4). The latter pattern persisted during both breeding and nonbreeding periods of year when distributions were generally inshore of 30 m bottom depth. On balance, the size composition of immature and adult female queenfish censused at 5-27 m depths at night resembled the composition of those censused at 5-10 m depths during the day (Fig. 4A, C), indicating that few very large imma- tures or females move offshore of 27 m at night. This moreover confirms that the queenfish seined offshore at night had resided at 5-10 m depths during the day and not in a region (e.g., shallower or deeper) that we did not census. Analogous data for adult males (Figs. 1, 4B) indicate that this may not be true for the largest males. However, the pattern of larger individuals farther offshore per- sisted for males as well as immatures and females during the nonbreeding period. The diel food habit data also are largely consis- tent with the hypothesis that queenfish disperse offshore at night to feed, despite several dis- crepancies. Certain prey are known to be much more abundant at either extreme of the queenfish depth distribution. The presence of shallow-living "marker" species such as Diastylopsis tenuis (Ta- ble 6) in the stomachs of queenfish collected offshore of the respective prey distribution likely reflects some feeding activity just prior to or dur- ing the dusk offshore emigration. The presence of some night-active meroplankton in stomachs of fish collected during the day probably represents the partial confounding of nighttime foraging by circumdiel gut residence times. We consider it un- likely that queenfish feed on prey such as D. tenuis during the day, as the nocturnal activity patterns of this and other species of demersal meroplankton "Barnett, A. M., A. E. Jahn, P. D. Sertic, and W. Watson. Long term average spatial patterns of ichthyoplankton off San Onofre and their relationship to the position of the SONGS cooling system. A study submitted to the Marine Review Committee of the California Coastal Commission, July 22, 1980. Unpubl. rep., 32 p. Marine Ecological Consultants of Southern Califor- nia, 531 Encinitas Boulevard, Encinitas, CA 92024. 181 FISHERY BULLETIN: VOL. 83, NO. 2 Table 6. — Prey taxa comprising ^1% of the diet (by IRD of immature, adult male, and adult female queenfish collected during the night from the mid-depth (11-16 m) block. Data for the two periods of onshore distribution are pooled; for further details of diet characteriza- tion see Table 4. Mean (and range) of body lengths (SL, mm) were immatures — 71 (42-86), adult males — 126 (83-183), and adult females— 147 (103-207). % N^ean Wt/ Type of prey IRI %A/ %wt %F0 prey (mg) Immatures (n = 1 4) Engraulis mordax clupeoid fish 25.2 2.6 27,2 42,9 16.8 Metamysidopsis elongata meroplanktonic mysid 23.3 7.3 9,3 71,4 2.1 Acartia tonsa holoplanktonic copepod 16.4 55.0 3,2 14.3 <0.1 Labidocera trispinosa holoplanktonic copepod 15.0 16,2 1,6 42.9 <0.2 Ampelisca brevisimulata ? demersal amphlpod 3.8 1,3 12,2 14.3 15.0 Diastylopsis tenuis meroplanktonic cumacean 3.5 3,3 2,9 28.6 1.4 Acanthomysis macropsis meroplanktonic mysid 3.1 1.3 4,2 28.6 5.2 Ampelisca cristata ? demersal amphipod 2.8 1,6 18,2 7.1 18.0 Blepharipoda occidentalis ? meroplanktonic juveniles 1.8 0,3 12.4 7.1 61.2 (anomuran) Hemilamprops californica meroplanktonic amphipod 1.6 2,3 0.6 28.6 0.4 All other prey 3.5 8,8 8.2 — 1.6 Adult males (n = 125) Engraulis mordax clupeoid fish 82.4 11,0 95.5 53.6 463 Labidocera trispinosa holoplanktonic copepod 5.8 20,8 ■0,1 19,2 <0.2 Metamysidopsis elongata meroplanktonic mysid 4.0 84 0.4 31,2 23 Diastylopsis tenuis meroplanktonic cumacean 3,2 12,2 0,3 17.6 1.3 All other prey 4.6 47,6 3,8 — 4.2 Adult females [n = 89) Engraulis mordax clupeoid fish 83.6 9 1 93,0 59.6 615 Metamysidopsis elongata meroplanktonic mysid 6,5 15,6 0,7 29.2 2.6 Diastylopsis tenuis meroplanktonic cumacean 2,5 11,6 0,2 15.7 1.2 Labidocera trispimsa holoplanktonic copepod 1,8 7,3 <0,1 18.0 <0.2 Hemilamprops californica meroplanktonic amphipod 1,0 8,1 <0,1 9.0 0.4 All other prey 4.6 48,3 6,0 — 7.5 are well recognized (Barnard and Given 1961; Hobson and Chess 1976). The significant amount of shallow-living prey such as D. tenuis and Labidocera trispinosa pres- ent in the stomachs of queenfish captured offshore at night (Table 6) nonetheless clearly illustrates that these fish had recently emigrated from depths nearer to shore. Numerous data characterize D. tenuis as largely restricted to within the 30 m isobath (Barnard and Given 1961; Parr and Diener'). Diastylopsis tenuis, in fact, declines >1 order of magnitude in abundance in benthic core ^Parr, T. D., and D. D. Diener San Onofre sand bottom benthic studies, San Onofre Nuclear Generating Station (SONGS) Units 2 and 3, pre-oi>eration monitoring results. Vol- ume 2. A study submitted to the Marine Review Committee of the California Coastal Commission, May 8, 1981. Unpubl. rep., 109 p. Marine Ecological Consultants of Southern California, 531 Encinitas Boulevard, Encinitas, CA 92024. samples between 8 and 15 m depths near San Onofre (Parr and Diener footnote 7). Labidocera trispinosa, a holoplanktonic copepod, also has been described as much more abundant inshore of 12-15 m bottom depths, both off La Jolla (Barnett 1974) and off San Onofre-Oceanside (Barnett et al.^). It seems less likely that nearshore forms such as D. tenuis andL. trispinosa are more avail- able as prey offshore at night, since they are markedly less abundant offshore. The presence of offshore prey in the stomachs of adult queenfish collected offshore obviously re- flects nocturnal foraging while in that region. ^Barnett, A. M., A. E. Jahn, P D. Sertic, and S. D. Watts. The ecology of plankton off San Onofre Nuclear Generating Station, Volume II. A study submitted to the Marine Review Committee of the California Coastal Commission, April 30, 1981. Unpubl. rep., 105 p. Marine Ecological Consultants of Southern California, 531 Encinitas Boulevard, Encinitas, CA 92024. 182 DeMARTINI ET AL: DISTRIBUTION PATTERNS OF QUEENFISH Neomysis kadiakensis, a mysid more abundant at depths corresponding to those frequented by adult queenfish at night (Clutter 1967; Bernstein and Gleye^), was a nontrivial component of the diet of adult queenfish that ranked third by weight in both males and females (although <1% of the total IRI for each sex). None of the immature queenfish that we sampled, however, had eaten any A^. kadiakensis, even its juvenile stages (which also occur offshore, Bernstein and Gleye footnote 9). Nearshore prey, such as L. trispinosa and the de- mersal meroplankter, D. tenuis, were generally more important by weight and frequency of occur- rence, if not numbers, in the diet of immature versus adult queenfish (Table 4). The tendency for immatures to remain closer to shore than adults and to feed on meroplankton (that are more abun- dant in sheltered areas closer to shore) has been noted for many species of nocturnal zooplank- tivorous fishes on tropical coral reefs (Hobson and Chess 1978). Both immature and adult queenfish are concen- trated nearshore during the day, probably in re- sponse to pressure from diurnal predators (Hobson 1978; Allen and DeMartini 1983). Several species of voracious carnivores including Pacific mack- erel. Scomber japonicus; Pacific bonito, Sarda chiliensis (Allen and DeMartini 1983); and California halibut, Paralichthys californicus, of piscivore-size (Plummer et al. 1983) are less abun- dant nearshore in the San Onofre-Oceanside re- gion. The California halibut is a known predator of queenfish (Frey 1971; Plummer et al. 1983). The kelp bass, Paralabrax clathratus, another species known to prey on queenfish (Young 1963; E. De- Martini^®), is most abundant in and near beds of giant kelp, Macrocystis pyrifera, that occur at 10-15 m bottom depths in the region (Larson and DeMartini 1984). Overall, our data indicate that the nocturnal offshore dispersal of adults and the less contagious distribution of immatures nearshore at night are primarily for feeding. Allen and DeMartini (1983) reviewed and discussed the possible advantages of dispersal for feeding in schooling, zooplank- tivorous fishes. To these we add the possible benefit (for adult queenfish) of foraging in regions where M kadiakensis, a species of large mysid, is more abundant. It is also likely that the rate at which individual queenfish encounter planktonic prey is enhanced by foraging in regions farther offshore, where longshore currents are, on aver- age, stronger (Reitzel"). The nighttime, nearshore distribution of small immature queenfish also may be due to either or both of the following factors. First, small queen- fish are undoubtedly incapable of making as ex- tensive diel migrations as adults because of body size limitations. Hence the nearshore daytime dis- tribution of immatures, probably set primarily by the influence of diurnal predators offshore, might limit the offshore movements of immatures at night. Second, predation pressure from nocturnal predators located farther offshore could restrict immature queenfish to nearshore regions. Poten- tial nocturnal predators of immature queenfish include California halibut (Allen 1982) and California scorpionfish, Scorpaena guttata (Hob- son et al. 1981). The nocturnal habits of other po- tential predators (Pacific mackerel. Pacific bonito, and Pacific barracuda, Sphyraena argentea) of small, immature queenfish are unknown. Large (>70 mm SL, Fig. 3) immature queenfish move offshore to some extent at night, which is also consistent with offshore dispersal to feed on larger prey, since size of prey is strongly related to queenfish body size (Tables 4, 6). Offshore disper- sal of large immatures is nonetheless consistent with relaxed predation pressure, since susceptibil- ity to predation must be inversely proportional to body size. Rigorous evaluation of offshore dispersal for spawning would require censuses of the onshore, offshore distribution of recent spawning products. However, we are at present unable to routinely distinguish queenfish eggs or yolk-sac larvae <2.2-2.3 mm long (~4 d old or less) (Watson^'^). In summary, we conclude that, as might be ex- pected, the distributional (including migration) patterns of queenfish have feeding, predator avoidance, and perhaps other functions such as breeding. 'Bernstein, B. B., and L. G. Gleye. The ecology of mysids in the San Onofre region. A study submitted to the Marine Review Committee of the California Coastal Commission, April 30, 1981. Unpubl. rep., 72 p. Marine Ecological Consultants of Southern California, 531 Encinitas Boulevard, Encinitas, CA 92024. '"E. DeMartini, Marine Science Institute, University of California, Santa Barbara, CA 93106. Unpubl. data. "J. Reitzel. 1979. Physical /chemical oceanography. In Interim Report of the Marine Review Committee of the Califor- nia Coastal Commission. Part II: Appendix of Technical Evi- dence in Support of the General Summary, March 12, 1979, p. 6-23. Unpubl. rep. Marine Review Committee Research Center, 531 Encinitas Boulevard, Suite 106, Encinitas, CA 92024. '^W. Watson, Marine Ecological Consultants of Southern California, Inc., 531 Encinitas Boulevard, Suite 110, Encinitas, CA 92024, pers. commun. May 1983. 183 Comparison With the Diel Migrations of Other Fishes Numerous other temperate (see Hobson and Chess 1976; Hobson et al. 1981; Allen and DeMar- tini 1983) and tropical (reviewed in Helfman et al. 1982) fishes are known to make horizontal migra- tions at dusk and dawn away from and back to reefs and other shallow areas. Such migrations have been characterized as a form of commuting between daytime resting/sheltering and night- time feeding areas (Hobson 1965, 1973). These be- haviors are most widely recognized for tropical coral reef-based fishes that forage on night-active benthic invertebrates in surrounding sandflats and seagrass beds or on nocturnal meroplankton in the water column (Hobson 1965, 1973; Domm and Domm 1973; Helfman et al. 1982; and others). The diel migration of queenfish certainly suggests feeding as a major, if not principal function. Both predator avoidance and feeding are probably major determinants of the nearshore distribution of immature queenfish. Feeding is probably the principal reason for the crepuscular onshore, off- shore migrations of adults. Offshore movement for spawning may be of secondary importance, but data are inconclusive. With the exception of the relatively short-range (within-reef) migrations observed for some tropi- cal wrasses (see Moyer and Yogo 1982 and others), we are unaware of any study of the diel migratory behavior of nearshore.temperate or tropical fishes that has demonstrated a primary spawning func- tion for the behavior We do not now believe that spawning is a major reason for the nocturnal off- shore movements of queenfish, although we still feel that spawning is partly involved. We strongly recommend that future studies of the diel migra- tory patterns of temperate and tropical fishes be watchful for possible spawning as well as feeding activity. ACKNOWLEDGMENTS We thank Art Barnett for graciously allowing us to cite some of his unpublished data and Jan Fox for typing the manuscript. An anonymous re- viewer helped us recognize the relative strengths and weaknesses of our arguments. This paper is the result of research funded by the Marine Re- view Committee (MRC), Encinitas, Calif The MRC does not necessarily accept the results, find- ings, or conclusions stated herein. FISHERY BULLETIN; VOL. 83, NO. 2 LITERATURE CITED ALLEN, L. G., AND E. E. DEMARTINI. 1983. Temporal and spatial patterns of nearshore distribu- tion and abundance of the pelagic fishes off San Onofre- Oceanside, California. Fish. Bull., U.S. 81:569-586. ALLEN, M. J. 1982. Functional structure of soft-bottom fish communities of the southern California shelf. Ph.D. Thesis, Univ. California, San Diego, 577 p. BARNARD, J. L., AND R. R. GIVEN. 1961. Morphology and ecology of some sublittoral cuma- cean Crustacea of Southern California. Pac. Nat. 2:153- 165. BARNETT, A. M. 1974. The feeding ecology of an omnivorous neritic copepod, Labidocera trispinosa Esterly. Ph.D. Thesis, Univ. California, San Diego, 233 p. BEAMISH, E W. H. 1966. Vertical migration by demersal fish in the northwest Atlantic. J. Fish. Res. Board Can. 23:109-139. BLAXTER, J. H. S. 1970. Light. In O. Kinne (editor). Marine ecology. Vol. 1, Environmental factors, Part 1, p. 213-320. Wiley-Inter- science, Lond. BLAXTER, J. H. S., AND B. B. PARRISH. 1965. The importance of light in shoaling, avoidance of nets and vertical migration by herring. J. Cons. 30:40- 57. BRAWN, V. M. 1960. Seasonal and diurnal vertical distribution of herring (Clupea harengus L.) in Passamaquoddy Bay, N.B. J. Fish. Res. Board Can. 17:699-711. CLAVIJO, E. J. 1983. Pair spawning and formation of a lek-like mating system in the parrotfish Scarus vetula. Copeia 1983:253-256. CLUTTER, R. I. 1967. Zonation of nearshore mysids. Ecology 48:200-208. DEMARTINI, E. E., AND R. K. FOUNTAIN. 1981. Ovarian cycling frequency and batch fecundity in the queenfish, Seriphus politus: Attributes representative of serial spawning fishes. Fish. Bull., U.S. 79:547-560. DOMM, S. B., AND A. J. DOMM. 1973. The sequence of appearance at dawn and disappear- ance at dusk of some coral reef fishes. Pac. Sci. 27:128- 135. FEDER, H. M., C. H. TURNER, AND C. LIMBAUGH. 1974. Observations on fishes associated with kelp beds in southern California. Calif Dep. Fish Game, Fish Bull. 160, 144 p. FREY, H. W. (editor). 1971. California's living marine resources and their utili- zation. Calif Dep. Fish Game, Mar. Res. Agency, 148 P- HELFMAN, G. S. 1978. Patterns of community structure in fishes: summary and overview. Environ. Biol. Fish. 3:129-148. HELFMAN, G. S., J. L. MEYER, AND W. N. MCFARLAND. 1982. The ontogeny of twilight migration patterns in grunts (Pisces:Haemulidae). Anim. Behav 30:317-326. HICKLING, C. F 1933. The natural history of the hake. Part IV. Age- determination and the growth rate. Fish. Invest., Minist. Fish. Agric. Food (G.B.) Sen 2, 13 (2):1-120. 184 DeMARTlNI ET AL: DISTRIBUTION PATTERNS OF QUEENFISH HOBSON, E. S. 1965. Diurnal-nocturnal activity of some inshore fishes in the Gulf of California. Copeia 1965:291-302. 1973. Diel feeding migrations in tropical reef fishes. Hel- gol. Wiss. Meeresunters. 24:361-370. 1978. Aggregating as a defense against predators in aquatic and terrestrial environments. In E. S. Reese and F. J. Lighter (editors), Contrasts in behavior, p. 219- 234. Wiley, N.Y. HOBSON, E. S., AND J. R. CHESS. 1973. Feeding oriented movements of the atherinid fish Pranesus pinguis at Majuro Atoll, Marshall Islands. Fish. Bull., U.S. 71:777-786. 1976. Trophic interactions among fishes and zooplankters near shore at Santa Catalina Island, California. Fish. Bull., U.S. 74:567-598. 1978. Trophic relationships among fishes and plankton in the lagoon at Enewetak Atoll, Marshall Islands. Fish. Bull., U.S. 76:133-153. HOBSON, E. S., W. N. MCFARLAND, AND J. R. CHESS. 1981. Crepuscular and nocturnal activities of Californian nearshore fishes, with consideration of their scotopic vis- ual pigments and the photic environment. Fish. Bull., U.S. 79:1-30. JUNE, F C. 1972. Variations in size and length composition of Atlantic menhaden groupings. Fish. Bull., U.S. 70:699-713. LARSON, R. J., AND E. E. DEMARTINI. 1984. Abundance and vertical distribution of fishes in a cobble-bottom kelp forest off San Onofre, California. Fish. Bull., U.S. 82:37-53. LUCAS, C. E. 1936. On the diurnal variation of size-groups of trawl- caught herring. J. Cons. 11:53-59. MILLER, D. J., AND R. N. LEA. 1972. Guide to the coastal marine fishes of California. Calif Dep. Fish Game, Fish Bull. 157, 235 p. MOVER, J. T, AND Y. YOGO. 1982. The lek-like mating system of Halichoeres melanochir (Pisces:Labridae) at Miyake-jima, Japan. Z. Tierpsychol. 60:209-226. PARRISH, B. B., J. H. S. BLAXTER, AND W. B. HALL. 1964. Diurnal variations in size and composition of trawl catches. Rapp. P-v Reun. Cons. Perm. int. Explor. Mer 155:27-34. PINKAS, L., M. S. OLIPHANT, AND I. L. K. IVERSON. 1971. Food habits of albacore, bluefin tuna, and bonito in California waters. Calif Dep. Fish Game, Fish Bull. 152, 105 p. PLUMMER, K. M., E. E. DEMARTINI, AND D. A. ROBERTS. 1983. The feeding habits and distribution of juvenile-small adult California halibut (Paralichthys callfornicus) in coastal waters off northern San Diego County. CalCOFI Rep. 24:194-201. QUINN, T P, B. S. MILLER, AND R. C. WINGERT. 1980. Depth distribution and seasonal and diel movements of ratfish, Hydrolagus colliei, in Puget Sound, Washing- ton. Fish. Bull., U.S. 78:816-821. SCOFIELD, W. L. 1951. Purse seines and other roundhaul nets in Califor- nia. Calif Dep. Fish Game, Fish Bull. 81, 83 p. SlEGEL, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill Book Co., 312 p. WILSON, K. V. 1956. A distribution-free test of analysis of variance hypotheses. Psychol. Bull. 53:96-101. WOODHEAD, R M. J. 1964. Diurnal changes in trawl catches of fishes. Rapp. P-v. Reun. Cons. int. Perm. Explor Mer 155:35-44. 1966. The behavior of fish in relation to light in the sea. Oceanogr. Mar. Biol. Annu. Rev 4:337-403. YOUNG, R H. 1963. The kelp bass iParalabrax clathratus ) and its fishery, 1947-1958. Calif Dep. Fish Game, Fish Bull. 122, 67 P- 185 REACTION OF DOLPHINS TO A SURVEY VESSEL: EFFECTS ON CENSUS DATA Roger P HEvmT^ ABSTRACT A field experiment is described in which a helicopter was used to observe the efficiency of shipboard line-transect sampling of dolphin populations in the eastern tropical Pacific Ocean. Nineteen dolphin schools were tracked; 13 of these were detected by observers aboard the ship and 5 of these reacted to the approach of the ship by altering the direction and/or the speed of their movement; however, only 1 school reacted prior to shipboard detection. The results suggest that dolphin schools only occasionally react to the approach of a survey vessel prior to their detection by shipboard observers and that the use of a monotonically decreasing detection function is adequate to minimize bias. Aerial and shipboard estimates of school size and species composition for six schools compared favorably. The Marine Mammal Protection Act of 1972 pro- hibits the harvest of marine mammals and specifies that the Federal Government may issue permits for their take only under special cir- cumstances. One such circumstance involves the incidental kill of dolphins associated with the yel- lowfin tuna fishery in the eastern tropical Pacific Ocean. Before issuing the permits, the govern- ment must first certify the viability of the affected dolphin populations. To meet this requirement, scientists at the Southwest Fisheries Center define stocks and monitor their population demography, reproductive output, and abundance. The vital statistics are derived primarily from specimens obtained from the tuna fishery. How- ever, to estimate abundance, surveys are con- ducted using ships and aircraft independently of the fishery. The surveys, using line-transect methods (Burnham et al. 1980), have yielded esti- mates of the density of dolphins in the eastern tropical Pacific Ocean (Holt and Powers 1982). A critical assumption in the application of the method is that the animals do not move, in reac- tion to the observer, prior to their detection. In practice, a detection function, which is relatively insensitive to nonrandom movement, is used to describe the probability of observing a school of dolphins given its position relative to the ob- server's transect. A field experiment. was designed with the following objective: ^Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, EO. Box 271, La Jolla, CA 92038. Msinuscript accepted April 1984. FISHERY BULLETIN: VOL. 83. NO. 2, 1985. 1) Test the assumption that the animals do not alter their movement in reaction to the ap- proach of a survey vessel prior to shipboard detection. During a survey the unit of observation is a school of dolphins. In addition, species composition and the number of individual animals in a school (school size) are estimated. Surveys routinely col- lect information to determine the precision of these estimates by recording independent obser- vations of several observers; however, determining their accuracy is more difficult and attempted less often (Holt and Powers 1982). Six schools Were closely approached and observed from both an air- craft and a ship with the following objective: 2) Compare shipboard and aerial estimates of school size and species composition. Although not an absolute determination of accu- racy, the comparison yielded estimates from two very different viewpoints (high-altitude plan view versus low-altitude profile view). A similar experiment was conducted using the NOAA Ship Surveyor and a ship-supported helicopter in 1977 (Au and Perryman 1982). They observed the reaction of eight dolphin schools to the approach of a ship; all eight schools swam away from the projected trackline of the ship. Au and Perryman also suggested that, in some cases, avoidance began beyond the visual range of ship- board observers. The present study was intended to collect additional data under a wider variety of conditions. 187 FISHERY BULLETIN: VOL. 83, NO. 2 METHODS The experiment was designed to observe the efficiency of shipboard survey operations by using a helicopter to track dolphin schools before, dur- ing, and after shipboard detection. This approach was an enhancement of the design employed by Au and Ferryman (1982) which focused only on the behavior of the dolphins. A simulated survey op- eration was included in the experiment for the following reasons: 1) It was not reasonable to assume that move- ment of a dolphin school and the probability of detecting it are unrelated (i.e., it may be easier to see a school in full flight than one at rest). Therefore, associated data on move- ment and shipboard detection were collected for each school. 2) It was necessary to separate random move- ment from directed movement toward or away from the survey vessel. To do so unam- biguously, the ship could not be directed to- ward a school detected by the helicopter, but rather had to continue searching along a predetermined transect. were tracked for about an hour's time until one of three events occurred: 1) the school passed abeam of the ship; 2) the school passed beyond the visual range of shipboard observers; or 3) the aer- ial observers lost sight of the school and had to terminate the track prematurely; in all of the latter cases the presence of the animals was obscured by deteriorating sea state. During a track, the helicopter was positioned over the school at a minimum altitude of 1,200 ft (370 m); the radar range and bearing to the helicopter were determined from the approaching survey vessel about every 4 min (an interval suffi- cient to record the appropriate navigational data and still provide continuity in the track). A tran- sponder, mounted on the aircraft, facilitated accu- rate radar measurements. In addition, OMEGA navigation positions were recorded from dual sys- tems aboard the helicopter and the ship. As the track progressed, field notes were taken on visual observations of school behavior and associated birds and fish. The tracking altitude appeared to SHIP TRANSECT t From the experience gained on the 1982 survey (Holt 1983), we expected 80% of the sighting cues to be within 3 nmi of the transect line and <5 nmi ahead of the vessel. Furthermore, the Au and Fer- ryman observations on eight schools suggested that dolphins may react to a ship 6 nmi away. With these considerations and prior experience in mind, the following field procedure was employed. The ship proceeded at 12 kn in a direction selected so as to minimize glare from the sea sur- face. Two observers maintained constant watch through 25 power binoculars, mounted on the port and starboard sides of the flying bridge (11m above the water); search patterns extended from the bow to the beam of the ship on each side. Records were kept of searching effort and sighting details. With the exception of selecting the transect direction, these are the same methods employed during pre- vious dolphin surveys (e.g.. Holt and Fowers 1982). The helicopter searched a distance of 8 to 12 nmi ahead of the ship and 2 nmi to either side of the transect line, at right angles to the direction of the ship's travel (Fig. 1). Search altitude was 1,200 ft and speed was 60 kn. When a school was sighted by the helicopter, shipboard radar tracking began. The observers on the flying bridge were not aware of a track in progress until its termination. Schools I HELICOPTER PATH j-2 nm.-J '8-12 nm. SHIP Figure l. — Port and starboard search patterns (shaded areas) and path of helicopter (sohd Hne) during transect (dashed line) surveys for dolphins. 188 HEWITT DOLPHINS" REACTION TO A SURVEY VESSEL be sufficient so as not to elicit a response from the animals. The dolphins appeared to be swimming calmly throughout the tracking; similar experi- ence was reported by Au and Ferryman (1982). It also placed the helicopter above the shipboard ob- servers' vertical field of vision and therefore did not prematurely cue them on a school. Two oil drums were released and tracked at the beginning of the cruise to test the procedure: The resolution of radar measurements was 1-2° in bearing and 0.1 nmi in range; at 1,200 ft (370 m) altitude we were able to maintain visual contact with aim object; and the shipboard observers were not aware of the helicopter until it was within 1 nmi of the ship, where the noise signaled its presence. Shipboard observers were questioned periodically through- out the experiment as to their cognizance of the helicopter; answers were always in the negative except when the binoculars were purposefully di- rected above the searching field. Observers were aware that looking for the helicopter would com- promise the experiment and did not do so. At the finish of a track, the helicopter descended to a lower altitude for additional photography and to estimate school size and species composition. The ship approached a limited number of schools to enable close-range shipboard estimates of the same school parameters. After school size and species composition were determined, normal sur- vey operation resumed, with the helicopter search- ing ahead of the vessel and the shipboard observers actively scanning and recording search effort. Relative motion radar plots were maintained. Apparent change in the relative direction of dol- phin school movement was used as an indication of avoidance; field notes of aerial observations of be- havior supplemented this information. The criteria defining reaction was a change of 30° or more in the direction of relative motion that was sustained over 2 or more subsequent fixes (Fig. 2). The experimental design was opportunistic and only specifically designed to compare between a steam-powered survey vessel (NOAA Ship Sur- veyor) and a diesel-powered survey vessel (NOAA Ship David Starr Jordan). The experiment was conducted within a 100 square nmi area to the north and east of Clipperton Island (lat. 10°N, long. 110°W) during March and April 1983.^ Ob- servations were conducted with the Surveyor from 10 March through 17 March; the ship then ported at Manzanillo, Mexico, to take on fuel and sub- sequently met the David Starr Jordan, which had just completed a marine mammal survey^ on 26 March at Clipperton Island. Observations were conducted in the same area with the David Starr Jordan until 7 April. RESULTS Avoidance Tracks were started on a total of 26 dolphin schools, 5 in front of the Surveyor and 21 in front of ^Cruise Report NOAA Ship Surveyor Cruise RP-12-SU-83 dated May 24, 1983, on file at the Southwest Fisheries Center, National Marine Fisheries Service, NOAA, P.O. Box 271. La Jolla, CA 92038. ^Cruise Report NOAA Ship David Starr Jordan Cruise DS- 83-01 dated May 6, 1983, on file at the Southwest Fisheries Center, National Marine Fisheries Service, NOAA, RO. Box 271, La Jolla, CA 92038. 300 270' SCHOOL 8 Visual cue (birds) at 1306 Dolphins sighted at 1306 300 090° 270° SCHOOL 23 Visual cue (birds) at 0926 Dolphins sighted at 0954 -090° Figure 2. — Relative motion plots of dolphin school #8 and school #23. School #8 appeared to react to the approach of the ship; the sighting cue was reported after the dolphins' initial reaction. School #23 did not appear to react to the survey vessel. 189 FISHERY BULLETIN: VOL. 83, NO. 2 the David Starr Jordan; a summary of the obser- vations is listed in Table 1. Seven of the tracks were terminated prematurely, and of the remaining 19, 6 schools passed undetected by shipboard observ- ers. These 6 schools did not appear to adjust their direction of movement in reaction to the survey vessel. Of the 13 schools sighted by ship, 1 school altered its direction of movement in reaction to the ap- proaching ship, prior to the detection of a sighting cue by the shipboard observers, and 12 schools did not appear to react before detection by the ship. One of the 12 schools was composed of rough toothed dolphins, Steno bredanensis, which are not a target of abundance surveys. Thus, from the results of this experiment, it is expected that 8% ( V12) of the target schools encountered on a survey will have moved (in reaction to the observer) prior to detection. This does not imply a corresponding degree of survey bias. Nonrandom movement, prior to detection, will alter the distribution of sighting distances and the detection function fit to the distribution; the survey will be biased to the extent that the functional form is sensitive to the data (see Discussion). Survey bias may also exist as a result of schools that react to the ship and are subsequently never seen by shipboard observers; if these schools would have been ob- served (the expectation is certain if they are on the transect line, less certain if they are off the line), then the bias is proportional to the fraction of schools that escaped detection. As stated above, no schools were observed to react to the ship and avoid detection. The data suggest that dolphin schools may alter their direction of movement in reaction to the ap- proach of a survey vessel. Thirty-eight percent (V13) of the schools which were tracked by helicop- TabLE L — Summary of dolphin school tracking data. Interpolated Number Closest Radial radar position Beaufort of point of Reaction sighting at time of School sea Indivi- approach distance distance Relative Sighting sighting Vessel number state Species composition duals (nml) (nml) (nml) bearing cue (range/bearing) Surveyor 1 1 Sfeno bredanensis 100% 9 1.3 F' 2.5 317" animals 1.8/335° 2 1 Stenella attenuata S. longirostns 50% 50% 175 7.0 F' F2 3 3 S. attenuata 100% 53 2.5 F' F2 4 5 Unidentified dolphins 100% 100 2.0 F' 2.0 030 splashes 4 0/032° 5 5 Unidentified dolphins 100% 15 F3 Jordan 6 4 Unidentified dolphins 100% 22 F3 7 4 Unidentified dolphins 100% 35 F3 8 4 S. attenuata S. longirostns Unidentified dolphins 25% 5% 70% 300 1.5 2.5 1.5 024' birds 1.6/030° 9 4 Unidentified dolphins 100% 25 F3 10 4 S. attenuata S. longirostris 20% 80% 150 0.5 1.7 6.0 003° birds 6.3/002° 11 4 S. attenuata 100% 25 5.0 F' 6.8" 023 = birds 7.2/019° 12 4 S. attenuata S. longirostris 1 5% 85% 65 7.0 F' F2 13 4 S. attenuata S. longirostris 65% 35% 175 1.3 2.2 6.8 356° birds 6.2/357° 14 4 S. attenuata S. longirostris 90% 10% 50 2.5 F' 6.8" 000° birds 8.1/354° 15 4 Stenella spp 100% 150 F3 16 4 S. attenuata 100% 35 1.2 1.5 6.0 357° birds 7.0/359° 17 3 Unidentified dolphins 100% 40 F3 18 3 S. coeruleoalba 100% 160 F3 19 3 S. attenuata 100% 45 3.0 F' F2 20 0 S. attenuata S. longirostns 15% 85% 260 1.7 F' 6.8" 355° birds 6.7/353° 21 2 S. attenuata S. longirostns 91% 9% 230 6.4 F' F2 22 1 S. attenuata S. longirostris 50% 50% 180 2.1 2.1 6.8" 340° birds 6.7/336° 23 1 S. attenuata S. longirostns 50% 50% 155 1.5 F' 6.8" 004° birds 8.0/357° 24 1 S. coeruleoalba 100% 29 01 F' 1.8 020° animals 1.8/018° 25 1 S. attenuata S. longirostris 40% 60% 410 2.0 F' 4.0 015° birds 50/010° 26 1 S. attenuata 100% 85 3.0 F' F2 'School did not appear to react to the approach of the survey vessel. 'School passed undetected by shipboard observers ^Track prematurely tenninated. "Cue observed on the horizon. 190 HEWITT DOLPHINS' REACTION TO A SURVEY VESSEL ter and detected by shipboard observers appeared to react to the ship. Spotted dolphins, Stenella attenuata, and spinner dolphin, S. longirostris, reacted at a distance of 0.5 to 2.5 nmi and were able to maintain a separation of 0.5 to 2.0 nmi from the ship; one school of striped dolphins, S. coeruleoalba. was successfully tracked and these animals stayed on a collision course with the ship until they were only a few hundred meters away. In all cases but one (school 8), the schools were detected by shipboard observers at distances far greater than the reaction distance. None of the four dolphin schools successfully tracked in front of the Surveyor appeared to react to the approach of the ship. Five out of 15 schools appeared to react to the approach of the David Starr Jordan. Estimates of School Size and Species Composition Six schools were approached at close range by the David Starr Jordan so that shipboard observ- ers could make estimates of school size and species composition using the same techniques that were used on previous abundance surveys. Estimates of school size and species composition were made in- dependently by four to six shipboard observers and averaged, giving each an equal weight. These es- timates compared favorably with estimates made by a single aerial observer stationed in the heli- copter (Table 2 ). Shipboard estimates of school size ranged from 65 to 134% of the aerial estimates and averaged 101% (mean difference = 1.167; Pr = 0.713, paired t test of mean difference = 0); ship- board and aerial observers agreed on the species composition for all six schools compared, although there was some variation in the proportion as- signed to each species. DISCUSSION The density estimator used in line-transect ap- plications, formally derived by Burnham and An- derson (1976), and used to estimate the density of dolphin schools by Smith (1981) and Holt and Pow- ers (1982), is: D Nf(0) 2L where D is the estimated density of dolphin schools in the survey area based on the number of schools observed, A^, over transect length L. The function fix) is a probability density function fit to the observed perpendicular sighting distances and estimating its value at zero distance, /"(Oi, is the critical concern in the application of line-transect methods (Burnham et al. 1980). The frequency distribution of observed perpen- dicular sighting distances reflects both the detec- tion abilities of the observer and the reactions of the observed (Burnham et al. 1980). Dolphin schools are more difficult to see with distance from the track line and avoidance, prior to detection, may cause fewer schools to be seen close to the track line and more schools to be seen further from the track line. The school that did move away from the transect line before shipboard detection (#8) would have been sighted at 0.1 nmi off the transect line if it had not altered the direction of its move- ment. Instead it was detected at 1.0 nmi off the transect line. If the sample size was larger, such information could be used to dissect the frequency distribution of perpendicular sighting distances into that component which is the result of decreas- ing visibility with distance from the transect line and that component which is the result of dolphin schools adjusting their natural spatial disposition Table 2. — Comparison of shipboard and aerial estimates of dolphin school size and species composition. Vessel estimate Helicopter estimate Number Estimated number Number Estimated School of of individuals Species proportions of number of number observers (standard error) (range) observers individuals Species proportions 20 5 248 (24) S attenuata S longirostris 0 14(0 05-0.20) 0.86 (0 80-0.95) 1 260 S attenuata 0.15 S longirostns 0 85 22 4 241 (40) S. attenuata S. longirostris 0.96(0.90-1.00) 0.04 (0.00-0 10) 1 180 S attenuata 0.50 S longirostris 0.50 23 4 139(20) S. attenuata S. longirostris Unidentified 0 62 (0 50-0.73) 0.35 (0.22-0 50) 0.03(0.00-0.12) 1 155 S attenuata 0.50 S. longirostris 0.50 24 6 36(6) S. coeruleoalba 1 1.00(1.00-1.00) 1 29 S coeruleoalba 1.00 25 5 393(61) S. attenuata S. longirostris Unidentified 0.55 (0.40-0.70) 0.39 (0.30-0.60) 0.06 (0.00-0.30) 1 410 S. attenuata 0.40 S. longirostris 0.60 26' 5 55(9) S. attenuata 1.00(1.00-1.00) 1 85 S. attenuata 1 .00 'Not detected by shipboard observers while In survey mode; ship was directed to school by aerial observer. 191 FISHERY BULLETIN: VOL. 83, NO. 2 in response to the ship. There are, however, other factors (such as glare and sea state) which are seldom constant long enough to allow for accumu- lation of a reasonably precise frequency distribu- tion, such that the effects due to school movement would not be overwhelmed by the effects due to sighting conditions. The results of this experiment suggest that 1) dolphin schools occasionally react to the approach of a survey vessel prior to their detection by ship- board observers and 2) the expected rarity of the event implies that a considerable amount of addi- tional data would be required to quantify its effect. Any directed movement prior to detection biases the frequency distribution of perpendicular dis- tances and may bias the function, fix), fit to these data. In the absence of information regarding movement, Burnhametal. (1980) suggested choos- ing a function which is relatively insensitive to data contaminated by movement, i.e., a function that monotonically decreases with distance from the transect line. Their simulations suggest that in situations where "undetected movement is rela- tively minor, then use of an estimator based on a monotonically decreasing function will minimize bias in D," (Burnham et al. 1980:130). The small sample size of the present experiment was suffi- cient to qualify undetected movement as relatively minor but not sufficient to quantify its effect on the distribution of perpendicular distances. Although the work reported here was conducted in the same geographic area (Clipperton Island, lat. lO'N, long. 110 W) as the Au and Ferryman (1982) observations, the two experiments are not strictly comparable. Au and Ferryman used the ship and helicopter to search for schools and col- lected data on their reaction to the ship without regard to the effect on survey operations; in four of the eight schools they studied, the ship was turned toward the school during tracking. They were in- terested in describing the behavior of dolphin schools and combining the description with a search model to quantify survey bias. The present experiment did not assume that the two processes (reaction and detection) were independent and was less ambitious because there was no intention to generalize dolphin behavior Indeed, the results presented here may only be relevant to this area and for these sighting conditions. Both the reac- tion distance and the sighting distance may be affected by environmental conditions and may vary between geographic areas with the degree of animal naivete. The comparisons of aerial and shipboard results suggest that school-size estimates may be more reliable than those of species composition. Al- though neither observation platform can be con- sidered to yield estimates without error, they do provide unique vantage points with very different views of the dolphin school. All shipboard observ- ers, after exposure to observation conditions in the helicopter, agreed that they could more confi- dently estimate school size from the air than from a vessel. The helicopter provides an opportunity to observe the entire school over an extended period of time, making it easier to estimate that portion of the school which is submerged and not com- pletely visible. Species proportions are more diffi- cult to estimate and it is not clear which platform is better; indeed, in the case of school 22, all four shipboard observers reported similar proportions which were quite different than that estimated from the air One explanation may be that it is more difficult to identify animals in plan view than in profile view; alternately, the fluid charac- ter of school structure may combine with the lim- ited view of the school from a ship to preclude accurate estimates of species proportions; a third possibility is that both are inaccurate because of species-specific behaviors which make the ani- mals less visible from above and/or the side. Estimates of the density of dolphin schools are multiplied by the area of the survey, the average school size and the species proportions to estimate species abundances (Holt and Fowers 1982). Be- cause they affect the abundance estimates di- rectly, biases in the latter two parameters may be more serious than the effect of school movement prior to detection. As an example, consider the six schools compared during this experiment: the av- erage number of S. atten uata per school, estimated by shipboard observers, was 2T^'( greater than that estimated from the helicopter data, the shipboard estimate of S. longirostris was 34% less than the helicopter estimates, and the estimate of S. coeru- leoalba was the same for both platforms (Table 3). Although these differences should only be consid- ered as variability between two estimates, they illustrate the direct dependence of abundance es- timates on accurate estimates of species propor- tions. Avoidance affects density estimates less dramatically; its affect on /'(O) may be somewhat offset by using a function that is relatively insen- sitive to predetection movement. The application of line-transect methods re- quires that along the transect line all schools are seen with certainty. Any departures from the as- sumption of perfect detection, either because of 192 HEWITT DOLPHINS' REACTION TO A SURVEY VESSEL Table 3. — Average dolphin school composition. Vessel Helicopter Average school size (number of individuals) Average species proportions S. attenuata 185.3 0.545 186.5 0.425 S. longirostris S coeruleoalba 0.273 0.167 0.408 0.167 Unidentified dolptiin Average school composition' S. attenuata 0.015 102.5 79.3 S. longirostris S. coeruleoalba 51.3 31.5 76.1 31 1 'Unidentified dolphins distributed proportionately among identified dolphins following Holt and Pow/ers (1982). generously gave their time and advice during the design of the experiment; and the conscientious and competent field observers included G. Fried- richsen in the helicopter, A. Jackson, W. Irwin, and M. Noel on the NOAA ship Surveyor, and J. Cotton, J. Doxey, M. Henry, M. Graybill, R. Pitman, and G. Yee lead by W. Parks on the NOAA Ship David Starr Jordan. R. Holt, T. Jackson, W. Perrin, and P Vergne reviewed the manuscript. The final draft benefited from the comments of two anonymous reviewers. movement or visibility effects, will introduce a negative bias in the density estimate that is pro- portional to the decrease in apparent density along the transect line (Smith 1979). The sample size was insufficient to test this assumption rigor- ously; only one school was observed on the transect line (school 14) and it was detected well beyond any of the reaction distances observed. It is recommended that future fieldwork include additional comparisons of estimates of school size and species proportions. In addition, the assump- tion of certain detection along the transect line should be tested. Biases in school composition and detection on the transect line affect the abundance estimates directly and present a greater potential for inaccuracy than the degree of directed move- ment prior to detection observed during this experiment. ACKNOWLEDGMENTS This work was accomplished with the help and collaboration of several people: D. Au, D. Chap- man, P Hammond, J. Laake, and W. Perryman LITERATURE CITED Au, D., AND W. L. PERRYMAN. 1982. Movement and speed of dolphin schools responding to an approaching ship. Fish. Bull. U.S. 80:371-379. BURNHAM, K. P., AND D. R. ANDERSON. 1976. Mathematical models for nonparametric inferences from line transect data. Biometrics 32:325-336. BURNHAM, K. P., D. R. ANDERSON, AND J. L. LAAKE. 1980. Estimation of density from line transect sampling of biological populations. J. Wildl. Manage. Monogr. 72, 202 p. HOLT, R. S. 1983. Report of eastern tropical Pacific research vessel marine mammal survey, May 15- August 3, 1982. U.S. Dep. Commer., NOAA Tech. Memo. NMFS-SWFC-29, 151 p. HOLT, R. S., AND J. E. POWERS. 1982. Abundance estimation of dolphin stocks involved in the eastern tropical Pacific yellowfin tuna fishery deter- mined from aerial and ship surveys to 1979. U.S. Dep. Commer., NOAA Tech. Memo. NMFS-SWFC-23, 95 p. SMITH, G. E. J. 1979. Some aspects of line transect sampling when the target population moves. Biometrics 35:323-329. SMITH, T. D. 1981. Line transect techniques for estimating density of porpoise schools. J. Wildl. Manage. 45:650-657. 193 FIN EROSION AMONG FISHES COLLECTED NEAR A SOUTHERN CALIFORNIA MUNICIPAL WASTEWATER OUTFALL (1971-82)* Jeffrey N. Cross ^ ABSTRACT In the Southern Cahfornia Bight, fin erosion is most frequently encountered among fishes collected near municipal wastewater outfalls. This paper presents an analysis of the trends in the incidence of fin erosion among fishes collected by otter trawls near Los Angeles from 1971 through 1982. About 24% of the 122 species offish and 9% of the more than 170,000 individuals collected had the disease. Flatfish (Pleuronectidae, Bothidae, and Cynoglossidae) and rockfish (Scorpaenidae) accounted for 66% of the affected species and 99% of the affected individuals. Dover sole (Pleuronectidae: Microstomus pacificus ) accounted for 89% of the affected individuals. The incidence of fin erosion was highest close to the outfalls and declined with increasing distance. The number of species with the disease declined from 1971 to 1982. The incidence of the disease also declined in two of the three most affected species (Dover sole and rex sole, Glyptocephalus zachirus). The contemporaneous decline in the number of species and the proportion of individuals with the disease and the decline in surface sediment contaminant levels suggest that the magnitude of contami- nation and the incidence of fin erosion are directly related. The effect of fin erosion on the Dover sole population was examined. Dover sole recruit to the study area when they are 40-50 mm SL; the incidence of fin erosion was negligible in new recruits but increased rapidly with increasing fish size. No significant differences were detected in the length- weight relationships or size-at-age data between Dover sole with and without the disease. Survival rates of Dover sole with and without fin erosion were similar until age 3; thereafter, the survival rate of diseased fish was significantly lower. Fin erosion in the Southern California Bight is most prevalent in fish collected near major munic- ipal wastewater outfalls. Of the major outfalls, the disease is most frequently encountered around the Joint Water Pollution Control Project (JWPCP) outfalls on the Palos Verdes shelf (Mearns and Sherwood 1974, 1977; Sherwood and Mearns 1977). About 20% of the 151 species offish collected in the Southern California Bight between 1969 and 1976 were affected by the disease. Flatfish (Pleuronec- tidae, Bothidae, and Cynoglossidae) and rockfish (Scorpaenidae) accounted for 60% of the affected species and 97% of the affected individuals (Mearns and Sherwood 1977; Sherwood 1978). Municipal wastewater discharge on the Palos Verdes shelf began in the late 1930's. The mass emission of suspended solids increased steadily from about 17,000 metric tons (t) in the early 1940's to a peak of 167,000 t in 1971 (Wilson et al. 1980). Awareness of the problem of discharging excessive amounts of solids and associated con- 'Contribution No. 196, Southern California Coastal Water Re- search Project. ^Southern California Coastal Water Research Project, 646 W. Pacific Coast Highway, Long Beach, CA 90806. taminants by JWPCP in the early 1970's resulted in the installation of new equipment and more effective treatment procedures to reduce solid emissions. By 1981, discharge of suspended solids had declined to 84,000 t. The mass emission of contaminants also declined from 1971 to 1981 (Schafer 1982). Otter trawling, as part of a regular monitoring program of the animals on the Palos Verdes shelf, began in 1971. The purpose of this paper is to determine trends in the incidence of fin erosion among fish collected by otter trawls near the JWPCP outfalls from 1971 through 1982. The prevalence of the disease over the size range of individuals collected and the ef- fect of the disease on growth and survival are examined for the most affected species. METHODS The data analyzed in this study were collected by the Los Angeles County Sanitation District (LACSD) during regular monitoring cruises on the Palos Verdes shelf; station and transect identifica- tions used herein are LACSD designations. Day- time trawls were made at three depths (23, 61, and Manuscript accepted April 1984. FISHERY BULLETIN: VOL. 83, NO. 2, 1985. 195 FISHERY BULLETIN: VOL. 83, NO. 2 137 m) at each of seven transects (Fig. 1) with an otter trawl towed along a depth isobath at 1.1 m/s for 10 min. A 7.3 m (headrope length) trawl was used from 1971 to 1974, when it was replaced by a 7.6 m net; a 1.25 cm mesh cod end liner was used in both nets. From 1971 through 1978, two samples — one between April and June and one between October and December — were collected annually at each depth; occasionally, additional trawls were made and these were included in the analyses. Quarterly trawling began in 1979 and has continued to the present. Sampling was dis- continued at transects T2, T3, and T6 in 1977. Trawl catches were sorted by species and pro- cessed on board ship. The standard length of each individual was determined on a measuring board (BSL = board standard length). External abnormalities (e.g., fin erosion, parasites, tumors) were recorded along with length. Trends in the incidence of fin erosion per 10-min trawl were determined from linear regressions of the proportion ip) offish with the disease (trans- formed to arcsin \ p) versus time [numbered in consecutive months from 1971 through 1982 (i.e., 1, 2, 3, ..., 143, 144)]. Collections where only one individual of the species of interest was caught were dropped from the analysis because the result- ing transformed datum (either 0 or 90) often had a large effect on the residual sum of squares. Trends in the total catch per 10-min trawl were deter- mined from linear regressions of numbers [transformed to logio (x -I- D] versus time. The null hypothesis that the regression function (slope) was equal to zero was tested with a ^-test. The null hypothesis of equality of the regression functions was tested by analysis of covariance (ANCOVA). If the null hypothesis was not accepted, significant differences were detected with a Newman-Keuls multiple range test (Zar 1974). Calico rockfish, Sebastes dalli, rex sole, Glyptocephalus zachirus, and Dover sole. Microstomas pacificus, accounted for 96.6% of the fish with fin erosion (Table 1) so the trend analysis was performed on each species in- dividually. Less than 1% of the fin eroded fish came from the 23 m stations; these stations were then dropped from the analyses. The quarterly trawl data (1979-82) from tran- sects T4 and T5 were examined for seasonal trends in the total catch of Dover sole, the number with fin erosion, and the proportion with fin erosion with the following model: Y,=fiT^,S,) + e^ where Y = observed abundance (or proportion) in period t,T^ = trend factor of time series in period t, S^ = seasonal factor of time series in period t, f = function relating observed abundance (or pro- portion) to the trend and seasonal components, and e^ = irregular factor of time series in period t (Bowerman and O'Connell 1979). Multiplicative and additive models were fitted to the quarterly trawl data after transformation [logio (x -I- 1) and arcsin \ p ]. Multiplicative models gave the best fit (lowest residual sum of squares) so only those re- sults are presented. After determining the quar- terly trends, the original transformed data were "deseasonalize'd" by dividing each value by the corresponding seasonal factor. Trends in the de- X .'b'^ <^ % ^o. Figure l. — Location of sampling transects on the Palos Verdes shelf. Three depths (23, 61, and 137 m) were sampled at each transect. Joint Water Pollution Control Project outfalls are located between transects T4 and T5. Net current flow is northwest. 196 CROSS: FIN EROSION AMONG FISHES Table l. — Taxonomic list of fish affected with fin erosion collected in 622 otter trawls on the Palos Verdes shelf from 1971 through 1982. Frequency of occurrence (%) Percent of Number in trawl Percent with all fish with Common name Scientific name collected collections fin erosion fin erosion spotted cusk-eel Chilara laylori 191 16.7 0.5 <0.1 blackbelly eelpout Lycodopsis pacifica 2.629 20.6 <0.1 <0.1 shortspine thornyhead Sebastolobus alascanus 312 4.3 1.9 <0.1 calico rockflsh Sebastes dallii 9.153 23.6 9.9 5.8 shortbelly rockfish Sebastes jordani 3,247 25.1 0.2 <0.1 stripetail rockfish Sebastes saxicola 18,938 54.5 <0.1 <0.1 Vermillion rockfish Sebastes miniatus 382 17.1 0.5 <0.1 pink rockfish Sebastes eos 37 1.1 5.4 <0.1 greenstriped rockfish Sebastes elongatus 263 13.8 2.3 <0.1 splitnose rockfish Sebastes diploproa 6,973 24.0 <0.1 <0.1 sablefish Anoplopoma fimbria 711 15.6 0.3 <0.1 shortspine combfish Zaniolepis frenata 794 20.6 0.1 <0.1 longspine combfish Zaniolepis latipinnis 891 20.7 0.4 <0.1 barred sand bass Paralabrax nebulifer 40 4.0 2.5 -0.1 white croaker Genyonemus lineatus 9,062 20.9 1.6 0.9 white seaperch Phanerodon furcalus 848 12.1 0.2 <0.1 shiner perch Cymatogaster aggregate 9,478 27.1 <0.1 <0.1 Pacific pompano Peprilus simillimus 30 2.0 20.0 •'0.1 California tonguefish Symphurus atricauda 1,091 25.4 1.7 0.1 Pacific sanddab Citharichttiys sordidus 11,698 61.3 0.5 0.4 speckled sanddab Citharichthys stigmaeus 15,491 39.6 <0.1 • 0.1 bigmouth sole Hippoglossina stomala 148 14.6 2.1 <0.1 C-0 sole Pleuronichthys coenosus 187 10.8 1.1 <0.1 curlfin sole Pleuronichthys decurrens 3,774 40.4 2.4 0.6 hornyhead turbot Pleuronichthys verticalis 458 25.0 1.1 0.1 English sole Parophrys vetulus 1,256 44.7 0.9 <0.1 rex sole Glyptocephalus zachirus 4,452 28.0 6.7 1.9 slender sole Lyopsetta exilis 4,674 26.0 3.3 1.0 Dover sole Microstomus pacificus 41.627 62.2 33.5 88.9 seasonalized data were determined by linear re- gression. The effects of fin erosion on the population of Dover sole on the Palos Verdes shelf were examined using the original data of Mearns and Harris (1975)^ that consisted of length, weight, sex, and age (from otoliths) of 328 Dover sole col- lected in 1972 and 1973. Size-frequency distributions between Dover sole of the same age with and without fin erosion were compared with a Kolmogorov-Smirnov two sample test (Siegel 1957). A one-tailed test was used be- cause the disease might be expected to reduce the growth rate of affected individuals. Weight-length relationships among male and female Dover sole with and without fin erosion were compared with the geometric mean regres- sion log w = log a + b (log /) where w = weight, I = length, and a and b are fitted constants (Ricker 1973). The regression co- Mearns, A. J., and L. Harris. 1975. Age, length, and weight relationships in southern California populations of Dover sole. Tech. Memo. 219, 17 p. Southern California Coastal Water Research Project, Long Beach. efficients were compared statistically using the method of Clarke (1980). The survival rate of Dover sole with and without fin erosion was compared. Using an age-length key developed from the data, ages were assigned to all Dover sole captured in LACSD monitoring trawls on the Palos Verdes shelf from 1972 through 1975. Survival rate (S) was calculated from age t (in years) to age ^ -I- 1 from S = N t + i N. where A^ = the number caught (Ricker 1975). Sur- vival rates of fish with and without the disease were compared with a ^-test after the data were transformed to the reciprocal (l/x) to stabilize the variance. A one-tailed test was used because the disease might be expected to reduce the survival rate of affected individuals. RESULTS From 1971 through 1982, LACSD made 622 monitoring trawls on the Palos Verdes shelf. Fin erosion was reported for 15,680 individuals (9.2% of all individuals collected) representing 29 species (23.8% of all species collected) (Table 1). 197 FISHERY BULLETIN: VOL. 83, NO. 2 Flatfish (Pleuronectidae, Bothidae, and Cynoglos- sidae) and rockfish (Scorpaenidae) accounted for 65.5% of the affected species and 99.2% of the affected individuals; Dover sole (Pleuronectidae) accounted for 88.9% of the affected individuals. The number of species affected by fin erosion declined from a high of 18 in 1971 to a low of 3 in 1981 (Fig. 2); the decrease w^as highly significant (r = -0.766, n = 12, 0.002

0.25). REX SOLE Seven percent of the rex sole, Glyptocephalus zachirus, had fin erosion; as a species, they ac- counted for 2% of all fish with the disease (Table 1). More than 99% of the rex sole collected were caught at 137 m. Less than 1% of the fish collected at station TO-137 m had fin erosion. The incidence of fin erosion among rex sole at the remaining 137 m stations declined significantly (Table 4). ANCOVA did not detect a significant difference among the regression coefficients for these collec- tions (F = L05,P > 0.25). The number of rex sole caught in a 10-min trawl at the 137 m stations did not change over the study period (Table 4), and the regression coefficients for these collections were not significantly different (F = 1.43, 0.10 0.50 T1-61 m Y = 1.681 - 0.001 5X 27 -0.0115 0.0085 P > 0.50 T4-61 m Y = 1 888 - 0.0098X 26 -0.0194 -0.0002 0.02- . P ■ 0.05 15-61 m Y = 0.756 -^ 0.0021 X 26 -0.0076 0.0118 P >0.50 198 CROSS: FIN EROSION AMONG FISHES of these collections (F = 17.84, P < 0.0001) (Table 6). The total number of Dover sole collected in a Table 3. — Results of Newman-Keuls multi- ple range test for equality of regression coeffi- cients (6) for the incidence of fin erosion among calico rockfish. s = station; un- derscored stations are not significantly dif- ferent. b: s: 0.678 T4-61 m 0410 T5-61 m 0.125 T1-61 m 0010 TO-61 m 10-min trawl declined significantly at three of the eight stations (Table 5). ANCOVA detected a sig- nificant difference among the regression coeffi- cients of these collections (F = 3.43, 0.001 < P < 0.003) (Table 6). Size-Frequency Distributions The size distributions of Dover sole with and without fin erosion were examined (Fig. 3). Most Dover sole recruit to the study area when they are 40-50 mm BSL (broad standard length). Fin ero- TABLE 4. — Linear regressions of A) the proportion (p) of rex sole, Glyp- tocephalus zachirus, in one 10-min trawl with fin erosion (transformed to arcsin \ p) and B) the total number of rex sole caught in one 10-min trawl [transformed to logio (J: -i- D] against time (numbered in consecutive months from 1971 through 1982). See Table 2 for explanation of column headers. Station Y =a +bX LI L2 A) TO- 137 m T1-137 m y = 10.217 - 0.085X T4-137 m y = 18.259 - 0.151X T5-137 m y = 10.733 - 0.087X B) TO-137 m y = 1.828 - 0.0038X T1-137m y = 1.326 - 0.0012X T4-137 m y = 0.976 + 0.0023X T5-137 m y = 1.182 - 0.0013X [only 7 of 1,520 fish (0.5%) had fin erosion] 33 -0.132 -0.038 0.002 < P < 0.005 35 -0.239 -0.064 0.001

0.50 36 -0.0020 0.0066 0.20 < P < 0.50 32 -0.0064 0.0038 P > 0.50 Table 5. — Linear regressions of A) the proportion (p) the Dover sole, Micros- tomas pacificus, in one 10-min trawl with fin erosion (transformed to arcsin \ p) and B) the total number of Dover sole caught in one 10-min trawl [transformed to logio (x -I- 1)] against time (numbered in consecutive months from 1971 through 1982). See Table 2 for explanation of column headers. Station y =a + bX n LI L2 P A) TO-61 m y = 0.861 - 0.0076X 29 -0.0148 -0.0004 0.02 < P < 0.05 TO-137 m y = 0.907 - 0.0047X 33 -0.0120 0.0026 P = 0.20 T1-61 m y = 29.526 - 0.240X 24 -0.308 -0.172 P <•: 0.001 T1-137m y = 28.048 - 0.201X 33 -0.274 -0.128 P <.. 0.001 T4-61 m y = 54.520 - 0.332X 29 -0.511 -0.154 P • 0.001 T4-137m y = 48.157 - 0.224X 36 -0.336 -0.112 P < 0.001 T5-61 m y = 54.732 - 0.374X 31 -0.482 -0.266 P < • 0.001 T5-137m y = 38.618 -0.214X 30 -0.300 -0.128 P < < 0.001 B) TO-61 m y = 1.601 - 0.0075X 37 -0.0128 -0.0022 0.005 < P < 0.01 TO-137 m y = 1.981 - 0.001 7X 34 -0.0066 0.0032 P > 0.50 T1-61 m Y = 1 .603 - 0.0077X 34 -0.0136 -00018 0.01

0.50 T5-61 m y = 1.354 - 0.001 8X 39 -0.0067 00031 0.20 < P < 0.50 T5-137m y = 2.592 - 0.0037X 31 -0.0082 0.0008 P =0.10 Table 6. — Results of Newman-Keuls multiple range test for equality of regression coefficients (6) for A) the proportion of Dover sole. Microstomas pacificus, with fin erosion and B) the total catch of Dover sole, s = station; underscored stations are not significantly different. A) b: s: b: s: -0.374 T5-61 m -0.332 T4-61 m -0.240 T1-61 m -0.224 T4-137m -0.214 T5-137m -0.201 T1-137m -0.008 TO-61 m -0.005 TO-137 m B) -0.0043 T4-61 m -0.0007 T4-137m -0.0017 TO-137 m -0.0018 T5-61 m -0.0037 T5-137m -0.0072 T1-137m -0.0075 TO-61 m -0.0077 T1-61 m 199 FISHERY BULLETIN: VOL. 83, NO. 2 20 60 100 140 180 220 260 300 BSL (MM) FIGURE 3. — Size distributions of Dover sole, Microstomas pacificus, with fin erosion (dashed line) and without fin erosion (solid line) by 20 mm size class (i.e., 20-39 mm, 40-59 mm, ..., 300-319 mm) for all collections (1971-82) combined. The number of individuals is presented to the right. BSL = board standard length. sion was observed in 0.3% of the fish between 40 and 59 mm BSL, 3.8% of the fish between 60 and 79 mm BSL, and 17.7% of the fish between 80 and 99 mm BSL. The incidence of fin erosion peaked in fish 120-139 mm BSL at stations close to the out- falls, and in fish 140-179 mm BSL at more distant stations. The proportion of fish with fin erosion in a particular size class was greatest at stations close to the outfalls and declined progressively with increasing distance (Fig. 4). Seasonal Trends Time series analyses of the quarterly trawl data (1979-82) for Dover sole at T4 and T5 showed con- sistent seasonal peaks in the total catch and in the number of individuals with fin erosion, but not in the proportion of individuals with the disease (Fig. 20 60 100 140 180 220 260 300 BSL (mm) FIGURE 4. — Percent of Dover sole, Microstomus pacificus, with fin erosion in each 20 mm size class (i.e., 20-39 mm, 40-59 mm, . . . , 280-299 mm) at 61 m and 137 m on the sampling transects. Data for each station are 12-yr totals. BSL = board standard length. 5). The seasonal indices of total catch and number with fin erosion were highest in the second (April-June) and third (July-September) quarters at 61 m and in the third and fourth quarters at 137 m. The magnitude of the seasonal swing was greater at 61 m. Examination of the deseasonalized data re- vealed that there was no change in the total catch of Dover sole, the number of individuals with fin erosion, or the proportion of individuals with fin erosion at 61 m between 1979 and 1982 (Table 7). At 137 m, there was a significant decline in the number of Dover sole with fin erosion at T4 and T5, and a significant decline in the total catch and proportion of individuals with fin erosion at T4 (Table 7). Size at Age The Mearns and Harris (footnote 3) data were examined for differences in the size-frequency dis- tributions between Dover sole with and without fin erosion at a particular age. There were no signi- ficant differences for fish age 2 (x^ = 0, df = 2, P > 0.90), age 3 (x^ = 1.27, df = 2, 0.50

0.50 T4-137 m y = 2.467 - 0.036f 16 -0.069 -0.003 0.02

0.50 T5-137 m y' = 2.509 -0.04U 16 -0.115 0.033 0.05 < P < 0.10 B) T4-61 m y = 0.370 + 0.043f 16 -0.028 0.114 0.20

0.50 T5-137 m y = 2.082 - 0.067f 16 -0.121 -0.013 0.01 < P < 0.02 C) T4-61 m y = 22.09 + 0.350f 16 -1.616 2.316 P > 0.50 T4-137 m y = 48.45 -1.470f 16 -2.819 -0.121 0.02 < P < 0.05 T5-61 m y = 29.32 - 0.267f 16 -2.055 1.521 P > 0.50 T5-137 m y = 40.87 - 0.982f 16 -2.635 0.671 0.20 < P < 0.50 Weight-Length Relationships The Mearns and Harris data were examined for differences in the weight-length relationships among males and females with and without fin erosion. There were no significant differences in the regression coefficients between males with and without the disease (T12 = 1.587, df = 38, 0.10

0.50) (Table 8). There was a significant differ- ence between males and females without the dis- ease (T12 = 3.189, df= 64, 0.002

- o z HI a LU CC U- LU < LU o CC LU 20 10 I I rii I I I I I I I I I i~i I I I 0.5 1.0 1.5 2.0 2.5 Ne/No 99.9 99.0 90.0 50.0 10.0 - 1.0 - 0.1 20 - 10 B I I r -0.5 20 - 10 Jl •■ WNo). crease. C. Distribution of (NqINq)^'^, the realized factor of in- (Ng/No) 1/6 99.9 99.0 90.0 50.0 - 10.0 - 1.0 0.1 -1 — I — I — I — I — I — I — I — I — I I r 0.0 0.5 1.0 LOGeCNg/No) 99.9 99.0 90.0 50.0 10.0 - 1.0 - 0.1 ■i — I rii I — I — I — r~i — I — I I I ii r 0.9 1.0 1.1 1.2 > o 2 LU Zi O Hi CO Li. LU o < LU o CC m CL LU > 3 ID o 211 FISHERY BULLETIN: VOL. 83, NO. 3 Table 6. — Accuracy of the 95% confidence limits (C.L.) on popula- tion size estimated by the logarithmic and root transformations of the distribution of total population size. For each transformation, the estimated lower and upper confidence limits are shown for pro- jections of the population given in Table 1 for 2, 5, and 10 time steps. The columns labeled "Proportion beyond C.L." give the actual pro- portion of 10,000 stochastic projections using program SLT which fall below the estimated lower limit and above the estimated upper limit for each transformation. Each set of projections was replicated 3 times. The root transformation estimates the 95% confidence in- terval on population size more accurately, especially for short pro- jections. Transformation No. of time steps to projection Logar thmic Root Estimated 95% C.L. Proportion beyond C.L. Estimated 95% C.L. Proportion beyond C.L. 2 Lower Upper 199 309 0.0311 0.0317 0.0295 0.0160 0.0181 0.0199 197 307 0.0256 0.0255 0.0231 0.0204 0.0214 0.0228 5 Lower Upper 187 380 0.0290 0.0285 0.0318 0.0227 0.0219 0.0215 184 378 0.0251 0.0242 0.0260 0.0245 0.0235 0.0235 10 Lower Upper 178 489 0.0266 0.0280 0.0286 0.0226 0.0211 0.0217 175 486 0.0238 0.0252 0.0257 0.0234 0.0221 0.0226 example projection are shown in the columns on the right side of Table 3. The mean and the 95% con- fidence interval for the total population size and for the realized factor of increase are given for each time step. As the population vector approaches the stable age distribution, the ratio between successive mean total population sizes approaches the asymptotic value 1.0240. The mean realized factor of increase shown in Table 3, which is computed relative to the initial population, does not converge on this asymp- totic value; nor can the mean realized factor of in- crease be computed from the ratio of the mean final population size to the initial population size. Instead, the mean and variance of the realized factor of in- crease are computed by methods described above. The probability that the total population size will have increased over its initial value is also shown for each time step in the last column of Table 3. In this particular example, since we did not begin with the stable age distribution, this probability decreases at first and then increases. As a further check, program SLT computes the proportion of cases in which the final population was greater than the initial popula- tion, and this answer (0.7954, Table 5) is close to the probability computed analytically by program SPP assuming that the realized factor of increase is nor- mally distributed (0.7990, Table 3). Given a popula- tion whose age structure and dynamics conform to the values given in Table 1, therefore, we can make the statement that there is an 80% chance that the population will be larger 6 time steps from now and a 20% chance that it will be smaller. Example 2. The results of the stochastic projection of the northern fur seal population by program SPP are given in Table 7 and Figure 2. Table 7 shows that after 5 yr, the expected (mean) number of 9-yr-olds, for example, is 6,188 with a standard deviation of 333. The expected total population size is 147,982 with a standard deviation of 8,832. The mean and standard deviation of the realized factor of increase are 1.0812 and 0.0129, respectively; from these values we compute the 99% confidence interval on population size to be from 126,410 to 171,930. Note 212 GERRODETTE ET AL.: CONFIDENCE LIMITS FOR POPULATION PROJECTIONS Table 7.— Results of the 5-yr stochastic pro- jection of the northern fur seal population, bas- ed on the age structure and vital rates given in Table 2. Probability that the final population > initial population = 0.999 -t-. Expected Age population Standard class size deviation 1 26,071 1,672 2 21,179 1,619 3 17,205 1,519 4 14,058 1,402 5 11,553 1,280 6 9,655 972 7 8,205 720 8 7,087 516 9 6,188 333 10 5,432 121 11 4,743 106 12 4,075 92 13 3,406 76 14 2,767 62 15 2,170 49 16 1,611 36 17 1,121 25 18 717 16 19 415 9 20 209 5 21 87 2 22 26 1 23 2 0.05 Total 147,982 Total Factor of population increase Low/er99% C.L. 126,410 1.0480 Expected value (mean) 147,982 1.0812 Upper 99% C.L. 171,930 1.1145 that, as will generally be the case, the confidence in- terval for total population size is not symmetric about the mean value. In the last line of Table 7, the probability of an in- creased population is shown to be very close to 1.0. In other words, it is virtually certain that the popula- tion will have increased in size after 5 yr. Figure 2 presents the results for total population size graph- ically. The 95% and 99% confidence limits computed by program SPP are shown for each time step. The confidence limits grow nearly geometrically. DISCUSSION Fishery and wildlife management often involves predictions of population size, and, owing to im- perfect knowledge of the world, these predictions are uncertain. Accordingly, a practical analysis attaches estimates of confidence intervals for any given prediction. The programs described in this paper carry out the computation of confidence intervals for projections of age- structured populations, if we can specify the statistics of the variation in the age- specific vital rates. Realistically, we do not expect there to be be many examples where the statistics of the variation in vital rates are genuinely known with substantial precision, for these rates are difficult to measure in natural populations. Nevertheless, in an imperfect world, management decisions must be made with imperfect data. A considerable compo- nent of the uncertainty in a population prediction will be owing to the phenomena treated in this paper. Thus, even the use of very rough guesses at the sta- tistics of the variation in the age-specific vital rates, in order to estimate confidence intervals in a popula- tion projection, is preferable to neglecting this source of variation entirely. At the very least, incor- poration of speculative estimates in this applied con- text will allow the exploration of "what if " questions in a fashion that can indicate priorities for future data gathering. In many fish and aquatic invertebrate species, there is an enormous variation in the success of year classes. In such cases the population dynamics may be dominated by the overwhelming abundance of one 180 |- 95% Confidence limits 99% Confidence limits ■g 160 0 12 3 4 5 YEARS IN FUTURE Figure 2. -Confidence limits for future total female population size for the northern fur seal, based on the schedule of vital rates in Table 2. The solid line plots the mean population trajectory. 213 FISHERY BULLETIN: VOL. 83, NO. 3 or two cohorts. The environmental factors which lead to such huge variations in recruitment are as yet imperfectly understood for most species. In order to predict future population sizes, the year-to-year variation could be incorporated into the variances of the effective fecundity terms in the first row of the Leslie matrix. This will lead to enormous (but realistic) confidence limits for predicted future population sizes of such stocks. A more fruitful use of the results of this paper, however, would be to separate recruitment uncertainty from survival uncertainty and to calculate a confidence interval on future population size given recruitment success for a particular cohort. Among harvested species such a conditional forecast could be used to incorporate environmental variation into management recom- mendations. In keeping with the fact that applied management may often depend on very elementary quantities, we also calculate a particularly important special statistic of the distribution of projections- the prob- ability that the population will increase under the specified conditions. In the first example, the prob- ability of an increased population was found to be about 0.8. In the second example, the fur seal popula- tion projection, there is a higher probability that the population will increase. Starting with the female population of 100,000, the calculations indicate 99% certainty that the population will have increased to between 126,410 and 171,930 in 5 yr. Our simulations of stochastic population growth differ from previous efforts by Boyce (1977) and Slade and Levenson (1982) by allowing all vital rates to vary simultaneously, rather than only one at a time, and by permitting correlations among the vital rates to be specified. In the stochastic growth models of Cohen (1977, 1979a) and Tuljapurkar and Orzack (1980), at each time step the population finds itself in one of several possible environments. Within each environment vital rates are fixed. By contrast, here we model a single variable environment whose condi- tions, as reflected in the population's vital rates at any point in time, are never precisely duplicated. The results of Example 1 verify the results for the mean and variance of future population vectors and show that the mean and variance for total ultimate popula- tion size can be computed from Sykes' formulae. Our results confirm the conclusions of Cohen (1977), Tul- japurkar and Orzack (1980), and Slade and Levenson (1982) that the expected mean value of a stochastic population projection with no serial correlation in vital rates is equivalent to the value projected deter- ministically from mean vital rates. Cohen (1979a, b) and Tuljapurkar (1982) address the more general question of serial correlation in vital rates. All of the work cited above has been concerned v^dth the state of the population at a time in the future much greater than will generally be useful in management. In this paper we have examined the stochastic behavior of the population at a shorter time in the future. Example 1 has verified that the distribution of ultimate population sizes from stochastic population projections is approximately lognormal (Tuljapurkar and Orzack 1980). From the perspective of fitting the tails of this distribution for a small number of time steps t, however, it appears better to assume that the 1/tth power of the distribu- tion is normally distributed. In either case the distri- bution of ultimate population sizes is skewed (with long tails at the higher values), and the skew becomes more pronounced as t increases. An impor- tant property of such a distribution is that the most likely or modal population value will always be smaller than the mean. How much smaller depends on the number of time steps t, and on the variances and covariances among the survival and fecundity rates. An interesting theoretical and practical problem is to find a descriptor of population growth under stochastic conditions which characterizes the skewed distribution of ultimate population size. Cohen (1979a) has proposed two measures of long-run popu- lation growth: A, the ensemble average of realized factors of increase; and pi, the factor of increase need- ed to realize the ensemble average of final population sizes. The first is a measure based on growth rates, while the second is based on population sizes (Cohen 1979b). The average realized factor of increase calcu- lated here is analogous to A. If the Leslie matrix of mean vital rates is known, pi is easily calculated as the dominant eigenvalue of that matrix. The prob- lem, as we have seen, is that under stochastic condi- tions the mean of the population sizes is not very in- formative and may, in fact, be misleading. Tul- japurkar (1982) has proposed a growth measure a which leads to the approximate median population size. The two measures proposed here -namely, E [(iV^/A^o)'"] a.ndE[\og{N,/N(f)]-are close approxima- tions to the rate of growth leading to the modal population size. As such, they may loosely be said to describe the most probable trajectory of the popula- tion under stochastic conditions. ACKNOWLEDGMENTS The work was supported by a National Research Council Fellowship to the first author and by NOAA 214 GERRODETTE ET AL.: CONFIDENCE LIMITS FOR POPULATION PROJECTIONS Contract 80-ABC-00207 to the second. We thank Douglas Chapman for his helpful comments. Typing assistance was provided by Lorraine Prescott and her staff at the Southwest Fisheries Center. LITERATURE CITED AlTCHISON, J., AND J. A. C. BROWN. 1957. The lognormal distribution with special reference to its uses in economics. Cambridge Univ. Press, Cambridge, 176 p. BOYCE, M. S. 1977. Population growth with stochastic fluctuations in the life table. Theor. Pop. Biol. 12:366-373. Cohen, J. E. 1977. Ergodicity of age structure in populations with Marko- vian vital rates, III: Finite-state moments and growth rate; an illustration. Adv. Appl. Prob. 9:462-475. 1979a. Long-run growth rates of discrete multiplicative processes in Markovian environments. J. Math. Anal. Appl. 69:243-251. 1979b. Comparative statics and stochastic dynamics of age- structured populations. Theor. Pop. Biol. 16:159-171. Daley, D. J. 1979. Bias in estimating the Malthusian parameters for Leslie matrices. Theor. Pop. Biol. 15:257-263. Gerrodette, T., D. Goodman, and J. Barlow. 1983. Two computer programs to project populations with time-varying vital rates. Natl. Mar. Fish. Serv. Tech. Memo. N{)AA-TM-NMFS-SWFC-28, 56 p. (Copies are available through NTIS or from the Southwest Fisheries Center, National Marine Fisheries Service, NOAA, P.O. Box 271, La JoUa, CA 92038.) Goodman, D. 1981. Life history analysis of large animals. In C. W. F'owler and T. D. Smith (editors). Dynamics of large mammal popula- tions, p. 415-436. Wiley, N.Y. Keyfitz, N. 1968. Introduction to the mathematics of population. Addi- son-Wesley, Reading, Mass., 450 p. Leslie, P. H. 1945. On the use of matrices in certain population mathema- tics. Biometrika 33:183-212. Lewontin, R. C, and D. Cohen. 1969. On population growth in a randomly varying environ- ment. Proc. Natl. Acad. Sci. (U.S.) 62:1056-1060. Slade, N. a., and H. Levenson. 1982. Estimating population growth rates from stochastic Leslie matrices. Theor. Pop. Biol. 22:299-308. Sykes, Z. M. 1969. Some stochastic versions of the matrix model for popu- lation dynamics. J. Am. Stat. Assn. 64:111-130. Tuljapurkar, S. D. 1982. Population dynamics in variable environments. II. Correlated environments, sensitivity analysis and dynamics. Theor. Pop. Biol. 21:114-140. Tuljapurkar, S. D., and S. H. Orzack. 1980. Population dynamics in variable environments. I. Long- run growth rates and extinction. Theor. Pop. Biol. 18:314- 342. 215 FISHERY BULLETIN: VOL. 83, NO. 3 APPENDIX 1. Calculation of the mean and variance of the realized factor of increase, assuming it is normally distributed. Let A, the realized factor of increase, be defined as the ^th root of the ratio of the population size at time t to the initial population size: nJ or Let ^ be the mean and v the variance of A. The mean and variance of A' are given by formulae in the Methods section. The problem is to find the mean and variance of A. Let F (/^,v) be a function which gives the ^th moment of A: F(m,v) = E(A'). Similarly let G{^a,v) be a function which gives the variance of A' in terms of the ^th and 2tth moments of A: G(m,v) = F(A2')- [F(A')]2. Now assume that A is normally distributed. Appendix 2 gives a recursive algorithm which allows any moment of a normally distributed variate to be calculated. From the tth and 2tth moments of A, the functions F and G can be computed from the equations above. Generally, F and G will be tth and 2tth order polynomials in fu and V. Then, with F and G known, we have a system of two equations F(m,v) - Eil^) = 0 G(m,v) - Var (A' ) = 0 in two unknowns. Given initial estimates of ^ and u, a two-variable version of Newton's method, or any similar iterative technique, can be used to converge on a simultaneous solution. 216 GERRODETTE ET AL.: CONFIDENCE LIMITS FOR POPULATION PROJECTIONS APPENDIX 2. A recursive algorithm for computing the higher order moments of the normal distribution. The moment generating function for the normal distribution is 2 where ^ is the mean and v is the variance of the normal variate x. The nth moment of x is found by evaluating, at ^ = 0, the nth derivative ofM^{t) with respect to t. The wth differentiation with respect to t leads to the series (m + vO"M^(0 + ... + A(m+ vt)" v'^ M^{t) + 5(m + vty-^v^^^M^it) + ..., which, evaluated oXt = 0, gives where A and B are coefficients and a and ft are exponents such that a + 2/3 = w. The next [(n + l)th] dif- ferentiation of the middle terms gives A(m + vtf^^ v^M,{t) + Aa{yL + vty-^ v^^^ MJt) + B{yL + vt)"-^ v^*^ M,{t) + B{q - 2)(m + vt)'-^ v^^^ M^(t) = ... + (Aa + 5)(m + vty-^ v'^^^ M,{t) + ... which, evaluated at ^ = 0, gives ... + (Aa + 5V"^ v^^l + ... . Thus the coefficient of each term of the series of the {n + l)th moment can be computed from the two terms in the series of the nth moment "before" and "after" it. The exponents of ^ and v follow the regular pattern shown. 217 LIFE HISTORY CHARACTERISTICS OF PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS IN PENOBSCOT BAY, MAINE David K. Stevenson^ and Fran Pierce^ ABSTRACT A number of life history characteristics of two species of pandalid shrimp from Penobscot Bay, ME, were inferred from length-frequency and relative abundance data collected on five occasions during a bottom trawl survey in 1980-81. Pandalus montagui is a sequential hermaphrodite. Sex transition occurs throughout the year, but most transitional individuals were observed in late March. Most individuals change sex shortly before or after reaching age 2, but some do so either a year earlier or a year later. Ovigerous females were observed from late November through January; eggs are apparently produced during the sec- ond, third, and fourth years. Fifteen percent of the 0 age-group caught in the fall of 1980 were females which may never have functioned as males. Growth was rapid in the spring and summer and negligible in the late fall and winter. Females which changed sex at age 1 were larger than females which changed sex a year later. Dichelc/pandalus leptocerus is not hermaphroditic. Ovigerous females were collected primarily in late November and early December. Some females produce eggs during their first and second years, but most do so only during their second year. None of the females caught during this study appeared to be older than age 2; a few large males remained in the population during their third year of life. Females of both species were larger than males of the same age-group, a distinction which was attributed to differences in growth rate and, for P. montagui, was associated with earlier sex transition. Larger shrimp of both species migrated down the Bay into deeper water as the winter progressed. The Pandalidae are a family of boreal, subarctic shrimp composed of 2 genera and about 20 species. Four species {Pandalus borealis, P. montagui, P. propinquus, and Dichelapandalus leptocerus) are common in offshore waters of the Gulf of Maine (Wigley 1960). Of these, P. borealis is the largest, reaching a maximum total length of 17-18 cm (7 in), and has been the object of a directed winter fishery in coastal waters of the Gulf of Maine since the late 1930s (Scattergood 1952) and in coastal and offshore waters since 1958^. This species is also exploited commercially on the west coast of North America, in the Canadian Maritime Provinces, on the west coast of Greenland, in the Norwegian and North Seas, and in the northwest Pacific (Balsiger 1981). Pandalus montagui and D. lefptocerus are smaller species (maximum length 10 cm or 4 in), which are harvested incidentally with P. borealis in the Gulf of 'Zoology Department, University of Maine, Orono, ME 04469 and Maine Department of Marine Resources, West Boothbay Har- bor, ME 04575. ^Maine Department of Marine Resources, West Boothbay Har- bor, ME 04575. ^Stickney, A. P. 1980. A characterization of the northern shrimp fishery of Maine. In C. J. Walton (editor). Fisheries management and development. Vol. Ill, Element D: Character- ization of the shellfisheries, p. 244-293. Completion report to the State Planning Office, Oct. 1, 1978-Sept. 30, 1979, Maine Depart- ment of Marine Resources, Augusta. Maine, but have little or no market value because of their size. Pandalus montagui is also harvested as an incidental species in the Gulf of St. Lawrence (Balsiger 1981), and for many years was the object of several localized commercial beam trawl fisheries in the southern North Sea and in Morecambe Bay, northwest England, until declining stock sizes led to the demise of the fisheries in the Thames estuary (described by Mistakidis 1957) and Morecambe Bay in the 1950s and 1960s. Warren (1973) described a fishery for P. montagui in the Wash on the east coast of England which was still active in the early 1970s. Pandalus propinquus is also smaller than P. borealis and is generally restricted to deeper water (165-330 m in New England waters according to Wigley 1960); consequently it is rarely taken in Gulf of Maine commercial catches. Pandalus montagui is differentiated taxonomically into two subspecies: P. montagui tridens in the North Pacific and P. montagui montagui in the North Atlantic from the Arctic south to the British Isles and Cape Cod (Simpson et al. 1970) or Rhode Island (Rathbun 1929). According to Simpson et al. P. montagui montagui is found in estuaries, coastal waters, and offshore in depths of 5 to over 700 m, but is more common in shallow waters (20-90 m); at depths > 90 m it is gradually replaced by P. borealis. Manuscript accepted August 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 219'^^ Squires (1968) reported that P. borealis occurred together with P. montagui in depths < 200 m in the Gulf of St. Lawrence and southwest of Newfound- land, but at depths between 200 and 300 m with a smooth detritus bottom and temperatures of 4°-6°C, only P. borealis were caught; in colder temperatures (-1° to 3° C) in this same depth range, P. montagui were more abundant. Pandalus montagui was described as a more eurythermal and eurybathic species than P. borealis. Of all the pandalids in the northwest Atlantic, P. montagui is the only one which inhabits colder Arctic waters < 1.5°C (Squires 1966). The Atlantic subspecies of P. montagui has been the subject of several biological studies (Mistakidis 1957; Allen 1963; Couture and Trudel 1969a, b). Dichel&pandahis leptoceru^ is distributed in the northwest Atlantic from Newfoundland to North Carolina (Rathbun 1929). It has not been reported from the northeast Atlantic and is rare in the north- ern Pacific (Squires 1966). During a November 1956 bottom trawl survey in New England waters, D. lep- tocerus was much more widely distributed than P. montagui or P. borealis (Wigley 1960). Dichelopan- dalus leptoceru^ was also found over a broad depth range (33-340 m), but was common between 35 and 145 m and at temperatures (in November) of 5°-20°C, whereas P. montagui occurred primarily between 70 and 135 m and at temperatures of 6°-10°C. Dichelopandalu^ leptocerus was also col- lected in areas where bottom sediments contained low, medium, and high quantities of organic matter, whereas P. montagui appeared to be associated with sediments with relatively low organic content. Thus, in several ways, D. leptocerus appears to have less restricted habitat requirements than P. montagui (or P. borealis). No detailed biological studies of D. lep- tocerus have been published. The Maine Department of Marine Resources con- ducted an exploratory bottom trawl survey to deter- mine the abundance and distribution of pandalid shrimp populations in Penobscot Bay (Figure. 1) dur- ing 1980-81. During the course of this survey, biological data were collected from about 10,000 shrimp. The objective of this paper is to describe im- portant life history characteristics of P. montagui and D. leptocerus in Penobscot Bay (the Bay); these include breeding seasons, female sizes and ages at maturity, sex transition, growth, longevity, and migratory behavior. Aspects of the life cycle and reproductive biology of each species were examined as functions of time of year, depth, and location within the Bay. FISHERY BULLETIN: VOL. 83, NO. 3 METHODS The survey was conducted over the course of a 12-mo period from late November 1980 to early October 1981. Samples were collected during five distinct periods of time at 19 different stations located from Cape Jellison in the northern end of Penobscot Bay to Mark Island, a distance of about 37 km (23 mi) (Fig. 1). Stations were established at depths ranging from 12 m (40 ft) to 84 m (280 ft) and were located in areas of trawlable bottom. Since a primary objective of the survey was to stimulate commercial shrimp fishing, no attempt was made to randomly select station locations, depths, or sam- pling times. Attempts were made, however, to return to each station as often as possible so as to determine the seasonal variation in the relative abundance of different sexes, reproductive stages, and size groups of each species at individual locations over the course of the year. Adjacent, well-defined, length groups were assumed to represent successive age-groups. A total of 45 successful tows (i.e., tows that were not aborted because of bottom obstructions, damage to the trawl, or gear malfunction) were made during the entire survey. Of these, 37 tows which could be assigned to a specific area, depth range, and sam- pling period were selected for data analysis. Area 1 was defined as the upper Bay, area 2 as west of Islesboro, area 3 as south of Islesboro, and area 4 as east of Islesboro; depth ranges were defined as shallow (12-25 m), moderate (25-50 m), and deep (50-85 m) (Table 1). The distributions of sampling ef- fort between stations by sampling period, area, and depth range are shown in Table 2. No data were Table 1.— Definitions of coded sampling periods, areas, and depth ranges, 1980-81 Penobscot Bay shrimp survey. Sampling periods 1 2 3 4 5 Areas 1 2 3 4 Depth ranges 1 2 3 20 November-2 December 1980 21-29 January 1981 24-31 March 1981 16 July-18 September 1981 5-6 October 1981 Upper Bay: stations 2, 3, 4, 6, 18 West of Islesboro: stations 1, 5, 9, 10, 14 South of Islesboro: stations 7, 8. 12, 15, 16, 17, 19 East of Islesboro: stations 11. 13 12-15 m (shallow) 25-50 m (moderate) 50-85 m (deep) 220 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERVS Figure l.-Map of Penobscot Bay. ME, showing bottom trawl stations visited during 1980-81 survey. 221 FISHERY BULLETIN: VOL. 83, NO. 3 Table 2. — Distribution of sampling effort (number of tows per station) by sampling period, area, and depth range, 1980-81 Penobscot Bay shrimp survey.' Sampling period Area Depth Station 1 2 3 4 5 1 2 3 4 1 2 3 1 2 1 6 9 9 3 1 2 2 4 1 1 5 1 1 6 1 1 7 1 1 1 3 3 8 1 1 9 1 2 1 1 10 1 3 5 5 11 1 1 2 1 1 12 1 1 13 1 1 1 15 1 1 16 1 1 1 3 3 17 1 1 18 2 2 2 19 1 1 Totals 9 10 9 6 3 6 17 11 3 6 17 14 'See Table 1 for definitions of coded sampling periods, areas, and depth ranges. available from stations 2 and 14. Trawling opera- tions were limited to the area between Northport and Islesboro in the summer of 1981, since the only vessel available at that time was not equipped to work elsewhere in the Bay. The trawl used was a semiballoon shrimp try-net with a 7.6 m (25 ft) headrope and 9.5 m (31 ft) foot- rope with no rollers; mesh size was 38 mm (IV2 in) in the body and 31 mm (IV4 in) in the cod end. The trawl was also equipped with a 12.5 mm (V2 in) liner. The net was rigged on 1.2 m (4 ft) legs with ark floats on the headrope and 2/0 chain on the footrope. The trawl doors were hardwood, 76 x 41 cm (30 x 16 in), with iron bracings and a wide shoe. The net was fished on a single trawl wire attached to a 30.5 m (100 ft) wire net bridle. All tows lasted 30 min and were made at speeds varying from 1.5 to 2.5 kn. Loran bearings were recorded at the beginning and end of each tow, and depth was recorded either as a single reading or at the beginning and end of each tow. Location and depth could not be determined in the summer since the vessel used then did not have sonar or navigational equipment. Although three dif- ferent vessels were used during the course of the survey, the gear was identical and was fished the same way during the entire survey. If catches were small (under 1 kg), the entire catch was generally brought to the laboratory and frozen for later analysis; otherwise, the catch was sub- sampled aboard the vessel. In some cases, large samples were further subsampled in the laboratory after they were thawed. Inasmuch as was possible, all samples and subsamples were randomly selected. Samples (or subsamples) of 200-900 g were sorted (after removing extraneous "trash") by species ac- cording to morphological characteristics described by Rathbun (1929). Biological data were compiled for a total of 7,259 D. leptocerus and 2,475 P. montagui; numbers of P. borealis were inadequate for data analysis. Each individual shrimp was sexed (male, female, or transitional) using external morphological characteristics for the genus Pandalics originally described by Wollebaek (1908), Berkeley (1930), Jagersten (1936), and Leloup (1936) and summarized by Mistakidis (1957). The females were further grouped as ovigerous or non-ovigerous depending on whether or not the eggs had "dropped" and were being carried on the pleopods; the non-ovigerous females were further subdivided into two groups - those which had never carried eggs before and those which had - based on the presence or absence of ster- nal spines. This characteristic of non-ovigerous females was originally described by McCrary (1971) for three pandalid species (Pandalus borealis, P. goniurus, and P. hypsinotus) in Alaska. Stage I females were defined as those which had not carried eggs before and Stage II females as those which had; there was no way to distinguish between females which had carried eggs only once before and those which had carried eggs more than once. Carapace lengths were measured between the eye socket and posterior dorsal edge of the carapace and recorded to the nearest 0.1 mm. For each species, the numbers and lengths of shrimp in each biological category (sex, with or without eggs. Stage I or II) were compiled by sam- pling period, geographic area, and depth range. Length frequencies were expressed as numbers of shrimp per 0.5 mm dorsal carapace length. Since nearly all of the samples collected in areas 1-3 were also collected in specific depth ranges (i.e., all 6 samples from area 1 were from shallow water, 16/17 samples from area 2 were from moderate depths, and all samples from area 3 were from deep water), length frequencies were presented for appropriate area/depth combinations. Length-frequency data for P. montagui collected in October 1981 were not presented since so few individuals were captured. RESULTS AND DISCUSSION Breeding Seasons and Female Sizes (Ages) at Maturity'* Nearly all the ovigerous female D. leptocerus were 222 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS caught in November-December 1980 (Table 3), although a few remained in January and March. It was therefore apparent that most eggs hatched dur- ing a relatively short period of time in late December and early January. Although only the larger size group was carrying eggs (Fig. 2D), the presence of a were reported for the same populations as late February through April with peak activity in April. According to Couture and Trudel (1969b), ovigerous females were observed in Grand-Rivifere, Quebec, beginning in July and accounted for the greatest percentage of the population in October (no samples Table 3.— Percent total number of male and female Dichelopandalus leptocerus collected at all locations and depths in Penobscot Bay during five sampling periods in 1980-81. (Females are categorized by reproductive stage.) 11/20-12/2 1/21-1/29 3/24-3/31 7/16-9/18 10/5-10/6 Sex/Stage 1980 1981 1981 1981 1981 Total Total males 49.7 65.3 59.4 47.6 53.0 53.2 Females/Stage 1 32.0 33.5 37.8 50.2 45.3 38.8 Females/Stage II 1.4 0.3 1.5 2.2 1.5 1.5 Total non-ovigerous females 33.4 33.7 39.3 52.4 46.8 40.2 Ovigerous females 16.8 1.0 1.3 0.0 0.2 6.6 Total females 50.3 34.7 40.6 52.4 47.0 46.8 Total no. individuals 2,694 729 1,151 1,107 1,577 7,259 few Stage II non-ovigerous females in roughly the same size range (Fig. 2C) indicated that some females produced eggs a year earlier as well. It could not be determined from the samples collected during this study whether the younger females spawned earlier or later than the older group. The fact that so few Stage II females were captured in the Bay at any time of year indicated that most of the spawning population was made up of first time spawners. Ovigerous females were collected at all depths and in all areas (Tables 4, 5), but made up a greater percent- age of the catch at moderate depths in area 2. Nearly all of the ovigerous female P. montagui were collected in November-December and late January (Table 6); the fact that 50% of the females in late January were still ovigerous suggests that eggs hatched over a more prolonged period then was true for D. leptocerus, possibly from November at least through February. Females belonging to two North Sea P. montagui populations were reported to carry eggs primarily between November and February (Mistakidis 1957; Allen 1963), although ovigerous females were observed from mid-October to April in the Thames estuary by Mistakidis. Hatching times *Since no internal sexual characteristics (such as oocyte develop- ment) were examined in this study, the breeding season was defined as the period of time when ovigerous females were observed and sizes (ages) at maturity as the sizes (ages) when females produce eggs. No comparable information for males (i.e., mating times or sizes (ages) at maturity) was available. As used in this paper, the breeding season was, strictly speaking, the period of time between spawning and hatching when eggs were incubated. Table 4.— Percent total number of male and female Dichelo- pandalus leptocerus collected at all depths and times of year in four areas in Penobscot Bay during 1980-81. (Females are categorized by reproductive stage.) Sex/Stage Area 1 Area 2 Area 3 Area 4 Total Total males 54.4 49.3 57.4 70.6 53.0 Females/Stage 1 Females/Stage II Total 42.2 1.0 39.7 1.8 36.7 1.1 26.1 1.1 38.9 1.4 non-ovigerous females 43.2 41.5 37.8 27.2 40.3 Ovigerous females 2.4 9.2 4.8 2.2 6.6 Total females 45.6 50.7 42.6 29.4 47.0 Total no. individuals 1,047 3,626 2,419 92 7,184 Table 5.— Percent total number of male and female Dichelo- pandalus leptocerus collected in all areas and times of year by depth range in Penobscot Bay during 1980-81. (Females are categorized by reproductive stage.) Sex/Stage Shallow (12-25 m) Moderate (25-50 m) Deep (50-85 m) Total Total males 54.4 47.7 80.3 53.3 Females/Stage 1 Females/Stage II Total 41.5 1.1 40.4 2.0 32.4 0.9 37.6 1.4 non-ovigerous females 42.6 42.4 33.3 39.0 Ovigerous females 3.0 9.9 6.4 7.7 Total females 45.6 52.3 39.7 46.7 Total no. individuals 812 3,015 2,299 6,126 223 FISHERY BULLETIN: VOL. 83, NO. 3 120- 100 80- 60- 40- 2 0- 180- N0V20 -DEC 2 1980 160- Males N = l 33 1 140- 120- h 100- — 1 8 0- H ^ 60- , — , 40- U 20- ^ 1 T 1 1 T ^ A Non- o vige r ou s Females Stage I N-863 8 0- 60- 40- 20- JAN 21-29 1981 Males N = 476 ^ ^^ , s ^ E 10 Non -o vige rous Females Stage II 5 N = 38 1 ^^ r^ . "H r^^ c 80- 6 0- 40- 2 0- ■ 8 0- p_,_^ Ovigerous Females 6 0- N-4 5 4 J n 40- 2 0- • r^ , , ^ ° 40- 3 0- f\ Non -ovigerous Stage 1 N=24 4 Females 20- H 10- r^ ^r^ ■F=^ 1 r- F 10 0- MARCH 24 -31 1981 8 0- 6 0- pJ Males N = 684 4 0- 1—1 20- ^ ^ G -T 1 — — r- J~L Non-ovigerous Females Stage 1 r^^ 1 N-436 r u No n - o vige f ous Females Stage II N = l 7 I Ovigerous Females N = l 4 J 5 6 7 8 9 10 11 12 13 14 15 16 17 5 6 7 8 9 10 11 12 13 14 15 16 17 CARAPACE LENGTH (mm) CARAPACE LENGTH (mm) Figure 2.-Dichelopnvdalus leptoren/s numerical lenjjth- frequency distributions by sex and reprcKJuctive stafje. Table 6. — Percent total number of male, transitional, ancj female Pandalus mon- tagui collectecj at all locations an(j depths in Penobscot Bay during five sampling periods in 1980-81. (Females are categorized by reproductive stage.) 11/20-12/2 1/21-1/29 3/24-3/31 7/16-9/18 10/5-10/6 Sex/Stage 1980 1981 1981 1981 1981 Total Total males 73.2 85.2 68.4 37.3 34.3 68.9 Transitionals 0.3 2.5 15.4 5.7 1.5 6.9 Females/Stage 1 7.8 6.1 15.9 50.4 61.2 18.2 Females/Stage II 0.3 0.0 0.0 6.7 3.0 1.2 Total non-ovigerous females 8.1 6.1 15.9 57.0 64.2 19.4 Ovigerous females 18.4 6.2 0.4 0.0 0.0 4.8 Total females 26.5 12.3 16.2 57.0 64.2 24.2 Total no. individuals 332 871 800 405 67 2,475 224 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS JULY 16 -SEPT 18 1981 Males N»52 7 Non -0 viger ouB Femal»« Slaga 1 r n N-5Se H L r ' ■ -^ 1 ■ . r Non-ovlgaroua Females Stage II N-24 M 1 OCT 5-6 1981 Males N-83e N Non-ovlgeroua Females Stage I N> 7 1 5 -_X 0 -1 — — ^ Non- o vigerous Females Stage II N>2 3 —n . r^ —r 1 1 1 r- a 9 10 11 12 13 14 15 16 CARAPACE LENGTH (mm) — I » 17 ia were collected between November and April). A few females were still carrying eggs the following June. Hatching began in the winter and continued through June. Pandalus montagui populations studied in Penobscot Bay and Grand-Riviere spawned primari- ly in their second and third years. Females in the North Sea, on the other hand, were fully mature dur- ing their first year (Mistakidis 1957; Allen 1963), but there was no evidence that individual age-groups spawned more often there than at Grand-Riviere or in Penobscot Bay. At least two age-groups of ovigerous P. montagui were apparent in the winter samples (Fig. 30, G), whereas only a single age-group of ovigerous D. lep- tocenis was observed (Fig. 2D). Length-frequency data suggested that older female P. montagui (age 2 + ) spawned before younger females (age 1); 57% of the ovigerous females collected in November- December were > 15 mm CL (Fig. 30), whereas only 30% remained in the same size group in late January (Fig. 3G). Sample sizes were much too small, however, to clearly indicate how many spawning age-groups were present or whether older femaleo spawned earlier than younger ones. Earlier comple- tion of spawning by older females was reported by Mistakidis (1957) in the Thames estuary. In Penobscot Bay, the relative abundance of ovigerous females was higher in moderate and deep waters and in areas 2, 3, and 4 (Tables 5, 6). The capture of a single 10 mm OL ovigerous female P. montagui in January (Fig. 3G) indicates that a few females mature and reproduce during their first year. This shrimp could have started life as a female or could have changed sex in the first year and therefore never functioned sexually as a male. Mistakidis (1957) reported that some 0 age-group males in the Thames estuary changed sex and func- tioned as females during their first year. Some in- dividuals in both the Thames estuary and North- umberland began life as females and matured in their first year (Mistakidis 1957; Allen 1963). Sex Transition Unlike most other Pandalid species, the Penobscot Bay population of D. leptoceru^ was not her- maphroditic. Not a single transitional individual was identified in the over 7,000 shrimp which were ex- amined. Furthermore, males and females recruited to the > 5 mm OL population in October of their first year in nearly equal numbers (Fig. 2N, P). The ratio of males to females for the entire survey period was 53:47 (Table 3). Pandalus montagui, on the other hand, is a protandric (sequential) hermaphrodite. Nearly 7% of the 2,475 individuals examined ex- hibited external morphological characteristics typical of transitional P. borealis (Allen 1959). The sex ratio was 69% males to 7% transitionals to 24% females (Table 6). Although P. montagui is clearly protan- dric, some individuals in Penobscot Bay either begin life as females or assume external female character- istics by late November of their first year.^ These ^Individuals of protandric pandalid shrimp species which begin life as females are referred ta as primary females; those which change 225 FISHERY BULLETIN: VOL. 83 NO. 3 NOV 20 -DEC 2 1980 Males N=242 Non-0 vigerous Females Stage I N=26 rL Transitionals N=22 — 1 T" 1 0- 5 • _^^ Non- 0 ¥ige r o u s Females Stage I N-53 F r 1 ■=• ■ i 1 0 5 • Ovtgerous Females N-54 Figure 3.-Pandalus montagui numerical lenp^th-frequency (iistribu- tions by sex (including transitionals) anti repnxiuctive stage. 10- 5 30 20- 10- 5 ■ MARCH 24 -' 51 1981 60- pJ Males 50- 1 — N-547 40- 30- 20- -^ LJ^ 1 0- 5 1 ' r 1 1 1 r 1 1 1 r^* r- H Transitionals N-' 23 Transtttonais N=23 Non- ovigerous Females Stage I N = 204 1 2 0- r^ — 1 Non-0 vigerous Fema Stage 1 les 10- 5 ^ N>I 2 7 1 , 1 •—\ , r— "=1 J 1 , , — . 30- JULY 16- SEPT 18 1981 Males 20- N = 1 S 1 1 0- '-l 5 • — t I - — . . . — — ,__ K r4on - o vigerous Females Stage II N=27 CARAPACE LENGTH (mm) females were obvious as a distinct size-group in the November-December length-frequency data at 6-9 mm CL (Fig. 3B). At this time, these females made up 15% of the newly recruiting 0 age-group. The relative abundance of 0 age-group females was considerably higher in the North Sea: 29-37% of most samples in the Thames (Mistakidis 1957) and about 50% in Northumberland (Allen 1963). On the sex In their first year following the repression of male sex characteristics and never function as males are called secondary females; and those which function first as males and then change sex are called hermaphroditic females (Mistakidis 1957). Since no distinction could be made in this study between the three types of female P. mxmtagui, we have avoided the use of this termin()lc)gy altogether and simply distinguish between individuals which re- mained as males during their first year, those which were females when they were first captured in November-December of their first year, and those which apparently changed sex during their first winter. other hand, only 11.2% of the age-1 P. montagui col- lected in May 1965 in Grand-Rivifere were females (Couture and Trudel 1969b). Size (Age) at Sex Transition Transitional P. montagui were collected during all five sampling periods, but were most abundant in late March (Table 6) following the end of the breeding period. Sex transition apf)arently began in Januar\', peaked in late March, and continued through the summer and early fall, reaching a minimum in late November. The rapid decline in the relative abundance of males after January 1981 and the accompanying increase in females after March (Table 6) indicated that shrimp, which functioned as males in the previous breeding season and became 226 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS females prior to the subsequent breeding season, assumed external female characteristics during the winter and spring. The appearance of a distinct size- group of 9-11 mm CL females in late January (Fig. 3F) which was not present 2 mo earlier (Fig. 3B), suggests that transition was well underway by late January. The two size-groups of non-ovigerous females in January and March (Fig. 31) were as- sumed to belong to the same age-group, the smaller females being those which did not function as males in their first year (they may have started life as females) and the larger females being those which were still males in November-December (Fig. 3A). Transitional shrimp made up a larger percentage of samples collected in moderate and deep waters and in areas 3 and 4 (Tables 7, 8). Transition of the younger age-group which was first captured in November-December (Fig. 3A) was incomplete since a great many shrimp remained as males for another entire year before undergoing Table 7.— Percent total number of male, transitional, and female Pandalus montagui collected at all depthis and times of year in four areas in Penobscot Bay during 1980-81. (Females are categorized by reproductive stage.) Sex/Stage Area 1 Area 2 Area 3 Area 4 Total Total males 82.3 63.4 67.1 77.9 68.1 Transitionals 1.6 3.0 11.5 12.3 7.1 Females/Stage 1 Females/Stage II Total 14.0 CO 25.4 2.6 16.8 0.4 5.5 0.0 18.8 1.3 non-ovigerous females 14.0 28.0 17.2 5.5 20.0 Ovigerous females 2.2 5.7 4.2 4.3 4.7 Total females 16.1 33.6 21.4 9.8 24.8 Total no. individuals 186 1,037 827 326 2,376 Table 8. — Percent total number of male, transitional, and female Pandalus montagui collected in all areas and times of year by deptfi range in Penobscot Bay during 1980-81. (Females are categorized by reproductive stage.) Shallow Moderate Deep Sex/Stage (12-25 m) (25-50 m) (50-85 m) Total Total males 82.4 61.6 75.3 68.7 Transitionals 1.1 5.8 7.1 5.9 Females/Stage 1 14.2 24.0 10.9 18.1 Females/Stage II 0.0 2.7 0.1 1.5 Total non-ovigerous females 14.2 26.6 11.1 19.6 Ovigerous females 2.3 5.9 6.6 5.8 Total females 16.5 32.6 17.6 25.4 Total no. individuals 176 1,013 732 1,921 transition during their third spring (at age 2). Given the fact that a few male and transitional shrimp > 15 mm CL were sampled at various times of year, the possibility that a few individuals do not change sex until their fourth year (age 3) could not be ruled out. Even though the transition of younger (age 1) shrimp in the spring was incomplete, a sizable number of non-ovigerous Stage I shrimp which com- pleted transition in their first year were collected in March (Fig. 3J). These new females were consider- ably larger (by about 2 mm CL) than their male counterparts, suggesting that it was the larger, faster growing, individuals which underwent transi- tion at age 1. Allen (1963) also reported that the largest 0 age-group males changed sex first. Older shrimp which changed sex in their second year had not yet appeared as females in March (Fig. 3J), suggesting that sex transition in older shrimp was delayed; it may also have been less rapid, par- ticularly since growth was considerably reduced after the first year. Earlier studies of P. montagui in- dicated that sex transition in two locations in the North Sea persisted for most or all of the year. Mistakidis (1957) noted that sex transition occurred from May to December in the Thames estuary whereas Allen (1963) collected transitional in- dividuals throughout the year in Northumberland, but primarily in June. The timing of minimal sex transition in Penobscot Bay and Northumberland was identical (November-December); Allen (1963) reported that this was when males were sexually ac- tive. Couture and Trudel (1969b) reported that most sex transition occurred in October at Grand-Riviere; a few transitionals were collected in July and August but none in May and June. Sex transition in the North Sea, as reported by Mistakidis (1957) and Allen (1963), was accelerated in comparison with Penobscot Bay; some individuals changed sex in their first year, but most did so in their second year. Ages at sex transition at Grand-Rivifere, on the other hand, were the same as in Penobscot Bay, i.e., some in their second year, most in their third year, and some in their fourth year. Seasonal Changes in Size (Age) Composition Two size-groups of male D. leptocerus were ob- served in the Penobscot Bay during the winter (Fig. 2 A, E, G) and only one in the summer (Fig. 2K); similarly, female length-frequency distributions in November-December were bimodal (Fig. 2B), but a single size-group was dominant in the summer (Fig. 2L). Presumably, most of the older (age 2) males either die or migrate out of the upper Bay in the 227 FISHERY BULLETIN: VOL. 83 NO. 3 spring since the younger age-group made up the en- tire population in the summer. Most of the females apparently spawn once and die after their eggs are hatched since very few older females were found in the winter. The presence of a few larger males in October (Fig. 2N) suggested that a few survive into their third fall and mate twice (or three times if they mature during their first year). Further evidence that most female mortality occurs following hatching while male mortality is delayed until later in the spring was indicated by the seasonal changes in relative abundance of males and females (Table 3): females decreased from 50 to 35% of the population between early December and late January while males decreased more slowly from 65% in January to 48% in the summer. Age-2 P. montagui which were either in transition or were still males in March had mostly become females by the summer (although a few transitionals and large males still remained). The single large size- group of first- maturing Stage I females in the sum- mer (Fig. 3M) presumably included age-1 females at a modal length of 13 mm CL and age-2 females at about 15 mm CL. At the same time, there appeared to be at least two size-groups of Stage II females in July-September (Fig. 3N) which had carried eggs the previous winter (Fig. 3C, G). Ovigerous females cap- tured in November-December 1980 (Fig. 3C) presumably included first- time spawners at 12-15 mm CL and one or two groups of repeat spawners at 15-19 mm CL. Excluding the single female at 10 mm, two or three age-groups of ovigerous females were apparent in January (Fig. 30). As indicated earlier, the relative abundance of the different age- groups in November-December and January showed that repeat spawners may have accounted for a larger percentage of the ovigerous females earlier in the winter. Growth and Longevity Male and female D. leptocents which hatched in the winter of 1979-80 reached 8.0-8.5 mm CL by Octo ber of their first year (Fig. 2N, P) and grew relative- ly slowly during their first winter; by March they had reached 7-10 mm CL and the females were slightly larger than the males (Fig. 20, H). This difference in size-at-age was also discernible in January (Fig. 2E, F). Orowth was rapid during the spring of the second year prior to the beginning of the breeding season: males increased about 3 mm in carapace length by the summer while females increased by 4 mm (Fig. 2K, L). The difference in modal lengths between males and females had increased further by October (Fig. 2N, P) as growth increased modal carapace lengths by an additional 0.5-1.0 mm for both sexes. Orowth between early October and late November when ovigerous females were first sampled was negligible, if October 1981 data can be compared with November 1980 data. During this same short period of time the relative abundance of the younger, newly recruited, age-group (males and females) in- creased dramatically. (A reduction in somatic growth can be expected at a time of rapid egg development since female growth ceases once their eggs have "dropped" and they are unable to molt). This species reached a maximum observed size of 19 mm CL (not included in compiled length frequencies), but most in- dividuals did not exceed 16 mm CL. Unless older in- dividuals migrate completely out of Penobscot Bay and were therefore not sampled during this survey, the maximum lifespan of Z). leptocents in the Bay ap- pears to be about 2 yr and 9 mo, although the bulk of the population apparently survives for only 2 yr. The first evidence of newly recruited 0 age-group P. montagui was in November-December 1980 (Fig. 3A, B). Individuals which remained as males during their second year grew from about 7-9 mm CL in November-December of their first year to 8-10 mm CL in March (Fig. 3H) and 10-13 mm CL in their sec- ond summer (Fig. 3K). As was observed for D. lep- tocerus, the growth rate increased in the spring. Age-1 shrimp which became females during their second spring reached 10-12 mm CL in March (Fig. 3J) and 12-15 mm CL in the summer (Fig. 3M). Com- parison of November-December 1980 and January 1981 data (Fig. 3C, 0) with summer 1981 data sug- gested that growth of mature females in the fall was negligible; the same was true for the males. The maximum observed size was 19.5 mm CL; females as large as 17-18 mm CL were collected in the winter (Fig. 3C, 0). These results suggested that P. mon^ tagui in Penobscot Bay normally spend 1-2 yr as males and as many as 3 yr as females. The maximum lifespan is probably 4 yr since shrimp that remain males for 2 yr do not complete sex transition until their third year and function as females in their third and fourth years. Growth at Grand-Rivifere was sufficiently slower that males there were 2-3 mm CL smaller by the end of their first year than they were in Penobscot Bay. This difference in growth could be a result of lower summer bottom temperatures in the Gulf of St. Lawrence. A temperature range of - 1°C (in May) to 3°C (in October) was reported in 54 m at Grand- Rivifere in 1965 (Couture and Trudel 1969a). Temperatures recorded in lower Penobscot Bay dur- ing the same months of the year at 40-60 m were 228 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS considerably higher, i.e., 3°C west of Islesboro in May 19706 and 11°-12°C at various stations in the lower Bay in August 1976''. In Penobscot Bay, P. montagui were smaller after their first year's growth than in the two North Sea locations (Mistakidis 1957; Allen 1963) but attained approximately the same size by the end of the second year. Males in the Thames estuary measured 10 mm average CL by November of their first year, and in Northumberland (at 40-60 m depth) they averaged 9.5 mm CL by October; tran- sitionals reached 12.5 mm CL by November of their second year in both locations as compared with 11-13.5 mm CL males of the same age in Penobscot Bay (Fig. 3A), while females in Northumberland reached 14.8 mm CL by November of year 2 as com- pared with 13-15 mm CL at the same age in Penobscot Bay (Fig. 3C). One-year-old females in Northumberland averaged 10.8 mm CL in October. Allen's (1963) explanation for this difference be- tween male and female lengths-at-age was that shrimp which mature as females in their first year do so 3-4 wk after males of the same age-group; since growth virtually ceases in the fall and winter, the dif- ference in length attained by the females in the first year is maintained into the third year of life. In Penobscot Bay, on the other hand, very few females mature in their first year; however, if males mature in their first year (this was not determined) and stop growing in the fall before the females, Allen's ex- planation might apply. It seems more likely that sex transition is a function of size, not age, and that the faster growing 0 age-group shrimp complete sex transition in their first year. Another possible ex- planation for the difference in size of females which change sex in their second and third years is that there may be two distinct periods of larval produc- tion and/or survival. Length-frequency data collected at two different periods during the winter (Fig. 3C, G) did suggest that older females may have spawned earlier than younger females. A 5-yr study of P. borealis in the Sheepscot River of Maine* failed, however, to reveal any consistent bimodality in lar- val production during February-April even though at least two age-groups of ovigerous females are com- «Muirhead, C. R., and J. H. Wartha. 1971. Temperature- salinity observations, Penobscot Bay, Maine, 1970. Open Data Rep. NOS DR-13, U.S. Dep. Commer., NOAA, Natl. Ocean Surv., Off. Mar. Surv. Maps, Oceanogr. Div., Descr. Oceanogr. Sect., Rockville, MD. 'Central Maine Power Co., unpublished data, courtesy Richard Birge, Environmental Studies Department, CMP, August, ME. *Stickney, A. P. Environmental physiology of northern shrimp, Pandalus borealis. Maine Dep. Mar. Resour., West Boothbay Har- bor, ME, Annu. Rep. 1981-82, 15 p. monly observed in commercial catch samples^. We contend, therefore, that the most plausible explana- tion for differences in the sizes of shrimp which become females in their second and third years is a difference in growth rates, especially since a similar difference in size was observed between male and female D. leptocerus in which sex remains fixed throughout life. Pandalus montagui which either began life as females or became females early in their first year were smaller than their male counterparts by November-December of their first year (Fig. 3A, B). It therefore seems probable that the smaller Stage I (6.5-8.5 mm CL) females captured in late January (Fig. 3F) did not grow as rapidly as the larger females in the same age-group which apparently completed sex transition in late December and early January or as shrimp which remained as males for the entire year (Fig. 3D). These differences in length-at-age between 0 age-group males and both groups of females were also evident in late March (Fig. 3H, J). The accelerated growth rate of 0 age- group individuals which changed sex during their first year contrasts with the reported faster growth of P. montagui in the North Sea and at Grand- Rivifere which began life as females (Allen 1963; Couture and Trudel 1969b). Winter Migration During November-December 1980, younger male and non-ovigerous female D. leptocerus were predominant in shallow water in area 1 (Figs. 4A, 5A) while the older age group predominated in deeper water (Figs. 4B, C, 5B, C). By late January, the older females were no longer being caught, and the older males had disappeared completely from depths < 50 m (Fig. 4D, E), but accounted for about 50% of the males collected in deep water in area 3 (Fig. 4F). There were older males in areas 2 and 3 (moderate and deep water) in March (Fig. 4G, H). These results suggested that older male shrimp migrated down the Bay into deeper water as the winter progressed and as bottom water tempera- tures dropped from about 8°C in early December to 1°C in late February and early March in the upper Bay.i° Their disappearance from the catches. 9Diodati, P., S. H. Clark D. Mclnnes, R. Tichko, and D. Sampson. 1983. Gulf of Maine northern shrimp stock status -1983. Northern Shrimp Technical Committee, November 1983, 9 P- i°Birge, R. P. 1982. Surface and bottom water temperatures, upper Penobscot Bay, Maine, March 1975 - December 1981. Cen- tral Maine Power Co., Environmental Studies Department, Report SI-82-3, 45 p. 229 FISHERY BULLETIN: VOL. 83, NO. 3 50- 40- 3 0- 20- 10- 10 0- 8 0- 6 0- 4 0- 2 0 NOV. 20- DEC 2 1980 n Area 1 Shallow N = 325 -1 1 1 1 1 1-" Area 2 Moderate Depth N-505 r — f ■ T r » f ~1 ' r- 40- Area 3 ^ 30- Deep N-230 ^ 2 0- - 10- ^T '— 1 J ■ r ^-HlH ^ - r^T Area 2 Moderate Depth N-55 -=^ 1 40 30 20 10- N0V20-DEC2 1980 8 0- 60 40- 20 Area 2 Moderate Depth N=385 JAN 21- 29 1981 3 0- p. Area 1 20- 1 — n Shallow N=1 1 7 10- Ln D 1 f T r r- 20- Area 3 Deep 1 0- N = 97 1 — ' L- r- J 1 ^ JAN 21-29 1981 —I 1 1- Area 3 Deep N-2 7 3 T 1 1 1 Area 2 Moderate Depth N-52 E T 1 1 1 I ?n- Area 3 , — ' Deep 10- N = 9 1 - r^ L^ --- ,- F 1 T — 1 80- 60- 4 0- 2 0- 2 0- 10- — I 1 ' ' 1 \ 1 ' r~ MARCH 24-31 1981 Area 2 Moderate Depth N=423 MARCH 24-31 1981 Area 2 Moderate Depth N=290 2 0- 10- r^n Area 3 Deep N-102 -. F=»— . H . » ' r- -i-F=i- 5 6 7 8 9 1011 12131415 16 17 CARAPACE LENGTH (mm) CARAPACE LENGTH (mm) Figure 4. -Male DichelopandaLus ieptoceru.s numerical length- frequency distributions by area and depth range. 230 STEVENSON and F'lERCE: PANDALUS MONTAGU! AND DICHELOPANDALUS LEPTOCERUS however, may still have been a result of mortality rather than migration. The depth-area length- frequency data also revealed that slightly larger shrimp of both sexes and age-groups were captured in deeper water further down the Bay, suggesting that larger individuals are more likely to migrate than smaller ones. Older male P. montagui also migrated down the Bay into deeper water during the winter: older age-1 males were more abundant than younger 0 age- group males in areas 2 and 3 (moderate and deep water, respectively) in November-December 1980 (Fig. 6A, B, C) and by late January nearly all the older males were in deep water in area 3 (Fig. 6D, E, F), but were not as abundant as the younger males in deep water in area 4 (Fig. 6G). There was no clear evidence that the larger males in either age-group 2 0J NOV 20- DEC 2 1980 Area 1 ~~ Shallow 10- r^ L-, r N=10 0 A ^ 11 ZO- , — =1 — I — ^ r^ n Area 2 Moderate Depth IC- N = 9 1 ^ B 1 ■ — 1 40- 30 20 10 Area 2 Moderate Depth N=255 ^^TL Area 4 Deep N=ei ■^ 1 r1 I n 10 11 12 13 14 15 16 17 CARAPACE LENGTH (mm) Figure 6. -Male Pandalus montagui numerical length-frequency distributions by area and depth range. were more frequent in deeper water and not enough females were collected during any single sampling period to permit an analysis of differential distribu- tion of different age-groups by depth. Allen (1963) reported an offshore migration of age-1 males to deeper water in the spring and sum- mer in Northumberland waters; Couture and Trudel (1969a) observed the same phenomenon in Grand- Rivifere in the summer as the temperature exceeded 6°C and reported that it was triggered by the onset of maturity; earlier maturing males migrated sooner than later maturing males, indicating that migration was a function of size, not age. Mistakidis (1957) and Allen (1963) both reported an offshore migration of P. montagui females in the fall; Allen, however, reported that the smaller females stay behind to spawn in shallow water and that there was no "massive" return migration in the spring whereas Mistakidis reported a general offshore fall migration of females and a return migration in the spring. Allen (1963) reported that females in Northumber- land which survived to spawn a third time were only found in depths > 100 m. CONCLUSIONS This study of the life histories of Pandalus mon- tagui and Dichelopandalus leptocerus in Penobscot Bay has revealed some notable differences in repro- ductive characteristics and (apparently) in longevity between the two species. On the other hand, growth rates and migratory behavior were similar. Most importantly, P. montagui is hermaphroditic; D. leptocerus is not. In 1980-81, some transitional P. montagui were observed during all five sampling periods, but were most common in the early spring; most individuals change sex at the end of their sec- ond year but some do so during their first year and a few may not change sex until their third year. Some individuals apparently begin life as females; 15% of the 0 age-group which recruited to trawl catches at 5-10 mm CL in November-December 1980 were females. Ovigerous P. montagui were collected over a longer time period, owing, at least in part, to the presence of more age-groups in the spawning popula- tion. Most D. leptocerus females spawn during their second year, although a few also spawn in their first year; ovigerous females were collected primarily in late November-early December. Pandalus montagui spend 1-2 yr as males; individuals which undergo sex transition in their second year may function as "emales during their second, third, and fourth years /hile those which change sex in their third year may only function as females for 2 yr. Judging from the 231 FISHERY BULLETIN: VOL. 83, NO. 3 length-frequency data collected during this study, it is unlikely that very many P. montagui survive in the Bay beyond age 4. No D. leptocenis females older than age 2 were caught in the Bay although some males remained until their third fall (age, 2 yr and 9 mo). Conclusions concerning longevity were com- plicated by the fact that larger individuals of both species migrated down the Bay into deeper water as the winter progressed and were not captured and by the difficulty of inferring age from length data for the larger size-groups, particularly for P. montagui. Both species were similar with respect to growth and migratory behavior. Growth decreased with in- creasing age and was seasonal, i.e., rapid in the spring and summer of the first year and the spring of the second year and negligible in the fall and winter. Males of both species reached 7-9 mm CL at age 1 and 11-13 mmCL at age 2. The data suggested that early (age 1) transitional P. montagui also grow con- siderably more rapidly than the remainder of their age group which undergo sex transition at age 2. The slowest observed growth rate was for P. montagui which apparently begin life as females and therefore never undergo sex transition. For D. leptocerus, females grow more rapidly than males and differen- tial growth is a "fixed" sexual attribute. For P. nwn- tagui, the fact that the faster growing individuals change sex a year earlier and therefore function as females for an additional year (assuming that fast- and slow-growing shrimp have identical lifespans) means that more rapid growth and early sex transi- tion increase the reproductive potential of the population, as long as enough males remain in the population to mate with the females. Female maturi- ty is clearly a function of size, not age. The life cycle of P. montagui in Penobscot Bay was quite different in several respects from the life cycles of populations which have been studied in the North Sea and at Grand-Riviere, Quebec. Pandahis mon- tagui populations in two locations in the North Sea (Mistakidis 1957; Allen 1963) appeared to grow more rapidly in their first year of life than in Penobscot Bay and were composed of a considerably greater proportion of early maturing females, many of which never functioned as males. Growth over the entire lifespan was considerably more rapid in Penobscot Bay and the North Sea than at Grand-Riviere (Couture and Trudel 1 969b), but the relative scarcity of females which do not function first as males and the delay of most sex transition until the third year were common to the Canadian and United States populations. In addition to a difference in the timing of maximum sex transition from age 2 (Penobscot Bay and Grand-Riviere) to age 1 (North Sea), the seasonal intensity of sex transition was not the same in three of the populations. Most transitionals were observed in March in Penobscot Bay, in June in Northumberland (Allen 1963), and in October at Grand-Riviere (Couture and Trudel 1969b). Eggs were carried by females in both North Sea locations and in Penobscot Bay during the winter (November- March); at Grand-Riviere most ovigerous females were observed in October. In Penobscot Bay and the North Sea, sex transi- tion tended to follow the end of the breeding season, whereas in Grand-Riviere, maximum sex transition coincided with the time when most females were carrying eggs (unless sex transition was more com- mon later in the fall when no samples were collected), suggesting that there was a 12-mo inter- val between the appearance of external female characteristics and spawning at Grand-Riviere, and a 6-9 mo interval in Penobscot Bay and Northumber- land. More rapid growth rates in the latter two loca- tions would explain the shorter time intervals between sex transition and spawning. The reproduc- tive cycle in Grand-Riviere was seemingly con- tinuous, beginning in July and ending in June (Couture and Trudel 1969b). Although an offshore migration of larger male P. montagui was observed in Penobscot Bay, Grand- Riviere, and Northumberland, this migration oc- curred in the winter following the end of the spawn- ing season in the Bay and in spring and summer, prior to spawning, in the other two locations. Similar movements of larger females have been noted in both North Sea populations in the fall. Unlike the other migrations, the one observed in Penobscot Bay was not a spawning migration and may instead have been a response of older shrimp to declining winter temperatures in the shallower waters of the upper Bay. The departure of significant numbers of older shrimp from the Bay could certainly affect any con- clusions concerning the size or age structure of either population and their estimated maximum lifespans. ACKNOWLEDGMENTS The authors wish to acknowledge Frank Spencer, Director of the Fisheries Technology Division, and Penn E stab rook. Director of the Bureau of Marine Development, both of the Maine Department of Marine Resources (DMR), whose support made this project possible. Thanks are also due to Mike Brown, formerly a DMR employee and leader of this project; to Mike Dunton, the captain of the RV Explorer and to the late Paul DeRocher, captain of the RV 232 STEVENSON and PIERCE: PANDALUS MONTAGUI AND DICHELOPANDALUS LEPTOCERUS Challenge; to Curt Crosby for assisting with the field work; to Margaret Hunter, who supervised the com- puter analysis of the data; to Pat Hoyt and Vicki Averill, who typed many copies of the manuscript; to Patti Millette and Kim Knowlton who drafted and photographed the figures; to Richard Birge of Cen- tral Maine Power Company who kindly supplied us with temperature data for Penobscot Bay; and to Steve Clark of the Northeast Fisheries Center Woods Hole Laboratory, National Marine Fisheries Service, for his constructive review comments. This research was supported by NOAA Grant NA- SI AA-D-00035 to the University of Maine and the University of New Hampshire Sea Grant College Program and by the Fisheries Technology Division of the Maine Department of Marine Resources. LITERATURE CITED Allen, J. A. 1959. On the biology of Pandalus borealis Kr(?yer, with refer- ence to a population off the Northumberland coast. J. Mar. Biol. Assoc. U.K. 38:189-220. 1963. Observations on the biolog>' of Pandalus ymmtagui [Crustacea: Decapoda]. J. Mar. Biol. Assoc. U.K. 43: 665-682. Balsiger, J. W. 1981. A review of Pandalid shrimp fisheries in the northern hemisphere. Proc. Int. Pandalid Shrimp Symp., Univ. Alaska Sea Grant Rep. 81-3:7-35. Berkeley, A. A. 1930. The post-embryonic development of the common Pandalids of British Columbia. Contrib. Can. Biol. Fish., New. Ser. 6:79-163. Couture, R., and R. Trudel. 1969a. Biologie et ecologie de Pandalus mimtag^n Leach (Decapoda Natantia). I. Distribution et migrations, a Grand- Riviere (Gaspe), Quebec. Nat. Can. 96:283-299. 1969b. Biologie et ecologie de Pandalus montagui Leach (Decapoda Natantia). IL Age, croissance et reproduction. Nat. Can. 96:301-315. Jagersten, G. 1936. Uber die Geschlechtsverhaltnisse und das Wachstum bei Pandalus. Ark. Zoo). 28A(20):l-26. Leloup, E. 1936. VL - Les transformations des gonades et des carac- t^res sexuels externes chez Pandalus montagui Leach (D^capode). Bull. Mus. Hist. Nat. Belg. 12(19):l-27. McCrary, J. A. 1971. Sternal spines as a characteristic for differentiating between females of some Pandalidae. J. Fish. Res. Board Can. 28:98-100. Mistakidis, M. N. 1957. The biology of Pandalus montagui Leach. Fish. Invest, Minist. Agric, Fish. Food, (G.B.), Ser. II, 21(4), 52 p. Rathbun, M. J. 1929. Canadian Atlantic fauna. 10. Arthropoda. 10 m. Deca- poda. Biol. Board Can., Atl. Biol. Stn., St. Andrews, N.B., Can., 38 p. Scattergood, L. W. 1952. The northern shrimp fishery of Maine. Commer. Fish. Rev. 14(1):1-16. Simpson, A. C, B. R. Howell, and P. J. Warren. 1970. Synopsis of biological data on the shrimp Pandalus montagui Leach, 1814. FAO Fish. Rep. 57:1225-1249. Squires, H. J. 1966. Distribution of decapod Crustacea in the northwest Atlantic. Ser. Atlas Mar. Environ., Am. Geogr. Soc. Folio 12. 4 p. 1968. Some aspects of adaptation in decapod Crustacea in the north-west Atlantic. Fish. Res. Board Can. Stud. 1260, p. 215-223. Warren, P. J. 1973. The fishery for the pink shrimp Pandalus montagui in the Wash. Minis. Agric, Fish. Food., Lab. Leafl. (New Ser.) 28, Suffolk, 46 p. Fisheries Laboratory, Lowestoft, Engl. Wigley, R. L. 1960. Note on the distribution of Pandalidae (Crustacea, Decapoda) in New England waters. Ecolog>' 41:564-570. Wollebaek, a. 1908. Remarks on decapod crustaceans of the North Atlantic and the Norwegian fjords. Bergens Mus. Aarb. 12:1-77. 233 I i VISUAL THRESHOLD FOR SCHOOLING IN NORTHERN ANCHOVY ENGRAULIS MORDAX John Hunter and Ragan Nicholl' ABSTRACT The visual threshold fur schooling was determined for two groups of 50 adult northern anchovy in the laboratory. The index of dispersion and the mean distance to the nearest neighbor were used to measure changes in schooling as a function of light intensity. The threshold light intensity for schooling, (i x 10" " Wcm''-(2.6 X 10"'' mc), was estimated to occur at a depth of 30 m on a starlit night and at 38 mduringa full moon, when the chlorophyll concentration is 0.2 mg Chi a ni"'^. At 2.0 mg Chi a ni"-^ the threshold oc- curs at a depth of 8 m on a starlit night and at 20 m under full moon light. Sufficient light appears to exist at night within the upper 10 m for schooling to occur in most of the habiUit of the anchovy. The vertical distribution of newly spawned anchovy eggs indicated that the maximum depth of spawning may be similar to the maximum depth of schooling and that the visual threshold for schooling could be used to forecast max- imum spawning depth in the sea. Vision plays a primary role in the maintenance of most fish schools in the sea. Other sense organs, par- ticularly the lateral line, are important in coor- dinating movements and spacing of fish within the school (Pitcher et al. 1976), but it is unlikely that lateral line sense alone is sufficient for maintaining the integrity of schools at night in the sea. In fact, a large number of laboratory studies indicate that if light is sufficiently reduced, fish no longer maintain schools (Whitney 1969; Blaxter 1970). Thus the visual threshold for schooling and the depth of pene- tration of light probably determine the maximum depth at which pelagic fishes are able to school at night in the sea. Our objective was to determine the visual thresh- old for schooling in adult northern anchovy, Engraulis mordax, and to use this information to forecast the maximum schooling depth for anchovy at night in the sea. This calculation is of ecological in- terest because the maximum depth for schooling is probably also the maximum depth for spawning and for nocturnal feeding. Anchovy spawn only at night and visual recognition of other fish is probably as essential for spawning as it is for schooling. During what we believe was spawning behavior, several males rapidly pursued a female over an irregular path, a tactic probably not possible using senses other than vision. This is a casual laboratory observa- tion and requires further documentation, however. 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La .Jolla, CA 92038. Northern anchovy feed at night as well as in the day (Loukashkin 1970; Hunter and Kimbrell 1980). Light may not be necessary for filter feeding, but it is essential for particulate feeding, although somewhat higher light levels may be required for feeding than for schooling (Hunter 1968). Anchovy migrate toward the surface at sunset (Mais 1974), and schools are less frequently detected at night using sonar and not detected with echo sounders (Smith 1970). Nevertheless, no doubt exists that schooling con- tinues at night because the anchovy fisher^' is typi- cally a night fishery and because profiles of schools are detectable at night owing to the bioluminescent disturbances their movements produce (Squire 1978). METHODS Apparatus and Laboratory Procedures Two groups of 50 northern anchovy (group 1, mean length = 10.5 cm SL; group 2, = 9.8 cm SL) were maintained in a 4.6 m diameter tank supplied with running seawater (group 1 , mean temperature = 16.9°C; group 2, 21.0°C). To simplify photo- graphic analysis the school of 50 fish was constrained to a somewhat two dimensional form by maintaining them in water 45 cm deep. The fish were fed adult Artemia at the daily time of 1000, and the tank was cleaned 1 h after feeding. The tank, which was constructed of blue vinyl, was located in a light tight rectangular enclosure in which the walls and ceiling were covered with white vinyl Manuscript accepted August 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 235 FISHERY BULLETIN: VOL. 83, NO. 3 to diffuse the Jight. Four light sources were equally spaced around the periphery of the tank; the top of each being just below the tank rim. Each source con- sisted of a 30 W tungsten microscope lamp with reflector enclosed in a tube. On top of the tube were two color filters; a green acrylic plastic filter (#2414, Rohm and Haas), and a sealed petri dish containing a 5% CUSO4 solution (by weight). To diffuse the light a translucent white acrylic filter (#W-2447, Rohm and Haas) was placed on top of the color filters and light transmitted by the white filter entered a white opal glass globe (13 cm diameter) that formed the top of the source. The lamps were operated for 5 h (10% of lamp life) before they were used to produce test levels of irradiance. These sources were used to pro- duce four test levels of irradiance during a 12-h night. Additionally, four tungsten 100 W household lamps (unfiltered) with reflectors were used for the daytime level of irradiance. These lamps were placed at regular interv-als around the perimeter of the tank near the ceiling. Light from these lamps reflected off the ceiling providing a uniformly diffuse illumination. The spectrum produced by the four sources (Table 1) resembles the greenish spectrum tj^jical of anchovy habitat, but the spectrum used in the day was not dif- ferent from a standard curve for a tungsten lamp and consequently contained an unnaturally high pro- portion of longer wavelength energy. Our term for the condition when all lamps were off was darkness; under these conditions light was not detectable by a dark adapted human observer and the irradiance was below the sensitivity of a 931 A photomultiplier which can detect about 5 x lO-*' mc (meter candle). To record the effect of light on the schools the fish were photographed from above the tank using a 35 mm automatic camera and flash attachment. The camera was controlled by a timer, and photographs were taken at 30-min intervals for 5 h during the 12-h day, at night in darkness, and at night at the test levels produced by the four sources. Night photographs were taken during a 5-h period com- mencing 2 h after the end of the 12-h day. Ten photographs were usually analyzed at each light level for each group, but in several tests, 1 or 2 photo- graphs were not analyzed because not all 50 fish could be seen. Two indices of schooling were calculated for each photograph: an index of dispersion (Pielou 1969), and the mean distance to the nearest neighbor (Hunter 1966). The dispersion index was calculated by super- imposing a grid containing 326 quadrats over the projected image of the tank and counting the number of fish occurring in each quadrat. The variance mean ratio (s'^/x) for the number of fish per quadrat was the index of dispersion. The index was calculated for each photograph, and an average index was com- puted for each light treatment (n = 8-10 photo- graphs). A dispersion index of 1 indicates a random distribution, whereas higher values indicate aggrega- tion (Pielou 1969) and imply the existence of school- ing. Values < 1 imply a uniform distribution over the grid. The mean distance to the nearest neighbor was computed for a random subsample of 10 fish in a photograph. All 50 fish in a photograph were numbered and the subsample of 10 was selected by drawing the fish numbers from a table of random numbers. For each of the 10 fish in the subsample the distance in centimeters to its nearest neighbor was measured (distance between heads), a mean distance calculated for each photograph, and means Table 1.— Spectral functions used to estimate the depth of occurrence of the visual threshold for schooling under the various water types and incident irra- diances including spectral irradiance in the laboratory apparatus, moonlight at 3 m below water surface, starlight at the earth's surface (Munz and McFarland 1977), and the relative sensitivity of the dark adapted anchovy eye (Engraulis encrasicholus, Protasov 1964). Wavelengths Energy per 25 nm interval (nm) (W cm") Relative sensitivity Interval Laboratory Moonlight Starlight Mean luminaires luminaires at 3 m depth at surface anchovy eye 400 400-412 3.411 X 10-" 425 413-437 1.346 X 10" 6.109 X lO'" 1.978 X 10" 0.16 450 438-462 3.402 X 10" 6.689 X 10'° 2.418 X 10" 0.53 475 463-487 6.822 X 10 " 6.820 X 10 '° 2.291 X 10 " 0.85 500 488-512 9.631 X 10 " 7.072 X 10-"> 2.374 X 10" 1.00 525 513-537 1.069 X 10'° 7.067 X 10'° 2.449 X 10" 0.75 550 538-562 9.618 X 10 " 7.283 X 10 '° 5.125 X 10" 0.42 575 563-587 5.297 X 10 " 6.881 X 10 '° 2.838 X 10" 0.22 600 588-612 '9.494 X 10'^ 5.481 X 10'° 3.874 X 10" 0.08 'Wavelength Interval = 588-600 nm. 236 Hl'NTKR and N'lCHOLL: NORTIIKRN ANCHdXV SCIIOOI.INC TIIKKSIIOLD for each lig-ht treatment. These measurements are indices of only one characteristic of a fish school, the tendency of individuals to maintain contact with each other and thereby remain in a social group. The polarization of individuals in a school is frequently in- cluded in definitions of schooling as the cohesive movements of a school require that fish maintain relatively constant headings and individual distances which gives the polarized appearance of moving schools. This characteristic of fish schools was not measured in our study, thus the visual threshold we estimated was one for the maintenance of schooling in the broadest sense, that is the existence of a group maintained by visual attraction. Radiometric Procedures, Calibrations, and Computations Radiometric equipment used in this study con- sisted of an Optronics spectroradiometer (model 714-V) (calibrated against a radiometric standard) and a photometer (Gamma Model 700). The spectro- radiometer was used to measure the spectral irra- diance produced by the sources at full lamp output but the treatment levels of irradiance were below the sensitivity of the spectroradiometer. Absolute measurements of light intensity were made 25 cm above the water surface (the difference between this position and within the water would be < 10%). The light treatments were varied by placing neutral den- sity filters in each of the light sources; the neutral density filters were calibrated on an optical bench. Test levels we used were computed from the filter factors for the neutral density filters. The photo- meter was used to check irradiance levels prior to a test, but we believe the computed values to be more accurate. Irradiance distribution in the tank was mapped using the photometer and the treatment values weighted by tank area so that they represent- ed the average irradiance 25 cm above water surface. Our laboratory estimates of the visual threshold for schooling were used to calculate the maximum possible depth of schooling in the sea for various levels of incident irradiation and water types. Threshold values in W cm"^ were converted to an- chovy effective units (W cm-'^^nch. eff.) by weighting the spectrum in the apparatus by the relative sensi- tivity of the scotopic curve of the anchovy Engraulis encrasicholus from an electroretinogram by Pro- tasov (1964) (Table 1). Two levels of night illumina- tion were used, full moon at 3 m depth (2.78 x 10 ~^ W cm -2), and starlight at the earth's surface (1.08 x 10" 10 W cm"^) (both measurements from Munz and McFarland 1977). The depth (meters) in the sea (Z) at which a given threshold (E,) value occurred was calculated using the equation of Baker and Smith (1982): In where £■„ is the incident radiation (full moon or star- light) in anchovy effective units. K, is the wave length specific attenuation coefficient and is the sum of coefficients for pure water (K,,), dissolved organic matter iK,j), and chlorophyll a (K,). Tables of coeffi- cients, and equations for calculating these attenua- tion coefficients, are given by Baker and Smith (1982). In our calculations we assumed that the dissolved organic matter was constant at 0.7 mg 1 ~ ' which is typical of the anchovy habitat. We calcu- lated K^ for a range of chlorophyll (Chi) a concentra- tions ranging from 0.1 to 10 mg Chi a m -^ and at 25 nm intervals from 425 to 600 nm for each Chi a con- centration. Each K, value for 25 nm increments was weighted by the appropriate anchovy scotopic sensitivity, and the average anchovy weighted ^t anch. eff. ^as used in the final calculation of Z. Many uncertainties and possible biases exist in such an extrapolation from laboratory- to sea condi- tions: Cloud cover was not considered nor were possible effects of bioluminescence; spectral irra- diance values for full moon and starlight of Munz and McFarland (1977) may not be representative of con- ditions in the anchovy habitat although they are relatively close to those given in photometric units by Brown (1952); variation in dissolved organic mat- ter is not considered; the radiance distribution over 360° in the tank probably does not resemble that in the sea (only downwelling irradiance was considered here); use of the action spectrum based on an electro- retinogram of a dark adapted E. encrasicholus eye instead of one for schooling of E. mordax; and of course, the usual statistical uncertainties. Despite these uncertainties and biases we believe our esti- mates of schooling depth are the most accurate to date thanks to the models developed by Baker and Smith (1982). RESULTS AND DISCUSSION The schooling threshold based on the index of dispersion was between 4.8 x 10"^^ and 7.8 x 10" '^ W cm-2 (Fig. 1; Table 2). At the lower value and in darkness the index of dispersion (s^/x fish per 237 FISHERY BULLETIN: VOL. 83. NO. 3 2.4 - 2.3 1.7 - 1.6 1.5 - 1.4 13 1.2 1.1 1.0 0.9 •- 2xSE • -Group 1 o- Group 2 ci SCHOOLING THRESHOLD (Geometi TIC mean) T 1 ir RANDOM "Dark" 10" '2 10"" 10''° 10'^ 10-8 "Daylight" IRRADIANCE Wcm"^ FiciKK I.-\'isual threshold for schooling in northern anchovy. Schooling is indicated by the index of dispersion (s^/x)- A ratio of 1 implies no schooling as it indicates a random distribution. Each point is a mean calculated from 8-10 photographs and bars are ± 2 X standard error of the mean. No error bars are given for one value (2.34 ± 0.47) because it falls far beyond the rest of the values. Mean dispersion is shown for four test levels of downwelling irradiance (log scale), "dark" (below the sensitivity of a 931 A photomultiplier). and "daylight" (1.496 x 1 0 -■'' W cm " 2) which was the norma! day- time level of irradiance in the apparatus. quadrat) did not differ from unity indicating that the fish were randomly distributed. Hence no schooling existed at the lower irradiance value and in darkness, whereas at the higher value the fish were clearly aggregated. These values delimit a region of about 2 log units of irradiance where one cannot be certain if schooling occurs or not. The actual threshold for schooling must fall somewhere in that region, and we have arbitrarily considered the threshold value to be the geometric mean irradiance of the above two irra- diances (6 x 10"^^ W cm"-) thereby reducing the uncertainty in the threshold value to about 1 log unit. In subsequent tables and figures we give the upper and lower bounds of the region as well as the threshold value, however. The variance of the dispersion index, a measure of the variation in school dispersion among photo- graphs, increased sharply at irradiances above the threshold indicating a wide variation in the disper- sion of fish among photographs. This can be expected because schooling fish react to fright stimuli, feed- ing, and many other conditions by altering interfish distances, thereby changing the cohesion or degree of dispersion of the school (Blaxter and Hunter 1982). At light levels below the visual threshold, fish are unable to respond socially to such stimuli, hence the variation among photographs is low. Mean distance to the nearest neighbor followed the same pattern as we have described for the index of dispersion. Values in darkness and at the lower Table 2.— Mean and standard deviation of the Index of dispersion and mean distance to the nearest neighbor for various irradiance levels. Schooling indices Mean distance Dispersion to nearest Irradiance (W cm 2) ir (- dex 5'/X) neighbor (cm) Number of Group X S X S photographs 1.496 X 10' ("daylight")' 1 1.28 0.29 17.50 3.71 10 1 2 1.52 1.28 0.30 0.11 12.15 24.61 4.10 4.94 10 10 '8.92 X 10' 1 2 1.25 2.34 0.10 0.74 12.19 19.49 2.52 5.50 10 10 M.079 X 10' 1 2 1.17 1.34 0.09 0.27 16.65 21.72 3.86 7.14 10 10 M.777 X 10 '" 1 2 1.07 1.49 0.15 0.21 18.48 18.05 3.86 6.62 9 10 '7.785 X 10 " 1 1.05 0.09 21.87 5.39 8 2 0.99 0.07 30.40 5.14 10 1 1.04 0.12 21.20 4.16 9 Dark' 1 1.04 0.12 19.61 4.83 9 2 0.98 0.07 34.12 6.36 10 'Unfiltered tungsten lamp. 'Filtered lamp - spectrum given In Table 1. 'Below sensitivity of 931A photomultiplier. 238 HINTER an.1 NRHOl.l.: NdRTHKRN ANCHOVY SCIIOOl.INC THRESHOU) 40 r <- tr UJ o O 03 5 X < O 30 ?^Z 20 10 •-Group 1 o-Group 2 :2xSE •1^ "Dark" 10" 10" 10 10 10" 10"^ "Daylight" IRRADIANCE W cm-^ Figure 2. - \'isual threshold for schooling in northern anchovy is in- dicated by changes in the mean distance to nearest neighbor. Each f)oint is a mean calculated from 8-10 photographs and bars are ± 2 X standard error of the mean. Mean distance to the nearest neighbor is shown for four test levels of irradiance (log scale), "dark" (below the sensitivity of a 931 A photomultiplier), and "daylight" ( 1 .5 X 10 ~ ■"• W cm - ") which was the normal daytime level of irradiance in the apparatus. type and incident li^ht intensity. An order of mag'ni- tude decline in incident irradiation can be expected under the darkest storm clouds (Brown 1952); under these conditions schooling may not be possible at the highest chlorophyll concentrations. It seems appropriate to use these visual thresholds as estimates of the maximum depth of spawning because spawning probably also depends upon the ability of anchovy to see one another. We calculated the vertical distribution of newly spawned anchovy eggs (0-4 h old, type "S") using data from the un- published vertical distribution study of Pommeranz and Moser (1983). We selected sets of vertical sam- ples at two stations for which surface Chi a concen- trations had been measured and then calculated a mean Chi a concentration for an inshore and offshore series of net hauls. We then estimated the maximum depth for schooling assuming that the surface Chi a was equivalent to an integrated value for the water column as required by the Baker and Smith 1982 model. Spawning occurred closer to the surface at the inshore station which had a high Chi a concentra- irradiance values were not statistically different. Above the threshold the mean distance to nearest neighbor was lower than that in darkness indicating closer spacing among pairs, but no trend with light intensity seemed to exist above the threshold range (Fig. 2). For the purpose of estimating a threshold, however, we believe the index of dispersion is prefer- able because the criterion for randomness is well defined and the dispersion index takes into account all 50 fish, whereas we used only 10 random pairs per photograph for the nearest neighbor measurements which reduced its precision. Our calculations indicate that in water of low chlorophyll concentration (0.2 mg Chi a m'-^) the threshold irradiance for schooling occurs at a depth of 38 m during a full moon and at a depth of 30 m on a starlit night (Fig. 3). The method of calculation is described in the Methods section on radiometric pro- cedures. Light attenuates rapidly as Chi a concentra- tion increases up to about 2 mg m"'^ total chloro- phyll; at 2.0 mg Chi a m"'^ the schooling threshold occurs at a depth of only 8 m on a starlit night and at 20 m under full moonlight. Above 2 mg Chi a m"^ light attenuates more slowly with increasing Chi a concentration with the threshold at 10 mg Chi a m"'^ falling at 5 m in starlight and at 12 m in full moon- light. These calculations indicate that sufficient light exists at night for northern anchovy to school within the upper 10 m of nearly all habitats under clear skies, but the maximum possible depth of the school- ing would be expected to vary greatly with water Surfacer 10 'g 20 a. 30 LU Q 40 50 Surface 50 L^ FULL MOON LIGHT SCHOOLING NO SCHOOLING STAR LIGHT SCHOOLING 0.1 0.5 1.0 1.5 2.0 2.5 Chi a (mgm"3) 3.0 10.0 Figure 3. - Maximum depth of schooling of northern anchovy in waters of various chlorophyll concentrations (Chi a) under starlight and full moon. Coefficients used in calculations are in Table 1 and in Methods section. Darkly shaded area indicates proportion of water column where no schooling is expected, lightly shaded area indicates depth range of schooling threshold. Centra! dotted line is the geometric mean. 239 FISHERY BULLETIN: VOL. 83, NO. 3 tion (1.5 mg m"-^) than at the offshore station which had a lower concentration (0.24 mg Chi a m"'^). At the onshore station only 4% of the eggs occurred below 20 m, whereas at the offshore station 31% were below 20 m. This difference is particularly striking because the inshore samples were taken under a full moon, whereas the moon was in the first quarter when the offshore station was occupied. At both stations the predicted maximum depth for schooling was close to the observed maximum depth for newly spawned eggs (Fig. 4). We may have underestimated the depth of schooling for the off- shore (low Chi a) station as we used a starlight value of Munz and McFarland (1977) because no data ex- isted for 1/4 moon. Spawning occurred prior to moonset since spawning occurs between the time of 1800 and 2400 and moonset varied from about the time of 2130 to 0200 (19-25 March 1980). In addi- tion, the offshore station had a deeper mixed layer (about 35 m) than the inshore station (about 10 m) and vertical distribution of anchovy eggs and larvae also may be affected by the depth of the mixed layer (Ahlstrom 1959). Regardless of these uncertainties, these data indicate that underwater visibility may set the maximum depth for spawning of anchovy. although other factors, such as low temperature, might constitute an additional barrier to spawning schools. Thus fish visual thresholds may be a conve- nient way to establish a general function for esti- mating the maximum depth of spawning for anchovy and perhaps other pelagic spawning clupeoids in all habitats. Such a general function, that could account for much of the variation in the maximum depth of eggs, could be quite useful in three dimensional models of larval transport or predation. A spawning- depth, water-type function based on visual thresholds seems particularly attractive owing to the considerable cost of accurately measuring the ver- tical distribution of eggs and larvae even in a single habitat let alone the cost for estimating it for all possible spawning habitats of the population. To compare the northern anchovy schooling threshold to literature values we converted our radiometric measurements to lux or meter candles (mc), by weighting the spectral irradiance in the apparatus by the human photopic curve, as the literature values are largely in photometric units (see reviews by Whitney 1969 and Blaxter 1970). The visual threshold for anchovy schooling (2.6 x 10^^ mc, Table 3) is about an order of magnitude higher Surfacer 10 ^ 20 E t 30 UJ Q 40 50 1.5 mg Chi a m 3 FULL MOON SCHOOLING THRESHOLD 0.24 mg Chi a m"3 FIRST 1/4 MOON ^ SCHOOLING J THRESHOLD NO SCHOOLING ' ' ' ' ' ' ' ' ' 0 20 40 60 80 NO SCHOOLING I I ' I 0 20 40 60 PERCENT OF NEWLY SPAWNED EGGS ^ Depth of mixed layer Figure 4. -Comparisons of the estimated depths of schooling of northern anchov^' and the observed depths of spawning. Estimated depth of schooling calculated from visual threshold estimates (W cm-2^^^|^ ^^^^^ an assumed dissolved organic matter of 0.7 mg 1 ~ ', and the average Chi a concentration and moon phase at the station (1/4 moon phase assumed to be equivalent to starlight) using the model of Baker and Smith (1982). Observed spawning depths at the two stations are indicated by a frequency histogram for newly spawned anchovy eggs where the y axis indicates the depth stratum of the plankton tow and the x axis indicates the percentage of newly spawned eggs taken at each of the 10 m vertically stratified tows. Data are from Pommeranz and Moser (1980) atifl are for the total number of newly spawned eggs taken over a 4-8 d interval. 240 HUNTEK and NICHOLL: NORTHERN ANCHOVV SCHOOl.INC THRESHOLD Table 3. — Upper and lower bound and geometric mean for the visual threshold for schooling of adult northern anchovy, Engraulis mordax, in the various energy units. Energy units Schooling' No schooling^ Geometric mean^ 7.785 X 10 " 6.051 x 10 Radiometric (W cm ') 4.777 x 10 Anchovy effective' (W cm 'anch. eff.) 3.079 x 10 '° 5.018 x 10 '^ 3.900 x 10 " Photometric* (mc) 2.048 x 10' 3.337 x 10' 2.594 x 10' 'Lowest irradiance level at which schooling occurred. 'Highest irradiance level at which anchovy failed to school. 'Geometric mean of the irradiance at the upper and lower bounds of the threshold. 'Weighted by ERG action spectra for Engraulis encrasicholus dark adapted retina (413-612 nm) (Protasov 1964). ^Weighted by the 1964 human photopic response (413-600 nm). than that for jack mackerel (3.5 x lO"'' mc, Hunter 1968), a species associated with anchovy in the California Current. Visual thresholds for schooling in fishes range from about 1 x 10"-^tol x 10"' mc with about 90% (14/16) of the literature values being higher than anchovy (Blaxter 1970). We do not at- tach much importance to such specific differences because criteria for schooling differ widely and radio- metric procedures in the older studies were primitive by today's standard. We suspect the threshold for jack mackerel may have been lower than the north- ern anchovy because of use of a uniform and highly reflective background in the apparatus and the use of photometric brightness as a unit of measurement. In our work the brightness to the sides and below was much lower than the downwelling irradiation whereas this was not the case in the jack mackerel experiment. ACKNOWLEDGMENTS We thank Mike Sokol (Southampton College, NY) for constructing the apparatus and for conducting some of the experiments and Sandor Kaupp (Univer- sity of California at San Diego) who provided advice and assistance throughout the study. We also thank Tilman Pommeranz and Geoffrey Moser for permit- ting us to use their unpublished data on vertical distribution of anchovy eggs, and Paul Smith, Tilman Pommeranz, Roger Hewitt, and J. H. S. Blaxter for reviewing the manuscript. LITERATURE CITED Ahlstrom, E. H. 1959. Vertical distribution of pelagic fish eggs and larvae off California and Baja California. U.S. Fish Wildl. Serv., Fish. Bull., 60:107-146. Baker, K. S.. anjd R. C. Smith. 1982. Bio-optical classification and model of natural waters. 2. Limnol. Oceanogr. 27:500-509. Blaxter, J. H. S. 1970. 2. Light 2.3 Animals 2.32 Fishes. In 0. Kinne (editor). Marine Ecology, Vol. 1, Part 1, p. 213-320. Wiley-Inter- science, Lond. Blaxter, J. H. S., and J. R. Hunter. 1982. The biology of clupeoid fishes. Adv. Mar. Biol. 20:1- 223. Brown, D. R. E. 1952. Natural illumination charts. U.S. Navy Bur. Ships Pro- ject NS 714-100, Rep. No. 374-1, Wash., D.C.. 11 p. Hunter, J. R. 1966. Procedure for analysis of schooling behavior. .J. Fish. Res. Board Can. 23:547-562. 1968. Effects of light on schooling and feeding of jack mack- erel, Trachurus symmetricus. J. Fish. Res. Board Can. 25:393-407. Hunter, J. R., and C. A. Kimbrell 1980. Egg cannibalism in the northern anchovy, Enffraulij^ mordax. Fish. Bull.. U.S. 78:811-816. Loukashkin, A. S. 1970. On the diet and feeding behavior of the northern anchovy, Engraulis moi-dux (Girard). Proc. Calif. Acad. Sci., Ser. 4, 37:419-458. Mais, K. F. 1974. Pelagic fish surveys in the California Current. Calif. Dep. Fish Game, Fish Bull. 162, 79 p. Munz, F. W., and W. N. McFarland. 1977. Evolutionary adaptation of fishes to the photic environ- ment. In F. Crescitelli (editor). The visual system in verte- brates. Vol. 7 (Part 5), p. 193-275. Handb. Sens. Physiol. PlELOU, E. C. 1969. An introduction to mathematical ecology. Wiley- Interscience, N.Y., 286 p. Pitcher, T. J., B. L. Partridge, and C. S. Wardle. 1976. A blind fish can school. Science 194:963-965. Pommeranz, T., and H. G. Moser. 1983. Vertical distribution study R/V Ellen B. Scripps March 19-April 4, 1980. Admin. Rep. LJ-83-01, 368 p. Protasov, V. R. 1964. Some features of the vision of fishes. [In Russ.] A. N. Severtsov Inst. Anim. Morphol., Acad. Sci. U.S.S.R., p. 29- 48. Marine Laboratory, Aberdeen, U.K., Transl. No. 949. 241 FISHERY BULLETIN: VOL. 83. NO. 3 Smith, P. E. Squire, J. L., Jr. 1970. The horizontal dimensions and abundance offish schools 1978. Northern anchovy school shapes as related to problems in the upper mixed layer as measured by sonar. In G. B. in school size estimation. Fish. Bull., U.S. 76:443-448. Farquhar (editor). Proceedings on biological sound scattering Whitney, R. R. in the ocean, p. 563-600. MC Rep. 005, Maury Cent. Ocean 1969. Schooling of fishes relative to available light. Trans. Sci., Dep. Navy, Wash., D.C. Am. Fish. Soc. 98:497-504. ! 242 DISTRIBUTIONAL PATTERNS OF FISHES CAPTURED ABOARD COMMERCIAL PASSENGER FISHING VESSELS ALONG THE NORTHERN CHANNEL ISLANDS, CALIFORNIA Milton S. Love.i William Westphal,' and Robson A. Collins^ ABSTRACT We surveyed fishes taken aboard commercial passenger fishing vessels alon^ the four northern Channel Islands (San Miguel, Santa Rosa, Santa Cru?, and Anacapa) within the Southern California Bight. P'ourteen species declined in abundance along the Northern Channel Island chain. Colder water forms decreased to the east, while temperate species declined to the west. In the shallowest depth interval (0-36 m). the mean lengths of four rockfish species increased toward the west. In general, the size of these four species also in- creased with depth. We believe these phenomena are linked to the differences in water temperature be- tween the islands- with cold, California Current water dominant in the west, and warmer Southern Califor- nia Bight water entrained in the east. The mainland coast of California is distinguished by two faunal provinces: A warm-temperate Californian Province lies south of Point Conception and a cold- temperate Oregonian Province exists to the north (Seapy and Littler 1980). In shallow waters, the fish fauna of the Californian Province is a mixture of eurythermic temperate and subtropical species, while the Oregonian Province is predominantly a colder temperate region, with few subtropical species present. Recent studies examining the distributional pat- terns of marine intertidal invertebrates (Littler 1980; Seapy and Littler 1980), algae (Murray et al. 1980), and seabirds (Hunt et al. 1980) around south- em California islands imply there is a replication of these two mainland faunal provinces along the 88 km, east-west lying, northern Channel Islands (San Miguel, Santa Rosa, Santa Cruz, Anacapa) (Fig. 1). Oregonian Province species dominate the western end of the chain, while the fauna of the eastern end is more Californian. There is little published on the biogeography of fishes around the northern Channel Islands. Ebeling et al. (1980a, b) examined the fish populations of Santa Cruz Island kelp beds, and Hubbs (1967, 1974) stated that the fish communities of San Miguel were closely related to those of central California, while about Santa Cruz fish were typical of southern Cali- iVANTONA Research Group, Moore Laboratory of Zoology, Occidental College, Los Angeles, CA 9004 L ^Marine Resources Branch, California Department of Fish and Game, 1301 W. 12th, Long Beach, CA 90813. fornia. No other work has been published on this topic. In this paper, we describe one aspect of the north- ern Channel Islands' fish fauna, utilizing data gathered by the California Department of Fish and Game in their Commercial Passenger Fishing Vessel creel census. This census (fully described in Methods) counted, measured, and noted the location and depth of capture of fishes taken by hook and line on sport- fishing passenger vessels in southern California. Data from this study could not give an unbiased estimate of species composition. Most angling in- volved fishing with live bait (primarily northern an- chovies, Engraulis mordax) or with lures simulating fishes, and angling techniques were similar along the island chain. Thus, the sample was biased toward relatively large-mouthed, piscivorous species. However, the purpose of this study was to ascertain distributional patterns of whatever species were taken by these methods, rather than attempting to describe entire fish communities. METHODS Fishes taken aboard commercial passenger fishing vessels (CPFV) were sampled by the senior author and by California Department of Fish and Game per- sonnel from April 1975 to December 1978. The sam- pling units (trips) were chosen randomly, and the population sampled consisted of all regularly sched- uled trips by CPFV's operating south of Point Con- ception to the Mexican border. The sampler assigned to each boat boarded the Manuscript accepted August 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 243 FISHERY BULLETIN: VOL. 83, NO. 3 SANTA. BARBARA Figure 1 . - Location and block numbers of sampling sites about the northern Channel Islands, C A. boat at the beginning of the trip and remained aboard, measuring and identifying all fish caught by the passengers, until the boat returned to dock. A particular effort was made to measure every fish landed, even those returned alive to the water because of undesirability or in compliance with bag or size limit regulations. Also noted were the number of anglers aboard the vessel, the hours of fishing, and the location and depth of fishing effort. Each fish was placed on a plastic measuring sheet held in a two-sided aluminum frame and the length marked. Total length (tip of snout to tip of depressed caudal fin) was recorded for all fish except members of the jack (Carangidae) and mackerel (Scombridae) families, from which fork length was taken. A portion of the caudal fin was clipped from each fish measured so that the fish could be recognized later. When fish were brought aboard too rapidly for all to be measured, samplers gave immediate priority to those being returned to the water and measured the remaining fish at the end of fishing or during a break in activity. When samplers were uncertain of the identification of a fish, they retained it for positive identification. After completing a trip, the samplers tallied and recorded by species the data col- lected. Individual fish lengths were measured on the plastic sheet with a meter stick. The California Department of Fish and Game has divided marine waters off California into numbered blocks. For this study, we utilized data from block numbers 684-690 and 707-712 (Fig. 1). Care was taken to remove data from mainland fishing sites in block number 684. Block 710 was Santa Rosa Island alone. We measured fish abundances by catch per unit effort, which was defined as number of fish taken per angler hours (where angler hours = number of anglers x number of hours fished). RESULTS One hundred and nineteen trips were made, and the catches from 3,712 anglers were sampled. A total of 23,089 fishes of 78 species were tallied, of which 49 are listed in Table 1. Rockfishes (particu- larly bocaccio, Sebastes paucispinis; blue rockfish, S. mystimcs; and olive rockfish, 5. serranoides) and kelp bass, Paralabrax clathratus, were numerically dominant. Among the 20 most abundant species, only 4 {Paralabrajc clathratus; chub mackerel. Scom- ber japonictts; lingcod, Ophiodon elongatus; ocean whitefish, Caulolatiliis princeps) were not rock- fishes. We stratified our data by depth (36 m intervals). 244 LOVE ET AL.: DISTRIBUTIONAL PA'ITKRNS OF KISHKS Table 1.— Partial list of fishes taken aboard commer- cial passenger fishing vessels during sampling from April 1975 to December 1978 around the northern Channel Islands. Only those species where 10 or more individuals were tal' and archeology of the California Islands. In R. N. Philbrick (editor). Pro- ceedings of the Symposium on the Biology of the California Islands, p. 337-341. Santa Barbara Botanic Gardens, Santa Barbara, CA. 1974. Review and comments. Marine zoogeography. Copeia 1974:1002-1005. Hunt, G. L., .Ik.. R. L. Pitman, and H. L. Jones. 198(1. Distrihulioti and abundance of seabirds breeding on the California ('hannel Islands. In U. M. Power (editor), The California islands: Proceedings of a Multidisciplinary Sympo- sium, p. 443-460. Santa Barbara Mus. Nat. HisL KuLl'ACK, R. L. 1971. Oceanography of the Santa Barbara Channel. In R. L. Kolpack (editor). Biological and oceanographical survey of the Santa Barbara Channel oil spill 1969-1970. II. Physical, chemical and geological studies, p. 90-180. Allan Hancock Found., Univ. S. Calif., Los Angeles, CA. Littler, M. M. 1980. Overview of the rock intertidal systems of southern California. In D. M. Power (editor). The California islands: Proceedings of a Multidisciplinary Symposium, p. 265-306. Santa Barbara Mus. Nat. Hist. Love, M. S. 1978. Aspects of the life history of the olive rockfish, Sebajites serranoides. Ph.D. Thesis, Univ. California, Santa Barbara, 185 p. Miller, D. J., and J. J. Geibel. 1973. Summary of blue rockfish and lingcod life histories; a reef ecology study; and giant kelp, Macrocystif pyrifera, experiments in Monterey Bay, California. Calif. Dep. Fish Game, Fish Bull. 158, 137 p. Miller, D. J., and R. N. Lea. 1972. Guide to the coastal marine fishes of California. Calif Dep. Fish Game, Fish Bull. 157, 235 p. Murray, S. N., M. M. Littler, and I. A. Abbott. 1980. Biogeography of the California marine algae with emphasis on the southern California Islands. In D. M. Power (editor). The California islands: Proceedings of a Multidisci- plinary Symposium, p. 325-340. Santa Barbara Mus. Nat. Hist. Neushul, M., W. D. Clarke, and D. W. Brown. 1967. Subtidal plant and animal communities of the southern California Islands. In R. N. Philbrick (editor). Proceedings of the Symposium on the Biology of the California Islands, p. 37-55. Santa Barbara Botanic Gardens, Santa Barbara. CA. Radovich, J. 1961. Relationships of some marine organisms of the north- east Pacific to water temperatures. Particularly during 1957 through 1959. Calif. Dep. Fish Game, Fish Bull. 1 12, 62 p. Reid, J. L., Jr., G. I. Roden, and J. G. Wyllie. 1958. Studies of the California Current system. Calif. Coop. Oceanic Fish. Invest. Rep. 5:27-56. Seapy, R. R., and M. M. Littler. 1980. Biogeography of rocky intertidal macroinvertebrates of the southern California Islands. In D. M. Power (editor). The California islands: Proceedings of a Multidisciplinary Sympo- sium, p. 307-324. Santa Barbara Mus. Nat. HisL SOKAL, R. R., and F. J. ROHLF. 1969. Biometry; the principles and practice of statistics in biological research. W. H. Freeman, San Francisco, CA, 776 P- Stephens, J. S., Jr., and K. E. Zerba. 1981. Factors affecting fish diversity on a temperate reef. Environ. Biol. Fishes 6:111-121. Stephens, J. S., Jr, K. E. Zerba, P. Morris, and M. S. Love. In press. Factors affecting fish diversity on a temperate reef II: The fish assemblage of Palos Verdes Point, 1974-81. Environ. Biol. Fishes. 251 MORPHOLOGICAL DEVELOPMENT, IDENTIFICATION, AND BIOLOGY OF LARVAE OF PANDALIDAE, HIPPOLYTIDAE, AND CRANGONIDAE (CRUSTACEA, DECAPODA) OF THE NORTHERN NORTH PACIFIC OCEAN Evan B. Haynes> ABSTRACT All published descriptions of pandalid, hippolytid, and crangonid larvae of the northern waters of the North Pacific Ocean are summarized. Included are recent changes in nomenclature, definitions of terms used in describing the larvae, and procedures for preparing larvae for examination. The general morphology of lar- vae of the three families is reviewed, and development of the morphological characters used for their iden- tification is discussed. Principal morphological characters and number of larval stages of known larvae in each family are tabulated. Pandalid larvae are keyed to species and stage of development. A synopsis of the most important morphological characters used for identification is given for larvae of each family, genus, and species. Biologj' of the larvae is reviewed. Larvae of the Pandalidae, Hippolytidae, and Cran- gonidae (order Decapoda, tribe Caridea) are common inhabitants of the neritic meroplankton of the north- ern (temperate and arctic) waters of the North Pacific Ocean. About 135 species of shrimps are found in these waters, and larvae have been de- scribed, at least in part, for 46 species. Many of these descriptions, especially of hippolytids and cran- gonids, are scattered in various foreign scientific journals. This report summarizes the morphology of de- scribed larvae of the Pandalidae, Hippolytidae, and Crangonidae of the northern North Pacific Ocean and gives instructions for examining them. Develop- ment of the characters used for identification is discussed, and a generalized key to stage is given. Larvae of each family, genus, and species are charac- terized morphologically, and the principal mor- phological characters and the number of the larval stage are tabulated. Illustrated keys to species and stages are provided for 9 of the 13 pandalid species recorded from the northern North Pacific Ocean. De- scriptions of larvae of the remaining four species of pandalids have not been published although their probable morphology has been discussed (Haynes 1980a). References to the published descriptions of larvae of each species and a review of the biology of the larvae are provided. In the synopses of species, I have selected the most distinguishing larval characters; however, these >Northwest and Alaska Fisheries Center Auke Bay Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 210155, Auke Bay, AK 99821. characters may not always be repeated for similar species, and additional characters may be needed for a specific diagnosis. Other distinguishing characters for the larvae discussed here can often be found in the original published descriptions. An annotated listing of published descriptions precedes the synopsis of each species. When two or more descriptions are listed, the most complete description is given first. Whenever a "?" appears after a species name or stage in the listings, the author of the original description was uncertain of the identification. In these cases, references to the corrected or verified identification are included in the listing. For a few species, descriptions of larvae are based on specimens from the Atlantic Ocean. It should be noted that morphological characters of lar- vae of the same species from different geographical areas may vary somewhat (Haynes 1978a). Taxonomic Nomenclature I used Rathbun's (1904) nomenclature of Panda- lidae except for Pandalus tridens. After considering both the larval and adult morphology of P. tridens (see Rathbun 1902), I give this species full specific rank (Butler 1980; Haynes 1980a) rather than sub- specific rank as P. montagui tridens, a Pacific sub- species of P. montagui Leach 1814. Nomenclature of the Hippolytidae follows Hol- thuis' (1947) revision of the genus Spirontocaris sensu lato. Holthuis' revision, based on adult mor- phology, has been verified from larval morphology (Pike and Williamson 1961; Haynes 1981). Nomenclature of the Crangonidae is based on the Manuscript accepted August 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 253-^^ FISHERY BULLETIN: VOL. 83, NO. 3 revision by Kuris and Carlton (1977), with one excep- tion. I use only Crangon, rather than Crangon and Neocrangon, because I agree with Butler (1980) and consider Zarenkov's (1965) separation of the genus Crangon into two subgenera, Crangon s.s. and Neocrangon, to be invalid. The synopsis of characters of Crangon septem- spinosa larvae also applies to the characters of C. af- finis larvae. Needier (1941), Kurata (1964b), and Tesmer and Broad (1964) have described these as two species, but according to Makarov (1967), the two species are synonymous. Descriptions of the lar- 0. 5 mm FiGi'RE 1.- Lateral view and body parts of a diagrammatic decapod shrimp larva. (A) A,, antennule; A^, antenna; C, carapace; Md, mandi- ble; Mx,. maxillule; Mx.^, maxilla; Mxp,, maxilliped 1; Mxp^, maxilliped 2; Mxp.,, maxilliped 3; P,, pereopod 1; P^, pereopcxi 2; P,, pereopf)d 3; Pj, pereopod h; Pl.^, pleopfxl 2; Pl^,, pleopod .5; S,, somite 1; S^, somite 6; T, telson; U, uropod; ad, anteroventrail denticles; as, antenna! spine; 254 HAYNES: PANDALIDAE, HIPPOLYTIDAK, ANDCRANGONIDAE LARVAE vae of C. affmis and C. septemspinosa are very similar. More comments on nomenclatxire for certain species are given in the synopses. Definition of Terms I follow Williamson's (1969) terminology for decapod larvae and Haynes' (1979) terminology for larval appendages (Fig. 1). The terms are defined as follows: abbreviated development- less than five zoeal stages, carapace length -straight-line distance between the 0.25 mm 0. 5 mm Figure I.- Continued -ds, dorsal spine; pi, posterolateral spine; pt, pteryRostomian spine; so, supraorbital spine; r, rostrum; rt, rostral teeth. (B) A,, antennule; ac, aesthetascs; ap, antennule peduncle; if, inner flagellum. (C) A^. antenna; f, flagellum; ps, plumose setae; sc, scale; ss. scale seg:ments. (D) Md, mandible; cl, curved lip; ip, incisor process; mp, molar process; p, palp; sp, subterminal process. (E) Md. mandible; Im, lacinia mobilis. (F) Mx,, maxillule; b, basipodite; c, coxopodite; en, endopodite; st, subterminal seta. (G) MXj. maxilla. (H) Mxpj, maxilliped 1; ep, epipodite. 255 FISHERY BULLETIN: VOL. 83, NO. 3 posterior margin of orbit and the middorsal pos- terior margin of the carapace. denticles -toothlike projections on anteroventral margin of the carapace. developed pereopods- segmented pereopods direct- ed vertically under cephalothorax. juvenile -young form, usually small, sexually im- mature, and generally resembling adult. larva -a free- swimming phase in the life cycle of the individual whose morphology (such as body form, appendages, and spination) and habit are different from the adult. The term applies to both zoea and Figure \.-Continued-(l) Mxp^, maxilliped 2. (J) Mxp,, maxilliped 3; b, basipodite; c, coxopodite; ca, carfjopodite; da, dactylopodite; ex, ex- opodite; en, endopodite, is, ischiopodite; me, meropodite; pp, protopodite; pr, propodite. (K) P,, pereopod 1. (L) P2, pereopod 2; ch, chella. 256 HAYNES: PANDALIDAE, HIPPOLYTIDAE, ANDCRANGONIDAE LARVAE megalopa. (For shrimp, the change from larva to adult is usually somewhat gradual and may include more than one molt.) megalopa- larva with fully setose natatory pleopods^ on some or all abdominal somites. setation formula of telson- setae or spines along the terminal margin of the telson are numbered begin- ning at the middle of the telson. Thus, 7 + 7 means ^In the Decapoda. the development of setose pleopods does not always provide a convenient and clear distinction between zoeal and postzoeal stages. Several species of Pandalidae metamorphose gradually into the postzoeal stage (see Haynes 1976). The term "megalopa", therefore, may include a single stage or several stages depending upon the species. In this paper, the number of larval stages includes all stages before the juvenile stage, regardless of whether the megalopa has one or more stages. that the telson has seven pairs of setae along the terminal margin. The first pair is the medial pair, setose - having setae (bristles), spine -a sharp, pointed projection, usually long and narrow, spinose-with many spines, spiniform - shaped like a spine, spinule - small spine, spinulose - with small spines, stage - intermolt. subchelate - the dactylopodite (finger) folds against the preceding segment (propodite), as in the first pereopod of crangonid adults, length -total body length: distance (mm) from the anterior tip of the rostrum to the posterior tip of 0.5 mm 0. 5 mm Figure l.-Contmued-{M) P^, pereopod 4. (N) Pleopod; ai, appendix interna; en, endopodite; ex, exopodite. (0) Tail fan; al, anal spine; en, endopodite; ex, exopodite; ts, telsonic spines. 257 FISHERY BULLETIN: VOL. 83, NO. 3 telson, excluding telsonic setae or spines. unabbreviated development -five or more zoeal stages. undeveloped pereopod-unsegmented pereopod directed anteriorly under cephalothorax. zoea- larva vi^ith natatory setae on maxillipeds, wfith- out setose natatory pleopods on some or all abdom- inal somites 1-5. Examination Procedure It is usually necessary to dissect the animal and mount certain appendages on a slide before the iden- tification characters can be used. Visibility of segmentation is often improved by clearing speci- mens for several days in 10% KOH or full-strength lactic acid. Larvae can be dissected with pins designed for mounting small insects. (The pins are available from most biological supply companies.) After dissecting the larva, place the appendage in a drop of mounting medium (I use Turtox^ CMC red mounting medium) and cover with a cover glass. Gently press the cover glass to splay hairs and setae and make them easier to examine and count. After mounting the appendages, examine them using a dis- secting microscope. GENERAL MORPHOLOGY OF LARVAE Pandalid, hippolytid, and crangonid larvae have three major body regions (Fig. 1): head, thorax, and abdomen. The head and thorax are coalesced and are dorsally covered by a common, unjointed cephalo- thoracic shield, the carapace. The body is divided into 19 true somites which, with their appendages (Fig. 1), are grouped as follows: 1) The head, five indistinguishable fused somites, is covered by the anterior portion of the carapace (C) and has the first five pairs of appendages: antennules (or first antennae) (AJ, antennae (A2), mandibles (Md), maxillules (Mxj), and maxillae (Mx,). 2) The thorax is composed of eight somites that are dorsally fused with, and covered by, the carapace. Each somite has a pair of appendages: Somites 1-3 each have a pair of maxillipeds (Fig. lA, Mxpi.;j); somites 4-8 each have a pair of pereopods (Fig. lA, P,.5). 3) The abdomen is composed of six somites (Fig. 1 A, Si.^) and a terminal segment, the telson (T). The 'Reference to trade name does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. first five somites each have a pair of pleopods (Fig. lA, Pl?.^). The sixth somite has uropods (U). The uropods, when present with the telson, com- prise the tail fan. CHARACTERS USEFUL FOR IDENTIFICATION OF LARVAE Understanding the development of morphological characters is necessary for identifying genus, species, and stage of larval development. The follow- ing discussion emphasizes the characters most useful for identification. It should be noted, however, that these characters are based on relatively few species of a limited number of genera. For instance, in the family Hippolytidae, the subterminal seta is ap- parently absent in larvae of Hippolyte clarki from British Columbia (Needier 1934) but present in lar- vae of H. inermis and H. varians from European waters (Williamson 1957a). The seta may also be present in undescribed Hippolyte larvae from the northern North Pacific Ocean. Characterization of the family Pandalidae is based on only two genera, Pandalapsis and Pandalus. In the northern North Pacific Ocean, described larvae of these two genera develop exopodites only on pereopods 1-2 or 1-3, never on pereopods 1-4. Other genera of the family (e.g., Plesionika) may develop exopodites on pereo- pods 1-4 (Williamson 1957b). Additional descriptions of decapod larvae from the northern North Pacific Ocean will undoubtedly modify the morphological characterizations given here. Rostrum In pandalid larvae, the rostrum (Fig. lA, r) is always long (at least one-third the carapace length). For most species of Pandalidae, the rostrum is styli- form in Stage I and does not have teeth (Fig. lA, rt) until about Stage III. The exceptions are Pandalop- sis coccinata, P. dispar, and Pandalus platyceros. In these species, the rostrum is curved in Stage I and has teeth in all larval stages (Berkeley 1930; Kurata 1964a; Price and Chew 1972). In Stage I Pandalus preTisor, the rostrum curves ventrally between the eyes (Mikulich and Ivanov 1983). The rostrum of hippolytid larvae may be absent, or, if present, from minute to long. The rostrum is usually without dorsal or ventral teeth in all stages. European specimens oiEualus gaimardii, an excep- tion, have two dorsal rostral teeth in the last zoeal stage (Stage V) (Pike and Williamson 1961). If the rostrum is short to long in Stage I, it is also styli- form, except in Stage I Lebbeus groenlandicus (Fig. 258 HAYNES: PANDALIDAE, HIPPOLYTIDAE. ANDCRANGONIDAE LARVAE Figure 2. - Stage I zoeae of Lehbeus groenlandicus showing slightly sinuous rostrum and posterolateral spines on abdominal somites 4 and 5. 2) and L. polaris, which have a sHghtly curved rostrum (Haynes 1978b, 1981). In Stage I Hepta- carpus camtschaticus, the rostrum is minute and curves slightly dow^nward following the dorsal con- tour of the eyes (Haynes 1981). In crangonid larvae, the rostrum is long, spini- form, and without teeth in all larval stages except for Stage I zoeae of Sclerocrangon zenkevitchi and S. boreas, which lack a rostrum, and Stage I zoeae of Paracrangon echinata, which have a spinulose rostrum (Birshteyn* and Vinogradov 1953; Kurata 1964b; Makarov 1968). Spines on the Carapace The presence or absence of certain spines on the carapace is useful for distinguishing between families and identifying one or more stages. The lar- val carapace (Fig. lA, C) may have supraorbital spines (so), pterygostomian spines (pt), antennal spines (as), and anteroventral denticles (ad). Supra- orbital spines are absent in all larval stages of the Crangonidae. For pandalid and hippolytid larvae, supraorbital spines are usually absent in Stage I and the megalopa, but present in the intermediate zoeal stages. However, there are some exceptions to these generalizations. Larvae oiPandalus hypsinotus have supraorbital spines only in Stages II and III, and lar- vae of P. kessleri have supraorbital spines only in Stage II (Kurata 1955; Haynes 1976). Larvae of P. ••Translator's spelling of "Birstein". prensor are without supraorbital spines in all larval stages (Mikulich and Ivanov 1983). In the Hippolyt- idae, Spirontocaris spinus, S. lilljeborgii, and prob- ably S. phippsil have supraorbital spines in all larval stages (the spines are minute in Stage I). Spirorir tocaris ochotensis has minute supraorbital spines in Stage I; whether supraorbital spines develop later is unknown. Lebheus groenlandicus, a species with abbreviated development, has supraorbital spines only in Stage II and the megalopa (Stage III) Pike and Williamson 1961; Haynes 1978b, 1981). In all three families, pterygostomian spines are usually present in all larval stages. Antennal spines are often absent in Stage I but usually develop in later stages. Anteroventral denticles are most prevalent in the early stages and usually, but not always, disappear during larval development. Branchiostegal and hepatic spines are rarely, if ever, present in the lar- val stages. Eyes Development of the eyes is the same for nearly all members of the three families. The eyes are com- pound and sessile in Stage I and are stalked in later stages. The exception. Stage I Pandalopsis cocci- nata, has compound eyes that are only partially attached to the carapace (Kurata 1964a). Antennules In the Pandalidae and nearly all hippolytid and 259 FISHERY BULLETIN: VOL. 83, NO. 3 crangonid larvae, the peduncle of the antennule (Fig. IB, ap) is unsegmented in Stage I and becomes three segmented later, usually in Stage II. The exceptions are Stage I Sderocrangon boreas, which has a 3-seg- mented antennule; Stage II Eualiis suckleyi and E. fabricii, which have 2-segmented peduncles; and both Stages I and II Lebbeiis polaris and L. groenlan- dicus, which have unsegmented peduncles. In Stage I pandalid and hippolytid larvae, the inner flagellum (Fig. IB, if) of the antennule is usually a plumose seta, whereas in Stage I crangonid larvae, it is usually a setose spine. The only described excep- tions are Stage I larvae of Pandaltis prensor, Sdero- crangon boi'eas, S. salebrosa, and 5. zenkeiitchi. In Stage I Pandalus prensor, the inner flagellum is spine shaped and has a few simple setae medially. In Stage I S. boreas, S. salebrosa, and S. zenkevitchi, the inner flagellum is an oblong projection that is naked except for a few minute, simple setae termi- nally (Birshteyn and Vinogradov 1953; Makarov 1968; Haynes 1978b, 1981; Mikulich and Ivanov 1983). Antennae Segmentation of the tip of the antennal scale (Fig. IC, sc, ss) is an important characteristic for distin- guishing crangonid larvae from pandalid and hippo- lytid larvae. In crangolid larvae, the scale tip is unsegmented in all stages. In pandalid and hippolytid larvae, the scale tip is unsegmented in only four species: PayidalopRis coccinata, Pandalus kessleri, P. prensor, and possibly Heptacarptis ( = Spirontocaris) tridens (Needier 1934; Kurata 1955, 1964a, b; Mikulich and Ivanov 1983). The absence of segmen- tation of the scale tip of Pandalopsis coccinata, Pan- dalus kessleri, and P. prensor is related to the ex- tremely precocious development of these species. Hepta<:arpus tridens, however, has unabbreviated development (Needier 1934) and, presumably, a segmented tip. Needier (1934) may not have ob- served segmentation of the scale in H. tridens because she based her description on unstained lar- vae (staining emphasizes segmentation (Haynes 1976)). Mandibles Mandibles (Fig. ID) are described for most pan- dalid larvae, but descriptions of mandibles for hippo- lytid and crangonid larvae in the northern North Pacific Ocean are usually limited to Stage I. I have supplemented these limited descriptions with infor- mation on larvae from other areas, particularly the North Sea. Descriptions of late stage larvae from the northern North Pacific Ocean are needed, however, to verify development of mandibles in hippolytid and crangonid larvae. Zoeae of Pandalidae, Hippolytidae, and Crangoni- dae have similar mandibles in all stages, and both molar and incisor processes are present. In Stage I, the incisor processes of the left and right mandibles are typically biserrate or triserrate. The number of teeth increases in later stages. In some species, the left mandible also has a subterminal tooth and a lacinia mobilis (movable spine adjacent to incisor pro- cess; Fig. IE). The subterminal tooth and lacinia mobilis are usually, if not always, absent on the right mandible. In the Crangonidae, the incisor process eventually becomes a molar process, usually at the megalopa. The most distinctive character of the mandible is the absence of a palp in the zoeal stages. This palp first appears in the megalopa or first juvenile stage of Pandalidae and in the megalopa or later stages of the Hippolytidae. In one exception, Pandalopsis coc- cinata, the palp is present and segmented as early as Stage I (Kurata 1964a). The palp is absent in all stages of Crangonidae, including the adults. The mandibular palp of the Hippolytidae may develop somewhat later than the mandibular palp of the Pandalidae. For instance, in some species of Hippolytidae, the palp may not appear until as late as the third or fourth juvenile stage (Lebour 1936), and in some genera, such as Hippolyte, the absence of the palp in the adult may mean the palp is absent in the larvae also. The palp eventually becomes three segmented in the Pandalidae and two segmented in the Hippolytidae. Maxillules Lebour (1930) stated that Pandalus larvae have no subterminal seta (Fig. IF, st) on the basipodite of the maxillule (F^ig. IF, Mx,, b). Yet, the seta is present in some or all larval stages of Pandalus kessleri, P. tridens, P. stenolepis, P. borealis, P. goniurus, P. jor- dani, P. hypsinotus, and Pandalopsis coccinata. The seta is also present in the early stages of Pandalus montagui, P. projnnquus, Pandalina brevirostris, and Dichelopandahis bonnieri that were collected from waters off Great Britain (Needier 1938; Kurata 1955, 1964a; Pike and Williamson 1964; Modin and Cox 1967; Lee 1969; Haynes 1976, 1978a, 1979, 1980a). The subterminal seta is absent in hippolytid and crangonid larvae described from the North Pacific Ocean but is present in larvae of Hippolyte in- ermis and H. varians from waters off Great Britain (Lebour 1931). According to Gurney (1942) and Pike 260 HAVNKS:1'ANDAL11)AK. llll'I'OI.VTlDAK. ANDCRANGtWIDAE LARVAE and Williamson (1964), the seta is probably the vestige of an exopodite; however, Williamson (1982) regards it as a vestigial epipodite or pseudoepipodite. Maxillae Development of the scaphognathite (exopodite) of the maxilla (Fig. IG) is related to development of the larvae. Most species that lack precocious develop- ment have a scaphognathite that is not lobed proxi- mally and has only a few (usually < 12) plumose setae only on its outer margin. The scaphognathite gradually becomes lobed proximally in subsequent stages, and the outer margin becomes fringed with many plumose setae. In species with precocious development, the scaphognathite is lobed proximally and fringed with many plumose setae in Stage I. The number of plumose setae is sometimes used for distinguishing Stages I or II of similar species. For instance. Stage I zoeae oiPandalus horealis and P. goniums are similar morphologically, and in these species the scaphognathite has 12 and 5 plumose setae, respectively. In later stages, however, the number of plumose setae fringing the scaphogna- thite becomes too great to be practical for distin- guishing zoeae of similar species. Maxillipeds The number of natatory setae on the exopodite of each maxilliped (Fig. IH-J) is helpful for distinguish- ing Stage I hippolytid and crangonid zoeae from Stage I pandalid zoeae. All Stage I hippolytid and most Stage I crangonid zoeae have 4, 5, 5 natatory setae on the exopodites of maxillipeds 1-3. In the Pandalidae, all Stage I pandalid zoeae, except Pan- dahis stenolepis, have ^ 8 natatory setae on the ex- opodites of at least two pairs of maxillipeds. Stage I P. stenolepis, however, cannot be differentiated from Stage I hippolytid and crangonid zoeae based only on natatory setae because Stage I P. stenolepis also has 4,5,5 natatory setae on the exopodite of each max- illiped. The absence or reduction in numbers or size of natatory setae on the exopodites of maxillipeds is associated with markedly precocious development. This is especially true for Pandalopsis coccinata, Sclerocrangon boreas, and S. zenkevitchi. Each of these species has only one zoeal stage before molting to the megalopa. In P. coccinata, the natatory setae are absent from the third maxilliped. In 5. boreas, the number of natatory setae on maxillipeds 1-3 is 2, 3, 4, respectively; and the setae are reduced in size. Sclerocrangon zenkevitchi zoeae do not have nata- tory setae on the maxillipeds. Apparently, the absence or reduction in size of natatory setae prevents zoeae from being planktonic. Zoeae of S. boreas and S. zenkevitchi (collected at sea) cling to the pleopods of the adult (Birshteyn and Vinogradov 1953; Makarov 1968). Zoeae of P. cocci- nnta are rarely, if ever, taken in plankton tows (Kurata 1964a). Pereopods The presence of exopodites (Fig. IJ, ex) on certain pereopodal pairs is an important morphological character for identifying shrimp larvae. Exopodites are present on pereopods 1, 1 and 2, 1-3, or 1-4, depending on genus or species (Fig. IK-M). Species with unabbreviated development usually develop an exopodite on each pereopod. In most species with > 5 zoeal stages, the exopodites are characteristically small, naked, and nonfunctional at Stage I but functional (have natatory setae) at Stage II or III. Development of exopodites on pereopods tends to be suppressed in species with < 5 zoeal stages. In the Pandalidae, species that have segmented pereopods directed vertically under cephalothorax (i.e., developed pereopods) in Stage 1- Pandalopsis coccinata, Pandalopsis dispar, Pandnlus kessleri, Pandalus danae, Pandalics hypsinotus, and Pari'- d/ilus prensor- also have exopodites or vestigial exo- podites on pereopods 1 and 2. Species that have unsegmented pereopods directed anteriorly under cephalothorax (i.e., undeveloped pereopods) in Stage I -Pandalus tridens, P. stenolepis, P. borealis, P. goniurus, and P. jordani -a\so have exopodites on pereopods 1-3 (the exopodites are undeveloped in Stage I and develop in later stages). An exception is P. platyceros, which in Stage I has developed pereo- pods and exopodites on pereopods 1-3 (Haynes 1980b). Of the Hippolytidae, only larvae of the genus Lebbeus have developed pereopods in Stage I. Lebbeus polaris has vestigial exopodites on pereopods 1 and 2 in Stage I and on pereopod 1 in Stage II. Lebbeus groenlandicus has vestigial exo- podites on pereopods 1 and 2 in Stage I and no exo- podites on pereopods in Stages II or III (Haynes 1978b, 1981). In the Crangonidae, most species with developed pereopods in Stage I {Argis crassa, A. lar, A. deno- tata, Sclerocrangon boreas, and S. salebrosa) are either without exopodites or have rudimentary exo- podites. An exception is S. zenkevitchi, which has an exopodite on pereopod 1 in Stage I (Birshteyn and Vinogradov 1953). 261 FISHERY BULLETIN: VOL. 83. NO. 3 Which pereopods have exopodites can differ for different species of a genus. For example, the genus Pandahis includes larvae that have exopodites on pereopods 1 and 2 or 1-3 (Haynes 1980a). The genus Eualus includes larvae that have exopodites on pereopods 1-3 or 1-4 (Haynes 1981). Larvae of Crangon typically have an exopodite only on pereopod 1; however, larvae of C. franciscorum angustimana have exopodites on pereopods 1 and 2 (Haynes 1980b). Abdomen The presence or absence of posterolateral spines (Fig. lA, pi) on the abdomen is often an important character for identif^'ing the families of caridean lar- vae. Specimens of pandalid larvae from the northern North Pacific Ocean do not have posterolateral spines. Crangonid larvae, however, usually have posterolateral spines on somite 5, except for larvae of Sclerocrangon boreas and S. zenkeintchi (Birsh- teyn and Vinogradov 1953; Makarov 1968). Larvae of Hippolytidae also have posterolateral spines. Posterolateral spines are present on somites 4 and 5 in Lebbeiis larvae and on somite 5 in Hippolyte lar- vae, but are absent in Heptacarpus larvae. The number of abdominal somites with postero- lateral spines is not always the same for all species of a genus. For instance, Spirontocaris larvae and most Eualus larvae have posterolateral spines on somites 4 and 5, or 5. In some species oiEualua, the postero- lateral spines may be absent. Most pandalid and hippolytid larvae lack dorsal spines or teeth on the abdomen. The only known exception is Spirontocaris spinus, which has a distinct dorsal tooth on the posterior margin of abdominal somite 3 in the megalopa (Stage VI) (Pike and Williamson 1961).^ Of the described crangonid larvae of the northern North Pacific Ocean, only Crangon septemapinosa, C. affinis, C. alaskensis, and C. franciscorum angusti- mana have a dorsal spine (Fig. lA, ds) on somite 3 (Makarov 1967; Loveland 1968; Haynes 1980b). Paracrangon echinata has dorsal spines on somites 1-5 (Kurata 1964b). Some zoeae have spinules on the posterior margins of abdominal somites. These spinules are present in zoeae of Pandaliis platyceros, P. tridens, P. steno- iepis, Eualus suckleyi, E. fabricii, Argis crassa, A. dentata, and Crangon communis. The number and size of spinules decrease in later stages. *This spine is present in ail subsequent stages (juvenile and adult) and should probably not be regarded as a larval character. Telson The shape of the telson is useful in determining the stage of development of caridean larvae. For most Stage I-III pandalid and hippolytid larvae, the posterior margin of the telson is about twice the width of the anterior margin. At about Stage IV, the shape of the telson narrows posteriorly, and from Stage IV on, the posterior margin of the telson is noticeably less than twice the anterior width. Even- tually, the telson narrows posteriorly, as in the adult. Although the telson remains triangular in all stages of crangonid larvae, it is somewhat narrower in the megalopa or first juvenile stage than in earlier stages. For caridean larvae with unabbreviated develop- ment, the typical number of telsonic setae (Fig. 1 0, ts) is 7 -I- 7 in Stage I and 8 -i- 8 in later stages. These numbers of telsonic setae are seldom exceeded in later stages and are often reduced by either loss or transformation of certain pairs (usually pairs of 2 or 3) into small setae or hairs. A larger number of telsonic spines are more com- monly associated with abbreviated development than with unabbreviated development (Gurney 1942; Pike and Williamson 1964), and this is generally true for caridean larvae of the North Pacific Ocean. For ex- ample, Pandalus kessleri, a species with four larval stages, has 16 -h 16 telsonic setae in Stage I, and Pandahpsis coccinata, a species with three (or two) larval stages, has 28 + 28 telsonic setae in Stage I (Haynes 1980a). Examples can also be found in the other families of Caridea. In the Crangonidae, Sabinea septemcarinata has 16 -(- 16 telsonic setae in Stage I and three larval stages, whereas Sr/frorraw- gon salehrosa has 22 + 22 telsonic setae in Stage I and one larval stage (Williamson 1960; Makarov 1968). In the Hippolytidae, Lebbeus polaris has 9 + 9 telsonic setae in Stage I and four larval stages; L. groenlandicus has a total of 21 telsonic setae in Stage I and three larval stages (Haynes 1978b, 1981). All larvae of Hippolytidae except larvae of the genus Hippolyte have an anal spine (Fig. 10, al). For Hippolyte larvae, the anal spine is absent in all described stages. When pandalid and crangonid lar- vae have unabbreviated development, the anal spine usually appears at about Stage II in pandalid larvae and about Stage IV in crangonid larvae. However, the presence of an anal spine has little value in the identification of pandalid and crangonid larvae with abbreviated development. In pandalid larvae with abbreviated development, the anal spine first appears at different stages in different species. 262 HAYNES: PANDALIDAE, HIPPOLYTIDAE. ANDCRANGONIDAE LARVAE For example, the spine is present in Stage I Pan- dahis hessleri, but absent in P. hypsinotus until Stage III. The stage at which the anal spine first ap- pears in crangonid larvae with abbreviated develop- ment has not been reported. KEY TO STAGE OF DEVELOPMENT Larvae of most of the species in this report can be keyed to stage based on development of the eyes and tail fan (Key I; Table 1). Key I is mostly applicable to species whose larval development is not markedly abbreviated (i.e., those with > 5 zoeal stages). In species with unabbreviated development, exopodites on pereopods are characteristically undeveloped in Stage I and usually have natatory setae beginning at Stage II or III. Most of the species excluded from the key (Table 2) have < 4 zoeal stages, and exopodites on pereopods are either absent in all stages, vestigial in Stages I and II, or have natatory setae as early as Stage I. Key I is limited because it does not differen- tiate between the latest stages and uses only one or two characters, which may be absent in damaged specimens. For pandalid shrimp. Key I can be supplemented by keys to stages for each species (i.e.. Keys IV and VI-XI, which have characters not given in Key I and distinguish between the latest stages). With a few Table 1.— Species included In keys. Pandalidae Haworth 1825 HIppolytidae Bate 1888 Crangonidae White 1947 Pandalopsis Bate 1888 P. cocclnata Urita 1941 P. d/spar Rathbun 1902 Pandalus Leach 1814 P. boreal is Krdiyer 1838 P. danae Stimpson 1857 P. gon/urus Stimpson 1860 P. hypsinotus Brandt 1851 P. jordani Rathbun 1902 P. /(ess/en CzernlavskI 1878 P. platyceros Brandt 1851 P. prensor Stimpson 1860 P, stenolepis Rathbun 1902 P. tridens Rathbun 1902 Eualus Thallwltz 1892 E. barbatus (Rathbun 1899) £. fabricii (Kr6yer 1841) E. gaimardii (H. Milne Edwards 1837) E. herdmani (Walker 1898) E. macilentus (Krciyer 1841) £. pusiolus (Krdyer 1841) E. sucl 1 pair of lateral spines (Stage FV on). The principal morphological characters and number of larval stages of known larvae of pandalid shrimp of the northern North Pacific Ocean are sum- marized (Table 3, modified from Haynes 1980b). Genus Pandalopsis Bate Larval development abbreviated; larvae relatively large (> 10.0 mm in Stage I). In all described stages, rostrum with teeth; carapace without denticles; antennal flagellum as long as or longer than body; thoracic appendages noticeably long and thin; ab- dominal somites without spines or spinules, somites not flared laterally. In Stage I, pereopods seg- mented; telson with at least 12 -i- 12 setae; telson jointed with abdominal somite 6. Pandalopsis coccinata Urita Only Stage I described, known parentage; figure 7 in Kurata (1964a). Eyes partially fused with carapace; antennal fla- gellum same length as body; mandibular palp with 3 segments; seta on basipodite of maxillule; rudimental exopodite on maxilliped 3 and pereopods 1 and 2; telson discoid with 55 or 56 setae (Fig. 3). Length: Stage 1, 15.5 mm. Range: Japan (Kurata 1964a), no depth given. Pandalopsis dispar Rathbun Possibly 7 larval stages. Stage I, known parentage; Stages II, III, and V?, from plankton; figures 11-13 in Berkeley (1930). 1 . 0 mm Figure 3. -Telson, Stage I zoea of Pandalopsis coccinata. In all stages, basipodite of maxillule without sub- terminal seta. In Stage I, eyes sessile; antennal fla- gellum about 1/4 longer than body. Telson fan shaped in early stages. Until at least Stage V, man- dibles without palps; developed exopodites on maxil- liped 3 and pereopods 1-2. Length: Stage I, 10.0 mm. Range: Pribilof Islands, Bering Sea to Man- hattan Beach, OR; depth, 46-649 m (Butler 1980). Key II. — Larval stages of Pandalopsis dispar. 1. Rostrum curves dorsally (Fig. 4); exopodites on pereopods without natatory setae . . . Stage I 265 FISHERY BULLETIN; VOL. 83, NO. 3 Figure 4. -Carapace, Stage I zoea of Pandalopsis dispar. i 1 . 0 mm Rostrum horizontal; exopodites on pereopods with natatory setae 2 2. Uropods absent; length of larvae < 13.0 mm Stage II Uropods present; length of larvae • 13.0 mm 3 3. Endopodite of uropod noticeably shorter than exopodite; length of larvae = 16.0 mm Stage III Endopodite and exopodite of uropod same length; length of larvae > 16.0 mm Stage IV and later^ Genus Pandalus Leach Most species with unabbreviated development. In all stages, thoracic appendages not especially long or thin. In Stage I, larvae usually 5-6 mm long; rostrum usually without teeth; pereopods usually undevelop- ed and tucked under cephalothorax; telson with 7 -t- 7 setae; telson not jointed with abdominal somite 6. In early stages of some species, anteroventral margin of carapace with denticles; posterior margin of abdominal somites with spinules. Pandalus borealis Kr<>yer' Six larval stages. •'A later larval stage, probably Stage V or VI, characterized by rostrum with 25 dorsal spines: 1 spine near tip, 14 dorsal spines, 10 ventral spines. Chela of pererjjKxl 2 (Fig. 1, P^,, ch) developed, and carpopodite (Fig. 1, ca) with a few faint segments; abdominal somites and pleopods essentially adult; telson narrows posteriorly and has 7 + 7 slightly plumose, terminal setae and 6 pairs of single lateral spines. Length: .30.0 mm (Berkeley 1930). 'Larvae of ParuidLuK harealia and Pandaliuf goniurus are often Stages I-V, VI (megalopa), and VII and VIII (ju- veniles); all stages from both known parentage and plankton; figures 1-7 in Haynes (1979). Stage I, known parentage; Stages II-VII from plankton; figures 1-3 in Kurata (1964a). Stage I, known parentage; Stages II-VI from plankton; figures 7 and 8 in Berkeley (1930). Stages I-V from plankton; described as "P. propin- quus (?)" by Stephensen (1912); figures 22-31 in Stephensen (1912). Stage III from plankton; described as "Spironto- caris larva Nr. 4" by Stephensen (1916); figure 11 in Stephensen (1916). Also described as P. propinquics, Stage VI? by Stephensen (1916); figure 17 (chelae only) in Stephensen (1916). Stage IV from plankton; described as "Dynuis typus" (?) by Kr0yer (1861, as cited in Stephen- sen 1935). No figures. "Post larval" from plankton; Plate VII in Sars (1900). Probably megalopa of P. borealis (see Haynes 1979). Not Stages I-VIII P. borealis as described by Sars (1900) and figured in Plates I-VI. Correct iden- tity, Caridion gordoni (see Berkeley 1930; Le- bour 1930). In all described stages, carapace and abdominal somites not flared laterally; basipodite of maxillule found together in plankton. They are esf)ecially similar in the early stages and are difficult to distinguish. For identification of these two species, I have included Table 4, which lists by larval stage the most readily observable differences for both species. In general, larvae of P. goniuni.'i are smaller than those of P. horeiilin. In SUiges I-III, P. goninruK larvae have fewer setae on the antennal scale and certain mouth parts than /'. horeali.^ larvae. From Stage IV to megalo()a, the rostrum of P. hureaU.^ has more dorsal teeth, pereopcni 2 is more developed, and the pleopods are fringed with more setae than larvae of P. jrowmrus (Haynes 1979). 266 HAYNKS: I'AiNDAI^lDAK, HM'POLYTIDAK, ANDCKANCONIDAK l.AKVAK with subterminal seta. Carapace usually with denti- cles on anteroventral margins in Stages I-V. Rostrum without teeth until Stage III; rostral tip bifid in Stage V. In Stage I, exopodites of maxillipeds 1, 2, and 3 with 5 or 6, 13 or 14, and 16 natatory setae, respectively; pereopods tucked under cara- pace; left mandible with a lacinia mobilis; basipodite of maxillule with 9 spinulose spines terminally; sca- phognathite of maxilla with 12 setae along outer margin. Length: Stages I-VI, 6.5-20.2 mm. Range: Sea of Japan to Chukchi Sea to Columbia River mouth (northwestern United States); Barents Sea to North Sea; western Greenland to Gulf of Maine; depth, 16-1,380 m (Butler 1980). Key HI. — Larval and first juvenile stages of Pandalus borealis (see footnote 7 and Table 4). 1. Eyes sessile; pleopods absent; telson with 7 + 7 setae Stage I Eyes stalked; pleopods present as minute buds; telson with 8 + 8 setae 2 2. Rostrum without teeth; pereopods 4 and 5 tucked under cephalothorax; uropods enclosed Stage II Rostrum with ^ 1 dorsal tooth at base; pereo- pods 4 and 5 extend ventrally; uropods free ... 3 3. Rostrum with 1 or 2 dorsal teeth at base; antenna] flagellum with 8 segments and same length as scale; endopodite of uropod about 1/3 length of exopodite Stage III Rostrum with ^ 4 dorsal teeth; antennal flagellum with > 15 segments and longer than scale; endopodite of uropod > 1/2 length of exopodite 4 4. Rostrum with 4-8 dorsal teeth (usually 6), without ventral teeth; tip of rostrum styli- form; pleopods with a few small setae at tip; endopodite of uropod about 2/3 length of exopodite Stage IV Rostrum with > 8 dorsal teeth and ^ 4 ven- tral teeth; tip of rostrum bifid; pleopods setose; endopodite of uropod nearly as long as expodite 5 5. Rostrum with 9-12 dorsal teeth and 4 or 5 small ventral teeth; carapace with supra- orbital spine; telson with 2 spines on each lateral margin; lateral margins of telson nearly parallel Stage V Rostrum with > 12 dorsal teeth and ^ 6 distinct ventral teeth; carapace without supraorbital spine; telson with ^ 4 spines on each lateral margin; lateral margins of telson taper posteriorly 6 6. Mandibles without palps; exopodites on max- illipeds and pereopods reduced in size; lateral margins of telson converge posteriorly but widen slightly at junction with posterior margin Stage VI (megalopa) Mandibles with palps; vestigial exopodites on maxillipeds and pereopods; telson with lat- eral margins converging to narrow tip, as in adult Stage VII (first juvenile) Pandalus danae Stimpson Six larval stages. Stage I, known parentage; Stages II-VI, from plankton; figures 3-5 in Berkeley (1930). In all described stages, carapace without denticles on anteroventral margin; basipodite of maxillule without subterminal seta; posterior margins of ab- dominal somites without spinules. In Stage I, thoracic appendages developed; naked exopodites on pereopods 1 and 2; telson jointed with abdominal somite 6. Length: Stages I-VI, 5.7-17.0 mm. Range: Resurrection Bay, AK, to Point Loma, CA; depth, intertidal to 185 m (Butler 1980). Key IV. — Larval stages of Pandalus danae. 1. Eyes sessile; carapace without supraorbital spine; exopodites on pereopods 1 and 2 with- out natatory setae Stage I Eyes stalked; carapace with supraorbital spines; exopodites on pereopods 1 and 2 with natatory setae 2 2. Rostrum without teeth; uropods enclosed . . . Stage II Rostrum with teeth; uropods free 3 3. Rostrum with 2 or 3 minute dorsal teeth at base; endopodite of uropods noticeably shorter than exopodite Stage III Rostrum with ^ 8 teeth dorsally; endopodite of uropods same length as exopodite 4 4. Pleopods small, slightly cleft buds; telson widens slightly posteriorly Stage IV 267 FISHERY BULLETIN: VOL. 83, NO. 3 Table 4— Morphological characteristics for distinguishing between larvae of Pandalus borealis and P. goniurus reared in situ in Kachemak Bay, Alasl 2 dorsal teeth; antennal flagellum extends beyond plumose setae of antennal scale; chela of pereopod 2 nearly or fully formed; lateral margins of telson nearly parallel or narrow posteriorly 5 5. Carapace with supraorbital spine; rostrum with 5 or 6 dorsal teeth, no ventral teeth; tip of rostrum styliform (may have undeveloped bifid tip); lateral margins of telson nearly parallel Stage V Carapace without supraorbital spine; rostrum with dorsal and ventral teeth; tip of rostrum bifid; lateral margins of telson narrow posteriorly 6 6. One or two setae betw^een several posterior dorsal teeth of rostrum; 2-segmented mandi- bular palp without setae; exopodites of maxil- liped 3 and pereopods present but reduced; carpopodite of left and right pereopods 2 with 20-25 and 7-9 joints, respectively Stage VI (megalopa) One or two setae between most, if not all, rostral teeth; 3-segmented mandibular palp with spiniform setae; no exopodites on maxil- liped 3 and pereopods; carpopodites of left and right pereopods 2 with 29 and 1 1 joints, respectively Stage VII Pandalus hypsinotus Brandt Seven larval stages. Stages I-VI, VII (megalopa), and VIII-IX (juven- iles), known parentage; figures 1-6 in Haynes (1976). Stage I, known parentage; Stages II-V, from plankton; figures 5-6 in Kurata (1964a). Stage I, known parentage; Stages II and III, from 269 FISHERY tU'LLETIN: VOL. 83. NO. 3 plankton; figures 6 (only Stage I figured) in Berkeley (1930). In all described stages, carapace without denticles on anteroventral margin; posterior margins of ab- dominal somites without spinules. In Stage I, thoracic appendages segmented, dactyli undevelop- ed; naked exopodites on pereopods 1 and 2; telson not jointed with abdominal somite 6. Beginning in Stage III, basipodite of maxillule with subterminal seta; anal spine present. Length: Stages I-VII, 5.5-12.8 mm. Range: Sea of Japan to western Ber- ing Sea; Norton Sound to Puget Sound, WA; depth, 5-460 m (Butler 1980). Key VI. — Larval stages of Pandalus hypsinotus. 1. Eyes sessile; naked exopodites on pereopods Stage I Eyes stalked; exopodites on pereopods with natatory setae 2 2. Rostrum without teeth; uropods enclosed . . . Stage II Rostrum with ^ 1 dorsal tooth at base; uropods free 3 3. Rostrum with dorsal tooth at base and no ventral teeth; exopodites on maxilliped and pereopods with natatory setae Stage III Rostrum with ^11 dorsal teeth and ^ 2 small ventral teeth; naked, vestigial exopo- dites on maxillipeds and pereopods 4 4. Rostrum with 11-13 dorsal teeth and 2 or 3 small ventral teeth; tip of rostrum not bifid; mandibular palps unsegmented; telson margins nearly parallel but widen slightly posteriorly Stage IV Rostrum with > 13 dorsal teeth and > 3 ven- tral teeth; tip of rostrum bifid; mandibular palp with 3 segments; telson slightly wider at midlength or narrows posteriorly 5 5. Bilobed pleopods without setae; telson slight- ly wider at midlength Stage V Biramous pleopods with setae; telson nar- rows posteriorly 6 6. Carpopodite of left and right pereopods 2 with 19 and 7 or 8 joints, respectively; pleo- pods with setae only at tip; telson with 3 pairs of dorsolateral spines Stage VI Carpopodite of left and right pereopods 2 with 24 or 25 and 10 joints, respectively; pleopods setose; telson with ^ 4 pairs of dorsolateral spines Stage VII (megalopa) Pandalus jordani Rathbun Thirteen larval stages. Stages I-XIII, known parentage; figures 1-7 in Modin and Cox (1967). Stages I-XIII, Stages XIV and XV (juveniles), all from both known parentage and plankton; fig- ures 1-15 in Rothlisberg(1980). Stages I-VIII, known parentage; figures 5-11 in Lee (1969). In all described stages, carapace and abdominal somites not flared laterally and lack denticles or spinules; basipodite of maxillule without subterminal seta (except possibly Stage I). Pereopods tucked under carapace in Stage I. Rostrum without teeth in Stage I; rostnun with undeveloped dorsal tooth in Stage II; rostral tip bifid beginning at Stage VIII. Length: Stages I-XIII, 5.1-16.3 mm. Range: Un- alaska. AK, to San Nicolas Island, CA; depth, 36-457 m (Butler 1980). Key VII. — Larval and first juvenile stages of Pandalus jordani. 1. Eyes sessile; telson and abdominal somite 6 not jointed Stage I Eyes stalked; telson and abdominal somite 6 jointed 2 2. Rostrum with precursor of first dorsal tooth; uropods enclosed Stage II Rostrum with ^ 1 movable dorsal tooth; uropods free 3 3. Rostrum with 1 movable dorsal tooth; endo- podite of uropod a bud with only a few setae " Stage III Rostrum with > 1 movable tooth; endopodite of uropod > 1/2 length of exopodite and se- tose 4 4. Posterior width of telson about twice anterior width Stage IV Posterior width of telson noticeably less than twice anterior width 5 5. Lateral margins of telson widen slightly posteriorly Stage V 270 IIAYNKS: I'ANDAl.lDAK. llll'l'(>l,> TIDAK. AM i CKANCONMDAK I.AKVAK Lateral niarjiins of telsoii parallel (ir narrow posteriorly 6 6. Rostrum with 5 developed dorsal teeth and 1 un(ieveloped dorsal tooth; antennal flaj^ellum with If) segments and same lenjj;th as scale Stage VI Rostrum with ^ 5 developed dorsal teeth and > 1 undeveloped dorsal tooth; antennal tla- gellum with > 20 segments and longer than scale 7 7. Rostrum with 5 developed dorsal teeth and styliform tip; anal spine absent Stage VII Rostnmi with > 5 developed teeth, tip of rostrum with precursor of sub terminal tooth; anal spine present 8 8. Rostrum with 7 developed teeth; lateral margin of telson with 2 spines Stage VIII Rostrum with > 7 developed teeth; lateral margin of telson with ^ 3 spines 9 9. Rostrum with 9 developed teeth; inner flagel- lum of antennule with 3 segments; outer flagellum of antennule with 2 segments Stage IX Rostnun with > 9 developed teeth; inner flagellum of antennules with ^ 4 segments; outer flagellum of antennule with 3 segments 10 10. Rostrum with 10 developed teeth; medial pair of terminal telsonic spines shorter than adjacent pair Stage X Rostrum with > 10 developed teeth; medial pair of terminal telsonic spines same length or longer than adjacent pair 11 11. Inner flagellum of antennule with 4 seg- ments; medial pair of terminal telsonic spines same length as adjacent pair Stage XI Inner flagellum of antennule with 5 seg- ments; medial pair of terminal telsonic spines longer than adjacent pair 12 12. Rostrum with 12 developed dorsal teeth and precursors of 3 ventral spines; terminal margin of telson straight Stage XII Rostrum with > 12 developed dorsal teeth and precursors of > 3 ventral teeth; terminal margin of telson convex 13 13. Carapace with supraorbital spine; pereopod 2 with unsegmented car|)us Stage XIII (last larval stage) Carapace without supraorbital spine; pereo- pod 2 with segmented carpus Stage XIV (first juvenile stage) Pandalus kessleri Czerniavski Voxxv larval stages. Stages I-IV and Stage V (first juvenile), known parentage; figures 2-6 (fig. 6, first juvenile stage) in Kurata(1955). (Stages II-IV have a mixture of zoeal and megalopal characters.) Abbreviated larval development. In all described stages, carapace without denticles along antero- ventral margin; antennal scale not jointed at tip; 2 lateral setae proximally on exopodite of maxilliped 1 . Stages II and III with transverse dorsal groove. Rostrum with teeth beginning in Stage II. Supra- orbital spines in Stage II only. In Stage I, antennal flagellum 3/4 length of body; antennal flagellum segmented throughout its length; telson with 30-34 setae; anal spine present; zoea longer than 8.1 mm. In Stage III, mandible with unjointed palp. Vestigial exopodites on pereopods 1 and 2 in Stages I and II. Pleopods with plumose setae in Stage II. Length: Stages I-IV, 8.1-10.8 mm. Range: Hokkaido, Japan (no depth given) (Kurata 1955). Key VIII. — Larval and first juvenile stages of Pandalus kessleri. 1. Rostrum without teeth; eyes sessile .... Stage I Rostrum with teeth, eyes stalked 2 2. Mandible without palp; uropods enclosed; carapace with supraorbital spine Stage II Mandible with palp; uropods free; carapace without supraorbital spine 3 3. Rostral tip not bifid; telson widens posteri- orly Stage III Rostral tip bifid; telson margins parallel or narrows posteriorly 4 4. Mandibular palp unsegmented; sides of telson parallel; telson with 2 pairs of lateral spines and 11-14 terminal seta Stage IV Mandibular palp segmented; telson narrows posteriorly; telson with 3 pairs of lateral spines and several terminal setae vestigial as hairs Stage V (first juvenile) 271 FISHERY BULLETIN: VOL. 83, NO. 3 Pandalus platyceros Brandt Five larval stages. Stages I-V and Stages VI-X (juveniles), known parentage; figures 1-6 in Price and Chew (1972). Stage I, known parentage; Stages II, IV?, and V?, from plankton; figures 9 and 10 in Berkeley (1930). In all described stages, rostrum with teeth, basi- podite of maxillule without subterminal seta. In Stages I-III, carapace flares laterally, lateral margins with denticles; abdominal somites flare laterally, lateral margins with spinules (Fig. 5). In Stage I, pleopods present as buds; telson jointed with abdominal somite 6. In Stages II and III, less flaring of abdominal somites and smaller denticles and spinules than in Stage I. Thoracic appendages devel- oped in Stage I, except naked endopodites on pereo- pods 1-3. Length: Stages I-V, 8.1-13.0 mm. Range: Sea of Japan, Hokkaido; Unalaska Island, AK, to off San Diego, CA; depth, intertidal to 487 m (Butler 1980). Key IX. — Larval and first juvenile stages of Pandalus platyceros. 1. Rostrum without ventral teeth (Fig. 6); eyes sessile; pereopods with naked exopodites Stage I Rostrum with ventral teeth; eyes stalked; pereopods with setose exopodites 2 2. Antennal flagellum with 6 segments and flagellum only slightly longer than plumose setae of scale; uropods enclosed Stage II Antennal flagellum with ^15 segments and flagellum at least twice length of scale; uropods free 3 3. Antennal flagellum about twice length of antennal scale; telson widens posteriorly; telson with 1 pair of lateral spines .... Stage III Antennal flagellum > 3 times length of antennal scale; telson margins parallel or converge posteriorly; telson with > 1 pair of lateral spines 4 4. Carapace with supraorbital spines; mandibles without palps; pereopod 2 with unsegmented carpopodite; telson with 2 pairs of lateral spines Stage IV Carapace without supraorbital spines; mandi- bles with palps; pereopod 2 with segmented Figure 5. -Dorsal view of body. Stage I zoea of Payi- dalles platyceros. L 1 . 0 mm Figure 6. - Rostrum, Stage I zoea Pandalus platyceros. carpopodite; telson with > 3 pairs of lateral spines 5 5. Telson margins nearly parallel, distal margin concave Stage V (megalopa) 272 HAVNKS: PANnAl.lDAK. mi'l'dl-VTIHAE. ANDCKANCONIDAK LAKVAK Telson narrows posteriorly, distjil margin convex Stage VI (first juvenile stage) Pandalus prensor Stimpson Five larval stages. Stages I-V (larvae) and VI-IX (juveniles), known parentage; figures 2-7 in Mikulich and Ivanov (1983). Abbreviated larval development characterized by marked heterochrony in development of appendages. In all described stages, supraorbital spine absent; antennal flagellum segmented; antennal scale not jointed at tip; pereopods developed. In Stage I, rostrum curves ventrally between eyes; telson discoid and jointed with abdominal somite 6; natatory setae on exopodites of maxillipeds 1-3 and pereopods 1-2. Length: Stages I-V, carapace length 1.42-1.75 mm. Range: coastal waters of southern Okhotsk Sea, Sea of Japan, and southeastern Siberia (Vladivostok and Possjet Bay); depth, 2-93 m (Holthuis 1976; Mikulich and Ivanov 1983). Key X. — Larval stages of Pandalus prensor. 1. Eyes sessile; rostrom bends ventrally be- tween eyes Stage I Eyes stalked; rostrom straight 2 2. Rostrum does not extend to anterior margin of eye; telson discoid; uropods enclosed Stage II Rostrum extends beyond anterior margin of eye; telson rectangular; uropods free 3 3. Pleopods uniramous and unsegmented; tel- son with 7 pairs of terminal spines .... Stage III Pleopods (pairs II- V) biramous and segment- ed; telson with ^ 5 pairs of terminal spines ... 4 4. Ventral rostrum with 4 teeth; telson with 5 pairs of terminal spines Stage IV Ventral rostrum with 6 teeth; telson with 3 pairs of terminal spines Stage V Pandalus stenolepis Rathbun Six larval stages. Stages I and II, known parentage; Stages III-VII (Stage VII, first juvenile), from plankton; figures 1 and 2 in Needier (1938); figure 73 in Gurney (1942; Page 208 verifies subterminal seta on maxillule). In Stages I-IV, carapace flares laterally, lateral margin with denticles; abdominal somites with spinules and flared laterally. P^laring, size of den- ticles, and spinules decrease in Stages II-IV. In Stage I, pereopods tucked under carapace; telson not jointed with abdominal somite 6; and flagellum of antenna longer than antennal scale. Pleopods absent until Stage III. Length: Stages I-VI, 6.0-14.0 mm. Range: Unalaska Island, AK, to Hecata Bank, OR; depth, 49-229 m (Butler 1980). Key XI. — Larval and first juvenile stages of Pandalus stenolepis. 1 . Eyes sessile; rostrum without teeth .... Stage I Eyes stalked; rostrum with teeth 2 2. Rostrum with only dorsal teeth (4-5 teeth); uropods enclosed Stage II Rostrum with dorsal and ventral teeth; uropods free 3 3. Rostrum with 8 or 9 dorsal teeth and 2 ven- tral teeth; pleopod buds only slightly cleft Stage III Rostrum with > 9 dorsal teeth and > 2 ven- tral teeth; pleopods biramous 4 4. Unjointed pleopods without setae .... Stage IV Jointed pleopods with setae 5 5. Right and left pereopods 2 with endopodites of same length; pleopods with a few setae; each endopodite without an appendix interna (Fig. IN, ai) Stage V Right and left pereopods 2 with endopodites of different lengths; pleopods setose; each endopodite with an appendix interna 6 6. Carapace with supraorbital spine; setose exopodite on each pereopod Stage VI (megalopa) Carapace without supraorbital spine; naked, vestigial exopodite on each pereopods Stage VII (first juvenile stage) Pandalus tridens Rathbun Probably 8 larval stages. Stage I, known parentage; Stages I-VII, from plankton; figures 1-7 in Haynes (1976). Stage I, known parentage; figure 1 in Ivanov (1971). 273 FISHERY BULLETIN: VOL. 83, NO. 3 In all described stages, carapace and abdominal somites not flared laterally; antennal scale relatively long and narrow (about 5-7 times as long as wide). In Stages I-III, but rarely in Stage IV, carapace with denticles along anteroventral and posteroventral margins; posterior margin of abdominal somites 1-5 fringed with spinules (Fig. 7). Rostrum sinuate, pro- jects somewhat upwards in Stages I-III, remains shorter than carapace as late as Stage VIII, without teeth until Stage IV. Antennal flagellum shorter than antennal scale through at least Stage V. Length: Stages I- VII, 3.1-13.0 mm. Range: Bering Sea to San Nicolas Island, CA; depth, 5-1,984 m (Butler 1980). Key XII.— Larval stages (Stages I-VII) of Pandalus tridens. 1. Eyes sessile; carapace without supraorbital spine; pereopods 1-3 without exopodites; telson with 7-1-7 setae Stage I Eyes stalked; carapace with supraorbital spine; exopodites on pereopods 1-3; telson with 8-1-8 setae 2 2. Uropods enclosed Stage II Uropods free 3 3. Endopodite of uropod < 1/2 length of exopodite Stage III Endopodite of uropod > 1/2 length of exopodite 4 4. Rostrum with 2 dorsal teeth; endopodite of uropod about 3/4 length of exopodite; telson widens posteriorly Stage IV Rostrum with > 2 dorsal teeth; endopodite of uropod nearly same length as exopodite; lateral margins of telson nearly parallel 5 5. Antennal flagellum with 5 segments and about 2/3 length of antennal scale; chela of pereopod 2 slightly developed (Fig. 8); pleo- pod 2 about 1/4 height of abdominal somite 2 Stage V Antennal flagellum with ^ 20 segments and as long as or longer than antennal scale; chela of pereopod 2 well formed (Fig. 9); pleopods 2 at least 1/2 height of abdominal somite 2 6 6. Rostrum with 6 dorsal teeth; pleopods without setae; telson slightly wider near center Stage VI Figure 7. -Abdomen, Stage II zoea of Pandalus tridens showing spinules on posterior margins of somites. 0.25 mm Figure 8. -Chela of pereopod 2, Stage V zoea of Pandalus tridens. 274 HAYNES: PANDALIDAE, HIPPOLYTIDAE, AND CRANGONIDAE LARVAE Rostrum with 7 dorsal teeth; pleopods tipped with a few setae; telson margins nearly parallel Stage VII Figure 9. -Chela of pereopod 2, Stage VI zoea of Pandalns tridens. 0.25 mm HIPPOLYTIDAE (Genera Eualus, Heptacarpus, Hippo lyte, Lebbeus, and Spirontocaris) In all described stages, rostrum absent to long, usually spiniform (slightly sinuate in species with abbreviated development); plumose seta rather than long setose spine on inner flagellum of antennule; exopodites on pereopods 1-2, 1-3, or 1-4; abdomen without dorsal spine or keels on somite 3 (megalopa of Spirontocaris spinus with a minute dorsoposterior spine on abdominal somite 3); posterolateral spines absent, on abdominal somites 4 and 5, or only on abdominal somite 5 (spines may be lacking in mega- lopa); pereopod 1 never subchelate; anal spine pres- ent in all stages (exception: at least Stage I of Hyp- poly te). Rostrum may have teeth in last zoeal stage (megalopa); supraorbital spine usually absent in Stage I. Stages I-III, antennal scale nearly always jointed or partially jointed at tip. Stage I, exopodites of maxilliped 1-3 with 4, 5, 5 natatory setae; about Stage V, setose setae on endopodite of each pleopod; telson does not widen posteriorly, has more than 1 pair of lateral spines. The principal morphological characters and number of larval stages of known larvae of hippolytid shrimp of the northern North Pacific Ocean are sum- marized in Table 5. Table 5. — Principal morphological characteristics and nunnber of larval stages of known larvae of hippolytid shrimp of the northern North Pacific Ocean. + = yes; - = no; ? = unknown. Pereopods Postero- Supra- bearing lateral orbital Pereopods an exopodite spines on Tel sonic No. of spine in In in later abdominal spi nes in larval Species Rostrum Stage 1 Stage 1 zoeal stages somites Stage 1 stages Eualus barbatus ? 7 + 7 ? E. fabricii + — '5 1-3 4,5 7 + 7 ? E. gaimardii + — M 1-3 5 7 + 7 6 E. herdmani — — — ? — 7 + 7 ? E. macilentus -1- = — — ? — 7 + 7 ? E. pusiolus + ^ ? — 1-4 — 7 + 7 6-7 E. suckleyi + — '5 1-3 5 7 + 7 ? Heptacarpus brevirostris — — — ? — 7 + 7 ? H. camtschaticus + ' — '5 1,2 — 7 + 7 ? H. paludicola — — — ? — 7 + 7 ? H. tridens — — — ? — 7 + 7 ? Hippolyte clarki -1- — '1 ? 5 7 + 7 ? Lebbeus groenlandicus + — 5 — 4, 5 9(10) + 10(11) 3 L. polarls + — 5 — 4, 5 9 + 9 4 Spirontocaris arcuata -1- — '5 1, 2 — 7 + 7 ? S. liUjeborgii -t- -1- '5 1, 2 4 7 + 7 6 S. murdochi + — '5 1,2 4, 5 7 + 7 ? S. ochotensis — -1- '5 1,2 4,5 7 + 7 ? S. phippsii 4- + '5 1,2 4,5 7 + 7 6 S. spinus + + '5 1,2 4,5 7 + 7 6 S. spinus var. intermedia + — ^5 1,2 4,5 7 + 7 ? 'Undeveloped pereopods. 'Minute. 275 FISHERY BULLETIN: VOL. 83, NO. 8 Genus Eualus Thallwitz In Stage I, rostrum absent to long; carapace without supraorbital spine; tip of antennal scale jointed; pereopods absent or, if present, undevelop- ed; anal spine present. Exopodites first appear on pereopods 1-3 or 1-4 in Stage III. Posterolateral spines absent, on abdominal somites 4 and 5, or only on abdominal somite 5. Eualus barbatus (Rathbun) Only Stage I described, known parentage; figure 3 in Ivanov (1971). Carapace without rostrum, supraorbital spine, or denticles; pereopods absent; abdominal somites with- out posterolateral spines or denticles but with isolated hairs on dorsal surface of abdominal somites 3 and 4; abdominal somite 3 with indistinct row of setae on dorsal surface. Length: 4.5 mm. Range: Pribilof Islands, AK, to Hecata Bank, OR; depth, 82-507 m (Butler 1980). Eualus fabricii (KrOyer) Only Stages I and II described, known parentage; figure 5 in Haynes (1981). Not "Spirontocaris larva Nr. 3, Spirontocaris fabricii!" as described by Stephensen (1916) (see Haynes 1981). Not "SpirontocarisAarvaie No. 3? Spirontocaris fabricii (Kr0yer)" as described by Stephensen (1935) (see Haynes 1981). Not Spirontocaris fabricii as described by Frost (1936) (see Haynes 1981). Not "Eualus fabricii (Kr<)yer)" as described by Pike and Williamson (1961) (see Haynes 1981). In all stages, posterolateral spines on abdominal somites 4 and 5. In Stage I, antennal flagellum about 1.5 times length of antennal scale; minute spinules along dorsoposterior margins of abdominal somites 4 and 5 (spinules absent in Stage II); supraorbital spine absent (small in Stage II). In Stages I and II, pereo- pods 1-3 with undeveloped exopodites. In Stage II, exopodites of maxillipeds 1-3 with 4, 9, and 11 nata- tory setae, respectively; telson not jointed with ab- dominal somite 6. Length: Stages I and II, 3.5-4.3 mm. Range: Sea of Japan, Okhotsk Sea; Chukchi Sea to British Columbia; in northwestern Alan tic, from Foxe Basin and West Greenland to Massachu- setts Bay (eastern United States); depth, 4-255 m (Butler 1980). Eualus gaimardii (H. Milne Edwards) Six larval stages. Stages I-VI, known parentage; also Stages I and 11 from plankton; figure 2 in Pike and William- son (1961). Last zoeal stage from plankton, described as "Spirontocaris-larva No. 2A" by Stephensen (1935), probably E. gaimardii forma gibba (see Pike and Williamson 1961, p. 198). No figure. Stage V (?), described as "Spirontocaris B" by Frost (1936), probably £". gaimardii {orma. gibba (see Pike and Williamson 1961, p. 198); figure 4 in Frost (1936). Stage I, known parentage; described as Spironto- caris gaimardii by Lebour (1940); figure 1 in Lebour(1940). Stage I, known parentage; figures 21-23 in Williamson (1957a: figures from Lebour 1940). In all described stages, rostrum long (about 1/3 carapace length); no subterminal seta on maxillule. Rostrum without teeth until Stage V. In Stage V, rostrum with 2 dorsal teeth; in Stage VI, rostrum with 3 dorsal teeth. Supraorbital spine in Stages II-V. Carapace with 3 or 4 denticles on anteroventral margin in Stages I-IV. Antennal flagellum does not extend beyond antennal scale until Stage V. Nata- tory setae on exopodites of maxillipeds 1-3: 5, 7, 7 natatory setae, respectively, in Stage II; 5, 9, 9 in Stage HI; and 5, 10, 10 in Stages IV and V. In Stages I-V, posterolateral spine on abdominal somite 5. Length: Stages I-VI, 2.9-5.4 mm. Range: cir- cumpolar, southward to North Sea; Cape Cod, MA; Sitka, AK; Siberia; depth, 10-900 m (Holthuis 1947). Eualus herdtnaHt (Walker) Only Stage I described, known parentage; describ- ed as "Spirontocaris herdmani" by Needier (1934). No figure. Stage I, known parentage; Pike and Williamson (1961: description from Needier 1934). No figure. Carapace without rostrum or supraorbital spine but with 3 anteroventral denticles; abdomen without posterolateral spines; anal spine minute. No length given. Range: Sitka, AK, to Puget Sound, WA; depth, 18-232 m (Butler 1980). Eualus macilentus (KrOyer) Only Stage I described, known parentage; figure 2 in Ivanov (1971). 276 IIAVNKS; I'AMiAl.lDAK. HII'I'Ol.Vl'IDAK. ANDCKANCONIDAK LAk\ AK No supraorbital spine; 4 denticles on anteroventral martjin of carapace; pereopods absent; abdomen witliout spines or denticles. Lenj^th: 3.0 mm. Range: West Greenland to Nova Scotia; Bering and Okhotsk Seas; depth. 150-540 m (Holthuis 1947). Eualus pusio/ijs (Kr0yer) Larvae described from Atlantic specimens. Seven or eight larval stages. Stage I, known parentage; Stages I-VII, from plankton; figure 3 in Pike and Williamson (1961). Last zoeal stage, from plankton, described as "Spirontocaris C" by Frost (1936); figure 5 in Frost (1936). Stage L known parentage; described as "Spironto- cay-is pusiola" by Bull (1938); figure 1 in Bull (1938). Stage I, known parentage; figures 27-30 in Williamson (1957a); figures from Bull (1938). It! all described stages, 3 denticles on anteroven- tral margin of carapace; no spines, denticles, or dor- sal setae on abdominal somites. Rostrum minute in Stage L only slightly larger in other stages, without teeth in all stages. Exopodites of maxillipeds 1-3, with 5,8, and 8 natator\' setae, respectively, in Stage II. In Stages V-VIII, pereopods 1-4 with setae. Length: Stages I-VIII, 2.2-4.8 mm. Range: Sea of Japan to Chukchi Sea and British Columbia; Gulf of St. Lawrence to Cape Cod, MA; Europe from the southwestern Barents Sea to Spain; depth, intertidal to 1.381 m (Butler 1980). Eualus suckleyi (Stimpson) Only Stages I and II described, known parentage; figTjres 3 and 4 in Haynes (1981). In Stage I, anteroventral margin of carapace with 3 or 4 denticles; undeveloped exopodites on pereo- pods 1-3. In Stage II, carapace with supraorbital spine, without denticles along anteroventral margin. In Stage II, maxillipeds 1-3 with 4, 5, and 5 natatory setae, respectively; pereopods 1-5 present but undeveloped. Length: Stage I, 3.0-3.5 mm; Stage II, 3.5-4.2 mm. Range: Okhotsk Sea to Chukchi Sea to about Grays Harbor, WA; depth, 11-1,025 m (Butler 1980)." Genus Heptacarpus Holmes Only Stage I described. Rostrum minute to absent; no supraorbital spine; pereopods absent or pairs 1 -5 present but undeveloped; abdominal somites without posterolateral spines; exopodites develop on pereo- pods 1 and 2; tip of antennal scale not always jointed. Heptacarpus brevirostris (Dana) Stage I, known parentage; described as Spironto- caris brevirostris by Needier (1934). No figures. Carapace without anteroventral spines; antennal scale partially jointed at tip. Length: Stage I, 1.5 mm. Range: Aleutian Islands, AK, to San Fran- cisco Bay, CA; depth, intertidal to 128 m (Butler 1980). Heptacarpus camtschaticus (Stimpson) Stage I, known parentage; figure 8 in Haynes (1981). Carapace without spines; minute rostrum curves slightly downward following dorsal contour of eyes; undeveloped exopodites on pereopods 1 and 2; abdo- men without spines or spinules. Length: Stage I. 2.9 mm. Range: Sea of Japan to Chukchi Sea and Strait of Georgia, WA; depth, intertidal to 108 m (Butler 1980). Heptacarpus paludicola (Holmes) Stage I, known parentage; described as Spironto- caris paludicola by Needier (1934); figure 1 in Needier (1934). Antennal scale partially jointed at tip; abdomen without posterolateral spines. Length: Stage I, 2.0 mm. Range: Tava Island, AK, to San Diego, CA; depth, intertidal to 10 m (Butler 1980). Heptacarpus tridens (Rathbun) Stage I, known parentage; described as Spironto- caris tridens by Needier (1934); figure 1 in Needier (1934). Carapace without anteroventral spines; antennal scale unsegmented. Length: Stage I, 3.0 mm. Range: Aleutian Islands, AK, to Cape Flatter^', WA; depth, intertidal to 110 m (Butler 1980). Genus Hippo lyte Leach Only Stage I described. Antennal scale without joints at tip. 277 FISHERY BULLKTIN: VOL. 8.S. NO. :>. Hippo lyte clark't Chace Stage I, known parentage; described as Hippolyte califomiensis by Needier (1934); figure 1 in Needier (1934). Rostrum long; carapace with 4 anteroventral den- ticles; bud of pereopod 1 present; small postero- lateral spines on abdominal somite 5. Length: 1.9 mm. Range: Sheep Bay, AK. to Puget Sound, WA; Santa Catalina Island, CA; depth, intertidal to 30.5 m (Butler 1980). Cape Cod, MA; Bering and Okhotsk Seas; Aleutian Islands, AK; depth, 0-930 m (Holthuis 1947). Genus Spirontocaris Bate In all described stages, rostrum absent to long; posterolateral spines on abdominal somites 4 and 5, or 5 only; no minute spines on posterior margins of abdominal somites. In Stage I, supraorbital spine present or absent; all pereopods present luit undevel- oped. Exopodites only on pereopods 1 and 2 in later stages (usually by Stage IV or V). Genus Lebbeus White In all described stages, posterolateral spines on abdominal somites 4 and 5 (Fig. 2). In Stage I, larvae relatively long (> 5.0 mm); rostrum long, slightly sinuate (Fig. 2), with no supraorbital spine. Abbre- viated development. Lebbeus groenlandicus (Fabricius) Three larval stages. Stages I-III from both known parentage and plank- ton; figures 1-3 in Haynes (1978b). Stage I. known parentage; figure 5 in Ivanov (1971). Not "SpironfocarisAarva. No. lA." as described by Stephensen (1935) (see Haynes 1978b). Somewhat more developed in each stage than lar- vae of L. polaris. In Stage II, no vestigial exopodites f)n pereopods. Telson with about 20 setae in Stages I and II, and 3 -(- 3 spines in Stage III (megalopa). Length: 6.4-7.6 mm. Range: Sea of Japan to Chukchi Sea to Puget Sound, WA; arctic coast of Canada; Greenland to Cape Cod, MA; depth, 11-518 m (Butler 1980). Lebbeus polaris (Sabine) Probably 4 larval stages. Stages I and II, known parentage; figures 1 and 2 in Haynes (1981). Neither Spirontocaris polaris (= L. polaris) as described by Stephensen (1916) nor "Spironto- mnls-larva No. 1" as described by Stephensen (1935) (see Haynes 1981). In Stages I and II, telson with 9 -i- 9 setae. In Stage II, vestigial exopodite on peropod 1. Length: Stage I, 5.2 mm; Stage II, 5.8 mm. Range: circum- polar, southward to the Skagerrak and Hebrides; Spirontocaris arcuata Rathbun Only Stage I described, known parentage; figiuv 6 in Haynes (1981). Rostrum short (about 1/7 carapace length), pro- jects downward following contour of eyes; 2 or 3 minute denticles on anteroventral margin of cara- pace; supraorbital spine absent; posterolateral spines on abdominal somites 4 and 5. Length: 4.1-4.4 mm. Range: Sea of Japan to Chukchi Sea to Juan de Fuca Strait, WA; Canadian Arctic; depth, 5-641 m (Butler 1980). Spirontocaris lilljehoriiii (Danielssen) Larvae described from Atlantic Ocean. Six larval stages. Stages I and II, known parentage; Stages I-\'. from plankton; Stages VI (megaloi:)a) and \1I (first juvenile), reared in laboratory from Stage V; figure 1 in Pike and Williamson (1961). Stage I. known parentage; described as>). sjii)ii(s var. lilljehorgi by Lebour (1937); figure 1 in Lebour (1937). Stage I; figures 14-16 iti Williamson (1957a). De- scription and figures from Lebour (1937). In all described stages, posteroanterior margins of carapace smooth; abdominal somite 4 with a dorsal tuft of short setae, without posterolateral spines. Rostrum long (about 3/4 length of antennular pedun- cle), deepens slightly in later stages but does not develop teeth until megalopa. Supraorbital spine rudimentary in Stiige I, clearly defined in later stages. F\)sterolateral spines on abdominal somite 5 becoming smaller in later stages and may be absent in Stages V and VI. Megalopa and first juvenile stage without dorsal tooth on posterior margin of abdominal somite 3. Length: Stages I-VI, 4.8-8.5 mm. Range: From Spitsbergen and southwestern 278 IIAVNES: I'ANDAI.IDAK. IIIl'rol.VI'IDAK, AMKKANCOMDAK I.AKNAK Barents Sea south to south coast of England; Ice- land; (ireenland; east coast of North America from Nova Scotia to Massachusetts Bay; arctic Alaska; depth, 20-1,200 m (Holthuis 1947). Spirontocaris murdochi Rathbun Only Stages I-III described, known parentage; figures 1-3 in Haynes (in press). Rostrum about 1/4 carapace length, supraorbital spine in Stage III; posterolateral spine on abdominal somite 5 longer than posterolateral spine on abdo- minal somite 4; dorsal surface of abdominal somite 4 without tuft of setae. Length: Stages I-III, 3.2-4.3 nun. Range: Arctic to southeastern Alaska, Kam- chatka, Sea of Okhotsk, Patience Bay (Sakhalin); depth, 18-50 m (Holthuis 1947; Haynes in press). Spirontocaris ochotetisis (Brandt) Only Stage I described, known parentage; figure 7 in Haynes (1981). No rostrum; carapace usually with only 1 denticle along anteroventral margin; supraorbital spine minute; posterolateral spines on abdominal somites 4 and 5. Length: 2.8 mm. Range: Sea of Japan to Bering Sea and western coast of Vancouver Island, British Columbia; depth, intertidal to 247 m (Butler 1980). Spirontocaris phippsii (Kr0yer) Larvae known from Atlantic Ocean. Stage II, from plankton; Pike and Williamson (1961). No figure. Identity assumed from distri- bution of adults. Stage III, from plankton; described as "Sprionto- cam-larva Nr. 2," ("Sp. turgidaT); figure 6 in Stephensen (1916). Not figure 7 in Stephensen (1916), "last stage"; probably Eualus macilentus (see Pike and Williamson 1961) (Pike and Williamson identified figure 7 as E. macilentus based on identity of S. spinus and S. lilljehorgii and distribution of Eualus spp. in Greenland waters). Stage V, from plankton; described as "Spironto- caris-larva No. 2 (? Sp. turgida (Kr0yer))" and ''Spirontocaris-\a.r\'s. No. 2B" in Stephensen (1935) (see Pike and Williamson 1961 for identi- fication). Spines on abdominal somites 4 and 5; abdominal somite 4 without dorsal tuft of setae. Length: Stage II, 6.0 mm. Range: circumpolar, southward to northern Norway; Cape Cod, MA; Shumagin Islands, AK; and Plover Bay, Siberia; depth, 11-225 m (Holthuis 1947). Spirontocaris spinus (Sowerby) Larvae described from Atlantic Ocean only. Six larval stages. Stages I and II, known parentage; Stages III-VII (Stage VII, first juvenile), from plankton; figure 1 in Pike and Williamson (1961). Stage IV, from plankton; described as "Spironto- caris A" by Frost (1936); figure 3 in Frost (1936) (see Pike and Williamson 1961 for identifica- tion). Stage I, probably from known parentage; Stage V, probably from plankton (see Pike and William- son 1961); figures 17-20 in Williamson (1957a). Larvae and juvenile stages very similar to those of S. lilljeborgii. In all described stages, abdominal somite 4 with a dorsal tuft of short setae; abdominal somites 4 and 5 with posterolateral spines; postero- lateral spines on abdominal somite 5 remain same size in all zoeal stages. Posterior margin of abdo- minal .somite 3 with distinct dorsal tooth in megalopa and first juvenile stage. Length: Stages I-VI, 4.3-8.0 mm. Range: Circumpolar, southward to the northern North Sea, Massachusetts Bay (eastern United States), Alaska Peninsula, and eastern coast of Siberia; depth, 16-400 m (Holthuis 1947). Spirontocaris spinns var. intermedia Makarov* Only Stage I described, known parentage; figure 4 in Ivanov (1971). Not S. spimis intermedia as described by Makarov (1967) (see Ivanov 1971). Rostrum long (> 1/3 carapace length); no supra- orbital spine; abdominal somite 4 with dorsal tuft of setae; posterolateral spines on abdominal somites 4 and 5. Length: Stage I, 5.0 mm. Range: (see S. spinus); depth, 9-1,380 m (Hayashi 1977). ^According to Ivanov (1971), V. V. Makarov, rather than Z. I. Kobjakova, is the author of the subspecies 5. spimis var. intervriedia based on Article .51(c) of Chapter XI of the International Code of Zoological Nomenclature (International Commission on Zoological Nomenclature 1964). The subspecies S. spinus var. intermedia. however, may be identical toS. spiv^is. Hayashi (1977) believed that the morphological criteria used by Kobjakova (1937) to distinguish 5. spinus var. intermedia from S. spimis were too small and variable to be valid. 279 FISHERY BULLETIN: VOL. 83, NO. 3 CRANGONIDAE (Genera Argis, Crangon, Mesocrangon, Paracrangon, Sabinea, and Sclero crangon) Rostrum nearly always present and long (at least 1/4 length of carapace), spiniform, and always with- out teeth (rostrum spinulose in Stage I larvae of Paracrangon echinata); supraorbital spine absent; in- ner flagellum of antennule a setose spine or oblong projection rather than a plumose seta; tip of antennal scale never segmented; exopodite usually only on pereopod 1 (rarely on pereopod 2); maxillule without subterminal seta on basipodite; dorsal spine may be on abdominal somite 3 or keels on both abdominal somites 2 and 3; usually posterolateral spines on abdominal somite 5; endopodite of pleopods undevel- oped; usually telson widens posteriorly, never with more than 1 pair of lateral spines. Pereopods 1 sub- chelate at about Stage V. Anal spine absent until about Stage IV. The principal morphological characters and number of larval stages of known larvae of crangonid shrimp of the northern North Pacific Ocean are summarized in Table 6. Genus Argis Kr^yer In Stage I, rostrum styliform; pereopods and pleo- pods developed but not functional; exopodite on pereopod 1 rudimentary or absent; abdominal somite 3 without dorsal spine; abdominal somite 5 with posterolateral spine (posterolateral spine absent in megalopa of A. dentata). Argis crassa'' (Rathbun) Only Stage I described, known parentage; figure 2 in Ivanov (1968). Not Stage I A. crassa as described by Makarov (1967) in figure 21. Antennal scale without distal spine on outer margin; endopodite of maxillule with 5 setae; sca- phognathite of maxilla with 9 setae; abdominal somites 2-5 fringed dorsally with small spinules; telson and abdominal somite 6 jointed; telson with 8 + 8 setae. Length: 7.5 mm. Range: Sea of Japan to Bering Sea to San Juan Islands, WA; depth, 4-125 m (Butler 1980). Argis dentata (Rathbun) Three larval stages. Stage I described, known parentage from Pacific Ocean; Stage I-III de- scribed, from plankton from Atlantic Ocean. Stages I-III, from plankton; figures 1-6 in Squires (1965). Stage I, known parentage; figure 3 in Ivanov (1968). Megalopa, from plankton; described as "Necto- 'Makarov (1967) described a crangonid larva from plankton that has a short rostrum flattened dorsoventrally. He assumed it was Argis (= Nectocrangon) crassa. According to Ivanov (1968), who reared Stage I A. crassa from known parentage, the larva described by Makarov is neither Argis crassa nor a later stage of Argis crassa. The short flattened rostrum, however, is typically a post- larval (juvenile) characteristic of crangonid shrimp. The specimen described by Makarov, therefore, is probably a juvenile rather than a larva. Table 6. — Principal morphological characteristics and number of larval stages of known larvae of crangonid shrimp of the northern North Pacific Ocean, -i- = yes; - = no; ? = unknown. Pereopods Postero- bearing lateral Pereopods an exopodite spines on Dorsal spine Telsonic No. of in in later abdominal on abdominal spines in larval Species Rostrum Stage 1 zoeal stages somite somite Stage 1 stages Argis crassa + 5 5 8 + 8 '2 or 3 A. dentata + 5 5 — 8 -1- 8 3 A. lar + 5 — 5 — 7 -1- 7 '3 Crangon alaskensis + — 5 3 7 + 7 5 C. communis + M 5 — 7 + 7 '5 C. dalli + ^5 5 — 7 + 7 5 C. franciscorum angustimana + M 1,2 5 3 7 + 7 '5 C. septemspinosa + — 5 3 7 + 7 5-6 Mesocrangon intermedia + '5 5 — 8 -1- 8 5 Paracrangon echinata + ? 1.2 1-5 1-5 7 7-1- Sabinea septemcarinata + ? 5 — 16 + 16 '4 Sclerocrangon boreas — 5 — — — 12 + 12 2 S. salebrosa -»- 5 5 22 -1- 22 1 S. zenkevitchi — 5 — — — — '2 'Estimated. 'Undeveloped pereopods. 280 HAYNES: PANDALIDAK, HIPPOLYTIDAE, ANDCRANCONIDAE LARVAE crangon larl, young stage" by Stephensen (1916); figure 3 in Stephensen (1916). In Stage I, antennal scale with distal spine on outer margin; endopodite of maxillule and scaphog- nathite of maxilla with 6 setae each; abdominal somites 3-5 fringed dorsally with small spinules (spinules not mentioned for Atlantic specimens); telson and abdominal somite 6 jointed; telson with 8 + 8 setae. In megalopa, short, pointed rostrum ex- tends to middle of eye; carapace with 2 dorsal teeth, ventral edge fringed with short plumose setae; ab- dominal somite 5 without posterolateral spines. Length: Stages I-III, 8.0-12.0 mm. Range: Sea of Japan to Anadyr Gulf, Gulf of Georgia, and San Juan Islands, WA; arctic Canada to Nova Scotia, Canada; depth, intertidal to 2,090 m (Butler 1980). Argis la I (Owen) Probably 3 larval stages. Stages I and II, from plankton; described cisNecto- crangon lar by Makarov (1967); figure 22 in Makarov (1967). Not Crangonidae "Species F" (described by Kurata 1964b). as assumed by Makarov (1967). i" In Stage I, telson and abdominal somite 6 not jointed; telson with 7 -i- 7 setae. In Stage II, telson with 8-1-8 setae. Abdominal somites without spinules. Length: Stages I and II, 6.2-7.5 mm. Range: Sea of Japan to Chukchi Sea to Strait of Georgia, WA; depth, 10-280 m (Butler 1980). Genus Crangon Fabricius (= Crago Lamarck) Five or six zoeal stages. Anteroventral margin of carapace denticulate in most if not all larval stages. In all described stages, rostrum about 1/3 carapace length, spiniform, without teeth; posterolateral spines on abdominal somite 5; exopodites develop on either pereopod 1 or pereopods 1 and 2; ab- dominal somite 3 usually with dorsal spine; telson always widens posteriorly, with setae and < 8 + 8 spines. Crangon alaskensis Lockington Five larval stages. Stages I-VII (Stage VI, first juvenile), known parentage; illustrations 1-79 in Loveland (1968). In all described stages, rostrum barely reaches beyond eyes; dorsal spine on abdominal somite 3. Length: Stages I-V, 2.0-3.3 mm. Range: Kuril Islands; Bering Sea to Puget Sound, WA; depth, intertidal to 275 m (Butler 1980). Crangon communis Rathbun" Only Stage I described, known parentage; de- scribed as Sclerocrangon communis by Ivanov (1968); figure 1 in Ivanov (1968). Not Stages I I-V C. communis from plankton; de- scribed by Makarov (1967) as Sclerocrangon communis.^^ Antennal flagellum about 3/4 length of antennal scale; antennal scale with 14 setae; abdominal somites without keels; spinules on posterior margins of abdominal somites 3-5. Length: Stage I, 4.8 mm. Range: Sea of Japan to Chukchi Sea to San Diego, CA; depth, 16-1,537 m (Butler 1980). Crangon da Hi Rathbun Five larval stages. Stage I, known parentage; Stages II-V and VI (first juvenile, "postlarval"), from plankton; figure 18 in Makarov (1967). Larvae figured in part but not described. Larvae thought to be identical morphologically to C. allmani larvae from the Atlantic Ocean (Makarov 1967). Stage I from plankton; figures 7-9 in Birshteyn (1938). Not "last (?) stage" as described by Birshteyn (1938) (see Makarov 1967). Typical unabbreviated crangonid development. In all described stages, carapace without dorsal, lateral, or supraorbital spines; anterior margin of carapace denticulate; rostrum spiniform, without teeth; '"Morphological differences are too great for Makarov'.s Necto- rrnngon lar larvae and Kurata's "Species F" larvae to be identical. Makarov's larvae lack exopodites on pereopods in all stages and, in Stages I and II, have posterolateral spines on abdominal somite 5. Kurata's "Species F" larvae, known only in Stage II, have an exo- podite on pereopod 1 and posterolateral spines on abdominal somites 5 and 6. •'Zarenkov (1965) proposed placing C. commuvis in a new subgenus, Neocrangov. Butler (1980) has shown that Zarenkov's diagnosis for Neon-angon is invalid, at least for British Columbia species. Based on Butler's findings, I have retained C. mmmuni.^ in the genus CraJigcn). 12 It is unlikely that Makarov's (1967) larvae and Ivanov's (1968) larvae are the same species because Makarov's larvae have keels on abdominal somites 2 and 3, whereas Ivanov's larvae do not. 281 FISHERY BULLETIN: VOL. 88. NO. 3 posterolateral spines on abdominal somite 5; abdo- minal somites without spinules or keels; telson always widens posteriorly, never with > 1 pair of lateral spines. In Stage V, pleopods uniramous, with buds of endopodites. Length: Stages I-V, 2.8-7.0 mm. Range: Sea of Japan to Chukchi Sea, to Puget Sound. WA; depth, 3-630 m (Butler 1980). Crangon franciscorum angustiwana Rathbun Only Stage I described, known parentage; figure 1 in Haynes (1980b). Rostrum extends beyond eyes; carapace without shallow transverse groove; antennal scale with 10 plumose setae including subterminal seta along outer margin; endopodite with 4 segments on maxilliped 1, and 5 segments on maxilliped 3; exopodites of maxil- lipeds not jointed; pereopods 1-4 present but un- developed; buds of exopodites on perepods 1 and 2; median dorsal spine on abdominal somite 3; postero- lateral spines on abdominal somite 5; fifth pair of telsonic spines about equal in length to fourth and fifth pairs. Length: Stage I, 2.8-3.3 mm. Range: Kachemak Bay, AK, to Tillamook Rock, OR; depth, 18-183 m (Butler 1980). Crangon septemspinosa Say Described from specimens from both Atlantic and Pacific Oceans. Five or six larval stages. Stages I and II, known parentage; Stages III-VI, from plankton; described as C. affinis by Kurata (1964b); Pacific specimens; figures 1-29 in Kurata (1964b). Stage I, known parentage; Stages II-V, from plankton; Atlantic specimens; described as Crago septemspinosufi Say by Needier (1941); figures 1 and 2 in Needier (1941). Stages I-X (Stage X, first juvenile stage), known parentage (larval series likely includes extra stages); Atlantic specimens; figures 1-51 in Tesmer and Broad (1964). Stages I-V, from plankton; Pacific specimens; figure 20 in Makarov (1967). Discrepancies among descriptions may result, at least in part, from geographical variations in mor- phology. The following synopsis is based on speci- mens from off Hokkaido, Japan (Kurata 1964b). In all described stages, shallow transverse groove in carapace; dorsal spine on abdominal somite 3, posterolateral spines on abdominal somite 5. Exo- podite only on pereopod 1. In Stage I, antennal scale with 11 setae, including 2 subterminal setae along outer margin; endopodites of maxillipeds 1-3 with 4 segments; exopodites of maxillipeds jointed; fifth pair of telsonic spines distinctly shorter than fourth or sixth pairs. Length: Stages I-V, 1.9-5.0 mm. Range: Prince Edward Island, Canada (Needier 1941); Beaufort, NC (Tesmer and Broad 1964); an estuarine, subarctic boreal species, Sea of Okhotsk (Makarov 1967); Hokkaido, Japan (Kurata 1964b); depth, 0-90 m, rarely to 440 m (Holthuis 1980). Genus Mesocrangon Zarenkov Largest larvae of Crangonidae with unabbreviated development. From Stage III on, posterior margin of telson straight or slightly concave. Mesocrangon intermedia (Stimpson) Five larval stages. Stages I-V, from plankton; described as Sclero- crangon intermedia by Makarov (1967); figure 24 in Makarov (1967). In Stage I, antennal flagellum about half as long as antennal scale; antennal scale with 11 setae; abdo- minal somites 2-3 with keels; abdomen apparently without spinules. Length: Stages I-V, 4.5-9.0 mm. Range: Sea of Okhotsk to St. Lawrence Island (Ber- ing Sea); depth, 18-180 m (Makarov 1967). Genus Paracrangon Dana In all described stages, rostrum long, spiniform, spinulose, directed upwards about 45°; carapace with denticulate anteroventral margin; basipodite of maxilliped without subterminal seta; exopodites on pereopods 1 and 2; exopodites of pereopods 1 and 2 with ^ 6 natatory setae; protopodite of antenna with 2 spines- one at base of flagellum, other a long spine at base of scale. Paracrangon echinata*^ Dana At least 7 larval stages. "A diagnostic character of adult Paracrangon is the absence of pereopods 2 (Rathbun 1904). Kurata's (19fi4b) description of Pnni- rrangon erhiriala shows pereopixi 2 fully (levelof)ed as late tLs the seventh larval stage. Either Makarov's(19t>7) identification of these larvae as P. fchmata is incorrect, or P. echimitu must have at least '.i or 4 more larval stages before pereopod 2 becomes reduced or absent. 282 HAYNES:PANI)A1,11)AK. HIl'I'OI.VTIDAE.ANDCRANCONIDAKLAKVAE Stapes II, IV-VII, from plankton; tentatively iden- tified as GlyphocrangoH sp. by Kuratii (19641)); figures 103-130 in Kurata (1964b). Stages II and IV; Makarov (1967) based identity on known distribution of adults and morphology of embryo of Glyphocrangon granulofsLs (see Bate 1888); figure 28 in Makarov (1967). Most spinose of crangonid larvae known from northern North Pacific Ocean (Fig. 10). Length: Stage II, 5 mm; Stage VII, 13.8 mm. Range: Sea of Japan; Okhotsk Sea; Port Etches, AK, to La Jolla, CA; depth, 7-201 m (Butler 1980). Genus Sabinea J. C. Ross Probably 4 larval stages. In all described stages, telson relatively wide, with shallow indentation; pereopods 2-5 without exopodites. Sahinea septemcarinata (Sabine) Probably 4 larval stages. Prezoeal telson and Stages I and III, probably known parentage; Atlantic specimens: plate V, figures 1-23, and plate VI, figures 1-13, in Sars (1890). Stages I and III, probably known parentage; de- scribed as Crangon septemcarinatus by William- son (1915); figures 167-172 in Williamson (1915). Williamson's figures from Sars (1890). Stages I and III, from plankton; figure 12, "last stage" (= Stage III), in Birshteyn (1938). Stage I (whole larva) and Stage III (telson); origin of specimens not given; figures 1 and 2 in Williamson (1960). Stages and origin of specimens not given, de- scribed as Myto gaimnrdi by Birshteyn (1938); plate 7, figure 1 (Kr0yer 1846 in Birshteyn 1938). In all described stages, anteroventral margin of carapace with about 7 denticles. Abdominal somites 1-4 with 1, 2, 1, and 1 ventral spines, respectively; abdominal somite 5 with posterolateral spine. Telson with 16-1-16 setae in Stage 1, 13 -i- 13 setae in Stage III. Length: Stages I-III, 7.7-11.5 mm. Range area of Iceland and Faroe Islands (Williamson 1960) Barents and Norwegian Seas (Williamson 1960) Chukchi Sea (Birshteyn 1938); eastern coast of North America from mouth of St. Lawrence River to Massachusetts Bay; Arctic Ocean to Point Barrow (Alaska), White Sea, and northern Europe (Williams 1974); depth, 10-240 m (Williams 1974). 0. 5 mm Figure 10. -Dorsal view of body, Stage I zoea of Para era ngon echinata. Genus Sclerocrangon G. O. Sars Not more than 2 larval stages. Appendages resem- ble adult except uropods enclosed and pleopods not fully setose. Pereopods without exopodites; pereopods 4 and 5 have characteristic sickle-shaped dactyli(Fig. 11). Sclerocrangon horeas (Phipps) Two larval stages. Late embryo extracted from egg; plate VI, figures 14-28, in Sars (1890, as cited in Williamson 1960). Stage not specified, known parentage (Makarov 1967). No figures. Stage I and Stage II ("postlarval"), known parent- age; figures 1-3 in Makarov (1968). 283 FISHERY BULLETIN: VOL. 83. NO. 3 1 , 0 mm Figure ll.-Pereopod 4 or 5 of Sclerocrangan larva with characteristic sickle-shaped daot\'l. Larvae not free living but cling to pleopods of female. In all described stages, carapace without anteroventral denticles; flagellum of antennule segmented; basal portion of antennule shaped as in adult; maxillipeds with undeveloped exopodites; exo- y)odites with a few feeble natatory setae; pereopods segmented, without exopodites. In Stage I, rostrum absent; carapace covers sessile eyes; telson and abdo- minal somite 6 jointed. In Stage II, rostrum short, triangular, flattened dorsoventrally. Length: Stage I, 9.0 mm; Stage II, 11.5 mm. Range: Sea of Japan to Chukchi Sea to Bare Island, WA; arctic Canada to Cape Cod, MA; North Atlantic Ocean and arctic Europe; Spitsbergen to Faroe Islands; depth, 0-366 m (Butler 1980). Sclerocrangon salebrosa (Owen) One larval stage. Stage I, from plankton; figure 27 in Makarov (1967). Embryos, from female; Stage I and Stage II ("post- larval"), from plankton; figures 1 and 2 in Makarov (1968). Larvae from plankton identi- fied by comparison with embryos dissected from eggs. Larvae free living. In Stage I, rostrum spiniform; carapace with anteroventral denticles; exopodites of maxillipeds fully developed, each exopodite with 5 natatory setae; telson exceptionally wide with 22 -t- 22 setae. Length: Stage I, 10.3-10.5 mm. Range: Okhotsk Sea; Hokkaido; no depth range (Kurata 1964b; Makarov 1967). Sclerocrangon zenkevitch't Birstein and Vinogradov Only Stage I described, known parentage; figure 5 in Birshteyn and Vinogradov (1953). Carapace nearly circular laterally, without ros- trum; cephalothorax and abdomen without spines or denticles; telson ovoid. Length: Stage I, 7.2 mm. Range: Bering Sea; depth, 2,995-3,940 m (Birshteyn and Vinogradov 1953). BIOLOGY Although pandalid, hippolytid, and crangonid lar- vae are common inhabitants of the neritic mero plankton of temperate and arctic waters of the North Pacific Ocean, only a few studies on their biology have been published. The most complete studies are those of Haynes(1983), Makarov (1967), and Rothlis- berg (1975). Haynes described the relative abun- dance and distribution of pandalid shrimp larvae in the lower Cook Inlet-Kachemak Bay area, Alaska; Makarov (1967) described the distribution of decapod shrimp larvae of the West Kamchatkan shelf; and Rothlisberg (1975) discussed larval ecology oiPan- dalus jordani off the Oregon coast. In this section, I review the findings of these authors and supplement their findings with information from the literature. To avoid redundancy of citation, only information in addition to that given by Haynes, Makarov, and Rothlisberg is cited by author and date. This section does not include every known facet of the biolog>' of decapod shrimp larvae of the northern North Pacific Ocean; however, more information can be acquired from the papers of Haynes, Makarov, and Rothlis- berg and from their bibliographies. Areas of high abundance of Stage I larvae ap- parently indicate areas where females are releasing larvae. For example, in Kachemak Bay in 1972, Stage I larvae of Pandalus borealis, P. goniunts, P. hypsinotixs, and Pandaiopsis dispar were most abun- dant in plankton samples collected in the same area 284 IIAVNKS. I'ANDAMDAl-:. llll'l'OLVTIDAK. AN1)CKAN(;()NIIIAK I.AKVAK where females were releasing larvae (for these 4 species, females were releasing larvae at depths of about 85, 35, 50, and 100 m, respectively, based on a trawling survey). Time of release of pandalid larvae varies with species. In Kachemak Bay in 1972, Stage I larvae of Panda Ins boreal i-x were not caught until the first half of April; Stage I larvae of P. goniurus and P. hyp- sinotus were caught later, in the latter half of April. In British Columbia waters, P. borealis larvae are also released earlier than larvae of either P. goni- urus or P. hypsinotus (Berkeley 1930; Butler 1964). Time of lar\'al release is also related to water temperature. F^or example, a residual layer of relatively cold (sometimes subzero) water remains on the central West Kamchatkan shelf at a depth of 50-150 m throughout the summer. Decapods living in this layer of cold water release their larvae later than decapods living in warmer waters to the north and south. In the western North Atlantic Ocean, pan- dalid shrimp also release their larvae later in colder waters than in warmer waters (Haynes and Wigley 1969). Depth distributions of larvae of P. borealis and P. goniurus in Kachemak Bay, 1972, were usually similar. Few larvae were in the 0-10 m stratum; most were between about 10 and 40 m. The abundance of larvae remained relatively constant below about 50 m. Numbers of Stage I P. borealis larvae, however, increased below about 70 m, possibly reflecting their recent release. These depth distributions differ from the depth distribution of P. jordani off the Oregon coast. Younger (Stages V-X) P. jordani larvae were found closer to the surface (0-10 m stratum) than older (Stages XI-XV) larvae (to 160 m). Water temperature has profound effects on larval survival, growth, and size at metamorphosis. For ex- ample, survival of P. jordani larvae (Stages I-III) is markedly less at 17° than at 5°C. For the oldest stage (Stages IX-XIII), the relation between survival and temperature is reversed, and survival is lowest at 5°C (Rothlisberg 1979). For larvae of P. platy- ceros, survival is reduced by sudden changes in temperature, particularly about 20 °C and below 9°C (Wickins 1972). At a given temperature (range 5°-14°C), growth increments for all larval stages of P. jordani decrease with increasing size; however, the higher the temperature, the more rapid the molting frequency (Rothlisberg 1979). Shrimp larvae can probably influence the direction and extent of their dispersal. For instance, in Kache- mak Bay, 1972 and 1976, pandalid shrimp larvae were released in the central portion of the outer bay. Some of these larvae were carried northward out of the bay in the direction of the current, but others were dispersed southwestward in a direction op- posite the current. In the southern area of the western Kamchatka shelf, Crangon larvae released close to shore with larvae of other species, such as king crab, Paralithodes camtschatica, remained close to shore. Larvae of the other species, however, were carried seaward. In the northern area of the western Kamchatka shelf, where currents are faster than in the southern area, Crangon larvae were carried seaward (Makarov 1967). The causes for dispersal of larvae against known water currents are unknown, but dispersal may be dependent, at least in part, on the swimming capability of the larvae. Some pandalid shrimp larvae migrate vertically in a diel cycle. In Kachemak Bay in 1972, Stages I and II larvae of P. borealis and P. goniurus were most abundant between the surface and 15 m during low light levels (1800-0800 h); however, during high light levels (1000 and 1600 h), they were most abundant between 30 and 60 m. Although present, a pro- nounced thermocline did not prevent larvae from moving vertically. Whether later stages of P. borealis and P. goniurus migrate similarly is unknown; however, in waters off Oregon, only Stages XII-XVI larvae of P. jordani migrate ver- tically in a diel cycle. During the day, these P. jor- dani larvae are distributed from the surface to 1 50 m by age: the deeper the water, the older the lar\'ae. At night, P. jordani larvae migrate upwards in the water column, and the stages remain somewhat uniformly distributed with depth. Foods of pandalid larvae have been determined during attempts to rear the larvae in the laboratory and from examination of shrimp stomachs. Larvae of P. jordani and P. platyceros have been reared on brine shrimp, A rtemia salina nauplii (Modin and Cox 1967; Lee 1969; Price and Chew 1972), P. hypsino- tus larvae have been reared on brine shrimp nauplii and algae (Haynes 1976), and P. kessleri larvae have been reared on small pieces of crab, shrimp, and mussel tissue (Kurata 1955). In 1976, I made a preliminary study (unpublished) on foods eaten by pandalid shrimp larvae in Kachemak Bay by examin- ing their gut contents. The larvae mostly ate diatoms, especially Coscinodisais types, and larval crustaceans. Many of the guts also contained black pigment and ommatidia. The assumption that pan- dalid larvae feed on eyes of other decapod larvae was subsequently confirmed when I observed a P. borealis zoea ingesting the eye of a live king crab zoea. Calcareous fragments (probably molluscs), coccolithophores, spines of larval echinoderms, and 285 FISHERY BULLETIN: VOL. 83, NO. :i bits of diatoms have been found in guts of Pandalus larvae from European waters (Lebour 1922). Some species of pandalid larvae are sustained by their internal yolk for several days after hatching without feeding; others must feed immediately after hatching or die. Pandalus platyceros larvae can live 11-13 d on stored yolk with no food (Price and Chew 1972); however, when food is offered, they feed immediately after hatching. In Price and Chew's (1972) study, the starved larvae ate their dead rela- tives, but did not actively prey on them. Larvae of P. jordani, however, if not fed soon after hatching, starve and die regardless of later increases in prey concentrations (Modin and Cox 1967). In another study, 40% of Stage I zoeae of P. borealis without prey died in 5 d, and 100% died in 13 d (Paul et al. 1979). The relationship between food and survival in cap- tivity has been determined for some North Pacific Ocean larvae; however, little is known about this relationship in nature. For instance, year-class strength may be influenced or even largely deter- mined by the quality and quantity of food available during the larval period. Unfortunately, there is vir- tually no information on the t\T3es and quantities of food needed for survival of shrimp larvae in nature. Until this information is available, the relation be- tween food and survival of shrimp larvae at sea will remain unknown. ACKNOWLEDGMENTS Donald I. Williamson of the University of Liver- pool, England, and T. H. Butler of the Pacific Biological Station, Nanaimo, British Columbia, read an earlier version of this paper and offered valuable suggestions for improvement. LITERATURED CITED Bate, C. S. 1888. Report on the Crustacea Marrura dredged by H.M.S. Challenger during the years 1873-1876. Rep. Sci. Results Voyage H.M.S. Challenger, Zool. 24, 942 p. Berkeley, A. A. 1 930. The post-embryonic development of the common panda- lids of British Columbia. Contrib. Can. Biol. Fish., New Ser. 6(6):79-163. BlR-SHTEYN, YA. A. 1938. Zemetka o Decapoda iz Planktona, Sobrannogo vo Vremya Plavaniya Ledokolov Sadko i Litke v Arktike. (Decapoda from plankton collected during voyages of the ice- breakers Sadko and Litke in the Arctic.) |In Russ.J Byull. M. 0-Va Isp. Frirody Otd. Biol. 47(3): 199-209. (Translated by A. L. Peabody, agency unknown, available from author, 15 p.) BiRSHTEYN, Ya. A. AND L. G. VINOGRADOV. 1953. Novyye Dannyye po Faune Desyatinogikh Rakoobraz- 286 nykh (Decapoda) Beringova Morya. [New data on the Deca- poda of the Bering Sea.] [In Russ., Engl. Summ.) Zool. Zh. 32:215-228. (Translated by Lang. Serv. Div., Natl. Mar. Fish. Serv., NOAA. Wash., D.C., 20 p.) Bull, H. 0. 1939 The newly-hatched larva of Spirnvtocarif: iniswla (Kroyer). Rep. Dove Mar. Lab. (3). 6:43-44. Butler. T. H. 1964. Growth, reproduction, and distribution of pandalid shrimps in British Columbia. J. Fish. Res. Board Can. 21: 1403-1452. 1980. Shrimps of the Pacific Coast of Canada. Can. Bull. Fish. Aquat. Sci. 202, 280 p. Frost, N. 1936. Decapod larvae from Newfoundland waters. Div. Fish. Res., Newfoundland, Rep. Faun. Ser. 1:11-24. GURNEY, R. 1942. Larvae of decapod Crustacea. Ray Soc. (Lond.) Publ. 129, 306 p. Hayashi, K.-L 1977. Studies on the hippolytid shrimps from Japan- VI. The genus Spirontocaris BATE. J. Shimonoseki Univ. Fish. 25: 155-186. Haynes, E. 1976. Description of zoeae of coonstripe shrimp, Fnndalns hypainotus, reared in the laboratory. Fish. Bull., I'.S. 74: 323-342. 1978a. Description of larvae of the humpy shrimp, PaxdaluK goniurus, reared in situ in Kachemak Bay. Alaska. Fish. Bull., U.S. 76:235-248. 1978b. Description of larvae of a hippolytid shrimp, Lrbbms groevlnndicus, reared in situ in Kachemak Bay. Alaska. Fish. Bull.. U.S. 76:457-465. 1979. Description of larvae of the northern shrimp, Pandalu-n borealis, reared in situ in Kachemak Bay, Alaska. Fish. Bull.. U.S. 77:157-173. 1980a. Larval morphology of Pnndalu,^ trideri.f and a sum- mary of the principal morphological characteristics of North Pacific pandalid shrimp larvae. Fish. Bull., U.S. 77:625-640. 1980b. Stage I zoeae of a crangonid shrimp, Crangon fnni- riscorum anguj-ttimann, hatched from ox'igerous females collected in Kachemak Bay, Ala.ska. Fish. Bull., U.S. 77: 991-995. 1981. Early zoeal stages ai Lehhe^is polaris, Eunhts surklei/i. E. fahricii, Spirontocaris arruatn. S. (trhotensis. and Heptn- carpus camtschaticus (Crustacea, Decapoda, Caridea, Hippo lytidae) and morphological characterization of zoeae of Spirontocaris and related genera. Fish. Bull., U.S. 79: 421-440. 1983. Distribution and abundance of larvae of king crab, Paralithodes camtschatica, and pandalid shrimp in Kachemak Bay, Alaska, 1972 and 1976. U.S. Dep. Commer., NOAA Tech. Rep. NMFS SSRF-765. 64 p. In press. Description of early stage zoeae of Spirontocaiis wurdochi (Decapoda, Hippolytidae) reared in the laboratory. Fish. Bull., U.S. 82:523-527. Haynes, E. B., and R. L. Wigley. 1969. Biology of the northern shrimp, Pandalus borealis, in the (kilf of Maine. Trans. Am. Fish. Soc. 98:60-76. HOLTHUIS, L. B. 1947. The Decapoda of the Siboga Expedition. Part IX. The Hippolytidae and Rhynchcx;inetidae collected by the Siboga and Snellius Expeditions with remarks on other species. Siboga Exped. 140, Monogr. 39a'*, 100 p. 1976. The identities of Pandalus gracilis Slimpson, 1860, and Pandalus prensor Stimpson, 1860 (Decapoda, Pandalidae). « HAVNLS: I'ANDALIDAE. Hll'l'OI.Y'l'lliAK. ANDCKANCOMDAK l.AKVAK Cnistaceaiia .S():4[491-r)4 . 1980. Shrimps and prawns of thf world, an annotated cata- lo^e of species of interest to fisheries. Vol. 1 , FAO species cataloKiio. FAO Fish. Syn. 125, FIR/SI 25 Vol. 1, 271 p. Internationai, Commission on Zooiak;ic:ai, Nomknclaturk. 1964. International ccxle of zoological nomenclature, adopted at the XV International Congress of Zoology. Int. Tnist Zool. Nomen., Lond., 176 p. IVANOV, B. G. 196,5. A description of the first larvae of the far-eastern shrimp (Piiiid(du,s i^otiiurn-t). [In Riiss., Kngl. summ.| Zool. Zh. 44:1255-1257. (Translated by U.S. Dep. Commer., NOAA. Natl. Mar. Fish. Serv., Off Int. Fish. Aff., Code No. F44.) 1968. Lichinki Nekotorykh Dal'nevstochnykh Krevetok Semeystva Crangonidae (Crustacea, Decajioda). (Larvae of certain far eastern shrimps of the family Crangonidae (Crus- tacea, Decapoda).] |In Russ., Engl, summ.] Zool. Zh. 47: 534-540. (Translated by A.L. Peabody, agency unknown, available from author, 8 p.) 1971. The larvae of some eastern shrimps in relation to their taxonomic status. |In Russ., Engl. summ.]. Zool. Zh. 50: 657-665. (Translated by U.S. Dep. Commer., NOAA, Natl. Mar. Fish. Serv., Off Int. Fish. Aff, Code No. F44.) KOBJAKOVA, Z. I. 1937. Systematisch Uebersicht der Dekapoden aus dem Ocho- tskischen und Japanischen Meere. Uchenie Zapiski Leningr. Univ.. (15). (). 93-154. [Not seen by author, cited in Hayashi 1977.1 KR0YER, H. 1846. Karcinologiske Bidrag. Naturhist. Tidsskrift. (Kroyer et Schiodte), Ser. 2, Bd. 1. [Not seen by author, cited in Birshteyn 1938.| 1861. Pp. 63-69 in Naturhist. Tidsskrift., Ser. 3, Vol. 1., Copenhagen. [Not seen by author, cited in Stephensen 1935.) Kl'RATA, H. 1955. The post-embryonic development of the prawn, Pnnda- liix kesslen. Bull. Hokkaido Reg. Fish. Res. Lab. 12:1-15. 1964a. Larvae of decapod Crustacea of Hokkaido. 3. Panda- lidae. Bull. Hokkaido Reg. Fish. Res. Lab. 28:23-34. (Translated by Fish. Res. Board Can., 1966, Transl. 693.) 1964b. Larvae of decapod Crustacea of Hokkaido. 4. Crango nidae and Glyphocrangonidae. Bull. Hokkaido Reg. Fish. Res. Lab. 28:35-50. Ki!Ris, A. M., AND J. T. Carlton. 1977. Description of a new species, Crarigor^ hatidi. and new genus Lissorrnruion, of crangonid shrimps (Crustacea: Caridea) from the California coast, with notes on adaptation in body shape and coloration. Biol. Bull. (Woods Hole) 153: 540-559. Lebohr, M. V. 1922. The food of planktonic organisms. J. Mar. Biol. Assoc. U. K. 12:644-677. 1930. The larval stages of Caridimi, with a description of a new species. C. utevetii. Proc. Zool. Soc. Lond., p. 181-194. 1931 . The larvae of the Plymouth Caridea. I. The larvae of the Cragonidae. II. The larvae of the Hippolytidae. Proc. Zool. Soc. Lond., [). 1-9. 1936. 6. Notes on the Plymouth species of Spircmtocaris (Crustacea). Proc. Zool. Soc. Lond., p. 89-104. 1937. The newly hatched larva of Spirontocari-t; spinus (Sowerby) var. lilljeborgi Danielssen. J. Mar. Biol. Assoc. U. K. 22:101-104. 1940. The larvae of the British sf)ecies oi Spirontocarix and their relation to Thor (Crustacea Decapoda). J. Mar. Biol. Assoc. U. K. 24:50.5-514. Lee, Y. J. 1969. Larval development nl. 80:1-94. Tesmer. C. a., and a. C. Broad. 1964. The larval development of Crangon septemspinosa (Say). Ohio J. Sci. 64:239-250. WlCKINS, J. F. 1972. Experiments on the culture of the spot prawn Pandahis platyreros Brandt and the giant freshwater prawn Mncro- brackium rosenbergii (de Man). Fish. Invest., Minist. Agric. Fish. Food (G.B.) Ser. 2, 27(5), 23 p. Williams, A. B. 1974. Marine flora and fauna of the northeastern United States. Crustacea: Decapcxia. U.S. Dep. Commer., NOAA Tech. Rep. NMFS Circ.-389, 50 p. Williamson, D. I. 1957a. Crustacea, Decapoda: Larvae V. Caridea, Family Hippolytidae. Fiches Identif Zooplancton 68. 5 p. 1957b. Crustacea. Decapixia: Larvae I. tleneral. Fiches Identif. Zooplancton 67, 7 p. 1960. Crustacea, Decapoda: Larvae VII. Caridea. Family Crangonidae, Stenopodidea. F"iches Identif Zooplancton 90, 5 p. 1969. Names of larvae in the Decapoda and Euphausiacea. Crustaceana 16:210-213. 1982. Larval morphology and diversity. In L. (i. Abel (editor). The biology of Cnistacea. Vol. 2. Embryology, niorpholog>', and genetics. |i. 4.'M10. Acad. Press. N.V. Williamson, H. C. 1915. Decapoden. I. Teil (Larven). Nordisches plankton IS: 315-588. (Also published in Nordisches plankton, Zoolo- gischer teil, Dritter Band: Crustacea, p. 315-588. Verlag von Lipsius and Tischer, Lepzig 1927, and by Neudruck A. Asher Co., Amsterdam 1964.) Zarenkov, N. a. 1965. Revision of genera Cramjon fiihrinns and Sclrnirnn/- gon G. (). Sars (Decapoda. Crustacea). |In Russ.. Engl. summ.] Zool. Zh. 44:1761-1775. (Translated from Russian. Fish. Res. Board Can. Transl. Ser. 1465. 43 p.) 288 WITHIN-SEASON DIFFERENCES IN GROWTH OF LARVAL ATLANTIC HERRING, CLUPEA HARENGUS HARENGUS Cynthia Jones' ABSTRACT Data (il)laitic'il from two previous studies of larval Atlantic herrinj^ Kff'Wth were compared, based on otolith intremeiU estimated age. These data, from the Gulf of Maine in 1976-77 and 1978-79, supported the hypothesis that larvae hatched early in the spawning season grew faster than larvae hatched late. Differ- ences were siffnificant under assumptions that increments were deposited in the otolith either daily or at 0.5 increments per day. Corroborative evidence indicated that otolith increments were formed daily at least dur- in.u; the early part of the spawning season. The otolith increment technique has been used to estimate age and growth in field-caught larval Atlan- tic herring, Clupea harengus harengus, in the Gulf of Maine by Townsend and Graham (1981) and by Lough et al. (1982). Use of the increment technique to estimate age usually assumes daily deposition of otolith increments. Uncertainty exists, however, regarding increment deposition rates in the otoliths of larval herring. Gjosaeter (1981) and 0iestad (1982) have observed daily increment deposition. In contrast, Geffen (1982) found that increment deposi- tion can be variable and a function of growth rate in larval herring, underscoring the problem in simply assuming that increments occur daily under field conditions. Growth calculations based on assump- tions of daily increment deposition in populations that experience variable increment deposition rate would result in inaccurate estimates of growth rates. In most cases where otolith increment deposition has been tested under suboptimal conditions, the deposi- tion rate has been found to be nondaily (for review see Jones 1984). Estimates of growth rates can be made, however, by expressing growth based on increment counts and with the use of corroborative evidence to determine periodicity of increment deposition. Das (1968) found that growth rates of larval Atlan- tic herring, measured by following the progression of length modes over time, were different within a spawning season. He stated that early-spawned lar- vae grew faster than late-spawned larvae and model- ed growth with curvilinear functions. Townsend and Graham (1981) also reported two different growth 'Graduate School of Oceanography, University of Rhode Island, Kingston. RI 02882-1197; present address: Department of Natural Resources, Fernow Hall, Cornell University, Ithaca, NY 14853. rates for Atlantic herring, one for larvae born prior to November 5 and one for larvae born later. Each group was modeled by two regression lines to emphasize that growth ceased in January and resum- ed in February. In their study, early- and late- hatched groups were analyzed separately and the comparison of growth between larvae hatched early versus late in the season was not statistically verified. This paper uses otolith increment data from Town- send and Graham (1981) and from Lough et al. (1982) to compare early-season versus late-season larval Atlantic herring growth. The comparisons are made using the assumptions of both daily and non- daily otolith increment deposition. METHODS Raw data for otolith counts and larval fish lengths used in these studies were obtained from Gregory Lough of the National Marine Fisheries Service, Northeast Fisheries Center, Woods Hole, MA, and from Joseph Graham and David Townsend of the Maine Department of Marine Resources, Boothbay Harbor, ME. Both data sets were used in the detec- tion of within-season differences in growth rates. Although the study of Lough et al. (1982) encom- passed a larger area, only data from the Gulf of Maine were included in the analysis (Table 1), in order that comparisons were made within the same area as for Townsend and Graham (1981). Methods employed for the collection of data were reported by Lough et al. (1982) and by Lough and Bolz (1979) for the 1976-77 data and by Townsend and Graham (1981) for the 1978-79 data. For each season (1976-77, 1978-79), data were analyzed in three ways: Manuscript accepted October 1984. FISHERY BULLETIN: VOL. 83. No. 3, 1985. 289 FISHERY BULLETIN: VOL. 83. NO. 3 TABLE 1. — Station information for Atlantic fierrlng samples from tfie Gulf of Maine area for the fall and winter of 1976-77 sampling program. (Data from Lough et al. 1982.) Vessel Cruise No. Stn. Time Lat. Long. (Night N W Date or Day) Annandale 76-01 Researcher 76-01 Mt. Mitchell 77-01 38 43°37' 69°22' 8 Oct. 0300 (N) 44 43°44' 68°50' 8 Oct. 1415(D) 59 44°25' 67°35' 9 Oct. 1515(D) 65 44°36' 67°07' 13 Oct. 0330 (N) 102 42°58' 70°00' 8 Dec. 1030 (N) 105 43°30' 69°30' 9 Dec. 1100 (N) 122 43°14' 70°01' 24 Feb. 1620(D) 123 43°00' 70°15' 24 Feb. 1933(D) 1) Hatch date was calculated on the assumption of daily increment deposition, and all data were considered. 2) Hatch date was calculated on the assumption of daily increment deposition only with larvae which had 60 or fewer increments included for analysis. This was done to determine whether growth differences were present in the earlier months of life. Also, since the range of incre- ment counts for the late-hatched larvae from 1976 to 1977 was greater than for early-hatch- ed larvae, use of a truncated data set resulted in more valid comparisons. 3) Hatch date was calculated on the assumption of nondaily deposition (0.5 increment/d). Date of hatching was calculated by subtracting the estimated age of each larva from its date of capture. This calculation, of course, depends on how age was estimated. According to the Lough et al. (1982) cal- culation, a larva with 10 otolith increments would be 29 d old: 22 d for the first 3 increments, plus 7 d to lay down the next 7 increments. According to the assumptions used by Townsend and Graham (1981), a larva with 10 otolith increments would be 15 d old, assuming that increment deposition began 5 d after hatch, and was daily thereafter. There is a difference of 14 d between these two estimates of age, and, therefore, estimated day of hatch. This does not af- fect the regression analysis, as long as the indepen- dent variable used is increment count, not age. The range of possible hatch dates for each in- dividual was also calculated, based on the considera- tion that deposition rates could vary from 0.5 to 1.0 increment/d (after Geffen 1982). Age could be equal to the number of increments plus a constant (5 d) or up to twice the number of increments plus a constant (5d). Larvae were classified as either early- or late- hatched within the spawning season. For 1976-77 the early-late division date was placed at the discon- tinuity in the frequency of hatching plot, which also occurred at the midpoint in the spawning cycle. Divi- sion date for the 1978-79 data set was placed at approximately the division of Townsend and Graham (1981) which they felt represented two different groups of larvae. For analysis of nondaily deposition, the data were partitioned to insure that there could be no overlap of early- and late-hatched classification of larvae, assuming deposition ranged from daily deposition to deposition of one increment every 2 d. Any late- hatched larva whose possible range of hatch dates overlapped the division date (for early-hatched vs. late-hatched classification) was eliminated from analysis. This resulted in a loss of data (e.g., the fish whose possible hatch date overlapped the division date) and decreased the ability to detect differences. Ordinary least squares linear regressions were fit to each data set. Bartlett's test for homogeneity of variance (Ostle and Mensing 1975) was applied to the data before each analysis. After regressions were fit, the residuals of length were plotted against pre- dicted length and examined for trends (Draper and Smith 1981). F-tests (Ostle and Mensing 1975) were applied to paired linear regressions, early-hatched versus late-hatched, to determine whether the slopes were significantly different. This test showed whether the data were better fit by two lines, one for early-hatched and one for late-hatched larvae, or whether a single regression line was preferable. In the regression plots the change in length is express- ed in millimeters per increment. The von Bertalanffy growth equation. L, = L^(1 ■k{t-l 0)) was also fitted to the data, using the nonlinear regression procedure (NLIN) within SAS (Statistical Analysis Systems, SAS Institute, Gary, NC). Esti- mates of the parameters {K, L^, ^,) of the von Ber- talanffy equations for early- and late-hatched larvae were compared with a Fisher-Behrens test (Hoenig 1982) to determine whether the vector of parameter estimates from the two classifications was signifi- cantly different. RESULTS Linear regression models fitted to larval length-at- increment count data showed significant differences between larvae hatched early and late in the spawn- ing season. Larvae hatched early had achieved greater length at a given increment count than those 290 JONES: niKFKKKNCKS IN I,AK\AI, IIKKKINC CKOW'I'II hatched later. Intercepts were not compared since the data sets did not contain any larvae with fewer than seven increments and inferences outside the range of the data should not be drawn. 1976-77 Study A frequency plot of hatching dates for the Gulf of Maine stations is shown in Figure 1 for age esti- mated on the assumption of daily ring deposition and in Figure 2 for age estimated on the assumption that deposition was daily or as little as one ring every other day. Differences in length-at-increment count between early- and late-hatched larvae was striking (Table 2). Regression plots are shown in Figure 3. Analysis of the data confirmed that the length-at-count data were modeled more accurately by two different regression lines (P < 0.01) and that the slopes of these two regressions were significantly different (P LARVAL HERRING 1976-1977 STUDY 15-1 UJ < > % 10 GC UJ m z o^ SEPT 1 1 SEPT 1 "^ OCT 1 NOV OCT NOV 1 DEC 1 DATE OF HATCHING Figure 1 . - Frequency of Atlantic herring hatching during the 1976-77 study. Upper scale gives the day of hatch based on the Lough et al. (1982) aging method, or, as discussed in the text. Lower scale gives the day of hatch based on Townsend and Graham's (1981 ) aging method as discussed in the te.xt. Arrow indicates division point between early- and late-hatched classification. Table 2. — Regression analysis of 1976-77 Gulf of Maine Atlantic herring data. et al. 1982.) (Data from Lough Otolith Hatch Slope Standard Probability Probability increment classifi- Sample regression error intercepts slopes count cation size Intercept line of slope R' equal equal All data Early 117 9.4 0.3284 0.0172 0.76 <0.01 <0.01 Late 64 15.8 0.0948 0.0047 0.87 60 or fewer Early 117 9.4 0.3284 0.0172 0.76 <0.01 <0.01 Late 44 14.6 0.1470 0.0274 0.41 Data were classified Into early- and late-hatched larvae. These two groups were compared by fitting ordinary least squares regression lines to 1) all the data within the two classifications, and 2) using only lengths from larvae with 60 or fewer increment counts. Slopes and intercepts were compared between early versus late for each group. 291 FISHERY BfLLETIN: VOL. 83. NO. 3 LARVAL HERRING 1976- 1977 15-1 LU < > DC 10 < cc LU CO INCREMENT EVERY 2 DAYS o 0 DAILY DEPOSITION \ I \ \ r MAY 1 JUN 1 JUL 1 AUG 1 SEP OCT 1 I \ \ \ f- MAY 1 JUN I JUL 1 AUG 1 SEP DATE OF HATCHING NOV OCT 1 NOV Figure 2. -Frequency of Atlantic herring hatching during the 1976-77 study, calculated under two assumptions of otolith increment count and age relationship. < 0.01). The average length achieved per increment was 0.33 mm and 0.10 mm for early- and late-hatch- ed larvae, respectively. Bartlett's test confirmed that variances were homogeneous. Analysis of residuals showed that the last three residuals, corresponding to the three largest larvae, were below the average. The exclusion of these points did not alter the results of the analysis. Analysis of the subset of larvae with fewer than 60 increments (Table 2) showed that data were better fitted with two different regression lines {P < 0.01) and that the slopes were significantly different {P < 0.01). Regression plots are shown in Figure 4. Change in length of early-hatched larvae was 0.33 mm/increment and 0.15 mm/increment for late- hatched larvae. Bartlett's test showed variances to be homogeneous and residuals showed no trends, ex- cept for the two youngest late-hatched larvae which fell below the regression line. Late-hatched larvae were slightly larger than early-hatched larvae for the lowest increment counts. Differences in length-at-increment count were apparent for data whose calculated hatch dates in- cluded deposition rates of from 0.5 to 1.0 incre- ments/d (Table 3). The change in length of early- hatched larvae was 0.33 mm/increment compared with 0.17 mm/increment for late-hatched larvae. Data were again better fit with two regression lines (P < 0.01) whose slopes were significantly different (P < 0.01). The von Bertalanffy growth equation fit the late- hatch larval data well (Table 4). L^ was estimated at 29.81 mm, with a 95% confidence interval of 26.41 to 33.22 mm. Fit to the early-hatched larval data was poor. L^ was estimated at 35.59 mm, with a con- fidence interval of 17.76 to 53.41 mm. These data were adequately fit with a straight line, and there is little justification for fitting with a curvilinear func- tion other than it has been traditionally used for adult fish. Beverton and Holt (1954), however, stated that the von Bertalanffy equation should not necessarily be used during the early life stages. Nevertheless, when the parameter estimates from the two curves were compared, they were signifi- FlGURE 4. -Regression plot of length-at-otolith increment count for Atlantic herring. Only lengths for larvae with 60 or fewer otolith increments have been included for analysis. Data from Lough et al. (1982). 292 JONES: DIKKKRENCES IN LARVAL HKKKlNCCKOWI'll 30 25 E E X I- 20H o z UJ 15- 10- LARVAL HERRING GROWTH 1976-1977 STUDY l^m.--^ A 1 AA A A A A A TIME OF HATCH EARLY • LATE A 0 20 40 60 80 100 INCREMENT COUNTS -1 \ I \ 1 120 140 Figure 3. - Regression plot of length-at-otolith increment count for Atlantic herring. Complete data set represented. Data from Lough et al. (1982). 25-1 E 20 - E I I- O z LU -I 15H 10^ LARVAL HERRING GROWTH 1976-1977 STUDY TIME OF HATCH EARLY • LATE A 10 n \ \ r 20 25 30 35 INCREMENT COUNTS 40 45 50 293 FISHERY BULLETIN: VOL. 83, NO. 3 Table 3. — Regression analysis of 1976-77 Gulf of Maine Atlantic herring data based on two otolith increment deposition assumptions. (Data from Lough et al. 1982.) Otolith increment count Hatch classifi- cation Sample size Intercept Slope of regression line Standard error of slope R' Probability intercepts equal Probability slopes equal All data Early Late 117 39 9.4 14.2 0.3284 0.1711 0.0172 0.0364 0.76 0.37 <0.01 <0.01 Legend: Data were classified into early- and late-hatched larvae. Two dates of hatch were calcu- lated: 1) Age equalled increment count plus a constant, and 2) age equalled twice the increment count plus a constant. This resulted in a range of potential hatching dates. Any late-hatched larva whose range of hatch date overlapped the division date (Text Fig. 1) was eliminated from the analysis. Table 4.— Estimation of von Bertalanffy growth parameters for larval Atlantic her- ring from the Gulf of Maine. Hatch classi- Para- Estimate of Standard error of 95% confide nee interval Year fication meter parameter estimate Low High 1976-77 Early K 0.01865 0.00939 0.00008 0.03723 L^ 35.6 9.0 17.8 53.4 h -12.3 S.OOj -22.2 -2.41 Late K 0.01530 0.00457 0.00616 0.02443 Loo 29.8 1.7 26.4 33.2 to -38.01 12.46 - 69.92 -13.10 1978-79 Early K 0.00262 0.00158 - 0.00050 0.00575 Loo 113.2 48.8 16.3 210.1 tn -42.28 9.94 - 62.00 - 22.57 Late (convergence criteria cou Id not be me t) LARVAL HERRING 1978-1979 STUDY lU < > a. < QC UJ CD SEPT 1 OCT 1 NOV 1 DEC 1 SEPT OCT I NOV I DEC DATE OF HATCHING JAN Figure 5. -Frequency of Atlantic herring hatching during the 1978-79 study. Upper scale gives the day of hatch based on the Lough et al. (1982) aging method as discussed in the text. Lower scale gives the day of hatch based on Townsend and Graham's (1981) aging method as discussed in the text Arrow indicates division point between early- and late-hatched classification. 294 JONES: DIFFERENCES IN LARVAL I IKRHING GROWTH cantly different (P < 0.01)- early-hatched larvae grew faster than late-hatched larvae. 1978-79 Study A frequency plot of hatching dates under the assumption of daily increment deposition for larvae sampled in the Sheepscot estuary is shown in Figure 5. Analysis (Table 5) showed that the data were better fit with two lines {P < 0.01) and that the slopes were different (P < 0.01). The change in length was 0.21 mm/increment and 0.18 mm/incre- ment for early- and late-hatched larvae, respectively (Fig. 6). However, the results should be interpreted with the knowledge that Bartletf s test showed the variances to be heterogeneous. This could have been caused by actual heterogeneity of variances, or by nonnormality in the data. The F tests used in these analyses assumed equal variances between the hatch classifications. Cochran (1947) reported, however, that lack of homogeneity would decrease the power of an F test to discern true differences when they did, in fact, occur. Since differences were statisti- cally significant, not meeting this assumption did not hinder analysis (the use of various transformations Table 5.— Regression analysis of 1978-79 Gulf of Maine Atlantic herring data. (Data fro send and Graham 1981.) im Town- Otolith Hatch Slope of Standard Probability Probability increment classifi- Sample regression error intercepts slopes count cation size Intercept line of slope m equal equal All data Early 102 13.3 0.2134 0.0661 0.92 <0.01 <0.01 Late 198 14.2 0.1793 0.0060 0.82 60 or fewer Early 42 9.4 0.3378 0.0189 0.89 <0.01 <0.01 Late 53 11.4 0.2434 0.0203 0.74 Data were classified into early- and late-hatched larvae. These two groups were compared by fitting ordinary least squares regression lines to 1) all the data within the two classifications, and 2) using only lengths from larvae with 60 or fewer increment counts. Slopes and intercepts were compared between early vs. late for each group. 60 -n 50 E E 40H I O m 30- 20- )0 - LARVAL HERRING GROWTH 1978-1979 A A, W/ : TIME OF HATCH EARLY • LATE A T 20 40 80 100 T 60 80 100 120 140 160 INCREMENT COUNTS ~1 180 Figure 6. - Regression plot of length-at-otolith increment count for Atlantic herring. Complete data set represented. Data from Townsend and Graham (1981). 295 FISHERY BULLETIN: VOL. 83. NO. 3 did not result in homogeneity of variances). Except for the residuals for three small larvae, analysis for residuals showed no trends. For larvae with 60 increments and fewer (Fig. 7), Bartlett's test showed homogeneity of variance. These data were better fitted by two lines (P < 0.01); the slopes were significantly different (P < 0.01). The change in length was 0.34 mm/increment and 0.24 mm/increment for early- and late-hatched lar- vae, respectively. The von Bertalanffy growth equation fit the early- hatched larval data poorly (Table 4). L^ was esti- mated at 113.22 mm, with a 95% confidence interval of 16.37 to 210.06 mm. The von Bertalanffy growth function could not be fitted (solution would not con- verge) to the late-hatched larval data. The 1978-79 data could not be tested under assumptions that increment deposition could vary from 0.5 to 1.0 increment/d. Almost all of the calcu- lated hatch dates for late-hatched larvae, estimated on deposition rates of 0.5 increment/d, overlapped the classification division date. Too few points were left for analysis. DISCUSSION Evidence from the Gulf of Maine supports the hypothesis that increase in length for herring larvae hatched early in the spawning season is greater than for larvae hatched late in the season. These differ- ences were evident both under assumptions of daily otolith increment deposition and for deposition of one increment every other day. Before these dif- ferences are assumed to be due to differences in growth, however, there are other hypotheses which should be considered that could explain these obser- vations. Differences could be the result of within- season changes in otolith increment deposition rates, or of differential mortality due to selective predation. If there are within-season changes in otolith incre- ment deposition rates, growth (change in length at age) could actually be similar, but the calculated growth rates would appear to be different because they are expressed as change in length per incre- ment count. In order for this hypothesis to explain the above results, larvae born early in the season would be required to put down fewer increments per time period than would larvae born late in the season. The data allow a test of the hypothesis that larvae lay down fewer than 1 increment/d during the early part of the year. When estimated hatching dates are calculated for larvae caught early in the season, under the assumption that one increment was deposited every other day, some of these larvae 35 -1 LARVAL HERRING GROWTH 1978-1979 TIME OF HATCH EARLY • LATE A 10 20 I r \ 30 40 INCREMENT COUNTS 50 60 Figure 7. - Regression plot of length-at-otolith increment count for Atlantic herring. Only lengths for larvae with 60 or fewer otolith incre- ments have been included for analysis. Data from Townsend and Graham (1981). 296 .lONKS: DIFKKKKNCKSIN LAIUAL HKKKINC OKOWl'H would have had to appear in the plankton in the mid- dle of the summer (Fig. 3). However, newly hatched larvae are not found in significant numbers in the plankton before September (Boyar et al. 1973; Col- ton et al. 1979). It is far more plausible that larvae hatched early in the season, when growing condi- tions are more nearly optimal (Sherman and Honey 1971; Cohen and Lough 1983), deposit increments with close to daily periodicity. Hence, in order for this hypothesis to be true, late-hatched larvae would have to deposit increments at a rate greater than 1 increment/d. There is no evidence in the literature to support this for larval herring. Difference in population growth rates within a spawning season could also result from a shift in size- specific mortality during the season. The observed differences in growth rate could result if early-hatch- ed larvae have higher cumulative mortalities for slower growing individuals, while late-hatched larvae have higher mortalities for faster growing indivi- duals. Progressively, fewer and fewer of the selec- tively predated larvae would be seen in older ages. This would result in differences in population growth rates that are not apparent for individuals within the population. Although differential mortality cannot be dis- missed with the available data, the most plausible explanation for the differences in length-at-incre- ment count is an actual difference in larval growth rate over the spawning season. Such differences in population growth rate can be important for larval herring survival. Since greater time spent in the lar- val stage is thought to be related to increased mor- tality, it is interesting to note that an early-hatched larva from the 1978 study would require, on the average, 80 d to reach 30 mm, compared with 88 d for a late-hatched larva. For the 1976 study, it would take, on the average 63 d for an early-hatched larva to reach 30 mm compared with 157 d for a late- hatched larva to reach this size. It has been shown that in both years, late-hatched larvae are larger than early-hatched larvae at the time of first increment formation. This could result from larger eggs being produced in the winter (Cushing 1967), or from different growth rates from hatch to the age of larvae covered in this study. Without further evidence of differences in egg size or actual growth rates between hatch and the age these studies began, neither hypothesis can be sup- ported. Differences in growth rate within the spawning season can contribute to error when using an age- length key to age larvae. For a given length, samples containing early-hatched larvae would yield different ages than samples containing late-hatched larvae. For the 1978-79 study (under the assumption of daily increment deposition), a 25 mm larva would average 60 increments for early-hatched larvae versus 56 for late-hatched larvae. For the 1976-77 study a larva of this length would average 47 versus 102 increments, respectively. This additional variation should be taken into consideration when using age-length keys for larvae. Differences in growth during the spawning season might be due to changes in the environment when a species of fish spawns over a protracted time period, such as Atlantic herring which spawns from late August through November (Boyar et al. 1973; Col- ton et al. 1979). Early in the season copepods, the main food for larval herring (Sherman and Honey 1971; Cohen and Lough 1983), are more abundant than late in the spawning season (Sherman et al. 1983). Temperatures average 12°-16°C early in the season and < 8°C later in the season (Colton 1968; Colton and Stoddard 1972). Day length and metabolic demand may also vary over the spawning season. Alternately, differences in growth between larvae hatched early and late in the season could be the result of genetic differences if early and late spawners are from different stocks. ACKNOWLEDGEMENTS I thank R. G. Lough, D. Townsend, and J. J. Graham for providing their data; and B. E. Skud, R. G. Lough, J. J. Graham, and K. R. Hinga for their suggestions and review of this manuscript. LITERATURE CITED Beverton, R. J. H., AND S. J. Holt. 19.54. On the dynamics of exploited fish populations. Fish. Invest. Minist. Agric, Fish., Food (G.B.) Ser. II, Vol. XIX, 533 p. Boyar. H. C, R. R. Marak, F. E. Perkins, and R. A. Clifford. 1973. Seasonal distribution and growth of larval herring {Clupea harengns L.) in the Georges Bank-Gulf of Maine area from 1962 to 1970. J. Cons. Int. Explor. Mer 35:36-51. Cochran, W. G. 1947. Some consequences when the assumptions for the ana- lysis of variance are not satisfied. Biometrics 3:22-38. Cohen, R. E., and R. G. Lough. 1983. Prey field of larval herring Clupea harengus on a Con- tinental Shelf spawning area. Mar. Ecol. Prog. Ser. 10:211- 222. Colton, J. B., Jr. 1968. Recent trends in subsurface temperatures in the Gulf of Maine and contiguous waters. J. Fish. Res. Board Can. 25: 2427-2437. 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. 297 FISHERY BULLETIN: \()L. 83. NO. .S Geogr. Soc. Folio 22, 10 p. CoLTON, J. B., Jr.,"' W. G. Smith, A. K. Kendall, .Jr., P. L. Berrien, and M. P. Fahay. 1979. Principal spawning areas and times of marine fishes, Cape Sable to Cape Hatteras. Fish. Bull., U.S. 76:911-915. Gushing, D. H. 1967. The grouping of herring populations. J. Mar. Biol. Assoc. U.K. 47:193-208. Das, N. 1968. Spawning, distribution, survival, and growth of larval herring {Clupea harengus L.) in relation to hydrographic con- ditions in the Bay of Fundy. Fish. Res. Board Can., Tech. Rep. 88, 162 p. Draper, N. R., and H. Smith. 1981. Applied regression analysis. 2d ed. John Wiley and Sons, Inc., N.Y., 709 p. Geffen, A. J. 1982. Otolith ring deposition in relation to growth rate in herring (Clupea harengus) and turbot (ScaphthalmTis inaxir WM-s) larvae. Mar. Biol. 71:317-326. Gjosaeter, H. 1981. Dagsonelesing som metode i aldersstudier pa fisk, med eksempler pa anvendelse pa tropiske og boreale arter. Ph.D. Thesis, Univ. Bergen, Bergen, Norway, 172 p. HOENIG. N. A. 1982. A study of seasonal growth models for fishes. M.S. Thesis, Univ. Rhode Island, Kingston, 91 p. Jones, C. M. 1984. The otolith increment technique: Application in larval fish. Ph.D. Thesis, Univ. Rhode Island, Kingston, RI, 123 p. Lough, R. (J., and G. R. Bolz. 1979. A description of the sampling methods, and larval her- ring {Clupea harengus L.) data for surveys conducted from 1968-1978 in the Georges Bank and Gulf of Maine areas. Natl. Mar. Fish. Serv., NOAA, Northeast Fish. Cent., Woods Hole, Mass., Lab. Ref Doc. 79-60, 230 [>. Lough, R. G., M. Pennington, G. R. Bolz, and A. A. Rosenburg. 1982. Age and growth of larval Atlantic herring, Clupea harengus L., in the Gulf of Maine-Georges Bank region based on otolith growth increments. Fish. Bull., U.S. 80:187-199. 0IESTAD, V. 1982. Application of enclosures to studies on the early life history of fishes. In G. D. Grice and M. R. Reeve (editors). Marine mesocosms, p. 49-62. Springer-Verlag, N.Y. OSTLE, B., and R. W. Mensing. 1975. Statistics in Research: basic concepts and techniques for research workers. Iowa State Univ. Press. Ames, Iowa, 596 P- Sherman, K., J. R. Green, J. R. Goulet, and L. E.isymont. 1983. Coherence in zooplankton of a large northwest Atlantic ecosystem. Fish. Bull., U.S., 81:855-862. Sherman, K., and K. A. Honey. 1971. Seasonal variations in the food of larval herring in coastal waters of central Maine. Rapp. P. -v. Reun. Cons. int. Explor. Mer 160:121-124. Townsend, D. W., and J. J. Graham. 1981. Growth and age structure of larval Atlantic herring. Clupea harengus harengus. in the Sheepscot River Estuary, Maine, as determined by daily growth increments in otoliths. Fish. Bull., U.S. 79:123-130. 298 SEASONAL CYCLES OF FAT AND GONAD VOLUME IN FIVE SPECIES OF NORTHERN CALIFORNIA ROCKFISH (SCORPAENIDAE: SEBASTES) Patrick J. Guillemot, ' Ralph J. Larson,^ and William H. Lenarz^ ABSTRACT Seasonal changes in visceral fat volume and gonad volume are compared in five species of rockfish from northern and central California: Sebastes entomelas, S. paucispinis, S. goodei, S. pinniger, and S. flairidus. In these species, visceral fat was deposited between spring and fall, at the same time as gametogenesis. Visceral fat declined in volume between fall and spring, coinciding with the decline in volume of testes and preceding the release of embryos in females. We suggest that increased feeding during the summer upwell- ing season provides the energy for simultaneous fat accumulation, gametogenesis, and perhaps somatic growth. During subsequent seasons of presumed food storage, these rockfishes may utilize visceral fat reserves for maintenance. This pattern of fat deposition and utilization may contribute to the long life and repeated reproduction of rockfishes, at the expense of current fecundit\' and growth. Our data also suggest that rockfishes which spawn only once during the reproductive season have fat cycles of greater magnitude than those spawning more than once a year. Of the species that we studied, the apparent single spawners S. entomelas and S. flaindus have more northerly geographic distributions and larger fat cycles than the multiple spawners S. goodei and 5. paucispinis. It is possible that the shorter and more pronounced productive season in the north leads to a greater need for fat reserves during winter and makes the wintertime production of additional batches of eggs energetically difficult. The seasonal storage and utilization of lipid (and nonlipid) reserves are important in the metabolic activities and overall life histories of many animals (Love 1970; Shul'man 1974; Derickson 1976a and associated papers). While cycles of lipid storage and utilization are generally associated with seasonal changes in food availability (Derickson 1976b) or metabolic demands (Lawrence 1976), the functions of lipid storage are varied. In many fishes, reserves are used primarily in reproduction, as indicated by complementary cycles of lipid content and reproductive activity (Lasker 1970; Shchepkin 1971a, b; Schevchenko 1972; Shul'man 1974; Tyler and Dunn 1976; Wootton and Evans 1976; Lasker and Smith 1977; Diana and MacKay 1979; Delahunty and de Vlaming 1980; Patzner 1980; Hunter and Leong 1981). The seasonal accumulation of sufficient reserves may be a prerequisite for sexual maturity in some fishes (lies 1974), and the amount of material stored may in- fluence fecundity (Tyler and Dunn 1976). Reserves 'Southwest Fisheries Center Tiburon Laboratory, National Marine Fisheries Service, NOAA, Tiburon, CA; present address: 3298 Madera Avenue, Oakland, CA 94619. ^Department of Biological Sciences, San Francisco State Univer- sity, San Francisco, CA 94132. ^Southwest Fisheries Center Tiburon Laboratory, National Marine Fisheries Service, NOAA, 3150 Paradise Drive, Tiburon, CA 94920. Manuscript accepted August 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. may also be used in migration (Robertson and Wex- ler 1960; Dotson 1978; Glebe and Leggett 1981 a, b), and, when used in spawning migrations, may con- tribute indirectly to reproduction. Slobodkin (1962) and Calow (1977), however, noted that fat deposition may actually detract from reproduction, particularly when fat deposition and reproduction are concurrent. In such cases, reserves are often used instead for maintenance during periods of food scarcity (Calow and Jennings 1977), enhancing the opportunity to reproduce in the future. Some fishes seem to use reserves both for reproduction and maintenance, when spawning oc- curs during periods of food scarcity or fasting (Wilkins 1967; MacKinnon 1972; lies 1974; Newsome and Leduc 1975; Foltz and Norden 1977; Dawson and Grimm 1980; Pierce et al. 1980; Glebe and Leggett 1981 a, b). The interpretation of such cases is complex, since reproduction and maintenance may be competing concurrently for reserves. Roberts (1979) noted that fat was deposited seasonally around the viscera of two species of shallow-water rockfish, Sebastes mystinus and S. melanops. He suggested that the cycle of fat deposi- tion and utilization was related to seasonal changes in the abundance of food and to reproduction. In this paper we examine the seasonal relationship 299 FISHERY BULLETIN: VOL. 83, NO. 3 between visceral fat volume and gonad volume in five offshore species of rockfish: Sebastes entomelas, S. paucispinis, S. goodei, S. pinniger, and 5. flavidus. Based on Roberts' suggestion and the general literature on fat cycles in fishes, we expected to find complementary cycles of fat and gonad volume, indicating that reserves are used in repro- duction. We used visceral fat volume as a convenient index of the timing and magnitude of seasonal changes in reserves, explicitly assuming that reserves stored elsewhere would change coinciden- tally (Delahunty and de Vlaming 1980). Since we did not measure total reserves, our data on visceral fat volume cannot be applied to quantitative studies of energ\' budgets, but are most useful for examining seasonal changes and making comparisons within a group of closely related species. The use of gonadal volume as an index of reproduc- tive activity requires some explanation. In males, gonadal volume is a good index of the timing of gametogenesis. In females of this live-bearing genus, however, gametogenesis is often reflected only in the initial increase of gonadal volume. Following fertili- zation (which may occur some time after copulation, due to sperm storage [Sorokin 1961; Echeverria 1981''; Boehlert and Yoklavich 1984]), ovarian volume continues to increase during a gestation period that lasts a month or so (Moser 1967). There may be some postzygotic nutrition of embryos dur- ing this time (Boehlert and Yoklavich 1984), but much of the increase in volume is due to the ac- cumulation of water (Moser 1967). In some species of Sebastes that spawn more than once a year, gametogenesis takes place again late in the repro- ductive season of females. Among the species con- sidered here, S. paucispinis and 5. goodei are clearly multiple spawners (Moser 1967; MacGregor 1970), and Echeverria^ has presented some evidence for multiple spawning in S. pinniger. MATERIALS AND METHODS Party boat and commercial rockfish catches were sampled weekly, as a part of a large-scale study, by personnel from the National Marine Fisheries Ser- vice (Tiburon Laboratory) and the California Depart- *Echeverria, T. 1981. Maturity in six species of roci 9 NS 9 > o- NS 9 > o" NS NS 9 > CT NS Sebastes pinniger Cycles of visceral fat volume were poorly defined in S. pinniger. In females, only the summer and fall fat regressions differed significantly, suggesting a peak in fat content during the fall (Fig. 1). In males. no two consecutive seasons differed significantly in fat content, and the only two seasons that differed significantly at all were spring and fall (Fig. 1). Thus, males may also have had peak fat content during fall, but their fat cycle was not pronounced. The visceral fat content of males and females did not differ signi- ficantly during their apparent fall peaks (Table 3). The ovarian cycle of S. pinniger was also poorly defined in our data. Ovaries increased significantly in volume between spring and summer (Fig. 2). No other seasons differed significantly, but a peak in winter is indicated in Figure 2. In males, gonadal volume increased dramatically from spring to a peak in summer, remained fairly high in fall, and declined in winter (Fig. 2). Sebastes flavidus Visceral fat volume in female .S. Jlavidtis increased from spring through fall, and then declined in winter (Fig. 1). Ma\e S.Jlavidits exhibited a gradual increase in fat content between spring and fall, followed by a decline to minimal fat levels in winter (Fig. 1). Fat content of females exceeded that of males during their fall peaks, and the difference in fat content dur- ing their winter lows was not significant (Table 3). Ovarian volume in S. flavidus showed a pattern similar to most of the rest of the species: an increase 304 CUILLEMOT ET AL.: SEASONAL CYCLES IN CALIFORNIA ROCKFISH Females S. entomelas Males < o C5 S. paucispinis S. goodei S. pinniger S. flavidus Summer Winter Spring Summer Winter Spring Fall Figure 2. - Seasonal variation of gonad volume in adults of five species of Sebastes from northern and central California, as in Fi^ire 1. Averages of In (total length, mm) used in estimating gonadal volume are S. ev- tomdas females, 6.14, and males, 6.06; S. paucispinis females, 6.24, and males, 6.17; S. goocki females, 6.09, and males, 5.88; S. pintiiqrr females, 6.20, and males, 6.13; and S. flavidus females, 6.09, and males, 6.04. Seasons: spring (March-May 1981), summer (.lune-August 1980), fall (September-November 1980), and winter (December 1980-Februar>' 1981). 305 FISHERY BULLETIN: VOL. 83, NO. 3 from low volume in spring to peak volume in winter (Fig. 2). In males, the volume of testes increased from spring to a peak in summer, declined slightly in fall, and fell greatly in winter (Fig. 2). Summary and Comparison of Adult Fat Cycles In all species studied, visceral fat volume was high, if not at a peak, during fall. In males, fat content was usually high during summer as well. In male S. en- toTnelas, S. paucispinis, and S. Jlavidus, summer and fall fat content did not differ significantly, and in male S. goodei the fat content in summer was actual- ly greater than in fall. Summertime fat content was relatively high in female S. entomelas, S. goodei, and S. flavidus, but in females of all species the fat con- tent increased to a peak in fall. Winter and spring were usually the seasons of low fat volume in both sexes. Little pattern existed in S. pinniger, except that fat volume appeared to reach a peak in fall. Both the relative amplitude of fat cycles and peak fat volume differed among the species and sexes. In Table 4.— Relative amplitude of fat cycles in five spe- cies of Sebastes. The ratio of estimated peak season : low season fat volume is presented for eacti species and sex, and is used as an index of cycle amplitude. Ratios are the antilogs of the differences between estimated In (fat volume) at the average of In (total length) in peak and low fat seasons (Fig. 1, Table 3). Amplitude Species Females Males S. entomelas 3.50 3.10 S. paucispinis 1.83 1.44 S. goodei 3.20 2.91 S. pinniger 1.67 1.84 S. flavidus 3.61 2.64 male and female S. entomelas, S. goodei, and S. flavidus, the fat volume of average-sized fish changed by about a factor of three during the year (Table 4). The amplitude of the fat cycle was lower in S. paucispinis and S. pinniger, where there was less than a twofold change in fat content between peak and low seasons (Table 4). Among females, S. en- tomelas and S. Jlavidus generally had higher peak fat volumes than the other species (Table 5). Among the other three species, peak fat content of females seemed to decrease from S. pinniger to S. paucispinis toS. goodei (Table 5). Females of S. ento- melas and S. Jlavidus, then, possessed high-volume, high-amplitude fat cycles, while S. pinniger and S. paucispinis had low-volume, low-amplitude fat cycles, and S. goodei showed low volume and rela- tively high amplitude. The peak fat volume in maleS. entomelas far exceeded that of the remaining species, most of which had relatively similar fat- length relationships during peak seasons (Table 5). Thus among males, only S. entomelas showed a high- volume, high-amplitude fat cycle. Males of S. Jlavidus and S. goodei possessed fat cycles of low volume and moderately high amplitude, and S. paucispinis and S. pinniger had cycles of low volume and low ampli- tude, like the females of these species. Fat Cycles in Juveniles Fat cycles were examined in juveniles of male and female S. paucispinis, and of female 5. pinniger and iS. Jlavidus, where seasonal sample sizes were large enough to permit analysis (Table 6). In 5. paucispi- nis, juvenile females appeared to reach a peak in fat content during winter, but in juvenile males there were no significant differences among the seasonal regressions (Fig. 3). In S. pinniger, juvenile females Table 5.— Comparisons of the peak-season regressions of In (fat volume) on In (total length) for different species of Sebastes. Comparisons were made by analysis of covariance. For significantly differing regressions (P < 0.05), the comparative fat volumes are indicated (based on the positions of the regressions, cf. Fig. 1, Table 1), unless intersecting lines made position difficult to interpret. NS denotes regressions that did not differ significantly. S. paucispinis S. goodei S. pinniger S. flavidus Females S. entomelas S.e. > S.pa. S.e. > S.g. NS NS S. paucispinis S.pa. > S.g. NS S.pa. < S.f. S. goodei S.g. < S.pi. S.g. < S.f. S. pinniger S.pi. < S.f. Males S. entomelas S.e. > Spa. S.e. > S.g. S.e. > S.pi. S.e. > S.f. S. paucispinis intersection S.pa. < S.pi. intersection S. goodei NS NS S. pinniger NS 306 GUILLEMOT ET AL.: SEASONAL CYCLES IN CALIFORNU ROCKFISH possessed little fat in the spring, and higher levels in summer through winter (Fig. 3). Fat content in juveniles of female S. flavidus peaked in fall (Fig. 3). Table 6.— Seasonal regressions of In (visceral fat volume, mL) on In (total length, mm) in juveniles of three species of Sebastes from northern and central California. Species and sex Season Sample size Intercept Slope r S. paucispinis females Spring 140 - 19.637 3.510 0.325 Summer 138 - 26.793 4.669 0.455 Fall 55 - 23.626 4.189 0.419 Winter 67 - 29.508 5.205 0.719 males Spring 92 -10.309 1.976 0.157 Summer 86 - 22.626 3.969 0.362 Fall 10 - 48.866 8.324 0.599 Winter 21 - 27.486 4.818 0.766 S. pinniger females Spring 41 -21.012 3.655 0.402 Summer 76 -21.794 3.859 0.624 Fall 37 - 36.539 6.383 0.875 Winter 20 -18.568 3.345 0.651 S. flavidus females Spring 21 - 22.796 4.080 0.388 Summer 70 - 8.502 1.642 0.260 Fall 53 - 36.450 6.558 0.639 Winter 19 - 9.906 1.848 0.361 Thus, in three of the four groups analyzed, fat con- tent of juveniles varied during the year. The timing of the apparent cycles was similar to that of adults. The amplitude of the fat cycle in juveniles of female S. paucispinis and S. pinniger (less than twofold change in fat volume) was similar to adult amplitude. In female S. Jlavidus, the amplitude of the fat cycle in juveniles (less than twofold change) was lower than in adults. As adults, S. flavidits showed high- amplitude fat cycles, while 5. paucispinis and S. pin- niger showed low-amplitude cycles. DISCUSSION Clear cycles of visceral fat volume and gonadal volume occurred in both sexes of S. entomelas, S. paucispinis, S. goodei, and S. flavidus. The fat cycles of male and female S. pinniger were less well- defined, as was the gonadal cycle of female S. pin- niger. Many of the female S. pinniger we examined were relatively small adults. This may account, at least in part, for the small and ill-defined fat and gonad cycles, as the fat cycles of female 5. flavidus increased in amplitude from juveniles to adults. Visceral fat volume generally increased propor- S. paucispinis Females S. paucispinis Males S. pinniger Females S. flavidus Females < U. ^ ^^-" • . 1 Summer Spring Fal Winter Summer Spring Fall Winter SEASON Figure 3. -Seasonal variation of visceral fat volume in the juveniles of five species of Sebastes, as in Figure 1. Averages of In (total length, mm) used in estimating fat volume are S. paucispinis females, 6.08, and males, 6.04; S. pinniger females, 5.88; and S. flavidus females, 5.78. Seasons: spring (March-May 1980), summer (June-August 1980), fall (September-November 1980), and winter (December 1980-February 1981). 307 FISHERY BULLETIN: VOL. 83, NO. 3 tionately with fish volume, since most slopes of In (fat volume) on In (length) were near 3. Slopes smaller than 3 occurred in those seasons when fish of all sizes had little fat. Some slopes during seasons of high fat volume were significantly greater than 3, in- dicating that larger fish had disproportionately greater fat volume than smaller fish. This was parti- cularly true of, but not limited to, the species and sexes with high-magnitude fat cycles (males and females of 5. entomelas and females ofS.flavidus). It would be interesting to discover whether larger females of 5. pinniger have high-amplitude fat cycles, since the slopes of the fat regressions for female S. pinniger were relatively high. The dependence of fat volume on fish size and season of collection, while applicable on average, does not apply to every individual. The scatter about the regressions of fat volume on length was large. Some of the scatter was probably due to errors in measurement, but a large part was real. We have observed large differences in the amount of visceral fat among fish of the same size and sex in the same collection. The reasons for such variation may be dif- ficult to resolve, since it is difficult to know the history of individuals collected in the field. However, our data were adequate in most species to detect seasonal changes on average. We wanted to compare the timing of fat cycles and gonad cycles, and thus learn something of the function of fat storage in rockfish. Our initial hypothesis was that stored fat would be used in reproduction, based on Roberts' (1979) observations. The data for males do not support this hypothesis. Fat and gonad cycles were nearly coincident, with peaks in both cycles occurring in fall and/or summer and lows in winter/spring. Thus the energy for gonadal growth was probably derived from current ingestion, not from material previously stored as fat. Maximum somatic growth also coincides with the summertime growth of testes and fat deposits (e.g.. Love and Westphal 1981), so all three processes may be competing for energy consumed during this time. The situation is more complicated in females. Some fat was deposited between spring and summer in females of S. entomelas, S. goodei, Sind S. Jlavidus, and females of all species gained fat between sum- mer and fall. Some gonadal growth took place be- tween spring and summer in females of all species except S. paucispinis. In all but 5. pinniger, however, ovarian volume increased steadily between summer and winter. The growth of ovaries through fall was due largely to vitellogenesis, while the greater growth of ovaries into winter was due primarily to embryogenesis and hydration of ova (Moser 1967). Additional vitellogenesis probably occurred during winter in the multiple spawners, 5. paucispinis and S. goodei (Moser 1967; MacGregor 1970). In females, then, fat deposition usually began in summer, perhaps slightly before the initiation of vitellogenesis, but continued into the main period of vitellogenesis between summer and fall. Thus, like males, females of these species deposited fat more or less concurrently with gonadal maturation and somatic growth. The depletion of visceral fats oc- curred between fall and spring in females of most species, during and after gestation. It is possible that fat reserves are used for the maturation of additional ova in multiple spawners or are involved in the nutri- tion of embryos (Boehlert and Yoklavich 1984), but as in males, reserves are not used in the initial devel- opment of gonads. There were two main differences in the fat cycles of males and females: females usually had larger fat cycles than males, and the peak fat volume of females occurred in fall, while fat volume in males usually reached a plateau that spread through sum- mer and fall. It is possible that courtship activity in males (Helvey 1982) draws energy from fat deposi- tion between summer and fall, or that this activity draws time from feeding. As a result, females may continue to fatten after fat deposition ceases in males. It is also possible that females require more reserves in winter and they somehow are able to ac- quire these extra reserves. The synchronous depletion of reserves in males and females, however, indicates a common function for such reserves. We suggest two possible func- tions: First, some rockfish may migrate during the period of fat depletion. Love (1981) presented evidence for seasonal movements in SebaMes paucispinis and S. entomelas off southern California, but had no data on the extent or direction of move- ment. Females of S. alutus undertake seasonal migrations covering as much as 300 m of depth (sum- marized in Gunderson 1977). Lenarz (pers. obs.) found evidence of seasonal movement in female S. jordani. Several species of rockfish undertake little or no seasonal migration, including ^S. ./7a rvV/M.s from shallow waters off Alaska (Carlson and Barr 1977). Since seasonal migrations in rockfish may not be long and seem to occur primarily in females, we sug- gest that other possibilities be ruled out before ac- cepting migration as a major use of fat reserves. Second, we suggest that fat reserves are used for maintenance during wintertime periods of reduced food availability. Fattening, as well as growth and 308 (;U1LLKM()TETAL.:SKAS()NAL CYCLES IN CALIFORNIA ROCKFISII initial gametogenesis, occurs during and shortly after the spring-summer upwelling period, and deple- tion occurs during and after a time of lower primary production (Bolin and Abbott 1963). Roberts (1979) noted that several species of inshore rockfish near Monterey consumed more food during the summer, when macroplankton (euphausiids) and juvenile rock- fish were abundant. During the nonupwelling season, juvenile rockfish and macroplankton were less abundant, and adult rockfish consumed less food. Love and Westphal (1981) found less food in the stomachs of S. serranoides during the nonupwelling season near Morro Bay. Gunderson (1977) noted a "summer feeding season" in 5. alutus. Hobson and Chess', however, found the gelatinous prey of 5. mystinns to be more abundant during nonupwelling periods off northern California. There is only limited information on the seasonal food habits of the five species studied here. Sehastes paucispinis is largely piscivorous (Phillips 1964), while S. entomelas, S. goodei, S. pinniger, and S. fiavidus feed on macro- plankton and small fish (Phillips 1964; Pereyra et al. 1969; Brodeur 1982). Brodeur (1982) examined seasonal changes in the diets of S. fiavidus and 5. pinniger, and found that food consumption declined during winter in 5. pinniger, but not in S. fiavidus. However, Carlson and Barr (1977) found a pro- nounced decline in activity during winter in 5. fiavidus off Alaska, suggesting seasonal changes in feeding. This hypothesis can be examined with con- current data on food consumption and fat volume. It is not mutually exclusive with other functions of fat reserves, since some fat could also be used for migra- tion, nourishment of embryos, or maturation of addi- tional ova in multiple spawners. The wintertime use of reserves for maintenance, however, is consistent with the overall pattern of long life and repeated reproduction in rockfish (as discussed also by Gunderson [1977]). As suggested by Slobodkin (1962) and Calow (1977), summertime fat deposition may exact a cost in current reproduc- tion (if more gametes could be packed into rockfish abdomens) or in future increases in gamete capacity (through additional growth), but may help to ensure survival. An interesting result of our study was the differ- ence in magnitude of fat cycles among species, which was particularly evident in females. These differ- ences are correlated with both the frequency of spawning and geographical distributions of the ^Hobson, E. S., and J. R. Chess. 1981. Seasonal patterns in trophic relationships of the blue rockfish, Sebaates myatinu-ti. |Abstr.| Proceedings of the 1981 Western Groundfish Conference. species. Females of S. entomelas and S. Jlaviduji had fat cycles of high amplitude and volume, probably spawn only once a year (Westrheim 1975; Eche- verria footnote 5), and have northerly distributions (Phillips 1957, 1964; Gunderson and Sample 1980; Adams 1980). On the other hand, 5. paucispinis and S. goodei had fat cycles of low amplitude and/or volume, spawn more than once a season, at least in southern California, and have southerly distributions (Phillips 1957, 1964; Gunderson and Sample 1980; Adams 1980). The case with S. pinniger is unclear. Females seemed to have low-amplitude fat cycles, but the small sample and small sizes of females in the samples make conclusions tenuous. Echeverria (foot- note 5) has some evidence for multiple spawning in S. pinniger. If so, its northerly distribution would be inconsistent with the pattern seen in the other species. We will restrict our discussion to the remain- ing four species, although further studies of S. pin- niger may be instrumental in evaluating the sugges- tions we make below. The small fat cycles of multiple spawners are prob- ably not caused by their potentially greater repro- ductive output, since fat deposition stops long before the maturation of additional batches of ova. It seems more likely that the magnitude of the fat cycle, fre- quency of spawning, and geographical distribution are all related. Boehlert and Kappenman (1980) sug- gested that year-round spawning in southern popula- tions of 5. diploproa served to increase reproductive output in response to annual variation in the survival of planktonic larvae. Multiple spawning in southern populations and species may also be a response to the more even seasonal distribution of upwelling in southern vs. northern areas of the northeastern Pacific (Bakun et al. 1974; Parrish et al. 1981; Smith 1978), as it affects planktonic larvae. However, the more restricted upwelling season in northern waters may also lead to larger and better defined seasonal variation in the food supply of adults. A more pro- nounced summertime pulse of food may enable northerly species like 5. entomelas and S. fiavidus to store large amounts of fat. Similarly, a more pro- nounced decline in food during winter may make the accumulation of such reserves necessary, and preclude the maturation of additional batches of ova. On the other hand, a more even seasonal distribution of food in the south may enable southerly species like S. paucispinis and S. goodei to produce additional batches of ova during winter, and may also reduce the need for wintertime reserves. These suggestions are speculative, but it should be possible to compare the seasonal variation of food consumption in species with northern vs. southern distributions. 309 FISHERY BULLETIN: VOL. 83. NO. 3 In summary', we suggest that seasonal changes in the availability' of food are rather important in the life histories of rockfish. The summer storage and winter use of fat reserves, in near synchrony with growth and decline of gonad volume, indicate a wintertime paucity of food that is compensated by the use of material that was stored at the expense of fecundity or growth. Further study of seasonal feeding, fat deposition, and reproductive cycles, categorized by size and sex, would be useful in evaluating this hypothesis. Further, geographical comparisons of these cycles may illuminate some causes of differences in the life histories of rockfishes. ACKNOWLEDGMENTS For their assistance during the course of the study, we thank Tina Echeverria, Pete Adams, Connie Ryan, Nancy Wiley, and many other employees of NMFS and the California Department of Fish and Game. Pete Adams, Tom Niesen, and George Boeh- lert suggested improvements in the manuscript. Diane Fenster drafted the figures. This paper is derived from the senior author's Master's thesis at San Francisco State University. The work took place at the Southwest Fisheries Center Tiburon Labora- tory, National Marine Fisheries Service, CA, under a NMFS student appointment. LITERATURE CITED Ar)AM.s. p. B. 1980. Morphologj' and distribution patterns of several impor- tant species of rockfish ({jt'nus Sehastes). Mar. Fish. Rev. 42(.3-4):8()-82. Bagenal, T. 1978. Aspects of fish fecundity. In S. L). Gerkinjj (editor), Ecology of freshwater fish production, p. 7.5-1(11. Biackwell Scientific Publications, Oxford. Bakun, a., D. R. McLain, and F. V. Mayo. 1974. The mean annual cycle of coastal upweliing off western North America as observed from surface measurements. Fish. Bull.. U.S. 72:843-844. BOEHLERT, G. W., AND R. F. KaFPENMAN. 1980. Latitudinal growth variation in the ji;enus Sehnsfcs from the northeast Pacific Ocean. Mar. Ecol. Prog. Ser. .3:1 -10. BOEHI.ERT. G. W.. AND M. M. YoKLAVICH. 1984. RepHKiuction. embryonic energetics, and the maternal- fetal relationship in the viviparous genus Sebastes (Pisces: Scor{)aenidae). Biol. Bull. (Woods Hole) 167:354-370. BoLiN. R. L., AND D. P. Abbott. 1963. Studies on the marine climate and phytoplankton of the central coastal area of ('alifornia, 1954-1960. Calif. Coop. Oceanic Fish. Invest. Rep. 9:2.3-45. Brodeur, R. D. 1982. Food habits, dietary overlap and gastric evacuation rates of rockfishes (Genus Sebn.sles). MS Thesis, Oregon State Univ., Corvallis, 102 p. Calovv, p. U)77. Ecology, evolution and energetics: a study in metabolic adaptation. Adv. Ecol. Res. 10:1-62. Calow, p., AND J. B. Jennings. 1 977. Optimal strategies for the metabolism of reserve mater- ials in microbes and metazoa. J. Theor. Biol. 65:601-603. Carlson, H. R., and L. Barr. 1977. Seasonal changes in spatial distribution and activity of two species of Pacific rockfishes, Sebn^ften flnndns and S'. ciliatvs in Lynn Canal, southeastern Alaska. Mar. Fish. Rev. 39(3):23-24. Dawson, A. S., and A. S. Grimm. 1980. Quantitative seasonal changes in the proteins, lipid and energy content of the carcass, ovaries and liver of adult female plaice. Plfumn/'ctf's |ll^t('!<^ta L. J. Fish Biol. 16:493- 504. Delahi'NTY, G., and V. L. de Vlaming. 1980. Seasonal relationships of ovary weight, li\er weight and fat stores with body weight in the goldfish, ( 'a nissius iinriitiis (L.). J. Fish Biol. 16:.5-13. Derickson, W. K. 1976a. Introduction to the symposium: lipi 0 (7) (Again, we must remark upon the limitations of the present calculation; certainly the shelf water is not thoroughly mixed and the density of the outflowing larvae is much more complicated and perhaps smaller on the whole than this formula would sug- gest. We hope that our results will spur further modelling and observational efforts to assess the pro- cesses we have been forced to represent so crudely.) When all of these simplifications are gathered together, the approximate equations for the average density of larvae N{x,t) become 10 for Fo < 0 VoN ] = -1^ for Fo > 0 316 Fl.lKRLAN'U\VKOKLh;\VSKl:\VAKiVU'OKK(;ri.FSIKKAMKlN(;s dU _ ^0 dx Y with the boundary condition iV(0,0 = iVo(0. STATIONARY EDDIES (8) (9) In nature, there are pulses of larvae entering the domain as the fish spawn. In addition, the shelf-edge velocities V^){x,t) are changing as mesoscale eddies and Gulf Stream warm core rings impinge upon the shelf. We shall present in the section on moving ed- dies several numerical solutions of Equations (8) and (9), simulating this complex situation. However, in order to fully understand the importance of the rings and eddies in determining the fish larvae's spatial distribution, it is first useful to consider some simpler, analytically tractable cases. We shall begin by discussing the distributions which occur when the shelf-edge flows are not changing with time, i.e., the eddies are stationary. This problem also has bearing on the real situation south of Long Island, where rings may often stop for considerable lengths of time. As a first example, consider the larval fish distribu- tion which would occur in the absence of any biolo- gical loss processes (f^ = 0) and when the source term A^Q is independent of time. The resulting equations dx iUN) = N Fo — foryo>0| Y for F„ < 0 au dx Y (10) can be solved readily where f/,, is the longshore velocity and N,, is the (time-independent) population density at the upstream boundary x = 0. We can now see explicitly the effects of the physics alone upon the larval fish distribution. In the regions where the flow is onto the shelf (Equation (11a)), the shelf break boundary con- tribution to Equation (10) is zero. But the effects of the tlow field are still felt in that the along-shelf flow is divergent. U increases downstream as water comes onto the shelf, spreading out the larvae and reducing their average density. In contrast, when the flow is offshore (Equations (lib) or (lie)), there are direct loss terms due to larvae being carried off the shelf. Some of the water flowing into a section are diverted offshore while some continues down the shelf, with the larvae separating in the same propor- tions. Thus, although there is a decreased flux down the shelf, this does not affect the density since there are no biological losses which need to be balanced by this flux. The net effect is that the physics by itself does not change the population density in regions of offshore flow (Equation (lib)). The only exception would occur when the offshore transport (/ F,, dx) is sufficiently strong so that all of the normal alongshore flow (f/,, Y) is diverted off the shelf. In this case (Equation (lie)), the flow in regions farther down the shelf is reversed and the water moves up the shelf. Since this water is from regions without . sources of larvae, the population density is zero. By putting together these two results, we can con- struct a picture of the density of larvae in continental shelf water flowing past a stationary ring centered a.tx = D. This is shown in Figure 3. For these calcu- lations we have used n = ,x- D ■A exp [2 2 L' ] (12) with A = 20 cm/s the peak offshore velocity) and L = 20 km (so that roughly 80 km along the shelf is strongly influenced by the ring currents). This figure A^ = 'A^o Nn U N,a o"-^o C/n S Va 0 Y if Fo < 0 if Fo > 0 and U^ > if F„ > 0 and Uo < Va Y (11a) (lib) (lie) 317 FISHERY BULLETIN: VOL. 83, NO. 3 ^1 No < > < O z < o z m < 0) H.; DISTANCE - DOWNSTREAM X (KM) 500 20 OFFSHORE cm/s ONSHORE -20 Figure 3. -The steady state abundance of fish larvae with distance aion^ the shelf. This abundance is expressed as a frac- tion of the number of larvae continuously being produced at the spawning site, Nq. There is a 80 km wide, stationary eddy at the shelf edge, inducing onshore and offshore flows of 20 cm/s. The longshore velocity Uq of the shelf water is 5 cm/s. Biological losses (f.() are set equal to zero. The dotted line shows the steady abundance of fish larvae with distance down the shelf when there is no eddy present. shows only the spatial distribution of larval density in the water moving down the shelf, as affected by flow convergences or divergences associated with the physics of the ring. The flux of larvae off the shelf (not shown) is given by V',, hN,^ in the regions where Vo is greater than zero and amounts in total to i Vq h N^^ dx ^ A hL N^e }k 66% of the flux into the domain (L^,, A^,, hY) at x = 0. Next, we shall see that the physics and biology ac- tually interact to produce a greater net impact than when either is considered separately. For this second model problem, we shall still use a steady onshore and offshore flow pattern, but now include the biological loss term and the time-depen- dence in the source function Nq. When the flow is off- shore or zero, the population distribution is given by N{x,t) = NS- T)e' (13a) whej-e the variable t measures the length of time necessary to reach the point x from the upstream edge of the domain. In general r is given by T = s ' rfo:' ., U{x) (13b) where U{x) can be found by integrating the mass conservation equation U{x) = f/o- \ dx —. i (13c) In the absence of ring-induced onshore-offshore flows (V,) = 0), however, t is just equal to xlU^^ and N{x,t) = iVo [t 1 ilL\ (13d) The population at any downstream point lags that at the origin by the travel time xIUq and has also decayed exponentially during its travel. This solution is an important base case for understanding the distributions in a spawned patch which has not been impacted by rings. When water is being drawn off the shelf, the along- shelf decay in concentration is again purely due to travel time, since the effect of losses off the shelf on the density is compensated for by the convergence. However the spatial density of the larvae is still noticeably altered by the offshore flow because the travel time necessary to reach any point is increased. This occurs because U is decreasing with i- as a result of the advection of water off the shelf (as shown in Equation (13c)). Since U is less than Uq, the travel time T in Equation (13b) is necessarily greater than that in the absence of the ring (xlU^^). We have sketched t(x) for the three possible signs of F,, in Figure 4a. These results suggest that there will be an enhanced spatial decay rate of larval density in the regions where the flow is offshore. 318 KLIKKL AND V\ KOBLliW SKI: WARM ('(IKK ( ;r 1 ,K STKKAM KINGS 100 o OL 2 T 2 (days) cr t- 0 \ + 1 dU for Vq < 0. (15) U U dx We have plotted these as functions of x in Figure 4b using /^ = 10""^ s""', U^) = 5 cm/s, Vq = ±20 cm/s, and Y = 200 km. With this value for fu, two-thirds of the larvae disappear from the population because of the various biological causes within 4 mo from hatching. Most values of ^x in the literature (e.g., Sissenwine et al. 1983) tend to be higher (see, however, Peterson 319 FISHERY BULLETIN: VOL. 83. NO. 3 and Wroblewski 1984), but it is important to remember that these also include the advective losses. We have therefore chosen a smaller value of ^u to reflect only biological processes; alternate values N of /i will be considered shortly. The graph shows the extreme situation where the inflow or outflow is uniform over the whole downstream distance. The decay rate with distance travelled is always increas- ed for offshore flow. For onshore flow, the decay rate can be reduced below the "no-ring" case but only very far downstream (x > 500 km) where the flow rate down the shelf is huge {U = 30 cm/s). Since on- shore flows of 20 cm/s over a 500 km stretch of shelf are not likely to occur, we can conclude that the spatial decay rate will be enhanced in both the regions of onshore flow and the areas with offshore flow. The net result, when a stream of larvae moving down the shelf and declining in density due to biolog- ical losses encounters a stationary eddy, can be calcu- lated by combining the result for offshore flow in the region 0 < x < Z> with the one for onshore flow in x > D where the ring is centered at the point x = D.ln Figure 5 we compare the solution without the eddy N exp -fXX (16) to that with the eddy exp(-MT) UiD) forO 0.10) and the coefficient of variation was low (CV = 4.2%). The ratio of the number of batch eggs to recruitment eggs was also independent of subsample location (P > 0.5). More- over, I also directly counted the total number of batch eggs in four of the females used in the above analysis; in each case estimated batch fecundity was within 10% of the true value. Field Observations of Spawning Frequency The frequency of spawning in a natural population of Menidia menidia was inferred from daily, direct field observations of mating. The Atlantic silverside spawns in large groups of fish that broadcast milt and eggs amongst vegetation in the upper intertidal zone of salt marshes within 1 or 2 h of the daylight high tide (Middaugh 1981; Middaugh et al. 1981; Conover and Kynard 1984). At such times, spawning can be easily observed. My observations were con- ducted at a major spawning site in Salem Harbor, MA. Daily assessments of spawning intensity were conducted throughout the spawning season by count- ing the number of aggregations of spawning fish sighted during high tide. Methodological details are provided in Conover and Kynard (1984). Laboratory Observations of Spawning Frequency and Egg Production Spawning frequency and egg production were also assessed by confining fish in laboratory tanks. A summary of the experimental procedure, described fully in Conover and Kynard (1984), follows. A large group of M. menidia were captured at Salem Harbor on 5 May 1979 and transported to the University of Massachusetts marine laboratory at Gloucester, MA. One female and two male fish were placed into each of the four 74 L tanks at room temperature on a natural photoperiod. Four males and four females were also placed into each of two circular plastic pools (diameter 1.5 m, depth 0.3 m). These pools were located outdoors. All fish were fed fresh, chop- ped seaworms (Nereis) and amphipods in excess of daily consumption. Spawning substrates consisted of a small tuft of synthetic aquarium filter floss, an- chored to the bottom of each tank or pool. The floss was checked several times daily for the presence of eggs. When eggs were discovered, the floss was replaced. All eggs were preserved and enumerated later. Eggs from each female were usually deposited in a distinct clump on the floss, providing a means for determining the number of females that had spawn- ed in the previous interval. The experiment was allowed to continue until spawning ceased (27 July). Length and weight of spawners was measured at the conclusion of testing. RESULTS AND DISCUSSION Gonadal Analyses In late fall, just prior to the offshore winter migra- tion (Conover and Murawski 1982), ovaries repre- sented about 1% of total body weight (Fig. 2) and contained only small (< 0.1 mm), clear, transparent eggs. Upon return of fish to the shore zone the following April, the ovarian GSI was about 4% and many opaque, white, immature ova (< 0.5 mm) were present. Of the 25 females captured on 6 May, 92% contained numerous immature ova and a clearly definable batch of maturing ova. The first female carrying ripe egges was collected on 12 May. The GSI peaked in both sexes on 25 May and declined thereafter until the end of July. The first fish in spent condition (no maturing egg class, recruitment eggs, if present, degenerating, GSI < 5%) was captured on 22 June. The proportion of spent fish was 23% on 334 CdXinKK: I'ATTKKXSIX KKCI Nni'I'V ( >K ATI.ANTH ' SIIAKKSIDK O < z o 28- 13 A. FEMALE 24- 43 32 20- 14 25 - 16 16- 12- 8- 1 10 > - - - i 4- 10 1 r 4< . — 1 - 8 1 =t^ 28- 24- 20- 16 12 8- 4- 27 18 10 + >. " I -f B. MALE 18 8 12 ^ >* I 111 I I I I > 17 16 29 A 12 25 6 22 6 13 26 OCT APR MAY JUN 1976 1977 JUL Figure 2.-Gonadosomatic index (gonad weight expressed as a percentage of total weight) for Menidia menidia collected during 1976-77 in Essex Bay, MA. The horizontal lines represent means, the vertical lines represent one standard deviation, and the sample size is given above the datum for each collection. 6-13 July and 100% on 26 July. Hence, the breeding season in Essex Bay began sometime after 6 May and was over by 26 July during 1977. The potential annual fecundity of M. menidia may be represented by the total number of eggs (recruit- ment + maturing ova) within females just prior to the breeding season (i.e., 6 May), if additional im- mature eggs are not continually added to the recruit- ment pool as the spawning season progresses. If this premise is true, then there should be a continuous decline in recruitment fecundity and total fecundity during the breeding season (although not necessarily in batch fecundity). Comparison of fecundities between sample dates was facilitated by the following observations. Total number of eggs per female was linearly related to ovary-free body weight (Fig. 3). Batch fecundity was also a simple linear function of ovary-free body weight (Fig. 4) and the rates of increase in batch fecundity, recruitment fecundity, and total fecundity with increase in female weight were generally similar among sample dates (i.e., regression slopes O CO o o UJ q: LiJ CD 14 12 10 8 6 4 MAY 6 • • Y = III3.4(X)-H25I9.5 n = 23 r = .78 < o tv^ 3456789 10 OVARY- FREE FEMALE WEIGHT(g) Figure 3. -Relation between total number of eggs (recruitment plus mature) and ovary-free female body weight for Atlantic silver- sides captured just prior to the beginning of the spawning season (ti May 1977) in Essex Bay, MA. differed little, t-test, P > 0.05). Correspondingly, relative batch fecundity, relative recruitment fecun- dity, and relative total fecundity (relative fecundity = no. eggs/g ovary-free body weight) were each in- dependent of body weight in nearly all tests (linear correlation, P > 0.05), suggesting that females of all sizes allocated about the same proportion of energy to reproduction. Hence, fecundity was adequately described and compared between dates if expressed as a proportion of ovary-free body weight, rather than as a function of weight. Batch fecundity, recruitment fecundity, and total fecundity (no. eggs/g ovary- free body weight) during the spawning season are presented in Figure 5. Three patterns are evident. First, total fecundity and recruitment fecundity monotonically declined (Fig. 5 A, B). Total fecundity was 1,609 ± 126 (95% C.L.) on 6 May and declined to 876 ± 177 by the second week of July (Fig. 5A) while recruitment fecundity was initially 1,430 ± 128 on 6 May and declined to 716 ± 164 in July (Fig. 5B). Second, batch fecundity differed significantly between sample dates, being maximal during the middle of the breeding season (266 ± 34 and 267 ± 23 on 6 and 22 June, respective- ly) and minimal at the beginning and end of the breeding season (179 ± 21 and 181 ± 28 on 6 May and 6-13 July, respectively; Fig. 5C). Third, many recruitment eggs remained in ovaries near the end of the spawning season (Fig. 5B) and most of these were probably resorbed soon thereafter because all females captured on 26 July contained only small (< 0.10 mm), transparent oocytes. Two females from the 6-13 July collection contained only immature eggs that appeared to be in a state of resorption and had no maturing egg group. 335 FISHF:RV BILLETIN: VOL. Ki. NO. .S , Y = 244 40{X)-33I.57 n-23 r = .83 iiJ E 3 Q Z) o UJ o < CD 2000 r ^ 1000 y^» • Y = 329.44(X)-3I4.36 n = 23 r=.83 i_/^M I 1 1 1 1 1 1- 3000 _ JUNE 22 • 2000 • • "^^ • -'^ ^» 1000 ^y^ • • ^ Y = 238 37(X) + I57 46 -U 1 1 n = 29 r = .65 1 1 1 1 1 2000 1000 JULY 6-13 i/- Y = l44.42{X) + 249.65 n=l4 r = .6l I I I I I 3456789 10 OVARY- FREE FEMALE WEIGHT(g) FiGi'RE 4. -Relation between batch fecunditv' (no. of eggs in the most advanced size class) and ovary-free female body weight for Atlantic silversides captured on four occasions during the 1977 breeding season in Essex Bay, MA. UJ x: "2 jQ < 3 I- o o UJ 000 500 0 ^ 1500 1000 5 5 q: _ q: ct 500 UJ ^ 0 CO o o _ 300 UJ £L liJ cn q: 1) ^ < >> 200 5 X) o a: J3 g IS" 100 D " E3 i B k23) (23) (29) (17) 1 1 MAY JUNE JULY Batch fecundity as estimated above assumes that the number of eggs in the most advanced mode is not reduced by atresia as they grow and are eventually shed. I noted few eggs which appeared to be atretic or in a process of resorption (except near the very end of the breeding season) and females which had recently spawned usually contained few. if any, residual ripe eggs. Similar observations were reported by Clark (1925) for the atherinid Leuresthes t£nuis. Moreover, if some eggs cease growing and are resorbed before reaching maturity, there should Figure 5. -Relative fecundity (no. eggs/g ovary-free female body weight) for Atlantic silversides captured on four (x-casions during the 1977 breeding season in Essex Bay. MA. The horizontal lines represent means, the vertical lines represent one standard devia- tion, and the rectangles represent 95% C.L. Sample sizes are in parentheses. All fish were spent on 26 July. A) ToUil number of eggs. B) Number of recruitment eggs (recruitment fecundity)- C) Number of mature eggs (batch fecundity). be a negative correlation between batch size and the mean diameter of eggs in the batch (i.e., no. batch eggs/g ovary-free body weight should generally be 336 CONOVER: PATTERNS IN FECUNDITY OF ATLANTIC SILVERSIDE lower in fish nearly ready to spawn than in fish where batch eggs are still maturing). Linear correla- tions of batch size and mean diameter of the maturing egg batch for each of the four dates on which fecundity was measured were nonsignificant (P > 0.05), suggesting that the number of eggs in a batch does not decline much as the oocytes grow to maturity. Assuming that the recruitment pool of immature eggs is fully formed prior to the breeding season, an estimate of the actual number of eggs produced an- nually can be derived from the above data. Because recruitment eggs remain at the season's end, actual egg production is best represented by the total number of eggs present just prior to the beginning of the spawning season minus the number of recruit- ment eggs retained when the spawning season ends. For the above data, this provides a value of (1,609 ± 126) - (716 ± 164) = 893 ± 197 eggs/g ovary-free body weight (±95% C.L.). The mean body weight of females during the breeding season was 5.6 g so that the average female would have spawned about 5,001 eggs in a season. The mean batch fecundity over the spawning season was 223 eggs/g ovary-free body weight or about 1,249 eggs/ female. Hence, if the ini- tial assumption is correct, the average female must spawn about four times during the breeding season. phic variation in M. menidia also support this conten- tion (Johnson 1975). Daily observations of the number of spawning ag- gregations sighted during high tide at Salem Harbor in 1979 showed that populations of Atlantic silver- sides breed on a semilunar periodicity coinciding with new and full moons (see figure 2 in Conover and Kynard 1984). Middaugh (1981) has reported similar observations based on a 3-yr study of populations in South Carolina. Within each semilunar spawning period of 1979 in Salem Harbor, the majority of spawning (40-90%) occurred on a single day, sug- gesting that females spawn, at most, once per semi- lunar cycle. This conclusion is further supported by the observation that sex ratios in spawning aggrega- tions were highly male biased, whereas during non- spawning intervals, the sex ratio was near unity (Conover 1984). The sex ratio data is explained if females spawn, at most, once per cycle, while males spawn each day of a spawning period. During 1979 in Salem Harbor, there were five semilunar spawning periods, although the first in late April was very light. In Essex Bay during 1977, there were four semilunar phases during the period defined as the breeding season. Hence, estimated spawning frequency based on gonadal analysis and direct observations of spawning fish agreed closely. Frequency of Spawning in Nature To test the prediction of spawning frequency derived above, spawning periodicity was determined from direct, visual observation of breeding events in the field. In 1978, 1 discovered a large spawning site in Salem Harbor, MA, where numerous eggs of M. menidia were deposited amongst mats of filamen- tous algae in the upper intertidal zone. Salem Harbor is located 20 km southwest of Essex Bay; however, individuals in each area are likely members of the same population because extensive mixing occurs during the offshore winter migration (Conover and Murawski 1982). Electrophoretic studies of geogra- Spawning Frequency and Egg Production in the Laboratory Females held in laboratory tanks, whether housed individually indoors or outdoors in groups where natural day and night illumination was present, did not maintain a fortnightly spawning periodicity. In- stead, spawning occurred much more frequently: the interval between spawnings averaged about 4 d/female (Table 1; see also figure 4 in Conover and Kynard 1984). Batch fecundity averaged 99-187 eggs/g ovary-free body weight among different tanks. Total egg production averaged 1,425-3,375 eggs/g ovary-free body weight. Hence, the experi- Table 1. — Egg production by Menidia menidia on unlimited food rations in the laboratory. Field data are also provided for connparison. Mean 9 Days from No. of Days No. eggs/g Total Total eggs/g body 1st to egg between 9 body eggs 9 body No. of weight last batches spawn weight per per 9 weight females (g) spawning per 9 per 9 egg batch (no.) (no.) Pool 1 4 6.32 63 15.25 3.9 99.0 9,551 1,511 Pool 2 4 5.82 51 11.75 3.7 121.2 8,299 1,426 9 A 1 3.9 72 20.0 3.6 169.5 13,218 3,389 9C 1 3.6 '28 '7.0 4.0 187.0 '4,710 '1,308 Field — — 60-75 4.0 14-15 180-266 5,000 893 ±197 'Low values reflect the fact that 9 C died before the conclusion of the experiment. 337 FlSHKKVBrLLF.TI\;V()l., h:1N().3 mental fish responded to laboratory conditions by reducing batch fecundity somewhat, but spawning more frequently, and thereby producing about twice the number of eggs as in nature (Table 1). The daily rate of egg production was 24-47 eggs/g female body weight per d in the laboratory, but averaged about 14 eggs/g female body weight per d in the field. At the termination of the experiment, four of the eight females in the outdoor pools were spent, three contained only recruitment eggs, and one had both recruitment and maturing eggs. Female A died of unknown causes after its last spawning on 22 July. Female C also died (9 June) before cessation of spawning by jumping out of the tank. The total weight specific egg production for the experimental fish was generally within the range of total eggs available prior to the beginning of the breeding season. The one exception was female A which produced about twice the total number of eggs that a fish of its size should have had available at the beginning of the spawning season (see Figure 3). Hence, under certain laboratory conditions, females may be capable of producing new oocytes from oogonia during the breeding season, as recruitment eggs become depleted. These laboratory observa- tions show that the reproductive patterns of egg maturation and spawning which are highly synchro- nized with and influenced by environmental factors in the field, easily become disrupted when individuals are removed from their natural habitat. CONCLUSIONS This study indicates that annual fecundity in Menidia menidia, and perhaps certain other fishes, can be estimated from the difference between total number of eggs (recruitment plus maturing) prior to the breeding season and recruitment egg retention near the end of the breeding season. Dividing the estimated total number of eggs shed per female by mean batch fecundity provided an estimate of spawning frequency. The accuracy of this value for spawning frequency was tested and found to agree closely with the spawning fre(]uency inferred from direct field observations of breeding fish. Previous estimates of the fecundity of M menidia were about 3-10 times less than that rejwrted here because spawning frequency was not determined (Bayliff 1950; Jessop 1983). The studies of Hunter and his coworkers on northern anchovy, Engraulis mordax (Hunter and Goldberg 1980; Hunter and Macewicz 1980; Hunter and Leong 1981), and DeMartini and Fountain (1981) on queenfish, Seriphus politus, have amply demonstrated that estimates of annual fecun- dity can be in error by over an order of magnitude when multiple spawning is ignored. The estimation of fecundity from the difference be- tween total prespawning fecundity and recruitment egg retention is dependent on the assumption that new oocytes are not simultaneously produced from oogonia and added to the reservoir of recruitment eggs as mature eggs are spawned. Agreement be- tween predicted and observed spawning frequency suggests that this may be true in M. menidia. Many more recruitment eggs were present in ovaries at the beginning of the spawning season than were ac- tually spawned in nature. Evidently, the recruitment egg pool is largely formed before the breeding season in Menidia, as is believed for some other seasonal spawners (Tokarz 1978; Jones 1978; Baggerman 1980). However, the generality of this pattern in other multiple spawning temperate or tropical fishes is not clear. Clark (1925) noted that the relative abundance of mature, intermediate, and immature eggs in Leuresthes tenuis was relatively constant during the breeding season and concluded from this that new oocytes must be continuously pro- duced to replenish those spawned. Taylor and DiMichele (1980) reached a similar conclusion based on the relative abundance of different developmental stages of oocytes during the semilunar spawning cycle of Fundulus heteroclitiis. However, analyses based on relative proportions do not take into ac- count that gonad weight (GSI) generally declines as the season progresses (e.g., Fig. 2) and that number of eggs in the most advanced mode is not necessarily constant during the breeding season. Comparison of the relative abundance of egg sizes from sections of an ovary may not reflect changes in absolute number. For example, the relative abundance of recruitment eggs in M. menidia during 1977 was 0.88 on 6 May, 0.78 on 6 June, 0.76 on 22 June, and 0.79 during 6-13 July. Hence, the relative proportion of recruitment eggs did not consistently decline dur- ing the breeding season even though the absolute number of eggs declined by a factor of 2.4. In any event, too little is known about patterns of oocyte growth in fishes to recommend that the annual fecundity of multiple spawners can generally be determined by monitoring the decline in the standing stock of ova as was done here. For instance, in tropical species that breed most of the year recruit- ment eggs may be produced continuously. Whenever possible, the results of several different api)roaches to estimating fecundity should be compared. The results of the laboratory study demonstrated that M. menidia is physiologically capable of spawn- ing much more frequently and over a shorter interval 338 rONO\ KK I'AITKKNS IN KKdNDI TV I )K ATLANTIC SIlAKkSlDK than normally occurs in the field. The reasons for the higher spawning frequency an(i cumulative egg pro- duction for fish in captivity are probably several. Fecundity may have been increased because ration size was unlimited. Fecundity is dependent on the food supply in many species (Wootton 1979). In the stickleback, Gasterosteus aculeatus, (Wootton 1977) and the convict cichlid, Cichlasoma nigrofasciatum, (Townsend and Wootton 1984) experimental studies have demonstrated that the number of spawnings was positively related to food ration and the interval betu'een spawnings was inversely related to ration. In my experiments on M, nienidia, spawning fre- quency may also have been increased beyond that in nature due to the continuous availability of appro- priate spawning substrates and lack of tidal spawn- ing cues in the laboratory. Conover and Kynard (1984) noted that both marine and freshwater populations of Menidia spp. tend to spawn during midmorning, and speculated that spawning in nature may be restricted by the fact that suitable spawning substrates are covered by high tide during midmorn- ing only at fortnightly intervals. Correspondingly, a lacustrine population of M. beryllina spawns daily at midmorning (Hubbs 1976). Hence, in the laboratory where tidal cues are removed, spawning substrates are continuously available, and food is abundant, M. menidia responded by spawning more frequently. The high egg production of female A also suggests that if the supply of recruitment eggs is exhausted, new recruitment eggs can be formed. It is clear that estimates of fecundity in natural populations of multiple spawners based on laboratory studies alone should be interpreted with caution. Many aspects of the fecundity and spawning periodicity of M. menidia are paralleled in a west coast atherinid, Leuresthes tenuis. The California grunion has a well-known semilunar spawning cycle (Walker 1952). Clark (1925) conducted a detailed study of egg diameter frequencies in ovaries of L. tenuis and concluded that each female spawns once about every 15 d. Batch fecundity was very similar to that reported here for M. menidia. Although Clark measured batch fecundity in only a few indivi- duals, a 118 mm grunion contained 1,613 ova. I calculate that a 118 mm Atlantic silverside would be expected to have 1,704 ripening eggs during the mid- dle of the breeding season. Clark also found reten- tion of recruitment eggs at the end of the breeding season and presented histological evidence that re- tained eggs were being resorbed. Based on my estimate of the average annual fecun- dity of M. menidia (893 ± 197 eggs/g ovary-free body weight) and the wet weight of ripe eggs (0.8 g/1,000 eggs), an Atlantic silverside produces nearly 0.7 of its body weight in eggs during the breeding season in nature. In the laboratory, females pro- duced 1.1-2.7 times their body weight in eggs. Studies of other multi[)le spawners have yielded similar results. DeMartini and Fountain (1981) esti- mated that the queenfish could spawn about 114% of its body weight in a year. Experiments on several species of cyprinids indicate that they are capable of spawning 0.7 to 6.8 times the volume of the female in eggs, at least in the laboratory ((lale and Gale 1977; (iale and Buynak 1978, 1982; Gale 1983). Hubbs (1976) estimated that a freshwater population of Menidia beryllina spawned 6-8 times female weight in eggs, although his assumption that each female spawns daily throughout the length of the breeding season needs further documentation. Subseasonal trends in batch fecundity among multiple spawners have been examined by few inves- tigators. If trends in batch fecundity within the breeding season are the adaptive result of natural selection, then periods of maximum batch fecundity should reflect the period when the probability of off- spring survival is greatest. On the other hand, trends in batch fecundity could simply result from varying food conditions for adults. Three general relation- ships between the batch fecundity and the time of the breeding season have emerged from field studies with which I am familiar. These include 1) constant batch fecundity during the breeding season (Fig. 6, curve A), 2) a concave downward relation between batch fecundity and the breeding season (Fig. 6, curve B), and 3) a constant decline in batch fecundity during the breeding season (Fig. 6, curve C). Con- stant fecundity (curve A) might be expected where the optimal environmental conditions for reproduc- tion and offspring survival are constant or vary un- predictably during the breeding season. This pattern has been found in the queenfish (DeMartini and Fountain 1981), a pelagic spawner of the western North American coast where aperiodic upwelling events produce unpredictable variations in plankton productivity and potential larval survival (Lasker 1978). When seasonal environmental conditions change in a predictable manner, there may be an op- timal period for reproduction that occurs at roughly the same time each year, and batch fecundity would be expected to be maximal at that time (curve B). In M. menidia, the relation between batch fecundity and the breeding season was concave downward, suggesting that reproductive success is maximal dur- ing the middle of the breeding season. There is some independent evidence to support this hypothesis. Winter mortality during the offshore migration is 339 FISHERY BULLETIN: VOL. 83, NO. 3 >- O 2 O X u < CD 4- BEGINNING END BREEDING SEASON Figure 6. -Three hypothetical relationships between batch fecun- dity and time of the breeding season that have empirical support in the literature. A) Constant batch fecundity. B) Batch fecundity maximal during the middle of the breeding season. C) Batch fecun- dity maximal at the beginning of the breeding season and declining continuously thereafter. strongly size- selective in M. menidia (Conover and Ross 1982; Conover 1984): the largest young-of-the- year have the highest probability of surviving. Hence, there should be selection pressure to breed as early in the spring as physical conditions (such as temperature) permit, and perhaps before conditions are optimal. Any offspring that managed to survive early in the breeding season will ultimately benefit from having a longer growing season. Conversely, towards the end of the breeding season, energy placed into reproduction becomes wasted because these offspring will have almost no chance of grow- ing to a size that will permit winter survival. It follows that somewhere in the middle of the breeding season will be the optimal period for reproduction. Declining batch fecundity during the breeding season (curve C) has been reported for a population of the common mummichog, Funduhis heteroclitus, where batch fecundity was greatest at the beginning of the breeding season and became progressively less thereafter (Kneib and Stiven 1978). A continuous decline in batch fecundity may evolve when the value of putting energy into current reproduction, as op- posed to somatic growth, declines continuously as the breeding season progresses. Although few data are now available for comparing the subseasonal pat- terns of batch fecundity in multiple spawning fishes, such information may eventually prove useful in understanding the general reproductive strategies of fishes. ACKNOWLEDGMENTS I thank the staff of the University of Massachu- setts Marine Station for logistic support. D. Chevalier assisted in maintaining the laboratory fish and in counting eggs. F. Sutter helped with field sampling. The comments of two anonymous reviewers improved the clarity of the manuscript. During part of the this study, I received support from the Massachusetts Cooperative Fishery Research Unit, which is jointly funded by the Massachusetts Division of Marine Fisheries, the Massachusetts Division of Fish and Wildlife, the University of Massachusetts, and the U.S. Fish and Wildlife Service. LITERATURE CITED Bagenal, T. B. 1967. A short review of fish fecundity. In S. D. Gerking (editor), The biological basis of freshwater fish production, p. 89-111. Blackwell Sci. Publ., Oxf., Engl. Bagenal, T. B., and E. Braum. 1971. Eggs and early life history. In W. E. Ricker (editor), Methods for assessment of fish production in fresh waters, p. 166-198. IBP (Int. Biol. Programme) Handb. 3, 2d ed.; Blackwell Sci. Publ., Oxf., Engl. Baggerman, B. 1980. Photoperiodic and endogenous control of the annual reproductive cycle in teleost fishes. In M. A. Ali (editor). Environmental physiology of fishes, p. 533-567. Plenum Press, N.Y. Ball, J. N. 1960. Reproduction in female bony fishes. Symp. Zool. See. Lond. 1:105-135. Bayliff, W. H., Jr 1950. The life history of the silverside Menidia menidia (Linnaeus). Chesapeake Biol. Lab. Publ. 90, 27 p. Clark, F. N. 1925. The life history of Leuresthes tenuis, an atherine fish with tide controlled spawning habits. Calif. Fish Game Comm., Fish Bull. 10, 51 p. Conover, D. 0. 1984. Adaptive significance of temperature-dependent sex determination in a fish. Am. Nat. 123:297-313. Conover, D. 0., and B. E. Kynard. 1984. Field and laboratory observations of spawning periodi- city and behavior of a northern population of the Atlantic silverside, Menidia menidia (Pisces: Atherinidae). Environ. Biol. Fishes 11:161-171. Conover, D. 0., and S. A. Murawski. 1982. Offshore winter migration of the Atlantic silverside, Menidia menidia. Fish. Bull., U.S. 80:145-150. Conover, D. 0., and M. R. Ross. 1982. Patterns in seasonal abundance, growth and biomass of the Atlantic silverside, Menidia menidia, in a New England estuary. B^stuaries 5:275-286. DeMartini, E. E., and R. K. Fountain. 1981. Ovarian cycling frequency and batch fecundity in the 340 OONOVEK: PATTERNS IN FECUNDITY OK ATLANTIC SILVKRSIDE queenfish, Seriphits politus: Attributes representative of serial spawning fishes. Fish. Bull.. I'.S. 79:547-560. Gale. W. F. 1983. Fecunditj' and spawning frequency of caged bluntnose minnows- fractional spawners. Trans. Am. Fish. Sex-. 112: 398-402. Gale, W. F., and G. Buynak. 1978. Spawning frequency and fecundity of satinfin shiner (Notropis nnalostamis)-a fractional, crevice spawner. Trans. Am. Fish. Soc. 107:460-463. 1982. Fecundity and spawning frequency of the fathead minnow -a fractional spawner. Trans. Am. Fish. Soc. Ill: 35-40. Gale, W. F., and C. A. Gale. 1977. Spawning habits of spotfin shiner {Notntpis spilotenis) -a fractional, crevice spawner. Trans. Am. Fish. Soc. 106: 170-177. Hempel, G. 1979. Early life history of marine fish. Univ. Wash. Press, Seattle, 70 p. HiSLOP, J. R. G., A. P. ROBB, AND J. A. Gauld. 1978. Observations on effects of feeding level on growth and reproduction in haddock, Melarwgrammus aeglefinus (L.) in captivity. J. Fish Biol. 13:85-98. HUBBS, C. 1976. The diel reproductive pattern and fecundity of Menidm audens. Copeia 1976:386-388. Hunter, J. R., and S. R. Goldberg. 1980. Spawning incidence and batch fecundity in northern anchovy, Engraulis mordax. Fish. Bull., U.S. 77:641-652. Hunter, J. R., and R. Leong. 1981. The spawning energetics of female northern anchovy, Engraulis mordax. Fish. Bull., U.S. 79:215-230. Hunter, J. R., and B. J. Macewicz. 1980. Sexual maturity, batch fecundity, spawning frequency, and temporal pattern of spawning for the northern anchovy, Engrmdis mordax. during the 1979 spawning season. Calif. Coop. Oceanic Fish. Invest. Rep. 21:139-149. Jessop, B. M. 1983. Aspects of the life historj' of the Atlantic silverside (Menidia menidia) of the Annapolis River, Nova Scotia. Can. Manuscr. Rep. Fish. Aquat. Sci. 1694, 41 p. Johnson, M. S. 1975. Biochemical systematics of the atherinid genus Meni- dia. Copeia 1975:662-691. Jones, R. E. 1978. Ovarian cycles in nonmammalian vertebrates. In R. E. Jones (editor), The vertebrate ovary, p. 731-762. Plenum Press, N.Y. Kneib, R. T., and a. E. Stiven. 1978. Growth, reproduction, and feeding oi Fundulus hetero- clitvs (L.) on a North Carolina salt marsh. J. Exp. Mar. Biol. Ecol. 31:121-140. Lasker, R. 1978. The relation between oceanographic conditions and lar- val anchovy food in the California current: identification of factors contributing to recruitment failure. Rapp. P. -v. R6un. Cons. int. Explor. Mer 173:212-230. MiDDAUGH, D. P. 1981. Reproductive ecology and spawning peritxlicity of the Atlantic silverside, Memidm menidia (Pisces: Atherinidae). Copeia 1981:766-776. MiDDAUGH, D. P., G. I. Scott, and J. M. Dean. 1981. Reproductive behavior of the Atlantic silverside, Menir dia menidia (Pisces, Atherinidae). Environ. Biol. Fishes 6: 269-276. Snyder, D. E. 1983. Fish eggs and larvae. In L. A. Nielsen and D. L. Johnson (editors). Fisheries techniques, p. 165-197. Am. Fish. Soc, Bethesda, MD. Taylor, M. H., and L. DiMichele. 1980. Ovarian changes during the lunar spawning cycle of Fundulus heteroclitus. Copeia 1980:118-125. TOKARZ, R. R. 1978. Oogonial proliferation, oogenesis, and folliculogenesis in nonmammalian vertebrates. In R. E. Jones (editor). The vertebrate ovary, p. 145-179. Plenum Press, N.Y. TOWNSHEND, T. J., AND R. J. WOOTTON. 1984. Effects of food supply on the reproduction of the convict cichlid. Cichlasoma nigrofasciatum. J. Fish Biol. 24:91-104. Walker, B. W. 1952. A guide to the grunion. Calif. Fish Game 38:409-420. WOOTTON, R. J. 1973. The effect of size of food ration on egg pnxluction in the female three-spined stickleback, Gasterosteus acuLeatvs L. J. Fish Biol. 5:89-96. 1977. Effect of food limitation during the breeding season on the size, body components and egg production of female sticklebacks {Gasterosteus aeuleatus). J. Anim. Ecol. 46:823- 834. 1979. Energy costs of egg production and environmental determinants of fecundit>' in teleost fishes. Symp. Zool. Soc. Lond. 44:133-159. 341 PARASITES OF SKIPJACK TUNA, KATSUWONUS PELAMIS: FISHERY IMPLICATIONS R. J. G. Lester,' A. Barnes,^ and G. Habib^ ABSTRACT The numbers of 26 types of parasites were counted in 878 fish, of which all but 3 were from 1 4 areas in the Pacific. Data from the 22 most reliable parasites gave no evidence of discrete stocks of skipjack tuna in the Pacific, either when analyzed singly or when usinjj combinations of parasites in multivariate analyses. New Zealand fish carried many tropical parasites, particularly didymozoids, in numbers similar to fish caught in the tropics, indicating;: that the bulk of these fish had recently migrated from the tropics. The number oiTen- tirriilnria nnyphaenac. a larval tapeworm, was positively correlated to fish size in the tropics. In New Zealand, however, fish over 5.5 cm carried about the same number of T. coryphaenxw as fish 45 to 55 cm, suggesting they had left the tropics when they were 45 to 55 cm and had not returned. Analysis of the numbers of parasites from particular schools suggested that school members stayed together for several weeks i)ut not for life. The use of parasites to delineate stocks for manage- ment purposes is a well-established technique. For a comprehensive review of the many examples see MacKenzie (1983). The skipjack tuna, Katsuwonus pelamis, is one of the most valuable fishery resources of the central and western Pacific. At least 50 species of parasites have been reported from it. The distribution of only one, the hemiuroid digenean Hirudinella ventricosa, has previously been investigated. In the Atlantic, Watertor (1973) found it in 7% of skipjack tuna off West Africa, 40% off Brazil, and < 1% off Florida. In the Pacific, Nakamura and Yuen (1961) found it in 21% of skipjack tuna off the Marquesas and 34% of fish from Hawaii. Sindermann (1961) pointed out that analyzing the distributions of combinations of parasites may provide more information than the ex- amination of individual species. That, in general, has been our approach here. In addition, school-school variation in parasite numbers was studied to determine how long schools stayed together, and secondarily to evaluate the degree of permanence of the parasites. MATERIALS AND METHODS Of the 878 fish dissected, 386 were collected by the Hatsutori Mam on charter to the South Pacific Com- 'Department of Parasitology, University of Queensland, St. Lucia, Brisbane, Australia 4067. ^Faculty of Science, University of Queensland, St. Lucia, Brisbane, Australia 4067. ^Fisheries Research Division, New Zealand; present address: Southpac Fisheries Consultants, P.O. Box 7230, Auckland 1, New Zealand. mission (SPC), 246 by the New Zealand Ministry of Agriculture and Fisheries (NZ), and the remainder by other governments and fishing companies (see Acknowledgments). Fish were obtained from 15 areas (Fig. 1, Table 1). Gills and viscera were frozen and flown to Bris- bane for dissection. The SPC and NZ fisheries offi- cers sampled 5 fish/school from a maximum of 3 schools/d. Commercial companies were unable to sample from individual schools and usually supplied the head and the anterior ventral body, removed from frozen fish by a single slanting cut using a band saw. Fork length, if not supplied, was calculated Table 1.— Sources of fish dissected. Avg. No. length Area Date fish (cnn) A Palau. Helen R. Aug. 1980 35 41 B Ponape July 1980 45 59 C Papua New G uinea June 1981 30 50 D Papua New G uinea Nov. 1981 60 41 E Solomon Is. June 1980 30 46 F Coral Sea Jan. 1982 19 57 G Fiji Feb., Mar., Apr.. May 1980 100 50 H Norfolk Is. Mar. 1980 21 57 1 New South Wales Jan. 1981 103 47 J New Zealand, west Mar. 1980; Jan.. Feb. 1982 69 52 K New Zealand, east Jan. 1980; Jan., Feb. 1982 163 49 L Marquesas Dec. 1979; Jan. 1980 150 47 M California Aug. 1981 30 47 N Ecuador Jan. 1982 20 48 0 Atlantic Mar. 1981 3 50 (Puerto Rico) Manuscript accepted November 1984. FISHERY BULLETIN; VOL. 83, NO. 3, 1985. 343 FISHERY BULLETIN: VOL. 83. NO. 3 0 r> § o O \^ ^ ' "^^ J . 0 +3 < CN 7 — ' 'vrty*^ II - r^ Z /^^^Y~^-^ 0 ^^^^ Vv? \ i\ .-' fc"' J / /^ 3 II z .is c \ J^- — ' - -■— ^— ^-^ 11/ i---7 t. n / 0 \ / 1 ^"^^"""--^ 1 1 Ja \ y \ \ ^^"""""X^ 1 13 J^ 1 / / "-^^ / 0 5: ■ / / II in T3 C 1 / iS . — ■ — ~__^ / CO ~^~^^^^~-J / "i — ~--^ /o en ^^^CD 3 / 1 ^ S II 1 J -^ _J k" 1 and Eas ico). -1 — ■— — "~~ — 15 OJ 1 O; 0 N] iS s- '^' S Si •• ■ £ 3 Z ^ » i • • 0 1 • 0 ' - (O < II II • ^ ^ ' . ^ ■Hi l--g ^ . • g iS ^ 15 c= 5 • S 0 js Ed £ S - -S N 00 ^-— -^ .-t; CO s 0 A — 1 W^ 2 ''^ g.W 11 0 sf '^ ^ 1 '^-> P\> Z^ C to 1 \ "O -rl Qi 1 \ 1 a> 3 -3 . X °o > 0 ^ »\ :=■ V y \Jf were r apua N w Sout • , ^— — -"''^^ I \ -C Cm 01 p~~~~--~ LjJ* '_t --T'''''^ \ 1— 1 \ ,^ « 11^ ^•'•' u_ ^ 1. -^\\^ "Q Jl 1 m ^ 1 f^ \ w^ Ic T3 .- '^ OS C M '.■ f^ \A^\f\ ^s, A to 0, from i = Panape; C = Norfolk Isla / ~ \ / \ \ \ /^ .■5 . 1 ^ <*- f tV^f 1/ \ \ \^ 15: Ree Fiji; 1 ^ 0^.* ^ \ H^~-^ \ V^^ J= S II ^^ <-*! 1 0 " §3=^-"^^'^ CURE l.-T Palau, Hel( iral Sea; G / ■cr^r-^ — 0 Ci. II 0 344 LESTKK KT Al..; I'AKASITKS OK SKUMACK Tl'NA from head length using the formula 7.8 + 2.75 x (head length) for heads under 14.5 cm and - 1.7 + 3.3 X (head length) for larger heads (from measure- ments of 80 and 83 fish, respectively). Prior to dissection, fish were thawed overnight at 6°C. In general, all viscera parasites were counted whereas gill parasites were counted on one side only and the numbers doubled in the final tables. A didymozoid capsule was counted as one parasite though most contained two individuals. Representative parasites were fixed and stored in 10% Formalin"* except for nematodes which were fixed and stored in 70% alcohol. An additional set of data on the abundance of the larval cestode Tentacularia coryphaenae was col- lected at sea by SPC and NZ fisheries officers. They recorded the number of Tentacularia visible through the peritoneum in the wall of the body cavity of 1,529 fish. Besides some summary statistics, two types of statistical analysis were done: 1) investigation into the similarities and dissimilarities of the parasite fauna between the various areas sampled, and 2) a study of school integrity. The similarities and dissimilarities between areas were examined using a series of cluster analyses and multivariate canonical analyses (Mardia et al. 1979). Strictly speaking, canonical analyses require data which are normally distributed and which have a common variance. However, the frequency distribu- tions of the parasites were not normal. They showed considerable differences from one parasite to ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service. NOAA. another and most appeared to have two components: one which could be adequately approximated by a negative binomial distribution; and a second compo nent consisting of a disproportionately large zero category, presumably arising because some schools had not been exposed to infection. Precise trans- formations to normalize the data would thus have been complex and of doubtful accuracy considering the small size of the samples from each school. A single transformation for all species was therefore used: the natural logarithm of the number of parasites plus 1.0. To avoid possible biases due to associations be- tween parasite numbers and fish length, such as that shown in Figure 2, the transformed counts were then adjusted for fish length. This was done for each species by regressing log (parasite number + 1.0) on fish length, for all Pacific tropical fish (489), to esti- mate the magnitude of any relationship. This was used to adjust the transformed parasite numbers, ex- cept where this was zero, to that expected for a fish of a standard length of 50 cm. (This length was very close to the overall mean length of the fish.) The method could not be trusted to eliminate all effects of length, so, as an added safeguard, only fish 39.5 to 57.5 cm were used in the multivariate analyses (83% of the total). These are likely to have been 1 yr old (Uchiyama and Struhsaker 1981; Wankowski 1981). In a few instances a parasite was absent from all fish in one area. To allow matrix inversion in the canonical variate analyses, a random number be- tween -0.005 and -1-0.005 was added to the data. This did not influence the outcome. The results of the canonical variate analyses were displayed graphi- cally as plots of the first versus the second canonical 25- 20 15 10 5- 30 \ Total tropics (1017 fish) k New Zealand (512 fish) AO 50 60 Fish length, cnn Figure 2. -Relationship between number of T. coryphaenae and fish lenjrth. Mean ±2 SE. Each mean from minimum of 19 fish. In the tropics the number increased with length but this was not reflected in the New Zealand samples. 345 FISHERY BULLETIN: VOL. 83, NO. 3 axes. Confidence limits (95%) for the positions of dif- ferent areas on these plots are presented as circles with radius equal to the square root of 5.99/number of fish in sample (Mardia et al. 1979). Analyses on the same combinations of parasites were also done by calculating minimum spanning trees (Gower and Digby 1981), and dendrograms from nearest neighbor and centroid cluster analyses (Clifford and Stephenson 1975), basing similarity measures on logarithms of area means. Areas were grouped in a similar way by all methods. Using clustering algorithms which either ignored or allow- ed for matches between areas where parasites were not recorded did not significantly influence results. For these reasons, and because only canonical variate analysis provided some measure of reliabili- ty for its conclusions (confidence rings), only the results of the canonical analyses are presented below. School integrity was examined by comparing the variability in parasite numbers per fish between schools, to that within schools, for the two areas (Marquesas and east New Zealand) where the largest numbers of schools were sampled. This showed which parasites were strongly linked to schools, and also allowed tentative estimation of the length of time schools remained intact. In theory, for parasites to show strong school associations two conditions need to be met: the parasite must heavily infect some schools and not others, and its life span in the fish must be equal to or shorter than the life of the school. Parasites which showed strong school-school associa- tion were therefore likely to be shorter lived than those not showing such associations, and other evidence being equal, were considered less reliable as population markers than related species. Two methods were used to compare within and be- tween school variability in each of the two areas. First, a series of univariate analyses of variance of log (parasite numbers -f- 1.0) were done to calculate the ratio of between school to within school variances. The magnitude of these ratios, and the corresponding probabilities that they do not differ from 1.0, were interpreted as measures of school in- tegrity. A limitation of this method was that the data were only approximately normally distributed, par- ticularly for rare parasites, and thus the derived pro- babilities were also approximations. The second method, a median test, was based on the binomial distribution. The number of parasites of a particular species in each fish was transformed to a zero if it was less than or equal to the median number per fish for the area, and to a one otherwise. The zeros and ones of each school were then considered as a binomial sample. If these samples showed evidence of greater variation than expected by chance (i.e., too many schools with nearly all zeros or nearly all ones), then the schools differed with respect to the distribution of the parasite. A statistic, approximately distributed as a x^ random variable, was calculated using GLIM (Baker and Nelder 1978) to determine whether the binomial samples showed evidence of differences. Its associated probability was used as a measure of school integrity. The method had the useful property of being independent of the distribution of parasite numbers. For parasites with a median per fish of <1, the test was based on the presence or absence of the parasite, though ob- viously the rarer the parasite the less sensitive the test. It is possible that some schools were sampled twice. If this did happen, the results of both methods err on the conservative side. Only those species that gave consistent results by both methods were used to draw conclusions about school integrity. RESULTS Evaluation of Parasite Species Information was collected on 26 different types of parasites (species or species complexes) from 15 areas. A summary of the raw data unadjusted for fish length is given in Table 2. The parasite species were evaluated for their prob- able longevity on or in skipjack tuna. For them to be useful as markers they needed to be relatively long- lived, preferably surviving for the life of the fish. Nothing was known specifically about their longevity in skipjack tuna, though data were available on related forms (Table 3). In general, intestinal lumen dwellers appear to be more easily lost than larval forms encapsulated in the tissues. The 26 skipjack tuna parasites were divided into four groups, those considered "temporary", "semi-permanent", and "permanent", and those not used at all. Four parasites were not used in any analyses. Two of the nematodes, Ctena.searophis sp. and Spinitec- tus sp. (Nos. 23 and 24 in Table 2), were found in the gut of virtually every fish in which they were sought, from every area. Their small size meant that the number recovered was a function of the time spent searching. They were only counted in every fifth fish, as were the two larval cestodes from the large intes- tine, Scolex polymorphus (large) and S. polymorphus (small) (Nos. 25 and 26). Counting these was time consuming, their apparent abundance may have been inversely related to the state of preservation of 346 LESTER ET AL.: PARASITES OK SKllMACK TUNA Table 2 —Average numbers of parasites per fish in all skipjack tuna (878) from the areas listed in Table 1, unadjusted for length. The last column gives the correlation coefficient (r) for length against log (parasite number + 1) for Pacific tropical fish. No. Parasites B H I K M N 1 Caligus spp. 5 17 5 5 10 36 4 3 3 3 1 7 0 1 7 0.37 2 Didymocylindms filiformis 16 5 2 3 4 4 7 3 6 8 10 4 4 3 10 -0 14 3 Didymocylindrus simplex 16 7 4 6 13 11 14 12 18 26 18 14 15 6 3 -0.08 4 Didymoproblema fusitorme 4 1 0 1 1 1 4 1 4 2 3 3 3 1 1 -0.06 5 Lobatozoum multisacculatum 0.1 0.1 0.0 0.0 0.0 0.0 0.1 0.5 0.0 0.1 0.4 0.2 0.3 0.1 0.0 0.03 6 Syncoelium filiferum 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 13.1 0.0 0.0 0.0 0.0 0.0 0.0 — 7 Philometra sp. 1 3 4 29 5 P' 2 2 3 3 1 6 6 1 8 002 8 Anisakis type 1 1.0 0.2 2.7 1.0 0.7 1.6 0.2 0.5 0.9 2.1 1.5 0.6 0.2 0.1 2.7 0.13 9 Anisakis type II 0.2 0.0 0.0 0.1 0.0 0.1 0.0 0.4 0.1 0.8 0.2 0.0 0.4 1.2 2.3 -0.02 10 Terra nova sp. 0.0 0.1 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 0.2 0.1 0.0 0.0 0.06 11 Coeliodidymocystis sp. 1.3 2.1 0.3 0.2 0.9 0.2 1.2 0.5 0.3 0.7 1.3 0.8 0.1 0.7 0.0 0.03 12 Tentacularia coryphaenae 3 22 P' 3 4 19 8 6 4 5 6 10 3 P' P' 0.48 13 Oesophagocystis dissimilis 12 6 8 7 8 12 9 3 6 3 9 8 8 9 11 -0.05 14 Kollikeria 1 Didymocystis spp. 13 4 1 5 9 4 7 7 8 4 5 6 11 5 6 -0.11 15 Dinurus euthynni 55 9 19 35 66 1 2 3 0 0 0 15 0 0 0 -0.34 16 Dldymocystoides intestino- musculans' 30 27 26 37 49 39 54 18 15 12 16 44 134 17 64 -0.14 17 Hirudinella ventricosa 0.4 0.6 0.4 0.2 1.1 0.7 0.4 0.3 0.1 0.0 0.0 1.1 0.1 0.2 1.0 -0.10 18 Raorhynchus terebra 22 16 13 17 15 18 25 65 4 2 4 12 3 1 0 -0.00 19 Dldymocystoides intestino- muscularis' 14 3 3 9 3 8 5 8 6 6 6 7 13 3 2 -0.24 20 Lagenocystis 1 Univitellannulocystis spp. 76 40 29 29 22 45 43 16 38 17 30 61 178 34 41 -0.11 21 Tergestia laticollis 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.8 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.00 22 Rhipidocotyle sp. 0.0 0.3 0.0 0.2 0.2 2.3 2.4 0.0 1.3 0.0 0.0 0.1 0.0 0.0 0.0 0.08 23 Ctenascarophis type 35 7 6 2 18 22 38 49 7 17 21 33 4 1 108 24 Spmitectus type 10 7 20 2 12 5 9 10 5 6 18 13 5 3 10 25 Scolex polymorphus (large) 4 0.4 0.2 0 7 166 33 P' 122 101 27 10 161 9 7 26 Scolex polymorphus (small) 200 124 1,089 287 8,900 257 463 140 211 53 24 206 495 153 105 'P ' = present. ^N lo. 16— stomach; No. 19— intestine. the fish, and their longevity was doubtful. Philometra sp. (No. 7) was found predominantly in developed ovaries, which were present in less than half of the fish sampled. The data were used for com- paring school-school variability only. Seven parasites were considered "temporary". They appeared to be short-lived or easily lost from the fish. The caligoid copepods (No. 1, primarily Caligits productus in the tropics and C. bonito in temperate waters) were not permanently attached and probably moved from fish to fish (Kabata 1981). Syncoelium filiferum (No. 6) was common on the gills in New South Wales and New Zealand samples (I, J, and K), but was not recovered from anywhere in the tropics. It is common on fish endemic to New Zealand (D. Blair^). It was considered possibly a temperate short-lived parasite, at least on skipjack tuna, and this was verified by the school integrity study and by conventional tagging data (see later). Some hemiurids are known to be readily lost from the gut of other species of fish (Table 3). Margolis and Boyce (1969) observed that over half the Leci- thaster gibbosus were lost from salmon fingerlings *D. Blair, Deparlment of Zoology, University of Canterbury, Christx;hurch, New Zealand, pers. commun. September 1984. within 3 wk of bringing the fish into captivity. We found Dinurus euthynni (No. 15) in all tropical sam- ples from the central and western Pacific but not in the temperate samples I, J, and K. As it showed strong school associations and as the didymozoid data described later showed that New Zealand fish had a recent origin in the tropics, D. euthynni was evidently a short-lived tropical parasite that was lost as the fish migrated south. This also appeared to be true for Hirudinella ventricosa (No. 17) and possibly for two relatively rare gut-lumen digeneans, Terges- tia laticollis (No. 21) and Rhipidocotyle sp. (No. 22). In other fish, adult acanthocephalans may be short lived (Table 3). Moller (1976) found that over half the Echinorhynchus gadi in three species of fish were lost within 2 wk of the fish being brought into capti- vity. In our data, Raorhynchus terebra (No. 18) was present in reduced numbers in I, J, and K, sug- gesting it was lost in southern waters. All these parasites then were labelled "temporary". Didymozoid digeneans were considered "semi- permanent" parasites. In other fish, some didymo- zoids, or at least the remains of them, are believed to stay in the tissues for the life of the fish. Others, in- cluding some species found in the gonads or gills, are lost annually (Table 3). In general, therefore, skip- 347 FISHERY BULLETIN: VOL. 83, NO. 3 0) o c <0 cc CO c ♦^ o CO CO T3 C 3 o CD CO O T3 CD m CO v CO CO CO o. CO c CO o. CO E 3 E "x CO E « CO o I ffi < c to Q. CO CO o CD CO CO k_ CO t^ r^ t^ i^ -j- o> o> 03 flj 0) ^ o o o o CO c O) 05 CO CO C£> G) 0> CT) 0} (1) 0) o o o o o o CO CD CD ^•O T3 T3 ^^ 1- CO (O (O 05 Q) o o o ■"" cj (3) ra o) J: >- ^ b b o Q CO (0 (0 Q CD s 2 :s CO CO CO o> ^ ^ ^ o o - — oo 00 : O) CJ) ♦- fc_ ^ OJ Q) CD •«- in CO (U 0) (1) w _j _J C- ^'CD ._ c o> 9^ E - III -§ C (0 -I w 2 CO 3 S. cr CO 3 o o CO 3 CO s CO CO to CO •S 2 Q. 3 3 .C CO _s •~ CD CO k> s O -C >, 3 CO S (D 3 c: S OO 03 0) 0) O CD E E E E E CO CO CO CO CO CD (D (D d) d) C C C C C CJ CJ o 0} CO CO CO o o o ^ ^ "a "D "o o o C C C CO CO CO CD CD E £ .^ *: *j O o £ S £ CO w5 ■O 13 ■© CD CD CD O) O) O) CO CO ^ "co o -I CO Uu o < CO CO 3 3 -C j: o o c c ■c -c o o c c ■c -c o o UJ UJ CD CO (o to ^2 p o ■S o o -c S c i: CO O CJ UJ < CO 3 CO O Q lo 5.S o 5 CO 3 CO o « -Q 5:9 — B) 5) 3 ^ ^ .9 CD Cb E 3 O o i^ O CD •Q CD r- esI 3 rj) ^^-§ o o o o E E -~l TJ ^ CO cn CO S' O .^ QJ CU ■C C cS O Q. O CO O E E C is >~ - o o CD -(3 O E 2^ O S m Q) Q) 5 CO CO 3 ci CD n c o ■=^ ^ 1 1 II O H CO (T> ...i^ o ^—^ t^ a> c '- CO CO r^ E T3 C3> o C- O) j; C CO CO CO S Q. CO CO CB CD >% >> ^ J£ ^ ^ .^ o o o o o o CO CO CO 5 5 5 5 5 E E E E E E O i^ ^ CO « CM CO t^ ■>- oo C3> ID oo •.- o oo T— ,- > A CD CD CO CO 3 Cn CD cr 3 C •^ U. ■^ ■n CO CD O bo >~ >N >% >->.>. k_ k_ > CD CD QJ CD CD CD CO (D .^ O — C C C C C C £" " 0) CD (D 3 3 CD (D CD CD CD 0) O CQ S 2 S 2 2 S 5 CD CO c CD CO 3 C k^ — O -^ .C CO Q_ CO OC _ CD ■;= £ CJ Cb CO cn E 3 O CO CD CD > CD 348 I.KSTKKKTAL.: PAKASITKSOF SKIIMACK TUNA jack tuna didymozoids were thought to be in the fish probably for at least several months. However, there was some suggestion that 3 of the 10 skipjack tuna didymozoids had a shorter adult life span than the others. Didymozoid No. 16 was much less common in New Zealand waters than in the tropics (Table 4), and didymozoid Nos. 19 and 20 were also less com- mon and, in addition, showed strong school associa- tions (see later). These three didymozoids (possibly representing four species) were omitted from the analysis for Figure 3. The remaining four parasites (Nos. 8, 9, 10, and 12) were classed as "permanent". Larval cestodes and nematodes, particularly those found in the tissues, are generally believed to survive for several years, often for the life of the fish (Table 3). They have been used successfully many times as fish population markers (see MacKenzie 1983). In skip- jack tuna, the larva of a trypanorhynch cestode, Ten- tacuLaria coryphaenae, was found in the wall of the body cavity and occasionally in the viscera. No degenerating forms were seen, suggesting that it survived for an extended period and hence could be an excellent population marker, though counts were not available from areas C, N, and 0. Larval anisakids were found on the wall of the stomach or in the mesentery. The literature suggested that they should also be good long-term markers (Table 3). They were counted in all areas. Protozoan parasites have been used successfully to separate stocks of several species of fish. However, none has been reported from skipjack tuna, and we found none in this study. Table 4— Average number of didymozoids in New Zealand fish (all lengths) compared with fish caught m the tropical western Pacific (areas A, B. C, D, E, F, G, and L). In paren- theses, log (X -I- 1) length-adjusted means for fish 40 to 57 cm only. No.' Parasite New Zealand Trop ics 2 D. filiform is 9 (1.1) 6 (0.8) 3 D. simplex 20 (1.7) 12 (1.3) 4 D. fusiforme 2.7 (0.6) 2.5 (0.5) 5 L. multisacculatum 0.3 (0.1) 0.1 (0.0) 11 Coeliodidymocystis sp. 1.1 (0.4) 0.9 (0.3) 13 0. dissimllis 7 (1.3) 9 (1.9) 14 Kollikeria 1 DIdymocystis spp. 5 (1.2) 6 (1.5) 16 D. mtestinomuscularis' 15 (1.8) 41 (3.3) 19 D Intestinomuscularis^ 6 (1.3) 7 (1.4) 20 L. katsuwoni 1 U. kat- suwoni 26 (1.9) 47 (2.3) No. of fish 232 (213) 469 (364) 'Code no. from Table 2. ^Stomach. ^Intestine. Relationships Between Areas Analyses of individual distributions of permanent and semipermanent parasites showed that the abun- dances of individual parasites varied across the Pacific. However, these differences were inconsis- tent, the pattern established by one parasite being in conflict with that of a second, and so on. The data from the three anisakid nematodes and the seven didymozoids considered longest lived were therefore analyzed using canonical variate analysis. Because of the more permanent nature of these parasites and the completeness with which they were recorded from all areas, these data were considered the most reliable for statistically assessing the similarities and dissimilarities between areas. The first three canonical axes accounted for 75% of the Figure 3. -Results of multivariate analysis using 3 "permanent" and 7 "semipermanent" parasites (Nos. 2, 3, 5, 8, 9, 10, 11, 13, and 14). Values for first two can- nonical vectors plotted, and 95% confidence rings in- dicated for samples of more than 24 fish. The letters refer to the sampling sites indicated in Figure 1. 349 FISHERY BULLETIN: VOL. 83, NO. 3 variation in area-to-area differences in parasite numbers. A plot of the first two, accounting for 58% of the variation, showed the Atlantic fish (0) to be distinct from all the Pacific ones, even though only three fish from the Atlantic were dissected (Fig. 3). However, fish from California (M) and Ecuador (N) fell close to the western Pacific samples. They were separated out on the third axis (not shown), but nevertheless it is evident that they had a somewhat similar parasite fauna. The fish from western New Zealand (J) appeared distinct, and so too, to a less ex- tent, were the Papua New Guinea samples (C and D). There is no suggestion that fish from Ponape (A), Palau (B), Solomon Islands (E), Fiji (G), and the Mar- quesas (L) had distinct faunas of these long-lived parasites. H- Figure 4. -Results of multivariate analyses using 7 "temporary'" parasites (Nos. 1, 6, 15, 17, 18, 21, and 22). 95% confidence rings given for samples of more than 24 fish. In this analysis, Anisakis II had the most powerful discriminating properties, though at least 7 of the 10 parasites used were capable of substantial discrimi- nation in their own right. An analysis based on the 7 "temporary" parasites (Nos. 1, 6, 15, 17, 18, 21, and 22) produced a much greater separation of areas (Fig. 4). They are grouped into two broad classes: one containing New South Wales (I), New Zealand (J, K), and the eastern Pacific (M, N); and the other the western tropical areas. Each area in the latter group had a temporary parasite fauna that was distinct from most other areas. Over 83% of the variation was accounted for by the first two axes, and 90% by the first three. It is interesting to note that New South Wales (I) is more similar to east New Zealand (K) than to west New Zealand (J) (this was much more marked on the third axis, not shown, where I and K were pulled to one side), and that west New Zealand is similar to California (M) and Ecuador (N). Taken together. Figures 3 and 4 indicate that several distinct skipjack tuna parasite faunas existed within the tropical Pacific, and the longer lived parasites were more evenly distributed than the shorter lived ones. To check these results and to look for links be- tween the New Zealand fish and the tropical areas, the west Pacific data were reanalyzed using first the 10 "semipermanent" parasites (the didymozoids) and second the 4 "permanent" parasites (anisakids and T. coryphaenae). The average numbers of didymozoids in the New Zealand fish were almost identical to the overall average for the central and western tropics (Table 4). In the multivariate analyses, the temperate water samples fell to one side of the tropical samples (Fig. 5 Figure 5. -Results of multivariate analysis using 10 didymozoids only (Nos. 2, 3, 4, 5, 11, 13, 14, IK, 19, and 20). 95% confidence rings given for samples of more than 24 fish. 350 LESTKKKTAL.: I'AKASI TKSOK SKIIMACK Tl'NA - H, I, J, K), possibly because of the three didymo- zoids suspected of being relatively short-lived (Nos. 16, 19, and 20). The east and west New Zealand sam- ples (J, K) were identical on the first two axes, and separated only slightly on the third axis (not shown). There was no obvious link between New Zealand and any particular tropical area. Similarly, the larval nematodes and T. cory- phaenae (Nos. 8, 9, 10, and 12) did not suggest a link between New Zealand fish and those from any par- ticular tropical area (Fig. 6). However, west New Zealand (J) now appeared distinct from east New Zealand (K) and New South Wales (I). The separa- tion was due to areas having either high Anisakis I and II and low Terranova and T. coT^phaenae or low Anisakis I and II and high Terranova and T. cory- phaenae. West New Zealand (J) was at one extreme (high Anisakis) and the three most northwestern areas- Ponape (B), Fiji (G), and Marquesas (L)-at the other. Tentacularia coryphaenae and probably Terranova were picked up in the tropics. It seems likely that one or both of the Anisakis larvae were picked up predominantly in temperate waters, par- ticularly in west New Zealand. This may explain the separation of west New Zealand from the other areas in Figure 4. In summaiy, the New Zealand fish were not close- ly aligned with any particular tropical sample, and the eastern and western New Zealand fish were probably carrying similar parasite faunas when they arrived in New Zealand. Tentacularia coryphaenae Data on this parasite are presented in detail because we had more than for any other parasite and because potentially it was our most valuable marker. It also was the subject of many queries from skipjack tuna processors. The parasite was common through- out the south, central, and west Pacific (Table 3, parasite No. 12). The means of samples of over 22 fish within the length range 44 to 53.9 cm suggested an east-west cline across the Pacific, with twice as many parasites being found in fish from around the Marquesas (L) as around Papua New Guinea (C and D) (Fig. 7). A regression analysis of number of para- sites against longitude using tropical data on the number of parasites in 972 fish, transformed and ad- justed for differences in host length (data collected independently by the SPC), showed that the relation- ship was statistically significant, though it only ac- counted for about 7% of the fish-to-fish variation. Considering fish of all sizes, the number of T. cory- phaenae in the tropics increased with the size of the Figure 6. - Results of multivariate analysis using the four "permanent" parasites (anisakids and T. coryphaenae, Nos. 8, 9, 10, and 12). 95% confidence rings given for samples of more than 24 fish. fish (Fig. 2, solid circles). The increase around 47 cm is due to many of the Marquesas fish being this size and Marquesas fish tended to have more T. cory- phaenae. In New Zealand, smaller fish had about the same average number as fish from the tropics. How- ever, this number did not increase with size (Fig. 2, open circles). Thus, the 58 -t- New Zealand fish had fewer parasites than their peers in the tropics, and about the same number as the 45 to 50 cm fish. School-to-School Variation An analysis of variance, and a median test, were carried out on 30 schools from the Marquesas and 19 schools from eastern New Zealand (areas L and K, respectively. Table 5). The results of the two methods on each data set show close agreement. In the Marquesas, five parasites showed strong evidence of association with particular schools, i.e., the probability that schools differed was at least 0.95 with both methods. The parasites were Caligtis spp. (No. 1), D. euthynni(bio. 15), H. ventricosa{No. 17), D. intestinomuscularis (No. 19), and Lagenocystisl Univitellannulocystis spp. (No. 20). For these para- sites to show significant differences, they must have heavily infected some schools and not others, and their life span in the fish must have been equal to or shorter than the life of the school. The literature review suggested that the first three species could possibly be readily lost from fish, and this is vin- dicated by their strong school association. The evi- dent impermanence of the last two, however, was unexpected. It was as a consequence of this finding that they were not included in the analysis for Figure 3. 351 FISHERY BULLETIN: VOL. 83, NO. 3 0 5 (26) ^=:^ X- 7 7 (40) (156) 11 (309) Figure 7. -The average numbers ofT. coryjihaenne in skipjack tuna 44 to 58.9 cm long in samples of over 22 fish. Note that the iiumi)er in- creased to the east. (In parentheses, number of fish sampled.) Several other parasites thought to be short-Hved, such as R. tereba, did not show up in the test, presumably because their infective stages were relatively evenly distributed in the tropical Pacific. In New Zealand, parasites showing close associa- tion with particular schools (using both tests) were L. multisacculatum (No. 5), S. filiferum (No. 6), Philometra sp. (No. 7), Coeliodidymocystis (No. 11), T. coryphaenae (No. 12), R. terebra (No. 18), and D. intestinomuscularifi (No. 19). Syncoelium filiferum and R. terebra were both thought to be temporary parasites that could be gained in New Zealand or ad- jacent waters (Norfolk Island). The origin of the Philometra was unknown. Their number reflected the state of maturity of the fish and this varied be- tween schools. However, we were left with three didymozoids and T. coryphaenae, all of which dif- fered markedly between schools in eastern New Zealand. One of the didymozoids, L. multisaccula- tum, a normally rare tropical parasite, was found on all five fish from one school (numbers per fish 1, 2, 8, 3, and 1). As the three didymozoids and T. cot^- phaenae are essentially tropical parasites, the schools had evidently not fully mixed while in temperate waters. If this is true, these four parasites could not have been picked up uniformly across the Pacific. Evi- dence is given above that D. intestinmnuscularis (No. 19) was not picked up uniformly even within the Marquesas. For the other species, a comparison of their mean numbers per fish per school in different areas of the tropical Pacific showed that Coeliodidy- m.ocystis sp. and particularly T. coryphaenae were in- deed more abundant in some areas than others. Lobatozoum multisacculatum was too rare for any conclusions to be drawn in this respect. 352 LESTER ET AL.: PARASITES OF SKIPJACK TUNA Table 5.— Comparison of within and between sctiooi vari- ability in numbers of parasites per fish for two areas. no. Marquesas New Zealand Parasite Analysis of Median Analysis of Median (see Table 3) variance' test' variance test 1 . ..a * * * 2 * * 3 * 4 5 * * * * * 6 (no parasites found) * * • * * * 7 * * 8 * 9 10 (no parasites found) 11 * * 12 * * * * 13 * 14 15 * * * * * (no parasites found) 16 17 * * 18 * * * * * * 19 * * * * * * * * * * 20 * * * * * 21 * 22 'The probabilities that the ratio of the between and within school variances is no greater than one. (Based on transform- ed data, i.e., log (parasite no. + 1.0).) 'The probabilities that the proportion of fish with more than the area median is the same for all schools. 3* * • = P< 0.001 ; * * = P < 0.01 ; * = P < 0.05; blank = P > 0.05. Rate of Mixing of Schools To estimate the rate of mixing of schools we need- ed to know the distribution of the parasites among schools before, and after, some known time interval. This we did not have for any of the Marquesas samples. In New Zealand, however, some approximate calculations could be made because schools arrived from the tropics at different times. Sixteen of the 19 east New Zealand schools were of similar-sized fish and were all caught within 1 mo. These schools were divided into two groups: "early arrivals" and "recent arrivals". (This was done by ranking the schools using a combination of four parasites whose prevalences were positively correlated with each other, Nos. 16, 18, 19, and 20, and which were thought to be relatively short-lived parasites picked up in the tropics. Thus high numbers indicated a recently arrived school.) From catch data (Habib et al. 1980), we calculated that there was an average of 3 to 4 wk between the capture of 25% and 75% of the annual catch. This interval was taken as the ap- proximate period between the arrival times of the early group and the recent group. If mixing was occurring, one would expect that the school-school differences for tropical parasites would be greater when the fish first arrived (the recent arrivals) than after they had been there for a few weeks (the early arrivals). However, this we could not demonstrate. Our sample sizes at this point were rather small (eight schools in each category), and in fact the reverse appeared to be the case, the early schools having a generally higher variability than the recent arrivals. This suggested that the early arrivals had come from several areas (and still had not fully mix- ed), whereas many of the later arrivals had perhaps come from one area. DISCUSSION Ten of the 26 parasites counted were species of didymozoid trematodes. These are almost exclusively a tropical group. Yamaguti (1970), for example, found 84 different species of didymozoid in fish around Hawaii. None were recorded in checklists of parasites from New Zealand (Hewitt and Hine 1972) or Canada (Margolis and Arthur 1979). Thus, although skipjack tuna are caught in both tropical and temperate waters, their didymozoid infections are evidently picked up primarily in the tropics. Larval didymozoids have been found in small fish and in invertebrates. It is almost certain that the definitive host becomes infected by feeding on an in- fected intermediate host (Cable and Nahhas 1962; Nikolaeva 1965). In the tropics skipjack 40 to 60 cm in length feed largely on fish, squid, and stomatopods (Argue et al. 1983). In New Zealand, however, they feed almost exclusively on euphausids (Habib et al. 1980, 1981). This completely different diet in New Zealand, together with the fact that no endemic New Zealand fish are known to carry any didymozoids, lead us to the conclusion that few, if any, didymozoids are picked up in New Zealand waters. The occurrence of 10 species of didymozoids in skipjack tuna caught in New Zealand, in numbers very similar to fish of the same size caught in the tropics, thus indicates that New Zealand and tropical fish were found until recently in a similar tropical en- vironment. Almost certainly, the New Zealand fishery is based on fish that have recently migrated from the tropics, and not on fish recruited as post- larvae in temperate waters. This disagrees with tag- ging data which show that the bulk of New Zealand skipjack tuna of known origin were off New South Wales 10 mo earlier. However, the tagging inference is applicable to < 4% of the total New Zealand fish (Argue and Kearney 1983). Our conclusion is in agreement with Argue et al. (1983) who found no juvenile skipjack tuna in the stomachs of adults from 353 FISHERY BULLETIN: VOL. 83, NO. 3 subtropical waters, though juveniles formed a signi- ficant component of the adult diet in the tropics. The absence of degenerating T. coryphaenae and the positive correlation of parasite number and host length suggest that the parasite was long-lived and accumulated in the fish with age. The low numbers of Tentacuiaria in the 57 -i- cm fish caught in New Zealand indicate that these fish have had a different history from their peers in the tropics. The bulk of the skipjack tuna caught in New Zealand are 45 to 55 cm long. Less than 10% measure 60 cm or more (Habib et al. 1980, 1981). We have concluded above that the majority of New Zealand fish recently arrived from the tropics. The T. coryphaenae data in- dicate that the 57 -i- cm fish left the tropics at 45 to 55 cm long and have not returned. Evidently as fish age, they become less migratory. This was hypothe- sized by Kearney (1978). Large fish were not necessarily permanent resi- dents in New Zealand, however. Of 17 57-t- cm fish on which full dissections were carried out, 2 were carrying the acanthocephalan R. terebra, a parasite thought to be relatively short-lived (see above) and not picked up in New Zealand. Raorhynchus terebra was common in fish from Norfolk Island (area H). Thus some of the large fish may have recently come from areas as far away as Norfolk Island. The first two canonical variate analyses comparing all areas sampled suggested that fish 40 to 57 cm long had moved between areas and carried the longer lived parasites with them. Parasitologically, there was no evidence of more than one stock of skip- jack tuna in the Pacific. Richardson (1983) observed an east-west cline in the gene frequency of two en- zymes across the Pacific. From an analysis of 200 gene frequencies he proposed an "isolation by distance" model for skipjack tuna. In this, the degree of mixing of skipjack tuna genes was inversely pro- portional to the distance between the spawning areas. Tagging data have confirmed that there is some mixing of adult skipjack tuna in the central and western Pacific (Kleiber and Kearney 1983), though more than 95% of the tagged fish recovered during the SPC program were caught within 1,000 mi of their point of release (Kearney 1982). Schools of skipjack tuna have been observed to break up when feeding (Forsberg 1980). This and observations from aircraft where schools have been seen to merge and later separate (Habib unpubl. obs.) have led to the hypothesis that skipjack tuna do not remain in a particular school for more than a day or so. Certainly the pattern of recovery of SPC tags suggested that tagged skipjack tuna underwent con- siderable mixing amongst schools soon after release (Argue and Kearney 1983). However, using Mar- quesas data we found that several parasites showed strong school associations, particularly didymozoid Nos. 19 and 20 (D. intestinomfiuscularis and Lageno- cystislUnivitellanulocystis spp.). In another didymo- zoid, Neometadidymozoon helwis from the gills of Platycephalus fuscus, it takes up to a year for the worms to migrate through the tissues, pair up, mature, and die (Lester 1980). Though only a short migration is needed for didymozoids 19 and 20, as they are intestinal parasites, the worms are still like- ly to be in the skipjack tuna for a period of weeks. Thus, their strong association with particular schools suggests that school half-life is likely to be in terms of at least weeks rather than days. In New Zealand, the large school-school differ- ences observed in the numbers of T. coryphaenae and several other tropical parasites, especially in the early arrivals, indicate that at the time of catching, the New Zealand schools had not mixed sufficiently to mask their previously distinct tropical faunas. Do schools remain intact for an extended period, perhaps for the life of the fish? Sharp (1978) found evidence of genetic similarity between individuals in core schools, suggesting that some members of the school were siblings. However, none of L. multi- sacculatum, Coeliodidymocystis sp., or T. coinj- phaenae, three long-lived parasites that showed significant school-school differences in New Zealand, showed any significant differences in the Marquesas. This suggests that within the probable long life of these parasites, fish caught in the Marquesas had changed schools and had thus obscured any patchi- ness in the distribution of the infective stages of the parasites. The parasitological data, then, do not sup- port the hypothesis that fish stay in the same school for life. ACKNOWLEDGMENTS We are indebted to R. E. Kearney, A. W. Argue, and other officers of the Skipjack Program, South Pacific Commission, for much of our material, much of the Tentacuiaria data, and assistance in preparing this report. Other material was obtained with the help of David Bateman, Heinz Tuna Cannery, Eden, N.S.W.; Paul Dalzell, D.P.I. Fisheries Research, Ka- vieng, P.N.G.; Bernie Fink, Van Camp Sea Food Co., San Diego, CA; James Joseph, lATTC, La Jolla, CA; Ted Morgardo, Star-Kist PNG Pty. Ltd., Rabaul, P.N.G.; and Ronald Rinaldo, Southwest Fisheries Center, La Jolla, CA. Their cooperation is greatly appreciated. Taxonomic assistance was rendered by David 354 LKSTKKKTAl..: I'AkASITKS (IK SKllMACK 'ITNA Blair, Department of Zoology, University of Canter- bury. N.Z.; Rod Bray, British Museum (N.II.). Lon- don, U.K.; and Arlene Jones, Commonwealth Insti- tute of Parasitology, St. Albans, U.K. For help with the dissections we thank C. Boel, K. Couper, B. M. Heath, M. K. Jones, M. S. Kennedy, G. G. Lane, and A. G. West. K. MacKenzie, D.A.F., Scotland, kindly reviewed an earlier draft of the manuscript. The project was supported by a grant to H. M. D. Hoyte, Department of Parasitology, University of Queensland, from the Nuffield Foundation, U.K. LITERATURE CITED Argue, A. W., F. Conand, and D. Whyman. 1983. Spatial and temporal distributions of juvenile tunas from stomachs of tunas caught by pole-and-line fjear in the central and western Pacific Ocean. Tuna Billfish Assess. Pro- tjranime Tech. Rep. 9, 47 p. South Pacific Commission, Noumea, New Caledonia. .Arcte. A. W.. AND R. E. Kearney. 1983. An assessment of the skipjack and baitfish resources of New Zealand. Skipjack Survey Assess. Programme Final ("try. Rep. 6, 68 p. South Pacific Commission, Noumea, New Caledonia. Baker, R. J., and J. A. Nelder. 1978. Ceneralised linear interactive modelling. Numerical Algorithims (Jroup. Oxford. p]ngl., 184 p. BOYCE. N. P. 19(i9. F'arasite fauna of j.iink salmon (()nciirhynchui< gor- hu,' 20 m depth, and in 1982 long (10-30 min) dives were common in depths of 40-600 m. Variability in distribution and behavior presumably was related to availability of prey. Besides feeding and scx-ializing, we saw sporadic bouts of aerial activity (breaches, tail slaps, etc.) and log play. During 1981 and 1982 we observed young-of-the-year calves apparently waiting at the surface while adults fed below. In 1982, two such lone calves played with debris in the water. During near-surface skim feeding, whales often associated in V-shaped or echelon formations, with up to 14 animals staggered behind and to the side of each other, all moving in the same direction at the same speed, with mouths wide open. We hypothesize that such coordinated movement may increase the efficiency of feeding on concentrations of small invertebrates. In recent years, much has been learned about behavior of several species of baleen whales (e.g., Payne 1983). Most long-term studies of whales have been carried out during winter, when social inter- actions, mating, and calving occur more often than feeding. Recently, however, detailed studies have been conducted in summer, when whales are pri- marily feeding (e.g., Dorsey 1983). This paper describes the general behavior of bow- head whales, Balaena mysticetus, in the summers of 1980-82. A companion paper gives a quantitative description of the surfacing, respiration, and dive patterns (Wursig et al. 1984). This study was done to provide background data necessary to interpret observations of bowhead behavior in the presence of 'This paper has been reviewed by the U.S. Minerals Manage- ment Service (MMS) and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of MMS, nor does mention of trade names or commer- cial products constitute MMS endorsement or recommendation for use. ^Moss Landing Marine Laboratories, P.O. Box 450, Moss Land- ing, CA 95039. 'New York Zoological Society, Weston Road, Lincoln, MA 01773; present address: Center for Long Term Research, Inc., Weston Road, Lincoln, MA 01773. "LGL Ltd., Environmental Research Associates, 22 Fisher Street, King City, Ontario LOG IKO, Canada; present address: Sohio Alaska Petroleum Co., Pouch 6-612, Anchorage, AK 99502. ^LGL Ltd., Environmental Research Associates, 22 Fisher Street, King City, Ontario LOG IKO, Canada. offshore industrial activities (Richardson et al. in press). The Western Arctic population of bowheads winter in the Bering Sea, and migrate north and east to the eastern Beaufort Sea in spring. During sum- mer (late June to early September), most are off northwestern Canada in Amundsen Gulf and the eastern part of the Beaufort Sea (Fig. 1). In the com- mercial whaling era in the 19th century, many bow- heads apparently summered in the Chukchi and western Beaufort Seas off Alaska (Townsend 1935), but bowheads are no longer present in significant numbers off Alaska in summer (Dahlheim et al. 1980). The eastern Beaufort Sea is believed to be a major feeding area for bowheads (Fraker and Bock- stoce 1980), but previous to 1980 there had been no comprehensive studies of bowheads in that area. METHODS Aerial Observations We observed from a Britten-Norman*^ Islander air- craft based at Tuktoyaktuk (Fig. 1). The Islander has two piston engines, high wing configuration, and low ^ Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Manuscript accepted October 1984. FISHERY BULLETIN; VOL. 83, NO. 3, 1985. 357 FISHERY BULLETIN: VOL. 83, NO. 3 BEAUFORT SEA ^ / Tuktoyaktuk Kugmallit Bay N.W.T. 100 I30*» Figure l.-The eastern Beaufort Sea. Stall speed. An OnTrac VLF/Omega navigation system indicated the latitude and longitude. A hand- held color video camera (JVC-CV-0001 in 1980 and 1981, Sony HVC-2000 in 1982) connected to a por- table video cassette recorder (Sony SLO-340 in 1980 and 1981, Sony SL-2000 in 1982) was used through the side windows to record oblique views of bowheads. Our usual strategy was to search until we encoun- tered bowheads, and then circle over them as long as possible while making observations. If contact was lost, we searched for another group. We created a fixed reference point about which to circle when bowheads were below the surface by dropping a fluorescein dye marker. Near the start of most periods of circling above whales, a sonobuoy (AN/SSQ-41B or AN/SSQ-57A) was dropped to broadcast underwater sounds to the aircraft, where they were recorded. In 1980-82, we flew for a total of 340 h during 71 offshore flights. Of this time, we circled over bow- heads for 97.7 h during 46 flights. Flight duration was typically 4-5.5 h. Flights were made between 3 and 31 August 1980, 31 July and 8 September 1981, and 1 and 31 August 1982. We encountered bow- head whales during every day we flew in 1980, and during the majority of days in 1981 and 1982. We usually did not fly when wind speed exceeded 25 km/h; in more severe conditions whales are dif- ficult to detect and behavior cannot be observed reliably. While searching for whales, we usually flew at 457-610 m (1,500-2,000 ft) above sea level (ASL), 358 Wl'KSlCKT Al..; HKHAVIOK OK HI (WIIKAI > VVHAl.KS and at 185 km/h. While circling over whales, we reduced speed to 148 km/h. Bowheads rarely ap- peared to be disturbed by the aircraft when it re- mained at or above 457 m (Richardson et al. in press). The aircraft crew usually consisted of four biol- ogists and the pilot. Three biologists were seated on the right side of the aircraft, which circled clockwise during behavioral observations. Biologists seated in the right front (copilot's) seat and in the seat directly behind it described behaviors. These descriptions were recorded onto audiotape, onto the audio chan- nel of the video recorder, and, in 1981, directly onto data sheets by a biologist in the left rear seat. The biologists in rear seats videotaped whales, handled sonobuoys, and kept records. All personnel on board were in constant communication through an inter- com. While circling bowheads, we usually were able to obtain consistent records of 12 variables and types of behavior: 1) Location of sighting (and, therefore, water depth); 2) Time of day; 3) Individually distinguishing features, if any, on whales; 4) Number of individuals visible in area and number of calves; 5) Headings and turns of each whale in degrees true; 6) Distances between individuals (estimated in whale lengths); 7) Length of time at surface, and sometimes length of dive; 8) Timing and number of respirations or blows, including underwater blows; 9) Possible indications of feeding: mouth open, defecation, mud streaming from mouth; 10) Socializing, possible mating, probable nursing by calves; 11) Aerial activity: breaches, tail slaps, flipper slaps, lunges, rolls; 12) Type of dive: flukes out, peduncle arch, pre- dive flex. Descriptions of these behaviors appear later. In most parts of this paper we consider only the observations under "presumably undisturbed" condi- tions. Bowheads were considered "potentially disturbed" if our aircraft was at < 457 m ASL, if a boat was underway within 4 km, or if sonobuoys showed that industrial noise was readily detectable in the water. The "presumably undisturbed" n 1982 ^1981 □ 1980 TIME OF DAY (MDT) Figure 2. - Hourly distribution of behavioral observation time from the air of bowhead whales. w 20 LJ < '■% tr LjJ > o CO (T ZD O X 10 - 5- 7 6 5 4 ■ 3 2 -i I 0 I I 1982 ^1981 □ 1980 0 10 20 30 40 50 1 rrrr. h=H- o 200 400 DEPTH (m) 600 Figure 3. - Distribution of behavioral observation time from the air by depth of water of bowhead whales. The inset shows effort for the 0-50 m range by 10 m intervals. behavioral observations were distributed by hour of day and water depth as presented in Figures 2 and 3. Shore-Based Observations Shore-based observations were obtained in 1980 and 1981 from the east end of Herschel Island, Yukon (lat. 69°35'N, long. 138°51'W), and about 225 km west of Tuktoyaktuk (Fig. 1). A surveyor's theo- dolite was used from a high point (50 m ASL in 1980, 359 FISHERY Bl'LLETIN: VOL. 83. NO. 3 90 m ASL in 1981) on the coast. We used a Wild Tl theodolite with 6-s accuracy and 30-power optics in 1980, and a Nikon NT-2A with 20-s accuracy and 30-power optics in 1981. Horizontal and vertical bearings were later translated to x and y map coor- dinates. This transiting technique, developed by R. Payne, is described by Wiirsig (1978). The station was in use from 19 August to 1 September 1980, and 23 August to 13 September 1981. Locations of most whales within a 10 km radius of the theodolite station during fair weather and day- light hours were documented. Unfortunately, whales rarely approached Herschel Island closer than 5 km during the 1980 field season, so details of behavior were difficult to discern. In 1981, fewer whales were seen, but they were closer to shore, allowing more detailed behavioral observations. RESULTS The Surfacing-Dive Cycle In the Beaufort Sea in summer, nonmigrating bowhead whales typically alternate between dives of variable length, depending on activity, and sur- facings within which there are several respirations. This pattern differs slightly from that during migra- tion, when sounding dives (around 15 min long) are separated by periods when several brief surfacings, each with a single respiration, alternate with "series" dives about 15 s long (Rugh and Cubbage 1980; Car- roll and Smithhisler 1980). Presumably, migrating animals dive between respirations to avoid hydro- dynamic drag imposed by the air-water interface. No such submergence is necessary for a whale that is not moving rapidly through the water. However, the basic repertoire of breathing several times in relatively, closely spaced series and then not breathing for many minutes (during the long dive) is similar during both prolonged directed movement and more stationary activity. The pattern extends to some degree even to whales that remain at the sur- face for long periods (up to 30 min or more during surface skim feeding, socializing, or play). They generally breathe several times within a few minutes, and then cease breathing for a longer time, despite their near proximity to the surface and the availability of air. Similar patterns are seen in other whales, including right whales, Eubalaena glacialvi, (Kraus et al. 1982) and gray whales, Eschrichtiua robicstus, (Sumich 1983). Durations of surfacings and dives, intervals between successive blows, and number of blows per surfacing are described in Wur- sig et al. (1984). Surfacing and Respiring Whales in water deeper than about 30-45 m usual- ly surface head and blowhole first after a sounding dive, with the body oriented at some angle (such as 30°) from horizontal. When whales do not dive very deeply (as in shallow water), the surfacing is less due to active swimming upward, and the head and tail surface at approximately the same time. A blow is an exhalation of air by a whale. Blows can occur above or below the surface. Surface blows are usually visible as a white cloud of water spray, but may be so weak as to be undetectable. The first blow after a surfacing usually appears strong, prob- ably because it is a more forceful exhalation and because water is present above the blowholes during or just after surfacing. On calm days and when whales lie at the surface with the blowholes exposed, the blowholes are relatively dry, and blows may be difficult to detect. Blows of calves can also be dif- ficult to see. Surface exhalations of gray; humpback, Megaptera novaeangliae; fin, Balaenoptera physalus; and southern right whales, Eubalaena australis, are almost always followed immediately by an inhalation (B. Wiirsig, pers. obs.). Hence we suspect, following Scoresby (1820), that exhalations and inhalations generally occur together in bowhead whales as well. Diving and Associated Behavior The predive flex is a distinctive concave bending of the back seen several seconds before many dives. The whale flexes its back by about 0.5-1 m, so that the snout and tail disrupt the surface. Considerable white water is created at these two points. The whale then straightens its back and lies momentarily still before arching the back convexly as it begins its roll forward and down. The predive flex is seen from low vantage points as an abrupt lifting of the head, because the flukes apparently only touch the water surface from below. The predive flex was seen more often during 1980 than during 1981 or 1982. Although it occurred previous to dives well over 50% of the time in 1980, it occurred in only 8% of the observations (before 29 of 352 dives) in 1981. For 1982, we have especially detailed analyses of predive flexes. In that year, predive flexes occurred in presumably undisturbed noncalves before 32 of 132 dives (24.2%); flexes occurred more often in late August than earlier (Table 1). Dives following predive flexes were, on the average, about twice as long as dives without predive flexes (19.00 ± SD 7.877 min, n = 13, vs. 10.15 ± 360 WI'KSK; KTAL.: BKIIA\1()K(»F HOWHKAI) whalks Table 1. — Dives preceded by a predive flex among noncalf bowheads early and late August 1982. The frequency of occurrence is significantly highier after 19 August (chi-square = 4.29, df = 1, 0.025 < P< 0.05). Up to 19 Aug. 1982 After 19 Aug. 1982 Total Dives with predive flex Dives without predive flex Total 9 49 58 23 51 74 32 100 132 7.465 min, n = 36; Mann-Whitney U = 97.5, P < 0.01). Five dives were preceded by two predive flexes, with the flexes separated by a blow. Two dives were preceded by three flexes. We have no data on durations of dives following multiple flexes. During the dive, which can at times be predicted by the predive flex, the whale makes its body convex and pitches forward and down. If the angle of sub- mergence is steep, the tail is usually raised above the surface; if not, the tail may remain below or just touch the surface. Rarely do bowheads sink down without visibly arching the back. In 1982, 59 of 138 dives (42.8%) were preceded by raised flukes. Of the 32 dives preceded by one or more predive flexes, 21 also showed raised flukes. These two predive behaviors tended to occur together (x" = 3.94, P < 0.05, df = 1), and dives with raised flukes were significantly longer than those not preceded by raised flukes (18.67 ± SD 9.966 min, n = 12, vs. 10.05 ± 6.956 min, n = 38; Mann-Whitney U = 114, P< 0.01). There was no difference in durations of surfacings concluded with and without raised flukes. However, surfacings including predive flexes tended to be longer than those without predive flexes (3.09 + SD 1.038 min, n = 14, vs. 1.79 ± 1.284 min, w = 52; i = 3.50, df = 64, P < 0.001), probably because dura- tions of surfacings and dives are correlated (Wiirsig et al. 1984). The function of the predive flex is unknown. Flexes occur more often before longer dives (which may take the whales deeper in the water column). Raising the flukes before a dive appears related to the steepness of the dive; whales that roll forward while dropping the front of the body at least 30° below the water surface usually raise their flukes. The weight of the raised tail stock in the air must help propel the animal downward (much as human skin divers raise their legs above the surface during the initiation of a steep dive). Although raised flukes are common during steep dives in many whales, the predive flex has not been reported in other spe- cies. The Underwater Blow The underwater blow is a burst of air emitted underwater. The bubble burst is circular and up to 15 m in diameter when it arrives at the surface. Release of air underwater was recorded about 10 times via nearby (< 1 km away) sonobuoys; the noise was detectable for 3-4 s, but the white water and expand- ing concentric wave were visible much longer. On one occasion, we definitely saw that the air came from the blowhole rather than the mouth, and we believe that this is always true. We saw underwater blows immediately after whales dove and just before they surfaced, but more usually in the middle of the dive, when the whales were out of sight. Underwater blows were most frequent in 1980 during periods of pronounced feeding in water < 14 m deep (see Feeding section). In 1980, we saw 158 underwater blows in 30.4 observation hours; in 1981, 57 blows in 30.8 observations hours; and in 1982, only 6 blows in 36.5 observation hours. (The dif- ference between years is statistically significant; x^ = 189, df = 2, P < 0.001.) Concurrently, whales tended to be found in progressively deeper water from 1980 to 1982. Underwater blowing occurred more often in the morning and evening than around solar midday in both 1980 and 1981 (Fig. 4; solar noon occurs about 1500 MDT in the eastern Beaufort Sea). The midday "lull" in underwater blowing coincided with a peak in frequency of socializing, the main nonfeeding behavior observed (see Social Behavior section below). Nemoto (1970) suggested that baleen whales in general show a high level of feeding activity in the UJ Q. CO o _l ffl UJ I- < UJ o q: 3 O ^ 8 H z o i= 6 < If) CD o . 2 UJ < 34/3.8 50/7 9 65/11 2 0/0 3 16/154 12/9 3 6^*5 41/9 I 8 12 20 24 TIME OF DAY (MDT) Figure 4. - Number of underwater bowhead whale blows per aerial observation hour in relation to time of day, 1980 and 1981 com- bined. There were few underwater blows in 1982. The numbers at the top of each column are number of blows seen/number of obser- vation hours. 361 FISHERY BULLETIN: VOL. 83, NO. 3 morning and a lower level during midday, but we have no direct evidence of this in bowheads. Although underwater blows seem to occur more often in shallow water when whales may be feeding, we have not included this behavior as a definite part of feeding. There is only a general similarity to bursts of bubbles associated with feeding humpback whales in the North Atlantic (Hain et al. 1982), and the bubble nets reported for humpbacks by Jurasz and Jurasz (1979) are very different. Social Behavior Behavior was termed social when whales appeared to be pushing, nudging, or chasing each other, or when they were within half a body length of one another. Whales within half a body length almost always stayed close to each other, and oriented towards each other or interacted in some manner. Thus, our use of proximity as an indication of social- ity was appropriate. Interactions between mothers and calves, and between whales skim feeding in close proximity, were not included as social interactions in this analysis. Whales may, of course, communicate by sound, and thus may socialize over far greater distances than those described here. Our sonobuoys often detected bowhead calls while socializing was underway. However, we could not verify whether acoustic communication was occurring between any particular whales, so we restricted our definition of socializing to visible behavior. Synchronous diving and surfacing over areas many kilometers in diameter (see below) may represent a different form of social interaction from what we discuss in this sec- tion. Because groups of whales usually could not be reidentified positively from one dive to the next, we treated observations of social behavior at intervals of > 5 min as independent for the purpose of counting number of interactions. Conversely, we did not score social behavior by one group more than once in 5 min when counting frequency. Frequency of Socializing Social behavior was seen less frequently in late August-early September than in early August, both in 1980 and 1981 (Table 2). Rugh and Cubbage (1980) and Carroll and Smithhisler (1980) reported a higher incidence of social interactions during the spring migration around Alaska than we saw at any time. The apparent waning of social activity from early to late August may be part of a continuing decrease from a higher level in spring. Little socializing was observed in 1982. In presum- ably undisturbed whales, we observed only seven cases, all on 8, 19, and 23 August. Throughout August 1982, most whales were alone and making long dives. The overall socializing rate for each year (Table 2) demonstrates the dramatic decrease in socializing in 1982 compared with the two previous years. This decrease may be related to the increase in 1982 in the average distance from shore and depth of water at locations where bowheads were studied. However, we found no consistent trend for social- izing to occur more often in shallow water than in deep water within 1 yr. There was some indication of hour-to-hour vari- ation in amount of social activity in all 3 yr (Fig. 5). In 1980 and 1981, it peaked around 1400-1600 MDT, the noon period by sun time. In 1982, the few (7) cases were recorded from 1600 to 2000 MDT, some- what after solar noon (Fig. 5). In both 1980 and 1981 , there was another peak after 2000 MDT. Why whales should engage in more social activity around noon (and possibly in the evening) than at other times is unknown. However, diel rhythms are well known in several mammals (e.g., Saayman et al. 1973 for bottlenose dolphins; Matsushita 1955 for sperm whales; Schevill and Backus 1960 for hump- back whales). The increased level of socializing around noon may reflect a lowered level of feeding at that time, which Nemoto (1970) suggested for baleen whales in general. Physical Interactions During surface interactions with nearby whales, socializing whales often turned. In contrast, non- socializing whales often surfaced and dove again without changing direction. In the 3 yr, turns oc- Table 2.— Number of social interactions per aerial observation hour, divided into about 10-d periods, in 1980, 1981, and 1982. Only presumably undisturbed periods are included. Year 1-10 Aug. 11-20 Aug. 21-31 Aug. 1-10 Sept. 1980 28/7.0 = 4.0 1981 14/4.3 = 3.3 1982 1/1.5 = 0.7 6/2.9 = 2.1 12/5.5 = 2.2 3/7.6 = 0.4 8/7.7 = 1.0 9/3.3 = 2.7 3/12.8 = 0.2 4/4.0 = 1.0 Overall 42/17.6 = 2.4 39/17.1 = 2.3 7/21.9 = 0.3 362 Wl'KSlCKTAI,.: HKHAVIOKOK HOWUKAD WHALKS cr cr UJ 3 Q. O I (rt z z o o H (- o < < > q: cr UJ UJ H (r> Z ffl o _J _i < < O cr o UJ (O < 12 16 20 TIME OF DAY (MDT) Figure 5. - Number of bowhead whale social interactions per aerial observation hour in relation to time of day. curred during 53 of 133 (40%) surfacings with social- izing, and in 128 of 484 (26%) without sociaHzing (x^ = 9.04; df = 1,P< 0.005). When bowhead whales touched, they often appear- ed to push each other. Pushing or touching was usually done with the head, while oriented head to head, or head to tail. However, we also saw whales of adult size dive under the bellies of other whales and apparently nudge or push the other whales near their genital areas. At other times, whales dove under each other at very close range without any indication that they were touching. Apparent chase sequences involved two or three whales in a line, usually < 2 body lengths apart. Dur- ing these chases one whale often turned abruptly left or right, and the second (and third) followed. Move- ment was faster during chases than at all other times when we saw presumably undisturbed whales at the surface. Both touching and chasing may at times represent low levels of sexual activity, but this is unproven because we cannot determine the sex of a bowhead whale from a distance. Payne and Dorsey (1983) and Tyack and Whitehead (1983) described physically interacting right and humpback whales, respectively, which appeared to be engaging in social-sexual activity. Possible Mating In 1981, we twice observed apparent mating. The more prolonged observation was on 10 August 1981, within a 25 km- area where there were 20-30 whales whose main activity was socializing. Two whales interacted for over 1 h with chases, flipper caresses, belly-to-belly orientation, rolls toward and away from each other, head nudges to the genital area and to the rest of the body, tail slaps, and flipper slaps. One whale, a recognizable animal that we termed "Whitespot", was about 1-2 m longer than the other ("B") and was the more aggressive. Although B originally nudged the genital area of Whitespot, it was Whitespot who appeared to initiate flipper caressing and rolls toward B. The two whales rolled their ventral surfaces together for about 5 s, but B then rolled its ventrum in the air in an apparent attempt to avoid ventral contact with the larger animal. As it rolled away from Whitespot, B defecated, and when Whitespot moved its head toward the genital area of B, B defecated two more times in rapid succession. B then dove away from Whitespot, and Whitespot followed it at the surface in an apparent chase. Whitespot then stopped and, alone at the surface, rolled two times and tail slapped while on its back. It then dove, and the two appeared together again at the surface 4 min later, with no fur- ther energetic surface interaction. We do not know the sex of either animal, but it ap- peared that Whitespot was attempting to copulate with the reluctant animal. Some of us (Wiirsig and Payne) have observed southern right whale females frequently roll their ventra away from aggressive males, leaving their genital areas above the surface of the water, where the males cannot reach them. Everitt and Krogman (1979) photographed very similar behavior of a group of six bowheads off Bar- row, AK, in May. Our observations here were highly reminiscent of such behavior. Although adult females are slightly larger than adult males in both right and bowhead whales, we commonly see large southern right whale males in pursuit of smaller females, which attempt to avoid the males. On 25 August 1981, two bowheads briefly placed their ventral surfaces together and clasped each other with their flippers. After 1 min, they rolled apart, blew, and dove slowly as a third whale ap- proached. The mutual rolling and leisurely diving indicated that, if this was copulatory behavior, it was mutually undertaken by the two whales in contrast to the previous example. Group Structure and Stability Two observations of recognizable bowheads pro- vided evidence about group structure and stability. We observed a distinctively marked pair of adults, one accompanied by a calf, at about lat. 70°10'N, long. 133°50'W on 7 August 1980. We saw a similar- 363 FISHERY BULLETIN: VOL. 88. NO. 3 ly marked group of two adults and a caLf, almost cer- tainly the same whales, on 20 August at lat. 70°07'N, long. 131°30'W, which is about 85 km from the place they had been seen 2 wk earlier. This observation suggests that some groups of bowheads are main- tained for at least a few weeks. The observation also suggests that females with calves may sometimes be accompanied by escorts, as has been observed for wintering humpback whales (Herman and Antinoja 1977). Feeding Feeding appeared to occupy much of the time of the bowheads that we observed, but we had to rely on indirect clues, such as observations of swimming with open mouth, mud streaming from the mouth, or presence of feces in the water, to indicate that feeding had taken place. The four possible types of feeding behavior that we identified were 1) water- column feeding; 2) near-bottom feeding; 3) skim feeding; and 4) mud tracking. Of these, the first three rather clearly represented feeding, whereas the function of the last was less certain. As noted above, underwater blowing showed some association with feeding, but the connection was uncertain. In 1980, certain feeding behaviors occurred in par- ticular areas: only water-column feeding was seen near the Issungnak artificial island site (Fig. 1), whereas only skim-feeding was seen off the Tukto- yaktuk Peninsula near McKinley Bay. In 1981, there was less evidence for feeding, although we suspect that most feeding occurred in the water column. In 1982, when whales dove for long periods (up to 30 min), we su.spected water-column feeding to be occurring at almost all times. Water-Column Feeding Water-column feeding could not be observed directly. Whales were scored as feeding in the water column when they dove for long periods, and when, between long dives, there was much defecation and only slow forward motion. Defecation is simply an in- dication of prior feeding. However, particular behaviors such as a series of long dives usually con- tinued for many hours, so occurrence of defecations between long dives was considered indicative of on- going feeding in the water column. The frequency of apparent water-column feeding was not constant. In 1980, we saw bowheads water- column feeding from 3 to 22 August. Thereafter, few whales were present in the area where we had ob- served this behavior, and whales seen elsewhere did not seem to feed in the water column. In 1981, when we saw less defecation, we only scored as water- column feeding some adult whales that dove for pro- longed periods on 24 August, while calves remained at the surface. In 1982, most whales made long dives. These whales probably were feeding in the water column, even though we saw little defecation at the surface. Feeding below the surface may have occurred during many other dives besides those that we classified as dives with water-column feeding. Observations on 3 August 1980 typify water- column feeding behavior. On this date, bowheads were north of Kugmallit Bay where water depth was 18-38 m. The surface water was turbid, brackish water from the Mackenzie River, but beneath this surface layer, there was a second layer of clearer, saline Beaufort Sea water (Griffiths and Buchanan'^). The whales occurred in groups of 2-10 animals, and occasionally as individuals without others nearby. Group members showed a high degree of synchrony, often surfacing very close together and remaining close at least until they dived again. Not only did the members of a group surface and dive synchronously, but various groups spread over an area several kilometers in diameter all tended to be at the surface or beneath it at the same time. While the animals were at the surface, they moved slowly forward while taking a series of breaths. As each individual dived, it raised its tail clear of the water, and disappeared from view in the turbid water. Thus, these dives must have taken the whales well below the surface. When the whales were at the surface, they often disturbed the turbid surface layer, exposing dark patches of seawater from deeper depths. However, while submerged after a dive that was preceded by raised flukes, they did not affect the thin surface layer, indicating that they were probably feeding in the underlying clearer ocean water. Defecation was frequent, suggesting that feeding may have taken place recently. The feces clouds were red-orange. Bottom Feeding On 12 August 1980, we noticed clouds of mud suspended in the water about 25 km west of Issung- 'Griffiths, W. B., and R. A. Bachanan. 1982. Characteristics of bowhead feeding areas. In W. .1. Richardson (editor), Behavior, disturbance responses and feeding of bowhead whales Balaena myst}cetu.s in the Beaufort Sea. 1980-81, p. 347-455. Unpubl. Rep., 456 p. LGL Ecological Research Associates, Inc. Bryan. TX. for Bureau of Land Management, U.S. Department of the In- terior, Washington, DC. Available from Minerals Management Service Alaska DCS Region, P.O. Box 101159, Anchorage, AK 99510. 364 WCKSK; KTAl..; KKllAVIOKOF HOWUKAD WllALKS nak artificial island (Fig. 1). The clouds represented suspended mud and not plankton because the material was of the same color as mud dredged up by industrial activities. Whales surfaced with large amounts of muddy water streaming from their mouths, indicating they had been feeding from or near the bottom. This behavior occurred in 24-29 m of water and seemed very localized. We saw no in- dication of bottom feeding in the same area on 22 August 1980, but we had observed similar mud clouds nearby on 9 August, when prolonged observa- tions were not possible. On 25 August 1981, whales again surfaced with mud streaming from their mouths. The location was 15 km south of the position where we observed such behavior in 1980; water depth was only 10-13 m. These are, to our knowledge, the first published behavioral observations of apparent near-bottom feeding by bowhead whales. However, Johnson et al. (1966), Durham (1972), and Lowry and Burns (1980) detected pebbles and bottom-dwelling species in bowhead stomachs. Bottom-feeding whales were usually separated from other whales by 150-300 m when at the surface. On 12 August 1980, at least 10 whales were bottom feeding within an area of 3 km radius. Whether they were feeding on inbenthic or epibenthic inverte- brates we do not know. In the eastern Beaufort Sea, the average biomass of inbenthic animals greatly ex- ceeds that of epibenthic animals (Griffiths and Buchanan footnote 7). However, the latter may occur in dense swarms in certain places. For a balae- nid whale, such swarms would seem to be a much more suitable type of food than inbenthic organisms. Mud might be taken inadvertently along with epi- benthic animals. Skim Feeding The only feeding type that we observed directly was skim feeding. In the third week of August 1980, we observed whales moving slowly and deliberately at the surface with their mouths open wide. Usually the rostrum just broke the surface of the water, and was parallel to it. In these cases, the lower jaw was dropped to varying degrees, as could be seen from the depth of the white chin patch. In 1980, skim feeding was observed along the Tuktoyaktuk Penin- sula in water 12-22 m deep. Whales occasionally skim fed alone, but more often did so in groups of 2-10 or more individuals. During any one observation period, they stayed in the same general area by repeatedly turning and did not appear to make any net geographic movement. However, we found groups of skim-feeding whales in different locations on different days. During 1981 , we witnessed skim feeding on a large scale only on the evening of 18 August, 32 km NNW of Pullen Island in water 25 m deep. About 20-30 whales in the 25 km^ area were swimming with mouths open; they travelled slowly, usually just below the surface (~ 2-3 m deep). Copepods were unusually abundant in near-surface waters at this location and time (Griffiths and Buchanan footnote 7). On 23 and 24 August 1981, we saw one isolated example on each day of a whale feeding at the sur- face briefly (observed for < 1 min) in approximately the same area as on 18 August. Typically, skim-feeding whales were oriented with their backs at the water's surface. However, they occasionally swam on their sides with mouths open at an angle of about 60°, and once we saw two whales separated by three body widths swimming on their sides, belly to back. In one instance, a skim-feeding whale swam inverted for at least 3 min, with the underside of its chin at the surface. Frequently, the skim-feeding whales swam in echelon formation, each whale swimming just behind the preceding whale, but offset laterally by one-half to three body widths, reminiscent of geese in V for- mation (Fig. 6). At other times, they swam abreast and parallel to one another. Videotape from 18 August 1981 showed that whales within the echelons were a mean of 0.53 whale lengths apart (SD = 0.599, n = 66), or about 8 m. These distances were measured from different echelons or from the same echelon at intervals of at least 5 min. We videotaped a recognizable whale for almost 3 h on this day as it skim fed in changing echelon formations, usually taking the lead position. Echelons were clearly dynamic in terms of membership, size, and organiza- tion. In 1981, the mean echelon size was 4.7 animals (SD = 4.05, n = 23). While the largest such forma- tion observed in 1980 contained 5 individuals, the largest in 1981 contained 14 animals. We suspect that echelon feeding increases the feeding efficiency of those animals staggered behind and to the side of other individuals, perhaps by help- ing them to catch prey that escape or spill from the mouth of the whale in front, or by reducing the abili- ty of prey to escape to the side. Skim feeding in eche- lon may allow more effective exploitation of concen- trated patches of small prey than would be possible if whales were feeding alone. If so, the change in effi- ciency that accrues when echelons are formed may have an important cost/benefit effect on energy ex- pended per whale. The predominant prey types of bowheads include copepods and euphausiids (Lowry 365 FISHERY BULLETIN: VOL. «3, NO. S Flcn'REK.-Fivebowhead whales skim feeding in echelon formation. Drawing after a 35 mm photograph and video footage from the air. and Burns 1980). The latter are adept at avoiding most sampling gear because of their rapid move- ment. However, bowheads at times collect euphau- siids in very large numbers, despite the bowhead's slow swimming speeds relative to the better known euphausiid predators such as rorquals. Perhaps echelon feeding is especially helpful in catching fast- swimming prey such as euphausiids. During 1982, little direct evidence for feeding was noted. We saw no skim feeding at the surface, and noticed only nine isolated instances when a whale's mouth appeared to be open slightly. These brief, slight 366 WI'KSK; KT A1..: HKIl.WKiKOF HOVVMKAlt WIIAI.KS openings of the mouth contrasted sharjaly with the sustained large gajjes observed in 1980 and 1981, and probably did not represent feeding. In southern right whales, Payne (pers. obs.) has observed mouth open- ing that he interprets as yawning following sleep. Mud Tracking Mud tracking occurred when whales swimming in shallow water (< 12 m depth) disturbed the bottom sediments with each fluke beat, producing clouds of mud joined by a narrower trail of muddy water. These elongated clouds of mud were different from mud clouds produced during presumed bottom feed- ing. Although we often could not see the whales, in at least a few instances their mouths were open. We saw mud tracking during only three flights in the third week of August 1980. Mud tracking probably represented incidental disturbance of bottom sediments by a whale feeding near the bottom in shallow water. We saw no evidence that bowheads ever turned and swam back along a mud track made previously. The mud tracks tended to be straight, and some extended for well over 1 km. At certain times, clouds of mud streamed from the whale's body as it swam near the surface. In this case, we suspect that the whales had contacted the bottom, and that the mud had stuck to their bodies. Sometimes, mud-tracking whales exhaled while submerged, producing a characteristic burst of bubbles (see section on The Underwater Blow). Defecation Defecation usually was evident as a cloud (2-3 m diameter) of red-orange feces near the surface. Whales almost invariably were moving forward or diving when they defecated, and over 50% of the bowheads observed defecating in 1980 did so while the tail was arched up high out of the water just before the dive. The anus was thus close to or at the surface. No part of the body appeared to touch the feces cloud, which was visible at the surface for up to 10 min. When whales moved forward while defecat- ing, the feces were more dispersed and disappeared within 1-2 min. Brown (1868) noted that feces of eastern arctic bowheads were also red. Renaud and Davis* observed red clouds of feces off the Tuktoyak- tuk Peninsula in 1980. Defecation was seen more often in 1980 (23 cases during 30.4 h over whales) than in 1981 (11 cases during 30.8 h over whales). The difference is statis- tically significant (x^ = 4.39, df = 1, 0.025

oC. R. Evans, Biologist, and J. Hickie, Biologist, LGL Ltd., En- vironmental Research Associates, 22 Fisher St., King Citj-, Ontario LOG IKO, Canada, pers. commun. September 198L 369 FISHERY BULLETIN: VOL. 83. NO. 3 right whales (Payne 1972). Both involved lifting the object with the head, moving the object along the back, and patting it with the flippers. Two log-play- ing bowheads attempted to push the log under water with the head. This action was reminiscent of a motion commonly made by male right whales when attempting to mate with uncooperative females (Payne, pers. obs.). CALF PLAY. -On two occasions in 1982, lone calves at the surface interacted with debris in the water, and the actions had the appearance of play. The first incident occurred over 12.3 min on 19 August 1982, when a lone young-of-the-year calf followed a line of surface debris ■^ 2 m wide, prob- ably composed mainly of invertebrates. The calf stayed at or just below the surface and oriented directly along the windrow, changing course as the line meandered left or right. Although the calf ap- peared to have its mouth open slightly for brief periods, it did not appear to feed extensively, if at all. However, its movements thoroughly disrupted and dispersed the line of debris. The movements were rapid and jerky, reminiscent of any uncoordinated young mammal. The calf lunged forward while in the debris on three occasions, and slapped its tail onto the water surface twice. For -^ 30 s, it moved rapidly along the line, ventrum up, with rapid up-and-down movements of the tail for the entire time. The se- quence ended when the calf dove out of sight at the end of the windrow; we did not see it with an adult. Although the incident did not seem to represent con- certed feeding, this "play" by the calf may have been practice in skills required for feeding. A second incident of "calf play" occurred on 23 August 1982. This calf was first encountered hang- ing quietly just below the surface, or moving forward very slowly. During slow movement, it entered an area marked by dispersed fluorescein dye from one of our dye markers (see section on Methods). The dye covered an area about 40 m by 100 m. Immediately upon entering the area of bright green water, the calf became active. During the 22.3 min of associa- tion with the dye, the calf rolled ventrum up eight times for 5-20 s each time, and moved back and forth within, and to the edge of, the dye-clear water inter- face. Although not as active and not beating its tail as fast as the calf in the windrow, this calf made abrupt turns of > 90° on 25 occasions during its stay in the dye, reorienting itself at the dye's edge in order to remain within the dye. The calf ultimately moved out of the dye and oriented toward an ap- proaching adult. When the two joined, the calf ap- parently began nursing. Synchrony of Activity and Orientations There was often an impressive degree of syn- chrony of basic behaviors among members of quite widely spaced groups. We observed apparent syn- chronization of behaviors on time scales ranging from seconds to days. Synchrony in General Activity During 1980, we found that all or most bowheads in various areas did the same thing for up to several days. Some days later, the whales had usually moved, and whales were then found elsewhere engaged in different activities. For example, on 3 and 5 August 1980, whales north and east of Issung- nak artificial island were mainly engaged in water- column feeding, with frequent defecation. By 6 and 7 August, whales in this area shifted to more surface- active behavior, interacting in groups with pushes and apparent chases. We saw little defecation at this time. On 12 August, at least 15 animals about 30-40 km west of this area were all apparently bottom feeding. Whales were encountered in two additional areas in 1980: east of Pullen Island (19 and 20 August) and just west of McKinley Bay (19-22 August). In the Pullen Island area, all whales were mud tracking as described above. In the second area, mud also was evident, but there was much less underwater blow- ing. Some animals had mouths open at the surface. On 22 and 23 August 1980 almost all whales we en- countered were skim feeding in groups of 10-30 animals north of McKinley Bay. However, in the Issungnak area farther west, substantial numbers of whales were still water-column feeding. On 27, 29, and 31 August 1980, whales interacted in small groups of 2-5 individuals. Some small groups oriented SSW, perhaps indicating the beginning of migration. In summary, during 1980 (but not 1981 or 1982) we found that whales in various areas did much the same thing for up to '^ 5 d, but then shifted location, activity, or both. A partial explanation for the syn- chrony of behavior seen in 1980 may be that whales moved to exploit new food resources, and that the most appropriate feeding mode changed according to site-specific conditions. In the subsequent 2 yr, it ap- peared that whales were doing more water-column feeding in deeper water, perhaps because of a more consistent food supply. Synchrony in Dives and Surfacings We sometimes had the impression that all whales 370 WfKSIC, ETAL.: BKHA\I()K()K HOWllKAD WHALKS in an area were synchronizing? their surfacin^s and dives. Many were too far apart to be in visual con- tact. Ljungblad et al (1980) also reported synchrony among whales engaged in water-column feeding ~75 km east of Kaktovik, AK. They reported that "... whales were observed on the surface almost at regular intervals and gave the impression of resting between dives; then, suddenly, no whales would be seen in any quadrant for several minutes." Although synchrony in surfacings by animals far apart suggests acoustic contact between animals, it is not proof of communication over that distance. The synchrony could be established through indepen- dent responses to common external cues. It could also occur if the animals were close together and visually synchronized before observations began; the observed synchrony would then be a residual phenomenon that persisted because of whales diving and surfacing for similar lengths of time. None of these possible explanations -acoustic communica- tion, common external cues, or residual phenomenon - can be either proven or discounted at this time. Synchrony in Orientations Analysis of orientations provides additional evidence that widely separated whales at times syn- chronize their behavior during summer. Our best data were from three flights in 1980 when we flew in a straight line. At these times, we counted each in- dividual only once. Rayleigh and x^ tests (Batschelet 1972) show that whale orientations were significant- ly nonrandom (Table 4). For flights when we circled to make detailed behavioral observations, we analyzed orientations using the first heading noted for each surfacing of a whale. Because we were making repeated observa- tions on the same animal in some cases, any consis- tency in orientations during those flights is attrib- utable in part to different whales, and in part to subsequent surfacings of the same whale. In 1980 and 1981, the whales were oriented nonrandomly during 7 of the 11 flights with enough data for analysis (Table 4). The headings changed from day to day, however, and bore no apparent relationship to the general behavior of the whales. In 1982, no signi- ficant departures from uniformity were found during any of the five flights with sufficient data for analysis. The headings on the latest day with observations in 1980 and 1981 usually were not in the direction to be expected at the beginning of the westward migra- tion. On 31 August 1980, most bowheads observed while we circled north of the Tuktoyaktuk Peninsula were oriented north, east, or south (mean 121 °T (true). Table 4). However, later that day on a direct flight, we found other bowheads to be oriented toward the south (mean 189° T). In this same general area, Renaud and Davis (footnote 9) also recorded a slight eastward tendency for bowheads seen on 21-24 August 1980, but a significant southwestward tendency (236 °T) on 3-4 September 1980. On 8 September 1981, most whales west of Herschel Island were oriented toward the northeast (62 °T), again not the direction to be expected at the begin- ning of westward migration. These results support our impression that most of the whales we observed were not migrating. Table 4— Bowhead orientations, judged relative to true north from the air, 1980-81. Only during the direct flights was each observation known to represent different animals. During the circling flights, each whale was scored an unknown number of times (but only once per surfacing). Vector Chi-square No. of animals with these orientations mean direction Rayleigh test test Date N NE E SE S sw W NW Total P Direct flights 11 Aug 1980 16 1 3 0 5 2 10 6 43 321° <0.001 <0.001 12 Aug. 1980 7 5 16 5 7 6 9 2 57 bimodal n.s. <0.025 31 Aug. 1980 1 1 1 3 8 8 0 1 23 189° <0.001 7 Circling flights 31 Aug. 1980 4 4 6 3 11 1 0 2 31 121° <0.05 10 Aug. 1981 0 3 0 2 0 4 0 1 10 — n.s. 10 Aug. 1981 3 1 7 2 1 6 0 1 21 bimodal n.s. 13 Aug. 1981 12 9 11 1 1 0 1 1 36 43° «0.001 18 Aug. 1981 2 5 10 5 6 1 2 1 32 111° <0.001 18 Aug. 1981 3 0 0 0 0 1 6 0 10 289° <0.005 23 Aug. 1981 0 4 1 1 0 0 0 0 6 62° <0.02 24 Aug. 1981 1 0 3 2 5 8 10 5 34 243° <0.001 6 Sept. 1981 1 7 2 1 0 2 3 3 19 — n.s. 7 Sept. 1981 2 5 1 1 0 2 2 3 16 — n.s. 8 Sept. 1981 1 8 3 1 1 0 0 0 14 62° <0.001 '/ means cell sizes too small for a chl-square test. 371 FISHERY BULLETIN: VOL. Ki. NO. :i We do not know whether consistent orientations represented a type of social synchrony, or whether the whales independently reacted to environmental stimuli (such as currents or wave orientations). Norris et al. (1983) and Braham et al. (1984) reported gray and bowhead whales, respectively, that may have been feeding by stationing themselves against a current. Shane (1980) has reported a similar stationing against the current for bottlenose dolphins in Texas. Gray whales in lagoons have been observed to move in the same direction as the tidal current (Norris et al. 1977), but in that case move- ment may have been related to avoiding shallow water as the tide receded. Miscellaneous Observations Speed of Travel In 1980 and 1981, some data were gathered on bowheads visible from Herschel Island. The whales were usually > 3 km from shore, and detailed behavioral observations were infrequent. However, speed was sometimes measurable with a surveyor's theodolite. Whales rarely changed direction within any one 30 s period, so we calculated speeds from theodolite readings taken within 30 s of each other. This criterion was changed to 60 s for 30 August 1981 , when a whale was followed at the surface for a long period, and changed direction relatively little. For 1980, average speed was 5.1 km/h(w = 18, SD = 2.93) at the surface, and 4.3 knVh {n = 4, SD = 0.79) below the surface. The 1980 speeds are com- parable with the most reliable estimates derived by Braham et al. (1979) and Rugh and Cubbage (1980) for migrating bowheads: 4.8-5.9 km/h and 4.7 ± SD 0.6 km/h, respectively. However, based on additional data, Braham et al. (1980) estimated the mean speed at Point Barrow in spring to be 3.1 ± SD 2.7 km/h. Speeds during active migration along the coast of Baffin Island in fall were 5.0 ± SD 1.3 km/h (n = 22) based on theodolite observations from a cliff (Koski and Davis''). On 30 August 1981, an adult whale traveling east was observed continuously for 1.52 h. Its behavior was unusual- it did not submerge during the entire time. Its mean speed was 2.3 ± SD 1.26 km/h, con- siderably slower than the speeds mentioned above. Its mean blow interval was 10.0 ± SD 13.55 s (n = 420), significantly lower than the mean for all other undisturbed whales observed from Herschel Island (14.6 ± 9.56; n = 60; t = 2.54, P < 0.02). On 8 September 1981, a mother-calf pair was observed by theodolite for 1.8 h. The average speed of the calf was 8.9 ± SD 5.57 km/h (n = 28). During this rapid movement, the calf exhibited breaches, forward lunges, tail slaps, and flipper slaps. Associations of Bowheads with Other Species We saw several marine mammal species in the same general areas in which we observed bowheads: ringed seals, Phoca hispida; white whales, Delphi- napterus leucas; and a gray whale. There was no ob- vious interaction between these species and bowhead whales. The gray whale was about 500 m from the closest bowhead. The Canadian Beaufort Sea is the extreme northeastern limit of the gray whale's sum- mer range (Rugh and Fraker 1981). Flocks of up to 50 phalaropes {Pkalaroptis sp.) were often present near skim-feeding bowheads. These birds often alighted on water that had been disturbed by the whales, sometimes only a few meters from the whales. Phalaropes and bowheads probably feed on some of the same plankton species. The whalers used the presence of phalaropes as an indicator of where "whale feed" was present and, therefore, where whales were likely to be found (J. R. Bockstoce in press). Aside from phalaropes, we noticed glaucous gulls, Lamis hyperboreus; arctic terns. Sterna paradisaea; and unidentified gulls circling briefly over whales on eight occasions. DISCUSSION Activities of Bowheads in Summer and Other Seasons From 1980 through 1982 we observed a steady progression in the August distribution of bowhead whales near Tuktoyaktuk from shallow water near- shore to deeper water farther from shore (Fig. 3; Richardson et al.'^). Such a dramatic difference in distribution over the 3 yr may be due to many dif- ferent ecological and behavioral factors. Disturbance "Koski, W. R., and R. A. Davis. 1980. Studies of the late sum- mer distribution and fall mipration of marine mammals in NW Baf- fin Bay and E Lancaster Sound, 1979. I'npubl. Rep., 214 p. LGL Ltd., Toronto, for Petro-C'anada E^xplorations, Caljarary. Available from Pallister Resource Management Ltd., 700 - 6th Avenue S.W., Calgary, Alberta T2P 0T6, Canada. '^Richardson, W. J., K. A. Davis. C. K. Kvan.s, and P. Norton. 1983. Distribution of bowheads and industrial activity, 1980-82. In W. J. Richardson (editor). Behavior, disturbance responses and distribution of bowhead whales Balnenti ynysticetics in the eastern Beaufort Sea, 1982. Unpubl. Rep., p. 269-3.'J7. L(]L Ecological Research Associates, Inc., Bryan, TX, for U.S. Minerals Manage- ment Service, Reston, VA. Available from Minerals Management Service Alaska CX:S Region, P.O. Box 101159, Anchorage, AK 99510. 372 WI'RSIOETAL.: BEHAVIOR (IF HOWIIKADWHALKS by industrial activity in nearshore waters is also a possibility (see footnote 12). The fact that predomi- nant feeding modes shifted from year to year is con- sistent with the "variable food supply" explanation. In 1980, whales in shallow water exhibited bottom feeding and skim feeding, while whales in slightly deeper water apparently fed in the water column. In 1981, most feeding appeared to be water-column feeding and skim feeding. In 1982, whales made long dives and presumably were often feeding in the water column. Bowhead whales have finely fringed baleen, the longest of any whale species, and are adapted to strain small zooplankters from the sea. Stomach con- tents indicate that, at least in Alaskan waters, bow- heads feed mainly on copepods. euphausiids, and amphipods (Marquette et al. 1982). Summering bow- heads tend to occur at locations where copepod abun- dance is above average (Giiffiths and Buchanan foot- note 7). Lowry and Burns (1980) examined five whales killed off Barter Island, AK, in autumn and found about 60% copepods and about 37% euphau- siids in their stomachs. However, all five whales may have fed at least partially near the sea floor; about 3% of the stomach contents consisted of mysids, amphipods, other invertebrates, and fish. Durham (1972) also suggested, based on stomach content analyses showing mud-dwelling tunicates, vegeta- tion, silt, and small pebbles, that bowheads feed at times near the bottom. Lowry and Burns concluded from stomach content analyses that "... a feeding dive probably involves swimming obliquely from sur- face to bottom and back, feeding the entire time." Although this may be true at times, there is no direct information on underwater feeding behavior. We suspect that bowheads can detect concentrations of prey and open their mouths when appropriate. The bowhead whale is perhaps a more catholic feeder than once thought, capable of taking advantage of many different types of prey items at various posi- tions in the water column and near the bottom. Year- to-year changes in distributions and availability of prey may account for the distributional changes that we have observed, but data on yearly changes in prey are lacking. During spring migration around Alaska, bowhead whales appear to do little feeding; their stomachs usually are nearly empty (Marquette et al. 1982). On the other hand, feeding continues in autumn after bowheads have moved from the Canadian to the eastern part of the Alaskan Beaufort Sea (Lowry and Burns 1980; Marquette et al. 1982). Some feeding occurs in autumn as far west as the Point Barrow area (Lowry et al. 1978; Braham et al. 1984), and perhaps farther west off the Soviet coast (Johnson et al. 1981). Feeding is not the only activity of bowheads in summer. Socializing, perhaps with occasional sexual activity, is also important. In 1982, however, there was less socializing than in 1980-81. Whales were in close proximity to each other less in 1982. This year- to-year difference in proximity may be related to the difference in type of feeding. While skim feeding at the surface, whales are often in close echelons. The proximity necessary for echelon feeding offers more chance for socializing, and socializing before or after feeding in echelon may be important to that mode of feeding. When whales appear to feed in the water column, however, they usually do not stay as close together. Thus, this type of feeding may neither re- quire nor stimulate aggregations of animals, and the suspected predominance of water-column feeding in 1982 may explain the low socializing rate that year. Even when there is no close socializing, however, animals are often in a dispersed group within which acoustic communication is probably possible. Our observations of surfacing and dive synchrony by whales spread over distances of several kilometers indicate that they may have been in touch by acoustic communication. The primary mating period of bowhead whales occurs in spring, including the spring migration (Everitt and Krogman 1979; Carroll and Smithhisler 1980; Johnson et al. 1981; Nerini et al. 1984). We saw some evidence for sexual activity in the Cana- dian Beaufort Sea in both 1980 and 1981, but not in 1982. Even the active rolling at the surface that we observed in 1981, however, was not as boisterous as observed by Everitt and Krogman in spring. Also, we found an indication of less social activity in late August-early September than in early August. This apparent waning in social activity may be a contin- uation of the waning of sexual activity that started in late spring. Many calves are born in winter or spring before the whales reach Point Barrow, although some may be born in early summer (Davis et al. footnote 9). During summer, the activities of female bowheads with accompaning calves are closely coordinated with those of their calves, and differ in some details from the activities of other adult bowheads (this study; Wursig et al. 1984). At least some calves re- main with their mothers for the fall migration (Davis and Koski 1980). We know of no information con- cerning the age of weaning of bowhead calves, but in the closely related right whale, at least some calves remain with their mothers for 1 yr and ultimately separate from their mothers after returning to the 373 FISHERY BULLETIN: VOL. 83, NO. 3 wintering area (Taber and Thomas 1982). Aerial activity similar to what we observed in the eastern Beaufort Sea- breaches, tail slaps, pectoral flipper slaps, and rolls- has been observed in bow- heads during spring migration (Rugh and Cubbage 1980; Carroll and Smithhisler 1980). It appears that aerial behavior is more frequent during spring migration than on the summer feeding grounds, and this may be related to the high levels of social-sexual activities during spring. Comparisons with Other Baleen Whales Bowhead whales spend their entire lives in arctic and near-arctic waters. This habit separates them from all other baleen whales, which may move into temperate or subtropical waters (Lockyer and Brown 1981). However, behavior is in large part determined by feeding mode and related ecological factors (Gould 1982), and here similarities between bowhead whales and several other species are evi- dent. Gray, bowhead, and right whales are often found in shallow water, and all three species feed on small invertebrates. Gray whales usually feed near the bot- tom (Bogoslovskaya et al. 1981; Nerini and Oliver 1983), whereas right and bowhead whales may skim their food at or near the surface (see Watkins and Schevill 1976, 1979 for right whales). But all three species are adaptable in feeding behavior. Gray whales apparently will feed on mysids associated with kelp (Darling 1977) or on crab Pleuroncodes in the water column (Norris et al. 1983). Right whales also feed below the surface, probably straining swarms of copepods and other small invertebrates in the water column (Pivorunas 1979). While it has long been known that bowhead whales feed at the surface and in the water column (Scoresby 1820), it was recently established from stomach content analyses (Durham 1972; Lowry and Burns 1980), and by observing bowhead whales surfacing with muddy water streaming from their mouths (this study), that bowheads sometimes feed near or on the bottom. It is not surprising that there are many similarities in the behavior of these species. Bowhead and right whales, in particular, are morphologically and tax- onomically quite similar, and appear to obtain their food in very much the same ways. In fact. Rice (1977). mainly relying on a detailed comparison of morphology of bowhead and right whales, suggested that they be put in the same genus, Balaena. The sleeker rorquals (Balaenopterid whales) generally gather their food more actively by lunging through concentrations of prey, and at least in the case of humpback whales, have developed compli- cated behavioral strategies for confining and concen- trating their prey (Jurasz and Jurasz 1979; Hain et al. 1982). The rorquals are more often found in deeper water far from shore, and their behavior in general appears to be less similar to that of the bowhead whale than its behavior is to that of gray and right whales. Gray whales spend part of the winter in warm water, near the shores of Baja California, and most of the summer they feed in the northern Bering and southern Chukchi Seas. Western Arctic bowheads make much shorter migrations, spending their winter in the pack ice of the Bering Sea and their summer predominantly in the Beaufort Sea. The two species thus use the Bering Sea at different seasons and for different purposes -gray whales to feed in summer and bowheads apparently to mate and calve in winter. Like bowhead whales summering in the Beaufort Sea, the primary activity of gray whales summering in the Bering and Chukchi Seas is feeding. However, both bowheads and gray whales (Sauer 1963; Fay 1963) occasionally socialize during summer. Right whales, like bowhead whales, often appear to feed in the water column or at the surface (Watkins and Schevill 1976, 1979) and may stay in the same general area for days. While skim feeding, both species at times aggregate into echelons. In right whales, these echelons usually consist of only 3-6 whales (Payne, pers. obs.), while up to 14 bow- head whales have been seen skim feeding in echelon. However, Payne observed right whales during winter when little feeding occurs, so apparent dif- ferences in feeding details may be due to seasonal factors. Apparent differences between the social activity of bowheads and right whales may also be largely attri- butable to the different times of year when they have been studied. The same kinds of nudges and pushes have been observed for interacting whales of both species, but the winter-spring social activity of right whales is much more boisterous than the summer social activity of bowheads. Observations of bowhead whales in spring indicate that their social-sexual acti- vity at that season can be as boisterous as is seen in mating groups of right whales (Everitt and Krogman 1979; Carroll and Smithhisler 1980; Rugh and Cub- bage 1980; Johnson et al. 1981). The belly-up posi- tion of a female bowhead photographed in spring in the Alaskan Beaufort Sea (Everitt and Krogman 1979) indicates that females may attempt to evade potential mates who pursue them in large mating aggregations in the same way that female right 374 Wl'KSlCKTAl..: HKHAVK )K OK BOWIIKAD WHALKS whales evade males in Argentine waters (Payne and Dorsey 1983). A photograph showing a remarkably similar mating group of right whales is shown in the article by Payne (1976). The fact that similar-looking social aggregations are seen in both species argues for a similar social system, although it does not show that the social systems are similar in all details. Female right whales have young only at intervals of 3 yr or more (Payne, pers. obs.). The same appears to be true of bowheads (Davis et al. footnote 9; Nerini et al. 1984). This long calving interval may help to explain why bowhead and right whales have not made as dramatic a recovery from commercial exploitation as has, for example, the gray whale. Payne also found that right whale females that calve along the shore of southern Argentina in winter are usually not present in the years between calving. Each winter, a different segment of the population of mature females is present, in a 3-yr cycle. It is not known whether this cycling extends to the summer feeding grounds of these right whales. During the present 3-yr study, year-to-year variation in feeding and social behavior was dramatic, but we do not know whether this was due in part to some cyclic and synchronized activity of individual whales. We suspect that variable prey distibution was largely responsible. ACKNOWLEDGMENTS This project, including preparation of this paper, was funded by contracts from the Bureau of Land Management and the Minerals Management Service, U.S. Department of the Interior, to LGL Ecological Research Associates. We thank the Polar Continen- tal Shelf Project of the Department of Energy, Mines, and Resources (Canada) for logistical help. Dome Petroleum Ltd. and Esso Resources Canada Ltd. shared data and assisted with logistics. NORCOR Engineering and Research Ltd. provided the Islander aircraft, and J. Merilees was its capable pilot. Personnel of the Beaufort Weather Office were helpful. K. Finley, P. Tyack, and R. Wells helped with aerial observations. K. Hazard, G. Silber, S. Taber, P. Thomas, and M. Wiirsig collected data from Herschel Island. C. Greene of Polar Research Laboratory set up the sonobuoy system. S. Heimlich- Boran prepared the illustrations of bowheads in this paper. J. Bird, L. Guinee, and V. Rowntree of the New York Zological Society, and C. R. Evans, R. Wells, and M. Wiirsig assisted with data analysis. H. Braham provided helpful suggestions on an earlier draft. We thank all of these organizations and in- dividuals. LITERATURE CITED Batschelet, E. 1972. Recent statistical methods for orientation data. In S. R. Galler, K. Schmidt-Koenig, G. J. Jacobs, and R. E. Belle- ville (editors), Animal orientation and navigation, p. 61-91. NASA SP-262, Natl. Aeronaut. Space Admin., Wash., D.C. BOGOSLOVSKAYA, L. S., L. M. VOTROGOV, AND T. N. SEMENOVA. 1981. Feeding habits of the gray whale off Chukotka. Rep. Int. Whaling Comm. 31:507-510. Braham, H. W., B. D. Krogman, and G. M. Carroll. 1984. Bowhead and white whale migration, distribution, and abundance in the Bering, Chukchi, and Beaufort Seas, 1975- 78. U.S. Dep. Commer., NOAA Tech. Rep. NMFS SSRF-778, 39 p. Braham, H., B. Krogman, J. Johnson, W. Marquette, D. Rugh, M. Nerini, R. Sonntag, T. Bray, J. Brueggeman, M. Dahlheim, S. Savage, and C. Goebel. 1980. Population studies of the bowhead whale {Bakierwi mys- ticetus): results of the 1979 spring research season. Rep. Int. Whaling Comm. 30:391-404. Braham, H. B. Krogman, S. Leatherwood, W. Marquette, D. Rugh, M. Tillman, J. Johnson, and G. Carroll. 1979. Preliminary report of the 1978 spring bowhead whale research program results. Rep. Int. Whaling Comm. 29:291- 306. Brown, R. 1868. Notes on the history and geographical relations of the Cetacea frequenting Davis Strait and Baffin's Bay. Proc. Zool. Soc. Lond., 35:533-556. Carroll, G. M., and J. R. Smithhisler. 1980. Observations of bowhead whales during spring migra- tions. Mar. Fish. Rev. 42(9-10):80-85. Dahlheim, M., T. Bray, and H. Braham. 1980. Vessel survey for bowhead whales in the Bering and Chukchi Seas, June-July 1978. Mar. Fish. Rev. 42(9-10):51- 57. Darling, J. D. 1977. Aspects of the behavior and ecology of Vancouver Island gray whales, Eschrichtius glaucus Cope. M.S. Thesis, Univ. Victoria, 200 p. Davis, R. A., and W. R. Koski. 1980. Recent observations of the bowhead whale in the eastern Canadian High Arctic. Rep. Int. Whaling Comm. 30:439-444. Dorsey, E. M. 1983. Exclusive adjoining ranges in individually identified minke whales (Balaewyptera acutorostrata) in Washington state. Can. J. Zool. 61:174-181. Durham, F. E. 1972. Greenland or bowhead whale. In A. Seed (editor). Baleen whales in eastern North Pacific and arctic waters, p. 10-14. Pacific Search Press, Seattle, WA. EVERITT, R. D., AND B. D. KROGMAN. 1979. Sexual behavior of bowhead whales observed off the north coast of Alaska. Arctic 32:277-280. Fay, F. H. 1963. Unusual behavior of gray whales in summer. Psychol. Forsch. 27:175-176. FRAKER, M. a., and J. R. BOCKSTOCE. 1980. Summer distribution of bowhead whales in the eastern Beaufort Sea. Mar. Fish. Rev. 42(9-10):57-64. Gould, J. L. 1982. Ethology, the mechanisms and evolution of behavior. W. W. Norton & Co., N.Y. 544 p. 375 FISHERY Bl'LLETIN: VOL. Ki. NO. S Hain, J. H. W., G. R. Carter, S. D. Kraus, C. A. Mayo, and H. E. Winn. 1982. Feeding behavior of the humpback whale, Megaptera navaeangliae, in the western North Atlantic. Fish. Bull., U.S. 80:259-268. Herman, L. M.. and R. C. Antinoja. 1977. Humpback whales in the Hawaiian breeding waters: Population and pod characteristics. Sci. Rep. Whales Res. Inst 29:59-85. Johnson, J. H., H. W. Braham, B. D. Krogman, W. M. Marquette, R. M. Sonntag, and D. J. Rugh. 1981. Bowhead whale research: June 1979 to June 1980. Rep. Int Whaling Comm. 31:461-475. Johnson, M. L., C. H. Fiscus, B. T. Ostenson, and M. L. Barbour. 1966. Marine mammals. I-n N. J. Wilimovsky and J. N. Wolfe (editors). Environment of the Cape Thompson region, Alaska, p. 877-924. U.S. At. Energy Comm., Oak Ridge, TN. JuRASZ, C. M., and v. p. Jurasz. 1979. Feeding modes of the humpback whale, Megaptera nm^aeangliae, in Southeast Alaska. Sci. Rep. Whales Res. Inst. 31:69-83. Kraus, S. D.. J. H. Prescott, P. V. Turnbull, and R. R. Reeves. 1982. Preliminary notes on the occurrence of the North Atlan- tic right whale, EvhaUiena glacialis. in the Bay of Fundy. Rep. Int. Whaling Comm. 32:407-411. Ljungblad, D. K., M. F. Platter-Rieger, and F. S. Shipp, Jr. 1980. Aerial surveys of bowhead whales. North Slope Alaska: final report fall 1979. Nav. Ocean Syst Cent Tech. Doc. 314, 181 p. Naval Ocean Systems Center (Code 5131), San Diego, CA. Lockyer, C. H., and S. G. Brown. 1981. The migration of whales. In D. J. Aidley (editor). Animal migration, p. 105-137. Soc. Exp. Biol. Semin. Sen 13, Cambridge Univ. Press. LowRY, L. F., AND J. J. Burns. 1980. Food utilized by bowhead whales near Barter Island, Alaska, autumn 1979. Mar. Fish. Rev. 42(9-10):88-91. LowRY, L. F., K. J. Frost, and J. J. Burns. 1978. Food of ringed seals and bowhead whales near Point Barrow, Alaska Can. Field-Nat 92:67-70. Marquette, W. M., H. W. Braham, M. K. Nerini, and R. V. Miller. 1982. Bowhead whale studies, autumn 1980 -spring 1982: harvest, biology and distribution. Rep. Int Whaling Comm. 32:357-370. Matsushita, T. 1955. Daily rhythmic activity of the sperm whale in the Antarctic Ocean. Bull. Top. Soc. Sci. Fish. 20:770-773. Nemoto, T. 1970. Feeding pattern of baleen whales in the ocean. IvJ.U. Steele (editor). Marine food chains, p. 130-141. Oliver and Boyd, Edinburg. Nerini, M., H. Braham, W. Marquette, and D. Rugh. 1984. Life history of the bowhead whale. J. Zool. (Lond.). 204:443-468. Nerini, M. K., and J. S. Oliver. 1983. Gray whales and the structure of the Bering Sea benthos. Oecologia 59:224-225. NoRRis, K. S., R. M. Goodman, B. Villa-Ramirez, and L. Hobbs. 1977. Behavior of California gray whale, Eanhrichtius robua- tns, in southern Baja California, Mexico. Fish. Bull., U.S. 75:159-172. NoRRis, K. S., B. Villa-Ramirez, G. Nichols, B. Wursig, and K. Miller. 1983. Lagoon entrance and other aggregations of gray whales (Eschrichtms rotnistus). In R. Payne (editor). Communica- tion and behavior of whales, p. 259-298. AAAS Selected Symp. 76, Westview Press, Boulder, CO. Payne, R. 1972. Swimming with Patagonia's right whales. Nat Geogr. 142:576-587. 1976. At home with right whales. Nat Geogr. 149:322-339. 1983. Intnxiuction. In R. Payne (editor), Communication and behavior of whales, p. 1-6. AAAS Selected Symp. 76, Westview Press, Boulder, CO. Payne, R., and E. M. Dorsey. 1983. Sexual dimorphism and aggressive use of callosities in right whales (Eubalaena aristralis). In R. Payne (editor). Communication and behavior of whales, p. 295-329. AAAS Selected Symp. 76, Westview Press, Boulder, CO. Pivorunas, a. 1979. The feeding mechanisms of baleen whales. Am. Sci. 67:432-440. Rice, D. W. 1977. A list of the marine mammals of the world. U.S. Dep. Commer., NOAA Tech. Rep. NMFS SSRF-711, 15 p. Richardson, W. J., M. A. Fraker, B. Wursig, and R. S. Wells. In press. Behaviour of bowhead whales, Balaena myMicetua. summering in the Beaufort Sea: reactions to industrial activities. Biol. Conserv. Rugh, D. J., and J. C. Cubbage. 1980. Migration of bowhead whales past Cape Lisburne, Alas- ka. Mar. Fish. Rev. 42(9-10):46-51. Rugh, D. J., and M. A. Fraker. 1981. Gray whale (Earhrichtius robustus) sightings in eastern Beaufort Sea. Arctic 34:186-187. Saayman, G. S., C. K. Tayler, and D. Bower. 1973. Diurnal activity cycles in captive and free- ranging Indian Ocean bottlenose dolphins {Tvrsiopg aduncus Ehren- burg). Behaviour 44:212-233. Sauer, E. G. F. 1963. Courtship and copulation of the gray whale in the Bering Sea at St Lawrence Island, Alaska. Psychol. Forsch. 27:157-174. Schevill, W. E., and R. H. Backus. 1960. Daily patrol of a Megaptera. J. Mammal. 4 1 :279-28 1 . Scoresby, W., Jr. 1820. An account of the Arctic regions, with a histor\' and description of the northern whale-fisher>'. Archibald Con- stable and Co., Edinburgh: 2 Vols. Reprinted (1969) by David and Charles (Holdings) Ltd., Newton Abbot Devon, Engl. Shane, S. H. 1980. Occurrence, movements, and distribution of bottlenose dolphin, Tursiops truncatiis, in southern Texas. Fish. Bull., U.S. 78:593-601. SUMICH, J. L. 1983. Swimming velocities, breathing patterns, and estimated costs of locomotion in migrating gray whales, Eschrichtms robustus. Can. J. Zool. 61:647-652. Taber, S., and p. Thomas. 1982. Calf development and mother-calf spatial relationships in southern right whales. Anim. Behav. 30:1072-1083. Townsend, C. H. 1935. The distribution of certain whales as shown by logbook records of American whaleships. Zoologica (N.Y.) 19:3-50. Tyack, p., and H. Whitehead. 1983. Male competition in large groups of wintering hum|>- back whales. Behaviour 83:132-154. Watkins, W. a., and W. E. Schevill. 1976. Right whale feeding and baleen rattle. J. Mammal. 57: 376 WI'RSICKTAL.: BKHAVIOK OK HOWIIKAl i WIIAI-KS 58-66. dolphins. Ph.D. Thesis, State Univ. New York, Stony Brook, 1979. Aerial observation of feeding behavior in four baleen 326 p. whales: Eubalaena glacialvt, Balaenoptera borealis, Mega- WiJRSlG, B., E. M. DoRSEY, M. A. P'raker, R. S. Payne, W. .1. ptera nomeangliae, and Balaenoptera physalus. J. Mammal. Richardson, and R. S. Weels. 60:155-163. 1984. Behavior of bowhead whales, Balaena mysticetus, WuRSiG, B. summering in the Beaufort Sea: surfacing, respiration and 1978. On the behavior and ecology of bottlenose and dusky dive characteristics. Can. J. Zool. 62:1910-1921. 377 FOOD HABITS OF BAIT-CAUGHT SKIPJACK TUNA, KATSUWONUS PELAMIS, FROM THE SOUTHWESTERN ATLANTIC OCEAN Lisa Ankenbrandt' ABSTRACT Stomach cdntents ofskipjai'k tuna captured in 1981-82 by live jiole-aml-lino vessels off the southern coast of Brazil were analyzed for the presence of larval and juvenile skipjack tuna. The percentage frequency of occurrence, perc'ent number, and percent volume were evaluated. Of the 1,041 stomachs that were exam- ined for food. 48B were empty. The mean volume of focxi in all stomachs analyzed was 3ti.9 niL. of which 18.9 mL was bait and 18.0 mL was prey. The g-onostomatid Moiirolicii.'i muelleri and the euphausiid Euphauxin simHi^ were the principal foods. Other important food.s were the chub mackerel, Sramberjaponiem; the fripjate tuna, Auj-is thuziird; ^jem- pylids: trichiurids; and carangids. In the study area, adult skipjack tuna were not found to feed on their young. Kruskall-Wallis nonparametric one-way analysis of variance was used to test for differences in the mean volumetric ratios of food items in relation to skipjack size. The percentage oiE. xiwUv^ in the diet was found to decrease, while the proportion of M. muelleri was found to increase with increasing skipjack size. Seasonal variations in the diet were also examined and discussed. Apparently the anatomy of their gill raker apparatus allows skipjack to ingest a wide variety of prey ts'pes above a minimum size. These variations in the food can be attributed to the number and size of the prey species in an area. A Brazilian skipjack pole-and-line fishery has been developing in the Rio de Janeiro area since 1979 (Fig. 1). Because skipjack tuna, Katswonus pelamis, is one of the major tuna species harvested at maxi- mum sustainable yield in the tropical and subtropical oceans (Kearny 1976; Evans et al. 1981), estimation of the fishery potential requires information on the distribution and concentration of its spawning stock. One technique used to determine the existence of a spawning stock is to quantify the distribution of its larvae. Obviously, the presence of large numbers of larvae would indicate a spawning stock occupies an area. Knowledge of the distribution and abundance of juvenile skipjack tuna is limited. Occasionally, speci- mens have been found in experimental plankton hauls or in the stomachs of apex predators (Kearny 1976). From ichthyoplankton surveys, Matsuura (1982) and Nishikawa et al. (1978) reported larvae in warm tropical waters north of the study area (Fig. 1), and juvenile skipjack tuna have been found in the stomachs of adult skipjack tuna captured off west Africa and in the Caribbean (Suarez-Caabro and ■Southwest Fisheries Center La JoUa Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271. La Jolla, CA 92038; present address: School of Fisheries, University of Washington, Seattle, WA 98195. Duarte-Bello 1961; Klawe 1961; Dragovich 1970; Dragovich and Potthoff 1972). Their occurrence in the diet of central and south Pacific skipjack tuna caught by pole-and-line has been used to deduce their distribution and abundance (Waldron and King 1963; Nakamura 1965; Argue et al. 1983). — ~~ ~~ ~~ ■H "~ ^^^ BRAZIL ^ / 1 'M 5 w. ((• 1 i i /] / <^ tS:^-^ ) ■ Adult Occurrence A ^ Larval Occurrence 'TI J 1 II II 1 1 1 1 1 1 1 1 1 1 1 1 II 20°S 30°S 60°W 500W 40°W 30°W 40°S Manuscript accepted October 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. Figure 1.- Solid area indicates fishing localities from where skip- jack tuna stomachs were obtained. Hatched area shows larval occur- rence (Matsuura 1982). 379 FISHERY BILLETIN: \( )L. Ki. NO. -.i Dragovich (1969) reviewed existing information on the food habits of Atlantic skipjack tuna. Since that time food habits have also been reported in studies by Dragovich (1970) and Dragovich and Potthoff (1972) for skipjack from the East and West Atlantic and by Batts (1972) for skipjack in North Carolina waters. Zavala-Camin (1981) examined predator- prey interactions of fishes, including skipjack cap- tured north of the area in this study. The primary objective of this study was to discover if skipjack tuna feed upon their young. The presence of juveniles in bait-caught skipjack stomachs would verify the study area as a spawning-rearing ground. Knowledge of the prey and their relative importance also contributes to the understanding of prey- predator interactions, which affect population distributions and fluctuations. MATERIALS AND METHODS Stomach samples for this study were collected on a monthly basis from October 1981 to December 1982 from skipjack tuna caught off Rio de Janeiro (Fig. 1). National Marine Fisheries Service (NMFS) person- nel collected stomachs from frozen fish transhipped to Puerto Rico, and Superintendencia do Desenvol- viemento da Pesca (SUPEDE) personnel sampled fish landed locally in Rio de Janeiro. Fish from the Puerto Rican source were caught within 1 mo prior to sampling; fish from the Brazilian source were sam- pled 3 to 5 d after the recorded catch date. The sam- pling design required collecting about 15 stomachs from each 10 cm length group, measured to the nearest cm per month. However, the number of stomachs collected was dependent on the catch-size distribution. Once the stomach was removed from the fish, it was preserved in 10% buffered Formalin- and shipped to the Southwest Fisheries Center (SWFC) for analysis. Stomachs were examined from 1,041 fish between 44 and 81 cm fork length. In the laboratory each stomach was opened. The volume of the food bolus was measured, and the contents were identified to the lowest possible taxon. The taxonomic groupings were then measured by volumetric displacement. and the individuals counted. Whole undigested fish were identified by comparing external characters with those described in published keys or with iden- tified museum specimens from Scripps Institution of Oceanography, La Jolla, CA. Digested animals, par- ticularly juvenile scombrids, were identified by verte- bral, gill raker, and fin ray counts, as well as other skeletal characteristics, described by Potthoff and Richards (1970), Miller and Jorgenson (1973), and other published keys. Cephalopods were identified by comparing beak characters with published illustra- tions, descriptions, and keys (see Wolff 1981). Crus- taceans and other invertebrates were identified by specialists from Scripps Institution of Oceanography and SWFC. The occurrence of bait in the stomachs may have biased the relative importance of fish in the diet. The bait primarily consisted of Sardinella brasiliensis, Harengula jaguana, and Engraulis anchoita; however, other fish families may have been included in the captured bait. The sardines were readily iden- tifiable from their external characters and usually were undigested. The anchovies, in contrast, were often quite digested, creating difficulties in identifi- cation. Gary Nelson^ nevertheless was able to verify these fish as Engraulis anchoita. Although the least digested item in the stomach was usually the last meal (bait), stomachs were removed from a few days to 1 mo after capture, and presumably postmortem digestion occurred. As a result, the degree of diges- tion was not a reliable indication of distinguishing bait from natural prey. The time required for com- plete gastric evacuation of smelt fed to skipjack tuna is estimated to be 12 h (Magnuson 1969). Although the bait was captured in nets from bays and estuaries (Rinaldo4), Matsuura et al. (1978, 1981) have con- firmed that a spawning stock of £■. anchoita does ex- ist in waters inhabited by skipjack tuna. It is unlikely that the sardines served as prey for skipjack. However, I could not distinguish between E. an- choita consumed as natural food and as bait. There- fore, although these species were considered bait, some may have been ingested as natural food. Bait was not considered prey, and stomachs containing only these species were treated as empty. Two methods of analysis were employed to rank the food items in terms of availability and impor- tance to the skipjack tuna: 1) An index of relative importance (IRI) was calcu- lated for each prey type in terms of numbers, volumes, and fre(|uencies (Pinkas et al. 1971): IRI = {N + V)F ''Reference to trade names does not innply endorsement by the Na- tional Marine Fisheries Service, NOAA. 'G. J. Nelson, Department of Ichthyolojjx-. American Museum of Natural History, New York, NY 10024. pers. commun.. May 1982. ■•R. R. Rinakid, Southwest Fisheries Center La .lolla Laboratory, National Marine Fisheries Service, NOAA. I'.O. Box 271, La Jolla, CA 92038, pers, commun., .June 1982, 380 ANKKNBRANDT: FdOD HABITS OF SKIPJACK TINA v/hereN = numerical percentage V = volumetric percentage F = frequency of occurrence percentage. 2) The mean volumetric ratio measurement (An'RAT) was used to illustrate the biomass impor- tance of prey items without the numeric exaggera- tion implicit in the IRI (John Hedgepeth^). The MVRM was calculated from the volumetric analysis of individual stomachs with each prey item contribu- ting to the total stomach volume. MVRM for each food type is expressed as M\'RM = r^ X 100 = mean volumetric percentage of prey j to the total volume of n stomachs where N = number of stomachs in a given strata y, J = volume of prey type j in stomach i V: = Z V: , = total volume of stomach i j=i 'J y.- 'J i _ ratio of prey j to the total vol- V^ ume of stomach i '7 n = '=^^ _ mean volumetric ratio of n prey J to the total volume of n stomachs. Both the IRI and the MVRM, which examine dif- ferent aspects of the diet, were used to evaluate seasonal variations in skipjack tuna food habits. The IRI presents a biased estimate caused by the numerical percentage; the relative importance of small numerous organisms, like euphausiids, is exag- gerated in the IRI because of their high numbers, when actually they may represent the same food value as a few large fish. The M\^RM is an expression of frequency of occurrence and volume without a numeric bias, but does not provide any information on prey abundance. The IRI contains information on the availability of the prey in the environment in terms of numbers, while the MVRM provides an in- dication of its energetic importance to the fish. The MVRMs, were stratified by fish length and y. B. Hedgepeth, Southwest Fisheries Center La Jotta Laboratorj', National Marine Fisheries Ser\'ice, NOAA, P.O. Box 271, La JoUa, CA 92038, pers. commun., April 1982. annual quarter (Fig. 2), and tested with the Krus- kal-Wallis nonparametric one-way analysis of variance to evaluate differences in diet with changes m size. >- O z UJ 3 o UJ 300- — ALL MONTHS 200- 100- n - n 100- I I T T T QUARTER 4 75- 50- 25- 1 1 Ci - rn u 75- I I I I I I 1 I QUARTER 3 50- 25- n - n yj 100- I I I I I T T- QUARTER 2 75 - 50- 1 25- n -1 u 75- I 1 1 I 1 QUARTER 1 50- 25- 0 - ' — ^ ' — ^ r — '• i ., ^ * .^=^ 40 50 60 70 80 LENGTH (cm) Figure 2. - Length-frequency distribution of skipjack tuna from which stomachs were collected. 381 FISHERY BULLETIN: VOL. 83, NO. 3 RESULTS Food Composition Of the 1,041 stomachs that were examined, 436 were empty. The mean volume of food in all stomachs examined was 36.9 mL, of which 18.9 mL was bait and 18.0 mL was prey. A complete list of the stomach contents in terms of numbers, volume, and frequency is presented in Appendix Table 1. No larval or juvenile skipjack were found in the stomach contents. Overall contributions of each category are presented in Figure 3. In terms of the M\''RM, the gonostomatid Mauro- licwi muelleri was the major prey item (MVRM = 26.7%). The euphausiid Ewphausia similis, with the highest IRI, was also important {IRI = 1 ,998). These items were major constituents of the diet throughout the year. Other important fishes in terms of both the ALL MONTHS ME«N % VOLUME (ranki 17 4 7 3 5 2 25 1 14 8 4 0 PERCENT FREQUENCY OF OCCURRENCE FiGiiRE .3. -Index of relative importance (IRI) plots for selected food items of skipjack tuna caught during 1982. The food categories are ranked in terms of IRI and MVRM. JANUARY-MARCH 5 60 X o *o ^ Tl ^ e I SPECIES Euphausia similts Maurolicus muelleri Auxis thazard Scomber japonicus FISH CRUSTACEANS CEPHALOPODS IRI (rank) 1756 49 (I) 228 60 (3) 3805^^) 47 31 <5) 116 29^2) 58 92 (4) 1.57 (6) MEAN % VOLUME (rank) 19 80 (3) 14 IS (4) Included in fish 1008 (5) 29 55 (1) 23 47 (2) 3.02 (6) APRIL-JUNE • Auxis thazard included in fish for ranking ~ combined IRI = 309.66 70 73 24.3 22 6 27 PERCENT FREQUENCY OF OCCURRENCE S 60 O 40 r SPECIES IRI MEAN % (rank! VOLUME (rank) - 1 Euphausia similis 275206(1) 38.32(1) jLiS^ 2 Maurolicus muelleri 771 62 (2) 25 96 (2) „ •A**^ 3 Caranx ruber 33.8 1*(3) Included in fish / / 4 FISH 30.14*(3) 20 02 (3) S CRUSTACEANS 6.42 (4) 8 16 (4) 6 CEPHALOPODS 0.584 (5) 7 4 (5) - 4> 1 / < • Caranx ruber included in fish for ranking - combined IRI = 135.39 - '^X^^^ - ^ -^ 1 1 1 1 13.3 59 90 90 24 PERCENT FREQUENCY OF OCCURRENCE JULY-SEPTEMBER SPECIES Euphausia similis IRI (rank) MEAN % VOLUME (rank) 1037 32 (2) 9 60 (3) 3710 17 (1) 55 40 (1) 34 45 (4) 5 70 (5) 7 60*(3) Included in liah 7 1.76*(3) 22.87 (2) 7.31 (5) 6.42 (4) 0 036 (6) 0.01 (6) • Benthodesmus sp. included in fish for ranking combined IRI = 127.4 OCTOBER-DECEMBER MEAN % VOLUME (rank) 34 I 3.7 3 7 18 4 9 2 0.4 PERCENT FREQUENCY OF OCCURRENCE 17 2 23 2 19 3 33 5 17 2 9 2 PERCENT FREQUENCY OF OCCURRENCE Figure 4. -Index of relative importance (IRI) plots for selected food items of skipjack tuna. The complete data are divided into four 3mo quarters (I-IV). The food categories are ranked in terms of IRI and mean percent volume. 382 ANKHNHKANPT: R)()I) HAHl'I'S ( )K SKIIMACK Tl'NA IRI and M\'RM were chub mackerel, Scomber japonicus. and Thyrsitops lepidopoides*^ (Fig. 3). Crustaceans other than A\ fiirriilis occurred frequent- ly in the stomachs (F = 22.6%), but as a relatively low percentage of the total volume {V = 2.0%). Cephalopods were usually insignificant in the diet (see below). Pteropods, siphonophores, beetles, rocks, and unidentified materials were the consti- tuents of the miscellaneous category (App. Table 1). Seasonal Variations The data were divided into four quarters: January- March 1982 (I), April-June 1982 (II), July-September 1982 (III), October-December 1981 and 1982 (IV). The results (App. Tables 2-6) are illustrated in Figures 4 and 5 both with the IRI and the MVRM of •^Identified by Y. Matsuura, Universidade de Sao Paulo, Institiito Oceanografico, Sao Paulo, Brazil, October 1984. dominant food items in each quarter. Note that items important in one quarter may be negligible or absent in another. When evaluated in terms of MVRM, the prey ranks sometimes did not coincide with those determined by the IRI (Figs. 4, 5). Based on the IRI, E. similvi was the dominant food in the first quarter, followed by other fish and M. muelleri (Fig. 4). According to the MVRM, other fishes, other crusta- ceans, and E. similis were ranked first, second, and third in importance, respectively. The importance of E. similis in this quarter based on the IRI was exag- gerated by their high frequencies of occurrence. Scomber' japonicus and frigate tuna, Auxis thazard, were secondary in importance to M. muelleri as the main fish species consumed. The rankings of the food categories in the second quarter were the same for both the IRI and the MVRM (Fig. 4). Euphausia similis and M. muelleri were the dominant food items, followed by Caranx ruber. Figure .5. -Relative importance based on mean volumetric ratio of selected food items for skipjack tuna grouped by length for each quarter (I-IV) and all months. * indicates a significant difference (P < 0.05) in the mean percent volume of that food item by length when tested with the Kruskal-Wallis nonparametric one-way analysis of variance. 383 FISHERY BULLETIN: VOL. 83, NO. 3 The principal food item in the third quarter based on both measurements was M. muelleri (Fig. 4). The ranks for the other items did not correspond. Eu- phausia similis was second in importance according to the IRI, but ranked third next to other fish based on the MVRM. Scomber japonicus and Benthodes- mus sp. were the predominant species consumed in the fish category. The IRI and MVRM ranks in the fourth quarter corresponded with the exception of the principal prey type (Fig. 4). Again, E. similis ranked first ac- cording to the IRI, but Thyrsitops lepidopoides was the primary food item based on the MVRM. Cepha- lopods, mainly Argonauta sp., were consumed in significant proportions in this quarter {IRI = 31.7, MVRM = 7.2%). In summary, M. muelleri and E. similis predom- inated in the skipjack tuna diet during all quarters (Fig. 3). With the exception of the second quarter, S. japonicus was an important food item. Benthodesmus sp., C. ruber, A. thazard, as well as the cephalopod, Argonauta sp., also proved important in specific quarters. The importance of T. lepidopoides in Figure 3 was exaggerated by its predominance in the fourth quarter. Variations with Size As might be expected, basic dietary changes occur as the skipjack tuna grow. Nakamura (1965), Alver- son (1963), Batts (1972), Dragovich and Potthoff (1972), and Wilson (1982) observed a decrease in the relative importance of crustaceans and an increasing importance of fish in the diet, as the skipjack size in- creased. To evaluate the relationship between size and food habits, the skipjack were arbitrarily divided into seven 5 cm groups (Fig. 2). For each prey category the MVRMs were stratified by size group and (quarter (App. Tables 6-10, Fig. 5). Trends reported in the results for length groups > 70 cm may not represent feeding habits of skipjack tuna from the study area because the sample sizes were too small (Fig. 2. App. Table 6). There were no significant differences (P < 0.05) in diet and size in the first quarter except in the amount of other fish consumed (Fig. 5, App. Table 7). The MVRM of this category increased from 28.7% in skipjack 45.0-49.9 cm to 91.7% in fish 75.0-79.9 cm. In the second quarter the proportions of other fish, other crustaceans, E. similis, and M. muelleri signi- ficantly changed with size (Fig. 5, App. Table 8). The larger skipjack tuna ate more fish {MVRM = 1 .7% in the 45.0-49.9 cm size class toMVRM = 54.1% in the 75.0-79.9 cm size class) and more M. muelleri {MVRM = 15.4% to MVRM = 30.3%) than the smaller skipjack. As their size increased, skipjack decreased their consumption of E. similis from MVRM = 76.9% to MVRM = 4.5%. There was a significant difference between size classes {P < 0.05) in the MVRM of other crustaceans in the diet, but this difference seemed uncorrected to increases in size. In the third quarter there were no significant dif- ferences in the diet with increasing size except that the MVRM of E. similis decreased from 25.0% to 8.0% (Fig. 5, App. Table 9). Thyrsitops lepidopoides was eaten by the smaller skipjack (45-59.9 cm) only in the fourth quarter (Fig. 5, App. Table 10). Although there were significant differences {P < 0.05) in the diets between the seven size groups during this period, these differences again seemed unrelated to increasing size. In summary, when the data on T. lepidopoides were included with the rest of the fish data, there were no significant differences between size groups in the proportions of other fish consumed throughout the year (Fig. 5, App. Table 6). The MVRM of E. similis in the diet decreased from 42.5% in skipjack tuna 45.0-49.9 cm to 0.0% in skipjack 75.0-79.9 cm. There were significant differences in the percent- ages of M. muelleri and S. japoniciis between the size classes. There were no significant differences in the MVRM of other crustaceans with changes in size. As reported in the studies of the food habits of skipjack tuna referred to above, the stomach con- tents of skipjack from this area indicated the basic dietary changes associated with increasing size: a significant decrease in the proportion of E. similis, the predominant crustacean prey, and an increase in proportion of M. muelleri, the predominant fish prey. DISCUSSION Several studies have reported that skipjack tuna feed predominantly on euphausiids and gonostoma- tids. Dragovich and Potthoff (1972) reported that the gonostomatid Vinciguerrin nimbario contributed 44.7% by volume to the diet of skipjack tuna from the Gulf of Guinea. Zavala-Camin (1981) reported M. muelleri and euphausiids as dominant food items in 36 stomachs of skipjack caught off Brazil. Alverson (1963) reported skipjack tuna captured in the eastern tropical Pacific fed primarily on euphausiids (47% by volume, in 37% of the stomachs), followed in impor- tance by the gonostomatid Vinciguerria lucetia (10% by volume). The abundance of euphausiids in 384 ANKKNHKANltT: FOOH HABITS OK SKIIMACK TINA the stomachs of skipjack, compared with larger scombrids. may be a result of smaller gill raker gaps in skipjack (Magnuson and Heitz 1971). The importance of other fishes as food for western Atlantic skipjack tuna observed in this study has been previously reported. Dragovich (1970) found a predominance of fish in the stomachs of skipjack caught off the eastern United States and the Carib- bean. Suarez-Caabro and Duarte-Bello (1961) found that fishes constituted 75% of the total volume, followed by squid (23%) and crustaceans (2%), in the stomachs of Cuban skipjack. Zavala-Camin (1981) observed that fish constituted 38.9%, crustaceans 22.2%, and mollusks 2.8% of the total stomach volume of Brazilian skipjack. CONCLUSIONS The multiplicity of prey found in this as well as other studies indicates that tunas are perhaps non- selective feeders, and stomach contents are probably determined by prey availability (Hotta and Ogawa 1955; Alverson 1963; Batts 1972; Perrin et al. 1973; Argue et al. 1983). Therefore, if the larval and juvenile skipjack were available in significant numbers, then one would expect them to occur in the diet of the adults. Their absence in the diet was caused by two possi- ble results. First, the young remained among the unidentified portion of the stomach contents; however, skipjack tuna have distinctive vertebral characteristics which were probably not discounted in the analysis (Potthoff and Richards 1970). Second, the adults did not spawn in the study area. Young skipjack should be found in the stomach contents of spawning adults (Argue et al. 1983). Goldberg and Au^ found no evidence of spawning in skipjack col- lected from the Brazilian fishery. These results are consistent with the absence of larval and juvenile skipjack in the diet of the adults in this study. The southernmost distribution boundary for larval skipjack tuna is the 24 °C surface isotherm (Argue et al. 1983). Matsuura (1982) found no larval skipjack in ichthyoplankton surveys south of lat. 21 °S in this area, where temperatures range from 21° to 24°C (Evans et al. 1981). ^Goldberg, S. R., and D. W. K. Au. 1983. The spawning schedule of skipjack tuna from southeastern Brazil as determined from histological examinations of ovaries, with notes on spawning in the Caribbean. Prepared for the International Skipjack Year Pro- gram conference of the International Commission for the Conser- vation of Atlantic Tunas. June 21-25, 1983, Tenerife, Canary Islands, Spain, 31 p. Manuscript in preparation; Department of Biology, Whittier College, Whittier, CA 90608. These results are consonant with those of Argue et al. (1983); juvenile skipjack tuna were absent from samples of adult stomachs taken in subtropical south Pacific waters. The adult skipjack in this investiga- tion did not feed on their young. The absence of cannabilism suggests that larvae and juveniles were not significantly abundant to serve as forage of the adults, and therefore probably do not occur in this cooler southern water. ACKNOWLEDGMENTS I wish to thank Ronald Rinaldo for organizing the collection of stomachs from skipjack tuna landed in Puerto Rico and Brazil. Silvio Jablonski (SUDEPE, Brazil), Eugene Holzapfel (NMFS, Puerto Rico), and their employees were responsible for the collection and shipment of the skipjack tuna stomachs. Gareth Nelson, Richard Rosenblatt, Kurt Schaef- fer, and Betsy Stevens assisted in the identification of the fishes. The euphasiid was identified by Ed- ward Brinton. Robert Olson assisted in the identifi- cation of cephalopod beaks. Angeles Alvarino assisted in the identification of other invertebrates. Thanks to Andrew Dizon, John Graves, Ronald Rinaldo, Robert Olson, and Kurt Schaeffer for their criticisms and suggestions in reviewing this paper. LITERATURE CITED Alverson, F. G. 1963. The food of yellowfin and skipjack tunas in the eastern tropical Pacific Ocean. [In Engl, and Span.] Inter-Am. Trop. Tuna Comm. Bull. 7:293-396. Argue, A. W., F. Conand, and D. Whyman. 1983. Spatial and temporal distributions of juvenile tunas from stomachs of tunas caught by pole-and-line gear in the central and western Pacific Ocean. Tuna and Billfish Assessment Programme Technical Report No. 4. South Pacific Commis- sion, Noumea, New Caledonia. Batts, B. S. 1972. Food habits of the skipjack tuna. Katsuwomm pelamis, in North Carolina waters. Chesapeake Sci. 13:193-200. Dragovich, A. 1969. Review of studies of tuna food in the Atlantic Ocean. U.S. Fish Wild!. Serv., Spec, Sci. Rep. Fish. 593, 21 p. 1970. The food of skipjack and yellowfin tunas in the Atlantic Ocean. Fish. Bull., U.S. 68:445-460. Dragovich, A., and T. Potthoff. 1972. Comparative study of food of skipjack and yellowfin tunas off the coast of West Africa. Fish. Bull., U.S. 70:1087- 1110. Evans, R. H., D. R. McLain, and R. A. Bauer. 1981. Atlantic skipjack tuna: influences of mean environmen- tal conditions on their vulnerability to surface fishing gear. Mar. Fish. Rev. 43(6):1-11. HoTTS, H., AND T. Ogawa. 1955. On the stomach contents of the skipjack, Katsuwomis pelamis. Bull. Tohoku Reg. Fish. Res. Lab. 4:62-82. 385 FISHERY BI'LLETIN: VOL. 83. NO. 3 Kearney, R. E. 1976. Some hypotheses on skipjack (Katsuwonus pelamis) in the Pacific Ocean. South Pac. Comm., Occas. Pap. 7, 23 p. Klawe, W. L. 1961. Young scombroids from the waters between Cape Hat- teras and Bahama Islands. Bull. Mar. Sci. Gulf Caribb. 1 1 : 150-157. Magnlfson, J. J. 1969. Digestion and food consumption by skipjack tuna (Kat- suwonus pelamvi). Trans. Am. Fish. Soc. 98:379-392. Magnuson, J. J., AND J. G. Heitz. 1971. Gill raker apparatus and food selectivity among mack- erels, tunas, and dolphins. Fish Bull., U.S. 361-370. Matsuura, Y. 1982. Distribution and abundance of skipjack (Katsuwonus pelamis) larvae in eastern Brazilian waters. ICCAT/SCRS/ 81/70, Coll. Vol. Sci. Pap. Vol. XVII(SCRS-1981), p. 211-213. Matsuura. Y., K. Nakatani, G. Sato, and S. T. J. Tamassia. 1978. Exploracao e avaliacao de estoque de peixes pelagicos no sul do Brasil. Relat. Teen. Conv. FINEP/IOUSP, 46 P- Matsuura, Y., J. C. Amaral, S. T. J. Tamassia, and G. Sato. 1981. Ocorrencia de cardumes de peixes peligicos e a estru- tura oceanografica da regiao entre Cabo de Sao Tome (RJ) e Cananela (SP) em janeiro-fevereiro de 1979. Ser. Doc. Teen. PDP/SUDEPE 36:1-73. Miller. G. L., and S. C. Jorgenson. 1973. Meristic characters of some marine fishes of the west- ern Atlantic Ocean. Fish Bull., U.S. 71:301-312. Nakamura, E. L. 1965. Food and feeding habits of skipjack tuna {Katsiuronus pelamvi) from the Marquesas and Tuamotu Islands. Trans. Am. Fish. Soc. 94:236-242. NisHiKAWA, Y., S. Kikawa. M. Honma, and S. IIeyanagi. 1978. Distribution atlas of lan,-al tuna, billfishes, and related species. Results of larval surveys by R/V Shunyo-Maru and Shoyo-Maru, 1956-1975. Far Seas Fish. Res. Lab., S. Ser., 9:1-99. PiNKAS, L., M. S. Oliphant, and L. K. Iverson. 1971. Food habits of albacore, bluefin tuna, and bonito in California waters. Calif. Dep. Fish Game, Fish Bull. 152, 105 p. Perrin, W. F., R. R. Warner, C. H. Fiscus, and D. B. Holts. 1973. Stomach contents of porpoise, SteTiella spp., and yellow- fin tuna, Thunnua nlbacares, in mixed species aggregations. Fish. Bull., U.S. 71:1077-1092. POTTHOFF, T., AND W. J. RiCHARDS. 1970. Juvenile bluefin tuna, Thunnus thynnus (Linnaeus), and other scombrids taken by terns in the Dr>' Tortugas, Florida. Bull. Mar. Sci. 20:389-413. Suarez-Caabro, J. A., and P. P. Duarte-Bello. 1961. Biologia pesquera del bonito (Katsuwonus pelamis) y la albacora (Thunnus atlantieus) en Cuba \. Inst. Cub. Invest. Tecnol. Ser. Estud. Trab. Invest. 15, 151 p. Wilson, M. A. 1982. The reproductive and feeding behavior of skipjack tuna, Katsuwonus pelamis, in Papua New Guinea waters. D.P.I. Fish. Res. Surv. Br., Papua New Guinea, Rep. 82-4, 38 p. Wolff, G. A. 1981. A beak key for eight eastern tropical Pacific cephalopod species with relationships between their beak dimensions and size. Fish. Bull., U.S. 80:357-370. Zavala-Camin, L. a. 1981 . Habitos alimen tares e distribuicao dos atuns e afins (Os- teichthyes - Teleostei) e suas relacoes ecologicas com outras especies pelagicas das regioes Sudeste e Sul do Brasil. Ph.D. Thesis, Universidade de Sao Paulo, Sao Paulo, Brasil, 237 p. 1982. Datos historicos de areas de crecimiento de listado (Kat- suwonus pelamis) obtenidos por medio de examen de conten- ido estomacal de predadores (informe preliminar). ICC AT/ SCRS/82/48, Coll. Vol. Sci. Pap. Vol. XVII (SCRS-1981), p. 209-210. APPENDIX Appendix Table 1. — List of prey items and other ingested nnaterials found in ttie stomacfis of 1,041 skipjack tuna caugtit off souttiern Brazil, from October 1981 to December 1982. NumI bers Vol ume Occurrence Prey items No. % mL % No. % Crustacea Stomatopoda 12.0 0.003 4.5 0.024 6.0 0.576 Mysidacea Eucopiidae 220.0 0.057 42.4 0.226 22.0 2.113 Lophogastridae 24.0 0.006 1.7 0.009 3.0 0.288 Isopoda 1.0 0.000 0.1 0.001 1.0 0.096 Flabellifera 7.0 0.002 2.1 0.011 7.0 0.672 Amphipoda 22.0 0.006 1.5 0.008 20 0.192 Gammaridea 16.0 0.004 5.1 0.027 10.0 0.961 Euphausiidae 139.0 0.036 3.9 0.021 15.0 1.441 Euphausia sp. 50.0 0.013 0.6 0.003 1.0 0.096 Euphausia similis 368,632.0 94.785 4.895.3 26.122 172.0 16.523 Stylocheiron sp. 1.0 0000 0.3 0.002 10 0.096 Caridea 3.0 0.001 2.3 0.012 3.0 0.288 Macrura Scyllaridae 1.0 0000 0.2 0.001 1.0 0.096 Unid. Ptiyllosoma larvae 1.0 0.000 0.1 0001 1.0 0.096 386 ANKKNBRANnT: FOOD H AMITS OF SKIIMA( 'K Tl'NA APPENDIX Table ^. — Continued. Numbers Volume Occurrence Prey items No. % mL % No. % Brachyura 5.0 0.001 0.7 0.004 1.0 0096 Portunidae 1.0 0.000 0.1 0.001 1.0 0.096 Unid. megalops 102.0 0.026 32.0 0.171 27.0 2.594 Unid zooea 555.0 0.143 30.2 0.161 24.0 2,305 Unid Decapoda 7.0 0.002 0.6 0.003 2.0 0,192 Unid. Crustacea 96.0 0.025 11.0 0.059 26.0 2.498 Mollusca Gastropoda 1.0 0.000 0.1 0.001 1.0 0.096 Cavolina sp. 101.0 0.026 6.0 0.032 7.0 0.672 Cephalopoda Teuthoidea 4.0 0.001 26.0 0.139 4.0 0.384 Thysanoteuthldae 2.0 0.001 10.0 0.053 1.0 0.096 Ommastrephidae 35.0 0.009 55.1 0.294 18.0 1.729 Loliginidae 2.0 0.001 10.5 0.056 2.0 0.192 Histloteuthidae 1.0 0.000 1,0 0.005 1.0 0.096 Onychoteuthidae 13.0 0.003 6.0 0.032 2.0 0.192 Octopoda Argonautidae Argonauta sp. 20.0 0.005 75.9 0.405 11.0 1.057 Unid. Cephalopoda 4.0 0.001 1.3 0.007 3.0 0.288 Insecta Coleoptera 2.0 0.001 2.0 0.011 2.0 0.192 Siphonophora 1.0 0.000 0.9 0.005 1.0 0.096 Algalmidae 8.0 0.002 6.6 0.035 8.0 0.768 Pisces Gonostomatidae Maurolicus muelleri 13,438.0 3.455 8,619.3 45.994 181.0 17.387 Synodontidae 8.0 0.002 16.0 0.085 3.0 0.288 Paralepididae 1.0 0.000 0.5 0.003 1.0 0.096 IVIyctophidae 43.0 0.011 62.3 0.332 5.0 0.480 Exocoetidae 2.0 0.001 2.0 0.011 2.0 0.192 Exocoetus volitans 1.0 0.000 56.0 0.299 1.0 0.096 Scomberesocidae Scomberesox saurus 8.0 0.002 105.0 0.560 6.0 0.576 Belonidae 2.0 0.001 5,0 0.027 2.0 0.192 Macrorhamphosidae 5.0 0.001 6.9 0.037 4.0 0.384 Sygnathidae 1.0 0.000 1.0 0.005 1.0 0.096 Holocentridae 3.0 0.001 4.0 0.021 1.0 0.096 Holocentrus sp. 3.0 0.001 9.0 0.048 1.0 0.096 Carangidae 1.0 0.000 1.0 0.005 1,0 0.096 Selene vomer 8.0 0.002 17.0 0.091 8.0 0.768 Decapterus punctatus 2.0 0.001 46.0 0.245 2.0 0.192 Caranx ruber 70.0 0.018 360.0 1,921 17.0 1.633 Mullidae 11.0 0.003 25.0 0.133 6.0 0.576 Sconnbridae 16.0 0.004 2.5 0.013 3.0 0.288 Auxis thazard 474.0 0.122 223.7 1.194 23.0 2.209 Scomber japonicus 1,474.0 0.379 978.7 5.223 77.0 7.397 Sarda sarda 81.0 0.021 127.0 0.678 8.0 0.768 Gempylidae 176,0 0.045 100.7 0.537 20.0 1.921 Thyrsitops lepidopoldes 2,617.0 0.673 2,348.4 12.532 54.0 5.187 Trichiuridae 11.0 0.003 12.0 0.064 4.0 0.384 Benthodesmus sp. 19.0 0.005 80.9 0.432 8.0 0.768 Unid. Perciforms 24.0 0.006 63.7 0.340 16.0 1.537 Balistidae 1.0 0.000 5.0 0.027 1.0 0.096 Monacanthidae 27.0 0.007 198 0.106 19.0 1.825 Ostraciidae 1.0 0.000 0.5 0.003 1.0 0.096 Molidae Ranzania sp. 3.0 0.001 13.0 0.069 1.0 0.096 Triglidae 1.0 0.000 1.0 0.005 1.0 0.096 Peristedion sp. 1.0 0.000 1.0 0.005 1.0 0.096 Unid. fish 289.0 0.074 163.5 0.872 94.0 9.030 Unid. material 1.0 0000 26.4 0.141 5.0 0.480 Empty 436.0 41.882 Total 388,912.0 18,739.9 1,041.0 387 FISHERY BULLETIN: VOL. 83, NO. 3 Appendix Table 2.— List of prey items and othier ingested materials found in the stomacfis of skip- jack tuna caught during ( Juarter \. Prey items Numbers Volume Occurrence No. % mL % No. % Crustacea f^ysidacea Eucopiidae Lophogastridae Isopoda Flabellifera Amphipoda Euphausiidae Euphausia simills Caridea Brachyura Unid. megalops Unid. zooea Unid. Decapoda Unid. Crustacea IVIollusca Gastropoda Pteropoda Cavolina Cephalopoda Teuthoidea Thysanoteuthidae Ommastrephidae Loliginidae Octopoda Argonautidae Argonauta sp. Insecta Coleoptera Pisces Gonostomatidae Maurolicus muelleri Belonidae Holocentridae Carangidae Selene vomer Mullidae Scombridae Auxis thazard Scomber /aponicus Gempylidae Unid. Perciforms Monacanthidae Ostraciidae Triglidae Peristedior) sp Unid. fish Unid. material Total 112.0 0.0076 19.0 0.013 1.0 0.001 1.0 0.001 20.0 0,014 49.0 0.033 44,070.0 97.624 1.0 0.001 80.0 0.054 549.0 0.372 4.0 0.003 47,0 0,032 68.0 0.046 9.3 0.253 1.5 0.041 0.1 0.003 0.1 0.003 0.5 0.014 2.3 0.063 25.9 57.878 0.5 0.014 22.6 0.615 28.6 0.779 0.1 0.003 8.3 0.226 2.2 0.060 8.0 2.658 2.0 0.664 1.0 0.332 1.0 0.332 1.0 0.332 4.0 1.329 34.0 11.296 1.0 0,332 17.0 5,648 19.0 6.312 1.0 0.332 13.0 4.319 1.0 0.332 2.0 0.001 2.0 0.054 2,0 0.664 2.0 0.001 10.0 0.272 1.0 0.332 5.0 0.003 1.1 0.030 2.0 0.664 1.0 0.001 8.0 0.218 1.0 0.332 2.0 0.001 0.3 0.008 2.0 0.664 1.0 0.001 1.0 0.027 1.0 0.332 1,346.0 0.912 838.0 22.815 29.0 9.635 1.0 0.001 3.0 0.082 1.0 0.332 3.0 0.002 4.0 0.109 1.0 0.332 6.0 0.004 9.0 0.245 6.0 1.993 2.0 0.001 8.0 0.218 1.0 0.332 15.0 0.010 1,5 0.041 2.0 0.664 4270 0.289 189.7 5.165 21.0 6.977 504.0 0.342 225.2 6.131 22.0 7.309 76.0 0.051 48.5 1.320 11.0 3.654 9.0 0.006 7.0 0.191 5.0 1.661 13.0 0.009 9.5 0.259 8.0 2.658 1,0 0.001 0.5 0.014 1.0 0.332 1,0 0.001 1.0 0.027 1.0 0.332 1,0 0.001 1.0 0.027 1.0 0.332 137,0 0.093 76.5 2.083 35.0 11.628 1,0 0.001 26.3 0.716 4.0 1.329 147.577,0 3,673.1 301.0 388 ANKKNHRANliT: FOOD HAHITS OK SKIIMACK Tl'NA Appendix Table 3. — List of prey items and other ingested nnaterials found in the stomachs of skip- jack tuna caught during Quarter II. Numbe rs Volume Occurrence Prey items No. % mL % No. % Crustacea Stomatopoda 1.0 0.001 1.0 0.016 1.0 0.345 Mysidacea Eucopiidae 97.0 0.055 31.9 0.508 9.0 3.103 Lophogastridae 5.0 0.003 0.2 0.003 1.0 0.345 Isopoda Flabellifera 2.0 0.001 1.5 0.024 2.0 0.690 Amphipoda 2.0 0.001 1.0 0.016 1.0 0.345 Gammaridea 6.0 0.003 0.4 0.006 3.0 1.034 Euphausiidae 2.0 0.001 0.2 0.003 2.0 0.690 Euphausia similis 171.843.0 97.352 2,104.3 33.485 61.0 21.034 Caridea 1.0 0.001 0.8 0.013 1.0 0.345 Macrura Scyllaridae 1.0 0.001 0.2 0.003 1.0 0.345 Brachyura Unid. megalops 1.0 0.001 0.5 0.008 1.0 0.345 Unid. zooea 1.0 0.001 0.2 0.003 1.0 0.345 Unid. Decapoda 3.0 0.002 0.5 0.008 1.0 0.345 Unid. Crustacea 30.0 0.017 1.2 0.019 2.0 0.690 Mollusca Gastropoda Pteropoda Cavolina sp. 33.0 0.019 3.8 0.060 6.0 2.069 Cephalopoda Teuthoidea Ommastrephidae 6.0 0.003 8.5 0.135 5.0 1.724 Onychoteuthidae 13.0 0.007 6.0 0.095 2.0 0.690 Siphonophora Algalmldae 8.0 0.005 6.6 0.105 8.0 2.759 Pisces Gonostomatidae Maurolicus muelleri 4,287.0 , 2.429 3,548.0 56.458 38.0 13.103 Synodontidae 8.0 0.005 16.0 0.255 3.0 1.034 Myctophidae 24.0 0.014 61.3 0.975 4,0 1.379 Sygnathidae 1.0 0.001 1.0 0.016 1.0 0.345 Carangidae Selene vomer 2.0 0.001 8.0 0.127 2.0 0.690 Decapterus punctatus 2.0 0.001 46.0 0.732 2.0 0.690 Caranx ruber 70.0 0.040 360.0 5.729 17.0 5.862 Scombridae Auxis t hazard 46.0 0.026 30.0 0.477 1.0 0.345 Scomber japonicus 8.0 0.005 6.0 0.095 2.0 0.690 Unid. Perciforms 1.0 0.001 12.0 0.191 1.0 0.345 Balistidae 1.0 0.001 5.0 0.080 1.0 0.345 Monacanthidae 4.0 0.002 4.1 0.065 3.0 1.034 Unid. fish 9.0 0.005 18.1 0.288 6.0 2.069 Total 176,518.0 6,284.3 290.0 389 FISHERY BULLETIN: VOL. H3, NO. '.i Appendix table 4.— List of prey items and other ingested nnaterials found in ttie stomachs of skipjack tuna caught during Quarter IN. Num bers Vol ume Occurrence Prey items No. % mL % No. % Crustacea Mysidacea Eucopiidae 10.0 0.046 0.8 0.020 4.0 1.843 Isopoda Flabellifera 2.0 0.009 0.3 0.007 2.0 0.922 Euphausiidae 88.0 0.401 1.4 0.034 9.0 4.147 Euphausia sp. 50.0 0.228 0.6 0.015 1.0 0.461 Euphausia similis 15,414.0 70.236 196.3 4.796 30.0 13.825 Unid. Crustacea 5.0 0.023 0.4 0.010 4.0 1.843 Mollusca Cephalopoda Teuthoidea 1.0 0.005 3.0 0.073 1.0 0.461 Pisces Gonostomatidae Maurolicus muellerl 6,239.0 28429 3.289.2 80.369 74.0 34.101 Exocoetidae 2.0 0.009 2.0 0.049 2.0 0.922 Exocoetus volitans 1.0 0.005 56.0 1.368 1.0 0.461 Macrorhamphosidae 2.0 0.009 2.9 0.071 2.0 0.922 Carangidae 1.0 0.005 1.0 0.024 1.0 0.461 Mullidae 6.0 0.027 9.0 0.220 3.0 1.382 Scombridae Scomber laponicus 45.0 0.205 374.0 9.138 8.0 3.687 Sarda sarda 6.0 0.027 4.0 0.098 2.0 0.922 Trichiuridae 11.0 0.050 12.0 0.293 4.0 1.843 Benthodesmus sp. 19.0 0.087 80.9 1.977 8.0 3.687 Unid. Perciforms 4.0 0.018 35.0 0.855 3.0 1.382 fVlonacanthidae 1.0 0.005 1.0 0.024 1.0 0.461 Unid. fish 39.0 0.178 22.8 0.557 21.0 9.677 Total 21,946.0 4,092.6 217.0 390 ANKKNKKANDT: FOOD HAHITS OK SKIIMACK TINA Appendix table 5— List of prey items and other ingested materials found in tfie stomachis of skip- jack tuna caught during Quarter IV. Num bers Volume Occurrence Prey items No. % mL % No. % Crustacea Stomatopoda 11.0 0.026 3.5 0.075 5.0 2.146 Mysidacea Eucopiidae 1.0 0.002 0.4 0.009 1.0 0.429 Isopoda Flabellifera 2.0 0.005 0.2 0.004 2.0 0.858 Amphipoda Gammaridea 10.0 0.023 4.7 0.100 7.0 3.004 Euphausiidae Euphausia similis 37,305.0 87.017 468.8 9.996 47.0 20.172 Stylochelron sp. 1.0 0.002 0.3 0.006 1.0 0.429 Caridea 1.0 0.002 1.0 0.021 1.0 0.429 Macrura Scyllaridae Unid. Phyllosoma larvae 1.0 0.002 0.1 0.002 1.0 0.429 Erachyura 5.0 0.012 0.7 0.015 1.0 0.429 Portunidae 1.0 0.002 0.1 0.002 1.0 0.429 Unid. megalops 21.0 0.049 8.9 0.190 9.0 3.863 Unid. zooea 5.0 0.012 1.4 0.030 4.0 1.717 Unid. Crustacea 14.0 0.033 1.1 0.023 7.0 3.004 Mollusca Gastropoda Pteropoda Cavoliniidae 1.0 0.002 0.1 0.002 1.0 0.429 Cephalopoda Teuthoidea 1.0 0.002 21.0 0.448 1.0 0.429 Ommastrephidae 24.0 0.056 455 0.970 11.0 4.721 Loliginidae 1.0 0.002 2.5 0.053 1.0 0.429 Histioteuthidae 1.0 0.002 1.0 0.021 1.0 0.429 Octopoda Argonautidae Argonauts sp. 18.0 0.042 75.6 1.612 9.0 3.863 Unid. Cephalopoda 4.0 0.009 1.3 0.028 3.0 1.288 Insecta Coleoptera 1.0 0.002 1.0 0.021 1.0 0.429 Siphonophora 1.0 0.002 0.9 0.019 1.0 0429 Pisces Gonostomatidae Mauroltcus muelleri 1.566.0 3.653 944.1 20.130 40.0 17.167 Paralepididae 1.0 0.002 0.5 0.011 1.0 0.429 Myctophidae 19.0 0.044 1.0 0.021 1.0 0.429 Scomberesocidae Scomberesox saurus 8.0 0.019 105.0 2.239 6,0 2.575 Belonidae 1.0 0.002 2.0 0.043 1.0 0.429 Macrorhamphosidae 3.0 0.007 4.0 0.085 2.0 0.858 Holocentridae Holocentrus sp. 3.0 0.007 9.0 0.192 1.0 0.429 Mullidae 3.0 0.007 8.0 0.171 2.0 0.858 Scombridae 1.0 0.002 1.0 0.021 1.0 0.429 Auxis thazard 1.0 0.002 4.0 0.085 1.0 0.429 Scomber japonicus 917.0 2.139 373.5 7.964 45.0 19.313 Sard a sard a 75.0 0.175 123.0 2.623 6.0 2.575 Gempylidae 100.0 0.233 52.2 1.113 9.0 3.863 Thyrsi tops lepldopoldes 2,617.0 6.104 2,348.4 50.074 54.0 23.176 Unid. Percitorms 10.0 0.023 9.7 0.207 7.0 3.004 fVlonacanthidae 9.0 0.021 5.2 0.111 7.0 3.004 Molidae 1.0 0.001 0.5 0.014 1.0 0.332 Ranzania sp. 3.0 0.007 13.0 0.277 1.0 0.429 Unid. fish 104.0 0.243 46.1 0.983 32.0 13.734 Unid. material 0.0 0.000 0.1 0.002 1.0 0.429 Total 42,871.0 4,689.9 233.0 391 FISHERY BULLETIN: VOL. 83, NO. 3 APPENDIX Table 6.— Mean volumetric ratio of selected food items of skipjack tuna divided into 5 cm length groups for all months. Data are 7 ± SD with (n) in parentheses. Range is 0-100%. Length cm Other Other Euphausia Maurolicus Scomber Thyrsitops (n) fish crustaceans similis muelleri laponicus lepidopoides 45.0-49.9(67) 9.51 ±22.5 9.51 ±25.4 42.50 ±48.1 10.35 ±29.5 8.57 ±23.6 16.64 ±36.4 50.0-54.9(155) 10.53 ±27.7 12.62 ±30.9 26.73 ±43.6 13.99 ±33.4 12.44 ±29.5 20.05 ±38.5 55.0-59.9(162) 24.81 ±39.4 11.80 ±29.5 20.40 ±39.5 29.49 ±43.6 5.62 ±19.6 2.70 ±14.2 60.0-64.9(147) 23.48 ±39.2 8.03 ±24.7 18.83±36.5 42.90 ±48.2 2.12±12.2 — 65.0-69.9(55) 42.80 ±47.6 12.43±31.7 4.73±18.7 32.18 ±45.9 0.27 ±2.0 — 70.0-74.9(13) 48.30 ±48.1 9.88 ±27.9 8.57 ±21.9 23.36 ±43.7 2.20 ±79.2 — 75.0-79.9(4) 47.50 ±55.0 27.50 ±48.6 — 25.00 ±50.0 — — Total (603) 18.80 ±35.5 10.96 ±28.6 22.22 ±40.0 26.73 ±42.8 6.24±21.1 10.34 ±27.9 APPENDIX Table 7. — Mean volumetric ratio of selected food items of skipjack tuna divided into 5 cm length groups for Quarter I. Data are 7 ± SD with (n) in parentheses. Range is 0-100%. Length cm Other Other Euphausia Mauroiicus Scomber in) fish crustaceans similis muelleri laponicus 45.0-49.9(12) 28.70 ±33.4 35.15 ±42.5 9.18 ±28.7 7.48 ±25.9 13.89 ±38.4 50.0-54.9(53) 19.87 ±36.8 29.64 ±43.5 24.48 ±43.3 6.09 ± 19.6 16.12±33.5 55.0-59.9(45) 30.23 ±40.9 17.69 ±33.9 20.81 ±38.9 22.96 ±38.1 7.98 ±25.2 60.0-64.9(36) 34.79 ±44.1 16.96 ±35.4 20.14 ±38.9 20.92 ±40.1 19.84 ±46.5 65.0-69.9(4) 48.61 ±56.2 28.24 ±47.9 — — 3.71 ±7.4 70.0-74.9(3) 91.67± 14.4 8.33 ± 14.4 — — — 75.0-79.9(2) — — — — — Total (155) 29.55 ±41.0 23.47 ± 38.9 19.80 + 38.9 14.18±32.1 10.08 ±27.3 Appendix Table 8.— Mean volumetric ratio of selected food items of skipjack tuna divided into 5 cm length groups for Quarter II. Data are r ± SD with (n) in parentheses. Range is 0-100%. Length cm Other Other Euphausia Maurolicus Scomber (n) fish crustaceans similis muelleri laponicus 45.0-49.9(26) 1.65 ±8.4 5.00 ±20.3 76.92 ±43.0 15.38 ±36.8 — 50.0-54.9(12) 5.82 ± 11,5 14.72 ±33.0 65.71 ±48.6 — — 55.0-59.9(35) 20.89 ±38.7 18.13±36.4 35.65 ±49.9 14.71 ±33.9 0.14 ±0,80 60.0-64.9(45) 16.33 ±36.2 4.67 ± 18.5 28.92 ±45.7 45.54 ±49,7 0.32 ±2.1 65.0-69.9(23) 54.07 ±49.1 — 4.49 ±20.8 30.29 ±46.8 — Total (141) 20.02 ±38.1 8.16 + 25.2 38.32 ±48.6 25.96 ±43.3 0.14±1.3 Appendix Table 9. — Mean volumetric ratio of selected food items of skipjack tuna divided into 5 cm length groups for Quarter III Data are r ± SD with (n) in parentheses. Range is 0-100%. Length cm Other Other Euphausia Maurolicus Scomber in) fish crustaceans similis muelleri laponicus 45.0-49.9(4) 25.81 ±49.5 25.00 ±50.0 24.19 ±48.4 25.00 ±50.0 50.0-54.9(27) 11.17±32.0 1.24 ±6.4 11.21 ±32.0 61.56 ±48.5 14.81 ±36.2 55.0-59.9(36) 24.88±41.3 9.22 + 27.9 3.67 ±16.2 59.05 ±47.0 3.17±16.8 60.0-64.9(31) 24.09 + 40.5 3.39 ± 17.9 16.51 ±29.9 53.76 ±46.6 2.26 ± 12.6 65.0-69.9(17) 31.80 ±46.4 17.65 ±39.3 3.89 ±9.8 46.67 ±48.8 — 70.0-74.9(5) 31.27 ±45.5 — 8.00 ± 17.9 60.73 ±53.8 — Total (120) 22.87 + 40.1 6.42 ± 23.6 9.6 ±25.5 55.40 ±47.39 5.70 ±22.7 392 ANKKNBRANnTFOdDHAHrrsnFSKnMACKTUNA Appendix Table 10— Mean volumetric ratio of selected food Items of skipjack tuna divided into 5 cm length groups tor Quarter IV. Data are r ± SD with (n) in parentheses. Range is 0-100%. Length cm Other Other Euphausia Maurolicus Scomber Thyrsitops in) fish crustaceans si mi lis muelleri japonicus lepidopoides 45.0-49.9(25) 5.86 ±13.8 3.43+11.0 25.49 ±39.6 4.26 ± 16.9 12.30±25.1 44.60 ±48.4 50.0-54.9(63) 3.30 ± 13.4 2.77 ±13.0 27.84 ±43.4 2.91 ± 16.3 10.69 + 25.1 49.33 ±47.0 550-59.9(46) 22.45 ±37.5 3.25 ±15.3 21.48 ±40.2 24.01 ±42.9 9.40 ±22.5 9.49 ±25.7 60.0-64.9(35) 20.51 ±35.4 7.29 ±22.1 6.55 ± 19.5 52.48 ±49.9 1.71 ±85.6 — 65.0-69.9(11) 34.11 ±44.1 24.59 ±40.2 8.26 ±27.4 25.45 ±43.9 — — 70.0-74.9(5) 39.31 ±53.8 20.69 ±44.4 14 29±31.9 — 5.71 ± 12.8 — 75.0-79.9(2) 45.00 ±63.6 5.00 ±7.1 — 50.0 ±70.7 — — Total (187) 14.80 ±16.5 5.6 ±19.5 29.17 + 37.8 19.31 ±38.9 8.03±21.2 24.91 ±27.7 393 STOMACH CONTENTS OF YOUNG SANDBAR SHARKS, CARCHARHINUS PLUMBEUS, IN CHINCOTEAGUE BAY, VIRGINIA' Robert J. Medved,^ Charles E. Stillwell,^ and John J. Casey» ABSTRACT nurinp the summer of \WA the stcmiach contents of 414 sandbar sharks captured by j^ll nets, and nnl and reel fishing gear in Chincoteague Bay, Virginia, were examined. The blue crab, CaUinectes mpidui<, occur- red in 67.4% of the stomachs and Atlantic menhaden, Brevoortia tyrannus, occurred in 13.3% of the stomachs. Other species of small crustaceans and fishes were found in < 6.0% of the stomachs, and 17.9%i of the stomachs were empty. Data collected concerning the amount, stage of digestion, and number of food items in the stomachs indicated that feeding occurred during relatively short periods of time separated by long pericKis during which food was digested and no additional food was consumed. Sharks caught in gill nets were found to be in various stages of the feeding cycle and were more representative of the entire population than those caught by rod and reel. In the stomachs of these sharks, crustaceans accounted for nearly twice as much of the mean weight of food as did fish. The mean quantity of food in the stomachs was 0.96% of body weight (BW) and the maximum quantity was 5.28% of BW. The quantity of food in all stomachs was significantly less than the estimated maximum stomach capacity (13.0% BW). Sharks caught between 0130 and 0430 were found to contain considerably more food in their stomachs than sharks caught during other times of the day. The data collected from this study when combined with information concern- ing gastric evacuation will provide the basis for food consumption estimates in this species. Traditionally the management of commercially valuable fisheries has been based on single-species production models and the concept of maximum sus- tainable yield (Hennemuth 1979). Although general- ly accepted as an objective of management, the esti- mation and application of maximum sustainable yield have not provided satisfactory results and have, in fact, led to significant declines of some traditional and highly valued fisheries (Edwards and Henne- muth 1975; Hennemuth 1977; Holt and Talbot 1978). The poor results of single-species models in allocating fishing quotas may be due in part to the fact that they assume no interactions of the target species with other components of the ecosystem. In recent years it has become clear that this assumption is unrealistic and that variables such as competition, predation, and abiotic factors should be considered in any assessment of fishery productivity and potential yields to man. It has been pointed out(Gulland 1978, 1983; Mercer 1982) that the future success of our iMARMAP Contribution No. MED/NEFC 84-02. ^University of Rhode Island, Graduate School of Oceanography, Narragansett, RI; present address: Northeast Fisheries Center Narragansett Laboratory, National Marine Fisheries Service, NOAA, South Ferry Road, Narragansett, RI 02882. ^Northeast Fisheries Center Narragansett Laboratory, National Marine Fisheries Service, NOAA, South Ferry Road, Narragansett, RI 02882. attempts at managing fishery resources will depend, to a large extent, on our ability to develop multi- species production models that adequately account for interactions among species. An important compo- nent of these models is predator-prey interactions. In fact, collection of data on the diets of the major predators is considered absolutely necessary for the progress of multispecies assessment techniques (Hennemuth 1980^; Mercer 1982). Considering their position as one of the most abundant apex predators in the sea, predation by sharks undoubtedly plays a major role in the exchange of energy in the marine environment. In fact, a study by Jones and Geen (1977) has indicated that the spiny dogfish, Squalus acanthias, in British Columbia waters annually con- sumes over 5 times the commercial catch of herring and up to 44% of the total stock. The impact that sharks have on commercial fisheries can only be determined by knowing the diversity of prey items and the biomass of each consumed. While numerous publications on sharks incorporate lists of items found in their stomachs, very little is known about daily ration and the amounts of food consumed an- nually. ■•Hennemuth, R. C. 1980. Research needs for multispecies fisheries. Office of Technology Assessment Workshop, Seattle, WA., 21-23 April. Manuscript accepted October 1984. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 395 FISHERY BULLETIN: VOL. 83, NO. 3 In most investigations the food consumption of fishes has been studied by methods that involve laboratory techniques to estimate various parameters relating to growth, metabolism, diges- tion, and excretion (reviews by: Davis and Warren 1971; Mann 1978). These methods, however, are of limited value for fishes such as sharks that are dif- ficult to maintain in captivity. An alternate method for determining food intake that can be applied to fishes in the wild has been successfully used in several studies (Bajkov 1935; Swenson and Smith 1973; Eggers 1977; Thorpe 1977; Elliott and Persson 1978; Jobling 1981; Stillwell and Kohler 1982; Durbin et al. 1983). This approach requires in- formation concerning the quantity of food found in the stomachs of fishes sampled at regular intervals over 24-h periods and the rate at which food is evacu- ated from the stomach. The objective of the present study was to obtain the quantitative stomach content data needed to use this approach to estimate the daily food ration of the sandbar shark, Carcharhinus plumbeiis. The sandbar shark was selected for this study because it is one of the few sharks for which gastric evacuation data are available (Medved in press). It is also an abundant, widely distributed shark (Springer 1960; Casey 1976) known to feed on commercially valuable species (Medved and Marshall 1981). In addition, it is a member of a large family of sharks (Carcharhinidae) and data collected for this species will provide the basis for making preliminary estimates of food consumption for the other members of the family. METHODS During the summer of 1983, young sandbar sharks were collected from Chincoteague Bay, VA, for stomach content analysis. The study area is located within the summer distribution of this species and supports a relatively large number of young sandbar sharks from early June through September. The bay is about 40 km long and 8 km wide at its widest point, and the average water depth is 2 m. A tidal inlet connects the bay with the Atlantic Ocean, and the tidal range varies from 0.75 to 1.50 m. The area is also characterized by strong tidal currents, vast salt marshes, and brackish to seawater salinities. A 4.9 m outboard motor boat was used as a fishing platform, and sharks were caught using monofila- ment gill nets and rod and reel fishing gear. The gill nets were 91m long, 1.8 m deep, and had a stretched mesh size of 10.8 cm. They were anchored at both ends and were buoyed so the foot rope touched the bottom. Net retrieval was made every 1 to 2 h. The fishing rods were equipped with Penn^ reels of 3/0 size, and the terminal tackle consisted of two wire leaders, each with a 4/0 fishing hook baited with squid. The hooks were set 1 m off the bottom. Both types of gear were used during all hours of the day. Upon capture each shark was brought into the boat where it was sexed, measured, and weighed. The sharks were then cut open and the stomach contents were removed and stored on ice in plastic bags. In the laboratory each food item was identified to species and a length measurement was made when possible. Each item in the stomach was also assigned a stage-of-digestion value ranging from 1 to 6 with a higher number indicating a greater extent of diges- tion. The stage-of-digestion scale was based on a gastric evacuation study (Medved in press) in which sandbar sharks were fed preweighed meals of either blue crab, Callinectes sapidus, or Atlantic menhaden, Brevoortia tyrannus, and were maintained in an enclosure constructed in the natural environment. The range of water temperatures in the enclosure (22.0°-30.0°C) was close to that recorded during the present study (20.0°-27.3°C). The sharks were sacrificed at various time intervals after feeding, and the food remains were weighed and described. The food item descriptions were used to arbitrarily esta- blish six stages of digestion that were each one-sixth of the total evacuation time. Each stage of digestion was about 12 h long for crustacean prey and 15 h long for fish prey. After identification of food items and assignment of digestion values, the stomach con- tents of each shark were separated into fish and crus- tacean components that were weighed to 0.01 g after draining off excess water. Each sample was then dried at 80 °C to constant weight (about 72 h) and again weighed to 0.01 g. RESULTS During the study 414 sharks were captured for stomach content analysis. The numbers of sharks caught by each fishing method and during various time periods of the day are summarized in Table 1 . The number of male and female sharks collected was nearly equal (210 and 204 respectively), and they ranged in size from 40.0 to 80.0 cm fork length (FL) (x = 56.1, SD = 6.8). Body weights were obtained from 369 (89.1%) of these sharks, and ranged from 720.0 to 5,690.0 g (x = 1,885.5, SD = 738.8). The body weight of the sharks not weighed was esti- mated from a regression equation derived from the ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. 396 MKI)\ Kli KTAL : STOMACH CONTENTS OF SANDBARSHARKS Table 1. — Number of sharks caught for stomach content analysis during different time intervals and by the two capture methods. Time interval Rod and reel Gill net Total 2230-0130 0130-0430 0430-0730 0730-1030 1030-1330 1330-1630 1630-1930 1930-2230 Total 19 11 20 21 27 39 18 23 178 33 27 20 27 27 39 30 33 236 52 38 40 48 54 78 48 56 414 animals that were measured and weighed: Wt = 0.0123 (FL)-^'^'" (v = 369, R- = 0.97). Water temperature during the fishing periods ranged from 20.0° to 27.3°C {n = 172, x = 25.1) but 90% of the temperatures were between 23.9° and 26.4° C. Fifteen different food types were identified in the stomachs (Table 2). A relatively large number of stomachs (« = 74, 17.9%) were empty, and unidenti- fiable fish remains occurred in others (n = 21, 5.1%). The blue crab was the most frequently occurring food item and was found in 279 (67.4%) of the stomachs examined and in 82.1% of the stomachs containing food. Of the food remains that could be positively identified as individual blue crabs (n = 309), 88.0% of the crabs had recently molted and were soft. The crabs that could be measured ranged in size from 1.0 to 14.0 cm between the two points of the carapace (n = 136, x = 7.4). Although exact numbers were difficult to determine, it appeared that less than half of the blue crabs were consumed whole. The only other prey frequently found was the Atlantic menhaden, which occurred in 55 (13.3%) of the stomachs examined and in 16.2% of the stomachs with food. Of the 61 cases where it was possible to determine if the fish was consumed whole or in part, 28 (45.9%) of the menhaden were whole and ranged in size from 5 to 10 cm total length (TL) (x = 7.3). The estimated sizes of the partially eaten menhaden ranged from 5 to 17 cm TL (x = 8.6). All other prey items were found in < 6.0% of the stomachs examined. The distributions of stage-of-digestion values assigned to the food items in the stomachs of sharks caught by the two different fishing methods are shown in Figure 1. The distribution for sharks caught by rod and reel indicated that 71.8% of the food items were in either the first or last stage of digestion. In contrast, food items in the stomachs of sharks caught by gill nets were divided more evenly among all the stages of digestion. The two capture methods also differed in the proportion of sharks Table 2— Stomach contents found in a sample of 414 sand- bar sharks. Stomach content No. of stomachs found in Percent of stomachs found in Blue crab, Call nee ties sapidus Empty Atlantic menhaden, Brevoortia tyrannus Summer flounder, Paralichthys dentatus Unidentified fish IVIantis shrimp, Squilla emprisa American eel, Anguilla rostrata Spot. Leiostomus xanthurus Atlantic silverside, Menidia menldia Smooth dogfish, Mustelus canis Northern pipefish. Syngnathus fuscus Anchovy, Anchoviella mitchilll Squid, Loligo pealei Bluefish, Pomatomus saltatrix Calico crab, Ovalipes ocellatus Mummichog, Fundulus heteroclitus Northern seahorse. Hippocampus hudemius 279 74 55 24 21 18 15 14 9 7 6 5 5 3 1 1 1 67.4 17.9 13.3 5.8 5.1 4.4 3.6 3.4 2.2 1.7 1.5 1.2 1.2 0.7 0.2 0.2 0.2 caught with empty stomachs. The percentage of the 178 sharks caught by rod and reel with empty stomachs (22.5%) was significantly higher than that found for the 236 sharks caught by gill nets (14.4%; ^-test, P = 0.015). Of the 414 stomachs examined, 203 contained a single food item. A stage-of-digestion value of 5 or 6 was assigned to 89 (43.8%) of these items, indicating that many sharks went at least the time equivalent of 5 stage-of-digestion units between meals (48 to 60 h for crustacean prey or 60 to 75 h for fish prey). Multiple food items were found in 137 stomachs. The difference between the stage-of-digestion values NET ROD-REEL ^ 50-, N = 338 N = 195 iiiiiii Li 1 2 3 4 5 6 2 3 4 5 6 STAGE OF DIGESTION FiGt_iRE 1.- Distributions of stage-of-digestion values assigned to food items present in the stomachs of sharks caught with gill net and rod and reel. 397 FISHERY BULLETIN: VOL. 83. NO. 3 of the first and last eaten food items within each stomach was calculated to assess the amount of time that passed during consumption of multiple item meals. In 19 stomachs a food item in the sixth stage of digestion and one in the first stage of digestion were found. In these cases a time equivalent of 5 stage-of-digestion units had passed between con- sumption of the two food items, and since a stomach containing an item in stage 6 of digestion would be relatively empty, the two food items were considered to represent two different meals. Excluding the above 19 stomachs from analysis, the mean differ- ence between the stage-of-digestion values of the first and last eaten food items was calculated for stomachs containing from 2 to 5 items (Table 3). The overall mean difference was 0.60 stage-of-digestion units indicating that multiple food items in the stomachs were in similar stages of digestion. The quantity of food in each stomach examined was measured on a wet weight and drj' weight basis. Excluding empty stomachs from analysis, the total Table 3.— Number of stomachs for wfilchi tfie stage-of- digestion value of the first eaten food item minus the stage- of-digestion value of the last eaten food item was equal to the given difference. The data are broken down into groups based on the number of food Items present in the stomachs. Stomachs that contained an item at stage 1 of digestion and an item at stage 6 of digestion are not included (see text). No. of stomachs with indicated difference between stage-of-digestion values of first and last food items eaten No. of Difference Mean differ- stomach 0 1 2 3 4 N ence SD 2 44 25 9 0 1 79 0.60 0.78 3 13 17 0 0 0 30 0.57 0.50 4 2 3 2 0 0 7 1.00 0.76 5 2 0 0 0 0 2 0 0 Overall 61 45 11 0 1 118 0.60 0.78 dry weight of food (TDW) was found to be linearly related to the total wet weight of food (TWW) in the stomachs (TDW = -0.24 + (0.22) TWW; n = 318, R~ = 0.96). Since the two measurements were highly correlated (r = 0.98) and wet weight measurements have frequently been used in similar food studies on other species, it seemed valid to express the food quantity results in this paper on a wet weight basis. Table 4 summarizes the descriptive statistics of the quantity of food in 414 stomachs examined. The mean total weight of food found in the stomachs of sharks caught by gill nets was significantly higher than that found for sharks caught by rod and reel (18.91 and 13.09 g respectively; 2-test, P = 0.003). Similar results were obtained when food quantity was measured as a percentage of shark body weight (0.96 vs. 0.76%; z-test, P = 0.043). This result, in conjunction with the other differences between the two capture methods mentioned above, suggested that sharks caught by rod and reel may not have been representative of the entire population (see sec- tion on Discussion). Because the primary value of the stomach content data in this study will be in the esti- mation of food consumed by the population, the following results concerning the amount of food in the stomachs were based on sharks caught by gill nets since they were probably more representative of the entire population of young sandbar sharks in the study area. For sharks caught by gill nets, crus- taceans accounted for nearly twice as much of the mean total wet weight of food in the stomachs than did fish. The mean wet weight of crustaceans in the 236 stomachs (12.37 g) was significantly higher than the mean of 6.53 g found for fish (z-test, P < 0.001). Similar results were obtained when food quantity was expressed as a percentage of shark body weight (0.65 vs. 0.31%; z-test, P < 0.001). The mean weights of the two food components in the stomachs Table 4 —Summary statistics of the amount of food In the stomachs of a sample of 414 sandbar sharks. Sharks were captured with gill nets and rod and reel gear. The z-test statistic was used to test the equality of the indicated pairs of mean values. Stomach contents Capture method Mean N SE mean Max Min z-test Stat. P-value 2-tailed test Fish (g) Crustacea (g) Gill net Gill net 6.53 12.37 236 236 1.10 1.11 114.80 102.20 0 0 3.74 < 0.001 Fish (% BW) Crustacea (% BW) Gill net Gill net 0.31 0.65 236 236 0.04 0.06 3.93 5.28 0 0 4.84 < 0.001 Total (g) Total (g) Gill net Rod and reel 18.91 13.09 236 178 1.53 1.25 135.68 100.30 0 0 2.95 0.003 Total {% BW) Total (% BW) Gill net Rod and reel 0.96 0.76 236 178 0.06 0.07 5.28 6.92 0 0 2.02 0.043 398 MKDVKl) KT AL.: STOMACH CONTKNTS OK SANDBAR SHARKS were also calculated for each of eight consecutive 3-h time intervals of the day (P^ig. 2). The means ranged from 1.05 to 14.92 g for fish, from 7.51 to 19.72 g for crustaceans, and from 11.74 to 34.64 g for the total wet weight of food in the stomachs. When 95% con- fidence bounds were placed around the means, con- siderable overlap of the confidence intervals was observed (Fig. 2). However, the mean total wet weight in the stomachs of sharks captured between the time of 0130 and 0430 was considerably higher than the other means, and the confidence interval for the mean during this time period overlapped sub- stantially with only two of the remaining seven inter- vals. Similar results were obtained when food quanti- ty was expressed as a percentage of shark body weight. During the study one stomach was examined that contiiined a total wet weight of 444.0 g of food (10.3% BW (body weight)). This shark was not in- cluded in the results presented above because the quantity of food in the stomach was substantially greater than for any other shark. It is mentioned here because it does indicate that the stomach capacity of this species is considerably greater than the amount of food typically found in the stomach. In an attempt to estimate maximum capacity, the stomachs of 23 sharks were removed, ligated, and filled with water to the point at which they were about to burst. This point was determined by filling several stomachs until they burst and noting the changes that occurred in the stomach wall just before the bursting point. The average maximum capacity of the stomachs was found to be 13.0% of BW (range: 8.04 to 19.8%). For sharks caught by gill nets the mean quantity of food in the stomachs (0.96% BW) was 7.4% of maximum capacity and the largest quantity of food in a stomach (5.28% BW) was 40.6% of maximum capacity. DISCUSSION Several investigations conducted in other areas have reported the sandbar sharks' diet to consist of small crustaceans and fish (Bigelow and Schroeder 1948; Springer 1960; Clark and von Schmidt 1965; Bass et al. 1973; Lawler 1977). With the exception of squid in several stomachs, the prey items of the sand- bar sharks captured in Chincoteague Bay, VA, were also found to be small crustaceans and fish (Table 2) and agree with those reported by Medved and Mar- shall (1981) for this species in Chincoteague Bay. The studies above provided little specific information con- cerning the frequency of occurrence, size, relative amounts, or physical state of the food items. In the NUMBER ou- 40- — 30- 1 [ 1 < O 20- 1— 10- 1 . 1 ■ .1 0130 0430 0730 1030 1330 1630 1930 2230 0430 0730 1030 1330 1630 1930 2230 0130 TIME INTERVAL Figure 2. -Amount of food in the stomachs of sandbar sharks caught by gill nets during various time intervals of the day. Dots in- dicate mean wet weight in grams, and bars represent 2 standard errors on each side of the mean. The number of stomachs examined during each interval is given at the top of the figure. present study, small blue crabs that had recently molted were, by far, the predominant food item in terms of both weight and numbers. Small menhaden were also found to comprise a significant portion of the food consumed, but other species appeared to be of minor importance in the diet of the sandbar shark in the study area. The results of this study strongly suggest that the feeding behavior of sandbar sharks in the study area was characterized by relatively short periods of feed- ing activity separated by substantially longer periods of time during which stomach contents were digest- ed and no additional feeding occurred. An indication that this species may go relatively long periods of time without feeding was the high percentage (21.5%) of sharks that had a single food item in their stomach that was in a late stage of digestion (stage 5 or 6). Based on the approximate duration of each 399 FISHERY BULLETIN: VOL. 83, NO. 3 stage of digestion (12 or 15 h depending on prey type) these sharl 25 mm in shell length did not differ significantly from 1 : 1 . No evidence of hermaphroditism was observed. The soft-shell clam, Mya arenaria, was once popular with clam diggers in San Francisco Bay. During the early 20th century, owners of bay front property fenced off portions of the mud flats in order to ex- clude clam predators, thus insuring bountiful harvests of M. arenaria (Bonnot 1932). Today as the "trend toward the improvement of San Francisco Bay water continues, "^ the potential for a recreation- al shellfishery exists again. Agencies for communi- ties on the bay have begun to look at this potential. Recently (1982), the digging of clams in San Fran- cisco Bay received official clearance for the first time in 30 yr.3 The spawning cycle of the soft-shell clam has been studied extensively on the east coast. Ropes and Stickney (1965) examined populations from the Cape Cod-New England region. They did not encounter clams in the ripe stage of gametogenesis until May, and by September spawning was over. Brousseau (1978) reported a biannual cycle of spawning forM. arevjiria from Cape Ann, MA. The first spawning occurred between March and April and was of short duration. A separate, second spawning took place from June through July. Porter (1974) studied M. arenaria from populations at Skagit Bay, WA. He noted a single yearly spawning from late May to early September. 'Department of Biological Sciences, San Francisco State Univer- sity', 1600 Holloway Ave., San Francisco, CA 94132. 2Jones and Stokes Associates, Inc. 1977. San Francisco Bay shellfish: an assessment of the potential for commercial and recrea- tional harvesting. Prepared for the Association of Bay Area Governments, 171 p. ^Champion, D. 1982. Clam digging OK'd on part of San Fran- cisco Bay. San Francisco, Chronicle, 27 August 1982, p. 1. The climate of the San Francisco Bay area, and hence the seasonal water temperature fluctuations of San Francisco Bay, are much less extreme than that of the New England or Washington areas and most of the other locations from which M. arenaria has been examined. The question investigated in this study is whether the spawning cycle of M. arenaria from San Francisco Bay would differ from that of other areas reported in the literature. MATERIALS AND METHODS Description of Study Sites Specimens of M. arenaria were collected from September 1979 through December 1980 from four sites around central San Francisco Bay (Fig. 1): 1) Candlestick Point- adjacent to the causeway leading to the Candlestick Park Stadium (lat. 37°42'32"N, long. 122°23'28"W); 2) Burlingame Lagoon -just south of San Francisco International Airport (lat. 37°35'12"N, long. 122°20'10"W); 3) Foster City- immediately north of the San Mateo Bridge, off Third Street (lat. 37°34'20"N, long. 122°23'28"W); 4) Point Isabel -north of the Golden Gate Fields race track on the eastern shore of San Francisco Bay (lat. 37°53'59"N, long. 122°23'28"W). These areas were selected because they experience annual variations in temperature and salinity (Conomos 1979), factors which are known to affect bivalve spawning cycles (Loosanoff and Davis 1951; Swan 1952; Matthiessen 1960, Pfitzenmeyer 1962; Stickney 1964). Substra- tum conditions were classified according to field observations; no particle size analyses were con- ducted. Manuscript accepted October 1984. FISHERY BULLETIN: VOL. 83. NO. 3, 1985. 403 FISHERY Bl'LLETlN: VOL. ,S3. NO. .S The Candlestick Point site has a sandy substratum and is adjacent to a broad expanse of mud flat. Temperatures and salinities at this site reflect those of the central San Francisco Bay (Conomos 1979). Clams were collected high on the sandy beach front 0.6 m above Mean Lower Low Water (MLLW) in the areas exposed as the tide begins to recede. Burlingame Lagoon is separated from San Fran- cisco Bay by a levee with a narrow channel connect- ing to the open body of the Bay (Fig. 1). Salinity and temperature can vary dramatically with heavy rains (8 to 12 ppt) (pers. obs.). The area has a heavy clay- mud substratum. The collecting site was 0.6 m above MLLW. The Foster City site was selected because it reflects conditions more closely associated with the south bay (Conomos 1979). The substratum is sandy- mud with rocks and cobble intermixed. The area juts into the bay and is exposed to wave action and temperatures and salinities characteristic of the bay waters (Conomos 1979). It is adjacent to outflow from the tidal channel that winds through Foster City. The collecting site was 0.24 m above MLLW. The Point Isabel site is on the mud banks of a nar- row (4 m wide) tidal channel that bisects the Point Isabel promontory. The substratum is clay-mud with rock and debris intermixed. Clams were collected from 0.15 to 0.46 m above MLLW. The depths of the collection sites (relative to MLLW) at the four locations were dictated by the presence of M. arenaria at each site. Mya arenaria signals its presence by distinct siphon holes at the surface of the substratum. After an initial excavation of each collection site to a depth of 10 cm, it became obvious that clams were present only in the specific areas clearly marked by their siphon holes. There- after only these areas were sampled. Temperature and salinity were recorded with a field hydrometer (marked at ppt) and thermometer at each' site each month, beginning in October 1979 and continuing through November 1980. Collecting Methods A gardener's hand rake and careful hand digging was used for excavating the deep-dwelling M. arenaria. At least 25 clams were collected from each of the four sites each month for 1 yr (with the follow- ing exceptions: 4 October- Foster City, 12 collected; 3 November- Foster City, 20 collected; 25 Decem- ber-Foster City, 24 collected; 2 November- Candle- stick Point, 20 collected; 24 September- Point Isabel, 20 collected; 3 November- Burlingame, 7 col- lected). A total of 1,625 clams were examined in this study. Clams were collected, placed in a Thermos* jug of cool bay water, and returned to the laboratory. Processing Methods Analysis of gonadal stage was made by micro- scopical examination of histological preparations (Ropes and Stickney 1965). The presence and devel- opment of gametes was used to infer the spawning stage or readiness of the clam. Specimens were measured for shell length to the nearest 0.1 mm. The anterior one-third of the visceral mass (Ropes and Stickney 1965) was removed, labeled, and placed in Bouin's seawater fixative. During dissection, tissues were submerged in cool seawater to prevent drying or osmotic changes. The time between collection and preservation was under 3 h to prevent any gonadal changes. The tissues were subjected to standard histological procedures (dehydrated in alcohol and embedded in paraffin). Embedded tissues were thin sectioned (5 ^A) on a rotary microtome. Sections were mounted on glass slides, stained with Harris' hematoxylin and eosin, and examined using standard light micro- scopy. Each slide of gonadal tissue was studied to deter- mine the presence of male or female gametes and the condition of the gonadal tissues. This allowed clams to be placed into one of the five classes of spawning readiness (inactive, active, ripe, partially spawned, spent) employed and described by Ropes and Stick- ney (1965) for M. arenaria. Categories of Spawning Readiness (adapted from Ropes and Stickney 1965) Female Gonads INACTIVE PHASE. -Ropes and Stickney (1965) used the term "inactive" to describe this phase. Brousseau (1978) preferred the term "indifferent" because cellular activity is continuing although no gametogenic activity is obvious. The term "inactive" is employed here and refers to individuals which are not seen to be producing gametes whether due to seasonal (quiescence or immaturity. Thus in this research which presents pooled male and female datii, the "inactive phase" may contain sexually un- differentiated individuals along with inactive animals clearly recognizable as male and female. Females in the inactive phase exhibit small oocytes ^Reference to trade names does not imply endorsement hy the Na- tional Marino P^i.shories Service, NOAA. 404 ROSKNHLIM :in5CO^Tl-lOU5oto •'-•'-ejcMcoco'^'^mmtDtDN.h.oooo SIZE CLASS Figure 2. -Size-frequency histograms for the four populations studied. Size classes are (in mm) 10.0-14.9, 15.0-19.9, etc. sampled were inactive and 5% still showed a few mature gametes which would probably have been resorbed. All individuals were inactive by the end of December 1979. The first sign of gonadal activity appeared at the end of January 1980; by March, 82.1% of the clams sampled were active. In April, all individuals were either ripe or spawning. Peak spawning occurred 407 FISHERY BULLETIN: VOL. 83, NO. 3 CANDLESTICK POINT 100 z 111 o cr. UJ 0 SO N D JFMA POINT ISABEL lOOr t— I LU O GC UJ FOSTER CITY 100 m O tr UJ Q. 0"- S O N D JF M A M JJASOND BURLINGAME lOOr ui O GC UJ a. MONTHS INACTIVE [HO] ACTIVE ^RIPE ^ PARTIALLY SPAWNED | | SPENT Figure 3. - Percentage of the clams sampled that were in each of the five categories of spawning readiness. Male and female data are pooled. 408 ROSKNHU'M and NIKSKN: SPAWNINC.CVC'LK OF A/y'.4 ARKNAHIA during May and June. By the end of June, the sample was roughly divided into thirds among the partially spawned, spent, and inactive stages. Some spawning could still be seen through the end of August, but by late September 85% were inactive, 10% partially spawned, and 5% spent. Foster City September-October 1979 marked the end of the Foster City population spawning, with most clams (70%) being inactive during this period (Fig. 3). From early November through the end of December 1979, no activity could be found. The late January 1980 sample showed a 90% active:10% spent ratio with no intermediate stages represented. It is unclear whether this indicated a rapid maturing and spawning of a few precocious individuals or an over- wintering of residual gametes. By mid-March, every stage was represented, most (50%) being in the ripe stage. Peak spawning extended from the beginning of April through May, with a complete maturation- spawning season extending from late February through May. From July to late September no appre- ciable gonadal activity could be discerned. This lack of activity from mid- to late summer distinguished the Foster City population by its short spawning season relative to other populations (Fig. 3). Burlingame Lagoon Sampling of the Burlingame Lagoon population did not begin until November 1979. At the time 85% of the clams sampled were spent and only 15% were inactive (Fig. 3). This suggests a protracted spawn- ing in 1979, similar to that which was seen in 1980. By December all but 5% were inactive and by 1 January 1980 all were inactive. Gonadal activity ap- peared again by the end of January 1980. Peak spawning occurred during May and June; however, spawning continued well into September and Octo- ber with a full 30% of the clams of each sample being in the partially spawned stage. The Burlingame Lagoon population had the longest spawning season of the populations examined. Point Isabel Sampling of the Point Isabel population began in September 1979. The September and October sam- ples still contained ripe individuals (< 10%), but most of the clams sampled were spent or inactive. By November, 70% of the clams sampled were inactive and by the end of December all were inactive (Fig. 3). The 1979 spawning season had ended for the Point Isabel population by early November. Gonadal activity had resumed by late January. By March, 80% of the clams sampled were active. Only ripe and partially spawned individuals could be found in mid-April, and by May inactive individuals were being found (10%). Peak spawning occurred during May. Spawning continued through June and by July almost 80% were inactive and 20% partially spawn- ed or spent. Spawning appeared to be over; however, the August sample contained almost 35% partially spawned and 5% ripe individuals. It is tempting to suggest a possible second spawning in August, but the May and June samples lacked active or ripe in- dividuals, which suggest the August observation be attributed to sampling error. Spawning was still occurring in late September, as 15% were still in the partially spawned stage. At this time, however, 60% were inactive and 25% spent. The four study sites showed similar trends (Fig. 4) in temperature and salinity. Temperatures fell from November through mid-Januar^' and rose from mid- January through the beginning of April and then stabilized. Temperature then climbed again from mid-May through July. The July-September temper- atures were steadier at the Foster City and Bur- lingame sites than at Point Isabel or Candlestick Point, where they dropped markedly during this period. Salinity followed a similar trend, falling dur- ing the late winter months, and rising during spring and summer (Fig. 4). Salinity ranged from 8 to 33 ppt and temperature from 12° to 23°C. DISCUSSION The spawning cycle of M. arenaria in San Fran- cisco Bay in 1980 was an extended one. Gameto- genesis had begun by late January for three of the four populations sampled (Candlestick Point, Bur- lingame Lagoon, and Point Isabel), and by mid- March all five stages of gonadal development were represented in the Burlingame Lagoon, Point Isabel, and Foster City populations. Spawning had begun at all four sites by April; over 20% of the individuals from each sample were in the partially spawned or spent stages. The number of clams in these spawning stages reached a maximum during May and June 1980. Spawning continued through September and October and then ceased. Only a single clam < 25 mm in length was found to have active gonads. This suggests that 25 mm might be generally recognized as the size at first repro- duction for San Francisco Bay M. arenaria. 409 FISHERY BULLETIN: VOL. «3. NO. 3 T E M p E R A T U ft E 26H 24 2> 20* !•• U* 12 26< 24- 22< 20- !•• !«• U< 12< 26« 24« 22« 20* !•• 16* 14< 12 26< 2« 2> 2C^ IS* lb* I4< ^ g '^ L_1_SL ISABEL FOSTER CITY BURLINGAME ^^1 CANDLESTICK N M M JASON 36 32 >2t •24 20 >16 >12 •• F36 32 >2t •24 •20 >16 ppt •12 >• 36 •32 •2t •24 •20 •16 •12 >• t A L I N I T V Figure 4. -Temperature and salinity variations during the collection period for the four collection sites, 1979 and 1980. Although no growth or age measurements were per- formed in this study, the 25 mm shell length in- dicating a lower limit to sexual maturity corresponds to the east coast M. arenaria found to be late in their first year (Brousseau 1979). Of the 313 clams that were indistinguishable as to sex, only 28 were < 25 mm in length, therefore we cannot attribute this lack of discernable sex to im- maturity. Furthermore, clams of indeterminate sex were seen only during the fall and winter months (September-March) when most clams were found to be in the inactive stage. During the period of March- September or spring-fall when the active, ripe, par- tially spawned, and spent stages were well represented, all clams could easily be determined to be male or female. The difference between inactive male and female gonads is obvious and was seen in many clams, yet many clams which were larger than 25 mm in length and should have been sexually mature showed no signs of sexually distinguishable tissue at all. No evidence of even small oocytes or atypical spermatogenesis was seen in these clams. For the sake of simplicity, these clams were placed in the inactive stage. Perhaps this condition was a kind of "gonadal exhaustion" due to the prolonged spawn- ing period. The four study populations were dominated by clams ranging from 40 to 75 mm in length (P^ig. 2). This size range corresponds to the 1.5 to 4.0 year classes determined by Brousseau (1979) for M. arenaria from Gloucester, MA. While total corres- pondence in growth rates between Massachusetts 410 ROSKNBLUM and NIESEN: SPAWNINC CVCl.K OF An'/l /IftAW/IWM and San Francisco Bay populations cannot be as- sumed, the age classes can be used as a first estimate of approximate age with size. Studies of the spawning cycle of soft-shell clams from the east and west coasts of the United States reveal both similarities and differences in spawning pattern. Spawning on both coasts begins in early spring as the water warms from the lower winter temperatures. The majority of east coast populations studied show two separate spawnings each year, while populations studied on the west coast show a single more protracted spawning. Differences in the length of spawning and the number of separate spawning episodes are probably partially related to the phenology of water temperature change and the difference in the range of water temperatures that occur on either coast. Mya arenaria in San Francisco Bay, studied dur- ing the 1979-80 season, began ripening earlier than M. arenaria of the New England region studied previously. Ropes and Stickney (1965) encountered active clams in eastern Maine by late January; however, ripe clams were not discovered until mid- May, and at that time none appeared to have spawn- ed. Clams from their Booth Bay Harbor samples showed the earliest ripening, which was in April and May; by September spawning was over. Brousseau (1978) reported a biannual cycle of spawning for M. arenaria from Cape Ann, MA. The first spawning at Cape Ann occurred as early as that in San Francisco Bay (March); however, it was of short duration, being over by April. A separate sec- ond spawning took place during June through July. Brousseau's figures indicate water temperatures began to rise from a low of 1°C around Cape Ann as early as mid-February, but did not rise above 10° C before May. It is possible that the increase in temperature triggered an early spawning, but the continuing, relatively cold temperature prevented an adequate build-up of mature gametes to sustain a prolonged spawning. Once spawning had taken place, the clams may have had to undergo another period of gametogenesis prior to a second spawning. Brousseau (1978, page 159) stated, "The presence of cytolyzed unspawned gametes in the summer samples suggested that the same individuals had also been ripe earlier in the year. Thus the observed spawning pattern was due to repeated spawning by the same individuals rather than asynchronous spawning of individuals within the population." Pfitzenmeyer (1962) also reported two annual periods of spawning in M. arenaria at Solomons, MD. He noted that "The first umbone larvae of the year usually were found in May after the surface waters rise above 15°C or a mean temperature of 16.7°C." Salinities which remained constant throughout the spring remained near 10 ppt. This corresponds to the March temperature-salinity pat- terns in San Francisco Bay. During March, San Francisco clams were just beginning to spawn. Pfit- zenmeyer also noted the disappearance of larvae from the Maryland waters as the temperature rose above a mean of 21.4° C. The larvae did not reappear until temperatures had fallen below this point. He concluded an optimal temperature range exists dur- ing which spawning may occur. In San Francisco Bay, temperatures rose to or above 21 °C only in the following instances: Foster City- 11 April - 23° C, 2 July - 21°C, 25 August - 21°C; Candlestick Point- 9 April - 23°C, 30 June - 23°C; Point Isabel- 10 April - 23°C (Fig. 4). Porter (1974) noted a single yearly spawning from late May to early September among M. arenaria from Skagit Bay, WA. This is a shorter spawning season than seen among San Francisco Bay M. arenaria and may be a result of lower temperatures (4.8°-15.7°C) encountered in Washington. Simel^ reported a single spawning from late March through April for soft-shell clams from Humboldt Bay, CA. Generally, this more northerly part of California has a cooler climate than the San Fran- cisco Bay area. Simel indicated that the later stages of gametogenesis corresponded with a peak in the phytoplankton abundance. Studies of the spawning cycles of M. arenaria from the east and west coasts of the United States suggests a pattern of spawning behavior. Spawning begins as the water temperature rises in the spring. Pfitzenmeyer's (1962) work suggests M. arenaria's optimal spawning range falls between 15° and 21 °C. Differences between the spawning cycle of M. arenaria from San Francisco Bay and that of M. arenaria from New England, Canada, Washington, and northern California may be explained as the logical result of the different seasonal warming pat- terns and extremes encountered in the different areas. San Francisco Bay does not cool to the same temperatures as the other areas and has a much longer period of moderate water temperatures, ex- tending from late March through summer and into September and October. Consequently, M. arenaria's spawning season is equally protracted in San Francisco Bay. ^N. Simel, Humboldt State University, Areata, CA 94542, pers. commun. 1982. 411 FISHERY BULLETIN: VOL. 83, NO. 3 ACKNOWLEDGMENTS We thank James Sutton of the California Academy of Sciences for his assistance with the selection of our study sites. LITERATURE CITED BONNOT, P. 1932. Soft shell clam beds in the vicinity of San Francisco Bay. Calif. Fish Game 18:64-66. Brousseau, D. J. 1978. Spawning cycle, fecundity, and recruitment in a popula- tion of soft-shell clam, Mya armaria, from Cape Ann, Massa- chusetts. Fish. Bull. U.S. 76:155-166. 1979. Analysis of growth rate in Mya arenaria using the Von Bertalanffy equation. Mar. Biol. (Berl.) 51:221-227. CoE, W. R., AND H. J. Turner, Jr. 1938. Development of the gonads and gametes in the soft-shell clam (Mya arenaria). J. Morphol. 62:91-111. CONOMOS, T. J. 1979. Properties and circulation of San Francisco Bay waters. In T. J. Conomos (editor), San Francisco Bay, the urbanized estuary, p. 47-84. Pac. Div., Am. Assoc. Adv. Sci. Allen Press, Lawrence, KS. LOOSANOFF, V. L., AND H. C. DAVIS. 1951. Delaying spawning of lamellibranchs by low tempera- ture. J. Mar. Res. 10:197-202. Matthiessen, G. C. 1960. Observations on the ecology of the soft clam, Mya arenaria, in a salt pond. Limnol. Oceanogr. 5:291-300. Pfitzenmeyer, H. T. 1962. Periods of spawning and setting of the soft-shelled clam, Mya arenaria, at Solomons, Mary'land. Chesapeake Sci. 3: 114-120. Porter, R. G. 1974. Reproductive cycle of the soft-shell clam, Mya arenaria, at Skagit Bay, Washington. Fish. Bull., U.S. 72:648-656. Ropes, J. W., and A. P. Stickney. 1965. Reproductive cycle of Mya arenaria in New England. Biol. Bull. (Woods Hole) 128:315-327. Stickney, A. P. 1964. Feeding and growth of juvenile soft-shell clams, Mya arenaria. Fish. Bull., U.S. 63:635-642. Swan, E. F. 1952. The growth of the clam Mya arenaria as affected by the substratum. Ecology 33:530-534. 412 RECRUITMENT PATTERNS IN YOUNG FRENCH GRUNTS, HAEMULON FLAVOLINEATUM (FAMILY HAEMULIDAE), AT ST. CROIX, VIRGIN ISLANDS' W. N. McFarland,2 E. B. Brothers.^ j. C. Ogden," M. J. Shulman,^ E. L. Bermingham," and N. M. Kotchian-Prentiss' ABSTRACT During 1979 and 1980 the settlement of postlarval ji^runts (mostly French grunts, Haemulon flarolinentuw) from the plankton to inshore areas in Tague Bay, St. Croix, Virgin Islands, was monitored. Settlement occurred at all times of the year, but showed two distinct maxima during May-June and October-November. Summer settlement rates were approximately one-third the peak rates, whereas winter settlement rates were low but always present. A dominant, semilunar periodicity in the settlement of the postlarvae was clearly present, but smaller interspersed weekly peaks occurred. Calculation of the fertilization dates of recently settled postlarval grunts, derived from otolith ages, also established a dominant 1 5-day periodicity, but again interspersed with smaller weekly fertilization peaks. The phase of settlement and fertilization is most strongly associated with the quarter moons and/or intermediate daily excursions of the tides; the smaller weekly peaks are more closely associated with new and full moons. The pelagic existence of French grunts is about 15 days, suggesting that fertilizations that lead to successful recruitment to inshore areas precede settlement by 15 days. Two hypotheses -semilunar shifts in spawning versus a more constant daily rate of reproduction -are discussed as possible explanations of the dominant semilunar rhythm observed in the settlement patterns. Community structure is primarily determined by predation, competition, environment, and patterns of reproduction and recruitment (Ricklefs 1979). In most marine fishes the larval stages are planktonic, a circumstance especially true for tropical reef fishes (Breder and Rosen 1966; Sale 1980; Thresher 1984). In general, larger reef species produce numerous but relatively small eggs, which at spawning are dis- persed into the plankton; whereas smaller reef species produce fewer but larger demersal eggs, which are guarded until hatching when the larvae also "escape" into the offshore plankton community (Johannes 1978; Barlow 1981). The potential significance of this widespread reproductive strategy to affect coral reef fish assemblage structure is wide- ly recognized (see Helfman 1978 and Sale 1980 for reviews). Most ecologists have assumed that ^Contribution No. 70 of the West Indies Laboratory. ^Section of Ecology and Systematics, Division of Biological Sciences, Cornell University, Ithaca, Wi 14853. ^Section of Ecology and Systematics, Division of Biological Sciences, Cornell University, Ithaca, NY 14853; present address: 3 Sunset West, Ithaca, Wi 14850. "•West Indies Laboratory, Fairleigh Dickinson University, Chris- tiansted, St. Croix, VI 00820. ^Department of Zoology, University of Washington, Seattle, WA 98195. ^Department of Biology, University of Massachusetts, Boston, MA 02125. 'Department of Zoology, University of Maine, Orono, ME 04473. Manuscript accepted October 1984. FISHERY BULLETIN; VOL. 83, NO. 3, 1985. planktonic larval fishes provide an extensive reser- voir of potential recruits that settle to the reef whenever space becomes available (Sale 1977, 1978; Dale 1978; Smith 1978). Virtually all recent studies, however, emphasize that we know little of the ecology of larval fishes at sea and of their patterns of recruitment to benthic juvenile habitats (McFarland in press; McFarland and Ogden in press). Until more quantitative information on the early life history of a variety of species of reef fishes is available, models that "explain" fish community structure remain, at best, first approximations. Here we describe spatial and temporal patterns of recruitment in the French grunt, Haeviulonflavolin£atum, a dominant western Atlantic tropical reef species. MATERIALS AND METHODS Recently settled postlarval French grunts and white grunts, H. plumieri, standard length (SL) ca. 8.5 mm, are commonly observed in schools over grass beds or associating with coral clumps and gorgonians in Tague Bay, St. Croix, VI. Single in- dividuals and larger aggregations associate with structure and/or intermix with schools of mysids (McFarland and Kotchian 1982). There is little diffi- culty in recognizing and counting these very small grunts in the field because they lack the body colora- 413 tion patterns of slightly larger postlarvae (see table 1 in McFarland and Kotchian 1982). Postlarval grunts, however, cannot be identified by sight at the species level. We will refer to these smallest postlarvae as PL-1 grunts. These recently settled postlarvae are diurnal plankton feeders (McFarland 1980) as, presumably, is the preceding pelagic larval stage. Although the postlarvae show strong social tenden- cies to school during daytime, they are solitary at night (Helfman et al. 1982); the same pattern is found in older juveniles and adults (Ogden and Zieman 1977). The typical stereotyped twilight migrations of older juvenile and adult grunts (Hob- son 1968; Ogden and Ehrlich 1977; McFarland et al. 1979), however, are not present, nor are the agonis- tic behaviors typical of all later stages (McFarland and Hillis 1982). During collateral studies on the population dyna- mics of grunts, on agonistic behavior in juveniles (McFarland and Hillis 1982), and age determinations of grunts (Brothers and McFarland 1981), we noted that recently settled PL-l's appeared in pulses. To in- vestigate this periodicity, an extensive area of bot- tom in Tague Bay and a series of shallower discrete reef sites were censused repeatedly for PL-l's. In addition, subsamples of PL-l's were collected throughout the census period for size and age deter- minations. Tague Bay Census Measurements Sixteen flagged iron stakes were set 10 m apart along the bottom of Tague Bay parallel to the bay's barrier reef at a depth of ca. 5m. The bottom was characterized by sandhills produced by the burrow- ing activity of thalassinid shrimps, with stands of the seagrasses Thalassia testudinum, Syringodium filiforme, and Halodole wrightii between the mounds. Censuses of the total number of postlarvae were made using scuba, recording the numbers of grunts encountered along a 5 m wide transect. The census included all postlarval grunts observed over an area of 800 m^ of bay bottom. Because almost all juvenile grunts in the immediate vicinity were French grunts, we presume the census data mostly represent this species. Of 85 PL-l's collected on this site, all were identified as French grunts. The census began on 25 February 1979, and was con- tinued at variable intervals through 31 August 1980. Although numbers of older grunts were also record- ed, here we report only the numbers of the smaller and youngest postlarvae (mean SL = 8.5 mm). This census is hereafter referred to as the "sandhill" site. FISHERY BULLETIN: VOL. 83, NO. 3 Reef Census Measurements A series of 20 individual sites in a shallow sandy area (1-3 m depth) along a 200 m stretch of the Tague Bay barrier backreef were monitored for post- larval settlement. The sites varied somewhat in size and structure, but were composed of small clumps of Montastrea annularis and/or Porites porites. Six- teen of the sites were in depths of 1 to 2 m; four were in 2 to 3 m depth. Reef areas varied from 0.1 to 8 m^; vertical relief from 20 cm to 1.5 m. Daily censuses were obtained as often as possible from 25 April 1980 through 25 May 1981. The census schedule was intensified especially from 6 May through 27 Decem- ber 1980 (172 censuses over 236 d). Counts on each site on each census day included the total numbers of PL-l's, older postlarvae, juvenile grunts (see McFar- land and Kotchian 1982), damselfishes (all species lumped), and the common sea urchin, Diadema an- tillarum, within the spines of which the middle-sized juveniles often seek refuge (see Helfman et al. 1982). In this reef area, which is surrounded by coral sand and lacks seagrass beds, juvenile white grunts were never encountered, only French grunts. We con- clude therefore that recruits were all French grunts. Reproductive Activity and Aging of French Grunts Spawning in grunts has not been observed or reported in the literature (Breder and Rosen 1966; Hobson 1968; Johannes 1978; Lobel 1978; our per- sonal observations and field observations of P. Colin and of E. S. Hobson). An indirect method was used to provide information on whether grunts spawned in some periodic manner, as so many reef fishes do (Johannes 1978; Lobel 1978; Colin 1982). The age of French gnmts can be established in days, for example, by counting the number of micro- structural growth increments laid down in the lapillus (Brothers and McFarland 1981). The method is especially useful for aging the younger life history stages (< 100 d). By ascertaining the actual age of an individual grunt in days, it becomes possible to establish the specific date on which it was spawned. The method requires a correction, however, because the first "daily" growth increment deposited in the otolith does not coincide with fertilization of the egg. Our best "estimate" for the age at formation of the first distinct increment in the otoliths of French grunts is the third day after fertilization (for details see Brothers and McFarland 1981; this revised esti- mate is based on laboratory-reared porkfish, the con- 414 McFAKl.ANI) KT Al,.; KKCRUITMKNT PATTERNS IN FRKNCll CKUNTS familial Anisotremus virginicus, of known age, sup- plied by Martin Moe). Thus, to each "otolith age" ( = total increments counted) 2 d were added to establish the "absolute" daily age of an individual fish. This method was applied to postlarval French grunts, collected throughout the year in the vicinity of the 20 discrete census sites. On most census days individual PL-l's were collected with a fine mesh net and fixed and preserved in 95% ethanol. Each post- larva was measured (SL) and the otoliths were removed and placed in immersion oil. The number of growth increments was counted and corrected ( -i- 2 d) to the actual date of fertilization. Tides A tidal gauge (NOAA, Ocean Survey #9751224) operates at the West Indies Laboratory dock, about 0.5 km from our Tague Bay study site. Hourly tidal heights for the year 1980 were obtained from NOAA. Missing, due to malfunction of the gauge, are records from 26 June to 9 September 1980. These missing values were approximated from calcu- lated tidal data for San Juan, Puerto Rico. The phase of the tides in Puerto Rico matched closely the tides at Tague Bay (comparisons of dates before and after the missing records), but the actual excursion of the tides was less at St. Croix than calculated for Puerto Rico. RESULTS Tague Bay Sandhill Study Site During 1979, 27 censuses were executed during 300 possible days of sampling (25 February to 21 December). The mean interval between censuses was 1 1 .2 d ± 6.6 ( 1 SD), the intervals ranging from 4 to 23 d. Five population peaks were recorded, with the number of PL-l's counted between peaks often declining to < 100 individuals. The 1979 census clearly indicated that settlement was represented by a series of pulses, but the sampling intervals were too long to resolve periodicities of much less than 1 mo. Therefore, in 1980 the sampling resolution was improved by increasing the number of censuses to 47 over a possible 325 sampling days (6 February to 26 December 1980); mean sampling interval was 7.0 d ±7.7, the shortest interval being 1 d and the longest interval 49 d (i.e., the first interval). Eleven peaks, of which 8 are distinct, establish that the postlarvae settle in pulses, with the population numbers on the bay bottom often decreasing to 0 between pulses (Fig. 1). Although the pulses in population numbers 1950 Figure 1.- Numbers of recently settled postlarval grunts observed over 800 m of bottom on the sandhill site area of Tague Bay, St. Croix, V.I. are suggestive of a semilunar periodicity, the peaks do not coincide more closely with either quarter moons or with the new and full moons than do the population declines. Individual Barrier Reef Study Sites The mean sampling interval on the 20 individual census sites from 6 May to 27 December 1980 (possi- ble 236 sampling days) was 1.37 d ± 0.7 (1 SD). This shorter sampling interval vividly reveals the period- icity in the appearance of PL-l's on the sites (Fig. 2). Coincidence between these peaks and the peaks for the sandhill study sites is quite good (compare Figures 1 and 2) and indicates that the timing of the settlement of PL-l's is general over the entire area. Also, as in the sandhill area, the numbers of PL-l's declined to 0 between most pulses. Furthermore, the pooled data for the 20 sites indicate a bimodal seasonal influx of PL-l's, one in late spring and a sec- ond increase in fall, as seen also in the sandhill study site (compare Figures. 1 and 2). The summer settle- ment pulses involved about one-third the number of individuals associated with the bimodal peaks. Winter settlement was very low; only small numbers of PL-l's were observed during January and February 1981. Age and Reproductive Activity A total of 2,353 postlarvae from 141 collections spread over 391 d were captured between 20 Febru- ary 1980 and 16 March 1981 (Fig. 3). From each of these 141 samples 15 PL-l's were measured and aged, or fewer if the collected samples contained 415 FISHERY BULLETIN: VOL. 83, NO. 3 CO cc o 100 - jl.jllll I Jl j| A| SI O J 7 -(4) ill I I 14 - (2) 11 - (11) nH dH DC 900 J yk 15^ Figure 2. - Numbers of recently settled postiarval grunts observed on the individual census sites located in the shallows along the backreef of Tague Bay barrier reef. The three upper graphs represent the recruitment of PL- I's to specific sites?, 11, and 14. Numbers in parentheses to the right of the hyphens indicate on how many sites a similar seasonal pattern of settlement was observed. Three of the 20 census sites were like 7 or 14, but also recruited grunts in summer ( not shown). The lower graph represents the pooie cc < 2000 1500 1000 500 SETTLEMENT CO ^ O 14 ^ 100 I- u. O CC UJ 50 10 10 14 n FERTILIZATION I 5 T 0 (-) 14 10 5 0 5 10 DAYS FROM FULL MOON 14 (+) should be unimportant. This condition would prevail because the tidal states (springs, neaps, intermediate conditions) change phase relative to the lunar cycle as the year progresses (Fig. 6). In contrast, if a par- ticular state of the tide served as a primary trigger for fertilization and/or settlement then the moon's state would be unimportant. To test for the impor- tance of tidal and lunar state the data were analyzed using a log likelihood statistic (g-test, Table 3) by grouping the 22 settlement and 26 fertilization peaks into the 9 possible combinations of tidal and lunar state (e.g., spring, neap, and intermediate tides and full, new, and quarter moons). Clearly both the lunar cycle and tidal state have significant effects on settlement and fertilization, but their interaction, although large, is nonsignificant. Because 60 to 80% Table 3. — Summary of comparisons of lunar cycle and tidal state for fer- tilizations and settlement pulses of Frencfi grunts during 1980. Data from Table 2. Log likelihood test from Sokal and Rohlf (1981). g-value and associated chi-square Category Settlement data Fertilization data Total G' Moon alone' Tide alone' Moon-tide' 36.7 11.3 18.0 >x' 001(8) .01(2) > X' .001(2) = 1 3.8 7.4 ns < x' .05(4) = 26.1 22.5 > x' 01(8) = 20.1 = 9.9 7.3 > x' 05(2) = 6.0 10.2 > x' .01(2) = 9-9 = 9.5 4.9 ns < x' 05(4) = 9.5 'Single classification of all categories. 'Single classification of moon or tidal state alone. 'Two-way classification, moon phase versus tidal state. 420 MiFAKLANI) KT AI, : KKCKl'ITMKNT I'AITKRNS IN FRKNCH CRrNTS DAYS BEFORE AND AFTER FULL MOON 5 14 10 5 0 5 10 15 -»- J I I I I I I I I I \ I I I I I I I I I I ' I I ' ' I I ' I CO < lu > I- o LU CO O o 1 2 3 4 5 6 7 8 9 10' [7\ yS> A M © .<^^ ^ f^ 1 — I — I — rrr w © O f w • 1 2 ~3 ^v 4 5 6 \. y z 7 8 • 9 J 10 I — I — I 1 A I 1 — I — I o MOON PHASE T 1 TXT — I — r T 1 — I — r P'iGiiRE 6. - Relation between the periodic settlement pulses of postlarval grunts during 1980, the phases of the moon, and the procession of spring and neap tides. Shaded areas represent relatively uniform series of neap or spring tides; unshaded areas are either increasing or decreasing intermediate tides (see Table 2). The sequence of days reads from left to right across each row. The first day of each calendar month is indicated by the letters, e.g., A at top = April 1980 J = January 1981. Closed circles represent the large and intermediate-sized pulses (> 100 PL-l's); open circles are settlement pulses with < 100 individuals (see Table 2). Closed and open circles with the letter w are weekly recruitment pulses interspersed between the more common biweekly peaks (see Table 2). The two abruptly changing vertical lines in- dicate the actual dates of the quarter moons at different times of the year, as determined from the Nautical Almanac for 1980. The mean deviations of those pulses most closely associated in each month with the quarter moons are: first quarter moon = 1.7 d + 1.9 (SD). third (juarter moon = 0.9 d + 0.8 (SD). Maximum tidal excursions for spring tides occurred in May and June, and in October and November, as did the maximum recruitment of PL-l's to the census sites (Fig. 2, Table 2). of the settlement pulses and fertilizations were associated with quarter moons and intermediate tides (Table 2), attempts were made to isolate the significant categories by combining the data (springs -I- neaps versus intermediate tides; full + new moons versus quarter moons). Nonsignificance occurred in all combinations with the exception of tides and settlement (g-test = 9.91 > x^,.:,,,, = 3.84). The relationship between fertilization and settle- ment of French grunts and the state of the tides and lunar cycle is obviously complex and difficult to unravel without ambiguity. Larger sample sizes (data over several years) would help, but perhaps more important would be specific data on current regimes at actual spawning sites and in the general vicinity of settlement sites. The highly significant but independent effects of moon and tidal state likely point at the importance of favorable currents to successful larval recruitment. Settlement of Postlarvae in Space and Time From May through December 1980, when the cen- sus frequency was high, there were 16 consecutive quarter-moon periods. On the 20 census sites, there- fore, a total of 320 separate pulses (16 x 20) could have been observed if settlement did occur during the quarter moons. During this period 161 pulses were actually observed on the 20 sites during quarter moons. When the number of PL-l's settling was high, most sites recruited fish (90% for May- June, 75% for October-November). In winter, when the number of fish settling was low, recruitment nevertheless occurred on at least 4 or more sites dur- ing the quarter moons. The periodic appearance of PL-l's on 11 of the 20 sites was virtually continuous at each quarter moon, and similar in general pattern to that depicted by the pooled data (e.g., site 11, Fig. 2). On 4 sites, 421 FISHERY BULLETIN: VOL. 83. NO. 3 however, PL-l's settled in spring-summer or in summer-fall. These differences in the pattern of settlement between the census sites are highly signi- ficant (P « 0.001, "distribution-free" nonparametric test; Hollander and Wolfe 1973, p. 139-146). The patterns were so consistent, however, that we were soon able to predict with considerable success not only when, but which sites would attract PL-l's. We can provide no correlations with depth or area of a site, nor with the species of corals and other major residents (fishes, sea urchins, etc.), as to why some sites consistently "attracted" postlarval grunts and others varied, even though it is known that post- larval settlement can be partly governed by priority effects on reefs (Shulman et al. 1983). To understand the characteristics of a site that make it highly "attractive" to settling postlarval grunts will require sophisticated field manipulations. DISCUSSION The most conspicuous features concerning the recruitment of French grunts from the plankton are the rhythms in settlement and fertilization dates (Figs. 2-4). The continuing and short 15-d periods of these rhythms over an entire year contrast with other coral reef fishes where settlement, although rhythmic, is concentrated mostly around a particular period of the year (Williams and Sale 1981). Settle- ment in reef fishes is often episodic; it may show a lunar periodicity (Johannes 1978), it may be monthly but not coupled to a particular phase of the moon (Williams 1983 -mixed guild of pomacentrids), or a rhythm may not be conspicuous (Victor 1982, 1983 - Thalassoma bifasciatum). The short period between fertilization and settle- ment implies that pelagic existence in French grunts is of short duration (about 2 wk; Brothers and McFarland 1981). To estimate the days spent in the plankton, the average age for the smallest 100 PL-l's was calculated from the sample of 1,478 fish that were used to estimate the actual day of fertiliza- tion [mean age = 15.7 d ± 2.1 (SD). range = 13-20; mean length (SL mm) = 6.9 ± 0.67 (SD), range = 5.9-8.5]. If each of these fish had settled from the plankton over the previous day, then pelagic exist- ence (about 15 d) agrees with the periodicity of ferti- lization and settlement (Table 3, Fig. 4). In contrast, the pelagic existence of most other coral reef fishes investigated exceeds 15 d (Randall 1961; Johannes 1978; Sale 1980; Bariow 1981). This has been veri- fied by results for age at settlement as determined from otoliths (Victor 1982, 1983; Brothers et al. 1983; Brothers and Thresher in press and unpublish- ed; Thresher and Brothers in press). Most of these fishes settled at various ages: Thalassoma bifascia- tum, 40-72 d; Paragobiodon melanosoma, 39-47 d; Gobiodon sp., 27-38 d; several unidentified scarids, 34-58 d; and lab rids, 21-56 d. Direct aging of new recruits or otolith counts to presumed settlement marks rarely yield ages as low as found for the French grunt. Examples of species with pelagic phases of 20 d or less include the angel fish, Holacan- thus paru (Brothers and Thresher in press); several damselfishes, Glyphidodontops rollandi, Pomacen- fyits amboinensis, P. popei, and P. wardii; the blenny Petroscirtes mitratus; the nemipterid Scolopsis dubiosus (Brothers et al. 1983); and the goby Gobio- soma prochilos (Brothers, unpubl. data). These may be exceptional cases; all or at least most of these species represent families characterized by having larval durations moderately to considerably longer than found in grunts. At the family level, therefore, only the haemulids, and perhaps the lutjanids, pro- vide evidence of a short pelagic existence. In support of this conclusion is the lack of haemulids or lutjanids amongst the larval fishes collected offshore through- out the Caribbean (Richards 1981, footnote 8). Are haemulids programmed for short larval lives? We cannot be sure, but out of the 1,478 recently settled French grunts we have aged, not one exceeded 20 d from fertilization. Such a developmental process would be crucial in their survival; if they do not drift over suitable substrates on which to settle after 2 wk, they would perish offshore. Our observations of a strong semilunar periodicity in French grunt recruitment, coupled with what seems to be a relatively fixed or invariant larval duration, could be the result of a number of different combinations of spawning and survivorship. Because we can only determine fertilization dates for indivi- duals that have successfully recruited, we cannot be certain whether the apparent periodicity in spawn- ing is an accurate representation of the temporal pat- tern of grunt reproduction. At the other extreme, it may be the result of relatively continuous spawning activity, the products of which survive differentially with respect to semilunar environmental variables. Surprisingly, for such common fishes, little is known about spawning in haemulids. They produce pelagic eggs and larvae (Breder and Rosen 1966; Saksena and Richards 1975). Recently, paired spawning has been described for Hapalogenys mucronatus in aquaria (Suzuki et al. 1983). On six *W. Richards, Southeast Fisheries Center, National Marine Fisheries Service. NOAA. 75 Virjcinia Beach Drive, Miami. P^L 33149-1099, pers. commun. October 1980. 422 McFARLANP ET AI..: RKCKl'ITMENT PATTERNS IN FRENCH GRUNTS specific attempts to validate spawning in French grunts in the sea at St. Croix, two of us (McFarland and Shulman) failed to observe reproductive activity during dusk, but we did observe what appeared to be sporadic spawning by small groups of tomtate grunts, Haemulon aurolineatum, within large schools of these fish. These limited data suggest that grunts, like many reef fishes, cast pelagic eggs into the water column at dusk. We emphasize, however, that there are no data about their daily spawning habits. Two hypotheses offer explanations for the domi- nant semilunar periodicity of fertilization and settle- ment. Hypothesis 1. Assumption: Spawning follows a semilunar rhythm with breeding peaks closely cou- pled to the quarter moons (and/or intermediate monthly tides). If reproductive activities in French grunts follow a semilunar cycle, and pelagic life is programmed for 15 d, then settlement should occur most often during the quarter moons and interme- diate monthly tides, which it does (see Table 3, Fig. 6). This hypothesis, however, does not account for the weekly peaks in settlement and fertilization (Table 3; Figs. 2, 3, 5), which contradict the assump- tion of the hypothesis. Hypothesis 2. Assumption: Spawning is relatively constant from day to day, and larval existence restricted to about 15 d. Consistent daily reproduc- tive effort could produce a continuous pool of grunts in the plankton. As a result, only those larvae that are favored by "correct" currents that disperse them inshore around 15 d will settle, and/or successful recruitment may also depend on favorable currents dispersing the eggs at the time of reproduction. At St. Croix we suspect that current conditions are most favorable to settlement during the quarter moons and intermediate tides. The weekly peaks that are associated with full and new moons (Table 3) could represent recruitment in less favorable cur- rents than occur around the quarter moons. These currents, nevertheless, must allow some larv^ae through the "filter screens" that all recruits must pass through to join a reef community (Smith 1978). The semilunar rhythm of fertilization and settlement observed in settling grunts would, under this hypo- thesis, be explained by semilunar rhythms in cur- rents favorable to settlement. The time of day when settlement occurs and, espe- cially, the amount of darkness at night may also relate to successful settlement. Although data are scarce, some reef fishes settle from the plankton at night (e.g., acanthurids- Randall 1961; McFarland, unpubl. data). We do not have similar direct observa- tions for French grunts, but we suspect settlement is nocturnal because the number of PL-l's remained fairly constant on most census sites throughout each day. Nocturnal settlement behaviors would tempora- rily remove recruits from the attack of reef pisci- vores, especially if accomplished during the dark of the moon. During full moons, recruits presumably would be subject to higher rates of predation by planktivores than during other phases of the moon (Hobson et al. 1981). The number of grunts settling during full moons should be low, which it is (Fig. 4). In contrast, if length of darkness is a dominant factor to successful settlement, then the highest influx of recruits should coincide with the new moons, and intermediate numbers with quarter moons, which they do not (Fig. 4). The recruitment dynamics, nevertheless, indicate higher influxes of PL-l's when some degree of darkness occurs during each night (Fig. 4). We hypothesize that recruitment is depen- dent primarily on favorable currents, with predation possibly acting as a secondary selective force. Reversals in surface currents and oppositely flow- ing currents in the upper 30 m, which could affect the dispersal of larvae, are known to occur at St. Croix and at Puerto Rico (Gladfelter et al. 1978; Lee et al. 1978; Molinari et al. 1980). Their local patterns, however, remain unknown. Eddy formation to the west of the island of Barbados, produced by Karmen trails as the generally west-setting current passes the island, has also been postulated to retain the pela- gic stages of inshore species (Emery 1972). It is reasonably well established that seasonal shifts in local current gyres in the vicinity of Hawaii favor the settlement of various species of reef fishes (Sale 1970, 1980; Johannes 1978). In addition, spawning is often synchronized to disperse eggs and lan-ae away from reefs and into offshore currents (Johannes 1978; Lobel 1978). But specific currents do not always trigger spawning, as Colin (1982) reported for several reef fishes at Puerto Rico. In these in- stances, rhythmic spawning often can be related to the lunar cycle, but significantly, some species spawn every day. Depending on the time of spawning, dispersal routes for fish eggs and larvae can vary over short-time periods because of reversals in cur- rents. The daily behaviors and distributions of grunts at sea are unknown. Do larval grunts passively drift with currents? Or do they seek different depths at different times of the day? Active behaviors that would utilize differences in currents have been in- voked to explain the retention of pelagic larval fishes and invertebrates close to the island of Oahu (Leis 1982). Similar activities by larval French grunts 423 FISHERY Bl'LLETIN: VOL. 83, NO. 3 could explain their absence in collections taken off- shore (Richards 1981). We have provided substantial evidence that French grunts recently recruited from the plankton are fertilized about 15 d earlier, and that these pro- cesses most closely correlate with the quarter moons and intermediate tidal excursions. Is this a general pattern that occurs throughout the Caribbean and western Atlantic where French grunts are most abundant? Or is the lunar-tidal correlation the result of local conditions? We have no answer at present, because it requires repeating the investigation in other localities. Different current regimes and local hydrographic conditions in other regions might elicit different recruitment patterns. The much weaker weekly fertilization and settlement patterns, for ex- ample, that are associated with full and new moons (Table 3) might dominate recruitment in other locales. There is a seeming order in the rain of young French grunts from the plankton. Over the period of this study young grunts following a semilunar time- table appeared on over half of the census sites 70% of the time (range 59 to 94%). Although we could not discern any special characteristics of these sites that attracted grunts, the sites were never preoccupied by other species. Settlement certainly did not appear to be a random phenomenon. The recurrent order in the occupation of space by settling French grunts, however, may reflect only their high abundance in the reef communities at St. Croix (Gladfelter and Gladflelter 1978). Large populations produce large numbers of offspring and this alone might swamp available sites. In this regard we stress that the cen- sus settlement sites are not main reefs, but isolates adjacent to them. Indeed, large coral domes in the census area which sustained large populations of older juvenile French grunts never sustained popula- tions of PL-l's. Without manipulative studies of grunts on the settlement sites, however, it is impossi- ble to ascertain to what extent recruitment is in- fluenced by a resident population (Shulman et al. 1983). We can conclude only that the recruitment of French grunts from the plankton has high temporal and spatial predictability at St. Croix. Fishes in the family Haemulidae represent a domi- nant component of the tropical reef fish community and constitute a major part of the trap fishery in western Atlantic and Caribbean waters (Dammann 1980; McFarland 1980). The findings presented in this study provide insights that are critical to the management of any fishery for grunts. For example, the relatively short larval existence in French grunts, which also occurs in white and tomtate grunts (Brothers and McFarland 1981), implies that dispersal does not occur over very long distances. Whether local populations of grunts are self-sustain- ing or dependent on interisland transport is unknown. Answers will require precise knowledge of spawning habits of each species of grunt, careful analysis of local and general current regimes, and, if possible, the behavior of the larvae. If local popula- tions are self-sustaining, then overly lenient limits on the harvest of adult grunts could seriously limit recruitment. In contrast, if recruitment shows large spring and fall peaks in other species of grunts, as it does in French grunts, and the seasonality in recruit- ment is geographically widespread (e.g. Munro et al. 1973), then closure of a grunt fishery for a few mon- ths during periods of peak recruitment would prob- ably assure maintenance of the local populations. ACKNOWLEDGMENTS We thank Jack Sobel for his technical help in con- tinuing the daily census routines when we could not be present, and Kim Benson who assisted in the otolith analysis. Numerous students at the West In- dies Laboratoiy assisted as diving buddies; we thank them all. This project was supported by National Science Foundation Grant OCE-7918569. LITERATURE CITED Barlow. G. W. 1981. Patterns of parental investment, dispersal and size among coral-reef fishes. Environ. Biol. Fishes 6:65-8.5. Batschelet, E. 1965. Statistical methods for the analysis of problems in animal orientation and certain biological rhythms. Am. Inst. Biol. Sci., Wash., DC, 57 p. Breder, C. M., and D. E. Rosen. 1966. Modes of reproduction in fishes. Nat. Hist. Press, Garden City, NY, 941 p. Brothers, E. B., and W. N. McFarland. 1981. Correlations between otolith microstructure, growth, and life history transitions in newly recruited French grunts [Haemulon flavolmeatum (Desmarest), Haemulidae]. Rapp. P.-v. Reun. Cons. Perm, int Explor. Mer 178:369.374. Brothers, E. B., and R. E. Thresher. In press. Pelagic duration, dispersal and the distribution of Indo-Pacific coral-reef fishes. NOAA Symp. Ser. Undersea Res., Vol. 3. Brothers, E. B., D. McB. Williams, and P. F. Sale. 1983. Length of larval life in twelve families of fishes at "One Tree Lagoon", Great Barrier Reef Australia. Mar. Biol. (Berl.) 76:319-324. Colin, P. L. 1982. Aspects of the spawning of Western Atlantic reef fishes. NOAA Tech. Memo. NMFS-SEFC-80, p. 69-78. Dale, G. 1978. Money-in-the-bank: a model for coral reef fish existence. Environ. Biol. Fishes 3:103-108. 424 McFARLAND ET AL.: RECRUITMENT I'A'ITERNS IN FRENCH GRfNTS Dammann, a. E. 1980. Caribbean reef fish: fish traps and management. Proc. Gulf Caribb. Fish. Inst. 32:100-105. Emery, A. R. 1972. Eddy formation from an oceanic island: ecological effects. Caribb. J. Sci. 12:121-128. Gladfelter, W. B., and E. H. Gladfelter. 1978. Fish community structure as a function of habitat struc- ture on West Indian patch reefs. Rev. Biol. Trop. (Suppl. 1) 25:65-84. Gladfelter, W. B.. E. H. Gladfelter, R. K. Monahan, J. C. Ogden, and R. F. Dill. 1978. Environmental studies of Buck Island Reef National Monument. U.S. Dep. Inter., Rep. Natl. Park Serv., 124 p. Helfman, G. S. 1978. Patterns of community structure in fishes: summary and overview. Environ. Biol. Fishes 3:129-148. Helfman, G. S., J. L. Meyer, and W. N. McFarland. 1982. The ontogeny of twilight migration patterns in grunts (Pisces: Haemulidae). Anim. Behav. 30:317-326. HOBSON, E. S. 1968. Predatory behavior of some shore fishes in the Gulf of California. U.S. Fish Wildl. Serv., Res. Rep. 73, 92 p. HoBsoN, E. S., W. N. McFarland, and J. R. Chess. 1 98 1 . Crepuscular and nocturnal activities of Calif ornian near- shore fishes, with consideration of their scotopic visual pig- ments and the photic environment. Fish. Bull., U.S. 79:1-30. Hollander, M., and D. A. Wolfe. 1973. Nonparametric statistical methods. John Wiley and Sons, Inc., N.Y., 503 p. Johannes, R. E. 1978. Reproductive strategies of coastal marine fishes in the tropics. Environ. Biol. Fishes 3:65-84. Lee, T. N., R. S. C. Munier, and S. Chiu. 1978. Water mass structure and variability north of St. Croix, U.S. Virgin Islands, as observed during the summer of 1977 for OTEC assessment. Rosenstiel Sch. Mar. Atmos. Sci., Univ. Miami, Tech. Rep. 78004, 81 p. Leis, J. M. 1982. Nearshore distributional gradients of larval fish (15 taxa) and planktonic crustaceans (6 taxa) in Hawaii. Mar. Biol. (Berl.) 72:89-97. Lobel, p. S. 1978. Diel, lunar, and seasonal periodicity in the reproductive behavior of the pomacanthid fish, Centropyge potteri, and some other reef fishes in Hawaii. Pac. Sci. 32:193-207. McFarland, W. N. 1980. Observations on recruitment in haemulid fishes. Proc. Gulf Caribb. Fish. Inst., 32:132-138. In press. Overview: the dynamics of recruitment in coral reef- organisms. NOAA Symp. Ser. Undersea Res., Vol. 3. McFarland, W. N., and Z.-M. Hillis. 1 982 . Observations on agonistic behavior between members of juvenile French and white grunts-family Haemulidae. Bull. Mar. Sci. 32:255-268. McFarland, W. N., and N. M. Kotchian. 1982. Interaction between schools of fish and mysids. Behav. Ecol. Sociobiol. 11:71-76. McFarland, W. N., and J. C. Ogden. In press. Recruitment of young coral reef fishes from the plankton. NOAA Symp. Ser. Undersea Res., Vol. 3 McFarland, W. N., J. C. Ogden, and J. N. Lythgoe. 1979. The influence of light on the twilight migrations of grunts. Environ. Biol. Fishes 4:9-22. MoLiNARi, R. L., D. K. Atwood, C. Duckett, M. Spillane, and I. Brooks. 1980. Surface currents in the Caribbean Sea as deduced from satellite tracked drifting buoys. Proc. Gulf Caribb. Fish. Inst, 32:106-113. MtiNRO, J. L., V. C. Galt, R. Thompson, and P. H. Reeson. 1973. The spawning seasons of Caribbean reef fishes. J. Fish Biol. 5:69-84. Ogden, J. C, and P. R. Ehrlich. 1977. The behavior of heterotypic resting schools of juvenile grunts (Pomadasyidae). Mar. Biol. (Berl.) 42:273-280. Ogden, J. C, and J. C. Zieman. 1977. Ecological aspects of coral reef - seagrass bed contacts in the Caribbean. In D. L. Taylor (editor), Proc. Third Int. Coral Reef Symp., Vol. 1, p. 378-382. Rosenstiel Sch. Mar. Atmos. Sci., Univ. Miami, Miami, F^L. Randall, J. E. 1961. A contribution to the biology of the convict surgeonfish of the Hawaiian Islands, Acanthurus triostegns sandvicensis. Pac. Sci. 15:215-272. Richards, W. J. 1981. Kinds and abundance of fish larvae in the Caribbean Sea. Rapp. P.-v. Reun. Cons. Perm. int. Explor. Mer 178: 240-241. Ricklefs, R. E. 1979. Ecology. 2d ed. Chiron Press, N.Y., 966 p. Saksena, V. P., and W. J. Richards. 1975. Description of eggs and larvae of laboratory- reared white grunt, Haem.ulon plumi£ri. (Lacepede) (Pisces, Poma- dasyidae). Bull. Mar. Sci. 25:523-536. Sale, P. F. 1970. Distribution of larval Acanthuridae off Hawaii. Copeia 1970:765-766. 1977. Maintenance of high diversity in coral reef fish commu- nities. Am. Nat. 111:337-359. 1978. Coexistence of coral reef fishes- a lottery for living space. Environ. Biol. Fishes 3:85-102. 1980. The ecology of fishes on coral reefs. Oceanogr. Mar. Biol., Annu. Rev. 18:367-421. Shulman, M. J., J. C. Ogden, J. P. Ebersole, W. N. McFarland, S. L. Miller, and N. G. Wolf. 1983. Priority effects in the recruitment of juvenile coral reef fishes. Ecology 64:1508-1513. Smith, C. L. 1978. Coral reef fish communities: a compromise view. Environ. Biol. Fishes 3:109-128. SoKAL, R. R., and F. J. ROHLF. 1981. Biometry. The principles and practice of statistics in biological research. W. H. Freeman and Co., San Francisco, 859 p. SuztJKi, K., S. HiOKi, Y. Tanaka, and H. Kitazawa. 1983. Spawning and early life historj' of Hapalogenys mucro- natus (EYDOUX et SOULEYET) (Pisces: Pomadasyidae) in an aquarium. J. Fac. Mar. Sci. Technol., Tokai Univ. 16:183- 191. Thresher, R. E. 1984. Reproduction in reef fishes. T.F.H. Pub!., Neptune City, NJ, 399 p. Thresher, R. E., and E. B. Brothers. In press. Reproductive ecology and biogeography of Indo- West Pacific angelfishes (Pisces: Pomacanthidae). Evolu- tion. Victor, B. C. 1982. Daily otolith increments and recruitment in two coral- reef wrasses, Thalassoma bifasciatum and Halwkoeres biint- tatus. Mar. Biol. (Beri.) 71:203-208. 425 FISHERY BULLETIN: VOL. 8:^. NO. 3 1983. Recruitment and population dynamics of a coral reef Williams, D. McB., and P. F, Sale. fish. Science (Wash.. DC) 219:419-420. 1981. Spatial and temporal patterns of recruitment of juvenile Williams, D. McB. coral reef fishes to coral habitats within "One Tree Lagoon", 1983. Daily, monthly and yearly variability in recruitment of a Great Barrier Reef. Mar. Biol. (Berl.) 65:245-253. guild of coral reef fishes. Mar. Ecol. 10:231-237. 426 THE HARBOR PORPOISE, PHOCOENA PHOCOENA, IN FISH HARBOUR, NEW BRUNSWICK, CANADA: OCCUPANCY, DISTRIBUTION, AND MOVEMENTS David E. Gaskin and Alan P. Watson^ ABSTRACT The distribution, movements, and relative population abundance of harbor porpoises were studied in the Fish Harbour region of New Brunswick, Canada (lat. 44°59'30"-45°0r00"N, long. 66°54'00"-66°57'00"W), from 1970 to 1978. In any given year numbers of this species were highest in the region between late July and early September. This is also the period during which surface temperatures attain a maximum (10°-12°C) and the largest herring, Clupea harengus, catches are usually made. During July-September the porpoise population of the inner (western) part of the study area contained 63% mothers with calves. Changes in relative population abundance were most strongly related with time of year (increasing from early July), tidal amplitude (most present when amplitude is 6.5 m or less), and wind phase (most present during onshore winds). Observation of recognizable individuals revealed consistent specific "territories" and patrolling patterns. A marked decrease in relative abundance in the latter half of the 1970s was noted. This decrease was coincident with a decline in mean midsummer surface temperature over 1974-78 of about 1°C. Unfortunately it was not possible to determine from existing data if major changes in availability of prey species also occurred in the inner Quoddy region during the same period. Literature concerning ecology of the harbor por- poise, Phocoena phocoena, has been largely restricted to reports of nonquantitative or casual observations (Scheffer and Slipp 1948; M<)hl-Hansen 1954; Amun- din and Amundin 1974), although more recently Gaskin (1977), Prescott et al. (1981),^ Flaherty and Stark (1982),3 Kraus et al. (1983)," and Taylor and Dawson (1984) have presented information from surveys and some population estimates for limited areas of both east and west coasts of North America. A population estimate for the Copper River area of Alaska was provided by Matkin and Fay (1980). Habitat indices, relating occurrence of animals in coastal waters to various environmental factors have been given by Smith and Gaskin (1983) and Watts 'Department of Zoology. Universitv of Guelph, Guelph, Ontario NIG 2W1, Canada. 2Prescott, J. H., S. D. Kraus, P. Fiorelli, D. E. Gaskin, G. J. D. Smith, and M. Brander. 1981. Harbor porpoise (Phocoemi phocoena): Distribution, abundance, survey methodology and preliminary notes on habitat use and threats. Final report to U.S. Department of Commerce, NOAA, National Marine Fisheries Ser- vice Contract 80-FA-d-00009, 61 p. ^Flaherty, C, and S. Stark. 1982. Harbor porpoise (Phocoena phocoena) assessment in "Washington Sound". Final report for Subcontract 80-ABA-3584, NOAA, NMFS, National Marine Laboratory, Seattle, WA, 84 p. ^Kraus.'S. D., J. H. Prescott. and G. S. Stone. 1983. Harbor porpoise, Phocoena phocoena, in the U.S. coastal waters of the Gulf of Maine. A survey to determine seasonal distribution and abun- dance. A report to the National Marine Fisheries Service, Woods Hole, Massachusetts, July 1983, 15 p. Manuscript accepted October 1984. FISHERY BULLETIN: VOL. 83. NO. 3. 1985. and Gaskin (in press), and some radiotelemetric studies of distribution and movements by Gaskin et al. (1975) and Read and Gaskin (1983, in press). In order to gain insight into the nature of local distributions and movements of P. phocoena over an extended period, we selected a semi-enclosed area of limited size in southern New Brunswick, Canada. Preliminary observations had already shown that the species was common in the area, and locality (about 8 km- maximum) bounded approximately by lat. 44°59'30"-45°59'00"N, long. 66°54'00"-66°57'00"W on the east coast of Deer Island (Fig. 1) was easily ac- cessible for research and relatively sheltered. We were also aided by the occurrence in the area of several recognizable porpoises with surface disfigurements, large blemishes, or distinctive pigmentation patterns visible at some distance in favorable light; several of these animals were pres- ent in successive seasons. CHARACTERISTICS OF THE STUDY AREA (Figure 1) Average depths range from about 8 m in Lords Cove to about 22 m in Simpson's Passage (respective maxima 20 m and 60 m) at mean low water. The region is subject to anomalistic, semidiurnal tides of 427 FISHERY BULLETIN: VOL. 83, NO. 3 Figure 1. -The Fish Harbour region on the east coast of Deer Island region, New Brunswick, Canada (see inset). Legend: Straight solid lines indicate approximate limits of study area. Stipple indicates shallow shelf area (low water mark to 5 m); widely spaced vertical lines indicate areas of relatively gentle slope; narrowly spaced bars indicate steep scarps or slopes terminating in cliffs and the white areas (other than within island boundaries) are relatively deep basins and channels. Broken line shows regular track of boat with watch-stations 1 and 2. Place names are those used in the text. Scale = 1 km. 5.6-8.3 m (mean about 6.3 m) (Forrester 1960). Water turbidity is high (Secchi disk reading in July- August about 2 m). Most tidal flow from the area is by the northern end of Simpson's Passage. Salinities in the whole Quoddy region (Trites 1962) range from 29.57oo (March-May) to 32.3%o (September). Lower salinities in shallow inshore waters occur after heavy rain and may last for several days. While some year- to-year changes in mean surface temperature were recorded (Trites 1962, fig. 11), one of the most com- plete series (for 1978) provides a typical sequence of the relative changes from spring to autumn. Late May-early June surface temperatures ranged from 6.0° to 7.1°C, mid-July from 8.4° to 9.6°C, and mid- August-early September from 10° to 11.2°C. By early December the waters were 7°C again. Peak summer values of 11.3°-14.5°C were only obtained in mid-August at the head of Lords Cove in shallow water (< 5 m). At any given time surface tempera- tures were 0.5°-1.0°C warmer than those at 12 m despite the tidal mixing which obviously occurred. A general pattern of temperature profiles could be recognized in midsummer: The inner (western) part of Fish Harbour was vertically well-mixed through- out, but a rather steep horizontal gradient of about 1.0 °C was maintained at all depths on the outer margin of this zone. In the central region (near Fish Island) the direction of water movement showed up to 90° variation at different depths at any given time of rapid tidal flow. The outer parts of Fish Harbour, and Simpson's Passage, tended to be well-mixed ver- tically, but were always about 0.5° cooler at the sur- face on the ebb. North of Adam Island there was a shift from horizonal gradients on flood to vertical gradients on the ebb. The inshore water is probably "trapped" in the bight of the harbor, i.e., with an ex- tended residency time, while movement of water in and out the northern and southern entrances prob- ably creates some mixing in the outer and central part of Fish Harbour. More detail of the static and dynamic characteristics of the study area and adja- cent regions is provided by Smith et al. (1984). 428 GASKIN and WATSON: HARBOR PORPOISE METHODS Most observations were made from a 5 m two- seater kayak (weather permitting), which appeared to exercise a negligible effect on the behavior of animals even at short distances (Watson 1976). A small fishing boat was used above Beaufort wind force 2, but this craft lacked the silent maneuverability of the kayak. Few publishable photographs were obtained because this species is an exceedingly difficult photographic subject. The method of making and maintaining contact was con- sistent throughout the study period (1 June 1970-10 September 1978). A central route from the head of Lords Cove was followed (Fig. 1); unless weather was too poor for optimal sighting in Simpson's Passage, or there was specific intent to track a group within Fish Harbour, the boat continued on track to watch-station 2 (Fig. 1). If no animals were present in Fish Harbour and conditions were less than Beau- fort wind force 2, the boat would remain in the vicin- ity, but if the animals were already within Fish Har- bour, the boat usually returned to watch-station 1 in outer Fish Harbour, from which point most of the study area normally used by porpoises could be kept under eye or binocular surveillance. As weather per- mitted, or presence and movements of animals dic- tated, this search pattern was usually repeated dur- ing the day at invervals of several hours. When porpoises were located either visually or from the sound of their expirations, the position of the boat and the initial positions of animals relative to the vessel were recorded to within a few tens of meters by taking 3-5 bearings from the many surrounding landmarks. Subsequent positions were noted in those cases when movements were tracked for extended periods, providing the animals surfaced long enough for accurate bearings to be taken and had moved to a significant distance (ca. 100 m) from the previous location. The time, location, size, and apparent com- position of each porpoise group was recorded on and later transcribed from tape. The particular problems of quantifying sightings of this small cetacean have been considered by Gaskin (1977), Prescott et al. (footnote 2), Taylor and Dawson (1984), Kraus et al. (footnote 4, 1983), and Watts and Gaskin (in press). During the present study we used only data obtained in Beaufort Sea state 0-1, when visibility was unlimited within the study area and lighting was direct or diffuse, but uniform. In these ideal condi- tions some porpoises can be detected even from the air (250 m) at 650 m, although this is the least satis- factory method for sighting this species (Kraus et al. 1983). In a simultaneous comparison of effectiveness of stationary, mobile, and aerial observers (Kraus et al. 1983) the former, whether on land or stationary vessel, not only saw far more animals than the observers in the aircraft, but could detect them using binoculars in calm water at 1,000 -(- m. The prob- bability of contact is enhanced when animals tend to surface repeatedly. The present authors found that porpoises working near the Nub close to the Simp- sons Island (Fig. 1) could be clearly seen with binocu- lars from watch-station 1, at a distance of > 1 km in optima] conditions. Some natural constraints on distribution of por- poises within the study area reduced the effective survey region to only 4 km-. Virtually no animals ventured farther inshore than the shallow shelf edge (ca. 5 m) even during high water. Almost all ingress and egress was between Bean and Fish Islands (Fig. 1). Very few animals left by the southern passage from Lords Cove, and only two were ever noted to enter this way. Only one group of animals consistent- ly visited Lords Cove in any case. No movement was noted through the narrow gut west of Hardwood Island, and only infrequent movement (< 5%) through the passage north of Fish Island. Harbor porpoises in the Quoddy region appeared generally indifferent to boat traffic (Watts and Gaskin in press), with no noticeable attraction bias as noted for Phocoenoides dalli by Kasuya and Jones (1984). Because of the small size of the study area, the impossibility of making random transects, the highly clumped distributions, and nonrandom move- ments of animals (see latter), any kind of line- transect approach was inappropriate. From a combination of the initial strip census search and the stationary observations, we concentrated on obtain- ing a "best estimate" of the mean number of animals present during each observation period with reference to each hour of the day and each tide sub- phase within that period. All are minimum estimates because some animals were probably missed at the seaward periphery of the study area. We tried to avoid inclusion of repeat sightings in these estimates which would lead to an upward bias, except where we calculated simple sightings of individuals per hour. The "best estimate" data were used to give relative abundance with respect to various en- vironmental conditions, while sightings per hour were more appropriate for time-based comparisons such as relative abundance from year to year. When sighting conditions were particularly favor- able, the movements of specific groups were plotted, either by observation from one or other watch- stations which commanded a wide view with many landmarks, or by discreetly following them in the 429 FISHERY BULLETIN: VOL. 83. NO. 3 kayak on a parallel course at distances of 50-150 m. If only one observer was present in the kayak, these periods were only included in the main data base if very few animals were known to be in the study area. STATISTICAL ANALYSIS For statistical analyses of these data we con- sidered the potential application of log-linear and multivariate models, principle component analysis (PCA), factor analysis (FA), and a categorical data (x^) procedure. Considerable differences in the magnitude of x-variate variances would make results from PCA or FA suspect (Maxwell 1977, ch. 4). Log- linear and multivariate approaches were initially attractive, but both have disadvantages. Tests for in- dependence of x-variates are difficult in the latter, and while these are facilitated by the former, results obtained from log-linear models are often difficult to interpret (SAS Institute Inc. 1979, p. 222). Further- more, statistical advisers noted that use of the whole data base was contraindicated in either method because 1) there was of necessity inclusion of linear, nonlinear, and enumeration data types, and 2) there were significant numbers of empty data cells usually resulting from poor weather when operation would have been pointless. The consultants recommended use of the categorical data procedure, not only for the reasons outlined above, but also because the very nature of the x-variates (e.g., lunar cycle and tidal cycle) precluded the existence of complete indepen- dence. From the ecological point of view it was con- sidered more important to relate one dependent variable (relative abundance of porpoises) to a group of variables one at a time than to test for indepen- dence in the latter when the result would likely be spurious. We first determined (P = 0.13 -^ ) that no data set from any x-variate was significantly associated with one part of the "range" of any other by a series of simple paired x^ tests. The main analyses were then carried out on an IBM^ 360 mainframe computer using the PROC FREQ program (SAS Institute Inc. 1979) which used a generalized least squares model to generate x^ values for combinations of the categorical levels between variables. In this case the relative abundance was related to date, time of day, tidal amplitude, tidal phase, lunar phase, extant wind direction, and wind direction 24 h previously. To examine changes in spatial distribution within ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. the Study area, the location of 669 porpoise sightings collected in 1973-75 were plotted with respect to chronological time and time of tide. To avoid repeat sightings only the initial sighting was used in this context. To examine possible changes in distribution by depth and location, the study area was divided in- to three zones based on subsurface features: shallow shelving areas, subsurface slopes and scarps, and relatively deep basins and channels (Fig. 1). Such data cannot be used to deduce actual patterns of movement, but we were able to obtain some informa- tion on these through selective observations of iden- tifiable individuals. Five of these were known to be females (consistently accompanied by calves), one a male (seen briefly copulating with a known female), and one of unknown sex. Each recorded sequential position was plotted for movements within a specific observation period, and for four of the known females data on all observations were plotted and the apparent ranges in consecutive seasons examined. In each case the master sheets had a grid with ordinates 100 m apart. RESULTS Hours of Observation of Porpoises in Fish Harbour Region About 550 h of observation were made between June 1970 and September 1978, and 324 h of this were during an "intensive" study phase encompass- ing the summers of 1973-75. Estimates of changes in relative abundance with respect to environmental parameters were based on 181 h in optimal sighting conditions in this period (see section on Methods). Sporadic observations were maintained by our research group subsequent to 1975 or by auxilliary observers from Marine Research Associates of Lords Cove in most months except for late-December to mid-January. Changes in Relative Abundance During the Year From low relative abundance in the spring and late fall months, numbers of porpoises were highest dur- ing August or September in each year of the study period (Table 1). Highest values in ideal conditions were 8.00/h in Fish Harbour and 7.72/h in Simpson's Passage in September 1973 and August 1974 respec- tively. Observations by Marine Research Associates over the period 1971-77 in the same area, although largely of a casual nature, confirmed the animals can occur in small numbers in any month between Octo- 430 (lASKIN ami WATSON: HARBOR POKCOISK Table 1. — Sightings of harbor porpoises per hour in the Simpson's Passage and Fish Harbour areas, southern New Brunswick, 1970-78. Month 1970 1971 1972 1973 1974 1975 1976 1977 1978 Simpson's Passage Anril 'O 2 May 0.61 ±'0.07 0 0 — — — — — June 0 0.90 ±0.20 0.50 ±0.15 0 — — — 0 0.94 ±0.40 July 2.63 ±0.30 1.91 ±0.20 3.55 ±0.40 3.55 ±0.55 1.97 ±0.65 1.68 ±0.30 2.11 ±0.30 0.76 ±0.40 1.70 ±0.25 August 4.42 ±0.75 3.34 ± 0.60 6.44 ±0.55 4.50 ± 0.60 7.72 ±0.40 6.55 + 0.40 3.37 + 0.45 1.60 ±0.25 1.93 ±0.40 September — 7.46± 1.15 6.33 ±0.50 — — — — 4.44 ±1.00 7,47 ±0.50 October — — — — — — — — 4.24 ±0.02 November — — — — — — — — 0 Fish Harbour May 0 0 — 0 — — — — — June — 0 0 0.70 ±0.01 — — — 0 0 July 0.93 ± 1.70 0.18 ±0.05 3.44 ±0.45 0.21 ±0.04 0.46 ±0.03 0.26 ±0.04 0.92 + 0.08 0.36 ±0.15 0.22 ±0.01 August 4.00 ±0.78 1.04 ±0.40 6.45 ±0.40 4.48 ±0.62 2.08 ±0.70 1.94 ±0.50 2.05 ±0.30 1.17±0.30 0.66 ±0.08 September — 2.41 ±0.35 — 8.00 ±1.80 5.22 ±0.60 — — 3.08±1.12 0.89 ±0.05 October — — — — — — — — 0 November — — — — — — — — 0 'No animals recorded. 'No search effort. ^Standard error of the mean. ber and May, but probably rarely enter Fish Harbour (see next section). Arrival and Departure of Porpoises Each Year in Relation to Sea Temperatures First sightings in Fish Harbour were usually made in mid-late July when surface temperatures attained about 9°C, and never in Lords Cove until about mid- August despite 9°-ll°C being reached at the surface in mid-late July. First regular sightings in Simpson's Passage varied from mid-May to late June. Deter- mining the date of departure of the majority of animals from either sector was difficult because strong autumnal winds invariably interfered with observations from mid-September onwards when relative abundance was still high. In both 1977 and 1978, porpoises were still present in Fish Harbour until the last week of September and in Simpson's Passage until at least 15 October. Occasional animals may venture into the latter area in any month of the year since a very small population usually over- winters in the Quoddy region (Gaskin 1977), and one animal was sighted outside Fish Island on 7 Decem- ber 1982 (B. M. Braune^ and D. E. Gaskin, pers. obs.). We could find no evidence that the distribution of porpoises was directly or indirectly influenced by the rather small daily local variations in sea tempera- tures within Fish Harbour. The most frequently ^B. M. Braune, Department of Zoology, University of Guelph, Guelph, Ontario NIG 2W1, Canada. observed known animal and her consorts would regularly traverse the width of the area (see Figure. 9) and their preferred locations appeared to have specific topographic rather than temperature charac- teristics. Estimating Relative Abundance Because the species does not make long dives (mean submergence 1 min 44 s, Watson and Gaskin (1983)), the required minimum period of observation needed to search the study area was not excessive. From our records we selected 3 wk in August 1972-75, when the probability of animals being pres- ent was high. In a random sample of 40 (i.e., above the minimum size for a "large" statistical sample (Bailey 1959)) search periods in optimum conditions of varying length (5 min to 2 h), the percentage of time that one or more animals was recorded in- creased from 50% for 10-min periods to well over 80% for 15-min periods. All observations of < 15 min were therefore discarded from the data set. If we were only interested in presence or absence, as in the case of simple locations at a given time of day or tide, observations from shorter periods or in Beaufort wind force 2 + were still of some value. We have already outlined the methods for obtain- ing our "best estimates"; it is worth noting that various characteristics of the animals (e.g., short dive times, stereotyped movements, recognizable individ- uals) and of the study area (limited search area because of shallow water, many landmarks, shelter, and limited entry and exit points for animals) were of great assistance in reducing repeat sightings to a minimum. 431 FISHERY BULLETIN: VOL. 83, NO. 3 Changes in Relative Abundance Between 1970 and 1978 Table 1 gives sightings of individuals per hour by month for each year from 1970 to 1978. As some observation periods were eliminated (see above) the results sometimes differ slightly from values given by Gaskin (1977) for the earlier years. Results sug- gest a decline in relative abundance from 1972 on- wards in Fish Harbour and from 1974 onwards in Simpson's Passage. Because the 1970 values were based on a relatively short series of observations, the apparent rise from 1970 to 1971 may be spurious. The slight increase in 1978 is also suspect as most observers were less experienced than the teams used in 1970-77. The overall trend however, seems unmis- takable. Furthermore, a decline in each month of the July-September period is evident when data for the whole study area are pooled (Fig. 2). Results of the categorical data procedure tests for relative abundance of porpoises against the set of en- vironmental parameters are presented in Table 2 (Fish Harbour) and 3 (Simpson's Passage). As might be expected, large x^ and significant P values were generated from the sharp drop in overall numbers sighted per observation period (regardless of its length) from 1972 to 1975, and from the great in- crease occurring each year between July and September. Time of day appeared to exert no in- fluence on relative abundance in either section (P = 0.45, 0.20), nor was any statistically significant rela- tionship noted between relative abundance and 0) c i 1 1 \ 1 1 > 1 > 1 ! s ■ 1 1 1 t 1 1 \ i \ \ 1 1 1 1 ; -'-'rC^'^. Sept i. *\ ^ v^ 1 ^ \ , » Aug. \ i ! .■n July Oct. Figure 2. -Comparison of sightings per hour of porpoises in good condition in the Fish Harbour study area (Fish Harbour + Simpson's Passage), 1970-78 by month, from July to September. Octo- ber and November values available for 1978 only. Vertical dots and bars represent standard error about the means. - Nov. 2 -. » 1970 1971 1972 1973 1974 1975 1976 1977 1978 Year 432 GASKIN and WATSON: HARBOR I'ORF'OISE Table 2.— Results of categorical data procedure tests for observed numbers of fiarbor porpoises In comparison to ex- pected numbers withi different environmental parameters In thie FIsfi Harbour region of souttiern New Brunswick, 1972-75. 1. FIsfi Harbour -(- Lords Cove (see Figure 1). n = observation periods; x^ = two-way table statistic. Parameter measured n df Years (1972-75) Time of year' (July to early September) Time of day^ (0600- -I- 1900) Tidal amplitude' Tidal pfiases" (Start of falling tide to end of rising tide) Lunar phiase* (New moon to full moon) Wind extant^ (On sfiore, coastwise, or offshore) Wind 24 ti previous^ 249 249 249 249 249 3 2 4 3 39.57 0.0001* 63.69 0.0001 • 3.67 0.4520 7.80 0.0500* 3.31 0.8450 249 3 5.38 0.1450 249 2 17.58 0.003 249 2 12.22 0.0094 'Categorical division of ttiree periods (July, first 3 wk of August, last 1 V2 wk of August -1- early September) to balance effort. ^Four periods of 3 h eacti, post-1900 observations amal- gamated. ^Four subpfiases (> 5.5, 5.6-6.5. 6.6-7.5, ^ 7.6 m). 'Thie eigfit subptiases given In Figure 4 were used. ^Three subdivisions used. More subdivision tfian thiese resulted in many empty data cells. ^Four subdivisions used. More subdivision tfian tfiese resulted In many empty data cells. *Statistically significant at 0.05 level or better. Table 3.— Results of categorical data procedure tests for observed numbers of harbor porpoises in comparison to ex- pected numbers with different environmental parameters In the Fish Harbour region of southern New Brunswick, 1972-75. 11. Simpson's Passage, n ~ observation periods; X^ - two-way table statistic. Parameter measured n df Years (1972-75) Time of year' (July to early September) Time of day^ (1600- -I- 1900) Tidal amplitude' Tidal phases" (Start of falling tide to end of rising tide) Lunar phase* (New moon to full moon) Wind extant^ (Onshore, coastwise or offshore) Wind 24 h previous^ 132 132 132 132 132 3 2 4 3 7.58 0.050* 16.99 0.0002* 5.98 0.201 2.54 0.468 9.97 0.190 132 3 7.38 0.061* 132 2 0.95 0.620 132 2 0.93 0.628 'Categorical division of three periods (July, first 3 wk of August, last 1 V2 wk of August -1- early September) to balance effort. ^Four periods of 3 h each, post-1900 observations amal- gamated. 'Four subphase (> 5.5, 5.6-6.5, 6.6-7.5, > 7.6 m). "The eight subphases given in Figure 4 were used. ^Three subdivisions used. More subdivision than these resulted in many empty data cells. *Four subdivision used. More subdivision that these resulted In many empty data cells. 'Statistically significant. **Close to significance at 0.05 level. magnitude of tidal amplitude in Simpson's Passage. In shallower, semi-enclosed Fish Harbour however, significantly more porpoises were present (P = 0.05) at low tidal amplitudes (< 6.5 m) than high amplitudes (^ 6.6 m). Data for the most commonly sighted individual (a female with a large dorsal scar) corroborated this general finding; about 78% of all sightings of this animals were made when the ampli- tude was 6.5 m or less. Relative abundance did not alter (P = 0.84, 0.19) in either zone of the study area during the tidal cycle. Similarly no significant change occurred relative to the lunar phase in Fish Harbour, although in Simp- son's Passage the x^ value approached the arbitrary 0.05 level of significance (P = 0.061; n = 249). No relationship between extant or previous wind direc- tion was evident in Simpson's Passage {P = 0.62, 0.63; n = 132), but there appeared to be a significant association between wind direction and relative abundance of porpoises in Fish Harbour (P = 0.0003, 0.009; n = 249). In both cases far more animals were present during onshore wind directions than when winds were coastwise or offshore. Distribution and movements of porpoises on the fine scale is likely to be correlated with the presence or absence of food species which, in the Quoddy region, consist largely of juvenile herring, Clupea harengiis; mackerel; and small gadoids (Smith and Gaskin 1974). The dispersal of the former in this region is greatly influenced by current velocities (Jovellanos and Gaskin 1983). Unfortunately the beam width of available equipment was far too nar- row to permit us to cover the study area by acoustic survey thoroughly, or even representatively, at any given time or tide phase. Because one of our major aims at the outset of the study was to avoid disturb- ing normal behavior of the porpoises, it was also rarely possible to acoustically scan in their immedi- ate vicinity. We were however able to infer feeding behavior from diving patterns (e.g., "pattern B") car- ried out consistently in one location (Watson and Gaskin 1983). Sometimes fish were seen jumping at the surface in such areas (Fig. 3), and these zones were acoustically scanned on an opportunistic basis. Fish schools, recorded during bottom topographic acoustic mapping runs, tended to be concentrated at several locations in which porpoises were often sighted. Usually these traces were of herring school- type. While this species predominates in the Quoddy region in the summer months (Jovellanos and Gaskin 433 FISHERY BULLETIN: VOL. 83, NO. 3 Figure 3. - Locations in Fish Harbour at which fish were detected visually or by sonar in 1973-78; opportunistic records only. Stipple: species not identified; vertical bars: Atlantic mackerel: diagonal bars: Atlantic herring; circles: one or more porpoises simultaneously present with record; arrows: direction of fish movement. The size of hatched areas indicates approximate size of the school. 1983), some could have been of harbor pollack, which are also taken by this species. Changes in Distribution Within the Fish Harbour Region No differences in distribution of the 669 recorded sighting positions for 1973-75 could be detected when they were plotted either by four or eight time periods from 0600-2200, but considerable differences were evident when they were plotted against eight subphases (slack low, slow flood #1, fast flood, slow flood #2, slack high, slow ebb #1, fast ebb, and slow ebb #2) of the tidal cycle. This analysis ignores for the moment the fact that many animals move in a rather regular manner through the study area, since useful information can be obtained simply from noting where they are at any given time of tide. The results (Fig. 4) showed that up to 80% of porpoises were congregated over the shallow shelf margins, slopes, and scarp areas between the onset of flood tide and high water. Conversely, about the same pro- portion became concentrated over the basin and channel areas between onset of the ebb, and slack low water. The Nature of Movements Within the Fish Harbour Region It became evident that when the recognizable animals returned each year they reestablished "specific ranges" (for lack of a better term) in virtu- ally the same locations in Fish Harbour each summer (Figs. 5-8). There is some evidence that these "ranges" shifted progressively further east of Deer Island between 1973 and 1975, especially in the case of the scarred female mentioned earlier. Each of these areas in Fish Harbour, of which we were able to measure nine approximately, appeared to be about 1.0-1.5 km^ in extent. Even the most distinctive animals would disappear from the study area for ex- tended periods and then reappear again, just as Gaskin et al. (1975) noted for a female carrying a radiotelemetric package. The scarred female was once observed with a calf off the eastern coast of Simpson's Island, and on another occasion between Minke Ledges and Tinker Island, which lie between 1 and 3 km south of the main study area. This speci- men was present in the Fish Harbour region for about half the days of excellent visibility in 1973-75, was well known to all observers, and yet was only 434 GASKIN and WATSON: HARH( )K fOKPOISE 100 Figure 4. -Distributions of 669 harbor porpoises in Fish Harbour 1973-75 over subsurface topographic features, by eight subphases of the tide: Slack low water (SL), slow flood 1 (SFi), fast flood (FF), slow flood 2 (SF2), slack high water (SH), slow ebb 1 (SEj), fast ebb (FE), and slow ebb 2 (SE2). Solid line shows percentage over shallow shelf, scarps, and slopes; dotted line, percentage over basins and channels. 20 10- SFi FFi SF2 SH SEi FE SE2 Subphases of tide SL seen on the two above occasions during simultaneous surveys outside the present study area. Only in the case of this particularly distinctive animal were we able to obtain enough observations to map some of her patterns of movement over ex- tended periods. On 17 August 1973, for example, we tracked her for 2 h 12 min (57 position fixes. Fig. 9), noting that she tended to stay close to the slopes of the main basin-channel, with one foray around the small basin in Lords Cove. This kind of point-to-point travelling interspersed with short periods of submergences in one location is quite typical of this species. This animal repeated almost identical range movements on 13 and 29 August 1974. On 1 and 30 August 1974 (1 h 4 min and 3 h 3 min respectively) this animal spent far longer periods in relatively restricted locations (Fig. 10). Submergences were again of the "pattern B" type, and no surface resting was recorded. On 1 August many herring were seen jumping at the surface immediately after each submergence by the female. Her calf was often left at the surface during these bouts. "Systematic patrolling" of small areas, often in the lee of ledges or small islets was also recorded (Fig. 10). The movements illustrated were carried out by a school of three medium-sized animals accompanied by one small one on 2 August 1974. The group sychronously dove repeatedly while moving back and forth in one restricted scarp location, then abruptly travelled to the second location shown and repeated the pattern. DISCUSSION The distribution of some small odontocetes is known to be correlated with sea surface tempera- tures (Gaskin 1968; Wiirsig and Wiirsig 1980), but the relationship is almost certainly indirect, the result of influences exerted one or more levels fur- ther down the food chain. The entry of the main sum- mer population of harbor porpoises into the study area not only coincides with 9°-10°C surface temperatures, but also with the arrival of large numbers of juvenile herring which feed in the Quod- dy region during the summer months (Battle et al. 1936; Jovellanos and Gaskin 1983). Although trans- port of relatively small fish into the study area might be expected to be at a maximum during spring tides, 435 FISHERY BULLETIN: VOL. 83. NO. 3 Figure 5. -Broken lines encompass all sightings of recognizable animal (#1, scarred mother). 1973; solid wedges; 1974: open wedges; 1975: dots. The solid triangle, open triangle, and solid circle respectively represent the center of the "range" as determined by the mean of the latitudes and longitudes of each position, excluding single isolated sightings. Arrows indicate number of times this animal (and any consorts) were observed leaving the "range". In each case they were swimming at 4 + knots and outdistancing the kayak. Figure 6. - Broken lines encompass all sightings of known animal #2 (clipped fin). 1 974: dotted lines; 1974: dashes. Open circle and closed cir- cle for 1974 and 1975 respectively, and arrows indicating animal leaving the area are as in Figure 5. 436 GASKIN anil WATSON: HAKHOK PdKI'OISK Figure 7. - Broken lines encompass all sightings of known animal #3 (Simpson's Passage mother) except for single 1977 sighting. 1973: dot- ted line; 1974: dashed line. Other symbols as in Figure 5. 1974. ir \ 'z:^-'";--'.. 74 • . • ' • . • 75 I 8 Figure 8. - Broken lines encompass all sightings of known animal if 4 (Light brown mother). 1974: solid wedges; 1974: open wedges. Other symbols as in Figure 5. 437 FISHERY BULLETIN: VOL. 83 NO. 3 Figure 9. - Movements of scarred mother during 2 h 12 min period of 10 August 1973 accompanied by calf (57 position fixes). On some occa- sions they resurfaced close to an immediately previous position; about 75 surfacings are represented by these positions. Reverse arrow in- dicates initial contact point (just outside Lords Cove), solid arrow point where tracking was abandoned, and white circles position of a medium- sized animal which briefly joined the pair. we found relative abundance of porpoises conversely to be greatest during neap tides. About 63% of the sightings in 1973-75 were of mothers and calves. Possibly these animals stay out of semi-enclosed areas during periods of maximum water transport, especially in areas with turbulent shallows such as the Fish Harbour region. On the other hand, if there are larger numbers of herring schools moving into the outer part of the Quoddy region it may not be necessary to forage in inshore waters. Once the animals are in the study area however, the observed changes in distribution during the tidal cycle confirm the earlier subjective comment (Watson 1976) that porpoises seemed to move inshore with the flood tide and offshore with the ebb. The distribution of porpoise activity in the region appeared to be closely related to subsurface topography, particularly the margins of the basins and channels in areas of restricted water flow, the lee of islets, especially Fish Island and the Nub, and blind trench heads such as that just north of Hath- away Head. Because our meager acoustic records revealed fish traces in all these sites at one time or another (Fig. 3), we speculate that these may be areas of fish accumulation which the porpoises have learned to exploit. Each may have subsurface eddies or areas of relatively slack water against the lee side of the slope or near the bottom, in which fish can shelter from strong currents, and in which zooplank- ton might also accumulate. In such areas fish could not only avoid being carried further inshore, but could also feed (Hamner and Hauri 1977). Dispersal from offshore areas into the coastal belt is likely to be augmented by onshore winds, which can increase significantly the flow of water in the immediate sur- face layers. Local fishermen believe that offshore winds serve to hold "feed" away from the coast, while coastwise winds might reduce the possibility of entry into Fish Harbour by increasing surface flow past the relatively narrow passages. We conclude that light or moderate winds are unlikely to affect any small cetacean directly; when winds were strong enough that behavior might conceivably be affected, we were not usually operating. Large variation always occurs in data such as these, and it is to a large extent unpredictable. Such variation in relative abundance can be explained quite simply; these animals are large, highly mobile predators with sophisticated adaptations both for ac- tive and passive acoustic scanning for their equally 438 GASKIN and WATSON: HARBOR PORI'OISE FiGl'RE 10. -Movements of scarred female on 30 August 1974 (solid circles at left of figure) and 29 August 1974 (solid circles in center of tlgure). Solid squares show "patrolling" behavior of school of four animals recorded on 2 August 1974. Qualifying comments on position fixes and surfacings in Figure 9 apply here also. mobile prey. That the harbor porpoises in this area can shift tens of kilometers in a few hours cannot be questioned (Read and Gaskin in press); if few fish are present in one area, they may well move in a loosely synchronized group to forage in other localities. This degree of mobility and scanning ability is highly adapted not only to the mobility of their prey species, but also to the patchy nature of the distribution pat- terns of such fish; there is probably a significant ran- dom element in the dispersal of juvenile herring in the Quoddy region (Jovellanos and Gaskin 1983). The type of porpoise movement shown in Figure 9 was presumed to indicate that prey were dispersed, since the animal spent little time in any one location. In the other movement pattern (Fig. 10) the same animal was believed to be encountering prey in local concen- trations that merited prolonged submergences in one area. We have little evidence that the region might be significant either as a mating area or a calving ground. Females appear to have their calves off- shore, since, although the latter may be very small in late June and July, they already accompany females sighted in the outer part of the Quoddy region (Smith and Gaskin 1983). Many females with calves appear to remain on tidal convergence streaks up to 20 km offshore in the outer Quoddy region (Read 1983) and may not move into the coastal zone at all. It seems more likely that the study area functioned as a signi- ficant feeding area for this species rather than being a zone favoured for reproductive activity. The demonstrated existence of "specific ranges" and annual returns by individual harbor porpoises in this study is not surprising. Rather similar patterns, although sometimes on very different geographical scales, have been recorded for Delphinus delphis by Martin et al. (1971), Orcinus orca by Bigg (1982), Tursiops trun^.atus by Wells et al. (1980) and Wiirsig and Wiirsig (1979), and Lagenorhynchus obscurtis by Wiirsig and Wiirsig (1980). Periodic disappearances and abrupt reappearances of T. truncatics were also recorded by Wiirsig (1978), while studying the animal. With respect to the progressive decline in relative abundance in the peak part of summer, lack of con- sistent observations through late September-late October (because of high winds) prevented us from determining whether or not this resulted from a real population decrease in the area, or simply a shift of peak abundance from mid-August to mid-September during the decade. Consequently we adopted an in- direct approach to the problem, plotting sightings 439 FISHERY BULLETIN: VOL. 83. NO. 3 per hour for each month of summer for each year with sufficient data (Fig. 2). This suggests that the decline resulted from a combination of both factors, with a general overall decline in relative abundance since about 1973 and a progressive retardation of the peak of numbers since about 1975. Unfortunately there is insufficient information in Department of Fisheries and Oceans records to decide if any change in food supply could have occurred during those years, especially with respect to juvenile herring abundance (T. D. Iles^). Nevertheless, our surface temperature records for the inner Quoddy region reveal a consistent decline in the mean summer value (centered on mid-August) from above 12°C in 1971 to below 11°C in 1977 (Gaskin et al. 1979). The temperature curve is compared with the pattern of relative abundance in Figure 11; there is close cor- respondence from 1972 onwards. Such a decline might result from increase in relative dominance of ''T. D. lies, Biological Station, Department of Fisheries and Oceans, New Brunswick, EOG 2X0, Canada, August 1980. Nova Scotia current water in the Bay of Fundy (Sutcliffe et al. 1976) and perhaps a significant decline in the availability of zooplankton prey of juvenile herring. Alternatively the slightly cooler waters might be marginally less suitable for young porpoise calves. ACKNOWLEDGMENTS The authors are very grateful to J. Hines and G. Darlington of the Department of Mathematics and Statistics at the University of Guelph for much assistance and advice during the analysis of the en- vironmental data. G. J. D. Smith and A. J. Read of the Department of Zoology read the manuscript in its several drafts, and along with two anonymous referees gave much useful and constructive advice regarding improvements. Many former and current assistants and graduate students, too numerous to name, helped to collect the laboriously obtained field data. Marine Research Associates of Lords Cove and the late R. Thurber of St. Andrews, N.B., provided 7- 6- C 4- CO Figure 1 1 .- Comparison of midsummer (late •July-early September) sightings per hour of har- bor porpoises in the Fish Harbor region of New Brunswick between 1970 and 1978 with surface temperatures based on 8 stations (n = about 60/annum): Solid circles and dashed line (sight- ings per hour), open circles and dotted line (°C). Vertical dots and bars represent the standard error about the mean. 12.0 .9 .8 .7 .6 .5 .4 .3 .2 .1 1 1.0- .9 .8' I. \ I . \ \ ?. < . ■ I 9 2. 4. : \ ■■•, y ■» 6 T. \ i'- 11 — I 1 1 1 — I 1 — 1970 1971 1972 1973 1974 1975 — I I 1976 1977 1978 Year 440 GASKIN and WATSON: HARBOR PORF'OISE valuable logistic support. The program was funded through Natural Sciences and Engineering Research Council of Canada operating grant A8563 through- out, and in later stages by a subvention from Fisheries and Oceans Canada and a grant from the Canadian National Sportsmen's Fund. LITERATURE CITED Amundin, M., and B. Amundin. 1974. On the behavior and study of the harbour porpoise, Phocoena phocoena, in the wild. In G. Pilieri (editor). Inves- tigations on Cetacea, Vol. V, p. 317-328. Berne, Switzer- land. Bailey, N. J. T. 1959. Statistical methods in biology. Engl. Univ. Press, Ltd., Lond., 200 p. Battle, H. I., A. G. Huntsman, A. M. Jeffers, G. W. Jeffers, W. H. Johnson, and M. A. McNairn. 1936. Fatness, digestion and food of Passamaquoddy young herring. J. Fish. Res. Board Can. 2:401-429. Bigg, M. 1982. An assessment of killer whale (Orcinus orca) stocks off Vancouver Island, British Columbia. Thirty-second Rep. Int. Whaling Comm. SC/Jn81/KW4, p. 655-666. Forrester, W. D. 1960. Current meaisurements in Passamaquoddy Bay and the Bay of Fundy 1957 and 1958. J. Fish. Res. Board Can. 17: 727-729. Gaskin, D. E. 1968. Distribution of Delphinidae (Cetacea) in relation to sea surface temperatures off eastern and southern New Zealand. N.Z. J. Mar. Freshw. Res. 2:527-534. 1977. Harbour porpoise Phocoena phocoena (L.) in the western approaches to the Bay of Fundy 1969-75. Rep. Int. Whaling Comm. 27:487-492. Gaskin, D. E., G. J. D. Smith, and A. P. Watson. 1975. Preliminary study of movements of harbour porpoises {Phocoena phocoena) in the Bay of Fundy using radiotele- rnetn,-. Can. J. Zool. 53:1466-1471. Gaskin, D. E., K. I. Stonefield, P. Suda, and R. Frank. 1979. Changes in mercury levels in harbour porpoises from the Bay of Fundy, Canada, and adjacent waters during 1969-1977. Arch. Environ. Contam. Toxicol. 8:733-762. Hamner, W. M., and I. R. Hauri. 1977. Fine-scale surface currents in the Whitsunday Islands, Queensland, Australia: Effect of tide and topography. Aust. J. Mar. Freshw. Res. 28:333-359. Jovellanos, C. L., and D. E. Gaskin. 1983. Predicting the movements of juvenile Atlantic herring (Clupea karengus harengus) in the SW Bay of Fundy using computer simulation techniques. Can. J. Fish. Aquat. Sci. 40:139-146. Kasuya, T., and L. Jones. 1984. Behavior and segragation of the Dall's porpoise in the Northwestern Pacific Ocean. Sci. Rep. Whales Res. Inst. No. 35, p. 107-128. Kraus, S. D., J. R. Gilbert, and J. H. Prescott. 1983. A comparison of aerial, shipboard, and land-based survey methodology for the harbor porpoise, Phocoena phocoena. Fish. Bull., U.S. 81:910-913. Matkin, C. 0., and F. H. Fay. 1980. Marine mammal-Fishery interactions on the Copper River and in Prince William Sound, Alaska, 1978. U.S. Dep. Commer., N.T.I.S. Rep. PB80-159536, Springfield, VA, 81 p. Maxwell, A. E. 1977. Multivariate analysis in behavioural research. Chap- man and Hall, Lond., 164 p. M6HL-HANSEN, U. 1954. Investigation of reproduction and growth of the por- poise {Phocoena phocoena (L.)) from the Baltic. Vidensk Medd. Dan. Naturhist. Foren. Khobenh. 116:369-396. Read, A. J. 1983. Movements and distribution patterns of harbour por- poises {Phocoena phocoena) in the western Bay of Fundy. M.S. Thesis, University of Guelph, Guelph, Ontario, 87 p. Read, A. J., and D. E. Gaskin. 1983. The application of radio tracking techniques to the study of harbour porpoises {Phocoena phocoena) in the Bay of Fundy. Proceedings of the 5th International Conference on Wildlife Telemetry, Halifax, N.S., Canada, August 1983, 7 p. 1985. Radio tracking the movements and activities of harbour porpoise, Phocoena phocoena (L.), in the Bay of Fundy, Can- ada. Fish. Bull. U.S. 83(2). S.A.S. Institute, Inc. 1979. SAS User's Guide. Edited by J. T. Helwig and K. A. Council. SAS Institute Inc., Raleigh, NC, 494 p. Scheffer, V. B., AND J. W. Slipp. 1948. The whales and dolphins of Washington State wdth a key to the cetaceans of the west coast of North America. Am. Midi. Nat. 39:257-337. Smith, G. J. D., and D. E. Gaskin. 1974. The diet of harbour porpoises {Phocoena phocoena (L.)) in coastal waters of eastern Canada, with special reference to the Bay of Fundy. Can. J. Zool. 52:777-782. 1983. An environmental index for habitat utilization by female harbour porpoises with calves near Deer Island, Bay of Fundy. Ophelia 22:1-13. Smith, G. J. D., C. L. Jovellanos, and D. E. Gaskin. 1984. Near-surface biooceanographic phenomena in the Quod- dy region, Bay of Fundy. Can. Tech. Rep. Fish. Aquat. Sci. 1280, 124 p. SuTCLiFFE, W. H., Jr., R. H. Loucks, and K. F. Drinkwater. 1976. Coastal circulation and physical oceanography of the Scotian Shelf and Gulf of Maine. J. Fish. Res. Board Can. 33:98-115. Taylor, B. L., and P. K. Dawson. 1984. Seasonal changes in densit>' and behavior of harbour porpoise {Phocoena phocoena) affecting census methodology in Glacier Bay National Park, Alaska. Thirty-fourth Rep., Int. Whaling Comm., p. 479-483. Trites, R. W. 1962. Temperature and salinity in the Quoddy region of the Bay of Fundy. J. Fish. Res. Board Can. 19:975-978. Watson, A. P. 1976. The diurnal behavior of the harbour porpoise {Phocoena phocoena) in the coastal waters of the western Bay of Fundy. M.S. Thesis, LTniversity of Guelph, Guelph, Ontario, 94 p. Watson, A. P., and D. E. Gaskin. 1983. Observations on the ventilation cycle of the harbour por- poise Phocoena phocoena (L.) in coastal waters of the Bay of Fundy. Can. J. Zool. 61:126-132. Watts, P. W., and D. E. Gaskin. In press. Habitat index analysis of harbour porpoise {Phocoena phocoena L.) in the southern coastal Bay of Fundy, Canada J. Mammal. 64. Wells, R. S., A. B. Irvine, and M. D. Scott. 1980. The social ecology of inshore odontocetes. In L. M. Herman (editor). Cetacean behavior: mechanisms and func- 441 FISHERY HULLETIN: VOI,. H3, NO. 3 tions, p. 263-317. Wiley-Interscience Publications, N.Y. stops truncatm, in the South Atlantic. F'ish. Hull., U.S. 77: WUKSIG, B. 399-412. 1978. Occurrence and group organization of Atlantic bottle- 1980. Behavior and ecology of the dusky dolphin, La(;(?nor%rir nose porpoises {nirsiops trurwatus) in an Argentine hay. chus obsmrus, in the South Atlantic. Fish. Bull., U.S. 77: Biol. Bull. (Woods Hole) l.'')4:348-359. 871-890. WURSIG, B., AND M. WURSIC. 1979. Behavior and ecology of the bottlenose dolphin, Tur- 442 NOTES THE RELATIONSHIP BETWEEN TILEFISH, LOPHOLATILUS CHAMAELEONTICEPS, ABUNDANCE AND SEDIMENT COMPOSITION OFF GEORGIA Elucidation of the relationship between physico- chemical factors and fish abundance has long been of interest to fisheries biologists. For example, water temperature frequently exerts a strong influence on the abundance of many pelagic marine fishes (Rado- vich 1961; Laurs et al. 1977; Barkley et al. 1978; Moyle and Cech 1982), and this effect has been noted also for freshwater species (Magnuson et al. 1979; Moyle and Cech 1982). For benthic marine fishes, however, substrate composition may be a more im- portant factor affecting fish abundance and distribu- tion. Associations between abundance and substrate composition have been noted for a diverse group of fishes: agonids, bothids, cottids, pleuronectids, scor- paenids, and steichaeids (Day and Pearcy 1968; Powell and Schwartz 1977; Marliave 1977; Barton 1982). Where detectable, however, these associa- tions vary substantially in intensity. This is probably due to the fact that many physicochemical factors are intercorrelated and most fishes probably respond to intercorrelated suites of variables rather than to single factors alone. In this note we quantify the relationship between catch rate of a demersal species, the tilefish, Lopho- latilus chamaeleonticeps, and substrate composition. This species is commercially exploited throughout most of its range (Grimes et al. 1980; Low et al. 1983; Turner et al. 1983), although, prior to this study, tilefish resident to the continental slope off Georgia appeared to have been subjected to minimal exploitation (D. Harrington'). The elucidation of a substrate-abundance relationship for tilefish should aid in the management and harvest of this species. Methods A total of 19 bottom longline sets and 19 sediment samples were obtained during daylight hours, be- tween 5 May and 22 November 1982. Fourteen long- line sets, each comprising 1.6 km of line, and 12 sedi- ment samples (Table 1) were obtained from the RV Georgia Bulldog (University of Georgia Sea Grant Program vessel). Five sets(X ± 1 SD length = 0.31 ± 0.09 km) and seven sediment samples were col- lected aboard the RV Delaware II (National Marine Fisheries Service vessel). At least one of the authors was present during collections. Bottom longlining on the Georgia Bulldog was con- ducted using snap-on gangions (~ 0.5 m in length) spaced about 4 m apart, along a 6.3 mm diameter galvanized aircraft cable groundline. Gangions were equipped with 4/0 or 5/0 circle hooks and baited with either fish or squid. A similar system was employed on the Delaware II except that a much shorter groundline of 6.3 mm diameter hardlaid nylon was used (Table 1), with hook sizes ranging from 3/0 to 8/0. Substrate Analysis Substrate samples were collected with a 25 x 30 X 37.5 cm box dredge suspended from a power winch. The dredge was lowered to the bottom and then dragged across the substrate (typically for < 5 min). After retrieval, 1.2-2.0 kg of sediment were removed from the dredge and stored in plastic bags. It is assumed that these samples accurately reflect the composition of surface sediments. Sediment samples varied in their proximity to longline sets. Fourteen samples were taken at the end of longline sets. Of the remaining five samples, one was taken from the midpoint of a set, three were taken alongside sets within a distance of 0.2 km, and one was taken alongside a set at a distance of 0.6 km. The general area sampled (see Table 1 for loran C coordinates) has a relatively homogenous, low-relief bottom topography, and large variations in substrate composition probably do not occur over short distances (V. J. Henry^). To determine the fraction of each sample compos- ed of sand and silt-clay, a known amount of sediment (i.e., enough to yield a dry weight of between 60 and 100 g) was dried in a forced-air oven at 98°C until a constant weight was reached. The sample was then moistened with water which contained ~ 2 g of Calgon^ as a dispersant, and washed through a sieve which retained particles > 0.0625 mm (4^) (U.S. standard seive #230). Sediments retained by the sieve were then oven-dried to a constant weight to 'D. Harrington, University of Georgia Marine Extension Service, Brunswick, GA 31523, pers. commun. 1983. ^V. J. Henry, Dept. of Geology, Georgia State University, Atlan- ta, GA 30303, pers. commun. 1983. ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 443 Table 1.— Sediment composition and catch data for longline sets used to establish the relationship be- tween catch rates and sediment composition off Georgia. Minimum and Percent maximum Groundline Number Soak Tilefish Percent silt- Loran C depths (m) of length of time per 100 sand clay Date coordinates longline sets (km) hooks (h) hook-h-1 (»4|) ( 0.7). Statistical Analysis To determine the relationship between tilefish 444 catch rate and sediment composition, we used the silt-clay fraction of each substrate sample as an in- dependent variable (X) and catch rate (i.e., tilefish caught/100 hook-h per soak time) as the dependent variable (Y). Exponential and polynomial regression models were fit to the data using the SAS statistical programs (SAS Institute Inc. 1982). The best poly- nomial model was compared with the nonlinear ex- ponential model using R^ as the criterion for model performance. Similar patterns of variation were observed in plots of residuals from all models, hence R^ values are a reasonable criterion for model selec- tion. Results The size structure of tilefish caught off Georgia was typical of unexploited to lightly exploited tilefish stocks (Grimes et al. 1980; Turner et al. 1983). This size structure remained relatively constant for ~ 10 mo, after which a slight decrease in catch rates and a possible truncation of size structure were observed (authors' unpubl. data). These results confirm verbal reports that little exploitation has occurred off Georgia (Harrington footnote 1). Hence, the data used in this analysis were probably not influenced by prior exploitation. A total of 323 tilefish were taken on 19 longline sets (Table 1). Catch rates ranged from 0 to 5.34 tile- fish/100 hook-h. Parameter estimates for linear and quadratic terms of the polynomial regressions were significantly different from zero (Table 2). Inclusion of a cubic term, however, did not significantly im- prove {F = 0.75, P > 0.40) the fit which was obtained using a second-degree polynomial. The second- degree polynomial yielded a higher K~ value than the nonlinear exponential model (Table 2) and hence was deemed to be the model of best fit. The ?/-intercept of this model also was not significantly different than zero (Table 2, Fig. 1) which contributes to its biolog- ical realism. Using this equation, 74% of the varia- tion in catch rate could be accounted for by substrate composition alone. (O Table 2.— Comparison of regression models. Either F-tests (b-,), f-tests (too), or asymptotic confidence intervals (exponential model) were used to test the significance of parameters. Model '1 '0 /?2 y = 0.087X - 1.496 ** * 0.64 y = 0.155{e0058X) • ps 0.68 y = 0.002X2 _ 0.050X + 0.122 *** ns 0.74 ns = nonsignificant * = P < 0.05 ** = P< 0.001 *** = P< 0.0001 8 6 x: - Y=.002X2- r2=747o 049X-^. 122 • X 4 - %/ §3 - - JC 1 • -1 it • !• 1 1 1 1 1 p°0 10 20 30 Percent Sill 40 50 6( -Clay Figure 1.- Relationship between the silt-clay fraction of the sediments and tilefish catch rates off Georgia, U.S.A. Discussion Tilefish abundance, as estimated by catch rates off Georgia's continental slope, was strongly correlated with the silt-clay fraction of the substrate. This rela- tionship was nonlinear, and based on W- values, a second-degree polynomial regression provided the best fit to the data. Off the northeastern United States, tilefish also were most abundant on fine- grain sediments (Able et al. 1982), although they were also found in horizontal burrows in the sides of submarine canyons (Warme et al. 1977), and in boulder fields (Valentine et al. 1980). Because tilefish construct vertical burrows in the substrate (Able et al. 1982), they require sediments which possess suffi- cient stability to prevent the collapse of their bur- rows. It would appear that bottom areas off Georgia which contain a sand fraction > 60% do not support substantial tilefish densities (Table 1, Fig. 1). It is likely that such substrates are not stable enough to allow tilefish to construct burrows. Thus, the ob- served correlation between catch rate and substrate composition has a biologically realistic explanation: substrates with high silt-clay fractions are conducive to the construction and maintenance of tilefish bur- rows, while substrates with high sand fractions are not. A similar explanation, based on submarine observations, has been proposed by Able et al. (1982) to explain tilefish distributions off the northeastern United States. Although we have not observed tile- fish burrows off Georgia, they have been identified in soft bottom areas off South Carolina (R. Jones'*). While the relationship between catch rates and sediment composition is quite strong, several poten- tial sources of error exist in our data. First, catch rate data were collected from two different vessels using different gear. Pooling data from the different vessels, however, would tend to obscure the relation- ship between catch rates and sediment composition. Hence, if differences in sampling methods did have an effect on our data, it would make the estimates of the catch rate-sediment relationship conservative. Second, only one substrate sample was collected with each longline set. While quantification of *R. Jones, Harbor Branch Foundation, Fort Pierce, FL 33450, pers. commun. 1983. 445 geographical variation in substrate composition was beyond the scope of this project, low relief areas off Georgia generally do not display large variations in substrate composition (Henry footnote 2). Evidence to substantiate this point is presented in Table 1, as substrate samples from areas with similar loran coor- dinates typically possessed similar substrate com- positions. Third, while a seasonal component to catch rate has been observed off New Jersey (Grimes et al. 1980), our data for this analysis do not strongly display this trend (Table 1). In addition, because the area fished did not display evidence of significant prior exploitation, our results were not affected by the potentially confounding in- fluence of commercial fishing. In conclusion, approximately three-quarters of the variation in tilefish catch rate off Georgia could be attributed to variation in a single physicochemical factor: the silt-clay fraction of the substrate. At pres- ent, however, the generality of this relationship is unknown with respect to other geographical areas or locations with different exploitation histories. In addition, temperature also has been shown to affect the distribution of tilefish off the northeastern United States (Grimes et al. 1980). It is likely that within areas possessing suitable thermal regimes, substrate composition is a major factor affecting tile- fish abundance. While a variety of associations be- tween fish abundance and physicochemical factors have previously been identified (Moyle and Cech 1982), to our knowledge, none approach the intensity of the relationship described herein. We believe that identification of this relationship will aid fisheries biologists in the identification and demographic quantification of tilefish stocks as well as the location of new fishing grounds. ACKNOWLEDGMENTS We appreciate the assistance of the captains and crews of both the Georgia Bulldog and the Delaware II, without whom this work would not have been possible. The following individuals either reviewed the manuscript or aided in other aspects of the study: K. Able, C. Barans, P. Bartlett, E. Chin, D. Beau- mariage, D. Daniels, D. Erickson, M. Freeman, S. Floyd, C. Grimes, E. Guthertz, D. Harrington, V. Henry, J. Hill, B. Low, L. Parker, L. Pittman, M. Rawson, D. Stouder, G. Ulrich, and C. Wenner. Facilities and logistic support for this study were graciously provided by the University of Georgia Marine Extension Service at Brunswick. We also wish to acknowledge the continued support of our research by the Georgia Sea Grant College Program under contract #NA80AA-D-00091. The manuscript was ably typed by Bonnie Fancher. LITERATURE CITED Able, K. W., C. B. Grimes, R. A. Cooper, and J. R. Uzmann. 1982. Burrow construction and behavior of tilefish, Lophola- tiltis chamaeleonticeps, in Hudson Submarine Canyon. En- viron. Biol. Fishes 7:199-205. Barkley, R. a., W. H. Neill, and R. M. Gooding. 1978. Skipjack tuna, Katsuwonis pelamis, habitat based on temperature and oxygen requirements. Fish. Bull., U.S. 76: 653-662. Barton, M. 1982. Comparative distribution and habitat preferences of two species of stichaeoid fishes in Yaquina Bay, Oregon. J. Exp. Mar. Biol. Ecol. 59:77-87. Day, D. S., and W. G. Pearcy. 1968. Species associations of benthic fishes on the continental shelf and slope off Oregon. J. Fish. Res. Board Can. 25: 2665-2675. Grimes, C. B., K. W. Able, and S. C. Turner. 1980. A preliminarj' analysis of the tilefish, Lopholatibus chamaeleonticeps, fisherj' in the Mid-Atlantic Bight. Mar. Fish. Rev. 42(11):13-18. Laurs, R. M., H. S. H. Yuen, and J. H. Johnson. 1977. Small-scale movements of albacore, Thunnus alalunga, in relation to ocean features as indicated by ultrasonic track- ing and oceanographic sampling. Fish. Bull., U.S. 75:347- 355. Low, R. A., Jr.. G. F. Ulrich, and F. Blum. 1983. Tilefish off South Carolina and Georgia. Mar. Fish. Rev. 45(4-6):16-26. Magnuson, J. J., L. B. Crowder, and P. A. Medvick. 1979. Temperature as an ecological resource. Am. Zool. 19: 331-343. Marliave, J. B. 1977. Substratum preferences of settling larvae of marine fishes reared in the laboratory. J. Exp. Mar. Biol. Ecol. 27: 47-60. Moyle, P. B., and J. J. Cech, Jr. 1982. Fishes: an introduction to ichthyology. Prentice Hall, Englewood Cliffs, NJ, 593 p. Powell, A. B., and F. J. Schwartz. 1977. Distribution of paralichthid flounders (Bothidae: Para- lichthys) in North Carolina estuaries. Chesapeake Sci. 18: 334-339. Radovich, J. 1961. Relationships of some marine organisms of the north- east Pacific to water temperatures. Particularly during 1957 through 1959. Calif. Dep. Fish Game, Fish. Bull. 112, 62 p. SAS Institute Inc. 1982. SAS user's guide, stat SAS Institute Inc., Gary, NC. Turner, S. C, C. B. Grimes, and K. W. Able. 1983. Growth, mortality, and age/size structure of the fish- eries for tilefish, Lapholatilits chximaeleonticeps, in the Middle-Atlantic-Southern New England region. Fish. Bull, U.S. 81:751-763. Valentine, P. C, J. R. Uzmann, and R. A. Cooper 1980. Geologj' and biology of Oceanographer Submarine Can- yon. Mar. Geol. 38:283-312. Warme, J. E., R. a. Slater, and R. A. Cooper. 1977. Bioerosion in submarine canyons, /w D. J. Stanley and 446 G. Kelling (editors), Submarine canyon, fan and trench sedi- mentation, p. 65-70. Hutchinson and Ross, Dowdon, U.K. Gary D. Grossman Michael J. Harris Joseph E. Hightower School of Forest Resources University of Georgia Athens, GA S0602 THE DEVELOPMENT AND OCCURRENCE OF LARVAE OF THE LONGFIN IRISH LORD, HEMILEPIDOTUS ZAPUS (COTTIDAE). The subfamily Hemilepidotinae, endemic to the North Pacific Ocean, is one of the more generalized subfamilies within the Cottidae (Peden 1978). According to Peden (1978), the subfamily is compos- ed of three subgenera: 1) Calycilepidottcs which in- cludes Hemilepidottcs spinostts; 2) Hemilepidotus which includes H. hemilepidotus, H. jordani, H. zapus, and H. gilberti; and 3) Melletes which includes H. papilio. The early life histories of most species are inadequately known and separation of larvae in mix- ed samples is difficult. Gorbunova (1964) described a number of larval series which she labeled H. hemilepidotus, H. gilberti, H. gilberti zapus, H. jor- dani, and H. papilio,^ but these descriptions are in- complete as well as incorrect for several species (Peden 1978; Richardson and Washington 1980). Hattori (1964) described a series of//, gilberti larvae (7.1-32.5 mm), and Peden (1978) illustrated postlar- vae (> 18 mm) of //. hemilepidotus, H. spinosus, H. zapras, and H. jordani. Richardson and Washington (1980) described larvae of H. hemilepidotus and H. spinosus. We here provide the first complete descrip- tion of//, zapus larvae, and include comments on lar- val occurrence in the eastern Bering Sea. This work supplements and clarifies the work of previous researchers by providing diagnostic characters useful in distinguishing the species. MATERIALS AND METHODS Specimens of//, zapus used in this study were col- lected during ichthyoplankton surveys conducted in the Bering Sea by the Northwest and Alaska Fisher- ies Center between 1977 and 1980. Morphological measurements were made on 57 unstained larvae (6.7-22.0 mm SL) following Matarese et al. (1981), except depth at caudal peduncle which was measured at the point of least depth. Measurements were grouped by 1 mm SL intervals, and the means of the measurements within each interval were plotted as percentage of the mean of standard lengths or head lengths within the interval (Kendall and Vinter 1984). A computer-generated best nonparametric curve, based on all data points, was drawn to illus- trate relative growth trends. Counts of meristic structures were made on 18 specimens differentially stained according to Dingerkus and Uhler (1977) following procedures outlined in Matarese et al. (1981). Terminology of head spination generally follows Richardson and Laroche (1979) and Richard- son and Washington (1980). Illustrations were made by the junior author with a camera lucida, and all specimens were preserved in either 3% Forma- lin^ buffered with sodium borate or 100% glycer- in. Identification of Hemilepidotus zapus We have routinely collected three types of Hemi- lepidotinae larvae during ichthyoplankton surveys in the eastern Bering Sea (1977-80). According to Peden (1978), four species of adults occur in this area: Hemilepidotus hemilepidotus, H. jordani, H. zapus, and H. papilio. Although preflexion larvae of H. hemilepidotus and H. jordani cannot presently be separated, we can separate the two species at noto- chord flexion according to differences in external pigment along the posterior body. Hemilepidotus hemilepidotus larvae develop pigment above the notochord along the posterior body earlier and in greater density than larvae of//, jordani (Fig. lA, B). Initially, the third series of larvae (< 17.0 mm SL) was misidentified as //. papilio (see Waldron and Vinter^) based on the presence of urostyle pigment (Gorbunova 1964). With the acquisition of larger specimens, > 17.0 mm SL, the complete series was later identified as H. zapus based on a set of charac- ters taken in part from Peden (1978) (Table 1). Gor- bunova's (1964) specimen attributed to H. zapus lacks pigment on the urostyle; of her two illustra- tions of//, papilio (footnote 1) only the 10.7 mm SL ^Hemilepidotus papilio (= Melletes papilio from Gorbunova (1964)). ^References to trade names do not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. ^Waldron, K. D., and B. M. Vinter. 1978. Ichthyoplankton of the eastern Bering Sea. Unpubl. manuscr., 77 p. Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake Boulevard E., Seattle, WA 98112. FISHERY BULLETIN: VOL. 83, NO. 3, 1985 447 External pigment nternal pigment B External pigment nternal pigment External pigment nternal pigment 448 Figure 1.- Postanal pigment patterns in //emitepirfotw-s larvae: A) H. jordani, 13.7 mmSL; B)H. hemilepidotus, 12.7 mm SL; C) H. zapus, 12.6 mm SL. Table 1. — Comparison of some important differentiating characters in Hemilepidotus zapus and H. papilio. This Characters study H. zapus^ H. papilio^ Dorsal spines X-X1 XI (X1-X11) XII (X1-X111) Pectoral fin rays 16-17 16(15-17) 17-18 (16-18) Total soft fin rays; dorsal, anal, and both pectoral fins 71 67-76 69-74 Lateral line pores 54-56 52 (47-58) 59 (49-65) Number of vertebrae 37-38 37-38 40 Number of horizontal scale rows in ventral band ca. 8 8 or 9 ca. 4 Dorsal fin notch between third and fourth spine yes yes no 'Data are from Peden (1978); mean is followed by range in parentheses. larvae \sH. zajms, while the 13.7 mm SL larva lacks urostyle pigment and is probably H. jordani Early larvae of H. zajms (6.5-17.0 mm SL) were linked together by the presence of melanophores above and below the urostyle (Figs. IC, 2); such melanophores are lacking in all other known Hemilepidotinae larvae. Larvae undergoing noto- chord flexion can be distinguished from H. hemilepi- dotus by the lack of external pigment along the posterior body and from H. jordani by the presence of ventral midline pigment which curves up toward the urostyle (Fig. 1). After notochord flexion and through the juvenile period, counts of meristic structures as well as a com- bination of adult characters (Peden 1978) will allow separation of the three species. Postflexion larvae of H. zapus have scales on the caudal peduncle that will distinguish them from other, similar-sized Hemilepi- dotinae larvae. These larvae also have a characteris- tic notch in the first dorsal fin, between the third and fourth spine, that is present in adults of all Hemi- lepidotus except H. papilio (Fig. 2F). A summary of some diagnostic features of all known Hemilepidotus larvae is presented in Table 2. Larvae and juveniles of H. papilio remain unknown. General Trends of Development Pigmentation (Fig. 2) In the smallest larvae (6-7 mm SL), pigment ap- pears on the head dorsally over the midbrain and on the anterior forebrain. In larger larvae 7-9 mm SL, additional pigment appears at the base of the hind- brain, posterior to the eye and in the opercular area. In postflexion larvae, head pigment increases. Separate pigment patches appear posterior to the eye (usually about 5 or 6 spots) and on the operculum dorsoposterior to the preopercular bone. Larvae 6-7 mm SL have pigment on the nape and on the dorsal surface of the gut. Gut pigment increases laterally with development, and in larger postflexion larvae it becomes more internal than external. By 14-15 mm SL, nape pigment extends ventrally to the dorsal surface of the gut. There are five general areas of pigmentation in the postanal region: 1) an external row (appearing more or less double) of melanophores along the dor- sal midline extending from the nape to the last myomere; 2) a dorsolateral row of internal pigment along the notochord, extending from the nape to about the last 4-7 myomeres; 3) an external row of melanophores along the ventral body midline from midbody (about 1 1 myomeres after anus) to the last myomere; 4) a ventrolateral row of internal pigment along the notochord, beginning at about 4-6 myo- meres after the anus and extending to about 6 or 7 myomeres from the end of the tail; and 5) a few ex- ternal melanophores along the notochord in the caudal peduncle area, and external melanophores dorsal and ventral to the notochord at the posterior tail tip. Prior to notochord flexion, at about 9.0 mm SL, the anterior ventral midline pigment gradually becomes more internal. In postflexion larvae, this ventral midline row is comprised of < 10 melano- phores beginning about 17 myomeres posterior to the anus. By 16.7 mm SL, all the postanal pigment is internal except for the dorsal midline melanophores and a few spots in the caudal peduncle area. After about 17 mm SL, melanophores in the caudal pedun- cle are no longer visible. Morphology (Table 3; Fig. 3) Relative growth trends are summarized in Figure 3. Preanal length, head length, depth at pectoral fin, snout to anal fin length, and snout length increase with development. Eye diameter as a proportion of head length undergoes a gradual decrease with development. Depth at the caudal peduncle and the length from the snout to dorsal fin origin increase sharply with development in larvae about 16.0-19.0 mm SL and then decrease in larger specimens. Meristic Structures (Tables 4, 5) Branchiostegal rays have begun to ossify in our smallest specimens (7-8 mm SL), and the adult com- plement of six rays is completely ossified in larvae ^ 12-13 mm SL. 449 Figure 2. -Larvae of Hemilepidotus zapus: A) 6.7 mm SL; B) 8.7 mm SL; C) 11.0 mm SL; Fins -All fin rays show their initial ossification in larvae between 12 and 13 mm SL. Fin formation oc- curs in the sequence: dorsal spines and fin rays, anal fin rays, and principal caudal fin rays (12-13 mm SL); pectoral fin rays (13-14 mm SL); pelvic spine and fin rays (15-16 mm SL); and secondary caudal fin rays (16-17 mm SL). The pterygiophores supporting dor- sal fin rays begin ossifying at 16-17 mm SL, and those supporting anal fin rays begin ossifying at 17-18 mm SL. The largest specimen (20.0 mm SL) has completely ossified dorsal and anal pterygio- phores. Axial skeleton - Neural and haemal spines have started ossification in the smallest larvae 7-8 mm SL, and are fully ossified in larvae 15-16 mm SL. Abdom- inal vertebral centra are completely ossified in larvae 12-13 mm SL, but the caudal vertebral centra are not fully ossified until larvae are slightly larger at about 14-15 mm SL. Lateral line scales do not begin ossifying until larvae are 18.0 mm SL, and our largest specimen (20.0 mm SL) has a fully ossified complement of scales. Spination (Table 5) -The development of head spines is summarized in Table 5. The parietal and nuchal spines fuse in larvae > 13 mm SL and appear as a single spine in the larger larvae 18-20 mm SL. A postocular spine is ossified in larvae 12-13 mm SL but disappears by 18-20 mm SL. A small spine below the eye (infraorbital) ossifies by 14-15 mm SL, but is no longer visible in specimens 20 mm SL. 450 Figure 2.-Continued-D) 13.0 mm SL; E) 16.7 mm SL; F) 22.8 mm SL. Occurrence of Hemilepidotus zapus in the Eastern Bering Sea During 8 yr of sampling ichthyoplankton at a total of 250 stations in the eastern Bering Sea (in the ap- proximate area between lat. 60 °N and long. 174°W with more intensive sampling between the Pribilof Islands and Unimak Island), only 58 Hemilepidotus zajMS larvae were collected. The number of positive stations is indicated in Figure 4. Small H. zapus larvae (6.7-10.1 mm SL) were col- lected during winter in surface water north of the Pribilof Islands (Fig. 4 A). Only a single larva (11.0 mm SL) was taken in February 1978 at the surface in about the same area over the slope (Fig. 4A). Our largest collection of larvae (12.2-16.7 mm SL) was made in late April and early May (Fig. 4B). Most of the larvae were collected in neuston nets at stations over the slope, but in May a few larvae were taken over the continental shelf. Larvae primarily occurred south of the Pribilof Islands. The largest larvae (18-22 mm SL) were collected during June and July at scattered stations throughout the area including only one northwest of the Pribilof Islands (Fig. 4C). Although data are insufficient to document the exact time and geographical extent of spawning, the presence of small larvae indicates some spawning oc- curs in early winter north of the Pribilof Islands. Peden (1978) indicated that adult and postlarval H. zapus have been collected only along the Aleutian Islands. Further studies are needed to investigate whether larvae and juveniles move south to the Aleu- 451 <0 3 O s. d) a: c o T3 C CO c o Q. c 3 C3) C C 3 (U o (0 c o c o cu CO I m < c c o o cu o c o CO c o CO .E -o X3 ■D 3 •D jc "D ^ . 3 3 3 CO O CO O ""^ *-* o CO "D CO CX3 CO ■o CO 00 <" to (0 CO C CD §2^ c CO > 05 CO CO CO r: ^ !c CO o CO o c I' Oi 0° T3 O CO 0) O - 5.E ^E o 2 > i5 O CO ^ c c CB E C3) CD s: 5 < o.E c ■ — ■ O CD o.E c ^^ O CO — CD CO c CO — Q E —'—CD CD 05 C k_ ^_ — J5 §E CO 3 o CO 2 E E « C ^ C Q. E o lis o . c OJ JO OJ CD E E o^ E A ' D3 i- - E E 05 CD O > O CD >^ , 00 cE CD CO O - CD) E E E E 5> CO O ■> O CO >^ CD CD CQ JD ^ CO 3 "D O CO 3 (D CO E -^ o >> CD) O !r CO ° c cz Q. E .E CO CO o C Q. CO CC — " « c 2 O CO CD CD ^ "O E O CD c O o CO CO CI CD 3 CO Q) CO CD .E E ^ D) 9 CD in CD ^ ^ T— CD £ CO V t CD E 2 T- (D ■^ 13 CD E CIi CO o CO 1- C CD C3) CO CO ^ <'i O n I" <= 0) >, CO c ^ >- ° ^ CD CO O h (D E 2 T- - TD >-? CD E ■^ "O •D E CD h. E C T CD -"J- S Q) o >, O) CD c j: ?i '^ - O -= C CO .E o ffl E o co a: 5; u CO ^^ ■ ^ 3 E CO o :3 CD ;o E -C Q. CO i -S CD C o O CO *- CJ) CD ■^ X T3 0) CD CD C CO CD E O) 0) >> CO o cd' . i5 E - CD ^ i= (3) ™ lo E o 0) ■q. CO CO = ^ E c a> o ~ " o> 3 CD -D E i5 'q. C CO c "^ ID > c O O CO O O J2 CD O CO JD 3 o CO CO c= 0) 3 jD 5 E • o •= c CO .E o CD ^E 00 O CD E .^ CO CO ^ 3 C Q) c O) W) CO CD O o ^\ /-\ ^-' O) c ^ >. o) o o CD CD CO CD C o z. o E 3 ™ -c -c CO . 3 E CO 3: S > CO CD CO c CD CO C 3 — CD JD — -^ "-1 CO CO 3: ■£ CD CO 3 CL CO O to CO C - 3 CO CO c . •.- ^ £ « V CD c o CD c o z 0 c o c o CO c CO CT OJ CD J2 5 £ CO , . 3 C o C CD >> CO E E B Q- 00 CD I. ■Q T3 CD CO Q. CD CO CD n O c c (0 tJ c P u 0 K UJifr a. '^i 4 « \1 t'e 20 3&I -32 -'28- o < Q24 z < I— 20 :i6- u a: UJ12 a. DEPTH AT PECTORAL 12 16 20 64 t-60^ -■56 o < <=>52 z < I— en 48 :44 o iu40 0. 36: SNOUT TO ANAL FIN ? i il TS 20 STANDARD LEnCTH (MM) Figure 3. -Relative trends in selected morphometric features Table 4. — Development of meristic characters in larvae of Hemilepidotus Sample size Fin rays Size interval (mm SL) Caudal Dorsal Anal Pectoral Pelvic P rincipal Secondary Total 7-8 2 8-9 3 9-10 2 MO-12 — 12-13 3 X,11-21 17 8-10 1,0 12 3-5 15-17 13-14 1 Xl,21 17 16 1,3 12 9 21 14-15 3 X-XI-20-21 17 13-16 1,3 12 9 21 15-16 2 X,21 17 16 1.4 12 11-13 23-25 16-17 1 Xl,20 17 17 1,4 12 15 27 18 1 XI,2G 16 16 1,4 12 15 27 20 1 Xl,21 17 16 1,4 12 14 26 'Specimens in this size group did not accept alizarin stain. ^Haemal spines 23-24 are fused. 454 29.6 ■ a X SNOUT TO DORSAL •— FIN /\ z28.8 / N. UJ 1 \ _i n 1 \ " Q28.0 1 \ q: j \ < 1 \ o j ^27.2 / >— / a trt / u.26.4 1 o \/ ,_ \/ Z25.6 LLl O < larvae in the eastern Bering Sea: A) January-March (IMF80 and MF7801); B) April-May (MF77B5(6)); C) June-July (2MF79). Closed circles in- dicate neuston stations and closed diamonds indicate bongo stations. Station data on file at the Northwest and Alaska Fisheries Center, Seattle, WA. APRIL-MAY 12.2-16.7 nnmSL (n=29) JUNE-JULY 18.0-22.0 mm SL {n=8) 62 OON 60 00 - 58 00 56 00 54 00 - 52 00 160 00 180 00 62 OON 60 00 58 00 56 00 - 54 00 - 52 00 160 00 456 tians or whether adult H. zapus range further north. Acknowledgments We thank Jean R. Dunn and Arthur W. Kendall, Jr. (Northwest and Alaska Fisheries Center) for helpful discussions and review of the manuscript. Alex E. Peden (British Columbia Provincial Museum) kindly assisted in the identification of Hemilepidotus zaipus postflexion larvae. Literature Cited DiNGERKUS, G., AND L. D. UHLER. 1977. Enzyme clearing of alcian blue stained whole small vertebrates for demonstration of cartilage. Stain Technol. 52:229-232. GORBUNOVA, N. N. 1964. Razmnozhenie i razvitie polucheshuinykh bychkov (Cottidae, Pisces) (Breeding and development of hemilepido- tine sculpins (Cottidae, Pisces)). [In. Russ.) Tr. Inst. OkeanoL, Akad. Nauk SSSR 73:235-251. (Transl. by Isr. Program Sci. Transl., 1966; in T. S. Rass (editor), Fishes of the Pacific and Indian Oceans, biology and distribution, p. 249-266; available U.S. Dep. Commer., Natl. Tech. Inf. Serv., Springfield, Va., as TT 65-50120.) Hattori, S. 1964. Studies on fish larvae in the Kuroshio and adjacent waters. [In Jpn., Engl, synop.) Bull. Tokai Reg. Fish. Res. Lab. 40, 158 p. Kendall, A. W., Jr., and B. Vinter. 1984. Development of hexagrammids (Pisces: Scorpaeni- formes) in the Northeastern Pacific Ocean. U.S. Dep. Commer., NOAA Tech. Rep. NMFS 2, 44 p. Matarese, a. C, S. L. Richardson, and J. R. Dunn. 1981. Larval development of the Pacific tomcod, Microgadus proximus, in the Northeast Pacific Ocean with comparative notes on larvae of walleye pollock, Theragra chalcogramma, and Pacific cod, Gadus macrocepkalus (Gadidae). Fish. Bull. , U.S. 78:923-940. Peden, A. E. 1978. A systematic revision of the hemilepidotine fishes (Cottidae). Syesis 11:11-49. Richardson, S. L., and W. A. Laroche. 1979. Development and occurrence of larvae and juveniles of the rockfishes Sehastes crameri, Sebastes pinniger, and Sebastes helvamaculatus (family Scorpaenidae) off Oregon. Fish. Bull., U.S. 77:1-46. Richardson, S. L., and B. B. Washington. 1980. Guide to identification of some sculpin (Cottidae) larvae from marine and brackish waters off Oregon and adjacent areas in the northeast Pacific. U.S. Dep. Commer., NOAA Tech. Rep. NMFS Circ-430, 56 p. Ann C. Matarese Beverly M. Vinter NOAA, Natwnal Marine Fisheries Service Northwest and A laska Fisheries Center 2725 Montlake BouLeniard East Seattle, WA 98112-2097 AN APPROACH TO ESTIMATING AN ECOSYSTEM BOX MODEL Recent trends in ecosystem modeling have produced complex simulation models which are very data in- tensive (Andersen and Ursin 1977; Laevastu and Larkins 1981). However, in many situations the con- struction of a biomass budget for a box model of an ecosystem is relatively simple and can provide impor- tant information about the ecosystem standing stock and energy flow (Walsh 1981; Pauly 1982; Polovina 1984). The ECOPATH model is an analytical procedure to estimate a biomass budget for a box model of an ecosystem given inputs which specify the com- ponents of the ecosystem, together with their mor- tality, diet, and energetics value. A computer pro- gram for ECOPATH has been written in BASIC-80, version 5.21, by Microsofti (CP/M version). A listing of the ECOPATH computer program and a user's manual are available from the author. The ECOPATH model produces estimates of mean annual biomass, annual biomass production, and an- nual biomass consumption for each of the user specified species-groups. The species-groups repre- sent aggregations of species with similar diet and life history characteristics and which have a common physical habitat. The ECOPATH model is not a simulation model with a time component as are some more complex ecosystem models. It estimates a biomass budget for the marine ecosystem in a static situation under the assumption that the ecosystem is at equilibrium conditions. Equilibrium conditions are defined to exist when the mean annual biomass for each species-group does not change from year to year. This condition results in a system of biomass budget equations which, for species-group i, can be expressed as Production of biomass for species i - all predation on species i - nonpredatory biomass mortality for species i - fishery catch for species i = 0 for all i. (1) The ECOPATH model expresses each term in the budget equation as a linear function of the unknown mean annual biomasses {B,'s) so the resulting biomass budget equations become a system of simul- taneous equations linear in the B-s. The mean annual biomass estimates are obtained by solving the system of simultaneous linear equations. 'Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 457 The formulation of each term of the biomass budget equation will now be presented in detail. The Model Biomass Production Production (P) for a cohort of animals over 1 yr is defined as 1 J F = f N, — (w,) dt 0 dt and mean annual biomass (B) for the cohort is de- fined as B = J NtWfdt where iV, is the number of animals and w, the mean individual weight at time t. Allen (1971) investigated the production to bio- mass (PIB) ratio for a cohort over a range of mortal- ity and growth functions. For a number of growth and mortality functions, including negative exponen- tial mortality and von Bertalanffy growth, the ratio of annual production to mean biomass for a cohort is the annual instantaneous total mortality (Z,). For a species-group which consists of n cohorts or species, with instantaneous annual total mortality (Z,) for cohort or species i, where mortality is determined by a negative exponential function and growth by a von Bertalanffy growth function, the total species-group production (P) is the sum of the cohort production (P,) and can be expresed as P= I P,= 1 Z,B, (2) i:=i ( = 1 Under the assumption that the Z's are all equal to say Z, then total species-group production can be ex- pressed as P = Z ■ B where B is the mean annual species-group biomass. Allen (1971) has also shown that when growth in weight is linear, the PIB ratio is equal to the recipro- cal of the mean age for a range of mortality func- tions. For a number of other growth and mortality functions the ratio of cohort PIB can be the recipro- cal of the mean lifespan. Thus, for a range of growth and mortality functions, total species-group produc- tion can be expressed as P = C ■ B where B is the mean annual species-group biomass, and C is a parameter. In an application of ECOPATH to an ecosystem of French Frigate Shoals where there was very little fishing mortality, the PIB ratio for fishes and crusta- ceans was taken as the annual instantaneous natural mortality (M); whereas, for primary and secondary producers whose growth is more likely to be linear than the von Bertalanffy, the PIB ratio was esti- mated as the reciprocal of the mean age (Polovina 1984). Predation Mortality The predation mortality is the fraction of the biomass of a species-group which is consumed by all predators excluding fishing mortality. Two types of information are needed. First the food web or predator-prey relationships must be defined. A diet composition matrix DC,, must be specified where an entry DC,j from this matrix refers to the proportion (by weight) of prey j in the diet of predator (. The primary source of this information is the analysis of stomach contents data. At least in one study it has been shown that there is a high correlation between diet indices based on weight, volume, and percentage of occurrence for stomach content data, and thus either index may be used to generate the DC matrix (Macdonald and Green 1983). The second type of in- formation needed to ascertain predation mortality is the food requirements of the predator. The ECOPATH model requires the user to specify FR„ the ratio of annual consumption to mean annual biomass. The annual food required by the predator is the product of FP, and P,. Some values of daily food required as a fraction of body weight range from 0.005 to 0.02 (Laevastu and Larkins 1981). Based on these daily estimates a range of annual food required as a fraction of mean biomass (FPj is 1.8 to 7.3. Nonpredation Mortality All mortality attributable to causes other than predation and fishing is termed nonpredatory mor- tality. The ECOPATH model defines ecotrophic effi- ciency e, as the fraction of total production which is removed by fishing and predation mortality. This 458 was 0.95 in the French Frigate Shoals model. The nonpredator mortality rate is (1 - e,) • Z|, and the amount of production which goes to nonpredation mortality is (1 - e,) P, = (1 - e,) C, B,. For n species-groups the biomass budget Equation (1) becomes a system of n simultaneous equations as follows: C,5i - 1 {FR,)B,DC„ - (1 - e^)C,B, = catchj, k= 1 C,B, - I^ (FRk) B^ DC,, - (1 - e,) C,B, = catch. Schoals in the Northwestern Hawaiian Islands pro- vided the estimates for many of the input parameters required by the ECOPATH model as well as some estimates of biomass and production to serve to evaluate the estimates produced by the model. The estimates of biomass and production generated by the application of ECOPATH to French Frigate Shoals are given in Figure 1 . In general the model's estimates of biomass and production are in good agreement with the available field data (Polovina 1984). In the application of the French Frigate Shoals, the biomasses of the top level carnivores are treated as fixed inputs thus a particularly appropri- ate validation of the model is the comparison of the estimate of net benthic primary production with an independent estimate from field data. The model estimated net benthic primary production, adjusted to the total 1,200 km^ habitat of French Frigate Shoals, at 2.3 x 10*' kg • km-^ • yr-i while the esti- mate based on field data was 2.5 x lO'^kg • km"- ■ yr-i (Griggetal. 1984). C„B„ - I iFR,)B,DC,„ - (1 - e,;)C„B„ = catch„. *■=! With input estimates for parameters C„ FR„ DC,j, and e, for all i and j, and catches (catch,) if there is fishing, this system of equations is a system of n simultaneous equations linear in the unknown B-s. This system of equations can be expressed in matrix form as AB = C, where A is an n x n matrix of coefficients, B is an n-dimensional vector of mean an- nual species group biomass, and C is the vector of fishery catch where the ith element is the total catch of the ith species-group. If the matrix A is of full rank and if there are some fishery catches for some species so the vector C is not null, then there typically exists a unique nontrivial solution vector of biomass B. If there are no fishery catches then it is necessary to provide an estimate of at least one of the mean species group biomass 5, before there exists a unique nontrivial biomass vec- tor B which solves the budget equation. In the appli- cation of ECOPATH to an ecosystem at French Frigate Schoals where there was no fishing mortali- ty, the biomasses of three apex predators were esti- mated from field censuses and treated as known in- puts. In this application the ith element of C vector was the annual predation by the three apex predators on the ith species- group. Five years of field work targeting most of the com- ponents of the marine ecosystem at French Frigate The Computer Program The ECOPATH model has been implemented via two BASIC language programs. The "dialect" of the language used is BASIC-80, version 5.21, by Micro- soft (CP/M version). These programs are designed to be used interactively on a terminal or a hard-copy printer. The first program is the input parameter program which accepts the input parameters and formats them into a BASIC sequential file. The sec- ond program is the ECOPATH model itself. Literature Cited Allen, K. R. 1971. Relation between production and biomass. J. Fish. Res. Board Can. 28:1573-1581. Anojersen, K. p., and E. Ursin. 1977. A multispecies extension to the Beverton and Holt theory of fishing, with accounts of phosphorus circulation and primary production. Medd. Dan. Fisk. Havunders., New Sen, 7:319-435. Grigg, R. W., J. J. Polovina, and M. J. Atkinson. 1984. Model of a coral reef ecosystem: Part III Resource Limitation, Community Regulation, Fishery Yield, and Resource Management. Coral Reefs 3:23-27. Laevastu T., and H. a. Larkins. 1981. Marine fisheries ecosystem: its quantitative evaluation and management. Fishing News Books, Farnham, Surrey, Engl., 162 p. Macdonald, J. S., and R. H. Green. 1983. Redundancy of variables used to describe importance of prey species in fish diets. Can. J. Fish. Aquat. Sci. 40:635- 637. Pauly, D. 1982. Notes on tropical multispecies fisheries, with a short 459 I 57 SEABIRDS B = 15 P = 81 847 MONK SEALS SMALL PELAGIC S B = 1,836 P = 2,020 i 150 SHARKS, JACKS, SCOMBRIDS B = 536 P = 192 LOBSTERS AND CRABS B = 1,348 P = 701 B = 13,966 P = 20,949 ZOOPLANKTON B = 899 P = 3.6 X 10* t ) * 2.3X10' PHYTOPLANKTON B = 3.3 X 10^ P = 2.3 X 105 BOTTOM FISHES B = 94 P = 30 GREEN TDRTLES B = 15 P = 2 4.3 X 10* HETEROTROPHIC BENTHOS B = 1.7 X 10^ P = 5.1 X 10^ OS y g X to 2.4X10® BENTHIC ALGAE B = 2.0 X 10^ P = 2.5 X 10^ 47 Figure l.-Biomass budget schematic for major prey-predator pathways. Annual production denoted as P and mean annual biomass as B with values in units of (kg/km)'^ based on a habitat area of 1 ,200 km^. The values associated with the arrows represent the pro(iucti()n from the lower trophic level consumed by the higher trophic level (Polovina in press). bibliography of the food and feeding habits of tropical fish. In Report on the regional training course on fishery stock assessment, 1 September-9 October 1981, Samutprakarn, Thailand, p. 30-35 and 92-98. Tech. Rep. 1, Part II, SCS/ GEN/82/41, Manila. Polovina, J. J. 1984. Model of a coral reef ecosystem, Part I: ECOPATH and its application to French Frigate Shoals. Coral Reefs 3:1-11. Walsh, J. J. 1981. A carbon budget for overfishing off Peru. Nature 290: 300-304. Jeffrey J. Polovina Mark D. Ow Southwest Fisheries Center Honolulu Laboratory National Marine Fisheries Service, NOAA P.O. Box 3830 Honolulu., HI 96812 460 FOOD AND FEEDING OF THE TOMTATE, HAEMULON AUROUNEATUM (PISCES, HAEMULIDAE), IN THE SOUTH ATLANTIC BIGHT' The tomtate, Haemulon aurolineatum, is an abun- dant demersal fish in a variety of marine habitats in the South Atlantic Bight, the Gulf of Mexico, and the Caribbean Sea (Darcy 1983). They are a reef-associ- ated species (Parrish and Zimmerman 1977), and in the South Atlantic Bight they are most commonly found over hard or "live" bottom reefs in depths < 55 m (Struhsaker 1969; Manooch and Barans 1982; Sedberry and Van Dolah 1984). While occa- sionally taken in trawl catches over open, sandy habitats on the southeastern continental shelf (Wen- ner et al. 1980), they are much more abundant in trawls directed at sampling hard bottom, and generally rank in the top three demersal species by number or weight in trawl catches (Wenner 1983; Sedberry and Van Dolah 1984; Sedberry unpubl. data). Although they are frequently caught on hard bottom reefs in the South Atlantic Bight, the depen- dance of these fishes on hard bottom habitat for food is unknown. Previous investigations in the Carib- bean have indicated that tomtate are not obligatory reef dwellers and that they forage extensively in open sandy areas (see Darcy 1983 for review). Because of the importance of this species in the hard bottom ichthyofauna of the South Atlantic Bight and its importance to fisheries associated with hard bot- tom reefs, a knowledge of its food habits is important to our understanding the ecology of this habitat. Tomtate may be important in transferring energy from the expansive sand areas of the shelf onto the much more restricted hard bottom habitat, and their feeding behavior in the South Atlantic Bight may be important in maintaining the higher biological pro- ductivity of hard bottom areas, relative to the open sandy shelf. To determine foraging habitat of the tomtate, an investigation on food habits was conducted. The pur- pose of this note is to report the results of that study and to relate the feeding behavior to existing knowledge of the ecology of hard bottom areas in the South Atlantic Bight. Methods Tomtate were collected during seasonal cruises in 1980 (two cruises- one in winter and one in summer) and 1981 (four cruises- one each in winter, spring, summer, and fall) by trawl from eight hard bottom reef stations off South Carolina and Georgia. Sta- tions were located in each of three depth zones representing the inner shelf (16-22 m depth, three stations), middle shelf (23-38 m, four stations), and the outer shelf (47-67 m, one station). Detailed descriptions of station locations and habitat can be found in Sedberry and Van Dolah (1984) and Wen- ner et al. (1984). Each station was mapped using loran C and underwater television, and all sam- pling was conducted in hard bottom areas mapped by using this technique (Sedberry and Van Dolah 1984). Tomtate were measured (standard length, SL) at sea and their stomachs removed if not conspicuously empty. Stomachs were individually labeled and pre- served in 10% seawater-Formalin^. Stomachs were washed in tap water and transfer- red to 50% isopropanol in the laboratory, and con- tents of individual stomachs were sorted by taxa and counted. Colonial forms (e.g., hydroids, bryozoans) and algae were counted as one organism. Volume displacement of food items was measured using a graduated cylinder, or estimated by using a 0.1 cm^ grid (Windell 1971). Since the methods of food habits quantification are variously biased (Hynes 1950; Pinkas et al. 1971; Windell 1971), the relative contribution of different food items to the total diet was determined using three methods: 1) percent frequency occurrence (F), 2) percent numerical abundance (N), and 3) percent volume displacement (V). These three values were calculated for individual prey species, for prey grouped by higher taxonomic categories, and for higher taxonomic categories pooled for 100 mm intervals of standard length. To determine the dependance of tomtate on hard bottom prey organisms, stomach samples were compared with benthic samples using IvleVs index of electivity (Ivlev 1961), calculated as follows: E = ^1 A +^2 where P, is the percentage of the diet comprised by a given prey taxon and P2 is the percentage of the food complex in the environment (i.e., in benthic samples) comprised by the same prey taxon. Electivity values range from - 1 to -i- 1 . Negative values imply that the prey species is avoided by the predator or that it ^Contribution No. 179, South Carolina Marine Resources Research Institute, P.O. Box 12559, Charleston, SC 29412. ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 461 is unavailable to the predator. Positive values imply that the predator prefers the prey species or that it is feeding on prey species which occur in a different habitat than those sampled by the benthic sampler. A value near zero implies no selectivity by the predator; i.e., the fish is feeding on the prey in pro- portion to the prey's relative abundance. Benthic samples and stomach collections were pooled by depth zone (inner, middle, and outer shelf) for comparison; however, too few tomtate for ade- quate comparison were collected at outer shelf sta- tions. Benthic samples were obtained with diver- operated suction sampler at the seven inner and mid- dle shelf, hard bottom sites during the same time periods in 1980 and 1981 as the fish collections were made. The suction sampler is very effective at sam- pling macroinvertebrates on hard substrates (Chess 1979; Wenner et al. 1983). Five replicate benthic samples were taken during the six cruises at each reef that was sampled for fishes, and these samples (30 for each reef) are believed to be adquate repre- sentatives of the hard bottom invertebrate fauna in Table 1.— Percent frequency occurrence (F), percent number (N), and percent volume (V) of food items in Haemulon aurolinea- tum stomachs collected at tiard bottom areas in 1980 and 1981. Taxon Food item F N V Taxon Food item F N V Algae Polychaeta undetermined 5.3 0.4 3.1 Sargassum sp. 1.0 0.1 0.1 Progoniada regularis 1.0 0.1 <0.1 Cnidaria Psalmmolyce ctenidophora 2.1 0.2 1.5 Hydrozoa Sabellidae undetermined 1.0 0.1 0.2 Dynamena cornlcina 1.0 0.1 <0.1 Scoloplos rubra 1.0 0.1 0.2 LIctorella convallaria 1.0 0.1 <0.1 Sptiaerodoridae Sertularia sp. 1.0 0.1 <0.1 undetermined 1.0 0.1 <0.1 Total Hydrozoa 3.2 0.2 <0.1 Syllidae undetermined 2.1 0.2 <0.1 Anthozoa Syllis sp. 3.2 0.2 0.1 Actiniaria undetermined 5.3 0.4 0.6 Syllis regulata carolinae 1.0 0.1 <0.1 Platytielminthes Terebellidae undetermined 2.1 0.2 2.0 Turbellaria undetermined 5.3 1.0 0.7 Travisia parva 1.1 0.2 0.1 Annelida Total Polychaeta 46.3 8.7 14.6 Polyctiaeta Mollusca Ampharete sp. 1.0 0.1 <0.1 Gastropoda Amphinomidae undetermined 1.0 0.1 <0.1 Caecum pulchellum 1.0 0.1 <0.1 Arabella iricolor 2.1 0.2 0.5 Diodora cayenensis 1.0 0.1 <0.1 Arabellidae undetermined 1.0 0.1 0.1 Gastropoda undetermined 1.0 0.1 0.4 Armandia maculata 6.3 2.4 0.3 Naticidae undetermined 1.0 0.1 <0.1 Capitellidae undetermined 5.3 0.4 0.4 Total Gastropoda 3.2 0.3 0.5 Ceratonereis mirabilis 1.0 0.1 <0.1 Pelecypoda Chloeia sp. 1.0 0.1 0.2 Ervilia concentrica 5.3 38.2 6.3 Chloela vlridis 1.0 0.1 0.4 Mactra fragilis 1.0 0.1 0.2 Chone americana 2.1 0.2 <0.1 Pelecypoda larvae 1.0 0.1 "^O.l Diopatra cuprea 1.0 0.1 <0.1 Total Pelecypoda 7.4 38.3 6.5 Drilonereis sp. 2.1 0.2 0.1 Cephalopoda Eunice vittata 1.0 0.1 <0.1 Octopus sp. 1.0 0.1 0.9 Eunice websteri 1.0 0.1 <0.1 Pycnogonida Eunicidae undetermined 1.0 0.1 <0.1 Anoplodactytus insigniformis 1.0 0.1 <0.1 Exogone dispar 2.1 0.2 <0.1 Crustacea Glycera sp. 5.3 0.4 1.2 Copepoda Glycera americana 3.2 0.2 0.8 Catanopia americana 5.3 13.0 0.3 Glycera tesselata 2.1 0.2 0.3 Longipedia helgolandica 2.1 0.6 <0.1 Goniadides carolinae 1.0 0.1 <0.1 Microsetella norvegica 1.0 0.2 <0.1 Harmothoe sp. 2.1 0.2 <0.1 Saphirella tropica 1.0 0.1 <0.1 Lumbrinerides acuta 1.0 0.1 <0.1 Temora stytifera 3.2 0.4 <0.1 Lumbrineris coccinea 2.1 0.2 0.9 Temora turbinata 6.3 1.2 <0.1 Lumbrineris sp. 1.0 0.1 0.9 Undinula vulgaris 1.0 0.1 <0.1 Malanidae undetermined 3.2 0.5 0.2 Total Copepoda 13.7 15.4 0.4 Nephtyidae undetermined 1.0 0.1 <0.1 Stomatopoda Nephtys incisa 1.0 0.1 <0.1 Gonodactylus bredini 2.1 0.2 1.4 Notomastus americanus 1.0 0.1 <0.1 Lysiosquilla scabricauda 1.0 0.1 1.7 Notopygos crinita 1.0 0.1 0.1 Stomatopoda larvae 3.2 0.3 0.3 Onuphis sp. 1.0 0.1 <0.1 Stomatopoda undetermined 4.2 0.3 0.4 Onuphis eremita 1.0 0.1 <0.1 Total Stomatopoda 10.5 0.8 3.8 Onuphis nebulosa 1.0 0.1 0.4 Mysidacea Opfieliidae undetermined 1.0 0.1 <0.1 Bowmaniella portoricensis 3.2 0.2 0.1 Phyllodoce castanea 1.0 0.1 <0.1 Cumacea Phyllodoce groenlandica 1.0 0.1 0.1 Cumacea B 1.0 0.1 <0.1 Phyllodoce longipes 1.0 0.2 <0.1 Oxyurostylis smithi 3.2 0.2 <0.1 Phyllodoce sp. 1.0 0.1 <0.1 Total Cumacea 4.2 0.3 <0.1 Phyllodocidae undetermined 1.0 0.1 <0.1 462 each depth zone (Wenner et al. 1983, 1984). Details of benthic sampling and structure of the invertebrate communities are described elsewhere (Wenner et al. 1983, 1984). The electivity index was calculated for each species that were numerically dominant in fish stomachs or in benthic samples collected within the two depth zones (inner and middle shelf). Results and Discussion Haemulon aurolineatum had a generalized diet and fed on about 120 species of prey (Table 1). Polychaetes and amphipods were the most important food and were eaten with almost the same fre- quency. Polychaetes, however, made up a large volume of prey because of their large size. Decapods were also frequently consumed, but made up a small percentage of the volume or number of prey items. Pelecypods were the most abundant prey and cepha- lochordates, while infrequently consumed, made up a large portion of food volume because of their large size. Fishes also made up a large portion of food volume and copepods, though small in volume dis- placement, were often eaten in large numbers. Table 1.— Continued. Taxon Food item F N V Taxon Food item F N V Tanaidacea Decapoda larvae 1.0 0.1 <0.1 Apseudes sp. B 1.0 0.7 <0.1 Leptochela sp. 1.0 0.1 0.4 Isopoda Leptochela papulata 7.4 0.7 1.7 Carpias bermudensis 1.0 0.1 <0.1 Lucifer faxoni 3.2 0.4 <0.1 Erichsonella flliformis 1.0 0.1 <0.1 Lysmata sp. 1.0 0.2 <0.1 Eurydice littoralis 3.2 0.2 0.1 Natantia undetermined 4.2 0.4 0.6 Paracerceis caudata 1.0 0.1 <0.1 Neopontonldes beaufortensis 1.0 0.1 <0.1 Total Isopoda 6.3 0.4 0.1 Paguridae 1.0 0.1 <0.1 Amphiipoda Periclimenaeus schmitti 1.0 0.1 <0.1 Acanttionotozomatidae 1.0 0.1 <0.1 Periclimenes sp. 1.0 0.1 0.1 Ampelisca sp. 1.0 0.1 <0.1 Perlcllmenes longicaudatus 2.1 0.2 0.1 Ampelisca cristoldes 1.0 0.1 0.1 Processa sp. 4.2 0.3 0.9 Ampelisca schellenbergi 3.2 0.3 0.1 Processa hemphilli 1.0 0.1 0.2 Ampelisca vadorum 1.0 0.1 0.1 Synalpheus minus 1.0 0.1 0.1 Amphipoda E 3.1 0.2 <0.1 Synalpheus townsendi 1.0 0.1 <0.1 Amphipoda G 1.0 0.2 <0.1 Thor sp. 1.0 0.1 0.2 Amphipoda undetermined 2.1 0.2 0.1 Thor floridanus 1.0 0.1 <0.1 Caprella equlllbra 13.7 1.7 0.2 Trachypenaeus constrictus 2.1 0.2 0.1 Caprella penantis 3.2 0.4 0.1 Xanthidae 1.0 0.1 0.2 Cerapus tubularis 1.0 0.1 <0.1 Total Decapoda 33.7 4.9 6.2 Elasmopus sp. A 2.1 0.4 <0.1 Sipunculida Elasmopus sp. 1.0 0.1 <0.1 Sipunculida D 1.0 0.1 <0.1 Erichthonlus brasillensis 12.6 2.1 0.2 Bryozoa Gammaropsls sp. 2.1 0.4 <0.1 Amathia distans 1.0 0.1 <0.1 Lembos unicornis 1.0 0.1 <0.1 Crisia sp. 1.0 0.1 <0.1 Leucothoe splnlcarpa 1.0 0.1 <0.1 Diaperoecia floridana 2.1 0.2 <0.1 Liljeborgia sp. A 2.1 0.2 <0.1 Discoporella umbellata 1.0 0.1 <0.1 Luconacia incerta 2.1 1.5 0.1 Total Bryozoa 5.3 0.4 0.1 Lysianopsis alba 4.2 0.7 0.1 Echinodermata Melita appendiculata 2.1 1.0 0.1 Echinoidea Metharpinla floridana 1.0 0.2 <0.1 Clypeasteroidea Microjassa sp. A 1.0 0.2 <0.1 undetermined 1.0 0.7 <0.1 Monoculodes sp. 1.0 0.1 <0.1 Ophiuroidea Photis sp. 3.1 0.4 <0.1 Hemiptiolis elongata 1.0 0.1 <0.1 Photis pugnator 1.0 0.2 <0.1 Ophiotfirix angulata 3.2 0.4 0.4 Phtlsica marina 2.1 0.3 <0.1 Ophiuroidea undetermined 17.9 1.5 1.5 Rhepoxynius epistomus 2.1 0.2 <0.1 Total Ophiuroidea 22.1 1.9 1.9 Rudilemboides naglei 6.3 0.7 <0.1 Chaetognatha Stenopleustes sp. A 1.0 0.1 <0.1 Chaetognatha undetermined 1.0 0.1 <0.1 Stenothoe sp. 3.2 0.2 <0.1 Chordata Stenothoe georgiana 5.3 0.8 <0.1 Cephalochordata Synchelldium americanum 1.0 0.1 <0.1 Branchiostoma carlbaeum 4.2 12.4 41.6 Tiron tropakis 1.0 0.1 <0.1 Pisces Total Amphipoda 47.4 12.8 1.3 Bothidae undetermined 1.0 0.1 0.2 Decapoda Clupeidae undetermined 1.0 0.1 1.0 Albunea paretii zoea 1.0 0.1 <0.1 Hypleurochilus geminatus 1.0 0.1 0.4 Alpheus normani 1.0 0.1 <0.1 Teleostei larvae 1.0 0.1 <0.1 Brachyura megalopae 1.0 1.2 0.1 Teleostei undetermined 10.5 0.7 18.8 Brachyura undetermined 2.1 0.2 0.3 Total Pisces 14.7 1.0 20.5 Callianassa atlantica 1.0 0.1 0.9 Number of stomachs examined: Examined stomachs with food: 154 95 463 Ophiuroids were frequently consumed but were usually represented in stomachs by small arm fragments. Small (1-100 mm SL) tomtate had a diet domi- nated numerically by very small crustaceans (cope- pods) and volumetrically by fishes and decapods (Table 2). Amphipods were most abundant prey tax- on for 101-150 mm tomtate and polychaetes made up the greatest volume of food. Large (151-200 mm SL) tomtate primarily consumed pelecypods, which were the most abundant taxon, and cephalochordates, which were abundant in the diet and made up the greatest prey volume. Many hard bottom invertebrates that were abun- dant in suction samples at inner and middle shelf sites were not important in the diet of tomtate (Table 3). Of the eight dominant hard bottom invertebrate species, only two (the polychaete Chone americana and the corophoid amphipod Erichthonms brasili- ensis) at inner shelf sites and one (the caprellid amphipod Luconacia incerta) at middle shelf sites made up a greater percentage of the diet than they did of benthic samples. On the other hand, inverte- brates that were common in stomachs were general- ly not abundant in benthic samples and electivity values were usually positive. Tomtate are apparently not completely dependent on hard bottom habitat for prey. Some of the most abundant prey species are pelagic (e.g., brachyuran megalopae, copepods). Most benthic prey are in- faunal species that are restricted to soft sediments. Amnandia maculata, a dominant prey species on the inner shelf, is a deposit-feeding polychaete that bur- rows in soft sediments (Fauchald and Jumars 1979). Ervilia concentrica, an important prey species on the middle shelf, was not collected at any of the 1 1 hard bottom stations. This bivalve is common in soft sediments (Porter 1974). The cephalochordate 5rar^ chiostoma caribaeum, a common prey species on the middle shelf that was very rare in benthic samples, is also an infaunal sand bottom species (Hildebrand and Schroeder 1928). Thus, a large portion of the prey of Haemulon aurolineatum are not hard bottom epi- faunal species, suggesting that tomtate are not Table 2. — Percent frequency occurrence (F), percent number (N), and percent volume (V) of higher tax- onomic groups of food in the diet of Haemulon aurolineatum, by length interval. Length Intervals (mm SL) Prey taxon 1-100 101-150 151-200 F N V F N V F N V 4.8 0.4 2.6 6.2 0.3 <0.1 8.3 1.1 3.0 6.3 0.3 0.4 12.5 4.0 6.3 4.2 0.7 0.1 19.0 1.5 11.4 62.5 15.6 35.9 50.0 9.4 13.5 8.3 1.1 4.7 2.1 0.2 <0.1 4.1 0.6 <0.1 12.5 2.1 57.2 0.1 8.0 1.1 4.2 0.6 0.1 47.6 77.1 14.4 4.2 0.6 <0.1 4.2 0.3 <0.1 19.0 1.9 14.7 12.5 1.7 2.2 6.2 6.2 0.3 0.3 4.0 0.1 4.8 0.4 0.2 4.2 0.6 <0.1 6.2 0.3 <0.1 12.5 1.7 0.1 6.2 0.3 0.2 33.3 6.0 7.3 79.2 58.3 7.7 39.6 5.9 0.5 19.0 9.8 20.8 33.3 5.6 11.3 39.6 3.2 5.6 4.8 0.4 0.7 8.3 1.1 0.4 6.2 2.1 0.3 0.1 <0.1 <0.1 14.3 1.1 1.7 25.0 5.0 8.0 22.9 1.4 1.0 4.8 0.4 0.4 4.2 0.6 24 6.2 18.4 51.1 4.3 1.1 28 21 71.5 49-99 26.0 16.7 2.2 44 24 138.2 101-150 17.8 12.5 0.7 79 48 168.3 151-198 14.1 Algae Cnidaria Hydrozoa Anthozoa Turbellaria Annelida [ylollusca Gastropoda Pelecypoda Cephalopoda Arthropoda Pycnogonida Copepoda Stomatopoda Mysidacea Cumacea Tanaidacea Isopoda Amphipoda Decapoda Sipunculida Bryozoa Echinodermata Echinoidea Ophiuroidea Chaetognatha Chordata Cephalochordata Pisces Number of stomachs examined: Examined stomachs with food: Mean length of fish v*/ith food (mm SL): Length extremes in interval (mm SL): 464 restricted to hard bottom habitat for food resources. Although numerous in hard bottom areas (it ranked third in total number and second in total weight in trawl catches over all eight trawlable stations and six sampling periods combined in 1980 and 1981), Hae- mulon aurolineatum has been characterized as a reef-related species; i.e., it uses the reef for only part of each day (Parrish and Zimmerman 1977). Randall (1967) found sand-dwelling organisms in 16 tomtate stomachs he examined, but the habitat of the deca- pods, the predominant prey, could not be inferred from his results. Parrish and Zimmerman (1977) noted a diet dominated by sand-flat invertebrates for an unspecified number of tomtate collected in the Caribbean. Parrish and Zimmerman (1977) reported nocturnal foraging, with tomtate sheltering in the reef during the day. During extensive (about 70 dives) daytime scuba observations by the author off of South Carolina and Georgia, no tomtate that ex- hibited foraging behavior was seen, and large schools were often noted "stacked up" at the edge of rock ledges protruding out into sand areas. Apparently, nocturnal feeding behavior described for tomtate in the Caribbean is also typical for the species in the South Atlantic Bight. Tomtate forage, apparently at night, on sand bottom areas of the shelf or in sand patches often found adjacent to rock outcrops, returning to the reefs for shelter during the day. This behavior probably results in considerable energy transfer, in the form of feces, from open sand bottom areas of the shelf onto hard bottom reefs. The fact that two hard bottom invertebrate species (Erichthonitcs brasiliensis and Luconacia incerta) were common in tomtate stomachs and that many additional hard bottom species (e.g., hydroids, many amphipods, alpheid decapods, and bryozoans) are oc- casionally eaten indicates that tomtate also forage to a limited extent on hard bottom. The high diversity found in hard bottom inverte- brate communities (Wenner et al. 1983) could be attributed, in part, to predation by abundant and diverse fish communities (Petersen 1979). However, as noted by Wenner et al. (1983), available data in- dicate that few dominant species of hard bottom invertebrates are heavily preyed upon by fishes (Sedberry and Nimmich^). Tomtate, an abundant ^Sedberry, G. R., and T. A. Nimmich. Food habits of some fishes associated with live bottom habitat off the South Atlantic coast of the U.S.A. Manuscr. in prep. South Carolina Marine Resources Research Institute, P.O. Box 12559, Charleston, SC 29412. Table 3. — Relative abundance (percent of total number of individuals) and electlvlty index values (E) of dominant species in suction samples (Group A) and Haemulon aurolineatum stomachs (Group B) by depth zone. Dominant species included those that ranked in the five most abundant species within each Group (A or B) in either depth zone, for collections pooled for all years and seasons. Percent of total nun- iber of in( jividuals Inner shelf Middle shelf Benthic Fish Benthic Fish samples stomachs £ samples stomachs E Group A: Chone amerlcana 0.33 0.36 0.04 0.81 0.09 -0.79 Erichthonius brasiliensis 2.89 9.32 0.53 0.30 0.19 -0.24 Exogone dispar 3.71 0.72 -0.68 0.47 - 1.00 Filograna implexa 20.42 -1.00 63.87 -1.00 Luconacia incerta 3.27 0.36 -0.80 1.03 1.77 0.26 Malacoceros glutaeus 0.41 -1.00 0.81 -1.00 Podocerus sp. A 2.87 -1.00 0.27 -1.00 Syllis spongicola 2.15 -1.00 1.90 -1.00 Total 36.05 10.76 69.46 2.05 Group B: Armandia maculata 0.22 10.75 0.96 0.03 0.19 0.73 Brachyura megalopae 6.09 1.00 0.00 Branchiostoma caribaeum 0.00 0.01 15.69 0.99 Calanopia americana 0.36 1.00 16.34 1.00 Caprella equilibra 1.55 2.87 0.30 0.34 1.40 0.61 Erichtonius brasiliensis 2.89 9.32 0.53 0.30 0.19 -0.24 Ervilia concentrica 0.36 1.00 48.18 1.00 Luconacia incerta 3.27 0.36 -0.80 1.03 1.77 0.23 Melita appendiculata 0.43 4.66 0.83 0.27 -1.00 Temora turbinata 4.30 1.00 0.28 1.00 Total 8.36 39.07 1.98 84.04 Stomachs with food: 43 50 465 predator on hard bottom areas (Sedberry and Van Dolah 1984), do not serve as "keystone" predators (Paine 1969) which influence community structure of invertebrates on South Atlantic Bight hard bottom reefs. Acknowledgments I thank the many individuals of the South Carolina Marine Resources Center who were involved in all phases of the 3-yr BLM (contracts AA551-CT9-27 and AA551-CT1-18) and MMS (contract 14-12-0001- 29185) funded study of hard bottom habitat in the South Atlantic Bight. I especially thank D. M. Knott, R. F. Van Dolah, and E. L. Wenner for providing the data on macrobenthos that was used in the electivity analysis; C. B. O'Rourke and E. C. Roland (poly- chaetes), H. Porter (bivalves), and E. L. Wenner (decapods) for help in prey identification; C. A. Barans, C. A. Wenner, and the anonymous re- viewers for comments that improved the manu- script; and N. M. Beaumont for processing the manu- script. Literature Cited Chess, J. R. 1979. An airlift sampling device for in situ collecting of biota from rock substrata. Mar. Technol. Soc. J. 12(3):20-23. Darcy, G. H. 1983. Synopsis of biological data on the grunts Haemulon aurolineatum and H. plumieri (Pisces: Haemulidae). U.S. Dep. Commer., NOAA Tech. Rep. NMFS Circ. 448, 37 p. Fauchald, K., and p. a. Jumars. 1 979. The diet of worms: a study of polychaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. 17:193-284. HlLDEBRAND, S. P., AND W. C. SCHROEDER. 1928. Fishes of Chesapeake Bay. U.S. Bur. Fish., Bull. 48: 1-366. Hynes, H. B. N. 1950. The food of fresh-water sticklebacks (Gasterosteus aculeatus and Pygosteus pungitius), with a review of methods used in studies of the food of fishes. J. Anim. Ecol. 19:36-58. IVLEV, V. S. 1961. Experimental ecology of the feeding of fishes. Yale Univ. Press, New Haven, Conn., 302 p. Manooch, C. S., Ill, AND C. A. Barans. 1982. Distribution, abundance, and age and growth of the tomtate, Haemulon aurolineatum, along the southeastern United States. Fish. Bull., U.S. 80:1-19. Paine, R. T. 1969. The Pisaster-Tegula interaction: prey patches, predator food preference, and intertidal community structure. Ecol- ogy 50:950-961. Parrish, J. D., and R. J. Zimmerman. 1977. Utilization by fish of space and food resources on an off- shore Puerto Rican coral reef and its surroundings. In R. L. Taylor (editor). Proceedings: Third International Coral Reef Symposium, Vol. I: Biology, p. 297-303. Univ. Miami. Peterson, C. H. 1979. The importance of predation and competition in orga- nizing the intertidal epifaunal communities of Barnegat Inlet, New Jersey. Oecologia 39:1-24. PiNKAS, L., M. S. Oliphant, and I. L. K. Iverson. 1971. Food habits of albacore, bluefin tuna, and bonito in California waters. Calif. Dep. Fish Game, Fish Bull. 152:1- 105. Porter, H. J. 1974. The North Carolina marine and esturine mollusca-an atlas of occurrence. Inst. Mar. Sci., Univ. North Carolina, N.C., 351 p. Randall, J. E. 1967. Food habits of reef fishes of the West Indies. Stud. Trop. Oceanogr. (Miami) 5:665-847. Sedberry, G. R., and R. F. Van Dolah. 1984. Demersal fish assemblages associated with hard bottom habitat in the South Atlantic Bight of the U.S.A. Environ. Biol. Fishes 11:241-258. Struhsaker, p. 1969. Demersal fish resources: composition, distribution, and commercial potential of the continental shelf stocks off south- eastern United States. U.S. Fish Wildl. Serv., Fish. Ind. Res. 4:261-300. Wenner, C. A. 1983. Species associations and day-night variability of trawl- caught fishes from the inshore sponge-coral habitat, South Atlantic Bight. Fish. Bull., U.S. 81:537-552. Wenner, C. A., C. A. Barans, B. W. Stender, and F. W. Berry. 1980. Results of MARMAP otter trawl investigations in the South Atlantic Bight V. Summer, 1975. S.C. Mar. Resour. Cent, Tech. Rep. 45, 57 p. Wenner, E. L., P. Hinde, D. M. Knott, and R. F. Van Dolah. 1984. A temporal and spatial study of invertebrate commu- nities associated with hard-bottom habitats in the South Atlantic Bight. U.S. Dep. Commer., NOAA Tech. Rep. NMFS 18, 104 p. Wenner, E. L., D. M. Knott, R. F. Van Dolah, and V. G. BuRRELL, Jr. 1983. Invertebrate communities associated with hard bottom habitats in the South Atlantic Bight. Estuarine Coastal Shelf Sci. 17:143-158. Windell, J. T. 1971. Food analysis and rate of digestion. In W. E. Ricker (editor), Methods for assessment of fish production in fresh waters, p. 215-226. IBP (Int. Biol. Programme) Handb. 3, 2ded. George R. Sedberry South Carolina Marine Resources Research Institute P.O. Box 12559 Charlestm, SC 29^12 466 SEMILUNAR REPRODUCTIVE CYCLES IN FUNDULVS HETEROCLITUS (PISCES: CYPRINODONTIDAE) IN AN AREA WITHOUT LUNAR TIDAL CYCLES Although lunar spawning rhythms are relatively common in species of shallow-water fish, semilunar reproductive cycles have been reported in a small but growing number of species (Korringa 1947; Johannes 1978). Species with the best documented semilunar gonad and spawning cycles include the California grunion, Leuresthes tenuis, (Clark 1925); Atlantic silversides, Menidia menidia, (Middaugh 1981); the tropical coral reef saddleback wrasse, Thallassoma duperrey, (Ross 1983); two tropical damselfishes, Pom,acentrus flairicauda and P. wardi, (Doherty 1983), and gulf killifish, Funduhus grandis, (Greeley and MacGregor 1983); and the salt marsh mummichog, Fundulus heteroclitus, (Taylor et al. 1979; Taylor and DiMichele 1980). On the east coast of North America, F. heteroditics move up onto marsh surfaces during high spring tides to spawn either in empty mussel (Geukensia demissa) shells or in the outer leaves of salt marsh plants (Spartina altemiflora) where oxygen levels are relatively high and sedimentation rates are relatively low (Able and Castagna 1975; Taylor et al. 1977; Kneib and Stiven 1978; Taylor and DiMichele 1983). Its eggs are ex- tremely tolerant of desiccation and hatch on the next series of high spring tides. In addition, semilunar rhythms of larval hatching are well known for inter- tidal chironomid insects (Newmann 1978) and for several species of intertidal and estuarine crabs (see Christy 1982; Forward et al. 1982). In most cases, the proximal factors (sensu Giese and Pearse 1974) responsible for initiating and syn- chronizing these semilunar cycles are unknown. However, among the factors postulated as cues are tidal rhythms in water turbulence and hydrostatic pressures (Korringa 1947; Newmann 1978; Weld and Meier 1982; Ross 1983); tidal regimes in the habitat (Forward et al. 1982); photoperiod in combi- nation with tidal fluctuations in water temperature (Miller et al. 1981), in combination with time of high tide (Middaugh 1981; Middaugh and Takita 1983), or in combination with interrupted current velocity (Middaugh and Hemmer 1984); and moonlight (Saigusa 1980). In addition to its wide distribution in coastal marshes with distinct lunar cycles of spring and neap tides, F. heteroclitus is abundant in exten- sive shoreline habitats and marshes of Chesapeake Bay, where tidal ranges are small and changes in water level caused by barometric pressure and wind frequently and unpredictably override lunar tidal levels and obliterate semilunar tidal cycles. In the present study, we measured the reproductive activi- ty of F. heteroclitus in an area without lunar tidal cycles in order to determine if semilunar reproduc- tive rhythms occur. The occurrence of semilunar reproductive rhythms would suggest that the prox- imal cues regulating the reproductive cycles are not factors associated with changes in tidal levels, such as pressure, or turbulence of currents. Methods This study was conducted from May through August 1982 at a small tidal creek (Muddy Creek) which flows into the Rhode River (lat. 38°5rN, long. 76°32'W), a subestuary on the western shore of cen- tral Chesapeake Bay, located about 11.3 km south of Annapolis, MD. The creek bottom consists of fine clays and silts, and its banks are fringed by a cattail {Typha angvstifolia) marsh. Water level was measured with a Honeywell^ diffused silicon differ- ential pressure transmitter (accurate to ± 1.238 cm) at a station operated by the U.S. Geological Survey on the Rhode River near the mouth of Muddy Creek about 1,000 m from the site where fish were sam- pled. Changes in water level at the monitoring sta- tion accurately reflect water levels at the sampling site (D. L. Correll^ and R. L. Cory^). Measured daily high water levels were compared with predicted high tidal levels published for the Rhode River (National Oceanic and Atmospheric Administration 1981). Fundulus heteroclitus is the most abundant fish in the creek (Mines unpubl. data). Fish were sampled every 2-3 d during the hours of 0730-1230, using un- baited minnow traps set just above low water level. Each sample consisted of 8-12 males and 8-12 females which were tested for readiness to spawn, and another 8-12 of each sex were taken to deter- mine gonad indices. Readiness to spawn was deter- mined in the field by applying gentle pressure from anterior to posterior along the ventral surface of the fish. Release of sperm or eggs was interpreted as the fish being ripe. The gonad index for each fish in the sample was determined by dissecting out the testes or ovaries and by drying the gonad and body to con- stant weight at 60°C and weighing them to the near- 'Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. ^D. L. Correll, Assistant Director, Smithsonian Environmental Research Center, P.O. Box 28, Edgewater, MD 21037, pers. com- mun. September 1982. 3R. L. Cory, Oceanographer, U.S. Geological Survey, Smithsonian Environmental Research Center, P.O. Box 28, Edgewater, MD 21037, pers. commun. September 1982. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 467 est 1 X 10"'' g. The gonad index equals the (gonad weighty body weight) x 100. Results Although tides in the Rhode River subestuary ex- hibited an approximately semidiurnal rhythm, fluc- tuations in measured high tide level were not corre- lated with predicted high tide level (product moment correlation of log transformed data for daily predict- ed and measured high tide, r = 0.111; Student's ^test for r i^ 0, P > 0.2; Fig. 1). Moreover, time series spectral analysis (Thrall and Engelman 1981) showed that measured tides exhibited only a very weak peak in spectral density at a frequency of about 12.5 d rather than the strong peak of 7.5 d exhibited by predicted tidal cycles. Cross correlation between predicted and measured tides in bivariate spectral analysis showed low and variable coherence {P > 0.05). Similarly, linear association of the two vari- ables was weak, with variable phase shifts and vari- able coefficients necessary to fit one variable to the other. In addition, the occurrence of tides sufficiently high to flood the marsh fringing the creek (i.e., tides > 46 cm) did not occur more frequently at night than during day (x^ test, P > 0.2), nor more frequently during any particular lunar phase (x^ test, P > 0.4). High tides did not occur consistently during any 2-h period of the 24-h day at the quarters of the new or full moons (x^ test, P > 0.4). Thus, water levels in the estuary were neither predictable in amplitude within the tidal range nor related in any obvious cyclical pattern to the lunar cycle. However, both female and male F. heteroclitus showed distinct semimonthly cycles in readiness to spawn from May through August (Fig. 1). Females and males also exhibited a distinct semimonthly cycle of gonad size, although the cycle of male gonad index was not as pronounced as that of females, due to the small changes in size of the testes during spawning (Fig. 1). The cycles of spawning readiness in males and females were usually in phase with the lunar cycle. Comparisons of the observed frequencies of spawning readiness with frequencies predicted by cycles with perfect semilunar periods and the same amplitudes showed no significant differences in 27 of 37 d of observation for females and 33 of 37 d of observation for males (x^ test, P < 0.05). Observed frequencies of spawning readiness over the entire study period were not significantly different from frequencies predicted by the perfect semilunar cycles for either females or males (x^ test, P > 0.2). Although six out of six peaks of spawning readiness for males occurred during the 7 d surrounding new or full moons, for females only four of the six peaks occurred during the first or last quarters of the moon, indicating that for the small numbers of peaks occurring during a reproductive season, readiness to spawn does not coincide with new and full moons (Fisher's Exact Test for number of spawning peaks occurring within the specified period, P = 0.227; Fig. 1). However, if the timing of the peaks in readi- ness to spawn is considered to lag 3.5 d after the new and full moons, then all peaks for males and females occurred within the 7 d surrounding the lagged period, indicating a significant synchronous semi- lunar cycle (Fisher's Exact Test on number of peaks occurring within the specified period, P < 0.01). Female spawning was correlated with male spawn- ing (product moment correlation of arcsine trans- formed spawning frequencies, r = 0.695; Student's ^test for r ¥= 0, P < 0.001). However, neither female nor male readiness to spawn was correlated with the measured tides (product moment correlation of arc- sine transformed frequencies of spawning readiness with log transformed high tide measurements, r = 0.184 for females and r = 0.272 for males; Student's ^tests for r # 0, P > 0.2). Thus the semilunar cycles of spawning readiness of both sexes appeared to be synchronized, but not to be related to the tidal regime of the estuary. The reproductive season of F. heteroclitus in the Rhode River-Muddy Creek estuarine system oc- curred from late April to September, when water temperatures were above 17° C (pers. obs.). Mean gonad indices of female and male samples declined during the season, both at the peak and at the spent phases of the semimonthly cycles (Female Gonad In- dex = -0.124 Day + 14.1 and Male Gonad Index = - 0.035 Day + 3.83; Student's ^test for slopes ¥=0,P < 0.05; (Fig. 1)). The decline of mean gonad index during the season reflected two statistics. First, an increasing percentage of the population failed to in- itiate gonad recrudescence during successive cycles. Figure 1.- Lunar phase, tidal levels, and reproductive cycles of Fundulus heteroclitus from May to August 1982 in a tidal creek on central Chesapeake Bay. Full moons (open circles) and new moons (solid circles) are indicated. Daily high water levels above mean low water are shown for measurements at the study site and for tidal levels predicted by National Oceanographic and Atmospheric Ad- ministration (1981). Horizontal lines indicate water level which floods the marsh adjacent to the Creek. The percentages of female and male fish which were ready to spawn are shown for each sample throughout the spawning season (solid lines), along with hypothetical spawning cycles with the same amplitudes and perfect semilunar periods in phase with the new and full moons (dashed lines). Gonad indices (mean ± SE) for females and males are also plotted for each sample. 468 MOON O O O O 60-1 predicted ^^^ ^.v^ H- ^ 90H measured UJ O LU 60- •V. 9 d 9 cP° «te 10%), whereas many had undeveloped gonads on the July peaks (38% of the sample for gonad weights had individual indices < 5%, and 25% of the sample tested for spawning were not ripe). Second, gonad size of fish with de- veloped gonads declined during the season. Mean gonad indices of females with developed ovaries of the May peaks were significantly greater than those of the July peaks (15% versus 9%, respectively) (Student-Newman-Keuls test of arcsine transformed data, P < 0.05). Discussion In Fundulus heteroclitus and F. conjluentus, regulation of the annual reproductive cycle is ap- parently dependent on changing combinations of photoperiod and temperature (Harrington 1959; Day and Taylor 1982). The decline in gonad index during the season in the present study is consistent with a similar decline observed by Kneib and Stiven (1978) and Taylor et al. (1979) and suggests that, along with physical factors, energy availability may be limiting reproductive output late in the season. Weisberg (1981) found that supplemental food increased the gonad indices of F. heteroclitus in Delaware salt marshes. Histological evidence shows that primary oocytes in F. heteroclitus can undergo complete development in one biweekly spawning cycle (Taylor and DiMichele 1980), so lag time for recrudescence of spent gonads does not account for reduced repro- ductive activity late in the season. Funduhis grandis reproduces throughout the summer in some areas of the gulf coast (Greeley and MacGregor 1983), but shows bimodal reproductive activity in spring and fall with no reproduction in midsummer in shallow ponds (Waas and Strawn 1983). In addition to F. heteroclitus (Taylor et al. 1979; Taylor and DiMichele 1980), semilunar spawning cycles occur in some populations of F. grandis (Greeley and MacGregor 1983), F. similis (Greeley 1982), F. majalis (Tedesco et al."), and probably inF. notatus (inferred from observations of spawning behavior in Carranza and Winn 1954). Waas and Strawn (1983) measured a weak lunar, but not a semilunar cycle in populations of F. grandis in two nontidal ponds and a tidal creek with lunar tides fre- quently overridden by wind effects. The semilunar ^Tedesco, M., A. H. Hines, and L. A. Wiechert. 1983. Semi- lunar gonadal cycles mFunduliis maja/i-s (Pisces: Cyprinodontidae). Technical Report of Smithsonian Environmental Research Center, P.O. Box 28, Edgewater, MD 21037. spawning cycle in F. grandis has been induced in the laboratory during their reproductive season by inter- action of photoperiod and tidal changes in water temperature shifting on a lunar cycle (Miller et al. 1981). Daily disturbances (netting) of F. grandis also apparently induces circadian gonadal responses (Weld and Meier 1982). Tidal rhythms in water tur- bulence and hydrostatic pressure have been postu- lated for lunar and semilunar reproductive rhythms in fish (Korringa 1947; Ross 1983) and for semilunar hatching in intertidal chironomid insects (Newmann 1978). Middaugh (1981) suggested that the biweekly coincidence of a high tide at the time of sunrise and/or a lunar cue may serve as a synchronizer for spawning of Menidia menidia in a South Carolina estuary. In contrast to M. menidia, which showed a precise reproductive response to diurnal tidal and lighting schedules (Middaugh and Takita 1983), M. peninsulae exhibited a variable and labile response in which a combination of semidiurnal interruptions of current and diel light cues was optimal at inducing spawning synchrony (Middaugh and Hemmer 1984). However, this synchrony in Menidia spp. had a diel period, and the laboratory experiments were not run long enough to test for lunar rhythms. Semilunar rhythms of larval hatching are well documented for several species of intertidal and estuarine crabs (see Christy 1982; Forward et al. 1982). Hatching in the mud crab, Rhithropanopeus harrisii, followed a semilunar and diurnal rhythm in populations from an estuary with semidiurnal and lunar tidal cycles, whereas hatching was not associated with lunar rhythms or tidal levels in an estuary with aperiodic tides (Forward et al. 1982). Moveover, Forward et al. (1982) induced circatidal rhythms in larval release in crabs from an estuary with irregular tides by transplanting them to an estuary with semidiurnal and lunar tides. However, Saigusa (1980) showed that the larval hatching cycle of a semiterrestrial crab Sesarma haematocheir is entrained directly by a moonlight cycle. In the present study, in an area without a lunar periodicity in the tidal cycle, F. heteroclitus had a semilunar reproductive cycle which lagged the new and full moons by 3.5 d. Although the study area had tidal fluctuations with an approximately semidiurnal period, the diel timing of high tides with respect to photoperiod also did not appear to cue the semilunar reproductive cycle. Changes in currents associated with the diurnal tidal cycle may serve as a synchro- nizing cue for the time of day of spawning, as it does for Menidia menidia and M. peninsulae. However, while identity of the cue remains unknown, the pres- ent study indicates that tidally mediated factors such 470 as turbulence, hydrostatic pressure, temperature, and salinity are not responsible for synchronizing the semilunar rhythm. Hypotheses concerning the ecological consequen- ces and adaptive significance of semilunar cycles of larval release in Uca spp. and other estuarine crabs have been discussed recently by Christy (1982). His analysis indicates that convergence among estuarine crab species in the semilunar timing of larval release results in rapid seaward transport of larvae, which minimizes the effects of lethal combinations of high temperature and low salinities or intense predation in the upper estuary. However, Rhithropanopeus harrisii completes its larval development within the upper reaches of estuaries and shows a variable cycle of hatching, depending on tidal predictability, which may be advantageous for retention of larvae within the estuary (Forward et al. 1982). Lunar spawning cycles of insular coral reef fish may have evolved to solve similar problems of larval dispersal away from predators while assuring return of the recruits to the adult habitat (Johannes 1978; Ross 1983). Semilunar spawning cycles of Fundulus heteroditus, Menidia menidia, Leuresthes tenuis, and other fish which spawn in the upper intertidal zone (Taylor and DiMichele 1983; Middaugh 1981; Clark 1925) may have the adaptive advantage of removing spawning adults and eggs from subtidal predators. Although the eggs of F. heteroditus are tolerant of desiccation (Able and Castagna 1975; Taylor et al. 1977), the major advantage of the spawning site appears to be higher oxygen levels and reduced sedimentation than in the creek (Taylor and DiMichele 1983). How- ever, spawned eggs of F. heteroditus and M. menidia are usually found in areas inundated daily by high tides (Middaugh 1981). In the present study, 62% of the days had tides which flooded the marsh (Fig. 1), where eggs are probably deposited in the bases of leaves of Typha angustifolia plants (see Taylor and DiMichele 1983). Although there may be additional advantages in areas with predictable tidal fluctuations with lunar periodicity, the major advan- tage of semilunar spawning rhythms in F. heterodi- tus appears to be improved fertilization success af- forded by synchronized spawning. Acknowledgments We are grateful for the assistance of L. Wiechert, S. Hodgkins, C. Trowbridge, and K. Comtois. R. Cory and P. Dresler of the U.S. Geological Survey provided the data for measured tidal levels. R. Ross and M. Taylor provided helpful discussion. D. Cor- rell, T. Jordan, and two anonymous reviewers pro- vided suggestions for improvements in earlier drafts of the manuscript. This work was supported in part by the Smithsonian Work/Learn Internship Pro- gram, Maryland Department of Natural Resources, and the Smithsonian Environmental Sciences Pro- gram. Literature Cited Able, K. W., and M. Castagna. 1975. Aspects of an undescribed reproductive behavior in Fundul^(s heteroditus (Pisces: Cyprinodontidae) from Vir- ginia. Chesapeake Sci. 16:282-284. Carranza, J., AND H. E. Winn. 1954. Reproductive behavior of the biaclcstripe topminnow, Fundulus notatus. Copeia 1954:273-278. Christy, J. H. 1982. Adaptive significance of semilunar cycles in larval release in fiddler crabs (Genus Uca): Test of an hypothesis. Biol. Bull. (Woods Hole) 163:251-263. Clark, F. N. 1925. The life history of Leuresthes tenuis, an atherine fish with tide controlled spawning habits. Calif. Dep. Fish Game, Fish Bull. 10, 51 p. Day, J. R., and M. H. Taylor. 1982. Effects of temperature and photoperiod on the seasonal reproductive cycle of Fundulus heteroditus L. [Abstr.] Am. Zool. 22:866. DOHERTY, p. J. 1983. Diel, lunar and seasonal rhythms in the reproduction of two tropical damselfishes: Pomacentrus flavicauda and P. wardi. Mar. Biol. (Berl.) 75:215-224. Forward, R. B., Jr., K. Lohmann, and T. W. Cronin. 1982. Rhythms in larval release by an estuarine crab Rhithro- panopeus harrisii. Biol. Bull. (Woods Hole) 163:287-300. GiESE, A. C, and J. S. Pearse. 1974. Introduction: general principals. In A. C. Giese and J. S. Pearse (editors). Reproduction of marine invertebrates, Vol. 1, p. 1-49. Acad. Press, N.Y. Greeley, M. S., Jr. 1982. Tide-controlled reproduction in the long- nose killifish Fundulus similis. [Abstr.] Am. Zool. 22:870. Greeley, M. S., Jr., and R. MacGregor, III. 1983. Annual and semilunar reproductive cycles of the Gulf killifish, Fundulus grandis, on the Alabama Gulf Coast. Copeia 1983:711-718. Harrington, R. W., Jr. 1959. Effects of four combinations of temperature and day- length on the ovogenetic cycle of a low-latitude fish, Fundulus confluentus Goode & Bean. Zoologica (N.Y.) 44:149-168. Johannes, R. E. 1978. Reproductive strategies of coastal niarine fishes in the tropics. Environ. Biol. Fishes 3:65-84. Kneib, R. T., and a. E. Stiven. 1978. Growth, reproduction, and feeding of Fundulus hetero- ditus (L.) on a North Carolina salt marsh. J. Exp. Mar. Biol. Ecol. 31:121-140. Korringa, p. 1947. Relations between the moon and periodicity in the breeding of marine animals. Ecol. Monogr. 17:349-381. Middaugh, D. P. 1981. Reproduction ecology and spawning periodicity of the Atlantic silverside, Menidia menidia (Pisces: Atherinidae). Copeia 1981:766-776. 471 MiDDAUGH, D. P., AND M. J. HeMMER. 1984. Spawning of the tidewater silverside, Menidia penin- sula (Goode and Bean), in response to tidal and lighting schedules in the laboratory. Estuaries 7:139-148. MiDDAUGH, D. P., AND T. TAKITA. 1983. Tidal and diurnal spawning cues in the Atlantic silver- side, Menidia menidia. Environ. Biol. Fishes 8:97-104. Miller, C. A., J. M. Wilson, and A. H. Meier. 1981. Induction of semilunar rhythms of reproductive indices \n Fundulus grandis. [Abstr.] Am. Zool. 21:995. National Oceanic and Atmospheric Administration. 1981. Tide tables 1982, high and low water predictions, East Coast of North and South America including Greenland. U.S. Dep. Commer., 235 p. Newmann, D. 1978, Entrainment of a semilunar rhythm by simulated tidal cycles of mechanical disturbance. J. Exp. Mar. Biol. Ecol. 35:173-185. Ross, R. M. 1983. Annual, semilunar, and diel reproductive rhythms in the Hawaiian labrid Thallassoma duperrey. Mar. Biol. (Berl.) 72:311-318. Saigusa, M. 1980. Entrainment of a semilunar rhythm by a simulated moonlight cycle in the terrestial crab, Sesarma haemntocheir. Oecologia (Berl.) 46:38-44. Taylor, M. H., and L. DiMichele. 1980. Ovarian changes during the lunar spawning cycle of Fundidus heteroclitus. Copeia 1980:118-125. 1983. Spawning site utilization in a Delaware population of Fundulus heteroclitus (Pisces: Cyprinodontidae). Copeia 1983:719-725. Taylor, M. H., D. DiMichele, and G. J. Leach. 1977. Egg stranding in the life cycle of the mummichog, Fundulus heteroclitus. Copeia 1977:397-399. Taylor, M. H., L. DiMichele, M. M. Levitan, and W. F. Jacob. 1979. Lunar spawning cycle in the mummichog, Fundulus heteroclitus (Pisces: Cyprinodontidae). Copeia 1979:291- 297. Thrall, T., and L. Engelman. 1981. Univariate and bivariate spectral analysis. In W. J. Dixon (editor), BMDP statistical software, p. 604-638. Univ. Calif. Press, Los Ang. Waas, B. p., and K. Strawn. 1983. Seasonal and lunar cycles in gonadosomatic indices and spawning readiness of Fundulus grandis. Contrib. Mar. Sci., Texas A&M Univ. 26:127-141. Weisberg, S. B. 1981. Food availability and utilization by the mummichog, Fundulus heteroclitus (L.). Ph.D. Thesis, Univ. Delaware, Newark, 104 p. Weld, M. M., and A. H. Meier. 1982. Circadian gonadal responses to daily disturbances in gulf killifish. [Abstr.] Am. Zool. 22:866. Anson H. Hines Kenric E. Osgood Joseph J. Miklas Smithsonian Environmental Research Center P.O. Box 28 Edgewater, MD 21037 UNDERSEA TOPOGRAPHY AND THE COMPARATIVE DISTRIBUTIONS OF TWO PELAGIC CETACEANS Prey species are not uniformly distributed. Foraging efficiency, therefore, should be maximized when ef- fort is concentrated in areas where prey are concen- trated. Cetacean food is probably most concentrated in regions of high general productivity. Because the undersea topography may be a major influence on productivity, cetacean foraging patterns may be associated with the topographic patterns of the ocean floor (Hui 1979). I report here the occurrences of two species of pelagic odontocete cetaceans, the Pacific pilot whale, Globicephala Tnacrorhyrwhus, and the common dolphin, Delphinus delphis, relative to seafloor topography and to diet. Although it is not clear if the genus Delphinus in this region is com- posed of two species or one species with two sub- species, the vast majority are Delphinus delphis (Banks and Brownell 1969; Evans 1975). If data from more than one species are included in this study, it is assumed that any interspecies difference in distribution relative to substrate was not signifi- cant to the analyses. Methods and Results This study was conducted in the Southern Califor- nia Continental Borderland (Fig. 1) which consists of ridges, deep troughs, and basins (Chase 1968). There were 61 survey flights totaling 22,353 km. The flights were conducted at various times, all of them during midday (1000-1500) from 1968 through 1976. Totals of 1,057 pilot whales in 38 aggregations (in January, March, April, July, October, and Decem- ber) and 47,105 common dolphins in 142 aggrega- tions (in all months of the year) were observed. The survey methods have been previously described (Hui 1979). The data for each species were not divided into temporal subsets as in Hui (1979), but were treated as whole sets. The distribution of each species was examined relative to depth and relative to change in depth. The method of analysis has been presented in detail elsewhere (Hui 1979). Both pilot whales and common dolphins were distributed uniformly {P > 0.10) throughout the depth classes but not uniformly among the Contour Index (CI) classes {P < 0.005). For each species there were more observations than expected over the steepest bottom topography and fewer than ex- pected over the flattest (Table 1). I also compared the distribution of pilot whales 472 FISHERY BULLETIN: VOL. 83, NO. 3, 1985. among CI classes to the distribution of common dolphins. Due to the small number of pilot whale aggregations, the CI range was divided into four unequal classes to meet the statistical assumptions for chi-square analysis (Dixon and Massey 1969). The proportion of pilot whales in each of these classes 34^ 33^ 32* 120^ 119< 118' 117° Figure 1.- Sightings of pilot whales, Giobicephala macrorhyyichus, and common dolphins, Delphinus delphis. The study area is bounded on the west by long. 120°00'W and on the east by the coast of California between lat. 34°05'N and 32°24'N. Sightings may not accurately depict the geographic distribution of these species because the survey flights were not uniformly distributed spatially or temporally. Encounters occurring outside the study area are not shown. Table 1. — Distribution among contour index classes of observed and expected ag- gregations (sec text). Contour index class Number of aggregations (F./)2 Species Expected (F) Observed (0 F Pilot whales 0.01-19.99 13.3 1 11.38 20.00-39.99 6.3 5 0.27 40.00-59.99 5.2 4 0.28 60.00-79.99 7.4 10 0.91 80.00-99.99 5.8 18 X2 25.66 = 38.50 Common dolphins 0.01-19.99 48.0 18 18.75 20.00-39.99 23.2 27 0.62 40.00-59.99 19.3 22 0.27 60.00-79.99 27.2 21 1.41 80.00-99.99 22.2 54 X2 45.55 = 66.60 X2(df = 4, P = 0.005) = 14.86 473 was used to compute the expected number of com- mon dolphins in that class. The distribution of pilot whales was not the same as that of common dolphins among the CI classes (P < 0.005), particularly in the class for the lowest relief. Of the total chi-square value, 63% is due to proportionally more observations of dolphins than pilot whales in this one class (Table 2). The distributions of the two sets of survey flights used to collect data for the two species were com- pared, but the CI range in this case was divided into 10 equal classes. The two sets of survey flights were equally distributed among the CI classes {P > 0.10). Table 2. — Comparative distribution among contour index classes of pilot whales and common dolphins. Contour index Dolphin aggrega tions (F-/)2 class Expected (F) Observed {f) F 0.01-39.99 22.4 45 22.80 40.00-59.99 14.9 22 3.38 60.00-79.99 37.4 21 7.19 80.00-99.99 67.3 54 X^ 2.63 = 36.00 X2 (df = 3, P = 0.005) = 12.84 'Computed from pilot whale sightings; see text. Discussion In the California Bight, pilot whales and common dolphins are distributed similarly above undersea topography of high relief, but common dolphins oc- cur more frequently than pilot whales over areas of low relief. This difference may be linked to differ- ences in feeding habits. Pilot whales and common dolphins have significant morphological and physiological differences that are associated with their foods and feeding methods. Pilot whales are larger (7 m vs. 2 m in length) and have fewer but larger teeth (40 vs. 200) than com- mon dolphins (Orr 1972). Pilot whales are capable of diving to 610 m (Bowers and Henderson 1972) while common dolphins dive to 257 m (Evans 1971). The diet of the common dolphin in the Southern California Bight includes 19 species of fish, 2 species of squid, and miscellaneous crustaceans (Fitch and Brownell 1968; Evans 1975). In contrast to common dolphins, pilot whales ap- pear to be stenophagous, eating primarily squid. Atlantic pilot whales, GLobicephala melaena, eat primarily squid {Illex illecebrosus). The only fish reported eaten (cod, Gadus morhua) composes <10% of the diet (Sergeant 1962). There has been no study on the Pacific pilot whale comparable with that of Sergeant's (1962) on the Atlantic form; however, the stomachs of four wild Pacific pilot whales have been examined. They contained squid but no fish (W. A. Walker,! (] \y Woodhouse,^ D. J. Seagars^). In cap- tivity an Atlantic pilot whale rejected mullet (Mugili- dae) and blue runner (Carangidae) fish and accepted only squid (probably Loligo pealei) until it was trick- ed into eating some herring (Clupeidae); and then the indications were that "He did not seem to digest the fish as well" (Kritzler 1949). Squid distribution cannot be related to any particu- lar bottom topography along the California coast because squids are not easily collected with sampling methods used in distribution studies (Mais 1974). However, it may be inferred that the narrow range of seafloor topographies visited by pilot whales reflects the narrow range of their diet and the areas where squid can be most easily caught by pilot whales. Also concentrated over areas of canyons and escarpments are anchovies (Mais 1974), a major com- ponent of the dolphin diet (Fitch and Brownell 1968; Evans 1975). Common dolphins frequent these areas most. However, common dolphins are euryphagous. Some prey probably occur over seafloor of low relief, although this could not be confirmed from fish reports because bottom topography is not a para- meter which is recorded in fish distribution studies. If some prey do occur over areas of low relief, their distribution would partially explain why dolphins oc- cur over seafloor of low relief more frequently than do pilot whales. My analyses show that the daytime distribution patterns of these two pelagic cetacean species are not random but are related to bottom topography. Although the distributions are similar, they are not the same. Differences in distributions may be due to the different foraging patterns but no firm conclu- sion can be drawn until more information is available. Acknowledgments I thank G. A. Bartholomew, M. F. Platter-Rieger, F. G. Wood, and two anonymous reviewers for their helpful comments; also D. J. Seagars, W. A Walker, and C. W. Woodhouse for information on the 'W. A. Walker, Research Associate, Section of Mammalogy, Natural History Museum of Los Angeles County, Los Angeles, CA 90007, pers. commun. July 1980. -C. W. Woodhouse, Curator of Vertebrate Department, Santa Barbara Museum of Natural History, 2559 Puesta del Sol Road, Santa Barbara, CA 93105, pers. commun. July 1980. 'D. J. Seagars, Wildlife Biologist, National Marine Fisheries Ser- vice, Southwest Region, 300 S. Ferry St., Terminal Island, CA 90731, pers. commun. January 1981. 474 stomach contents of Pacific pilot whales. My special thanks to W. E. Evans of Hubbs Sea World Research Institute for his support and encourage- ment throughout this study, which was supported by ONR contract T0044 subproject RR0310201 (W. E. Evans, principal investigator), and NMFS agree- ment 01-6-200 11439. Literature Cited Banks, R. C, and R. L. Brownell. 1969. Taxonomy of the common dolphins of the eastern Paci- fic ocean. J. Mammal. 50:262-271. Bowers, C. A., and R. S. Henderson. 1972. Project Deep Ops: Deep object recoverj' with pilot and killer whales. NUC TP 306, 86 p. Naval Undersea Center, San Diego, CA 92152. Chase, T. E. 1968. Sea floor topography of central eastern PacijRc Ocean. Bur. Commer. Fish. Circular 291, 33 p. Dixon, W. J., and F. J. Massey, Jr. 1969. Introduction to statistical analysis. 3d ed. McGraw- Hill, N.Y., 638 p. Evans, W. E. 1971. Orientation behavior of delphinids: radio telemetric studies. Ann. N.Y. Acad. Sci. 188:142-160. 1975. Distribution, differentiation of populations, and other aspects of the natural history of Delpkinus delphis Linnaeus in the northeastern Pacific. Ph.D. Thesis, Univ. California, Los Angeles, 164 p. Fitch, J. E., and R. L. Brownell, Jr. 1968. Fish otoliths in cetacean stomachs and their importance in interpreting feeding habits. J. Fish. Res. Board Can. 25: 2561-2574. Hui, C. A. 1979. Undersea topography and distribution of dolphins of the genus Delphimis in the southern California bight. J. Mam- mal. 60:521-527. Kritzler, H. 1949. The pilot whale at Marineland. Natural History 58: 302-308, 331-332. Mais, K. F. 1974. Pelagic fish surveys in the California current. Calif. Dep. Fish Game, Fish Bull. 162, 79 p. Orr, R. T. 1972. Marine mammals of California. Univ. Calif. Press, Berkeley, 64 p. Sergeant, D. E. 1962. The biology of the pilot or pothead whale Globicephala melaena (Traill) in Newfoundland waters. Fish. Res. Board Can., Bull. 132, 84 p. Clifford A. Hui Chemistry and Biochemistry Branch, Code 521 Naval Ocean Systems Center San Diego, CA 92152-5000 LARVAL AND JUVENILE GROWTH OF SABLEFISH, ANOPLOPOMA FIMBRIA, AS DETERMINED FROM OTOLITH INCREMENTS The black cod or sablefish, A naplopoma fimbria, has been the subject of an intensifying fishery off the west coast of North America over the last decade. Biological information on this species, however, in- cluding data on spawning, early life history, age and growth, and population structure, is relatively meager. Sablefish are widely distributed in the northern Pacific, with adults most abundant at depths of 366-915 m (Hart 1973). Mason et al. (1983) suggested that eggs are spawned and developed in waters deeper than 300 m and colder than 6°C off Canada. Juveniles occur in shallow water, however, and larvae are almost exclusively neustonic (Kendall and Clark^). Thus larval development and growth oc- cur in much warmer water than that inhabited by later stages, particularly in the southern portion of the species range. Sablefish growth has been described by Heyamoto (1962) and Pruter (1954), among others, who used scale annuli to define the growth pattern. More re- cent work, however, has shown that the age esti- mates, particularly for older, mature fish, are in er- ror; growth is apparently much slower and longevity much greater than previously thought (Beamish and Chilton 1982). The warmer neustonic habitat of the larvae may result in different growth patterns in early life; ontogenetic changes in growth and habitat are relatively common among deeper living fishes (Boehlert 1982; Luczkovich and 011a 1983). The only observations on growth of young sablefish are those of Heyamoto (1962), who suggested that juveniles of 12-16 cm fork length (FL) were about 6 mo old. In the present study we report on the growth of field- collected larval and juvenile sablefish where age was estimated by enumerating growth increments on the otoliths. Materials and Methods Larval and juvenile sablefish were collected in 1981-83. Larvae were taken in 0.5 m neuston nets (Sameoto and Jaroszynski 1969) with 0.505 mm mesh, off the coasts of Oregon and Washington dur- ing May 1982 by the RV Posey don. Samples were im- mediately preserved in 80% ethanol. After sorting. ^Kendall, A. W., Jr., and J. Clark. 1982. Ichthyoplankton off Washington, Oregon, and northern California, April-May 1980. Processed Rep. 82-11, 44 p. Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake Blvd. East, Seattle, WA 98102. FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 475 larvae were stored in individual vials labeled with sample number and date. Additional larvae were col- lected with neuston nets in May 1983 by the RV Ekvator. Larger juveniles (> 70 mm standard length (SL)) were taken in a small mesh purse seine deployed from 24 to 40 km off of the Oregon-Wash- ington coasts during the summer months of 1981 (Fig. 1). Specimens were frozen on board and held until measurements and otoliths were taken. Fork lengths to the nearest millimeter were recorded for these larger juveniles and standard lengths to the nearest 0.1 mm were measured for all larvae and small juveniles. No corrections were made for poten- tial shrinkage from preservation of young larvae, but alcohol preservation causes no noticeable shrinkage in length of anchovy larvae (Theilacker 1980). For subsequent analysis, fork lengths were converted to standard length by the relationship SL (mm) = 0.91 FL (mm) -1.15 (n = 54, r^ = 0.999), which was based upon specimens 21.7 to 297 mm FL. Otoliths of larval sablefish were removed and cleaned under a dissecting microscope fitted with polarizing filters. Increments on otoliths from larvae < 26 mm SL were clearly visible from the focus to the margin (Fig. 2); these otoliths were left intact, af- fixed to microscope slides with histological mounting medium and cover slips, and increments were read in the sagittal plane (see Taubert and Coble 1977 for terminology). For larger larvae and most juveniles, a sagittal section of the otolith provided the clearest in- crements. The left otolith of every pair was mounted in histological medium on a microscope slide and the Figure 1.- Distribution of AnopUrpoma fimbria samples, along the Pacific coast, used for age and growth analysis. Circles represent the neuston samples taken during May 1982, triangles represent the purse seine samples taken during sum- mer 1981, and squares represent the 1983 neuston samples. 476 mm Figure 2. - Sagitta otolith from a larval Anoplopomafiynbria (22.0 mm SL; duplicate increment counts were 37 and 40 d). Scale bar = 0.1 mm. internal surface was ground until the focus was visi- ble. The microscope slide was heated and the section was turned over to expose the external surface. Grinding and polishing continued, while care was taken to insure that material was not lost from the margin. The result was a clear, thin section of the otolith in the sagittal plane. For some of the larger juveniles (> 100 mm SL), transverse sections were cut from the otoliths using a diamond saw, mounted on microscope slides, and ground to make the incre- ments clear. All otoliths were read under a com- pound microscope at 400 x or 1,000 x magnifica- tion. Two independent counts were made for each otolith. These counts were made at least 2 wk apart; the age assigned to each specimen was the mean of the two counts. Increments, comprised of adjacent light and dark ring pairs, were distinct and clear in the smallest otoliths (Fig. 2), but interpretation became more dif- ficult as the increments became progressively smaller and as changes in growth patterns occurred in the otolith structure of the older specimens. There was no evidence of subdaily patterns in the incre- ments, and each increment was assumed to repre- sent 1 d. Support for the daily deposition of incre- ments was provided by data on three live juvenile sablefish held in the laboratory (Table 1). The specimens were captured by neuston net off Newport, OR, and transported to the laboratory where they were fed to satiation daily on Artemia. A check, apparently associated with capture and transfer to the laboratory, was evident on the otolith of each fish. The numbers of increments past this check corresponded closely to days captive; the minor differences are attributed to counting error and/or difficulty in interpretation of the check (Table .1). We thus consider the increments to be deposited with a daily periodicity. Hereafter increment counts will be equated with days after first increment for- mation; as we discuss later, first increment forma- tion may occur at first feeding. Data from the 1982-83 larval collections and the 1981 juvenile collections were fitted separately with simple linear regressions. Nonlinear curves (expo- nential, logistic, and Laird-Gompertz) were fitted to combined data with the NLIN procedure on the SAS- statistical package (SAS Institute, Inc. 1982). Table 1. — Growth and increment formation in captive specimens of Anoplopoma fimbria. L^, L2: standard length (mm) at capture and death, respectively. Capture date '-I h Days captive Increment past check Total increments 2 May 1983 17 May 1983 24 May 1983 19.9 14.1 53.8 60.4 53.4 109.7 31 40 32 31 38 33 63 95 87 Results and Discussion This study considers a total of 105 individuals, in- cluding 71 larvae and juveniles (9.8 to 41.2 mm SL) from the 1982 neuston collections, 21 juveniles (102.8 to 259.6 mm SL) from the 1981 purse seine collections, and 13 larvae (10.4 to 25.3 mm SL) taken in the 1983 neuston collections. Mean increment counts ranged from 9 increments for the youngest larva to 180 increments for the oldest juvenile. The abundance of larval sablefish in the neuston (Kendall and Clark footnote 1) at such young ages suggests that larvae move rapidly after hatching from the deep spawning region rather than early growth oc- curring at depths as suggested by Mason et al. (1983). The difference between the two increment counts for each otolith increased with increasing count, but the coefficient of variation remained the ^Reference to trade names does not imply endorsement by the Na- tional Marine Fisheries Service, NOAA. 477 same for the two ranges. For the larvae collected in 1982, with an overall mean of 30.8 increments, the mean difference was 1.67 increments (n = 71, stan- dard deviation (SD) = 1.45). For the 1981 juvenile collections, with overall mean of 109.5 increments, the mean difference between the two estimates was 6.57 increments {n = 21, SD = 5.03). Growth rates of field-collected larval and juvenile sablefish differ considerably. The data for the 1982 larval collections is described by the line SL = 0.375 (age, d) + 5.27 n = 71, r2 = 0.838, suggesting a mean growth rate for small larvae of 0.375 mnVd and an intercept of 5.27 mm, which coin- cides with the size of newly hatched larvae (Mason et al. 1983). Similarly the 1981 juvenile data is de- scribed by the line SL = 1.469 (age, d) - 0.926 n = 21, r2 = 0.822, tain of these growth differences may have been a function of gear selection. If net avoidance is a func- tion of fish size, as for most other planktonic organisms (Barkley 1972), then the oldest specimens taken in the neuston gear may have been only the slow-growing members of that cohort. Alternatively, interruptions of increment formation, resulting in underestimates of age, may occur. This has been observed for some species by Geffen (1982). In the laboratory specimens, however, one individual (L2 = 60.4 mm SL, Table 1) ceased eating for 5-6 d, became emaciated, and died. The last five incre- ments near the margin were smaller than the re- mainder, but the 1 : 1 correspondence of days to incre- ments suggests that increment formation continued. Estimated age-at-length data from all years were combined to describe the growth of sablefish to an age of about 200 d. Comparing exponential, logistic, and Laird-Gompertz growth models, the best fit (as judged by residual sums of squares) was provided by the Laird-Gompertz growth model (Fig. 3) in the form: suggesting a mean growth rate of 1.47 mm/d. Cer- L( = L^{AJa){l - exp(- at)) 280 Figure 3. -Estimated age at length for all Aru/plorpcmm fimbria in the study. Specimens taken in neuston nets (n = 84, including the 13 from 1983) are represented by circles, 1981 juvenile specimens from purse seine collections (n = 21) are represented by triangles. The equation and line represent the least squares fit of the Laird-Gompertz growth model. 20 40 60 80 100 120 140 160 180 200 MEAN AGE (DAYS) 478 where L, = standard length (mm) at age t (d), L,, = initial length (^-intercept), and A^^ and a are fitted parameters (Table 2). This sigmoid curve suggests relatively slow growth to an age of about 50 d and a length of about 25 mm SL, followed by rapidly accelerating growth through the juvenile stage, an inflection point at 113.2 mm, and an asymptotic length near 307.8 mm. Since sablefish achieve lengths to 100 cm (Hart 1973), these results should not be extrapolated beyond the ages in the present study. Also, the predicted fit of zero age individuals (Lq) is 1.22 mm SL (Table 2; Fig. 3). This value does not accurately reflect the length of sablefish at hatching. Egg size in sablefish ranges from 1.8 to 2.2 mm and newly hatched larvae are 5 to 6 mm (Mason et al. 1983). If daily increments are first laid down at first feeding as in some other species (Laroche et al. 1982), then this intercept is clearly an underestimate. Mean egg size suggests a length at first feeding of about 8 mm (Shirota 1970). The smallest larva taken in the pres- ent study was 9.8 mm SL (Fig. 3). This part of the curve may be related to the inclusion of the older, slower growing neustonic specimens. Another factor may be effects of shrinkage; small specimens were preserved in ethanol, older juveniles frozen. The magnitude of shrinkage for A. fimbria is unknown, but capture and preservation of other fish larvae causes shrinkage which decreases with increasing age or size (Theilacker 1980). Thus increases in ac- tual length for small individuals may have been relatively greater, changing the fitted equation and possibly increasing the length at time zero (Fig. 3). Heyamoto (1962) estimated growth for young sablefish, suggesting that specimens 12.3 to 16.4 cm FL (11.1 to 14.8 cm SL) were 6 mo old. His data, however, were based upon estimating the age at collection by difference between capture and an assumed spawning season. In our study, 6-mo-old specimens were > 24 cm SL. The specimens cap- tured by Heyamoto (1962) were taken by trawl in 320 to 412 m, much deeper than the epipelagic juveniles in our study. Beamish et al. (1983) used daily increments as part of a study to validate an- nulus formation in sablefish. In nine specimens 23 to 27 cm FL (208 to 245 mm SL), they observed from 270 to 350 (mean 313) increments but suggested that the fish were 1 yr old due to the inability to count all increments. Based upon our growth curve (Fig. 3), their ages would be overestimates. Recent observations of laboratory growth are in substantial agreement with growth described by our curve. Shenker and Olla^ found average growth rates as high as 2.3 mm/d for juvenile sablefish fed ad Table 2.— Fitted parameters of the Laird- Gompertz growth model for larval and juvenile Anoplopoma fimbria in the present study. The curve is fitted to all larvae and juveniles (W = 105) based upon counts of otolith increments. Parameter Estimate Asymptotic standard error 1.2203 0.1084 0.0196 0.4675 0.0146 0.0015 libitum. These fish were near the lengths where our curve predicts fastest growth (2 mnVd, Fig. 3). High growth rates were also observed for fish smaller than 25 mm, where our data suggest relatively slow growth. Grover and Olla'* noted starvation of field- collected sablefish larvae based upon morphological criteria; thus food probably limits sablefish growth in the field. This species apparently has a great scope for growth given high laboratory rations or patches of high prey density in the field. The distribution of dates of first increment forma- tion were estimated by back calculating from the ages of all specimens in our study. Since larvae and juveniles were from different years and sampling gears, it is possible that differences would be ob- served in this distribution. Since the plankton gear selects for smaller larvae due to avoidance by later stages, the results could be biased if the entire spawning season were not sampled. The median dates for the 1982 larvae (8 April) and the 1981 juveniles (18 March), however, were similar. Thus all 105 samples were combined and the distribution of the dates of first increment formation plotted (Fig. 4). The distribution has a mode in early April. If the first increment is formed in association with first feeding, as in most other species studied (Brothers et al. 1976; Taubert and Coble 1977; Laroche et al. 1982), then the spawning dates would precede the distribution in Figure 4. Ware (1975) provided an egg size-incubation time relationship for fishes; sablefish, with a 2 mm egg, would have an incubation time of 13 d. If a similar time is spent in yolk absorp- tion before first feeding, peak spawning would occur in early March. This generally agrees with most other reports of the spawning season for A. fimbria. ^Shenker, J., and B. L. 011a. Laboratory growth and feeding of ]uwem\e ssb\eT\s\ Ano^lcrpoma fimhris. Unpubl. manuscr. ■•Grover, J., and B. L. OUa. Field evidence for starvation of larval sablefish, Anoplopoma fimbria. Manuscr. in prep. Northwest and Alaska Fisheries Center, Newport Field Office, National Marine Fisheries Service, NOAA, c/o Marine Science Center, Marine Science Drive, Newport, OR 97365 (direct correspondence to B. L. 011a). 479 I/) z Ui 2 o LlI a. CD 2 r> z 15 FEBRUARY 20 25 I i 10 15 20 25 MARCH 30 I 10 15 20 25 APRIL 30 10 15 MAY Figure 4. -Distribution of dates of first increment formation for A noplopam-a fimbrin, determined by back-calculations using age and collection date. Phillips (1958) defined the peak spawning season off California to be January-February. Bell and Gharrett (1945) suggested that the spawning season was around December off Washington based upon fisher- men's observations and the presence of spent females in January. Farther north, Thompson (1941) observed ripe females and fertilized eggs in March at Cape St. James (lat. 51°45'N). More recent work has shown that the spawning season off British Columbia occurs in January to February with the peak of spawning in February (Mason et al. 1983). Our observed growth rates for A. fimbria during the first 6 months of life are high for a temperate- subarctic species, yet are clearly below the potential growth rate as shown in the laboratory (Shenker and 011a footnote 3). Similar but lower laboratory growth rates (1.2 mm/d) were observed for 100 to 150 mm juvenile red hake, Urophycis chuss, by Luczkovich and 011a (1983). Both of these species contrast mark- edly with larval juvenile growth in other taxa. Boeh- lert and Yoklavich (1983), for example, summarized laboratory and field growth measurements for 13 species in the genus Sebastes and noted growth rates ranging from 0.092 to 0.590 mm/d. Young sablefish thus utilize the neustonic and pelagic environment to rapidly reach sizes at which migration to the benthic adult habitat occurs. Acknowledgments This work was supported by the Northwest and Alaska Fisheries Center, National Marine Fisheries Service, Seattle, WA, through contracts 81-ABC- 00192-PR6 and 83-ABC-00045. We thank A. W. Kendall, Jr., W. G. Pearcy, and J. Shenker for pro- viding specimens for this study, Chris Wilson for technical assistance with otolith preparation, and R. Methot for reviewing the manuscript. Literature Cited Barkley, R. a. 1972. Selectivity of towed-net samplers. Fish. Bull., U.S. 70: 799-820. Beamish, R. J., and D. E. Chilton. 1982. Preliminary evaluation of a method to determine the age of sablefish {A noplopoma fimbria). Can. J. Fish. Aquat. Sci. 39:277-287. Beamish, R. J., G. A. McFarlane, and D. E. Chilton. 1983. Use of oxy tetracycline and other methods to validate a method of age determination for sablefish. In B. R. Meteff (editor). Proceedings of the International Sablefish Sympo- sium, p. 95-116. Alaska Sea Grant College Program, Univ. Alaska, Fairbanks, Sea Grant Rep. 83-8. Bell, F. H., and J. T. Gharrett. 1945. The Pacific coast blackcod, Anoplopoma fimbria. Copeia 1945:94-103. Boehlert, G. W. 1982. Ontogenetic changes in growth and their relationship with temperature and habitat change. In G. M. Cailliet and C. A. Simenstad (editors), GUTSHOP '81. Fish food habits studies, p. 115-123. Proceedings of the Third Pacific Work- shop, December 6-9, 1981, Pacific Grove, Calif.; Wash. Sea Grant Prog., Univ. Wash., Seattle, WSW-WO 82-2. Boehlert, G. W., and M. M. Yoklavich. 1983. Effects of temperature, ration, and fish size on growth of juvenile black rockfish, Sebastes melaruyps. Environ. Biol. Fishes 8:17-28. Brothers, E. B., C. P. Mathews, and R. Lasker 1976. Daily growth increments in otoliths from larval and adult fishes. Fish. Bull., U.S. 74:1-8. Geffen, a. J. 1982. Otolith ring deposition in relation to growth in herring (Clupea harengus) and turbot {Scopthalmus maximtts) larvae. 480 Mar. Biol. (Berl.) 71:317-326. Hart, J. L. 1973. Pacific fishes of Canada. Fish. Res. Board Can. Bull. 180, 740 p. Heyamoto, H. 1962. Age of younj; sablefish, AvopUrpmna fimbria (Pallas) 1811. J. Fish. Res. Board Can. 19:1175-1177. Laroche, J. L., S. L. Richardson, and A. A. Rosenberg. 1982. Age and growth of a pleuronectid, Paraphrya vetulus, during the pelagic larval period in Oregon coastal waters. Fish. Bull., U.S. 80:93-104. LuczKovicH, J. J., AND B. L. Olla. 1983. Feeding behavior, prey consumption, and growth of juvenile red hake. Trans. Am. Fish. Soc. 112:629-637. Mason, J. C. R. J. Beamish, and G. A. McFarlane. 1983. Sexual maturity-, fecundity, spawning, and early life historj' of sablefish {Anoplapoma fvmbrixj) off the Pacific coast of Canada. Can. J. Fish. Aquat. Sci. 40:2126-2134. Phillips, J. B. 1958. The fishery for sablefish, A noplopoma fimbria. Calif. Fish Game 44:79-84. Pruter, a. T. 1954. Age and growth of the Oregon sablefish, Anoplopoma fitnhrta. Pac. Mar. Fish. Comm. Bull. 3:121-128. Sameoto, D. D., and L. 0. Jaroszynski. 1969. Otter surface sampler: a new neuston net. J. Fish. Res. Board Can. 26:2240-2244. Shirota, a. 1970. Studies on the mouth size of fish larvae. Bull. Jpn. Soc. Sci. Fish. 36:353-368. SAS Institi'te, Inc. 1982. SAS users guide. SAS Institute, Inc., Raleigh, NC, 494 p. Tai'bert. B. D., and D. W. Coble. 1977. Daily rings in otoliths of three species oi Lepfrmis and Tilnpia monsambira. J. Fish. Res. Board Can. 34:332-340. Theilacker, G. H. 1980. Changes in body measurement of larval northern an- chovy, Engrnuii.'< mordaj-, and other fishes due to handling and preservation. Fish. Bull., U.S. 78:685-692. Thompson, W. F., Jr. 1941. A note on the spawning of the blackcod (Anoplopoma fmihria). Copeia 1941:270. Ware, D. M. 1975. Relation between egg size, growth, and natural mortal- ity of larval fish. J. Fish. Res. Board Can. 32:2503-2512. George W. Boehlert College of Oceanography and Marine Science Center Oregon State University Nevyport. OR 97365 Present address: Southwest Fisheries Center Honolulu Laboratory National Marine Fisheries Service, NOAA P.O. Box 3830 Honolulu, HI 96812 Mary M. Yoklavich College of Oceanography and Marine Science Center Oregon State University Newport, OR 97365 481 INFORMATION FOR CONTRIBUTORS TO THE FISHERY BULLETIN y Manuscripts submitted to the Fishei'y Bulletin will reach print faster if they conform to the following instructions. These are not absolute requirements, of course, but desiderata. CONTENT OF MANUSCRIPT The title page should give only the title of the paper, the author's name, his affiliation, and mailing address, in- cluding ZIP code The abstract should not exceed one double-spaced page. In the text. Fishery Bulletin style, for the most part, follows that of the U.S. Government Printing Office Style Manual. Fish names follow the style of the American Fisheries Society Special Publication No. 12, A List of Com- mon and Scientific Names of Fishes from the United States and Canada, Fourth Edition, 1980. Tfext footnotes should be typed separately from the text. Figures and tables, with their legends and headings, should be self-explanatory, not requiring reference to the text. Their placement should be indicated in the right-hand margin of the manuscript. Preferably figures should be reduced by photography to 5% inches (for single-column figures, allowing for 50% reduction in printing), or to 12 inches (for double-column figures). The maximum height, for either width, is 14 inches. Photographs should be printed on high quality or glossy paper. Do not send original drawings to the Scientific Editor; if they, rather than the photographic reductions, are needed by the printer, the Scientific Publications Office will request them. Each table should start on a separate page Consistency in headings and format is desirable. Vertical rules should be avoided, as they make the tables more expensive to print. Footnotes in tables should be numbered sequentially in arable numerals. To avoid confusion with powers, they should be placed to the left of numerals. Acknowledgments, if included, are placed at the end of the text. Literature is cited in the text as: Lynn and Reid (1968) or (Lynn and Reid 1968). All papers referred to in the text should be listed alphabetically by the senior author's sur- name under the heading "Literature Cited." Only the author's surname and initials are required in the literature cited. The accuracy of the literature cited is the respon- sibility of the author. Abbreviations of names of periodicals and serials should conform to Serial Sources for the BIOSIS Data Base. (Chemical Abfttracts also uses this system, which was developed by the American Standards Association.) Common abbreviations and symbols, such as mm, m, g, ml, mg, °C (for Celsius), %, "/qq, and so forth, should be used. Abbreviate units of measure only when used with numerals. Periods are only rarely used with abbreviations. We prefer that measurements be given in metric units; other equivalent units may be given in parentheses. FORM OF THE MANUSCRIPT The original of the manuscript should be typed, double- spaced, on white bond paper. Please triple space above headings. We would rather receive good duplicated copies of manuscripts than carbon copies. The sequence of the material should be: TITLE PAGE ABSTRACT TEXT LITERATURE CITED TEXT FOOTNOTES APPENDIX TABLES (Each table should be numbered with an arabic numeral and heading provided). LIST OF FIGURES (Entire figure legends) FIGURES (Each figure should be numbered with an arabic numeral; legends are desired) ADDITIONAL INFORMATION Send the ribbon copy and two duplicated or carbon copies of the manuscript to: Dr. William J. Richards, Scientific Editor Fishery Bulletin Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA 75 Virginia Beach Drive Miami, FL 33149-1099 Fifty separates will be supplied to an author free of charge and 50 supplied to his organization. No covers will be supplied. Contents— Cont "■ GASKIN, DAVID E., and ALAN P. WATSON. harbor porpoise, Phocoena pho- coena, in Fisli Harbour, New Brunswick, \ ' a: occupancy, distribution, and movements 427 Notes GROSSMAN, GARY D., MICHAEL J. HARRIS, and JOSEPH E. HIGHTOWER. The relationship between tiiefish, Lopholatilus chamaeleonticeps, abundance and sediment composition off Georgia 443 MATARESE, ANN C., and BEVERLY M. VINTER. The development and occur- rence of larvae of the longfin Irish lord, Hemilepidotus zapus (Cottidae) 447 POLOVINA, JEFFREY J., and MARK D. OW. An approach to estimating an eco- system box model 457 SEDBERRY, GEORGE R. Food and feeding of the tomtate, Haemulon aurolineatum (Pisces, Haemulidae), in the South Atlantic Bight 461 HINES, ANSON H., KENRIC E. OSGOOD, and JOSEPH J. MIKLAS. Semilunar reproductive cycles in Fundulus heteroclitus (Pisces: Cyprinodontidae) in an area without lunar tidal cycles 467 HUI, CLIFFORD A. Undersea topography and the comparative distributions of two pelagic cetaceans 472 BOEHLERT, GEORGE W., and MARY M. YOKLAVICH. Larval and juvenile growth of sablefish, Anoplopoma fimbria, as determined from otolith increments 475 • GPO 593-096 K^^^'^'Co, Fishem Bulletin "^^IrES O^ ^ ''EB 7 1986 ^!^H2E3E!!ZMl5s: Vol. 83, No. 4 October 1985 PARRISH, R. H., D. L. MALLICOATE, and K. F. MAIS. Regional variations in the growth and age composition of northern anchovy, Engraulis mordax 483 JOHNSON, PHYLLIS T. Parasites of benthic amphipods: microsporidans oiAmpelisca agassizi (Judd) and some other gammarideans 497 OVERHOLTZ, WILLIAM J., and ALBERT V. TYLER. Long-term responses of the demersal fish assemblages of Georges Bank 507 WAHLEN, BRUCE E., and TIM D. SMITH. Observer effect on incidental dolphin mor- tality in the eastern tropical Pacific tuna fishery 521 SINGER, MICHAEL M. Food habits of juvenile rockfishes (Sebastes) in a central Califor- nia kelp forest 531 READ, ANDREW J., and DAVID E. GASKIN. Radio tracking the movements and activities of harbor porpoises, Phocoena phocoena (L.), in the Bay of Fundy, Canada 543 HOHN, ALETA A., and P S. HAMMOND. Early postnatal growth of the spotted dolphin, Stenella attenuata, in the offshore eastern tropical Pacific 553 BROWN, R. S., and N. CAPUTI. Factors affecting the growth of undersize western rock lobster, Panulirus cygnus George, returned by fishermen to the sea 567 JAMIESON, G. S., and A. CAMPBELL. Sea scallop fishing impact on American lobsters in the Gulf of St. Lawrence 575 WARLEN, STANLEY M., and ALEXANDER J. CHESTER. Age; growth, and distribu- tion of larval spot, Leiostomus xanthurus, off North Carolina 587 ALBERS, W D., and P J. ANDERSON. Diet of Pacific cod, Gadiis macrocephalus, and predation on the northern pink shrimp, Pandalus borealis, in Pavlof Bay, Alaska 601 BOEHLERT, GEORGE W, DENA M. GADOMSKI, and BRUCE C. MUNDY. Vertical distribution of ichthyoplankton off the Oregon coast in spring and summer months 611 AU, DAVID W. K., and WAYNE L. PERRYMAN. Dolphin habitats in the eastern tropical Pacific 623 FREEMAN, MARY C, NATE NEALLY, and GARY D. GROSSMAN. Aspects of the life history of the fluffy sculpin, Oligocottus snyderi 645 BARLOW, JAY. Variability, trends, and biases in reproductive rates of spotted dolphins, Stenella attenuata 657 {Continued on back cover) Seattle, Washington U.S. DEPARTMENT OF COMMERCE Malcolm Baldrlge, Secretary NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION John V. Byrne, Administrator NATIONAL MARINE FISHERIES SERVICE William G. Gordon, Assistant Administrator 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. Separates were issued as documents through volume 46; the last document was Na 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. Government Printing Office, Washington, DC 20402. It is also available free in limited numbers to libraries, research institutions, State and Federal agencies, and m exchange for other scientific publications. SCIENTIFIC EDITOR, Fishery Bulletin Dr. William J. Richards Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA Miami, FL 33149-1099 Editorial Committee Dr. Bruce B. Collette National Marine Fisheries Service Dr. Donald C. Malins National Marine Fisheries Service Dr. Edward D. Houde Chesapeake Biological Laboratory Dr. Merton C. Ingham National Marine Fisheries Service Dr. Reuben Lasker National Marine Fisheries Service Dr. Jerome J. Pella National Marine Fisheries Service Dr. Jay C. Quast National Marine Fisheries Service Dr. Carl J. Sindermann National Marine Fisheries Service Mary S. Fukuyama, Managing Editor The Fishery Bulletin (ISSN 0090-0656) is published quarterly by the Scientific Publications Office, National Marine Fisheries Service; NOAA, 7600 Sand Point Way NE, BIN C15700, Seattle, WA 98115. Second class postage is paid in Seattle, Wash., and additional offices. POSTMASTER send address changes for subscriptions to Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. Although the contents have not been copyrighted and may be reprinted entirely, reference to source is appreciated. 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 Manage- ment and Budget through 1 April 1986. i Fishery Bulletin CONTENTS Vol. 83, No. 4 October 1985 PARRISH, R. H, D. L. MALLICOATE, and K. F. MAIS. Regional variations in the growth and age composition of northern anchovy, Engraulis mordax 483 JOHNSON, PHYLLIS T. Parasites of benthic amphipods: microsporidans of Ampelisca agassizi (Judd) and some other gammarideans 497 OVERHOLTZ, WILLIAM J., and ALBERT V. TYLER. Long-term responses of the demersal fish assemblages of Georges Bank 507 WAHLEN, BRUCE E., and TIM D. SMITH. Observer effect on incidental dolphin mor- tality in the eastern tropical Pacific tuna fishery 521 SINGER, MICHAEL M. Food habits of juvenile rockfishes (Sebastes) in a central Califor- nia kelp forest 531 READ, ANDREW J., and DAVID E. GASKIN. Radio tracking the movements and activities of harbor porpoises, Phocoena phocoena (L.), in the Bay of Fundy, Canada 543 HOHN, ALETA A., and P S. HAMMOND. Early postnatal growth of the spotted dolphin, Stenella attenuata, in the offshore eastern tropical Pacific 553 BROWN, R. S., and N. CAPUTI. Factors affecting the growth of undersize western rock lobster, Panulirus cygmis George, returned by fishermen to the sea 567 JAMIE SON, G. S., and A. CAMPBELL. Sea scallop fishing impact on American lobsters in the Gulf of St. Lawrence 575 WARLEN, STANLEY M., and ALEXANDER J. CHESTER. Age, growth, and distribu- tion of larval spot, Leiostomus xanthurus, off North Carolina 587 ALBERS, W. D., and P J. ANDERSON. Diet of Pacific cod, Gadus macrocephalus, and predation on the northern pink shrimp, Pandalus borealis, in Pavlof Bay, Alaska 601 BOEHLERT, GEORGE W, DENA M. GADOMSKI, and BRUCE C. MUNDY. Vertical distribution of ichthyoplankton off the Oregon coast in spring and summer months 611 AU, DAVID W. K., and WAYNE L. PERRYMAN. Dolphin habitats in the eastern tropical Pacific 623 FREEMAN, MARY C, NATE NEALLY, and GARY D. GROSSMAN. Aspects of the life history of the fluffy sculpin, Oligocottus snyderi 645 BARLOW, JAY. Variability, trends, and biases in reproductive rates of spotted dolphins, Stenella attenuata 657 (Continued on next page) ; iV,b!ine Biotogfea! LahoratOfv f FEB 7 1986 j Seattle, Washington i I 1985 LJ^(2£^ Hole, Mass. { For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington DC 20402— Subscription price per year: $21.00 domestic and $26.25 foreign. Cost per single issue: $6.50 domestic and $8.15 foreign. Contents— Continued Notes PETERSON, CHARLES H, R BRUCE DUNCAN, HENRY C. SUMMERSON, and BRIAN F. BE AL. Annual band deposition within shells of the hard clam, Mercenaria mercenaria: consistency across habitat near Cape Lookout, North Carolina 671 SULLIVAN, LORETTA E, DENNIS A. EMILIANI, and K. NEAL BAXTER. Stand- ing stock of juvenile brown shrimp, Penaeus aztecus, in Texas coastal ponds 677 THOMAS, DAVID H. A possible link between coho (silver) salmon enhancement and a decline in central California Dungeness crab abundance 682 FFLDKAMP, STEVEN D. The effects of net entanglement on the drag and power out- put of a California sea lion, Zaloiphus califomiamts 692 PARSONS, GLENN R. Notes on the life history of the catshark, Scyliorhinus meadi 695 LIBBY, DAVID A. A comparison of scale and otolith aging methods for the alewife, Alosa pseudoharengus 696 MAULE, ALEC G., and HOWARD F. HORTON. Probable causes of the rapid growth and high fecundity of walleye, Stizostedion vitreum vitreum, in the mid-Columbia River 701 TAYLOR, D. M., R. G. HOOPER, and G. R ENNIS. Biological aspects of the spring breeding migration of snow crabs, Chionoecetes opilio, in Bonne Bay, Newfoundland (Canada) 707 CREED, ROBERT P., JR. Feeding, diet, and repeat spawning of blueblack herring, Alosa aestivalis, from the Chowan River, North Carolina 711 Index 717 Notices NOAA Technical Reports NMFS published during first 6 months of 1985. The National Marine Fisheries Service (NMFS) does not approve, recommend or en- dorse any proprietary product or proprietary material mentioned in this publication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales promotion which would indicate or imply that NMFS approves, recommends or endorses any proprietary product or proprietary material mentioned herein, or which has as its purpose an intent to cause directly or indirectly the adver- tised product to be used or purchased because of this NMFS publication. REGIONAL VARIATIONS IN THE GROWTH AND AGE COMPOSITION OF NORTHERN ANCHOVY, ENGRAULIS MORDAX R. H. Parrish,! D. L. Mallicoate,! and K. F. Mais^ ABSTRACT Data from the Sea Survey Program conducted by the California Department of Fish and Game were analyzed to describe regional variations in growth and age composition of northern anchovy, Engraulis mordax. Juvenile growth appeared to be greater at higher latitudes and in the offshore portion of the Southern California Bight. Adult growth was less variable; however, there were statistically significant differences between regions. In addition, the growth rate for the southern stock was markedly lower than that of the central stock. This difference in growth rates was used to characterize the area of overlap between the two stocks. Age composition varied with depth of water and geographical location within the Southern California Bight and with latitude Young-of-the-year and yearling fish were found in larger proportions in shallow water and in the southern and inshore areas of the Southern California Bight. Age compositions of northern anchovies sampled in the California and Mexican purse seine fisheries were compared with those of the Sea Survey Program. This comparison suggests that the present California area restrictions, which exclude the fishery from the nearshore area, greatly reduces the number of young fish in the catch. The purpose of this study is to describe regional variations in growth and age composition of the northern anchovy, Engraulis mordax. Data for the study were taken by the Sea Survey Program of the Cahfornia Department of Fish and Gama Mais (1974) described this program and analyzed the data for the most common species taken in the survey. This report is an extension of Mais' work and focuses on geographical variations in age composition and growth rates and on depth variations in age composition. Meristic and morphometric (McHugh 1951) and electrophoretic (Vrooman et al. 1981) studies on the stock structure of the northern anchovy suggest that there are three stocks (northern, central, and southern) and that the boundaries between the stocks occur in central California and central Baja California. There is also recent evidence (Parrish^) of a fourth stock which spawns in the fall in central California and in the northern and offshore areas of the Southern California Bight. Mais (1974) showed that the southern stock of northern anchovies was smaller at age than the cen- tral stock. In addition, northern anchovies are known 'Southwest Fisheries 'Center Pacific Fisheries Environmental Group, National Marine Fisheries Service, NOAA, P.O. Box 831, Monterey, CA 93942. ^California Department of Fish and Game, 1301 West 12th Street, Long Beach. CA 90813. ^Parrish, R. H. 1983 Evidence for a fall spawning anchovy stock. Paper presented at 1983 CalCOFI Conference Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. to be larger off central California than off southern California (Collins 1969; Mais 1974; Mallicoate and Parrish 1981), and they are larger in the offshore areas of the Southern California Bight than in the inshore areas (Mais 1974). These differences could be due to varying growth rates between regions, varying seasonality of spawning, varying age com- positions, size-specific migration, or a combination of these factors. Tkgging experiments have shown that northern an- chovies move from southern California to central California, from central California to southern California, and from southern California to Ensenada, Mexico; there is a northerly movement in summer and a southerly movement in winter (Haugen et al. 1969). Mais (1974) found northern an- chovies to be distributed more offshore in some years and more inshore in other years, and he found them concentrated closer to shore and in the northern part of the Southern California Bight during the fall months. Mais (1974) suggested that northern an- chovies begin an offshore and southeastward move- ment in late winter, which coincides with the onset of major spawning activity. These movements of anchovy may affect the measurement of growth rates and age compositions within the different regions. METHODS The data used in the study were taken from north- 483 FISHERY BULLETIN: VOL. 83. NO. 4 ern anchovies caught by midwater trawl. The gear and sampling procedures are described by Mais (1974). The data set covers the period 1966-1983 and consists of 101 cruises. Twenty-three cruises extend- ed north of Point Conception, 77 cruises occurred in southern California and northern Baja California, and 8 cruises extended into southern Baja Califor- nia. Several cruises extended into more than one region. There were a total of 4,166 trawl hauls, of which 3,017 contained anchovies. Standard lengths were normally taken from about 25 anchovies in each trawl haul in which they occurred; otoliths, for aging, were usually taken from a subsample of up to 10 fish. A total of 60,082 northern anchovies were measured, of which 20,772 were aged by California Department of Fish and Game personnel with methods developed by Collins and Spratt (1969). For purposes of determining age composition and growth rates of northern anchovies, it was assumed that February was the birth month of all fish sam- pled. Anchovies off California and Baja California have a peak in spawning in February- March; however, some spawning does occur all year (Ahlstrom 1966). The age determinations used in this report are, of course, not accurate to the month. They are based on the number of annuli, the seasonality of annuli formation, an assumed February birth month, and the month the fish were caught. Annuli formation occurs in May in California (Collins and Spratt 1969). A 1-yr-old anchovy would therefore be an anchovy caught in February with no annuli on its otoliths. A lV2-yr-old anchovy would be a fish with one an- nuli which was caught in August. An anchovy born in February would be 15 mo old when its first an- nuli was formed. An anchovy spawned in September would be 8 mo old if it formed its first annuli in its first May and 20 mo old if it formed its first annuli in its second May. Differences in growth between geographical regions were compared by linear regression analyses which included anchovies of 1 V2 yr of age and older. Evaluation of the relationship between age composi- tion and water depth in the area of capture was made by grouping the samples into depth strata. The strata were established partially on the number of observations. lb determine growth rates, the data were process- ed with a computer program (Mallicoate and Par- rish 1981) which calculates and plots the mean length, standard deviation of length, and length range by age and month. The program also tabulated the number of observations by age and month; these data were used for determining age composition. Our preliminary analysis of the sea survey data showed an alongshore cline in the mean length of northern anchovies (Fig. 1). It also revealed a bias in the selection of fish to be aged. Aged fish were generally larger at all latitudes than were the un- aged fish. This bias is apparently due to a consis- tent tendency for samplers to pick larger anchovies for the subsample which was aged. Anchovies < 100 mm SL were particularly susceptible to not being selected for aging (Tkble 1). On checking with the field biologists who took the data, we found a second source of bias which occurred only in trawl hauls of exclusively small fish. When trawl hauls were con- sidered by the sampler to be "obviously" all young- of-the-year fish, there was a tendency not to take otoliths for age determination. These sampling biases affect the analyses of age composition presented in this report but do not affect the grow^th analyses. Table 1.— Two types of length bias in sampling northern anchovies in the Sea Survey Program. No. fish Standard measured in No. length No. fish hauls sampled otolith (mm) measured for otoliths samples iBias 1 2Bias 2 <70 2,174 1,275 460 0.59 0.36 70-80 2,366 1,939 684 0.82 0.35 80-90 4,667 4,241 1,667 0.91 0.39 90-100 7,077 6,367 2,498 0.90 0.39 100-110 8,988 7,875 2,997 0.88 0.38 110-120 12,058 10,834 4,164 0.90 0.38 120-130 1 1 ,744 10,462 4,215 0.89 0.40 130-140 7,390 6,456 2,748 0.87 0.43 140 + 3,619 3,036 1,339 0.84 0.44 Total 60,082 52,484 20,772 'Bias 1 is the decision to sample for otoliths (i.e., the proportion of fish in trawl hauls which were sampled for otoliths). ^Bias 2 is the selection of larger fish by the sampler (i e., the proportion of fish in trawl hauls, which were sampled for otoliths, for which otolith samples were taken). 484 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY 140- FlGURE 1.— Mean length of aged and unaged northern anchovies and the number of an- chovies by half degree of latitude 30 LATITUDE The data were inadequate to calculate growth curves or age composition on the one-half degree of latitude interval used in Figure 1; therefore, geo- graphical regions were selected based partially on the number of observations. For example, the southern and central Baja California and central California areas, which had fewer samples, were more widely spaced than the southern California and northern Baja California areas. Nearly all of the an- chovy samples taken south of lat. 32°N or north of lat. 34°N were taken within one-half degree of the coast. The lat. 32°-34°N area included a large amount of samples taken further than one-half degree from the coast. This area, the Southern California Bight, contains numerous islands and basins; we, therefore, divided it into regions which approximate the natural basins described in Emery (1954) (Fig. 2). RESULTS Growth Earlier studies on northern anchovies from British Columbia (Pike 1951), northern California (Waldvogel 1977), central California (Clark and Phillips 1952), and southern California (Spratt 1975) showed considerable variation in their growth (Fig. 3A). Anchovies in British Columbia, lat. 49°N, and northern California, lat. 41°N, are of the northern stock (Vrooman et al. 1981) and have a summer spawning season. Anchovies in southern California, lat. 33 °N, are of the central stock, whereas those in central California, lat. 36°N, are considered to be a mixture of northern and central stocks (Vrooman et al. 1981); anchovies in both southern and central California have a late winter (February-April) spawn- ing season (Parrish footnote 3). Anchovies in British Columbia were the fastest growing of all the four areas in their first year of life, whereas anchovies in northern California were the slowest. Anchovies in central and southern California showed similar growth rates after their first year of life; however, the growth rate of anchovies in central California was greater than the growth rate of those in southern California in their first year. Our study shows that the growth of anchovies has a distinct geographical pattern. Anchovies sampled in the Central California region (CC) and the off- shore area of the Southern California Bight [i.&, San Nicolas (SN) and Tknner and Cortez Banks (TCB) regions] have the fastest juvenile growth (Fig. 3). An- chovies in these areas attain an average length of 120 mm before they are IV2 yr old. In the inshore areas of the Southern California Bight and in Baja California there is a continuous decline in the growth rate associated with decreasing latitude (Figs. 3, 4). Anchovies reach a mean length of 120 mm at about age 2 in the Santa Barbara Channel region (SBC) 485 FISHERY BULLETIN: VOL. 83, NO. 4 NO" 125" 120" II5» 110" Figure 2.— Geographical regions for which the growth and age composition of northern anchovies were determined. and in the remaining offshore region, Catalina Basin (CB). In the San Pedro Channel (SPC), Coronado Escarpment (CE), and Ensenada (E) regions an- chovies reach 120 mm at about age 3. In the Cape San Quentin (CSQ) and Sebastian Viscaino Bay (SVB) regions anchovies reach 120 mm at about age 4 or later. The Cape San Quentin (CSQ), Sebastian Viscaino Bay (SVB), and Southern Baja California (SB) regions include anchovies from both the central and southern stocks, and there are marked differences in their grov^h (Fig. 4). Anchovies from the southern stock appear to reach an asymptotic mean size of about 92 mm, whereas those from the central stock continue to grow throughout their lives. Note the oc- currences of 2- to 4-yr-old fish with monthly mean lengths of about 92 mm (Fig. 4 CSQ, SVB, SB). Assuming that stocks can be identified by size at age, the Cape San Quentin region appears to be dominated by the central stock; however, the 486 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY 3 4 5 nCE (YERRS) 3 4 RGE (YERRS Figure 3— Age-length relationships of northern anchovies taken in A. Earlier studies 1) British Columbia (Pike 1951), 2) Central Califor- nia (Clark and Phillips 1952), 3) Northern California (Waldvogel 1977), and 4) Southern California (Spratt 1975), CC. Central Califor- nia; SN, San Nicolas; TCB, Tknner and Cortez Banks; CB, Catalina Basin; SBC, Santa Barbara Channel. 487 FISHERY BULLETIN: VOL. 83, NO. 4 3 4 5 AGE (YEARS) 3 4 5 RGE (YEARS) Figure 4.— Age-length relationships of northern anchovies taken in SPC, San Pedro Channel; CE, Coronado Escarpment; E, Ensenada; CSQ , Cape San Quentin; SVB, Sebastian Viscaino Bay; and SB, Southern Baja California. 488 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY southern stock does extend into this region. The most northerly trawl sample which could be iden- tified, by size at age, as southern stock was a June 1971 sample taken at lat. 30.3°N. Southern and cen- tral stock anchovies broadly overlap in the Sebas- tian Viscaino Bay region; however, the region is occupied principally by the southern stock in the summer and by the central stock in the fall and winter. The Southern Baja California region is dominated by the southern stock; however, central stock anchovies, as identified by size at age, were taken as far south as lat. 26.5°N in November 1967. The fact that the central stock is the farthest south in winter and the southern stock the farthest north in summer suggests that the separation of the stocks is aided by different environmental preferences, possibly temperature The Central California region is an area of overlap between the northern and central anchovy stocks. In addition, as previously mentioned (Parrish foot- note 3), a fall spawning stock may occur in central California and the offshore areas of the Southern California Bight. Vrooman et al. (1981) did not have any samples south of Monterey in central Califor- nia or from the offshore and northern areas of the Southern California Bight. Samples from these areas may consist of a mixture of different stocks with dif- ferent spawning seasons. Our assumption of a February birth month in these areas must therefore be evaluated. Regression Analysis Growth in length of anchovies in the size range sampled by the Sea Survey Program (i.&, 40-180 mm SL) can be divided into two phases. The early, juvenile phase extends until the fish are about 1 yr old. Methot (1981) found that in the Southern California Bight juvenile growth averages about 9 mm/mo, and it is at a maximum when the fish are between 40 and 50 mm SL. The second, adult phase extends from a little over 1 yr old until death. Growth during this phase is more than an order of magni- tude less than the early phase (i.e, 0.48-0.69 mm/mo). Examination of the age-length relationships (Figs. 3, 4) in the central stock shows that growth in the adult phase is essentially linear. Therefore, we used linear regressions to compare growth of anchovies from different regions. These analyses included only fish of IV2 yr of age and older. This model has several advantages for comparing the growth dur- ing the two phases described above. By limiting the model to the linear segment of the age-length rela- tionships, the slopes of the regressions can be used to compare the regional variations in growth rate during the adult phase Another advantage is that the expected length at 172 yr of age can be used to compare regional variation in growth during the juvenile phase Juvenile Growth In the area occupied by the central stock, growth during the juvenile phase shows considerable varia- tion among regions. There were significant dif- ferences in growth to age IV2. Fastest growth oc- curred in the north, and the slowest was in the south (Tkble 2, Fig. 5). Mean standard length at IV2 yr of age was 123.6 mm in the Central California region, 113.4 mm in the San Pedro Channel region, and 103.6 mm in the Cape San Quentin region. The three southernmost Baja California regions contain mix- tures of the central and southern stocks; therefore, data from these regions were divided into central and southern stock sets, based on length and month, for the analyses. In contrast to those of the central stock, anchovies of the southern stock attain only 91.8 mm by age IV2. Within the Southern Califor- nia Bight there is also an inshore-offshore gradient, with faster juvenile growth in the offshore regions and slower grov^th in the inshore regions (i.e, lengths at age IV2 in the Coronado Escarpment, Catalina Basin, and Tanner and Cortez Banks regions were 110.5, 116.4, and 119.2 mm respectively). An analysis of covariance showed that the dif- ferences in size at age IV2 are not significantly dif- ferent {a = 0.01) in two pairs of regions. The San Nicolas region (120.4 mm at age IV2) and the Ikn- ner and Cortez Banks region (119.2 mm) are the first pair, and the Catalina Basin region (116.4 mm) and the Santa Barbara Channel region (116.2 mm) are the second pair. In all other pairs of regions, size at age IV2 is significantly different at the a = 0.0001 level. Adult Growth The regressions demonstrate that growth during the adult phase is relatively constant between the different regions occupied by the central stock, with slopes varying from a low of about 6 mm/yr to a high of about 8 mm/yr (Ikble 2). In contrast, anchovies of the southern stock have essentially no adult growth after age IV2 (i.e, slope = 0.07 mm/yr). Although the differences in adult growth between regions is not large, there are statistically significant differences (Tkble 3). The San Nicolas region and the Tknner and Cortez Banks region have adult growth 489 FISHERY BULLETIN: VOL. 83, NO. 4 Table 2.— Regression parameters and statistics for the relationship between age and length of northern anchovies older than IV2 yr: Central California (CC), San Nicolas (SN), Tanner-Cortez Banks (TCB), San Nicolas combined with Tanner-Cortez Banks (SN+TCB), Catalina Basin (CB), Santa Barbara Channel (SBC), Catalina Basin com- bined with Santa Barbara Channel (CB+SBC), San Pedro Channel (SPC), Coronado Escarpment (CE), Ensenada (E), Cape San Quentin (CSQ), Sebastian Viscaino Bay and Southern Baja California, central stock only (SVB+SB), and Southern Baja Califor- nia, southern stock only (S). Mean Mean Length at Adult SD of SDof age length age IV2 growth length at adult Area (Yr) (SL mm) (SL mm) (mm/yr) r age Vh growth N Central Stock CC 3.4 138.4 123,6 7.75 0.702 1,563 0.431 335 SN 3.3 131.1 120.4 5.97 0.613 0,713 0.205 1,410 TCB 3.2 130.8 119.2 6.69 0.623 0,975 0.286 860 SN+TCB 3.3 131.0 120.0 6.23 0.616 0,577 0.167 2,270 CB 3.1 126.7 116.4 6.29 0.582 0,633 0.192 2,092 SBC 2.9 125.6 116.2 6.83 0.636 0,689 0.224 1,374 CB-hSBC 3.0 126.3 116.4 6.50 0.606 0,465 0.145 3,466 SPC 2.8 121,3 113.4 6.07 0.575 0,435 0.146 3,497 CE 2.7 118.7 110.5 6.89 0.569 0.666 0.235 1,798 E 2.8 116.5 106.2 8.08 0.636 0.944 0.321 934 CSQ 2.5 111.1 103.6 7.54 0.649 0.776 0.291 923 SVB+SB 2.4 106.4 101.1 5.76 0,443 1.571 0.618 358 Southern Stock 8' 2.9 91.9 91.8 0.07 0.009 1.274 0.421 335 'Combined from three areas: CSQ, SVB, and SB. 150- 140- J 130 H 5 120 H o ^ 110 H 100- 90- 3 AGE 6 Figure 5.— Linear regressions showing the expected length at age (Tkble 1) of northern anchovies taken in CC, Central California; SN, San Nicolas; TCB, Tknner and Cortez Banks; CB, Catalina Basin; SBC, Santa Barbara Channel; SPC, San Pedro Channel; CE, Coronado Escarpment; E, Ensenada; CSQ, Cape San Quentin; SVB + SB, Se- bastian Viscaino Bay and Southern Baja California, central stock only; and S, Southern Baja California, southern stock only. 490 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY patterns which are not statistically different (a = 0.01), and since the two regions are adjacent we have considered them together. Also there is no signi- ficant difference (a = 0.01) in adult growth in the Catalina Basin and the Santa Barbara Channel regions, but since these regions are geographically separated we have retained them as separate regions. The relatively large standard deviations of the regression slopes (Ikble 2) for fish from the Cen- tral California and the Sebastian Viscaino Bay and the Southern Baja California regions may indicate that these regions are the most likely to have mix- tures of more than one stock. There was no significant difference {a = 0.01) be- tween growth of adults in the best growth regions (Ensenada, Cape San Quentin, and Central Califor- nia). The difference between the Coronado Escarp- ment and the Santa Barbara Channel regions is not significant at the a = 0.1 level nor are the differences between the Catalina Basin, combined San Nicolas- Tknner and Cortez Banks, San Pedro Channel, and combined Sebastian Viscaino Bay and Southern Baja California regions. Tknner and Cortez Banks, and Catalina Basin) dif- fer from the other regions in that they are dominated by fish 3 yr and older (Fig. 6A, Tkble 4). Central California and Tknner and Cortez Banks also have substantial percentages of young-of-the-year fish, whereas the other two regions do not. Anchovies in the three inshore regions of the Southern California Bight (Santa Barbara Channel, San Pedro Channel, and the Coronado Escarpment) have very similar age compositions (Fig. 6B, Tkble 4). One- and two-yr-old fish are the most numerous in these three regions. Young-of-the-year and 3 yr- olds are slightly less abundant than 1- and 2-yr-olds, and there are fewer older fish in the samples. Baja California is characterized by a dominance of young fish (Fig. 6C, Ikble 4). The age composi- tions in the Ensenada, Cape San Quentin, and Sebas- tian Viscaino Bay regions are very similar; young- of-the-year and 1-yr-old fish are the most abundant, and there is rapid decrease in the abundance of fish with increasing age. Anchovies in southern Baja California are principally from the southern stock. The age composition in this region shows a pre- Table 3. — Significance levels for differences in growth of adult northern anchovies from different geographical regions: Ensenada (E), Central California (CC), Cape San Quentin (CSQ), Coronado Escarpment (CE), Santa Barbara Chan- nel (SBC), Tanner and Cortez Banks (TCB), Catalina Basin (CB), San Nicolas combined with Tanner and Cortez Banks (SN+TCB), San Pedro Channel (SPC), San Nicolas (SN), Sebastian Viscaino Bay and Southern Baja California, central stock only (SVB+SB), and Southern Baja California, southern stock only (S). Geographic areas are arranged in order from largest to smallest slopes. Area E CC CSQ CE SBC TCB CB SN+TCB SPC SN SVB+SB S E CC 0.5573 CSQ 0.2217 0.6517 — CE 0.0019 0.0514 0.0754 — SBC 0.0015 0.0477 0.0629 0.9422 — TCB 0.0013 0.0331 0.0384 0.6516 0.6951 — CB 0.0000 0.0000 0.0000 0.0605 0.0628 0.2400 — SN+TCB 0.0000 0.0003 0.0002 0.0283 0.0280 0.0000 0.8141 — SPC 0.0000 0.0000 0.0000 0.0015 0.0033 0.0465 0.3560 0.4659 — SN 0.0000 0.0000 0.0000 0.0042 0.0042 0.0367 0.2547 0.3234 0.6902 — SBV + SB 0.0000 0.0085 0.0051 0.0826 0.0952 0.1616 0.3943 0.4372 0.6048 0.7379 — 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 — Age Composition by Area Geographical variation in age composition is one factor which could influence the observed variation in mean size of anchovies in the various regions; therefore, age composition was calculated for each of the regions used earlier (Fig. 2). Few anchovies < 60 mm SL occur in the data, and it appears that young-of-the-year fish are apparently not susceptible to capture by the midwater trawl gear used in the sea survey until they are about 6 mo old. Central California and the three offshore regions within the Southern California Bight (San Nicolas, ponderance of 1-yr-olds, and like the other Baja California regions there are few fish of age 4 or older. Age Composition by Depth Strata Anchovies primarily live within the upper mixed layer; they occur in the surface layer over the con- tinental shelf and over deepwater regions. Mais (1974) showed that the average size of anchovies sampled in offshore areas was greater than that of anchovies sampled near the coast. The percentage of trawl hauls containing anchovies was quite con- stant in areas with different water depths, varying 491 FISHERY BULLETIN: VOL. 83, NO. 4 50- 40- 30- B 1^ 0 1 — I — I — 2 3 4+ 0 I r I 2 AGE n — I — 3 4 + 1^ 0 n 1 r 2 3 4+ Figure 6.— Age composition of northern anchovies taken in the following geogra- phical regions: A. Central California and offshore areas of the Southern California Bight; Central California (CC), San Nicolas (SN), Tknner and Cortez Banks (TCB), and Catalina Basin (CB); B. Inshore areas of the Southern California Bight; Santa Barbara Channel (SBC), San Pedro Channel (SPC), and Coronado Escarpment (CE); C. Baja California; Ensenada (E), Cape San Quentin (CSQ), Sebastian Viscaino Bay (SVB), and Southern Baja California (SB). Table 4. — Percentage age composition of northern an- chovies by geographical region: Central California (00), San Nicolas (SN), Tanner-Cortez Banks (TOB), Oatalina Basin (OB), Santa Barbara Channel (SBC), San Pedro Channel (SPO), Coronado Escarpment (OE), Ensenada (E), Cape San Quentin (CSQ) Sebastian Viscaino Bay (SVB), Southern Baja California (SB). Age No. fish Area 0 1 II III IV V Vl-i- CC 19.3 17.2 14.7 23.9 17.7 5.7 1.6 436 SN 5.6 15.5 27.7 26.0 15.9 7.4 2.0 1,721 TOB 12.6 12.6 24.3 28.3 13.6 6.5 2.1 1,136 OB 3.7 18.7 29.2 29.2 12.8 5.2 1.1 2,538 SBC 21.8 29.6 22.2 15.2 8.2 2.4 0.6 1,989 SPO 21.6 28.6 25.4 15.6 6.3 1.9 0.5 5,439 OE 23.5 28.1 26.9 14.6 5.4 1.3 0.2 2,965 E 25.8 23.6 23.4 17.1 7.6 2.3 0.2 1,464 CSQ 33.8 33.7 15.5 11.2 4.4 1.2 0.1 1,779 SVB 38.1 30.8 12.8 13.5 4.5 0.3 — 896 SB 18.8 45.7 19.6 11.5 3.7 0.7 — 409 between 64 and 77% (Ikble 5). This is probably in- fluenced by the fact that trawling (in the upper 30 m) was carried out normally in areas where fish schools had already been detected by depth recorder or sonar. In southern California (lat. 32°-34°N) young-of-the- year anchovies comprise more than one-half of the anchovies sampled in the 5-25 fathom-depth stratum (Tkble 6A). One- and two-yr-olds comprise more than one-half of the anchovies in the 26-50, 51-150, 151-300, and 301-500 fathom strata. The most abun- dant age group in the 26-50 stratum is age 1; there are essentially equal numbers of age 1 and age 2 an- chovies in the 51-150 and 151-300 strata; 2-yr-olds are the most abundant age group in the 301-500 stratum; and 3-yr-olds are the most abundant age group in the 701-1- stratum. Age groups 3-6-f- each show an increasing percentage with increasing depth. Collectively they comprise about 11% of the 5-25 stratum, 31% of the 51-150 stratum, and 51% of the 701 -f- stratum. The same general pattern oc- curs in northern Baja California (lat. 29.5°-32°N); however, the percentage of young fish declines more Table 5.— Percentage of mid- water trawl hauls taking north- ern anchovies by depth strata. Trawl hauls were normally with- in 30 meters of the surface. % sets Total with Depth (fm) sets anchovies 5-25 704 75 26-50 828 69 51-150 554 69 151-300 546 76 301-500 752 77 501-700 437 73 701 -^ 345 64 Total 4,166 72 492 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY Table 6. — Age composition (%) of northern anchovies taken in shallow and deep- water areas (depth in fathonns). A Depth: Lat. 320-340N 1 Age 5-25 26-50 51-150 151-300 301-500 501-700 701 + 0 56.5 26.3 16.9 7.8 5.5 3.0 6.5 1 20.6 29.5 26.5 27.4 25.8 17.9 15.5 II 12.5 24.1 26.0 27.9 30.6 32.0 26.8 III 7.0 12.3 20.1 22.8 22.7 28.2 27.1 IV 2.5 5.8 6.8 9.1 10.5 13.3 15.3 V 0.8 1.5 3.2 3.6 3.8 4.9 7.2 VI + 0.2 0.5 0.6 1.4 1.1 0.6 1.7 n 1,579 1,492 1,102 2,199 3,704 2,091 1,086 B Depth: Lat. 29.50-320N Age 5-25 26-50 51-150 151 + 0 1 56.1 23.1 36.8 40.4 21.0 28.5 8.9 28.8 II 12.3 10.8 24.2 26.9 III 6.2 7.8 14.2 23.3 IV 1.9 3.6 9.4 8.7 V 0.3 0.6 2.5 3.0 VI + — — 0.2 0.3 n 935 619 480 1,189 C Aug.-Dec, Depth: Lat. 29.50-32''N Lat. 320-340N Age 5-25 26-50 51-150 151 + 5-25 26-50 51-150 151-300 0 62.8 52.6 38.7 24.9 66.2 37.8 26.4 16.0 1 18.2 32.3 29.2 43.0 15.6 29.0 24.7 21,0 II 11.7 10.6 19.9 24.4 12.1 22.5 25.8 32.4 III 5.7 4.0 8.2 6.4 5.0 8.6 16.1 21.2 IV 1.6 0.5 4.1 1.2 0.9 2.0 5.1 8.3 V — — 0.7 — 0.2 0.1 1.8 0.9 VI + — — — — — — — 0.2 n 806 378 267 405 1,286 1,013 546 990 slowly with increasing depth there than in southern California (Tkble 6B). In both southern California and northern Baja California, there is a direct rela- tionship between average age and depth of water in which fish were caught. In the period August- December when smaller (< 60 mm) anchovies can be caught by midwater trawls, there is a greater dominance of young-of-the-year fish in the shallower water (Tkble 6C). In the northern Baja California area, 63% of the 5-25 stratum and 53% of the 25-50 stratum were young-of-the-year fish. In southern California the corresponding percentages were 66 and 38. Sea Survey - Fishery Comparisons The purse seine fleets which harvest anchovies operate primarily out of San Pedro, California, and Ensenada, Mexico. The age composition of anchovies in the San Pedro fishery (Mallicoate and Parrish 1981) contains a smaller proportion of age 0 and age 1 fish than does the sea survey data for the San Pedro Channel region. We only had 2 years of age composition data for the Ensenada fishery available to us (Sunada and Silva 1980), but this limited in- formation shows the same dominance of younger an- chovies as in the sea survey data for this region. The San Pedro fishery had several regulations which reduced the numbers of young fish in the catch. These included a 5-in minimum size limit and a series of area closures which prevent the fleet from fishing in nearshore areas. The Ensenada fishery did not have regulations which influenced the age composi- tion of the catch. Tb evaluate the effects of the area closures and size limit on the San Pedro fishery, we broke the sea survey data into depth classifications, < 50 fathoms and > 50 fathoms. The > 50 fathom classification was intended to approximate the area of the fishery (i.e, the coastal strip is excluded). In this area the 493 FISHERY BULLETIN: VOL. 83, NO. 4 age composition of anchovies taken by the Sea Survey Program is very close to that taken by the fishery; conversely, the age composition of the fishery is unlike that taken in areas < 50 fathoms (Fig. 7). The California fishery no longer has a 5-in size limit; however, the closure of the nearshore area appears to be the dominant factor in reducing the catch of young anchovies. <50fm \kr ^^\ FISHERY 30- §20- 1 \ '\ >50(m UJ a. .1 \ ^ . 10- fl \\. 1 \\\ ! >v •• 0 1 2 3 4 AGE 5 6+ Figure 7.— Comparison of the age com- position of northern anchovies taken in the San Pedro purse seine fishery with those taken in areas with <50 fathoms and >50 fathoms of water in the mid- water trawl Sea Survey Program. DISCUSSION Our data show that the growth rate and the age composition of northern anchovies vary geographi- cally. The greatest differences in growth appear to occur during the juvenile stage; growth in adults shows much less regional variation. Juvenile growth is greatest in central California and in the offshore areas of the Southern California Bight. In the in- shore regions there is a trend toward reduced juvenile growth from central California to southern Baja California. Average size at age IV2 falls from 123.6 mm to 91.8 mm over this area. Growth in adult anchovies appears to be the greatest in northern California, and it is also relatively high in British Columbia (Pike 1951), central California, and north- ern Baja California. Adult growth appears to be relatively low in the Southern California Bight; this, however, may be an artifact as this area probably in- cludes resident fish plus slower growing fish which have moved into this region from the south. Age com- position showed a large variation among regions, and the pattern of this variation appears to be closely related to the gyral circulation within the Southern California Bight. There is also a strong relationship in age composition to the depth of water at trawl sites. Adult anchovies dominated the catches in the offshore, deepwater regions of the Southern Califor- nia Bight and in central California. Age also had a strong latitudinal gradient with adult fish domi- nating in the north and young-of-the-year and year- ling fish dominating in the shallow water areas off central and northern Baja California. Adult an- chovies appear to be concentrated in areas of the Bight where prevailing currents will result in southerly and inshore larval transport (Parrish et al. 1981). At recruitment, anchovies appear to be heavily concentrated in shallow water, and young fish ap- pear to be concentrated in the nearshore area where they will tend to be advected northward by the southern California gyra As vidll be discussed later, the interpretation of the regional differences in juvenile growth is dependent upon the stock structure in the various regions. Earlier studies (McHugh 1951; Vrooman et al. 1981) showed that the boundary between the southern and central stocks was in the northern Sebastian Vis- caino Bay area. This is supported by the present study, and, as previously mentioned, the boundary is further north in the summer and fall and further south in winter and spring. We feel that there is am- ple evidence that the southern stock had the smallest juvenile growth rate and that growth during the adult phase is minor. Vrooman et al. (1981) suggested that the boundary between the northern and cen- tral stocks occurs in the central California area; both northern and central stocks occurred in samples taken at San Francisco (lat. 37°50'N) and Monterey (lat. 36°50'N). Their data might be interpreted to suggest that a fourth stock occurred in the San Fran- cisco and Monterey samples, and in addition it has been suggested (Parrish footnote 3) that this fourth stock spawns during the fall in central California and the offshore areas of the Southern California Bight. Unfortunately the Vrooman et al. (1981) study did not have any samples from the region between Monterey (lat. 36°50'N) and Newport (lat. 33°30'N), nor were there any samples from the offshore areas of the Southern California Bight. It is therefore not presently possible to determine the amount of stock mixture over much of the accepted range of the cen- tral stock. Variation in juvenile growth of northern anchovies in the different regions may be due to genetic fac- tors, differences in the seasonality of spawning, or environmental factors. The northern stock has a relatively short spawning season with a strong peak in July (Richardson 1980). The central stock has a 494 PARRISH ET AL.: REGIONAL VARIATIONS IN NORTHERN ANCHOVY more extended spawning season with a broad peak from February to April, and there is some spawn- ing all year in the central stock region. It is not yet known if the anchovies that spawn in central and southern California during the summer and fall are from the central stock, northern stock, or a possi- ble fall spawning stock. If the anchovies in this region are predominatly from the central stock, the relatively high juvenile growth in central California and the offshore portion of the Bight might be due to a favorable feeding environment. Offshore por- tions of the Southern California Bight have been shown to have considerably more plankton and nutrients than the inshore portions (Reid et al. 1958; Owen 1980). If the anchovies in the area occupied by the central stock have a large component that are not central stock, the increased juvenile growrth could be due to the genetic differences, due to environmen- tal differences, or caused by the assumption of a February birth month. At our "assumed" age of IV2 yr, an anchovy spawned in the fall would be about 6 mo older than the "normal" central stock anchovy. If the growth that occurred during these additional 5 mo was at the normal adult rate (i.e, 0.48-0.69 mm/mo), there would be only a 2-4 mm difference in the size of the two fish. However, the difference in mean length between lV2-yr-old anchovies in the central California and Cape San Quentin regions is 20 mm. If, however, growth during the 5 mo is even one-half of the average juvenile rate (i.e, 9 mm/mo) the difference in size at "IV2" yr could be achieved. CONCLUSION The interpretation of regional variations in the growth and age composition of northern anchovies in the area between central California and central Baja California and the implications of this study for fisheries management are dependent upon the stock structure of the anchovies in the area. If a significant proportion of these fish are not from the central stock, this study suggests the following: 1. The observed regional variation in age composi- tion may be the result of mixtures of stocks with different mortality rates. 2. The juvenile growth rate of anchovies in the cen- tral stock is lower than that of anchovies from the northern stock(s). The reason for this lower growth rate could be either genetic, environ- mental, or dependent upon the seasonality of spawning. 3. The southern California and Mexican fisheries are based on different stock mixtures, and thus the interactions between these fisheries would not be as great as they would be if both were based entirely on the same stock. If essentially all of these anchovies are from the central stock, this study suggests the following: 1. The offshore regions of the Southern California Bight contain a disproportionate share of the adult anchovies; however, recruitment does not occur here to any significant extent. 2. Recruitment occurs largely in shallow water along the coast, and the northern Baja Califor- nia region has the largest percentage share of young-of-the-year anchovies. 3. Larvae and juveniles recruited from the offshore regions of the Southern California Bight tend to move or be advected south and inshore 4. The relatively high juvenile growth rates in cen- tral California and the offshore regions of the Southern California Bight are due to favorable environmental conditions. 5. As they grow older anchovies tend to move, or be advected, north and offshore 6. Mixing of adults is not complete; otherwise length at age and age composition would be the same everywhere 7. Due to the inferred tendency for recruitment to occur in the south, an extensive fishery on the central stock would reduce the proportion of older anchovies and result in fewer older an- chovies in the northern and offshore areas. 8. The combination of the large Mexican fishery, which has been associated with a reduction in the proportion of older anchovies (Mais 1982), and the continued closure of the nearshore areas where younger fish are concentrated will severe- ly impact the California fishery. ACKNOWLEDGMENTS We are grateful to Roy Mendelssohn, Alec Mac- Call, Janet Mason, and Rick Methot for their com- ments and editorial assistance In addition we wish to thank Susie Myers for her help in completing the manuscript. LITERATURE CITED Ahlstrom, E. H. 1966. Distribution and abundance of sardine and anchovy lar- vae in the California Current Region off California and Baja California, 1951-64: A summary. U.S. Fish Wildl. Serv., Spec Sci. Rep.-Fish. 534, 71 p. 495 FISHERY BULLETIN: VOL. 83, NO. 4 Clark, F. N., and J. B. Phillips. 1952. The northern anchovy (Engraulis mordax) in the California fishery. Calif. Fish Game 38:189-207. Collins, R. A. 1969. Size and age composition of northern anchovies {Engraulis mordax) in the California anchovy reduction fishery for the 1965-66, 1966-67, and 1967-68 seasons. Calif. Dep. Fish Game Fish Bull. 147:56-74. Collins, R. A., and J. D. Spratt. 1969. Age determination of northern anchovies, Engraulis mordax, from otoliths. Calif. Dep. Fish Game Fish Bull. 147:39-55. Emery, K. 0. 1954. Source of water in basins off southern California. J. Mar. Res. 13:1-21. Haugen, C. W., J. D. Messersmith, and R. H. Wickwire. 1969. Progress report on anchovy tagging off California and Baja California, March 1966 through May 1969. Calif. Dep. Fish Game Fish Bull. 147:75-89. Mais, K. F 1974. Pelagic fish surveys in the California Current. Calif. Dep. Fish Game Fish Bull. 162:1-79. 1981. Age-composition changes in the anchovy, Eng^-aulis mordax, central population. Calif. Coop. Oceanic Fish. In- vest. Rep. 22:82-87. Mallicoate, D. L., and R. H. Parrish. 1981. Seasonal growth patterns of California stocks of north- ern anchovy, Engraulis mordax. Pacific mackerel. Scomber japonicus, and jack mackerel, Trachurus symmetri- cus. Calif. Coop. Oceanic Fish. Invest. Rep. 22:69-81. McHuGH, J. L. 1951. Meristic variations and populations of northern anchovy {Engraulis mordax mordax). Bull. Scripps Inst. Oceanogr. 6:123-160. Methot, R., Jr. 1981. Growth rates and age distributions of larval and juvenile northern anchovy, Engraulis mordax, with inferences on lar- val survival. Ph.D. Thesis, Univ. California, San Diego, 388 p. Owen, R. W. 1980. Eddies of the California Current System: physical and ecological characteristics. In D. M. Power (editor). The California Islands: Proceedings of a multidisciplinary sym- posium, p. 237-263. Santa Barbara Mus. Nat. Hist. (Calif.). Parrish, R. H., C. S. Nelson, and A. Bakun. 1981. Transport mechanisms and reproductive success of fishes in the California Current. Biol. Oceanogr. 1:175-203. Pike, G. C. 1951. Age, growth and maturity studies on the Pacific anchovy {Engraulis mordax) from the coast of British Columbia. M.A. Thesis, Univ. British Columbia, Vancouver, 44 p. Reid, J. L., G. I. Roden, and J. G. Wyllie. 1958. Studies of the California Current System. Calif. Coop. Oceanic Fish. Invest. Rep., 1 July 1956 to 1 January 1958, p. 28-56. Richardson, S. L. 1980. Spawning biomass and early life of northern anchovy, Engraulis mordax, in the northern subpopulation off Oregon and Washington. Fish. Bull., U.S. 78:855-876. Spratt, J. D. 1975. Growth rate of the northern anchovy, Engraulis mor- dax, in southern California waters, calculated from otoliths. Calif. Fish Game 61:116-125. Sunada, J. S., and S. Silva. 1980. The fishery for northern anchovy, Engraulis mordax, off California and Baja California in 1976 and 1977. Calif. Coop. Oceanic Fish. Invest. Rep. 21:132-138. Vrooman, a. M., p. a. Paloma, and J. R. Zweifel. 1981. Electrophoretic, morphometric, and meristic studies of subpopulations of northern anchovy, Engraulis mordax. Calif. Fish Game 67:39-51. Waldvogel, J. B. 1977. Age, maturity and distribution of northern anchovy Engraulis mordax in Humboldt Bay, California. M.S. Thesis, Humboldt State Univ., Areata, CA, 36 p. 496 PARASITES OF BENTHIC AMPHIPODS: MICROSPORIDANS OF AMPELISCA AGASSIZI (JUDD) AND SOME OTHER GAMMARIDEANS Phyllis T. Johnson^ ABSTRACT Microsporidan infections were found in individuals of 11 species of benthic amphipods collected during a 2V2-year survey of populations on the continental shelf of the northeastern United States. Ampelisca agassizi (Judd) was the most numerous and broadly distributed species of amphipod. A microsporidan confined to the abdominal muscles was common in most populations of A. agassizi. It is provisionally assigned to the genus Thelohania. There were prevalences up to 37% depending upon the population surveyed, but the microsporidans did not seem to contribute to mortality in A. agassizi populations, with the possible exception of adult males. Microsporidans in other amphipod species parasitized various organs and tissues according to the amphipod species and type of microsporidan. The relationships of the microsporidans with the genera Thelohania, Stempellia, and Nosema are discussed. In the late 1970's, a monitoring program was developed within the National Oceanic and At- mospheric Administration (NOAA) to assess the presence of pollutants and their effects on the fauna and flora of the continental shelf of the United States. As a part of this plan, the Northeast Monitor- ing Program (NEMP) has been conducted on a seasonal basis from the Gulf of Maine to Cape Hat- teras by the Northeast Fisheries Center, National Marine Fisheries Service In connection with NEMP, studies have been made of types and prevalences of parasites, diseases, and other abnormalities of various populations of benthic gammaridean am- phipods. Samples were mainly from stations on the Georges Bank and Mid-Atlantic Bight. The results of the survey will be presented in a series of papers. This, the first report, discusses microsporidan parasites, particularly those of Ampelisca agassizi (Judd). Published information on parasites and patho- logical conditions of gammaridean amphipods is limited and concerns mainly the parasites of selected estuarine and freshwater species, particularly the microsporidan parasites (Bulnheim 1975). Data col- lected during the present survey concern a broad ar- ray of species of marine amphipods. Communities of benthic amphipods are unlike most animal com- munities because they are composed of numerous in- dividuals of several to many related species that live in very close proximity to one another. Indeed, it is common for a population to contain two or more ^Northeast Fisheries Center, Oxford Laboratory, National Marine Fisheries Service, NOAA, Oxford. MD 21654. species of a single genus. It is also common for a thousand or more individuals of a single species, together with varying numbers of other species, to be crowded onto one-tenth of a square meter of the bottom (Dickinson et al. 1980). This unique popula- tion structure makes studies of parasites and diseases of the amphipods of great general biological interest. The methods used for collecting and preparing the benthic amphipods are satisfactory for study of many facets of the host-parasite relationships that exist in these animal communities: effects of parasites on their hosts, host specificity of parasites, seasonal prevalence, and modes of passage of parasites through host populations. On the other hand, paraffin-embedded sections seldom allow specific identification of parasites. Depending on the parasite group, this may require examination of live animals or of whole specimens fixed and stained by special methods. It is hoped that the data presented here and else- where will serve as a framework for more definitive studies on the taxonomy, life history, and other aspects of the various parasite species. MATERIALS AND METHODS Amphipods were sampled 11 times over a 2V2-yr period from July 1980 to December 1982 on NEMP cruises (Tkble 1). The 35 stations where benthic am- phipods were collected are shown in Figure 1. Not all stations were visited on each cruise, being sam- pled from 1 to 10 times each during the survey. The 11 stations indicated by solid circles on Figure 1 had Manuscript accepted November 1984. FISHERY BULLETIN: VOL. 83, NO. 4, 1985 497 FISHERY BULLETIN: VOL. 83, NO. 4 Table 1.— Sampling cruises for benthic amphipods, July 1980- December 1982. Cruise Date NEMP cruise no. designation July 1980 AL80-07 A September 1980 AL80-09 B December 1980 DE80-09 C April-May 1981 KE81-04 D July 1981 AL81-07 E August-September 1981 AL81-10 F November 1981 DE81-07 G January-February 1982 AL82-01 H March-April 1982 AL82-03 1 August-September 1982 AL82-10 J November-December 1982 AL82-12 K consistent, and usually numerous, amphipod popula- tions and were sampled five or more times. They yielded the majority of data presented in this pa- per. Collections of bottom sediments and accompany- ing biota were made with a 0.1 m^ Smith-Mclntyre^ grab. Generally a single grab was taken at each sta- tion sampled. If the first grab contained few amphi- pods but was from a station where they usually were abundant, a second and sometimes a third grab was taken. Sediment contained in the grab was washed through a 1.0 mm sieve, and amphipods were either collected with forceps or gently scraped from the sieve and placed in a jar of 10% seawater Formalin. On cruises A, B, and E (see Tkble 1), supplemental specimens were collected at some stations by use of an epibenthic sled or scallop dredge Storage of samples was in Formalin except that amphipods were transferred temporarily into 30 ppt artificial seawater for identification and enumera- tion, inspection for various gross lesions, and for determination of life-history stages and microspori- dan infections of the muscle in the case ofAmpelisca agassizi (Judd). A stereomicroscope was used for these procedures. Up to 30 and occasionally more individuals of each species in the sample, depending on numbers present, were processed for histologi- cal examination by standard means. Finished tissue sections were stained with hematoxylin and eosin. Depending on size and number to be embedded, 1 to 12 amphipods of a single species from a single station were embedded on their sides in each paraf- fin block. Several serial sagittal sections were taken, first laterally and then near the midline of the amphi- pods. Because of unavoidable variations in size and depth of the amphipods in the block, not all were sec- tioned at the same levels. Parts of the hemocoel, ^Reference to trade names does not imply endorsement by tiie National Marine Fisheries Service, NOAA. skeletal muscle, and appendages of all amphipods were present in sections. Usually, parts of the gills, hepatopancreas and other parts of the gut, heart, brain, and gonads were also present. Other tissues and organs, particularly the antennal gland, hemo- poietic tissues, eyes, and ventral nerve cord, often were not included. Measurements of microsporidan spores were based on fixed material, either whole or embedded, sectioned, and stained. RESULTS The amphipod population sampled at any one time at a particular station was a mixture of up to 14 dif- ferent species. Commonly five to eight species were collected in a single grab except at station 23, which was strongly dominated by Ampelisca agassizi. Of eight samples from station 23, three contained only A. agassizi, and A. agassizi made up 94 to 99% of the remaining samples. Ibtals of the eight samples from station 23 were 2,788 individuals of A. agassizi and 23 individuals of other species (99% A. agassizi). Ampelisca agassizi was the most numerous and broadly distributed of the species investigated, and occurred at 17 stations including the 11 major ones. Certain information on the life history of this species is pertinent. It is an annual, tube-building species that produces a single brood of young (Bousfield 1973). Overwintering is in the juvenile stage Gonads of both sexes develop during the subadult stage Breeding begins in the spring, and newly ovigerous females are found from spring through autumn. Postovigerous females tend to remain in the popula- tion for an unknown period after the young are re- leased from the brood pouch. Adult males are pelagic or epibenthic, probably short-lived, and usually were missing from samples collected with the Smith- Mclntyre grab. Only the adult male has strongly developed transverse pleosomal muscles (muscles of the first three abdominal segments) (Fig. 2). Presumably, these muscles aid in swimming. The transverse muscles lie lateral to the longitudinal muscles and are developed during the subadult stage They can be seen in various stages of development through the translucent cuticle of subadult males. Females of the gammaridean, tube-dwelling amphipods so far studied leave their tubes to molt to the adult stage Mating and egg extrusion take place in the water column (Mills 1967). Population dispersal is presumed to occur either by ovigerous females settling away from their original location 498 JOHNSON: MICROSPORIDANS OF AMPEUSCA AGASSIZI MARYLAND i^ZZ^^-^.'" CAPE '^ • HATTERAS: ^ ., ,76' 50 100 150 200 KILOMETERS Figure 1.— Benthic stations of Northeast Monitoring Program at which populations of gammaridean amphipods were sampled dimng the survey, 1980-82. and probably in a less populated area (Mills 1967) or by emigrating juveniles (Bousfield 1973). Thus, juveniles and perhaps ovigerous and postovigerous females of A. agassizi and other species could at times be immigrants into locations with already established populations of older juveniles and subadult males and females. Microsporidans of Ampelisca agassizi Most of the populations of A. agassizi sampled were regularly infected by a species of microsporidan that attacks the longitudinal pleosomal muscles (Figs. 3, 4). Infected muscles were chalky white in fully developed infection, and easily visible through 499 FISHERY BULLETIN: VOL. 83, NO. 4 4K/ ^■*v x: sv Figure 2.—Ampelisca agassizi: Transverse pleosomal muscles of an adult male Bar = 0.2 mm. the translucent cuticle. Usually, only one to three muscles were infected. Inspection of 150 micro- sporidan-infected amphipods showed that in eight cases transverse pleosomal muscles were also in- volved, and in three cases, only the transverse muscles were infected. One postovigerous female, with microsporidan infection in muscle, also had what appeared to be the same organism in one of the ganglia of the ventral nerve cord. On the basis of a tissue section she examined, A. Cali^ determined that this microsporidan is a pansporoblastic organism with the clusters appear- ing to be in groups of eight. However, she said fur- ther that possibly some clusters contained more than eight spores. This is a point difficult to determine in sectioned material. Spores are oval and of fairly uniform size Ones dissected out singly from infected muscle (not paraffin embedded) measured approx- imately 3 fim X 1.5 ^m. In Call's opinion, the A. agassizi parasite is best provisionally placed in the genus Thelohania, without specific designation. Figure 3.—Ampelisca agassizi: Longitudinal pleosomal muscles infected by microsporidans (open arrows). Normal muscle is also present (closed arrow), g, midgut. Bar = 0.2 mm. Numerical information on this microsporidan is based on samples taken on cruises D-K, because determination of microsporidan infection was by study of sectioned material only from cruises A-C, and infections can be missed by this method. Con- sidering all stations on cruises D-K, juveniles had a lower prevalence of grossly visible infection than did male and female subadults and ovigerous females (Tkble 2), but this was not invariably the case in in- dividual samples. In 5 of the 38 samples with micro- Table 2. — Prevalence of microsporidan infec- tions in Ampelisca agassizi by life-history stages. All stations, cruises D-K. Life-history stage No. infections/ total collected (o/o prevalence) 'A. Call, Rutgers University, Newark, NJ 07102, pers. commun. 1983. Juveniles Subadults Ovigerous females Postovigerous females Adult males Totals 517/4,868 (11) 1 ,335/5,293 (25) 111/501 (22) 82/413 (20) 1/55 (2) 2,046/11,130(18) 500 JOHNSON: MICROSPORIDANS OF AMPELISCA AGASSIZI ft .m% Am ^ '1 ' ^ *H' ^* •^^ ji»' I.' m^ ^ V I 1% w^ ^ IP Cil Figure i.—Ampelisca agassizi: Groups of spores and prespores of the muscle-infecting microsporidan. Bar = 10 fjm. sporidan-infected A. agassizi, prevalence was equal to or slightly higher in juveniles than in subadults or ovigerous females. Prevalence was very low in the small sample of adult males, but varied considerably in individual samples of both males and postovi- gerous females (Ikble 3). The three stations with the most consistently numerous populations of A. agassizi also had the highest prevalences of micro- sporidan infection. These were stations 23, 33, and 48, with overall prevalences of 30%, 37%, and 22%, respectively. Overall prevalence at other stations ranged from 0 to 14%. The majority of infected hosts showed no reaction to the presence of the microsporidans. However, there was occasional melanization in heavily infected muscle, with invasion of hemocytes into the mass of spores, some encapsulation of spores and infected muscle, and lysis of many spores. In these cases, it appeared that the muscle had lost its integrity; that is, the sarcolemma probably was no longer entire Often, other infected muscle near the necrotic mass of spores and muscle showed no evidence of attack by host defense mechanisms. The second microsporidan of A. agassizi parasi- tized epithelial cells of the posterior half of the midgut. Juveniles, male and female subadults, and ovigerous and postovigerous females were infected. The parasite resembled Nosema, the spores being single and free in the cytoplasm of the host cell (Fig. 5). Spores were slightly oval and about 2 ^m in the greater dimension. Infected cells were hypertrophied (Fig. 6). In one heavy infection, many spores were free in the gut and apparently most infected cells had ruptured. There was no host reaction to infec- tion. This parasite occurred twice in individuals with microsporidan infection in abdominal muscle. Based on sectioned material, overall prevalence of the gut microsporidan was < 0.1% (25/2403). Prevalence in samples with one or more infected A. agassizi was 3.7% (25/678), range 1-6%. Amphipods with microsporidans in the gut epithelium were from stations 23, 33, 47, 48, 49, 50, and 51. Microsporidans in Species Other Than A. agassizi Males and females of Unciola species (probably all U. irrorata Say and U. inermis Shoemaker) were hosts to a microsporidan that infected longitudinal muscles of the pleosoma In three instances, a similar or the same microsporidan was found in a ganglion of the ventral nerve cord, and not in muscla Spores appeared similar to those of the A. agassizi parasite; they measured about 3 x 1.5 ^m; and there were eight or more spores per envelope Unlike the A. agassizi parasite, vegetative stages were often pres- ent along with developed and developing spores. Un- ciola species have an opaque cuticle, and infected muscle cannot be seen grossly. Based on sectioned Table 3. — Prevalence of microsporidan infections in Ampelisca agassizi by life-history stages. Stations 47 and 48, cruises E and F. Life-history stage No. infections/ total collected (% prevalence) No. infections/ total collected (% prevalence) Cruise E Cruise F Station 47 (depth 48 m) (depth 62 m) Juveniles 29/851 (3) 130/1,124 (12) Subadults 45/258 (17) 12/53 (23) Ovigerous females 14/84 (17) 11/53 (20) Postovigerous 0/24 (0) 6/29 (21) females Adult males 0/34 (0) 1/3 (33) Totals 88/1,251 (7) 160/1,262(13) Cruise E Cruise F Station 48 (depth 72 m) (depth 68 m) Juveniles 2/33 (6) 5/29 (17) Subadults 66/246 (27) 28/1 1 1 (25) Ovigerous females 0 (-) 11/51 (22) Postovigerous 0 (-) 3/27 (11) females Adult males 0/1 (0) 0/4 (0) Totals 68/280 (24) 47/222 (21) 501 FISHERY BULLETIN: VOL. 83, NO. 4 Mr Figure 5.—Ampelisca agassizi: Nosema-\ike microsporidan in the midgut epithelium. Bar = 10 ixm. A'- • ■ # » *0 « *:? .' — ,^' _\ _^^ Figure 6.—Ampelisca agassizi: Hypertrophy of midgut epi- theUal cells infected by a Nosema-Vike microsporidan. Infected epithelium to left, normal epithelium to right. Bar = 60 ^im. material, prevalence v^as 8.3% (23/277), considering only samples containing infected Unciola spp. Prevalence was 1.7% when considering all Unciola spp. that were sectioned and examined (Ikble 4). There was no host reaction to infection in the ganglia, but animals with muscle infection often showed some melanization and encapsulative response (Fig. 7). Scattered small melanized nodules were common in the hemocoel of infected Unciola spp., but it was not evident whether they had form- ed in response to microsporidans. Other amphipod species with microsporidan infec- tions are listed in Table 4. Prevalence was usually very low. Most of the parasites appeared like the muscle-infecting microsporidans of A. agassizi and Unciola spp. A Nosema-\ike parasite similar to the gut microsporidan of A. agassizi, but smaller (0.7 ixm), occurred in the hepatopancreatic epithelium of a specimen of Leptocheirtis pinguis (Stimpson). Another L. pinguis harbored a larger A^osema-like species in oocytes and heart muscle Infected oocytes were necrotic and encapsulated by hemocytes. The generalized muscle parasite of Melita dentata (Kr0yer) s. lat. was also A^osema-like. W n f # O t i t 1% # ^ ^t^^ m" "^W ^"^ 5"^ * # liiS#* Figure 7.— Unciola sp.: Host reaction to microsporidans in ab- dominal muscle Some groups of degenerating spores and prespores are surrounded by melanized material. A few nuclei of encapsulating host cells are visible around the mass of microsporidans. Bar =10 jim. 502 JOHNSON: MICROSPORIDANS OF AMPELISCA AGASSIZI Table 4. — Microsporidans in amphipods other than Ampelisca agassizi. No. infections/ Amphipod total examined 1 ^•ositive Type of species (% prevalence) ( stations Tissues infected microsporidan Unciola spp. 23/1,365(1.7) 33, 35, 38, Abdominal muscle. "Thelohania" (irrorata Say and 47, 48, 51, ganglia of ventral inermis Shoemaker) 110, 112 nerve cord Ampelisca 4/448 (0.9) 57, 66 Muscle, tegmental "Thelohania" vadorum Mills glands, oocytes Ampelisca 1/48 (2.1) 62 Connective tissue. "Thelohania" verrilli Mills muscle Ericthonius 1/436 (0.2) 38 Abdominal muscle "Thelohania" rubrlcornis Smith Eriopisa elongata 1/29 (3.4) 47 Ganglia of ventral "Thelohania" (Bruzelius) nerve cord Leptocheirus 1/913 (0.1) 47 Abdominal muscle "Thelohania" plnguis (Stimpson) 1/913 (0.1) 15 Oocytes, heart Nosema-Wke 1/913 (0.1) 20 muscle Epithelium of hepatopancreas Nosema-Wke Melita dentata 2/44 (4.5) 51 Generalized in Nosema-Wke (Kr^yer) s. lat. muscle Monoculodes 1/110 (0.9) 40 Abdominal muscle "Thelohania" edwardsi Holmes Photis dentata 4/301 (1.3) 33 Abdominal muscle. "Thelohania" Shoemaker ganglia of ventral nerve cord The microsporidan of Ampelisca vadorum Mills resembled that of Unciola spp., but fully developed spores were not seen (Fig. 8). Muscle, tegmental 8 4Mi •»• x*" ^-* glands, and oocytes were infected. Often, groups of microsporidans had "used up" the host tissue, and appeared like groups of extracellular, closely knit, vegetative and sporulating stages, but some of these groups were unmistakably in the shape of tegmen- tal glands and oocytes and were in the correct anatomical positions. In one case, muscle fibers were still present adjacent to the mass of microsporidans and in another, microsporidans infected a recogni- zable tegmental gland. Host reactions to the micro- sporidans had not occurred in the few infected A. vadorum available for study. More than 35 specimens each of the following species were sectioned and examined, but micro- sporidans were not found: Anonyx sarsi Steel & Brunei (36 specimens), Byblis serrata Smith (316 specimens), Casco bigelowi (Blake) (60 specimens), Corophium crassicorne (Bruzelius) (50 specimens), Harpinia propinqua Sars (116 specimens), Orcho- menella minuta Kr0yer (64 specimens), Phox- ocephalus holbolli Kr0yer (73 specimens), Pseuduni- cola obliquua (Shoemaker) (46 specimens), and Rhepoxynius epistomus (Shoemaker) (249 specimens). % i Figure 8— Ampelisca vadorum: Vegetative and prespore stages of a muscle-infecting microsporidan. Bar =10 jjm DISCUSSION Bulnheim (1975) and Sprague (1977) have hsted and discussed the various microsporidans reported from amphipods. Most of the hosts are freshwater and estuarine forms, and depending on the species 503 FISHERY BULLETIN: VOL. 83, NO. 4 of microsporidan, muscles, ovaries, connective tissues, and gut epithelia are infected. One of the microsporidans, a parasite of Gammarus pulex L., infects the longitudinal abdominal muscles of its host in the same manner as does the A. agassizi parasite, but is knovv^n to have a variable number of spores per envelope It was named Glugea muelleri (Pfeif- fer, 1895, in van Ryckeghem 1930), later named Thelohania giraudi (Leger and Hesse, 1917), and has been called Stempellia muelleri (Pfeiffer) by Bulnheim (1975) and Microsporidium giraudi (Leger and Hesse) by Sprague (1977). The relation- ship of "Glugea muelleri Pfeiffer" and the micro- sporidan from A. agassizi remains to be determined. Although "Glugea muelleri Pfeiffer" and the A. agassizi parasite are remarkably similar in being restricted to the longitudinal abdominal muscles of their hosts, the latter is probably significantly smaller. Fixed spores of the A. agassizi parasite are about 1.5 X 3 ^m, and fresh spores of "Glugea muelleri" are 2.2 x 4.5 jum. The method of transmission of the A. agassizi parasite is not known. Microsporidans are usually transmitted orally, but transovarial transmission also occurs in amphipods. Transovarially transmitted microsporidans of Gammarus duebeni Lilljeborg in- fect the ovary, and, depending on the species, cause complete or partial feminization of males (Bulnheim 1975, 1977). The parasite of A. agassizi did not in- fect the ovary, and because it was regularly found in normal males, it apparently does not cause feminization of males. Bulnheim (1971) successful- ly transmitted "G. muelleri Pfeiffer" to several species of Gam/marus, including euryhaline ones, by feeding of infected muscle Prevalence of the muscle parasite of A. agassizi apparently increases with age of the host, and it could be hypothesized that this microsporidan is transmitted orally, that the older the host the more chances it has had to become infected, and that the parasite does not contribute to increased mortality in the population. Adult males are active swimmers and might prove an exception because impaired mus- cle function could lead to increased predation. If this occurred, one would expect infected males to be preferentially removed from the population, leading to a lower prevalence of infection in this stage In- deed, prevalence in adult males was only 2%. However, relatively few males were collected during the survey, and the low prevalence could prove to be sampling artifact. Note that in the sample from cruise E, station 47 (Tkble 3), both postovigerous females and adult males were uninfected, but 2 mo later, at the same station, prevalence in post- ovigerous females was 21% and the only infected adult male found during the survey was also collected at that time The discrepancy in prevalence might be due to sampling of slightly different populations. As discussed below, there is no assurance that the same population vras sampled spatially, and temporal differences conceivably might also have complicated the results. Relationships of the microsporidans seen in the various species of amphipods could not be decided on the basis of material fixed and prepared as it was. It would be interesting to determine whether the parasites of Unciola spp. and Ampelisca vadorum are the same or different species, and what their relationship is to the A. agassizi parasite There were some differences in the habits and the developmen- tal stages present in the three amphipods. Vegetative stages were common in the case of the A. vadorum parasite and fairly common in Unciola spp., but usually rare or absent in A. agassizi. Several dif- ferent tissues were infected in A. vadorum, but ex- cepting a few infections in nervous tissue, only ab- dominal muscle was infected in Unciola spp. and A. agassizi. Previous investigators have found that microsporidan infection is well tolerated by am- phipod hosts, and that defense reactions against these parasites generally are limited and may come into play mainly when host tissue becomes necrotic (reviewed by Bulnheim 1975). The muscle-inhabiting microsporidan of A. agassizi is obviously a primary parasite of that species and is seldom attacked by the host. However, the similar parasite of Unciola spp. often either provokes attack merely by its pres- ence or damages the muscle so that a response oc- curs to the necrotic tissue In either event, it is possi- ble that this parasite is not fully adapted to Unciola spp., because arthropods are known to be less tolerant of non-adapted parasites (Salt 1970; Unestam and Weiss 1970) With exception of the muscle-infecting species from A. agassizi, microsporidans are not common parasites of benthic amphipods in the areas sampled, even considering that some infections must have been missed because not all would be seen in the limited number of sections examined from each amphipod. Sampling methods used in the survey do not lend themselves to precise studies on progression of parasitic infections through particular populations. Sampling cannot be done often enough to show if and when additions to populations (with perhaps dif- ferent prevalences of parasites) are provided by im- migrating juveniles or other stages of these short- lived animals. Further, populations may not be 504 JOHNSON: MICROSPORIDANS OF AMPELISCA AGASSIZI homogeneous over the area sampled at a single sta- tion. Sediment sampling with a grab is imprecise, as the different depths of samples taken at stations 47 and 48 on cruises E and F testify (Table 3). It is probable that return to an exact location was never accomplished. Even if populations were homo- geneous, predation by fish, and other short-term disturbances, may cause local impoverishment of populations or differences in population composition that would not be detected in the necessarily blind sampling done with a Smith-Mclntyre grab. A general pattern does emerga In the area surveyed, microsporidans are dominant parasites of the most numerous and ubiquitous species, A. agassizi, but are rare in all other species. This may be a reflection of the fact that only A. agassizi con- sistently occurred in dense populations at certain stations at all sampling times, a circumstance that would promote spread of a host-specific and horizon- tally transmitted parasite. ACKNOWLEDGMENTS Thanks are due to Frank Steimle and Robert Reid of the Northeast Fisheries Center Sandy Hook Laboratory, and Linda Dorigatti, Gretchen Roe, and Sharon MacLean of the Oxford Laboratory, who col- lected the amphipods. Ann Frame, Sandy Hook Laboratory, provided expert advice and training in identification of amphipods. Linda Dorigatti iden- tified material from cruises A-C, and along with Gretchen Roe, Dorothy Howard, and Cecelia Smith of the Histology Section, Oxford Laboratory, prepared the specimens for histological examination. Ann Call, Rutgers University, Newark, NJ, review- ed the manuscript and provided identification of the muscle-inhabiting microsporidan of A. agassizi. LITERATURE CITED BOUSFIELD, E. L. 1973. Shallow-water Gammaridean Amphipoda of New England. Cornell Univ. Press, Ithaca, N.Y., 312 p. BULNHEIM, H.-P. 1971. ijber den Wirtskreis der Mikrosporidie Stempellia miiUeri. Arch. Protistenkd. 113:137-145. 1975. Microsporidian infections of amphipods with special reference to host-parasite relationships: a review. Mar. Fish. Rev. 37(5-6):39-45. 1977. Geschlechtsumstimmung bei Gammarus duebeni (Crustacea, Amphipoda) unter dem Einfluss hormonaler und parasitarer Faktoren. Biol. Zentralbl. 96:61-78. Dickinson, J. J., R. L. Wigley, R. D. Brodeur, and S. Brown- Leger. 1980. Distribution of gammaridean Amphipoda (Crustacea) in the Middle Atlantic Bight region. U.S. Dep. Commer., NOAA Ifech. Rep. NMFS SSRF-741, 46 p. Leger, L., and E. Hesse. 1917. Sur les Microsporidies de la Crevette d'eau douce C. R. Seances Soc Biol. Fil. 80:12-15. Mills, E. L. 1967. The biology of an ampeliscid amphipod crustacean sibling species pair. J. Fish. Res. Board Can. 24:305-355. Salt, G. 1970. The cellular defence reactions of insects. Cambridge Univ. Press, Cambridge, 118 p. Sprague, V. 1977. Systematics of the Microsporidia. In L. A. Bulla, Jr., and T. C. Cheng (editors). Comparative pathobiology. Vol. 2. Plenum Press, N.Y., 510 p. Unestam, T., and D. W. Weiss. 1970. The host-parasite relationship between freshwater cray- fish and the crayfish disease fungus Aphanomyces astaci: responses to infection by a susceptible and a resistant species. J. Gen. Microbiol. 60:77-90. VAN Ryckeghem, J. 1930. Les Cnidosporidies et autres parasites du Gammarvs pulex. La Cellule 39:401-416. 505 LONG-TERM RESPONSES OF THE DEMERSAL FISH ASSEMBLAGES OF GEORGES BANK William J. Overholtzi and Albert V. Tylers ABSTRACT The resilience of demersal fish assemblages on Georges Bank was investigated with data from seasonal bottom trawl surveys conducted by the Northeast Fisheries Center, National Marine Fisheries Service, Woods Hole, Massachusetts, from 1963 to 1978. Cluster analysis proved to be a useful statistical method for delineating assemblage boundaries and associated species. Assemblages persisted over the long-term and changed spatial configuration only slightly on a seasonal basis. Declines in biomass, numerical densi- ty, and changes in relative abundance occurred ranging from mild to severe Assemblage changes were probably triggered by intense fisheries as well as inherent trophic dynamics of component species. Results have useful multispecies management connotations. The assemblage concept appears to be an appropriate operational or conceptual framework for further management and modeling applications. Most community ecological studies have necessari- ly concentrated on the short-term aspects or season- ality of assemblages. Typically 1 to 3 yr of field measurements are analyzed with information theory, niche breadth procedures, or multivariate statistical methods. Demersal fish assemblages in particular have been investigated in a number of locations [see studies by Tyler (1971), Oviat and Nixon (1973), Stephenson and Dredge (1976), Hoff and Ibara (1977), Gabriel and Tyler (1980), and Inglesias (1981)]. The recurrent theme in most of these studies centers around seasonally varying diversity because of environmentally induced migration, temperature usually acting as the dominant driving variable Unfortunately, many interesting questions cannot be addressed in these studies because of their short- term horizon. It is important to consider the long- term ramifications of fishery system responses. The temporal scale referred to here as "long-term" does not refer to geologic time, but rather ecological time, the span of years during which the actions of fishery ecologists evoke system responses. Fishery ecologists are limited in their ability to function within this time frame. For instance, a plant ecologist could predict with some certainty the type of forest that would eventually occupy a cleared site, if left undisturbed, but comparable knowledge for fishery 'Department of Fisheries and Wildlife, Oregon State Universi- ty, Corvallis, OR 97331; present address: Northeast Fisheries Center Woods Hole Laboratory, National Marine Fisheries Service, NOAA, Woods Hole, MA 02543. ^Department of Fisheries and Wildlife, Oregon State Universi- ty, Corvallis, OR 97331; present address: Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, BC V9R 5K6, Canada. systems is lacking, especially in the marine environment. Are fish assemblages stable? How do they respond to exploitation? Holling (1973) investigated sys- tem responses to man's activities, showing that in closed systems, such as freshwater lakes, the pro- pensity to remain stable is high, but not infallible Smith (1972) critiqued the Great Lakes experience, concluding that the activities of man, notably fishing and pollution, when coupled with biological inter- actions, caused significant community alterations in this system. Few marine studies, with the exception of Soutar and Isaacs (1969), Sutherland (1980), DeVries and Pearcy (1982), and some general over- view papers (Brown et al. 1976; Richards et al. 1978), have stressed the long-term temporal and spatial aspects of marine system response Longer term temporal and spatial questions were examined with data from research conducted at the Northeast Fisheries Center (NEFC) (Grosslein 1969). Concentrating on Georges Bank, we used cluster analysis to produce yearly fall and spring dendrograms for the period 1963-78 and 1968-78, respectively. Assemblages were defined, component species were identified, distributional maps plotted, and the information was examined to elucidate long- term temporal and spatial patterns. Further analyses led to trajectories of species catch-per-unit- effort (CPUE), assemblage total biomass, estimates of intra-assemblage diversity, and other measures of community response It is suggested that fishing, coupled with interspecific interactions, appeared to have played a major role in determining trends in the Georges Bank assemblages. Manuscript accepted November 1984. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 507- 5 FISHERY BULLETIN: VOL. 83, NO. 4 METHODS Georges Bank, a large, submerged, marine plateau, is located off the northeastern coast of the United States (Fig. 1). It has been the site of an intense fishery for several centuries, and a large interna- tional fleet exploited the area from the 1960's to the mid-1970's. The NEFC has conducted annual bot- tom trawl surveys on the Northwest Atlantic con- tinental shelf since the autumn of 1963. Annual spring surveys commenced in 1968 and, in addition, several summer and winter cruises have been under- taken on an intermittent basis. Surveys were con- ducted from Nova Scotia to Hudson Canyon from 1963 to 1966 and coverage was extended to Cape Hatteras beginning in 1967. Grosslein (1969) and Azarovitz (1982) described the details and justifica- tion for the surveys, but a brief summary is appropriate The objective of the surveys is to obtain statistical- ly meaningful abundance estimates of the offshore marine fish populations in the aforementioned areas. Secondary objectives included the collection of data for distribution studies, age and growth determina- tions, predator-prey interactions, and a host of special purpose investigations. The potential area was divided into zones (strata) based on depth and biological considerations. Stratified random samples were selected with allocation to each strata propor- tional to its area. A 30-min sample with a standar- dized research bottom trawl and a 1.25 cm cod end liner was accomplished. All fish, as well as major in- vertebrates, were sorted to species, weighed, and measured, and some fish were sampled for other MAINE 44' 42° 40" 74' 72* 70' 68' 66° 64" Figure 1.— Georges Bank and Gulf of Maine region with shoreline place names and other associated geographical landmarks. 508 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES analyses. Sampling frequency averages about one station for every 250 mi^ or roughly 300 locations in a normal survey from Cape Hatteras to Nova Scotia. Data from a selected portion of this time series was used in cluster analyses that defined demersal fish assemblages. Specifically, a group of 36 species representing the dominant fishes on Georges Bank, were chosen as the focus for the investigation (Ihble 1). This choice was based on a preliminary examina- tion of the data to determine which species were most important in terms of biomass and numerical density. Catches (kg) for each of the species from every station in a particular cruise were organized into a data matrix and processed with an agglomera- tive cluster analysis program (Keniston 1978). lb remove skewness in the species matrices, we trans- formed the data prior to clustering by using an In (x + 1) conversion. Station dissimilarities were calculated by using the Bray-Curtis dissimilarity in- dex, an ecological distance measure that is sensitive to dominant species (Clifford and Stephenson 1975; Boesch and Swartz 1977). Table 1.— Species cited by common name in the text. Common name Scientific name Spiny dogfish Winter skate Little skate Smooth skate Thorny skate Atlantic herring Alewife Offshore hake Silver hake Atlantic cod Haddock Pollock White hake Red hake Cusk American plaice Summer flounder Fourspot flounder Yellowtall flounder Winter flounder Witch flounder Windowpane Gulf stream flounder Atlantic mackerel Butterfish Bluefish Blackbelly rosefish Longhorn sculpin Sea raven Gunner American sand lance Atlantic wolffish Ocean pout American goosefish Short-finned squid Long-finned squid Squalus acanthias Raja ocellata Raja erinacea Raja senta Raja radiata Clupea harengus Alosa pseudoharengus Merluccius albidus Merluccius bilinearis Gadus morhua Melanogrammus aeglefinus Pollachius virens Urophycis tenuis Urophycis chuss Brosme brosme Hippoglossoldes platessoides Paralichthys dentatus Paralichthys oblongus Limanda ferruginea Pseudopleuronectes americanus Glyptocephalus cynoglossus Scophthalmus aquosus CItharichthys arctifrons Scomber scombrus Peprilus triacanthus Pomatomus saltatrix Helicolenus dactylopterus Myoxocephalus octodecemspinosus Hemitripterus americanus Tautogolabris adspersus Ammodytes americanus Anarhichas luptus Macrozoarces americanus Lopliius americanus lllex illecebrosus Loligo pealei The resulting dissimilarity matrix was used in a group average fusion strategy to combine stations with similar species distributions (Clifford and Stephenson 1975). These station combinations were displayed in dendrograms, which were examined and assemblage groups were chosen by two criteria: large-scale separations, as shown in Figure 2, and dissimilarity levels. Stations from these assemblage groups were plotted on cruise maps from the original sampling plan and areas were delineated. This pro- cess was repeated for all spring and fall cruises to provide a consecutive series of maps, which were then examined for continuity (Fig. 3). Finally, data from several consecutive years were pooled to delineate assemblages designated, based on nearby geographic features or depth zones. Species lists were prepared for the assemblages outlined in the pooled cluster results, and data were analyzed to further define the structure of each group. Length frequencies from species in the dif- ferent assemblages were used to separate life history stages and catch-per-tow data were used to in- vestigate trends in distribution and abundance Ex- amination of food habit data in the literature and NEFC documents gave further insight into assem- blage structure Trajectories of assemblage CPUE for selected species were plotted and examined for long-term trends. Ibta! assemblage CPUE was also investigated and compared with previous trends reported by other authors for the region. Gradient analyses were performed with the objec- tive of explaining species distributions based on a set of location, physical, and chemical variables. Canonical correlations, using information on latitude, longitude, depth, bottom temperature, bottom oxy- gen, and bottom salinity, were employed to define possible gradients that might be useful indicators of species distribution (Pimentel 1979). Data for the autumn cruise were obtained from measurements of bottom temperature and depth made aboard the RV Albatross IV {U.S.A.) 20 October to 5 November 1976, and corresponding information on bottom salinity and oxygen from the RV Anton Dohrn (Federal Republic of Germany) 14 November to 1 December 1976. Information for the spring cruise was procured from measurements of bottom temperature and depth from the RV Albatross IV, 17 April to 3 May 1978, and salinity and oxygen data that was obtained from the RV Argus (Union of Soviet Socialist Republics) from 13 to 28 April 1978; these two data sets were chosen because they cor- responded closely in time to the available station information. 509 FISHERY BULLETIN: VOL. 83, NO. 4 ^ ^ 1.0r .8 - .6 - .4 - .2 - 0 ^ -- in oj o o o o o m o m 1- o oooppoo oooQoP ooo ooo CO to -^ C\J o o o o (£> a O) . oOooooooooooooooooooooooOooooo ~' ro^(Ch-ro OfOinototficDO^ioto tmio-^cvjiOcvj — toT-^tofO - -ooo - ro cvj po iT) STATION Figure 2.— Typical dendrogram, autumn 1966, showing duster station groups and dissimilarities. Dashed line indicates a dissimilarity of 0.65. RESULTS Five important assemblage groups were present on Georges Bank from fall 1963 to 1978. For refer- ence, w^e name these groups: Slope and Canyon, Intermediate, Shallow, Gulf of Maine Deep, and Northeast Peak. A consistent spatial pattern emerged as consecutive fall cruises were examined and plotted. The same five groups appear to have been present in similar locations since 1963. These five assemblages were present at the mid- and end- points of the fall time series also (Fig. 3). The groups appear to change their spatial configuration slight- ly on an annual basis, but the general area of each group was maintained. Lists of the dominant species in each assemblage are given in Ikble 2. The total area that each assemblage encompass- ed through time (years) was delineated by pooling the observations from consecutive years. Figure 4 shows an example of a representative assemblage from the spring and fall, respectively. The groups overlapped surprisingly little through time with the exception of a few border stations along adjacent assemblages. Table 2.— Assemblage species associations from cluster results (demersal species only). Slope and canyon: Silver hake White hake Red hake Gulf stream flounder Offshore hake Fourspot flounder Blackbelly rosefish American goosefish Intermediate: Winter skate Little skate Red hake Silver hake Atlantic cod Haddock Sea raven American goosefish Ocean pout Longhorn sculpin Yellowtail flounder Shallow: Winter skate Little skate Silver hake Atlantic cod Haddock Pollock White hake Red hake Summer flounder Yellowtail flounder Winter flounder Windowpane Longhorn sculpin Sea raven Ocean pout Sand lance American goosefish Gulf of Maine Deep: Thorny skates American plaice Witch flounder White hake Silver hake Atlantic cod Haddock Cusk Atlantic wolfflsh Northeast Peak: Thorny skate Atlantic cod Haddock Pollock White hake Winter flounder Ocean pout Longhorn sculpin 510 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES , GULF OF MAINE DEEP ■^Nj: X NORTHEAST SHALLOW --V \ PEAK INTERMEDIATE SLOPE AND CANYON FALL 1963 Figure 3.— Georges Bank assemblages for three autumn surveys 1963, 1970, 1976. 43<= 42° - 41° - 40° 43° 42° - 4r 40' i/t' — 1 — I — 1 — ''^' ^ \ J.. •• . •• ') \ ,/• •• J- . . » • .* / 1 1 1 1 71' 70' 69' 68' 67' 66' Figure 4.— Sample pooled station distributions for the Slope and Canyon assemblage, spring 1968-73 (top panel) and the Shallow assemblage, autumn 1963-67. Data for all cruises were pooled by season and used to generate composite maps of general assem- blage areas, for the spring and autumn (Fig. 5). The Slope and Canyon assemblage appears to encompass a similar area regardless of season, while some of the other assemblages changed slightly. The Shallow assemblage covered most of Georges Bank in the spring (Fig. 5) and was slightly smaller in the fall (Fig. 5). The Intermediate assemblage is somewhat larger in the fall (Fig. 5), suggesting a migration of the species in this area to shallower water as the year progresses. Assemblages in the spring appear to follow depth contours resulting in the elongate shape of the groups at this time (Fig. 5). The Northeast Peak Interior (NPI) and Northeast Peak-Gulf of Maine Deep (NP-GM Deep) assemblages show definite seasonal spatial changes when compared with the Gulf of Maine Deep (GM Deep) and North- east Peak assemblages in the fall (Fig. 5). The general shape and location of the fall assemblages suggests that a different set of oceanographic and biological forcing factors are important in deter- 511 FISHERY BULLETIN: VOL. 83, NO. 4 69° 68° 67' 66' 42' 41" I 1 ■T I I ( NP-GM DEEP NPI -V v > "^^^^^^ >r\- / SHALLOW J f I 2-^:;^ ^ 100 m , \ s *" "^^^••'"''^NTERMEDIATE 1 \^^ SLOPE AND CANYON 1 GEORGES BANK SPRING ASSEMBLAGES 42° - 41° NORTHEAST PEAK, ''d^ «^ INTERMEDIATE 69° 68° 67° 66° GEORGES BANK FALL ASSEMBLAGES Figure 5.— Composite maps showing seasonal changes in the Georges Bank assemblages and their approximate areas. NPI = Northeast Peak Interior; NP-GM Deep = Northeast Peak-Gulf of Maine Deep; GM Deep = Gulf of Maine Deep. mining the distribution of fish. The Northeast Peak assemblage, for instance, spans several depth zones and encroaches on the Shallow assemblage, reducing its area during this part of the year. The assemblage maps presented in Figure 5 were useful for organizing the 36 species of Ikble 1 into their corresponding demersal subunits (Tkble 2). Four basic species categories were defined in the various assemblages. These included ubiquitous species, resident species, periodics, and those resi- dent species present in several assemblages during different parts of their life history. Ubiquitous species, such as ocean pout, goosefish, sea raven, and Atlantic cod, were found with regularity in almost all of the assemblages. Resident species, such as lit- tle skate, winter skate, longhorn sculpin, yellow tail flounder, winter flounder, American plaice, and witch founder, were present in only one or two assemblages in abundance Periodic or seasonal migrants include bluefish, butterfish, and mackerel, as well as short-finned squid and long-finned squid. These species moved in and out of the various assem- blages on a seasonal basis with temperature being a likely dominant force, and were often highly variable in terms of their abundance and were there- fore not included in Ikble 2. A number of species, including silver hake, red hake, white hake, and haddock, were present in more than one assemblage as different life history stages. Silver hake, for example, are found in the Slope and Canyon and Shallow assemblages, with adults on the average, occurring more frequently in the Slope and Canyon and Gulf of Maine Deep assemblage, while juveniles are more abundant in the Shallow assem- blage. It appears that for many of the abundant fish species on Georges Bank, adults occupy the deeper peripheral assemblages while juveniles of these same species occupy the shallower zones during much of the year. ASSEMBLAGE TRAJECTORIES Assemblage CPUE indices were calculated for several of the spring and fall assemblages and used for evaluating temporal trends in total catch and catch by species. Assemblage CPUE declined drama- tically in the mid-1960's to early 1970's in four of five Georges Bank assemblages in fall (Fig. 6). In par- ticular, research catches in the Shallow, Northeast Peak, and Gulf of Maine Deep assemblages reached all-time lows in the early 1970's, coincident with large increases in international effort and landings at that time (Figs. 6, 7). International effort, mea- sured in thousands of days fished, increased three- fold over the period 1960-69 (Fig. 7). Assemblage biomass showed some signs of recovery in the late 1970's when good year classes of Atlantic cod, had- dock, and other species occurred and international effort declined due to the Magnuson Fishery Con- servation and Management Act of 1976 (Figs. 6, 7). Research catch of silver hake, fourspot flounder, red hake, white hake, and black belly rosefish re- mained nearly stable over the spring period (1968-75), then increased abruptly after 1976 due to increases in the silver hake (Fig. 8A). Total catch for the fall time series was also stable for most years, until 1972 when silver hake and red hake abundance fluctuated (Fig. SB). Figure 8C shows the trends in percent by weight for the five species during fall indicating a change in biomass dominance for silver hake and red haka Blackbelly rosefish and fourspot flounder showed the 512 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES SLOPE AND CANYON 200 r 150 100 50 1 •_! 1 a^^-S • 200 150 100 50, INTERMEDIATE 0 200 r 150 100 50 _l L. SHALLOW _1 I 200 NORTHEAST PEAK / 150 1 — / / / 1 / / • 100 50 n 1 1 1 1 200 r 150 100 50 GULF OF MAINE DEEP 0 _i I I I I ^ 5 60 r 50 40 30 20 10 _l_ _i_ _l_ 60 62 64 66 68 70 72 YEAR 74 76 78 80 63 66 69 72 73 78 TIME Figure 7— Unstandardized effort data in thousands of days fish- ed for the Georges Bank demersal fishery, all countries, for 1961-79, expressed as 3-yr moving means. same trends as in the former case, but represented more of the catch on a percent weight basis in the later years of the fall time series (Fig. 8C). Gulf Stream flounder was actually one of the more im- portant species numerically during the mid-years of the series (Fig. 8D). The same general trend for red and silver hake, and the other species is apparent in the percent by numbers data (Fig. 8D). The shallow assemblage was much more diverse than the Slope and Canyon assemblage. The major species of importance were Atlantic cod, winter skate, longhorn sculpin, little skate, yellowtail flounder, and haddock. Mean catch per tow in the fall time series declined dramatically from 202 kg in 1963 to 22 kg in 1972 and subsequently rose to 99 kg in 1978 or about one-half the 1963 value (Fig. 6). Winter flounder, longhorn sculpin, and winter skate appear to have remained fairly constant in abundance over the spring time period, while Atlan- tic cod, windowpane flounder, and little skate displayed an increasing trend in biomass (Fig. 9A). Yellowtail flounder and haddock showed declining mean catches over this interval. The fall time series, since it is longer, clarifies some of the observed spring trends. Cod and winter flounder CPUE re- mained relatively stable over the fall period, while windowpane flounder, winter skate, and little skate appear to have increased from 1972 onward (Fig. Figure 6.— Mean catch per tow (kg) from NEFC Georges Bank bottom trawl surveys for autumn 1963-78 for the five assemblages. Dashed line indicates a 3-yr moving mean of the plotted data points. 513 FISHERY BULLETIN: VOL. 83, NO. 4 ais UJCC i-q: CC LU OUJ h-O Q.Q l/iZ tnz Ll-3 q: d -lO no a, a: Oli. ILU. 5X t/5I £ Q tJ3aiAinN lN30d3d O o o CM -^ S^ <1> A £ 3 C -o c c« 4J x; u a; > >i ..Q ^-j C V c; 1-. m a .w eS 3 E 3 o St h 00 J= t>- tn co Q CO T3 I— ( § c F u rs 3 Ol oi •>. -^ a> 01 > > o u tv «5 T* ■w 05 T1 OJ T-H i d E 3 C E 3 n o 1 to § a: 0) T3 00 (^HJMOI/HDIVO NV3y\l 1H9I3M lN3Dd3d S< be ra m C -" "^^ he E ^ 1^ ni ^- ^3 CO a o V ^ C I 3 C 00 1^ H C cS 05 g 0) Uh «t-i CO 514 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES c B 3 O .1 U s 3 C T3 C c« be I C o 0) S 3 I CO -o £ s: o fcr m' (^>I)M01/H01VD NV3IAI 1H9I3M lN3D^3d M fc. 5 W u o OS 5* != §§§ 515 FISHERY BULLETIN: VOL. 83, NO. 4 9B). Haddock CPUE, on the other hand, declined dramatically from 97.3 kg in 1963 to 0 in 1972, re- maining at very low^ levels of abundance in the later years. Yellowtail flounder fluctuated from 15 kg in 1963 to a low of 6 kg in 1966, increased from 1966 to 1969, and declined through 1978 (Fig. 8B). An examination of trends in cumulative percent by weight and number trajectories for the Shallow assemblage highlighted some interesting points. Atlantic cod comprised a fairly constant proportion of the species biomass for all the years except 1964 and 1965. Longhorn sculpin, yellowtail flounder, and to a lesser extent winter flounder, made up an in- creasing part of the biomass of this assemblage dur- ing 1966-71 and then all declined in importance (Fig. 9C). Haddock, as previously noted, experienced a pronounced decline in abundance from the early 1960's and was only present at very low levels from 1972 to 1978. Winter skates, little skate, and windowpane flounder accounted for an increasing percent of the biomass in this assemblage from the early 1970's onward (Fig. 9C). When cumulative percent by number was inves- tigated, silver hake and red hake became important (Fig. 9D). Silver hake was the numerical dominant through most of the mid- and late 1970's. This trend was due entirely to increased numbers of juvenile silver hake that represented a small amount of biomass. This same phenomenon applies to red hake, which enjoyed several periods of increased abun- dance as a proportion of the total numerical densi- ty from 1963 to 1978. Winter skate numbers remain- ed relatively unchanging from 1963 to 1976 and then rose slightly in the late 1970's. Trends for window- pane flounder, longhorn sculpin, little skate, yellow- tail flounder, and haddock follow the cumulative ab- solute and percent weight data (Fig. 9B, C, D). The other Georges Bank assemblages were inves- tigated using the same techniques, but on a much less intense scale TDtal mean catch/tow for the Inter- mediate, Gulf of Maine Deep, and Northeast Peak assemblages is displayed in Figure 6 for the fall surveys 1963-78. The trends in total CPUE follow the same basic patterns for all three groups, a high initial period followed by a decline and subsequent recovery in the mid- to late 1970's. General decreases in the catch of throny skates, haddock, and cod were responsible for the downward trend in CPUE for the Northeast Peak assemblage, but the recovery that occurred in the late 1970's was due primarily to increased haddock biomass (Figs. 6, 10). The Northeast Peak assemblage is fairly sim- ple in species composition, and although some fluc- tuations in cumulative percent by weight occurred. NORTHEAST PEAK ASSEMBLAGE CUMULATIVE PERCENT BY WEIGHT FALL 1963-1978 100 r POLLOCK HADDOCK 1963 1967 1971 1975 Figure 10.— Reponses of species from the Northeast Peak assem- blage demersal fish community expressed as cumulative percent by weight for autumn 1963-78. the same four species remained dominant over the period (Fig. 10.) Time sequence cluster analyses were useful as fur- ther indicators of temporal trends in these groups. Species biomass for the Slope and Canyon assem- blage did not appear to follow any clear long-term trend (Fig. 11). Enough fluctuation in CPUE oc- curred to mask any trend, and no clear pattern was established. This same analysis on the Shallow assemblage showed three distinct temporal clusters, composed of consecutive years (Fig. 11). Using this perspective and Figure 9, there appears to have been three periods of significant change in relative abun- dance during the fall time series; an initial period dominated by haddock, intermediate period with high yellowtail, longhorn sculpin, and winter flounder catches, and finally a group with little skate, winter skate, and windowpane flounder as the domi- nant species. GRADIENT ANALYSIS Gradient analyses of two selected Georges Bank data sets did not prove to be as useful as was hoped, but some information and insight were gained and the dimensionality of the large multivariate data sets involved was much reduced. The data set used in the fall 1976 canonical correlation analysis accounted for about 26% of the variation in species distribution for 32 selected species of interest. The variables included in the analysis were latitude, longitude, depth, bot- tom temperature, bottom salinity, and bottom oxy- 516 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES 1.0 r .8 .6 .4 1- .2 I Sr/J TION 1.0 r .8 .6 .4 .2 r^roioiD'j-cnoo-^OCDr^tDf^infO'}- STATION Figure 11.— Time sequence cluster analyses for autumn 1963-78 for the Slope and Canyon (top panel) and Shallow (bottom panel) assemblages on Georges Bank. gen. The first three canonical axes accounted for 73.9% of this total, a cumulative redundancy of 19.0% (Ikble 3). The first canonical variable (CV) reflects the importance of depth and to a lesser degree bottom salinity, in determining the distri- bution of these species. Although none of the corre- lation loadings for CVl are particularly high, the gadoids, as a group, show a positive trend. Many of the shallow-water species, such as little skate, winter skate, and most of the flounders, showed negative correlations with this canonical variable The other two CV's reflect the location variables of latitude and longitude as well as bottom oxygen and salinity. Since the gadoids and flounders appeared to show a group response to these distribution variables, we decided to use them in another analysis, excluding the other species (Ikble 3, Fall 1976 II). This data set explained 28.3% of the total variation in distri- bution for a selected set of 14 species (Tkble 3). The first canonical variable had a high correlation with latitude (r = 0.904) and the gadoids, as a group, were highly positively correlated with this CV (Ikble 3). It appears that although Georges Bank spans only about two degrees in total north-south latitudinal variation, this variable is useful for defining centers of gadoid biomass. The third analysis did not reveal any new trends, accounting for 32.6% of the variation in species distribution. In general, then, although significant orthogonal canonical axes were defined in each of three data sets, the amount of variation that was ac- tually explained was relatively small. There appear to have been trends in the distribution of some gadoid and flounder species, but the strength of these relationships was hardly firm. Most of the variation in species distribution was related to latitudinal, salinity, and depth differences. DISCUSSION Questions of community resilience (Pimm 1984) are meaningful because resource managers are faced with the dilemma of making decisions that may alter future community structure Fishery managers, in particular, are unable to deal with the long-term consequences of their management decisions because they lack specific knowledge of ecosystem reponses. This idea may apply particularly to areas such as Georges Bank, where landings of each species are part of a multispecies otter trawl fishery. In this case the application of single species manage- ment to assemblages of fishes may result in simpli- fication of the community such that less productive fish populations or those more vulnerable to fishing are reduced dramatically (lyier et al. 1982). If this occurs, important trophic linkages may be precluded, economic viability may suffer, and management op- tions may be removed indefinitely. At the present time the argument of these central issues is pro- ceeding slowly in the literature and few, if any, management agencies are considering these types of questions in their decisions. We need, therefore, to begin to investigate the long-term temporal scale of communities so that ecologists and managers can begin to function in terms of ecological time instead of just a framework for short-term reaction. Declines in total finfish abundance on the conti- nental shelf of the northeastern United States reach- ed unprecedented levels over the period 1965-74 (Brown et al. 1976). Not only had biomass declined. 517 FISHERY BULLETIN: VOL. 83. NO. 4 Table 3.— Canonical variable (CV) loadings for fall 1976 and spring 1978 gradient analyses, with canonical correlation coefficients (Re), amount of variation explained by each canonical axis (% variation), and total variation in species distribution explained by the environmen- tal data. Fall 1976 1 Fall 1976 II Spring 1978 1 CV1 CV2 CV3 CV1 CV2 CVS CV1 CV2 CV3 Spiny dogfish - 0.509 0.149 0.232 -0.089 0.105 0.111 Winter skate -0.263 0.342 -0.244 -0.176 -0.377 0.058 Little skate -0.428 0.113 -0.163 -0.015 -0.437 0.142 Smooth skate 0.552 -0.117 -0.144 0.333 0.210 -0.204 Thorny skate 0.387 0.017 0.063 0.534 0.416 0.188 Sea herring 0.003 0.081 0.058 0.084 -0.147 0.068 Alewife -0.023 0.097 0.161 0.433 0.086 0.104 Offshore hake 0.456 -0.408 -0.100 -0.147 0.613 0.284 Silver hake 0.285 0.203 0.536 0.607 0.332 0.027 -0.237 0.637 -0.328 Atlantic cod 0.059 0.544 0.195 0.600 -0.263 0.048 0.262 -0.534 0.177 Haddock 0.370 0.525 0.131 0.647 -0.069 -0.235 0.383 -0.300 0.072 Pollock 0.337 0.158 0.199 0.439 0.274 0.167 0.425 0.209 0.015 White hake 0.564 -0.022 0.341 0.447 0.614 0.083 -0.030 0.527 0.055 Red hake -0.109 -0.109 0.034 -0.220 -0.043 -0.460 0.303 0.616 0.057 American dab 0.281 0.048 0.362 0.395 0.365 0.141 0.409 -0.035 -0.026 Summer flounder -0.245 -0.124 -0.102 -0.246 -0.111 0.126 -0.427 0.309 0.303 Fourspot flounder -0.235 -0.265 0.210 -0.326 0.051 -0.588 -0.417 0.418 0.366 Yellowtaii flounder -0.158 0.324 -0.083 0.115 -0.419 -0.095 0.080 -0.442 0.144 Winter flounder -0.145 0.364 0.078 0.295 -0.324 0.114 0.109 -0.301 0.040 Witch flounder 0.179 -0.267 0.140 -0.107 0.407 -0.187 0.368 0.138 -0.225 Windowpane -0.326 0.165 -0.351 -0.087 -0.439 0.504 -0.527 0.041 0.251 Butterfish -0.279 -0.415 0.405 Blackbelly rosefish 0.359 -0.544 0.050 -0.295 0.529 -0.199 Longhorn sculpin -0.078 0.432 -0.136 0.101 0.478 0.068 Sea raven -0.034 0.432 -0.092 -0.500 0.034 0.194 Gunner -0.082 0.198 0.023 American sand lance -0.134 0.043 -0.079 0.130 -0.125 0.224 Atlantic wolffish 0.011 0.240 0.106 0.233 0.073 0.358 Ocean pout -0.043 0.265 0.122 -0.133 -0.316 0.034 American goosefish 0.145 -0.073 0.184 -0.261 0,414 -0.039 Short-finned squid 0.249 0.284 0.303 -0.275 0.169 0.072 Long-finned squid -0.471 -0.239 -0.379 -0.256 0.399 -0.077 Variables Latitude 0.306 0.922 0.074 0.904 -0.333 0.066 0.878 0.012 -0.131 Longitude -0.458 -0.164 0.597 -0.004 0.064 0.118 0.240 -0.032 0.474 Depth 0.885 -0.432 0.049 0.030 0.849 -0.071 -0.107 0.792 -0.448 Bottom temp. -0,179 -0.696 -0.233 -0.590 0.258 0.366 -0.252 0.720 -0.465 Bottom salinity 0.463 -0.547 0.422 -0.107 0.753 -0.586 -0.374 0.909 -0.133 Bottom oxygen -0.316 0.450 -0.581 0.041 -0.613 0.407 0.199 -0.908 0.144 RC 0.975 0.961 0.874 0.871 0.841 0.750 0.979 0.968 0.943 % variation 8.2 7.2 3.6 11.1 9.1 4.3 9.2 11.8 2.8 Significance P < 0.001 P < 0.001 P < 0.05 P < 0.001 P < 0.001 P < 0.05 P < 0.001 P < 0.001 P < 0.001 Total variation 25.7 28.3 32.6 but total effort on Georges Bank increased several times (Fig. 7). The assemblage trends examined in this paper can be linked to these high levels of effort. During this time period, seasonal bottom trawl surveys monitored trends in finfish abundance over the area from the Gulf of Maine to Cape Hatteras. This survey proved invaluable to fish stock assess- ment work because changes in the relative abun- dance of most of the commercial species were follow- ed closely and were highly correlated with commer- cial catch, effort, and other indices (Clark 1979). Other species of ecological, perhaps not commercial importance, were also routinely and closely monitored over this time The spring and fall bot- tom trawl survey provided an excellent means for assessing community or assemblage responses over this time period. Cluster analysis, with the Bray-Curtis dissimilarity index and group average fusion method, proved helpful for defining demersal fish assemblages on Georges Bank. Recent studies confirm the value and applicability of the Bray-Curtis index (Bloom 1981). This method provided a means for collapsing the multidimensional nature of the spring and fall Georges Bank survey cruises into smaller, more easily interpreted, units. It was then possible to investigate not only long-term temporal and spatial persistence questions, but also intraspecific responses within the particular assemblage of interest. 518 OVERHOLTZ and TYLER: DEMERSAL FISH ASSEMBLAGES Not only did seasonal Georges Bank assemblages maintain their temporal integrity over the periods 1963-78 in the fall and 1968-78 in the spring, but they also appear to have retained their spatial configura- tion for the most part as well. The results of this study indicate that although changes in species com- position and relative abundance occurred in varying degrees in all the assemblages, they remained con- tinuous in time and spaca Although many of the species on Georges Bank are found in several assemblages, it appears that each of the five groups has enough large-scale varia- tion in biomass, species composition, and relative abundance to make each of the assemblages uniqua Also at least one or two dominant Georges Bank species occupy each assemblage, for example, the bulk of the haddock stock occurs in the Northeast Peak group. Thus, even though some assemblages changed dramatically in terms of species richness and relative abundance, the spatial integrity of each complex was preserved over tim& The energy budget of Georges Bank can serve as a plausible explanation for the particular species distributions we found. Georges Bank is a dynamic ecosystem driven by a complex and unique nutrient advection system. Its shallow topography and geo- graphic location, with constant mixing of the water column and lack of stratification, does not lead to the usual nutrient limitation of primary productivi- ty (Sutcliffe et al. 1976; Cohen et al. 1982). Instead of the usual spring and fall pulse in primary produc- tion, the region is characterized by high productivi- ty from April to November. Yearly primary produc- tion levels are as high as 450 gC/m^ per yr in the shallow (< 100 m) zone of Georges Bank (Cohen et al. 1982). This shallow mixed zone encompasses the same area as the Shallow and Intermediate assem- blages delineated in our cluster analysis results. The area is dominated by fish that feed on invertebrates. Primary prey items for these species include euphau- siids, copepods, mysids, amphipods, and other ben- thic invertebrates. This part of the ecosystem is fairly closely tied to primary production, and its compo- nent species may compete for food resources dur- ing their early life history (Pitt 1970; Bowman 1981; Overholtz 1982). Predation, too, may be an impor- tant biological mechanism for determing trends in this assemblage (Overholtz 1982). The other assem- blages that we have described in this analysis occur along the fringes of Georges Bank at the shelf-slope interface These shelf break groups contain the ma- jor adult demersal fish stocks found in the area with the exception of yellowtail flounder. These peripheral assemblages are dominated by large predators that are generally piscivorous, with little, if any, dietary overlap (Langton and Bowman 1981). The gradient analyses suggest that about 25% of the total variation in species biomass distribution can be explained with the variables used in the study. This result was surprising at first, since we felt that the variables we used would explain much more of the variation than this. However, considering the fact that other important biological factors, such as predation, fishing, competition, and food preferences were not included in the analysis, it is probably a realistic percentage Perhaps an analysis that in- cluded the whole east coast would account for much more variation because a wider range of conditions would exist. Other studies that have successfully explained species distributions usually occur in habitats with very strong physical or chemical gradients, such as mountain forests or estuaries (Whitaker 1967; Mcln- tire 1973). Either the actual gradients were not strong enough to explain more than a small percent- age of the species distribution or those other factors were more important. The questions of resilience and stability of demer- sal fish assemblages that were defined and inves- tigated in this study have implications for the management of Georges Bank. This study provides a useful conceptual framework for managing many of the demersal fish stocks in this area. Not only were stable zones with specific resident fishes delineated, but they were present over the long-term record. Species components of fall assemblages are indicators of general distributions that represent the location of major fish stocks during the productive portion of the year. Long-term responses observed in the Georges Bank community indicate the pro- pensity for adjustment or resilience (Rolling 1973) that a particular assemblage might have Assem- blages on the periphery of Georges Bank might be less susceptible to changes in species composition and relative abundance because their component species are less trophically linked. The Shallow assemblage, on the other hand, appears to be par- ticularly vulnerable to fishing and perhaps inter- specific interactions. This type of knowledge will be helpful for understanding changes in fish abundance and community structure and for effectively man- aging fishery systems in the future ACKNOWLEDGMENTS We would like to dedicate this paper to the men and women of the Northeast Fisheries Center, Woods Hole, MA, for without their long-term plan- 519 FISHERY BULLETIN: VOL. 83, NO. 4 ning and tireless efforts in collecting the data, it would have been impossible to complete this work. We thank E. Beals for providing such good advice and helping with analyses. We are grateful for the friendship and assistance of G. Kruse, T. Hayden, and W. Gabriel, who provided much inspiration over the span of the last several years. We would also like to thank B. Brown for his help, direction, and patience LITERATURE CITED AZAROVITZ, T. R. 1981. A brief historical re\dew of the Woods Hole Laboratory trawl survey time series. Can. Spec Publ. Fish. Aquat. Sci. 58:62-67. Bloom, S. A. 1981. Similarity indices in community studies: potential pit- falls. Mar. Ecol. Prog. Ser. 5:125-128. BOESCH, D. F., AND R. C. SWARTZ. 1977. Application of numerical classification in ecological in- vestigations of water pollution. U.S. Environ. Prot. Agen- cy Ecol. Res. Ser. EPA-600 3-77-033, 115 p. Bowman, R. E. 1981. Food of 10 species of northwest Atlantic juvenile groundfish. Fish. Bull., U.S. 79:200-206. Brown, B. E., J. A. Brennan, M. D. Grosslein, E. G. Heyer- DAHL, and R. C. HENNEMUTH. 1976. The effect of fishing on the marine finfish biomass in the Northwest Atlantic from the Gulf of Maine to Cape Hat- teras. Int. Comm. Northwest Atl. Fish. Res. Bull. 12:49-68. Clark, S. H. 1979. Application of bottom trawl survey data to fish stock assessment. Fisheries (Bethesda) 4(3):9-15. Clifford, H. T., and W. Stephenson. 1975. An introduction to numerical classification. Acad. Press, N.Y., 229 p. Cohen, E. B., M. D. Grosslein, M. P. Sissenwine, and F. Steimle. 1982. Energy budget of Georges Bank. In M. C. Mercer (editor), Multispecies approaches to fishery management ad- vice, p. 95-107. Can. Spec Publ. Fish. Aquat. Sci. No. 59, 169 p. DeVries, T. J., AND W. G. Pearcy 1982. Fish debris in sediments of the upwelling zone off cen- tral Peru: a late Quaternary record. Deep-Sea Res. 29: 87-109. Gabriel, W. L., and A. V. TVler. 1980. Preliminary analysis of Pacific Coast demersal fish assemblages. Mar. Fish. Rev. 42(3-4):83-88. Grosslein, M. D. 1969. Groundfish survey program of BCF Woods Hole Comm. Fish. Rev 31(8-9):22-30. HOFF, J. G., AND R. M. iBARA. 1977. Factors affecting the seasonal abundance, composition, and diversity of fishes in a southeastern New England estuary Estuarine Coastal Mar. Sci. 5:665-678. HOLLING, C. S. 1973. Resilience and stability of ecological systems. Annu. Rev Ecol. Syst. 4:1-23. Inglesias, J. 1981. Spatial and temporal changes in the demersal fish com- munity of the Ria de Arosa (NW Spain). Mar. Biol. (Berl.) 65:199-208. Keniston, J. A. 1978. Program cluster: an aid to numerical classification using the CDC cyber computer. Greg. State Univ Sea Grant, Pleuronectid Proj. Tfech. Rep. No 2, 32 p. OSU Mar. Sci. Ctr., Newport, OR. Langton, R. H., and R. E. Bowman. 1981. Food of eight Northwest Atlantic pleuronectiform fishes. U.S. Dep. Commer., NOAA Ifech. Rep. NMFS, SSRF-749, 16 p. McIntire, C. D. 1973. The distribution of estuarine diatoms along environmen- tal gradients: a canonical correlation. Estuarine Coastal Mar. Sci. 6:447-457. Overholtz, W. J. 1982. Long-term temporal perspectives for the demersal fish assemblages of Georges Bank with implications for manage- ment and modeling. PhD. Thesis, Oregon State University, Corvallis, 259 p. Oviatt, C. a., and S. W. Nixon. 1973. The demersal fish of Narrangansett Bay: An analysis of community structure, distribution, and abundance Estuarine Coastal Mar. Sci. 1:361-378. PiMENTEL, R. A. 1979. Morphometries: the multivariate analysis of biological data. Kendal/Hunt Publ. Co., Dubuque, Iowa. Pimm, S. L. 1984. The complexity and stability of ecosystems. Nature (Lond.), 307:321-326. Pitt, T K. 1970. Distribution, abundance, and spawning of yellowtail flounder, Limanda ferrunginea, in the Newfoundland area of the northwest Atlantic J. Fish. Res. Board Can. 27:2261-2271. Richards, J., D. W Armstrong, J. R. G. Hislop, A. S. Jermyn, and M. D. Nicholson. 1978. Trends in Scottish research vessel catches of various fish species in the North Sea, 1922-1971. In G. Hempel (editor), North Sea fish stocks - recent changes and their causes, p. 211-224. Rapp. P.-V. Reun. 172, 449 p. Smith, S. H. 1972. Factors in ecological succession in oligotrophic fish com- munities of the Laurentian Great Lakes. J. Fish. Res. Board Can. 29:717-730. SOUTAR, A., AND J. D. ISAACS. 1969. History of fish populations inferred from fish scales in anaerobic sediments off California. Calif. Coop. Oceanic Fish. Invest. Rep. 134:63-70. Stephenson, W., and M. C. L. Dredge. 1976. Numerical analysis of fish catches from Serpentine Creek. Proc R. Soc Queensl. 87:33-43. Sutcliffe, W H., Jr., R. H. Loucks, and K. F. Drinkwater. 1976. Coastal circulation and physical oceanography of the Scotian Shelf and the Gulf of Maine J. Fish. Res. Board Can. 33:98-115. Tyler, A. V. 1971. Periodic and resident components in communities of Atlantic fishes. J. Fish. Res. Board Can. 28:935-946. Tvler, a. v., W. L. Gabriel, and W J. Overholtz. 1982. Adaptive management based on structure of fish assem- blages of northern continental shelves. In M. C. Mercer (editor), Multispecies approaches to fishery management ad- vice, p. 149-156. Can. Spec Publ. Fish. Aquat. Sci. No. 59, 169 p Whittaker, R. H. 1967. Gradient analysis of vegetation. Biol. Rev. (Camb.) 42:207-264. 520 OBSERVER EFFECT ON INCIDENTAL DOLPHIN MORTALITY IN THE EASTERN TROPICAL PACIFIC TUNA FISHERY Bruce E. Wahlen' and Tim D. Smith^ ABSTRACT Scientific observers placed aboard a sample of purse seine vessels collect data that are used to estimate the total number of dolphins killed incidentally in the eastern tropical Pacific tuna fishery. If the presence of these observers, who are not crew members, affects incidental kill levels, then the kill estimates will be biased, lb test for the existence of such an observer effect, we compared dolphin kill data that had been recorded by observers who differed in levels of obtrusiveness according to their purposes for data collection. Some observers were placed on board primarily to collect data for estimating the total number of dolphins killed annually. Other observers collected data both for that purpose and for monitoring com- pliance with dolphin-release regulations. Our results confirm that the presence of an observer does affect dolphin kill. The primary effect is an increase in the proportion of sets with no dolphins killed, and a decrease in the proportion of sets with one to nine dolphins killed. While the magnitude of the effect of observers cannot be estimated from our data, estimates of total dolphin mortality based on data col- lected by the scientific observers are biased downward. Schools of dolphins of several species, primarily Stenella attenuata and 5. longirostris, have been used since the late 1950s by purse seine fishermen in the eastern tropical Pacific Ocean (ETP) to locate and catch yellowfin tuna, Thunnus albacares. Per- rin (1969) described the process of deploying, or set- ting, the net around the tuna and dolphins, and then releasing the dolphins while retaining the tuna. Significant numbers of dolphins have been killed in- cidentally in this fishery by becoming entangled in the purse seines (Smith 1983). The National Marine Fisheries Service (NMFS) and the Inter-American Tropical Tbna Commission (lATTC) place scientific observers who are not crew members aboard a sample of tuna purse seine vessels to collect data related to dolphin kill. Both the NMFS and lATTC have used the data collected by these scientific observers to estimate the total number of dolphins killed annually by the entire tuna purse seine fleet (Lo et al. 1982; Hammond and Tbai 1983). Additionally, the NMFS uses these data to monitor dolphin kills relative to annual kill limits establish- ed for the U.S. registered fleet (Lo et al. 1982). Periodic estimates of the cumulative numbers of dolphins killed are compared with the annual limit. If the limit is exceeded, U.S. vessels must stop fishing on the affected populations for the remainder of the year. Data collected by the NMFS observers have also been used to monitor compliance of vessel operators with dolphin-release regulations, including the release of all live dolphins from the net (Federal Register 1977, 1980). Until recently data collected by an NMFS observer could be used as evidence to prosecute vessel operators for violations of these regulations. Observer effects have been defined in a general context as measurement procedures which influence and thereby change the behavior of the subject (Johnson and Bolstad 1973, p. 38). Researchers have encountered such effects in a variety of empirical sciences, including psychology (Johnson and Bolstad 1973), social science (Webb et al. 1966, p. 18), and biology (Ricker 1975, p. 87). We defined an observer effect on the number of dolphins killed as a differential in levels of dolphin kill between trips made with and without a scien- tific observer. The existence of such a differential would introduce a bias into estimates of the total number of dolphins killed (Smith 1983; Powers^). Large numbers of sets involving dolphins (dolphin sets) are made each year (Punsly 1983), so even a moderate observer effect could result in a substan- 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. ^Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA; pres- ent address: Northeast Fisheries Center Woods Laboratory, Na- tional Marine Fisheries Service, NOAA, Woods Hole, MA 02543. Manuscript accepted November 1984. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. ^Powers, J. E. 1979. A discussion of incidental mortality by unobserved United States purse seiners. Unpubl. manuscr., 7 p. Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 521 FISHERY BULLETIN: VOL. 83, NO. 4 tial bias in the estimates of annual dolphin kill. Directly testing for the existence of an observer effect on dolphin kills would require comparison of covert observations with observations by NMFS and lATTC observers. Based on the large difference be- tween the kill rate observed covertly by one crew member and the kill rates recorded by NMFS observers during other fishing trips made by the same operator and vessel, Smith (1983) speculated that a large observer effect existed. We investigated the significance of the difference in kill rates reported by Smith (1983) by grouping NMFS- observed trips into sequences of trips with common operator and vessel. A few of these sequences of NMFS-observed trips revealed between-trip kill rate differences as large or larger than in the sequence that included the covert observations. The existence of an observer effect can be indirect- ly tested without relying on data from covert observers. Johnson and Bolstad (1973) established the existence of an observer effect by comparing measurements made by observers with various levels of obtrusiveness to the human subjects whose responses were being measured. They concluded that the differences in the responses measured by observers with different levels of obtrusiveness im- plied that the observer's presence had affected the subjects' behavior. They noted, however, that the magnitude of an observer effect cannot be estimated using this approach. Following this indirect approach, we tested for the existence of an observer effect on the numbers of dolphins killed by comparing dolphin kill data col- lected by scientific observers who differed in their purposes of data collection, and hence, in their levels of obtrusiveness. DATA The scientific observers were placed aboard a ran- dom sample of U.S. registered tuna purse seine vessels (Lo et al. 1982). Assignment of an NMFS or lATTC observer to vessels in the sample was also made randomly, subject to the constraint that any vessel sampled twice within a calendar year would be accompanied by an NMFS observer on at most one trip (Ikble 1). Information collected for each NMFS- or lATTC- sampled fishing trip included departure date and data pertaining to each set (such as set type, date, and location), and for dolphin sets, the number of dolphins killed. Data available to the authors from NMFS-sampled trips included all of this information. However, data available from lATTC-sampled trips did not include departure date, and set dates were available only to the quarter of the year. While the data items collected by both types of scientific observer have been similar over the years, for NMFS observers the purposes of the data col- lection changed after March 1981. The primary pur- poses of data collection, as explained to each vessel operator at a placement meeting held prior to depar- ture, were as follows: 1) On NMFS-sampled trips begun from 1978 through March 1981, data were col- lected for estimating the annual kill of dolphins and for monitoring compliance with dolphin-release regulations; 2) on NMFS-sampled trips begun after March 1981 through the end of 1982, the data were still used for estimating dolphin kills but were no longer used to monitor compliance with dolphin- release regulations''; 3) on all lATTC-sampled trips, since the inception of that sampling program in 1979, data were collected for estimating total kill but were never collected for monitoring compliance with dolphin-release regulations. As described above, the data collected by both NMFS and lATTC observers to estimate total dolphin kill can be used by the NMFS to halt fishing by U.S. vessels on specific dolphin populations for the remainder of the year. The data collected before March 1981 by the NMFS observers for monitoring compliance with dolphin-release regulations, however, can be used by the NMFS as evidence to prosecute operators who failed to comply. Thus, the operators are likely to be more conscious of the presence of an observer who is collecting data both for estimating dolphin kill and for monitoring com- ■*The change in data collection purposes of NMFS observers after March 1981 was prompted by a court order forbidding the NMFS from using data collected by observers for monitoring compliance with dolphin-release regulations. No NMFS observers were placed on fishing trips begun from 1983 through part of 1984 because of a subsequent court order forbidding placement of NMFS observers without a search warrant. Table 1. — Number of observed fishing trips which made at least one dolphin set from 1978 through 1982, by observer type and year. NMFS totals are subdivided ac- cording to departure date of trips (previous year, Jan.-IVIar., Apr-Dec.) and exclude trips in which fishing gear research was conducted. Observer type 1978 1979 1980 1981 1982 NMFS Previous year 5 4 3 3 7 Jan.-Mar. 44 33 15 7 13 Apr.-Dec. 56 32 28 28 18 Total 105 69 46 38 38 lATTC 0 31 57 58 44 Total 105 100 103 96 82 522 WAHLEN AND SMITH: OBSERVER EFFECT ON INCIDENTAL DOLPHIN MORTALITY pliance with dolphin-release regulations than an observer who is collecting data only for estimating dolphin kill. That this is the case is implied by the constraint in the sampling procedure that any vessel sampled twice within a calendar year may be accom- panied by an NMFS observer on at most one trip. METHODS We tested for the existence of an observer effect on dolphin kills by comparing the number of killed dolphins recorded by more obtrusive observers with the number recorded by less obtrusive observers. We considered observers who collected data both for estimating dolphin kill and for monitoring com- pliance with dolphin-release regulations to be more obtrusive to vessel operators than observers who col- lected data only for estimating dolphin kill. Thus, we compared kills recorded by (la) NMFS observers before and after March 1981, and (lb) NMFS and lATTC observers before March 1981. As a control, we compared the number of killed dolphins record- ed by observers of equal obtrusiveness. That is, we compared kills recorded by (2a) lATTC observers before and after March 1981, and (2b) NMFS and lATTC observers after March 1981. The frequency distributions of numbers of dolphins killed were extremely skewed, with very long right tails (Fig . 1). Normality assumptions were violated so strongly by these skewed distributions that ANOVA tests for differences in means, particularly one-sided tests, would be difficult to interpret (Glass et al. 1972). Therefore, we tested for differences in the percent of dolphin sets in which no dolphins were killed (zero-kill sets). This percent relates directly to the regulation requiring release of all live dolphins, and is a dominant feature of the dolphin kill distributions. When comparing frequency distributions, we entertained the null hypothesis of equality of per- cent zero-kill sets. When comparing observers of dif- ferent obtrusiveness levels, we tested this hypothesis against a one-sided alternative that distributions from more obtrusive observers had a higher percent Before After >■ o z 111 3 O Ui oc u. lU > < lit flC Ma r. 1981 Mar . 1981 TRIPS 215 59 DOLPHIN SETS 4.780 1,634 MEAN 3.24 5.21 STD. DEV. 22.83 21.79 MAXIMUM 854 387 ^ ^ ■ — ^ — p^^ . — p i—V/Z/i 1-9 10-19 20-29 30-39 40-49 50-99 ^100 NUMBER OF DOLPHINS KILLED Figure L— Relative frequency distributions of number of dolphins killed incidentally during sets of NMFS-observed trips, 1978 through 1982, by trip departure data 523 FISHERY BULLETIN: VOL. 83, NO. 4 of zero-kill sets than distributions from less obtrusive observers. When comparing distributions from observers of equal obtrusiveness, we tested the null hypothesis against the two-sided alternate of in- equality (Tkble 2). Results from all of our tests are reported at the 0.05 significance level. Table 2. — Alternate hypotheses to the null hypothesis of equality of percent zero-kill sets for each of tour comparisons, where Before and After refer to before or after March 1981. See text for details. Comparison Alternate hypothesis 1a. NMFS before vs NMFS after Before > After 2a. lATTC before vs lATTC after Before # After lb. NMFS before vs lATTC before NMFS > lATTC 2b. NMFS after vs lATTC after NMFS # lATTC For two-sided tests of differences in percents, we used the standard chi-square (x^) statistic with one degree of freedom (df). For one-sided tests, we used the square root of the chi-square statistic (Z), which is approximately normal (Snedecor and Cochran 1980, p. 126-127). In some instances, the expected cell frequencies were less than the tradi- tionally accepted minimum of five. However, recent Monte Carlo results (Fienberg 1980, p. 172) suggest that the chi-squared distribution is an ade- quate approximation at the 0.05 significance level even when minimum expected values are as low as one. While the sampling of vessels was nearly random, the actual sample obtained may not have been representative of factors affecting dolphin kills. It has been demonstrated that within the ETP, dolphin kills vary among three geographic areas^ (Fig. 2) and by periods within the year (Lo et al. 1982). We divided the year into two periods: January-March and April-December. This division corresponds to the date of the change in data collection purposes of NMFS observers in 1981, and also tends to equalize sample size since vessels in this fishery are more active in the early part of the year. We stratified the data by area and period of the year to account for biases due to possible non- representativeness of the sample with respect to these two factors. When data on numbers of dolphins killed were available in all six area-period strata, we made overall two-sided tests for differences in per- ^K-T. T^ai, Inter-American IVopical Thna Commission, c/o Scripps Institute of Oceanography, La Jolla, CA 92093, pers. commun. December 1983. 40°N 20" - 20° - ACS 1 e \ 1 ) - NORTH 20" \ N. ^Vv/7 - OUTSIDE 5 o O 5"N. N0RTH^~-1 ") INSIDE S!^ / - 3 0 C T ► » 1 SOUTI 1 H 1 / lecw 140° 120" 100° 80° Figure 2.— The three areas of the eastern tropical Pacific used to stratify the data, bounded by lat. 40°N., long. 160°W., lat. 40°S., and the western coastline of the North and South American Continents. 524 WAHLEN AND SMITH: OBSERVER EFFECT ON INCIDENTAL DOLPHIN MORTALITY cent zero-kill sets (conditional on period of the year and geographic area) by summing chi-square values and degrees of freedom from each stratum. When observations were not available in one of the strata, or when the alternative was one-sided, an overall test based on the chi-square statistic was not possible In those cases, results of the tests within each stratum were considered separately. RESULTS The first two comparisons of frequency distribu- tions test for differences in the percent of zero-kill sets in data collected by observers on trips begun before compared with trips begun after the change in NMFS observer data collection purposes in March 1981 (Comparisons la and 2a, Tkble 2). The last two comparisons test for differences in the percent of zero-kill sets in data collected by observers on trips begun during the same time period (Comparisons lb and 2b, Table 2). Before versus After The percent of zero-kill sets for NMFS-observed trips was higher before March 1981 than after that date (Fig. 1), and within all area-period strata with complete data, the percent of zero-kill sets was larger before March 1981 (Tkble 3). The one-sided test of this difference (Comparison la, Tkble 2) was signi- ficant within four of the five area-period strata which had complete data, and was very nearly significant within the fifth (Tkble 3). Thus, the percent of zero- kill sets recorded by NMFS observers was signifi- cantly larger before March 1981. The significant difference in percent of zero-kill sets for NMFS observers before compared with after March 1981 could be due to the change in data collec- tion purposes of NMFS observers which occurred then. Alternatively, the difference could be due to a temporal decline begun before that date. Allowing for period, the data prior to March 1981 do not show a pronounced trend for any of the three areas (Fig. 3). Although there appears to be a decline in the South for Period 2, this is unreliable as it depends entirely on the 1980 and 1981 data points representing a total of only 17 sets. Similarly, there seems to be a declining trend for the two northern areas. However, for the North Inside area the Period 1 points show no decline, and the possible decline of Period 2 points depends on the 1980 Period 2 point. A 95% confidence interval about this point (observed percent +2 x standard error), however, is large relative to the difference between it and the Period 2 point of 1979. Further, any such declining trend in Period 2 points for the North Inside area is not reflected in the low 1978 point. A similar argu- ment can be made for North Outside area data to reject the alternative explanation of the difference in percent zero-kill sets before and after March 1981 being the result of a temporal trend begun prior to March 1981. That the differences in percent of zero-kill sets for NMFS-observed trips was not due to a temporal trend was also tested by comparing the percent of such sets for lATTC-observed trips before and after March 1981. The percent of zero-kill sets for lATTC- observed trips was higher before March 1981, but within the six area-period strata the differences were not consistent (Tkble 4). The two-sided test (Com- parison 2a, Tkble 2) was significant within only one of the six area-period strata (Period 1, South), and the sample size within that stratum was very small (Tkble 4). The overall conditional test given area and period was not significant. Thus, I ATTC -observed trips with dolphin sets from 1979 through 1982 did Table 3.— Numbers of dolphin sets (n) made during NMFS-observed trips, 1978 through 1982. Sets are classified by trip departure date relative to March 1981 (before or after) and to period (1 - Jan.-Mar, 2 = Apr-Dec), by area of set (North Inside, North Outside, South), and by numbers of dolphins killed (0, X)). Percents of column totals (%), expected frequencies (e), and the statistic Z are also tabulated. Values of Z > 1.64 are significant, as indicated by an asterisk. Period 1 Period 2 Tol North Inside Before After North Outside Before After South North Inside Before After North Outside Before After South tal Kill Before After Before After Before After 0 n 1,498 226 0 21 107 0 972 421 591 229 86 33 3,254 930 % 72.9 62.1 — 52.5 61.1 0.0 69.4 62.7 60.4 55.7 50.6 23.1 68.1 56.9 e 1,464.7 259.3 0 21 104.0 3.0 941.9 451.1 577.4 242.6 64.6 54.4 — — >0 n 558 138 0 19 68 5 429 250 387 182 84 110 1,526 704 % 27.1 37.9 — 47.5 38.9 100.0 30.6 37.3 39.6 44.3 49.4 76.9 31.9 43.1 e 591.3 104.7 0 19 71.0 2.0 459.1 219.9 400.6 168.4 105.4 88.6 — — Total n 2,056 364 0 40 175 5 1,401 671 978 411 170 143 4,780 1,634 Z 4.18* 2.75* 3.0- 1* 1.63 4.99* .1 'Computation of overall test statistic not possible because of one-sided alternative, and because of lack of data in one stratum (Period 1, North Outside). 525 FISHERY BULLETIN: VOL. 83, NO. 4 532 528 978 527 403 346 143 326 364 345 NIS 0 534 0 263 0 181 0 270 40 141 NOS 45 64 75 92 55 14 0 3 5 140 S 80 I- O K lU o. 1978 Figure 3— Number of dolphin sets (upper portion) and percent of zero-kill dolphin sets (lower portion) for each of 3 areas (NIS = North Inside, NOS = North Outside, S = South) by period within year (open symbol = Jan.-Mar., closed symbol = Apr.-Dec). Data are from NMFS-observed trips, 1978 through 1982. Vertical line separates data before and after March 1981. Table 4.— Numbers of dolphin sets (n) made from 1979 through 1982, during lATTC-observed trips. Sets are classified by date of set relative to March 1981 (before or after) and to period (1 = Jan.-Mar., 2 = Apr.-Dec), by area of set (North Inside, North Outside, South), and by numbers of dolphins killed (0, X)). Percents of column totals (%), expected frequencies (e), and the statistic x^ with degrees of freedom (df) are also tabulated. Values of x^ > 3.84 (1 df) or 12.59 (6 df) are significant, as Indicated by an asterisk. Period 1 Period 2 To North Inside Before After North Outside Before After South North Inside Before After North Outside Before After South tal Kill Before After Before After Before After 0 n 537 188 9 3 13 3 709 721 291 328 5 35 1,564 1,278 % 67.1 71.2 26.5 14.3 38.2 12.0 70.0 70.1 58.7 55.6 33.3 24.1 65.4 61.6 e 545.1 179.9 7.4 4.6 9.2 6.8 709.8 720.2 282.7 336.3 3.8 36.2 — — >0 n 263 76 25 18 21 22 304 307 205 262 10 110 828 795 % 32.9 28.8 73.5 85.7 61.8 88.0 30.0 29.9 41.3 44.4 66.7 75.9 34.6 38.4 e 254.9 84.1 26.6 16.4 24.8 18.2 303.2 307.8 213.3 253.7 11.2 108.8 — — Total n 800 264 34 21 34 25 1,013 1,028 496 590 15 145 2,392 2,073 X2 1.53 1. 13 5.02* 0.01 1.04 0.61 9.34 df 1 1 1 1 1 1 6 526 WAHLEN AND SMITH: OBSERVER EFFECT ON INDICENTAL DOLPHIN MORTALITY not differ significantly in their percent of zero-kill sets before or after March 1981. Same Time Period Before March 1981 the percent of zero-kill sets was higher for NMFS-observed trips than for lATTC- observed trips (Fig. 4), and within all area-period strata with complete data, the percent of zero-kill sets was larger for the NMFS observers (Ikble 5). The one-sided test (Comparison lb, Ikble 2) was significant within four of the five area-period strata which had complete data (Tkble 5). Thus, for trips making dolphin sets from 1979 through March 1981, NMFS observers recorded a significantly higher per- cent of zero-kill sets than did lATTC observers. According to our hypothesis, the difference in per- cent of zero-kill sets between NMFS- and lATTC- observed trips before March 1981 should have disap- peared after March 1981 when the purposes for data collection of NMFS observers became nearly the same as for lATTC observers. After March 1981 the percent of zero-kill sets was higher for lATTC- observed trips than for NMFS-observed trips (Fig. 5), but within the six area-period strata the dif- ferences were not consistent (Table 6). The two-sided test (Comparison 2b, Tab\e 2) was significant within only one of the six area-period strata (Period 1, North Inside), yet this one chi-square statistic was so large that the overall conditional test for all six strata was also significant (Ihble 6). It is difficult to interpret the overall result in this situation because of the extraordinary influence of one stratum. However, after March 1981 the percent of zero-kill sets on NMFS-observed trips was clearly not higher than on lATTC-observed trips. While one would expect the mean number of dolphins killed to decrease when the percent of zero- kill sets increases, this is not necessarily so because of the sensitivity of the mean of a sample to the max- imum value in the sample For instance, in Figure 4 the NMFS maximum is nearly twice that of the lATTC, resulting in a larger NMFS mean despite the higher percent of zero-kill sets in the NMFS sample >■ O z Ui o lU e 11. lU > UI flC NMFS lATTC w TRIPS 119 104 DOLPHIN SETS 3,080 2,392 MEAN 2.87 2.75 STD. DEV. 21.58 16.28 MAXIMUM 854 447 I ^ . — p^ T =p =f 0 1-9 10-19 20-29 30-39 40-49 50-99 ^100 NUMBER OF DOLPHINS KILLED Figure 4.— Relative frequency distributions of number of dolphins killed incidentally during sets made from 1979 through March 1981, by observer type 527 FISHERY BULLETIN: VOL. 83, NO. 4 >- u z lU 3 o lU tr u. UJ > u c NMFS lATTC m TRIPS 64 85 DOLPHIN SETS 1,675 2,073 MEAN 6.16 3.60 STD. DEV. 21.98 13.97 MAXIMUM 387 268 J yf/!A [ W— ~ 1-9 10-19 20-29 30-39 40-49 60-99 ^100 NUMBER OF DOLPHINS KILLED Figure 5.— Relative frequency distributions of number of dolphins killed incidentally during sets made after March 1981 through 1982, by observer type Table 5.— Numbers of dolphin sets (n) made from 1979 through March 1981. Sets are classified by observer type (NMFS, lATTC), by area of set (North Inside, North Outside, South), by date of set relative to period (1 = Jan.-Mar., 2 = Apr.-Dec), and by numbers of dolphins killed (0, >0). Percents of column totals (%), expected frequencies (e), and the statistic Z are also tabulated. Values of Z > 1.64 are signifi- cant, as indicated by an asterisk. Period 1 Period 2 North Inside NMFS lATTC North Outside NMFS lATTC South North Inside North Outside NMFS lATTC South Total Kill NMFS lATTC NMFS lATTC NMFS lATTC NMFS lATTC 0 n 820 537 0 9 96 13 942 709 267 291 62 5 2,187 1,564 % 72.7 67.1 — 26.5 53.9 38.2 76.8 70.0 60.1 58.7 59.6 33.3 71.0 65.4 e 793.9 563.1 0 9 91.5 17.5 904.0 747.0 263.6 294.4 58.6 8.4 — — >0 n 308 263 0 25 82 21 284 304 177 205 42 10 893 828 % 27.3 32.9 — 73.5 46.1 61.8 23.2 30.0 39.9 41.3 40.4 66.7 29.0 34.6 e 334.1 236.9 0 25 86.5 16.5 322.0 266.0 180.4 201.6 45.4 6.6 — — Total n 1,128 800 0 34 178 34 1,226 1,013 444 496 104 15 3,080 2,392 Z 2.64* - 1.68* 3.66* 0.46 1.92* 1 'Computation of overall test statistic not possible because of one-sided alternative, and because of lack of data In one stratum (Period 1, North Outside). DISCUSSION AND CONCLUSIONS We established the existence of an observer effect on the number of dolphins killed incidentally in the ETP yellowfin tuna fishery by following two lines of argument. First, we demonstrated a decrease in the percent of sets with no dolphins killed on NMFS- observed trips after March 1981, when monitoring compliance with dolphin-release regulations was removed as a data collection purpose (Ikble 3). We further showed that this difference was not due to a temporal trend in fishing conditions by examin- 528 WAHLEN AND SMITH: OBSERVER EFFECT ON INCIDENTAL DOLPHIN MORTALITY Table 6.— Numbers of dolphin sets (n) made from April 1981 through 1982. Sets are classified by observer type (NMFS, lATTC), by area of set (North Inside, North Outside, South), by date of set relative to period (1 = Jan.-f\/lar., 2 = Apr-Dec), and by numbers of dolphins killed (0, >0). Percents of column totals (%), expected frequencies (e), and the statistic x^ with degrees of freedom (df) are also tabulated. Values of x^ > 3.84 (1 df) or 12.59 (6 df) are significant, as Indicated by an asterisk. Period 1 Period 2 To North Inside NMFS lATTC North Outside NMFS lATTC South North Inside NMFS lATTC North Outside NMFS lATTC South tal Kill NMFS lATTC NMFS lATTC NMFS lATTC 0 n 193 188 1 3 1 3 455 721 249 328 13 35 912 1,278 % 55.6 71.2 14.3 14.3 10.0 12.0 66.4 70.1 56.1 55.6 15.9 24.1 57.9 61.6 e 216.4 164.6 1.0 3.0 1.1 2.9 470.3 705.7 247.8 329.2 17.3 30.7 — — >0 n 154 76 6 18 9 22 230 307 195 262 69 110 663 795 % 44.4 28.8 85.7 85.7 90.0 88.0 33.6 29.9 43.9 44.4 84.1 75.9 42.1 38.4 e 130.6 99.4 6.0 18.0 8.9 22.1 214.7 322.3 196.2 260.8 64.7 114.3 — — Total n 347 264 7 21 10 25 685 1,028 444 590 82 145 1,575 2,073 X2 15.53* 0.00 0.03 2.63 0.02 2.16 20.37* df 1 1 1 1 1 1 6 ing the data by period (Fig. 3) and by demonstrating the lack of a corresponding change in data collected by lATTC observers (Tkble 4). Second, we demonstrated that before March 1981 the percent of sets with no dolphins killed was higher for NMFS observers collecting data both for esti- mating dolphin kill and for monitoring compliance with dolphin-release regulations than for lATTC observers collecting data only for estimating dolphin kill (Tkble 5). Tb validate this comparison we also demonstrated that the difference disappeared, or perhaps was reversed, following March 1981 when monitoring compliance with dolphin-release regula- tions was removed from the NMFS observers' responsibilities (Tkble 6). Following Johnson and Bolstad (1973), these differences in the data collected by observers differing in their purposes of data col- lection, and hence in their obtrusiveness, imply the existence of an observer effect. In making these comparisons, we stratified the data to account for possible differences in fishing conditions in different geographic areas and throughout the year because both area and time of year are important determinants of dolphin mor- tality. Thus, the differences in the percent of zero- kill sets which we identified cannot be attributed to nonrepresentativeness of the data with respect to area and time of year. We did not attempt to test for other differences in the frequency distributions of kills, such as changes in the percent of moderate or large kill sets. Sets with large numbers of dolphins killed are rare, and are generally associated with unusual circum- stances, such as mechanical failures. The percent of sets v^th 1-9 dolphins killed appears to vary inversely with the percent of sets with zero dolphins killed (Figs. 1, 4, 5). Powers et al.^ showed that the use of some dolphin- release procedures significantly reduces dolphin mortality. Thus, more time and effort expended by the operator on release of dolphins could result in an increase in the frequency of sets with no dolphins killed, and a corresponding decrease in the frequency of sets with 1-9 dolphins killed. A greater tendency for vessel operators to take the additional time in the presence of an observer collecting data for monitoring compliance with dolphin-release regula- tions could account for the differences we have demonstrated. The significantly different relative frequency of zero-kill sets recorded by NMFS observers after March 1981 (Tkble 6) was not expected under our hypothesis. As noted above, this difference was localized to one area-period stratum, and the other five strata were consistent with the null hypothesis of no difference Either this difference is merely a sampling anomaly, or there are differences between observers in more recent years that we have not taken into account. Gulland (1983, p. Ill) described a method of testing for the existence of a tagging effect that is analogous to our indirect method of testing for an observer effect. He suggested comparing the pro- portions of tags returned from fish tagged under poor and good conditions. In both Gulland's and our methods, the absolute magnitude of the effects can- not be estimated. For instance, in Gulland's exam- ple improvement in the conditions under which tags are applied is unlikely to eliminate entirely the tag- ging effect. Similarly, the reduction in observer ob- «Powers, J. E., N. C. H. Lo, and B. E. Wahlen. 1979. A statis- tical analysis on effectiveness of porpoise rescue procedures in reducing incidental mortality. Southwest Fish. Cent. La JoUa Lab., Natl. Mar. Fish. Serv., NOAA, Admin. Rep. LJ-79-7, 29 p. 529 FISHERY BULLETIN: VOL. 83, NO. 4 trusiveness after March 1981 is unlikely to have eliminated entirely the observer effect because the data collected by scientific observers after 1981 were still used to monitor dolphin kills relative to annual kill limits. Observers collecting data that could not be used for monitoring kill limits would be even less obtrusive than the scientific observers, and covert observers would be, of course, completely unobtru- sive Based on our analysis, we would expect that the frequency of zero-kill sets would be lower on unobserved vessels than on vessels with a scientific observer. This lower frequency of zero-kill sets, cou- pled with an increased frequency of sets with 1-9 dolphins killed, suggests that the average kill rate on unobserved vessels would be higher. Estimates of total kill, based on the average kill rates from the scientific observers, would therefore be underesti- mated. ACKNOWLEDGMENTS We gratefully acknowledge the assistance of M. M. Groom in providing data and in early discussions of the problem. We also acknowledge the assistance of K. E. Wallace and J. S. Cole in providing data. We are indebted to D. G. Ghapman and W. F. Perrin for suggesting analytical methods. We also ap- preciate the helpful reviews by F. G. Alverson, I. Bar- rett, P. S. Hammond, R. S. Holt, N. G. H. Lo, N. A. Mendes, J. M. Michalski, G. T. Sakagawa, and K.-T. 'Rai, as well as the reveiws by two anonymous individuals. LITERATURE CITED Federal Register. 1977. Department of Commercei NOAA, Ikking of marine mammals incidental to commercial fishing operations; final decision and final regulations. Fed. Regist. 42(247);64548- 64560. Federal Register. 1980. Department of Commerce, NOAA, Taking of marine mammals incidental to commercial fishing operations- permits, eta Fed. Regist. 45(213):72178-72196. Fienberg, S. E. 1980. The analysis of cross-classified categorical data. 2d ed. MIT Press, Cambridge, MA, 198 p. Glass, G. V., R D. Peckham, and J. R. Sanders. 1972. Consequences of failure to meet assumptions under- lying the fixed effects analyses of variance and covariance Rev. Edua Res. 42(3):237-288. Gulland, J. A. 1983. Fish stock assessment: a manual of basic methods. John Wiley and Sons, N.Y., 223 p. Hammond, P. S., and K.-T I^ai. 1983. Dolphin mortality incidental to purse-seining for tunas in the eastern Pacific Ocean, 1979-81. Rep. Int. Whaling Comm. 33:589-597. Johnson, S. M., and 0. D. Bolstad. 1973. Methodological issues in naturalistic observation: some problems and solutions for field research. In L. A. Hamerlynck, L. C. Handy, and E. J. Mash (editors). Behavior change: methodology, concepts, and practice, p. 7-67. Research Press, Champaign, IL. Lo, N. C. H., J. E. Powers, and B. E. Wahlen. 1982. Estimating and monitoring incidental dolphin mortali- ty in the eastern tropical Pacific tuna purse seine fishery. Fish. Bull., U.S. 80:396-401. Perrin, W F 1969. Using porpoise to catch tuna. World Fishing 18(6): 42-45. PUNSLY, R. G. 1983. Estimation of the number of purse-seiner sets on tuna associated with dolphins in the eastern Pacific Ocean dur- ing 1959-1980. [In Engl, and Span.] Inter-Am. Trop. TUna Comm. Bull. 18:229-299. RiCKER, W E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191, 382 p. Smith, T D. 1983. Changes in size of three dolphin (Stenella spp.) popula- tions in the eastern tropical Pacific Fish. Bull., U.S. 81:1-13. Snedecor, G. W, and W. G. Cochran. 1980. Statistical methods. 7th ed. Iowa State University Press, Ames, lA, 507 p. Webb, E. J., D. T. Campbell, R. D. Schwartz, and L. Sechrest. 1966. Unobtrusive measures: a survey of non-reactive research in the social sciences. Rand McNally, Chic, 225 p. 530 FOOD HABITS OF JUVENILE ROCKFISHES (SEBASTES) IN A CENTRAL CALIFORNIA KELP FOREST Michael M. Singer^ ABSTRACT The diets and feeding morphology of juveniles of seven rockfish species (Scorpaenidae: Sebastes) were investigated in a kelp forest at Stillwater Cove, Carmel Bay, CA. The seven species could be divided into two groups, those which fed primarily on open water prey in the water column and those which fed on substrate-associated prey. Substrate-associated prey were generally larger than open water prey and were eaten by predators with relatively larger heads and mouths and shorter gill rakers. Comparison of juvenile diets and foraging patterns with those of adults showed that both foraged in similar manners and in the same general habitats. The absence of aggressive interactions within or among species and high intra- specific variability of foraging patterns suggests that little interference or exploitative competition was present. Most of the literature on rockfishes deals mainly with aspects of either adult or larval biology (DeLacey et al. 1964; Phillips 1964; Miller and Geibel 1973; Westrheim 1975; Larson 1980a, b, c). Little of this literature deals with the juvenile stage. Juvenile rock- fishes often use nearshore kelp beds as nursery grounds and, during certain times of the year, are the most abundant members of the kelp-forest fish community (Burge and Schultz 1973). The change from pelagic life to a nearshore, reef-dwelling ex- istence represents a major ecological transition. With so many juvenile rockfish species co-occurring in kelp forests, the potential for competitive inter- action is great. Gaining an understanding of these fishes' patterns of resource utilization (eg., food or space) may be important in assessing the mechanisms that affect their survival. The purpose of this study was threefold: 1) to in- vestigate the food habits of juvenile rockfishes oc- curring in a nearshore kelp forest, 2) to determine the extent to which diet differed within and among the various species present, and 3) to compare the relationship between diet and morphology in juvenile and adult forms. The juveniles of seven rockfish species were studied. The adults of six of these species usually occur nearshore, in association with rock reefs: blue rockfish, S. mystinus; black rockfish, S. melanops; olive rockfish, S. serranoides; copper rockfish, S. caurinus; gopher rockfish, S. camatus; and kelp rockfish, S. atrovirens (Hallacher 1977; Roberts 1979). Adults of the seventh species, the canary rock- fish, S. pinniger, occur deeper and usually offshore (Miller and Lea 1972). For purposes of this study, an individual was designated a juvenile upon first appearance in the kelp forest, where fish were usual- ly 25 to 30 mm SL. The fish used in this study generally ranged from 30 to 70 mm SL. Sebastes mystinus and S. pinniger are among the earliest species to recruit, usually appearing as early as April. Sebastes mystinus usually occurs in the water column above 6 m while S. pinniger occurs on the substrate usually near sand/rock interfaces. Sebastes caurinus begins recruiting around late April or May and is found initially among kelp fronds and near the surface Next to arrive are S. ser- ranoides and S. melanops which first appear in May and June Both these species occur mostly in the mid- water within the kelp forest. Sebastes camatus usual- ly begins to recruit in late June and July and, like S. caurinus, is found initially among kelp fronds at the surface Both S. caurinus and S. camatus move down from the canopy and take positions near the bottom after a couple of months (around June or July for S. caurinus and late August or September for S. camatus). The last of these species to recruit is S. atrovirens, which begins to appear in late July and August and occurs in the surface kelp canopy. METHODS AND MATERIALS ^Moss Landing Marine Laboratories, Moss Landing, CA; present address: VANTUNA Research Group, Occidental College, Moore Laboratory of Zoology, 1600 Campus Road, Los Angeles, CA 9004L Manuscript accepted November 1984. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. Study Site All observations and samples were taken between 531 May 1980 to October 1981 at a rock reef at Arrow- head Point in Stillwater Cove, Carmel Bay, CA (about lat. 36°33.6'N, long. 121°56.3'W). Bottom depths at the study site averaged about 12.5 m below MLLW (mean low low water). The reef was roughly tri- angular, bounded by sand flats which sloped into deeper water on all but the nearshore side (Fig. 1). The reef substrate was composed of both high and low relief basaltic rock interspersed with areas of coarse, granitic sand and a few patches of worm tubes (Diopatra sp.). Stillwater Cove is protected from the prevailing north and northwest swells and winds. However, it is exposed to the most severe southerly storms which commonly occur from Oc- tober to February. The rock substrate was covered with an extensive mat of coralline algae and sessile invertebrates. The dominant algal form was the giant kelp, Macrocystis pyrifera, which became very thick in the summer months and represented a major structural compo- nent of the reef. Except for corallines, understory algae (Pterogophora californica, Cystoseira osmun- dacea, and seasonally dense patches of Desmarestia ligulata) were relatively sparse and patchy. Des- marestia abundance increased in the fall as the sur- face kelp canopy decreased. Drift algae of several types entered the reef at times and often became a major microhabitat. It usually occurred along the sand/rock interface or in low patches in the interior of the reef. Field Collections All fish used for stomach content analysis were collected with almxlmxl.Sm opening-closing, diver-held net. The net was constructed of 1/8-in stretch-mesh nylon netting on a 1/4-in PVC frame mouth. The mouth of the net was hinged in the mid- dle with tygon tubing. This allowed easy operation by a single diver in close spaces. Collected fish were brought to the surface and the stomachs injected with 10% Formalin^. The fish were then preserved whole in 10% Formalin, then washed in freshwater, and stored in 70% ethanol. The majority of fish collected for gut analysis were taken from June to August of both 1980 and 1981 with some supplemental collections occurring in Oc- tober and November of 1980. Collections were made during both day and night. ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. FISHERY BULLETIN: VOL. 83, NO. 4 Laboratory Methods Each fish was blotted dry and weighed to the nearest 0.01 g. The entire visceral mass was re- moved, and the stomach and intestine were then separated from the remaining viscera. The contents of each stomach were emptied into a small dish and examined under a binocular dissecting scopa Food items were separated by general taxonomic category (eg., calanoid copepods, zoea larvae, mysid shrimp, etc). These general categories were used because not all prey items found were in good enough condition to identify to species, thus identifying some items further than others could be misleading. Each category was enumerated by number and percent volume The percent of the total stomach volume of each prey type was estimated by spreading the con- tents to a uniform thickness over a background grid and then estimating the area represented by each type Since contents were spread to an equal thick- ness, area and volume were considered directly pro- portional. Digestive state of contents was estimated on a subjective one to five scale (DeWitt and Cailliet 1972). In addition, a subset of each prey type was taken from stomach contents and measured along its longest axis to the nearest 0.005 mm using an ocular micrometer for later estimation of mean sizes of each prey type Analytical Methods A plot of cumulative number of prey types against randomly pooled number of stomachs was con- structed for each species in order to assess adequacy of sample sizes. An asymptotic leveling of this type of plot indicates a sufficient sample size The mean percent by estimated volume (%V) and by number (%N) of each prey type were calculated for each fish species as an average of all values for individual specimens. The mean frequency of occurence (%F0) was also calculated for each prey type in each predator as the number of times the prey type was seen divided by the total of stomachs exam- ined. The importance of each prey type was calculated using the index of relative importance (IRI) described by Pinkas et al. (1971). The IRI used the proportion by amount (%N), volume (%V), and occurrence (%F0) of each prey type (IRI = (%N + %V) X %F0). The diets of the seven species studied were compared using the percent similarity index (PSI) (Whittaker 1952): PSI = 1 - 0.5 I pih - pjh. Where pi and pj are the proportions by IRI of each prey type (h) in the two predators being compared. In calculating 532 SINGER: FOOD HABITS OF ROCKFISH Figure 1— Map of vicinity of sampling area showing Stillwater Cove and the location of the study reef. ASILOMAR STILLWATER COVE 4 () 30 m J I ARROWHEAD POJNT Cormel 533 overlap values %IRI values were used as the propor- tion, p. Morphological Comparison Several measurements were taken on a subsam- ple of fish of each species covering all sizes en- countered in this study. These included standard length, head length, and gape Mean length of the five gill rakers nearest the angle on the ventral limb of the first gill arch, and the gap between them, were also measured. The feeding related morphological characteristics chosen for statistical comparison were head length, gape, gill raker length, and gill raker spacing. Gape and gill raker spacing were compared as absolute measurements. Gill raker lengths were standardized to a proportion of head length and head length was standardized as a proportion of standard length for comparison. Gape was compared using a one-way analysis of covariance (ANCOVA). This analysis regressed the natural log of the ga]je against the natural log of fish standard length to allow comparison over a range of fish sizes and to achieve linearity (Sokal and Rohlf 1969; Chen 1971). Relative head length, gill raker spacing, and relative gill raker length were compared among species with a Model I, one-way analysis of variance (ANOVA). Multiple range comparisons were then made be- tween individual species to detect groupings. Regres- sion lines resulting from the ANCOVA were sub- jected to Newman-Keuls pairwise comparisons. Mean values obtained from ANOVAs were subjected to Student-Newman-Keuls (SNK) multiple range tests. All statistical procedures used followed those presented in Sokal and Rohlf (1969) and Snedecor and Cochran (1980). Prey Abundance Qualitative assessment of prey abundances was made from zooplankton tows done in the kelp forest. A standard 0.5 m diameter net with 0.333 ^m mesh was used. Tbws were done in sets of three, 3 min each: one within 0.5 m of the canopy, one in mid- water, and one within 0.5 m of the bottom. These were done during both day and night. Generally, only presence or absence of plankton types was noted, along with relative daytime vs. nightime differences in abundances. Absolute abundances were not estimated. Zooplankton samples were taken once in mid- June 1980 and once in mid- July 1981. A total of 12 samples were collected. FISHERY BULLETIN: VOL. 83, NO. 4 RESULTS A total of 265 juveniles of the seven species was examined for stomach content analysis: 27 S. melanops (53 to 67 mm SL, x = 57.9); 51 S. ser- ranoides (44 to 63 mm; x = 51.8); 63 5. pinniger (28 to 56 mm, x = 43.7); 23 S. mystiniis (46 to 72 mm, X = 59.1); 38 S. atrovirens (38 to 64 mm, x = 47.1); 42 S. caurinus (35 to 62 mm, x = 48.7); and 21 S. camatus (33 to 65 mm, x = 44.4). An often co- occuring cogener of S. camatus, S. chrysomelas, was not found during this study. Diet Analysis The cumulative prey type curves versus number of stomachs examined leveled off asymptotically, in- dicating that sample sizes for all species were suffi- cient to characterize their food arrays. All species were found to be zooplanktivores, consuming both open water and substrate-associated prey. Open water prey, which mainly float free in the water column, consisted mostly of juvenile copepods, zoea, and juvenile spionid polychaetes, while substrate- associated prey, which were generally found direct- ly associated (<0.5 m away) with a physical struc- ture (rock, kelp, eta), consisted mainly of gammarid amphipods, mysid and caridean shrimp, and two species of isopod (Ikble 1). A wide variety of prey sizes were encountered (Tkble 2). Prey found in open water, such as calanoid and harpacticoid copepods, zoea, and larvaceans, were mostly < 1 mm in length, whereas substrate- associated prey, such as amphipods and decapods, were 3 to 4 mm or larger. Mean number of prey items and prey taxa per in- dividual were highly variable within and among species (Tkble 3). As would be expected, species which fed on small open water prey had higher numbers of prey per stomach. Conversely, those species which fed on large substrate-associated prey, which were often large enough to fill a stomach with a single prey item, had much lower numbers of prey items per stomach. Sebastes mystimis fed mainly on the larvacean Oikopleura sp., with high occurrence of copepods, zoea, and juvenile polychaetes (Fig. 2). They had the most cosmopolitan diet and the highest intraspecific variability of prey types (a mean of almost six dif- ferent prey types per individual) (Tkble 3). Because its diet was composed mainly of small open water prey, this species also had the highest mean number of prey items per individual. Sebastes serranoides and S. melanops had very 534 SINGER: FOOD HABITS OF ROCKFISH Table 1.— Taxa of prey identified in the stonnachs of juvenile rockfish; names in all-capitals are categories used in stomach content analysis. (?) signifies incomplete or unsure identification. Microhabitat associations: OW = open water prey, S = substrate oriented prey. Open water or Prey type substrate association Algae Mostly reproductive Cystociera osmundacea S Occasional pieces of Macrocystis fronds Invertebrates Molluscs - Macoma sp. S Copepods - CALANOIDS: unidentified juveniles OW HARPACTICOIDS: unidentified juveniles OW/S OSTRACODS - Unidentified OW POLYCHAETES - larval/juvenile Spionidae OW Cirripidea - unidentified BARNACLE LARVAE OW ISOPODS - Dimonella globera; Idothea resicata S Amphipods - GAMMARIDS: Aoroides columbiana Ampithoe sp. fiafea transversa Najna kitmata (?) S CAPRELLIDS: Caprella sp. S Unidentified Protocerid HYPERIIDS: Unidentified S MYSIDS - Acanthomysis sp.; Neomysis sp. S Decapods - SHRIMP: Heptocarpus sp. Hippolyte sp. (?) S ZOEA LARVAE: unidentified decapods and OW brachyurans ANOMURANS: unidentified JUVENILE S HERMIT CRAB Larvaceans - Oikopleura sp. OW FISH unidentified hard parts s unidentified fish eggs OW Table 2. — Mean (+95% confidence interval) sizes of individuals of the nine major prey types of juvenile rockfishes. Measured items were taken from stomach samples. Size (mm) Species N X 95% C.I. Open water prey Oikopleura sp. 50 0.69 0.04 Harpacticoid copepods 50 0.88 0.05 Calanoid copepods 50 0.96 0.03 Juvenile polychaetes 35 1.47 0.71 Zoea larvae 50 1.63 0.11 Substrate-associated prey Mysids 35 3.48 1.16 Gammarid amphipods 35 4.01 0.77 Isopods 30 4.29 0.89 Caridean shrimp 30 13.95 1.47 similar diets (Fig. 2, Tkble 4). Both species ate mainly open water prey: calanoid copepods, harpacticoid copepods, and zoea larvae in order of importance These species also showed fairly high intraspecific diet variability— 5.3 and 4.4 prey taxa per individual for melanops and serranoides, respectively. The two species which showed major microhabitat shifts (Carr 1983), S. caurinus and S. camatus, also showed large dietary changes. While in the canopy (generally <45 mm in SL), 5. caurinus ate predominantly calanoid copepods, with harpacticoids and zoea also eaten fairly consistently (Fig. 2). However, after moving out of the canopy and down to the bottom near kelp stipes and rocks, its diet shifted to primarily caridean shrimp, with gammarid amphipods being the second most important prey (Fig. 2). Sebastes camatus showed much the same type of dietary shift accompanying its large habitat shift. While in the canopy, smaller S. camatus ate nearly exclusively calanoid copepods (92% of the diet), while large individuals, which were found on the bottom, fed on caridean and mysid shrimp and isopods (Fig. 2). This marked change in diet can also be seen in Tkble 4. Within each of these species, the diet similarity between size classes was low. However, similarities between the two species within each size class was quite high. Sebastes atrovirens was found in the kelp canopy and among kelp throughout all depths during the entire study (Carr 1983). This species fed mainly on gammarid amphipods, with calanoid and harpac- ticoid copepods and mysid shrimp also being very 535 FISHERY BULLETIN: VOL. 83, NO. 4 Table 3.— Mean (+95% confidence interval) number of prey items and prey taxa per stomach and total number of prey types for the seven species of juvenile rockfishes. Items Taxa S Species N X 95% C.I. X 95% C.I. Total Open water predators Sebastes mystinus 23 142.6 74.9 5.8 0.4 13 S. caurinus (<45 mm) 17 136.4 64.7 3.4 0.7 8 S. carnatus (<45 mm) 12 101.4 79.6 1.9 0.7 5 S. serranoides 50 100.2 33.5 4.4 1.7 9 S. melanops 27 87.0 31.2 5.3 2.3 8 Substrate-oriented predators S. pinniger 59 35.5 18.3 4.3 1.6 15 S. atrovirens 33 34.9 20.3 2.2 0.4 8 S. caurinus (>45 mm) 21 2.5 1.3 1.5 0.3 9 S. carnatus (>45 mm) 11 2.3 1.2 1.6 0.7 5 Table 4.— Percent similarity (PSI) overlap values of juvenile rockfish diets. Proportions used for calculations are %IRI. serranoides melanops mystinus caurinus carnatus pinniger <45 mm <45 mm atrovirens caurinus >45 mm carnatus >45 mm Sebastes serranoides S. melanops S. mystinus S. caurinus (<45 mm) S. carnatus (<45 mm) S. pinniger S. atrovirens S. caurinus (>45 mm) S. carnatus (>45 mm) 0.787 0.455 0.874 0.602 0.653 0.389 0.091 0.090 0.402 0.732 0.516 0.734 0.318 0.094 0.019 — 0.384 0.127 0.458 0.239 0.097 0.043 — 0.668 0.573 0.427 0.185 0.204 — 0.297 0.259 0.043 0.034 0.468 0.174 0.218 0.169 0.258 0.634 important in their diet (Fig. 2). Thus, this spe- cies' diet was similar to most other species {Table 4). Sebastes pinniger had a very diverse diet (highest number of overall prey taxa found; Ikble 3). In- dividuals were found mainly over sand areas and the sand/rock interface at the edge of the kelp forest, generally within a few meters of the bottom. This species' diet consisted mainly of copepods and zoea larvae (open water prey), but gammarid amphipods and mysid shrimp (substrate-associated prey) were also important (Fig. 2). Figure 2 shows that these seven species can be split into two basic categories: open water and substrate-associated predators (this categorization can also be seen in microhabitat differences [Carr 1983]). Open water predators— S. melanops, S. mystinus, S. serranoides, and small S. caurinus and S. carnatus— axQ those that ate mainly copepods and zoea larvae. Substrate-oriented predators— S. atrovirens and larger 5. caurinus and 5. carnatus— are those that ate predominantly amphipods and decapods. Sebastes pinniger is intermediate between these categories. Its microhabitat and behavior are that of a substrate-associated predator, but its diet is more similar to the open water predators (Tkble 4). Prey Distribution Midwater organisms such as copepods, zoea lar- vae, and polychaetes were very abundant in all parts of the water column. Calanoid copepods were the most abundant type during the day at all depths, while both calanoids and harpacticoids were very abundant at night. Amphipods, mysids, caridean shrimp, and isopods were very abundant in and around all substrate types (kelp canopy, stipes, rocks, and drift algae). Isopods and amphipods were most abundant in the canopy and stipes, while amphipods, mysids, and carideans were more often abundant near rock and drift algae and around kelp holdfasts. These invertebrates remained near substrate dur- ing the day, moving farther away at night. Morphological Comparisons ANOVAs of relative head length, relative gill raker length, and gill raker spacing among species were all significant at the P < 0.001 level (Tkble 5). Generally, the open water and substrate-associated classifications also held true for groupings by morphology. In general, open water predators had smaller heads and larger gill rakers. Sebastes mystinus, S. 536 SINGER: FOOD HABITS OF ROCKFISH en X a c o o cS o > -a c 0) a. >» u o c > 3 a o c o 'to C 30 cm and winds speeds exceeded 13 km/h (see also Dudok van Heel 1962; Andersen and Dziedzic 1964). ^Frost, K. J., L. F. Lowry, and R. R. Nelson. 1983. Investi- gations of belukha whales in coastal waters of western and north- ern Alaska, 1982-1983; marking and tracking of whales in Bristol Bay. Final Report, Contract NA 81 RAC 00049, 104 p. J__J LI- ACTIVITY LEVEL HIGH u. ■ 11 ■ ■ J I I I L JJ I l—L _■_■ I L Patterns of Activity In total, 39.2 h of ventilation sequences were recorded from four radio-tagged porpoises (RT-2, RT-3, RT-4, RT-7). These sequences comprised 4,680 individual dives, lasting from 2 to 195 s. Two types of signals were received from radio- tagged animals. The most common signal was brief (1-3 s) and indicated that the porpoise had surfaced and submerged in a continuous motion. Such action patterns are commonly referred to as rolls (Amun- din 1974; Smith et al. 1976). Other signals were more prolonged (4-100 s) and are referred to here as sur- face periods. Prolonged signals received from radio-tagged har- bor porpoises have previously been interpreted as near-surface swimming (Gaskin et al. 1975). However, visual observations of radio-tagged animals RT-2 and RT-3 indicated that such signals originated from porpoises resting motionless at the surface. The strength of the transmitted signal attenuated rapidly as the length of exposed antenna decreased, making it unlikely that signals could be received at any ■ ■ ■ ■ -60 seconds- ACTIVITY LEVEL LOW -60 secoods- FlGURE 5.— Examples of signal patterns used to derive activity states of radio-tagged harbor porpoises (each example represents a continuous record). Activity level was considered high when signals were dominated by rolls (signal duration 1-3 s). Activity level was considered low when signals were dominated by surface resting intervals (signal duration >3 s). The signal pattern used to demonstrate the high activity level (top) is characteristic of Pat- tern B respiration (Watson and Gaskin 1983). 549 FISHERY BULLETIN: VOL. 83. NO. 4 Radio-tagged porpoises exhibiting the high activity state expressed two ventilation patterns; these are described using the terminology of Watson and Gaskin (1983). Most data recorded in this activity state consisted of Pattern B, a series of long dives, each followed by a sequence of several rolls (see Figure 5). Less commonly observed was Pattern A, in which single rolls followed relatively short submergences (seldom exceeding 30 s in duration). Pattern A was exhibited for brief periods only (5-16 min) and comprised <4% of all signals recorded dur- ing high activity sequences. Ventilation data recorded from RT-2 and RT-4 were dominated by low activity sequences. However, low activity sequences were not recorded from either RT-3 or RT-7. Although RT-3 was frequently observed resting at the surface, the loose transmitter package (see above) caused the antenna to reflect backwards, allowing signal reception only during rolls. Thus, it was not possible to accurately monitor the duration of resting periods for this porpoise Data from RT-7 were acquired only during periods of high winds and heavy seas which precluded surface resting behavior. Because surface resting was the criterion on which determinations of activity levels were based, it was impossible to ascertain the activity level of radio- tagged porpoises in periods of high vdnds and heavy seas. Tb construct an activity budget, therefore, it was necessary to exclude data recorded during periods when surface resting was not possible A total of 10.5 h of ventilation sequences were record- ed under such conditions. In addition, data acquired from RT-3 were excluded because of the bias imposed by the transmitting system. After these data had been deleted, 24.5 h of ventilation sequences record- ed from RT-2 and RT-4 remained. Both RT-2 and RT-4 were relatively inactive from midnight until 0600, spending over 90% of this period in the low activity state Both porpoises spent a considerable portion of this time resting at the sur- face (Ikble 2). During this period of reduced activity, the porpoises were seldom located in nearshore areas, although they frequented such areas during other periods. The two porpoises were highly active for 35% (RT-2) and 36% (RT-4) of daylight and even- ing hours (0600 until midnight) (Ikble 2). DISCUSSION Movements and Ranges Radio-tagged harbor porpoises demonstrated con- siderable mobility within the study area, often moving distances of 15-20 km in a 24-h period. These results are similar to those previously reported from radio-tagged harbor porpoises in the region (Gaskin et al. 1975). Other inshore odontocete species exhibit daily movements of a similar magnitude For exam- ple, dusky dolphins, Lagenorhynchus obscurus, track- ed by Wiirsig (1982), travelled a "mean minimum distance" of 19.2 km each day. However, pelagic species apparently travel over much greater distances. A pelagic spotted dolphin, Stenella at- tenuata, tracked by Leatherwood and Ljungblad (1979), travelled over 100 km in a 12-h period, while common dolphins, Delphintcs delphis, may cover distances of 70-140 km each day (Evans 1971). The mobility exhibited by the majority of radio- tagged porpoises suggest that the ranges of these animals were similar to that calculated for RT-2 (210 km^). Only one other study has examined the areas of home ranges utilized by odontocete cetaceans. Wells et al. (1980) used resightings of naturally mark- ed animals to estimate the size of bottlenose dolphin, Tursiops truncatus, ranges in the coastal waters of western Florida. The mean home ranges of these dolphins varied with age and sex, and ranged from 15 to 41 km^. It is possible that the apparent dif- ference in the size of home ranges of these two species reflects the exploitation of different prey species. In the Bay of Fundy, harbor porpoises feed predominantly on juvenile herring, Clupea harengvs (Smith and Gaskin 1974), which exhibit a high degree of mobility (Jovellanos and Gaskin 1983). In contrast, Florida bottlenose dolphins are opportunistic predators on species such as mullet Mugil cephalus, which may be more sedentary in nature (Irvine et al. 1981). Patterns of Activity The patterns of activity observed in the present Table 2.— Activity patterns of radio-tagged harbor porpoises RT-2 and RT-4 in the western Bay of Fundy. The low activity state was characterized by frequent surface resting periods, which were in- frequent or absent in the high activity state. Only data recorded dur- ing calm conditions have been included. Observation Activity: Activity: At time high low surface Porpoise Time (min) (%) {%) (0/0) RT-2 0000-0559 352.9 2.0 98.0 31.4 0600-1159 274.8 14.9 85.1 18.8 1200-1759 435.2 46.0 54.0 11.2 1800-2359 165.2 41.0 59.0 12.2 Total 1,228.5 25.7 74.3 18.8 RT-4 0000-0559 116.0 7.0 93.0 18.5 0600-1159 37.0 100.0 0.0 0.0 1200-1759 0.0 — — — 1800-2359 90.7 9.9 90.1 13.6 Total 243.7 22.2 77.8 13.9 550 READ and GASKIN: RADIO TRACKING HARBOR PORPOISES study do not support previous contentions that the metaboHc requirements of harbor porpoises (see Kanwisher and Sundnes 1965) are such that in- dividuals must spend a large proportion of each day engaged in foraging behavior (Smith and Gaskin 1974; Watson and Gaskin 1983). Herbers (1981) has hypothesized that behavioral inactivity is a product of predation efficiency. As predation efficiency increases, less time is spent searching for and capturing prey, and more time is available for other behavior, including inactivity. Therefore, if harbor porpoises are efficient predators, it seems reasonable to suggest that only a small por- tion of their day would be spent engaged in foraging behavior. Many other mammalian predators are inactive for large portions of the day. For example, Serengeti lions, Panthera leo, are inactive for about 85% of each day (Schaller 1972). Similarly, spotted hyaenas, Crocuta crocuta, are inactive for 84% of the day (Kruuk 1972). Even the sea otter, Enhydra lutris, with a metabolic rate 2.4 times that predicted for a terrestrial mammal of equal size (Costa and Kooyman 1982), spends only 34% of each day for- aging (Loughlin 1979). The ventilation sequences recorded from RT-2 and RT-4 suggest that these harbor porpoises restricted the majority of their activity to daylight and even- ing hours (Tkble 2). If a circadian pattern of activity exists, it may be related to the schooling behavior of prey species. The structure of herring schools breaks down after dusk, as the visual cues used to maintain school structure become inoperative (Brawn 1960). Thus, the fish exhibit a dispersed distribution at night, presumably limiting prey cap- ture by predators such as the harbor porpoises, which rely on dense schools to maintain maximum capture efficiency. Other odontocete species exhibit various circadian patterns of activity. Observations of captive bottle- nose dolphins indicate that, like the harbor porpoise, Tursiops is relatively inactive at night (McBride and Hebb 1948; McCormick 1969; Saayman et al. 1973). In contrast, Hawaiian spinner dolphins, Stenella longirostris, rest during the day and feed almost ex- clusively at night (Norris and Dohl 1980). The prey of spinner dolphins undertake extensive vertical migrations (Perrin et al. 1973) and may be more available to the dolphins at night. We were interested in observing the nocturnal behavior of harbor porpoises (when they were presumably relatively inactive) under conditions of strong winds and heavy seas, when surface resting was not possible Ventilation data recorded from RT-7 during a 5-h period (0000-0500, 5 September 1982) of heavy seas consisted almost exclusively of Pattern B sequences. Watson and Gaskin (1983) have sug- gested that this ventilation pattern is expressed primarily by foraging porpoises, but it seems unlikely that RT-7 (a calf) was foraging for 5 consecutive hours at night. An alternative explanation is that the porpoise was resting underwater and rising to the surface for a series of breaths (see similar observa- tions by McBride and Hebb 1948; Layne 1958; McCormick 1969; Condy et al. 1978). It is possible, therefore, that harbor porpoises engaged in diverse behavioral activities may exhibit similar ventilation patterns. During the period of reduced activity (from 0000 to 0600) radio-tagged porpoises were often located in open water some distance from shore This may reflect a tendency for porpoises to rest in areas where the hazards of swift currents and shallow waters are minimized. Observations made in the in- shore waters of the Deer Island region confirm that porpoises seldom rest at the surface in nearshore environments (Watson and Gaskin 1983). ACKNOWLEDGMENTS We thank W. Kozak and the members of the Fundy Weir Fishermen Association for their assistance in this study. Sterling field assistance was provided by C. Thomson and members of the Fundy Cetacean and Seabird Research Group. Constructive criticism of earlier versions of this paper were provided by B. Braune, L. Murison, P. Watts, L. White, and two anonymous reviewers. This research was supported by Joint Contract UP-G-152 (Departments of Sup- ply and Services and Fisheries and Ocean Canada). Harbor porpoises were tagged under a permit issued by Fisheries and Oceans Canada. LITERATURE CITED Amundin, M. 1974. Functional analysis of the surfacing behaviour in the harbour porpoise, Phocoena phocoena (L.). Z. Saugetierkd. Bd. 39:313-318. Andersen, S., and A. Dziedzic. 1964. Behaviour patterns of captive harbour porpoise Pho- coena phocoena L. Bull. Inst. Oceanogr. Monaco 63(1316): 1-20. Anonymous. 1982. Canadian tide and current tables. Vol. 1. The Atlantic coast and Bay of Fundy. Department of Fisheries and Oceans Canada, Ottawa. Bailey, W. B., D. G. MacGregor, and H. B. Hachey 1954. Annual variations in temperature and salinity in the Bay of Fundy. J. Fish. Res. Board Can. 11:32-47. 551 FISHERY BULLETIN: VOL. 83, NO. 4 Bigg, M. A. 1982. An assessment of killer whale {Orcinus orca) stocks off Vancouver Island, British Columbia. Rep. Int. Whaling Comm. 32:655-666. Brawn, V. M. 1960. Seasonal and diurnal vertical distribution of herring (Clupea harengiis L.) in Passamaquoddy Bay, N.B. J. Fish. Res. Board Can. 17:699-711. CONDY, P. R., R. J. VAN AaRDE, AND M. N. BESTER. 1978. The seasonal occurrence and behaviour of kOler whales, Orcinus orca, at Marion Island. J. Zool. (Lond.) 184:449-464. Costa, D. P., and G. L. Kooyman. 1982. Oxygen consumption, thermoregulation, and the effect of fur oiling and washing on the sea otter, Enhydra lutris. Can. J. Zool. 60:2761-2767. DORSEY, E. M. 1983. Exclusive adjoining ranges in individually identified minke whales (Balaenoptera acutorostrata) in Washington State Can. J. Zool. 61:174-181. DuDOK VAN Heel, W H. 1962. Sound and Cetacea. Neth. J. Sea Res. 1:407-507. Evans, W. E. 1971. Orientation behavior of delphinids: radio telemetric studies. Ann. N.Y. Acad. Sci. 188:142-160. Forrester, W. D. 1960. Current measurements in Passamaquoddy Bay and the Bay of Fundy J. Fish. Res. Board Can. 17:727-728. Gaskin, D. E., G. J. D. Smith, and A. P. Watson. 1975. Preliminary study of harbor porpoises {Phocoena pho- coena) in the Bay of Fundy using radiotelemetry. Can. J. Zool. 53:1466-1471. Herbers, J. M. 1981. Time resources and laziness in animals. Oecologia 49:252-262. Irvine, A. B., M. D. Scott, R. S. Wells, and J. H. Kaufmann. 1981. Movements and activities of the Atlantic bottlenose dolphin, Tursiops truncatus, near Sarasota, Florida. Fish. Bull., U.S. 79:671-688. Irvine, A. B., R. S. Wells, and M. D. Scott. 1982. An evaluation of techniques for tagging small odon- tocete cetaceans. Fish. Bull., U.S. 80:135-143. Jovellanos, C. L., and D. E. Gaskin. 1983. Predicting the movements of juvenile Atlantic herring {Clupea harengus harengus) in the SW Bay of Fundy using computer simulation techniques. Can. J. Fish. Aquat. Sci. 40:139-146. Kanwisher, J., and G. Sundnes. 1965. Physiology of a small cetacean. Hvalradets Skr. 48:45-53. Kruuk, H. 1972. The spotted hyaena. A study of predation and social behaviour. Univ. Chicago Press, Chia, 335 p. Layne, J. N. 1958. Observations on freshwater dolphins in the upper Amazon. J. Mammal. 39:1-22. Leatherwood, S., and W E. Evans. 1979. Some recent uses and potentials of radiotelemetry in field studies of cetaceans. In H. E. Winn and B. L. Olla (editors). Behavior or marine animals. Vol. 3: Cetaceans, p. 1-31. Plenum Press, N.Y. Leatherwood, S., and D. K. Ljungblad. 1979. Nighttime swimming and diving behavior of a radio- tagged spotted dolphin, Stenella attenuata. Cetology 34, 6 p. LOUGHLIN, T. R. 1979. Radiotelemetric determination of the 24-hour feeding activities of sea otters, Enhydra lutris. In C. J. Amlaner and D. W Macdonald (editors), A handbook on biotelemetry and radio tracking, p. 717-724. Pergamon Press, Oxf. Macdonald, D. W, F. G. Ball, and N. G. Hough. 1979. The evaluation of home range size and configuration using radio tracking data. In C. J. Amlaner and D. W Mac- donald (editors), A handbook on biotelemetry and radio track- ing, p. 405-424. Pergamon Press, Oxf. McBride, a. F, and D. 0. Hebb. 1948. Behavior of the captive bottlenose dolphin, Tursiops truncatus. J. Comp. Physiol. Psychol. 41:111-123. McCoRMiCK, J. G. 1969. Relationship of sleep, resj; iration, and anesthesia in the porpoise: a preliminary report. Proc Nat. Acad. Sci., U.S. 62:697-703. NoRRis, K. S., and T R Dohl. 1980. Behavior of the Hawaiian spinner dolphin, Stenella longirostris. Fish Bull., U.S. 77:821-849. Perrin, W F., R. R. Warner, C. H. Fiscus, and D. B. Holts. 1973. Stomach contents of porpoise, Stenella, spp., and yellowfin tima, Thunnus albacares, in mixed-species aggrega- tions. Fish. Bull, U.S. 71:1077-1092. Read, A. J., and D. E. Gaskin. 1983. The application of radio tracking techniques to the study of harbour porpoises {Phocoena phocoena) in the Bay of Fundy. In D. G. Pincock (editor). Proceedings of the Fourth International Conference on Wildlife Biotelemetry, p. 346-352. Halifax, Nova Scotia. Saayman, G. S., C. K. Tayler, and D. Bower. 1973. Diurnal activity cycles in captive and free-ranging bottlenose dolphins {Tursiops aduncus Ehrenburg). Behaviour 44:212-233. SCHALLER, G. B. 1972. The Serengeti lion. A study of predator-prey relations. Univ. Chicago Press, Chic, 480 p. Smith, G. J. D., K. W. Browne, and D. E. Gaskin. 1976. Functional myology of the harbour porpoise, Phocoena phocoena (L.). Can. J. Zool. 54:716-729. Smith, G. J. D., and D. E. Gaskin. 1974. The diet of harbour porpoises {Phocoena phocoena (L.)) in coastal waters of eastern Canada, with special reference to the Bay of Fundy Can. J. Zool. 52:777-782. Smith, G. J. D., C. L. Jovellanos, and D. E. Gaskin. 1984. Near-surface bio-oceanographic phenomena in the Quoddy region. Bay of Fundy. Can. Itech. Rep. Fish. Aquatic Sci. No. 1280, 124 p. Smith, G. J. D., A. J. Read, and D. E. Gaskin. 1983. Incidental catch of harbor porpoise, Phocoena phocoena (L.), in herring weirs in Charlotte County, New Brunswick, Canada. Fish. Bull., U.S. 81:660-662. Springer, J. T. 1979. Some sources of bias and sampling error in radio tri- angulation. J. Wildl. Manage 43:926-935. Watson, A. P., and D. E. Gaskin. 1983. Observations on the ventilation cycle of the harbour por- poise Phocoena phocoena (L.) in coastal waters of the Bay of Fundy Can. J. Zool. 61:126-132. Wells, R. S., A. B. Irvine, and M. D. Scott. 1980. The social ecology of inshore odontocetes. In L. M. Herman (editor), Cetacean behavior: mechanisms and func- tions, p. 263-317. John Wiley and Sons, N.Y. WiJRSiG, B. 1982. Radio tracking dusky porpoises in the South Atlantic In Mammals in the seas, Vol. 4, p. 145-160. FAO Fisheries Series No. 5. 552 EARLY POSTNATAL GROWTH OF THE SPOTTED DOLPHIN, STENELLA ATTENUATA, IN THE OFFSORE EASTERN TROPICAL PACIFIC Aleta a. Hohn' and p. S. Hammond^ ABSTRACT Estimates of length at birth and early postnatal growth are made for the northern and southern popula- tions of the offshore spotted dolphin in the offshore eastern tropical Pacific Length at birth is estimated to be 85.4 cm for the northern population and 83.2 cm for the southern population. Analyses of series of monthly distributions of length revealed two cohorts born each year in the northern population, at least in the northern inshore part of its geographic range, but only one cohort born each year in the southern population. Growth curves fitted to the means of the monthly distributions of length gave estimates of length at 1 year of 126.2 and 132.6 cm and length at 2 years of 154.3 and 154.9 cm for the two cohorts in the northern population, and length at 1 year of 127.9 cm for the southern population. A growth curve fitted to lengths and ages (in dental growth layer groups) from the northern population gave estimates of lengths at 1 and 2 years of 123.0 and 143.0 cm, respectively. The average growth rate of individual animals in a population is an important characteristic because of its correlation with other population parameters. In fisheries biology, two commonly employed techniques used to estimate growth rates are the aging of a sam- ple of fish of known length and the following of a series of length distributions through time These techniques allow the relationship between length and age (or relative age) to be applied to a much larger sample of fish, provided that the aged sample is a representative ona For most species of fish, length-age relationships may be appropriate for the entire life of the animals, or at least for the period of interest to a commer- cial fishery. In marine mammals, however, length changes little, if at all, after attainment of physical maturity. Growth rates may change markedly even while the animal is maturing, being high for an ini- tial period after birth and then declining quite rapid- ly. In delphinids, the growth rate has been found to be high in the first year, with animals typically in- creasing by 50-70% of their birth length (Sergeant 1962; Kasuya et al. 1974; Kasuya 1976; Miyazaki 1977; Hohn 1980; Perrin and Henderson 1984), but then declined rapidly in the second year. During this 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, J^OAA, P.O. Box 271, La Jolla, CA 92038. ^Inter-American Tropical Tlina Commission, Scripps Institution of Oceanography, La Jolla, CA 92098; present address: Sea Mam- mal Research Unit, c/o British Antartic Survey, Madingley Road, Cambridge CBS OET, United Kingdom. period, growth rates are high relative to the varia- bility in age-at-length so that length distributions are distinguishable as separate age groups. Consequent- ly, length-age relationships for these animals are most useful from birth until about 2 yr. In this paper, we have used both the technique of following a series of length distributions from month to month and the technique of aging a sample of dolphins of known length to estimate the rate of growth in the spotted dolphin, Stenella attenuata, in the offshore eastern tropical Pacific (hereafter referred to as the offshore spotted dolphin). In neither of these two techniques did we have an ab- solute measure of age. Consequently, we have esti- mated length at birth independently and used this to fix time at birth. Growth curves were fitted to the length data by relative age and then length at birth was substituted in order to predict length-at- age THE SAMPLE The field data and specimens used in the follow- ing analyses were collected by National Marine Fisheries Service (NMFS) and Inter-American Tropical lUna Commission (lATTC) scientific techni- cians aboard commercial tuna purse seiners from 1968 to 1982. Procedures for collecting sample data and specimens have been described by Perrin et al. (1976). In all the following analyses, the data have been stratified into northern and southern popula- Manuscript accepted December 1984. FISHERY BULLETIN: VOL. 83. NO. 4, 1985. 553 FISHERY BULLETIN: VOL. 83, NO. 4 tions divided by lat. 1°S, based on a division selected by Perrin et al. (1S79). Areas 1 and 2 in Figure 1 correspond to the region occupied by the southern offshore spotted dolphin and areas 3-8 correspond to the region occupied by the northern offshore spot- ted dolphin. ESTIMATION OF LENGTH AT BIRTH An accurate estimate of length at birth is impor- tant because it establishes a point through which any growth curve should pass. This extra degree of freedom allows greater accuracy in fitting growth curves and estimating growth rates. Neither of our methods of relating length to age, described below, allows us to fix absolute age so it is essential here for us to calculate an independent estimate of length at birth. A commonly used method of estimating length at birth when a sufficient sample is available is to re- gress the percent postnatal at each length interval on length and to calculate the length at which 50% of the specimens are predicted to be postnatal. This method, using a linear model, has been employed to estimate length at birth for spotted dolphins (Per- rin et al. 1976), for striped dolphins, S. coeruleoalba, (Miyazaki 1977), and for spinner dolphins, S. longirostris, (Perrin et al. 1977), and using a nonlinear model for spinner dolphins (Perrin and Henderson 1984). Another method commonly employed when a small sample is available is to estimate the average length at birth as the mean length of known neonates or the mean length of full- term fetuses and small calves combined. This method has been used to estimate length at birth for long- finned pilot whales (Sergeant 1962), for spotted dolphins (Kasuya et al. 1974), for bottlenose dolphins, Tursiops truncatus, (Ross 1977; Hohn 1980), and for I'll \tO lEE lED MS MO tSS 130 IZS IZO IIS 110 lOE 100 35 Figure L— Areas inhabited by the offshore spotted dolphin. The numbered regions refer to strata investigated in analyses of monthly distributions of length. In all analyses, the southern population is from areas 1 and 2, and the northern population is from areas 3 through 8. 554 HOHN and HAMMOND: POSTNATAL GROWTH OF SPOTTED DOLPHIN the franciscana dolphin, Pontoporia blainvillei, (Kasuya and Brownell 1979). Methods The data used in this analysis were from all fetuses and calves sampled between 1973 and 1981 except for three specimens <68 cm identified as calves and one specimen of 91 cm identified as a fetus {n = 609). These four specimens were judged to have been misidentified. The data were stratified by area, north and south of lat. 1°S, and the northern sample was further stratified by the size of kill in each net set. Powers and Barlow (1979^) have shown that in net sets in which the kill of spotted dolphins from the northern offshore regions was <40 (small-kill sets), about twice as many calves were killed as a propor- tion of the total kill as in sets where the kill was >40 (large-kill sets). This would introduce a bias in the estimate of length at birth in the regression pro- cedure because about 90% of all northern specimens were from small-kill sets. The effect would be to underestimate length at birth because the ratio of calves to fetuses was too high in most of the sam- ple Td investigate the extent of the bias, we calculated average length at birth for all northern specimens, for specimens from sets with kill <40 and >40, and for specimens from sets with kill <30 and >30 because the sample size for sets with kill >40 was small. The small sample available for southern specimens prevented any further stratification of the data. Length at birth was estimated by fitting a logistic model to the percent postnatal at each length inter- val, weighted by the inverse of the binomial variance of each percentage, and estimated by calculating from the fitted curve the length at which 50% of the specimens were predicted to be postnatal. We also investigated linear and asymmetric logistic-type models. improve the fit and gave similar results. Ikble 1 shows that the estimates of length at birth for northern specimens using the unstratified data (82.0 cm) and using specimens from sets with kill <30 (81.6 cm) or <40 (81.6 cm) are lower than the estimates using specimens from sets with kill >30 (84.6 cm) or >40 (85.4 cm), demonstrating that the bias resulting from an overrepresentation of calves in small-kill sets is significant. Furthermore, the estimate for kill >40 is higher than that for kill >30, indicated that the proportion of calves in the sam- ple may still be a function of kill-per-set at this level. Further stratification to investigate whether or not estimates of length at birth continue to rise at higher thresholds of kill-per-set was not possible because of small sample siz& The estimate of length at birth for southern specimens is 83.2 cm. No stratification was possible because of the small sample Estimates of standard deviations of the estimates of length at birth are not given because, in fitting the logistic model, sums of squares were minimized for differences between observed and predicted per- cent postnatal and it was unclear how to calculate the standard deviations. Table 1.— Estimates of length at birth stratified by number of offshore spotted dolphins killed per set. The range of length classes includes the last 0% postnatal length class and the first 100% postnatal length class. Length at birth predicted from the Sample Range logistic model size (cm) (cm) Northern offshore spotted dolph in Unstratified data 586 71-92 82.0 Sets with kill <30 321 73-92 81.6 Sets with kill <40 384 71-92 81.6 Sets with kill >30 105 71-89 84.6 Sets with kill >40 36 78-89 85.4 Southern offshore spotted dolph lin Unstratified data 23 78-85 83.2 Results Figure 2 shows the length-frequency data for northern specimens from sets with kill <40, and the logistic model fitted to the data. Figure 3 is the equivalent for northern specimens from sets with kill >40. Tkble 1 gives the results for all stratifica- tions described above using the logistic model. Using the linear or asymmetric logistic-type models did not ^Powers, J. E., and J. Barlow. 1979. Biases in the tuna-net sampling of dolphins in the eastern tropical Pacific Doc SOPS/79/31, Status of Porpoise Stocks Workshop, La Jolla, CA, 27-31 August 1979. Unpubl. MS. Discussion Our estimate of length at birth of 82.0 cm from the unstratified northern data is similar to that of Perrin et al. (1976), who estimated length at birth at 82.5 cm, based on a sample of 73 northern specimens (calves and fetuses) grouped into 3 cm in- tervals from 74 to 92 cm. Estimates of length at birth in large-kill sets are less biased because of the overrepresentation of calves in small-kill sets. A future larger sample from large-kill sets may allow for additional stratification by kill-per-set, enabling estimates to be calculated 555 FISHERY BULLETIN: VOL. 83, NO. 4 > o z UJ o HI cc u. 26 r 20 - I FETUSES 0 CALVES 10 70 72 74 76 78 80 82 84 86 88 90 92 94 LENGTH (cm) 100 r 80 < < Z 0) o LU o K U a. 60 40 20 Average length at birth I I I I I I I I T I t I T I I I I I I'l I I I I I 70 72 74 76 78 80 82 84 86 88 90 92 94 LENGTH (cm) Figure 2.— Length-frequency data for specimens from sets with Itill <40 grouped in 1 cm inter- vals for 163 fetuses and 221 calves from the northern offshore population of spotted dolphins, and the logistic model fitted to the percentage of animals that were postnatal. using specimens from sets with higher levels of kill. It may then be possible to determine at what level of kill-per-set the estimate ceases to increase. Until additional data are available, we consider 85.4 cm to be the best estimate of length at birth in northern offshore spotted dolphins. Our estimate of length at birth in the southern off- shore spotted dolphin of 83.2 cm is more ques- tionable because we were unable to stratify by kill- per-set. As adults, southern specimens are about 2.5 cm shorter than their northern counterparts (Per- rin et al. 1979). This small, but statistically signifi- cant, difference may or may not imply that length at birth is smaller in the southern population. The small sample of 23 specimens used in our calcula- tion of length at birth raises doubts concerning the 556 HOHN and HAMMOND: POSTNATAL GROWTH OF SPOTTED DOLPHIN >■ o z UJ o UJ DC 12 8 4 0 - I FETUSES CALVES Ira 70 72 74 76 78 80 82 84 86 88 90 92 94 96 LENGTH (cm) 100 r < < 0) o a. LU o UJ a Average length at birth I I I I I I i"r I f I f I f ri I I I I I I I I I I 70 72 74 76 78 80 82 84 86 88 90 92 94 96 LENGTH (cm) Figure 3— Length-frequency data for specimens from sets with kill >40 grouped in 1 cm intervals for 21 fetuses and 15 calves from the southern offshore population of spotted dolphins, and the logistic model fitted to the percentage of animals that were postnatal. accuracy of this estimate For these reasons, we take the estimate from all southern specimens of 83.2 cm as our provisional best estimate of length at birth for southern offshore spotted dolphins while recognizing that this estimate may be biased downwards because of a possible overrepresentation of calves in the sample. ESTIMATION OF LENGTH-AT-AGE USING ANALYSES OF MONTHLY DISTRIBUTIONS OF LENGTH Perrin et al. (1976) used the technique of fitting a growth curve to the means of normal distributions fitted to length-frequency data by month to estimate the length of the offshore spotted dolphin at 1 yr of age Perrin and Henderson (1984) used the same technique for the spinner dolphin. The technique is based on the assumption that breeding in these dolphins is seasonal and that a cohort of animals born at approximately the same time is characterized by a distribution of lengths, identifiable as a mode in the overall length distribution, which can be followed from month to month as mean length of the cohort increases. If there are sufficient data in each month, mean lengths can be followed from birth until 557 growth slows to an extent that distributions of lengths from different cohorts cannot be distinguish- ed. A growth curve can then be fitted to the month- ly mean lengths. Since the analysis of Perrin et al. (1976), the sam- ple of measured lengths from offshore spotted dolphins has increased from about 3,500 to over 15,000. Consequently, we were able to analyze the available data more extensively than had been done previously. Methods Length measurements from all postnatal specimens, made between 1968 and 1982, were used in the analyses except for three specimens <68 cm which were judged to have been erroneously identified as calves. The data were stratified into eight areas based upon apparent hiatuses in distribution from examination of sightings and ef- fort data (Fig. 1). Areas 1 and 2 comprise the southern population and areas 3-8 the northern population. For the northern data, no consistency could be found in preliminary analyses of lengths when data from all areas were included. When area 3 was ex- cluded, consistency was much improved. When areas 4 and 5 were also excluded, consistency was improved further for the months of February through June This indicated that there were nonseasonal or seasonal but asynchronous elements in areas 3, 4, and 5 at least at certain times of the year. Conse- quently, in our analyses of northern data we used lengths from areas 6, 7, and 8 only for February through June and lengths from areas 4-8 for January and July through December. A similar situation occurred for the southern data where the elimina- tion of area 2 improved consistency for January through May. In our analyses of southern data, therefore, we used lengths only from area 1 for these months. The data were grouped in interval widths of 4 cm. This gave four possible ways of grouping the data because lengths were measured to the nearest whole centimeter. Each of these four groupings were in- vestigated, there being no reason to prefer a start- ing point of the first interval as, for example, 76, 77, 78, or 79 cm. A mixture of normal distributions was fitted to each data set using a version of the computer pro- gram NORMSEP (Hasselblad 1966). The program requires the number of distributions to be specified, and this was varied in order to determine the most likely number of distributions present. The model FISHERY BULLETIN: VOL. 83, NO. 4 selected as most representative of the length- frequency data was that which gave the highest x" value, and therefore the highest probability that a greater ^ value could be obtained by chance alone, and also gave biologically feasible results based on prior knowledge of delphinid growth. (Some model fits had a very high probability of a greater ^, but the mean lengths could not be accounted for by any reasonable regime of growth.) We chose Laird's (1969) form of the Gompertz (1825) growth equation to fit to the monthly mean lengths. A linear model is clearly inadequate to describe growth except over a very short time period. We also investigated the use of the von Bertalanffy (1934) growth equation but found it to be less flexi- ble than the Gompertz model. Each model of growth was fitted to the mean lengths using the midpoint of the first month as time zero. In fact, this is not necessarily the time of birth so we fixed time of birth by substituting our estimate of length at birth into the fitted equation. Lengths at age were then calculated by substituting that age plus the difference between the midpoint of the first month and our calculated time of birth into the fit- ted equation. Results Northern Population Figure 4 shows, as examples, the fitted mixture of normal distributions to the length-frequency data for August and October. The arrows indicate the positions of the means of the fitted distributions. Ihble 2 shows the estimates of mean length of the fitted normal distributions for each month. The estimates are presented so that the increases from month to month can be clearly seen. The two final columns of Tkble 2 are mean lengths of the two distributions to the right of the length-frequency plots. These mean lengths are consistent from month to month. The table shows that there are actually two series of mean lengths: one beginning at 86.7 cm in September and continuing through columns 2 and 4 of the mean lengths, and the other begin- ning at 84.5 cm in April (the estimate of 92.7 cm for March is an anomaly for which we have no explana- tion) and continuing through columns 1, 3, and 5. These represent two cohorts born each year about 6 mo apart in the spring and autumn. Note that each series of mean lengths continues only for about 24 mo. This is because after this time growth has slow- ed to an extent that it is not possible to distinguish distributions of length from different cohorts. The 558 HOHN and HAMMOND: POSTNATAL GROWTH OF SPOTTED DOLPHIN 60-1 70 SO 110 130 150 170 ISO LEhGTH (Cn) ^0 > u z o UJ u. 120n 100- 80- 60- 40- 20- 70 90 110 130150 170 LEhGTH 0.05). An analysis of variance on the time taken (days) between the last molt before exposure and first molt after exposure (intern lolt period) resulted in exposure being not significant. Exposure was also not signifi- cant for the following intermolt period. In the December 1979 exposure experiment (Group II), one animal died in the 0-, 15-, 30-, and 120-min exposure categories and two died from the 60-min exposure before the second molt after ex- posure In this experiment, exposure had a signifi- cant effect on the growth increment for both the first (P < 0.001) and the second (P < 0.01) molts after ex- posure. An analysis on the intermolt period for the first molt after exposure showed no significant ef- fect due to exposure, but exposure was significant (P < 0.05) for the following intermolt period, main- ly due to the low number of days between molts for animals in the 120-min exposure category. TAGGING TRIALS Materials and Methods There are two main events in the commercial fishery for P eygnus which follow molting by a large proportion of the population. The first event is in November-December when maturing 4-5 yr old pale- shelled animals known locally as "whites" move off- shore into deeper water, where the breeding stock is generally situated. During the "whites" fishery, about 40% of the total catch is taken (Morgan 1980b). The second event is in March-April when postmolt, dark-shelled, sedentary animals called "reds" are captured (George 1958; Morgan 1977). As already mentioned in Brown and Caputi (1983), ac- count had to be taken of these two periods when planning tagging trials as the migratory "whites" could be more mobile and in a physiologically dif- ferent state than nonmigratory "reds" and hence their growth could be affected differently by handling practices (i.e., causing exposure, displace- ment, and damage). With these possible differences in mind, three tagging trials were conducted at Two Rocks, Western Australia (lat. 31°29.7'S, long. 115°3rE), avoiding the period of the full moon, when catches are at their lowest (Morgan 1974, 1977): 1) migratory "whites" phase— 26 November to 10 December 1978; 2) nonmigratory "reds" phase— 19 February to 10 March 1979; and 3) migratory "whites" phase— 16 November to 9 December 1979. An area consisting of shallow limestone reefs (6-18 m depth) within 6 km of shore and stretching from Two Rocks Marina to the mouth of Moore River was fished with standard wire beehive pots without escape gaps (Bowen 1971; Morgan and Barker 1974). Pots were baited daily with a variety of fresh fish; heads of Australian salmon, Arripis trutta; and bullock hocks. Tagging Trial A An experimental area was established, consisting of a grid on which pots were set and undersize lobsters could be displaced distances of 0, 230, 460, 910, 1,370, and 1,830 m from a base Hne of ex- perimental pots where they were captured (Brown and Caputi 1983). Pots were pulled each morning during the trial, weather permitting. Undersize (66 to <76 mm) lobsters were tagged with a numbered western rock lobster tag (Chittleborough 1974b) and their carapaces measured to the nearest 0.1 mm. Also recorded was the animal's damage, sex, and the depth and bottom type where it was caught and released. Grid areas were generally fished only once to avoid recapturing previously tagged animals. Recaptures of tagged animals were made by com- mercial fishermen who were paid a reward for the tag and market value for the animal if it had molted to legal size See Brown and Caputi (1983) for com- plete details of experimental procedures. Number of releases for this trial was about 1,500. Tagging Trial B Procedure for trial B was the same as trial A ex- cept that the 1,370 m transect was not set and ex- posure categories of 0, 30, 60, 120, and 180 min were also examined. Exposed undersize lobster were placed in plastic prawn baskets (lug baskets; com- monly used by fishermen to sort their catch) and ex- posed to air for the desired period before release at one of the displacement transects. About 2,300 tag- ged animals were released. 569 FISHERY BULLETIN; VOL. 83, NO. 4 Tagging Trial C Similar procedures were followed as for trials A and B, but only the 1,830 m and a new 3,660 m transects were set and exposure categories of 0, 15, 30, 60 min were used. Because parasite infection was observed on some animals (6.3% of releases), its presence was recorded as it could affect growth. The infection is a combination of a fungus {Fusarium sp.) and a bacteria {Vibrio sp.), which causes black lesions in the exoskeleton, usually in areas that have been damaged (e.g., tail fans and appendages). Infections were scored on an ordered scale of 0-6, with 0 in- dicating no infection. About 2,900 tagged animals were released. Results Figure 1 shows mean size (carapace) increment related to month of recapture for each of three tag- ging trials. In tagging trials A and C (November- December releases), February was the first month when there was evidence of molting in animals recaptured (Fig. 1); therefore, subsequent analyses on size increment only used recaptures from February onwards. In tagging trial B (February- March releases). May was the first month when there was evidence of molting (Fig. 1), so only data from this month onwards was used for the analyses. The results of an analysis of variance (ANOVA) on size increment for each of three tagging trials is shown in Tkble 1. In this analysis recapture month, sex, color, displacement, and exposure were treated as factors while size at release, damage, and level of parasite infection were treated as covariates. The analysis enables the significance of these factors and covariates to be determined after effects of other fac- tors and covariates are taken into account. In tagging trial A, after taking the effects of other factors into account (eg., sex, recapture month, etc.), the size increment per appendage missing was smaller by 0.48 mm wdth standard error of 0.04. This is also evident from Figure 2 which shows the rela- tionship between mean size increment and numbers of appendages missing for all recaptures from February to June 1979. Size increment in tagging trial C was also smaller by 0.48 mm per appendage missing with standard error of 0.004 (see Figure 3), while that for tagging trial B was 0.33 mm smaller (standard error of 0.07). From the ANOVA, size increments of displaced compared with nondisplaced animals were smaller by 0.32, 0.65, and 0.27 mm for tagging trials A, B, and C respectively. Exposure in tagging trial C, after other factors were taken into account by the ANOVA, resulted in increments smaller by 0.83, 1.34, and 2.30 mm for 15-, 30-, and 60-min exposure respectively, when com- pared with the zero exposure category. This can also be seen from Figure 3 which shows the mean size increment for animals recaptured after February 1980 related to exposure and number of appendages missing. While the effect of exposure on growth was not significant for the February-March 1979 trial B, size increments were smaller by 0.62 and 0.73 mm for 30- and 60-min exposures respectively, compared with unexposed animals. Mean size increment of males was larger than females by 0.95, 1.34, and 0.76 mm for the three tag- ging trials. Although the size at release of all animals was between 66 and 76 mm, there was still a highly significant decrease in size increment due to size at release of 0.25, 0.36, and 0.20 mm for every 1 mm increase in size at release Difference in color, i.e, dark-shelled vs. pale-shelled animals, was significant in tagging trial A with pale-shelled animals having a greater size increment by 0.65 mm, but this was not evident in tagging trial C. In tagging trial C, level of parasite infection of the animals was found to have had a detrimental effect on growth. DISCUSSION AND CONCLUSIONS Exposure Exposing undersize rock lobsters to the at- mosphere was detrimental to their growth increment at the first molt after exposure in both laboratory and field tagging trials. In the December laboratory experiment (Group II) the effect of exposure was still significant at the second molt after exposure "Eg- ging trial B (February-March 1979) resulted in ex- posure not having a significant effect on growth despite the presence of a negative trend between ex- posure and size increment. This may have been due to the fact that only 7 of the 110 animals recaptured in May- June 1979 had been exposed; this is the period when many of the undersize lobsters would have undergone their first postexposure molt (Fig. 1). Since there is a closed season from 1 July to 14 November, no recaptures were made until the time of the second postexposure molt in November- December 1979, by which time the combined effects of two molts may have masked the effect of the ex- posure treatment. However, this was not the case in tagging trial C (November-December 1979), which was held at the start of the fishing season, where the effect of exposure on the first molt which occurs 570 BROWN and CAPUTI: FACTORS AFFECTING GROWTH OF ROCK LOBSTER 20 19 18 17 16 15 M* 13 12 Ell 1— 1 10 UJ cr S9 LU 1/1 8 < -I a. 7 < < < ^5 C. NOV- DEC 1979 RELEASES o e 39 279 407 l_N Dj^ 1978 M M _D_, J_ M M Ij 1979 1980 RECAPTURE MONTH Figure 1— The mean size (carapace) increment by month of recapture and the sample size involved in calculating it shown next to the points, for the three tagging trials at Two Rocks. around February was clearly evident in animals caught from February 1980 onwards (Fig. 3, Ikble 1). Damage Damage to undersize rock lobsters was clearly shown to have a detrimental effect on growth (Ikble 1), which was directly proportional to the level of damage (Figs. 2, 3). Chittleborough (1975) found that growth incre- ment of i? cygrnts was reduced under laboratory con- ditions when four or more legs were removed and repetitive loss of two or more limbs led to precocious molting with reduced molt increments leading to an overall reduction in growth. Davis and Dodrill (1980) and Davis (1981) undertook research on the effect of injuries Oimb loss) produced by amateur and pro- fessional fishermen and natural causes (e.g., predators, molting accidents, etc) on the growth rate of P. argus populations in the wild. They found that 571 FISHERY BULLETIN; VOL. 83, NO. 4 NUMBER OF MISSING APPENDAGES Figure 2.— Two Rocks, November-December 1978. The relationship between the mean size increment and number of appendages missing for recaptures from February to June 1979, with the sample size and standard error from the mean shown at each point. Table 1.— The results of the analysis of variance on size increment for the three tagging trials at Two Rocks with the level of significance denoted by: NS, *, **, *** meaning not significant, P < 0.05, P < 0.01, and P < 0.001 respectively, and NA means not applicable. Nov.-Dec. Feb.-Mar. Nov.-Dec. Factor/covariate 1978 1979 1979 No. missing appendages * * * * * * * * * Displacement * * Exposure NA NS * * * Recapture month * * * * * * * * * Sex * * * * * * * * * Size at release * * * * * * * * * Color * * NA NS Parasite infection NA NA • * * Sample size 687 335 636 the growth rate of injured animals was significant- ly lower than that of uninjured animals, due to reduc- tions in molt increment and an increase in intermolt period. Their research did not demonstrate any pro- portional relationship between the degree of injury and the degree of molt increment depression as had been shown for H. oregonensis and P. crassipes (Kuris and Mager 1975) and also in this study on P. cygnus. Davis (1981) stated that growth rate of P. argus with minor injuries, five or fewer appendages missing, was almost identical to the growth rate of more seriously injured animals that were missing up to nine legs and both antennae. Displacement The displacement of undersize rock lobster was also found to significantly affect size increment in each of the three tagging trials. This was probably due to movement of animals from their home range (Chittleborough 1974c) which could have interrupted their normal feeding behavior and thus may have contributed to a decrease in food intake and hence growth. Overall Effect In general the handling of undersize rock lobsters by fishermen which causes them to be exposed to the atmosphere, damaged and displaced beyond their home range, not only affects their survival (Brown and Caputi 1983) but also affects growth of those that survive As discussed in Davis (1981), this reduc- tion in growth may result in: 1) The undersize lobsters staying below the legal size for a longer period than necessary with some being subject to natural mortality in this extra period before entering the fishery. 2) Those animals which do enter the fishery would do so at a reduced size, hence harvestable yield would be reduced. 572 BROWN and CAPUTI: FACTORS AFFECTING GROWTH OF ROCK LOBSTER 0 MINUTE = 0 MINUTE EXPOSURE ETC 2 3 U b 6 1 NUMBER OF MISSING APPENDAGES Figure 3— Two Rocks, November-December 1979. The mean size increment related to exposure categories and number of missing appendages for animals recaptured from February 1980 onwards. Exposure and missing appendages categories with less than five individuals have been combined and are plotted at the mean appendage level. The sample sizes are shown next to the points. 3) Size of these animals on reaching maturity would also be reduced, which would cause a decrease in fecundity directly proportional to their reduced size (Morgan 1972). The time they would take to reach maturity would probably not be affected since age appears to determine maturity rather than size (Chittleborough 1974d). 4) Affected animals would remain undersize for longer, thereby increasing the possibility that they could undergo multiple capture and handling. Multiple handling would result in increased mor- tality and further reduced growth. These factors, when added to the estimate of 14.6% reduction in recapture rate (most likely due to mortality) of the returned undersize lobsters dur- ing the fishing season (Brown and Caputi 1983), con- stitute a serious loss to the fishery. In addition, any mortality and reduced growth which may occur as a result of lifting the animals to the surface and returning them to the sea would also need to be add- ed to the above loss. This loss could not be quantified as both experimentals and controls in the tagging experiments experienced this. As mentioned by Brown and Caputi (1983), use of more effective escape gaps and an education pro- gram to encourage fishermen to return their under- size rock lobsters immediately to the sea while their vessel remained in the immediate vicinity of where the pot was pulled would help to overcome this serious source of industry created wastage Both these approaches are currently being examined wdth a view to reducing the numbers of undersize lobsters that are handled and the time they are kept on board the vessels. ACKNOWLEDGMENTS The authors would like to thank J. Prince and J. Jenke for technical assistance during this work; Ron Duckrell, the skipper, and the crew of the Flinders for assistance during the tagging trials; D. A. Han- cock and N. Hall for critically reading the manu- script and offering many helpful suggestions; and 573 FISHERY BULLETIN; VOL. 83, NO. 4 M. Isaacs for typing the manuscript. This research was supported by a grant from the Australian Department of Primary Industry's Fishing Industry Research Trust Account. LITERATURE CITED AlTKEN, D. E. 1980. Moulting and growth. In J. S. Cobb and B. F. Phillips (editors), The biology and management of lobsters, Vol. I, p. 91-163. Acad. Press, N.Y. BowEN, B. K. 1963. Preliminary report on the effectiveness of escape gaps in crayfish pots. West. Aust. Dep. Fish. Fauna Rep. 2, 9 p. 1971. Management of the western rock lobster (Panulirus longipes cygnus George). Proa Indo-Pac Fish. Couna 14(II):139-153. 1980. Spiny lobster fishery management. In J. S. Cobb and B. F. Phillips (editors). The biology and management of lobsters. Vol. II, p. 243-264. Acad. Press, N.Y. Brown, R. S., and N. Caputi. 1983. Factors affecting the recapture of undersize western rock lobster Panulirus cygnus George returned by fishermen to the sea. Fish. Res. 2:103-128. Chittleborough, R. G. 1974a. Review of prospects for rearing rock lobsters. Aust. Fish. 33(4):4-8. 1974b. Development of a tag for the western rock lobster. CSIRO Div Fish. Oceanogr. Rep. 56, 19 p. 1974c Home range, homing and dominance in juvenile western rock lobster. Aust. J. Mar. Freshw. Res. 25:227-234. 1974d. Western rock lobster reared to maturity. Aust. J. Mar. Freshw. Res. 25:221-225. 1975. Environmental factors affecting growth and survival of juvenile western rock lobsters Panulirus longipes (Milne- Edwards). Aust. J. Mar. Freshw. Res. 26:177-196. 1976. Growth of juvenile Panuliriis longipes cygnus George on coastal reefs compared with those reared under optimal environmental conditions. Aust. J. Mar. Freshw. Res. 27:279-295. Davis, G. E. 1981. Effects of injuries on spiny lobster, Panulirus argus, and implications for fishery management. Fish. Bull., U.S. 78:979-984. Davis, G. E., and J. W. Dodrill. 1980. Marine parks and sanctuaries for spiny lobster fisheries management. Proc Gulf. Caribb. Fish. Inst. 32:194-207. George, R. W. 1958. The biology of the Western Australian commercial cray- fish Panulirus longipes. Ph.D. Thesis, Univ. Western Australia, Nedlands, 124 p. Hancock, D. A. 1981. Research for management of the rock lobster fishery of Western Australia. Proc Gulf. Caribb. Fish. Inst. 33: 207-229. KURIS, A. M., AND M. Mager. 1975. Effect of limb regeneration on size increase at molt of the shore crabs Hemigrapsus oregonensis and Pachygrapsus crassipes. J. Exp. Zool. 193:353-359. Morgan, G. R. 1972. Fecundity in the western rock lobster, Panulirus longipes cygnus (George) (Crustacea:Decapoda:Palinuridae). Aust. J. Mar. Freshw. Res. 23:133-141. 1 974. Aspects of the population dynamics of the western rock lobster Panulirus cygnus George II. Seasonal changes in the catchability coefficient. Aust. J. Mar. Freshw. Res. 25: 249-259. 1977. Aspects of the population dynamics of the western rock lobster and their role in management. Ph.D. Thesis, Univ. Western Australia, Nedlands, 341 p. 1980a. Population dynamics and management of the western rock lobster fishery. Mar. Policy 4:52-60. 1980b. Increases in fishing effort in a limited entry fishery - the western rock lobster fishery 1963-76. J. Cons. Int. Ex- plor. Mer 39:82-87. Morgan, G. R., and E. H. Barker. 1974. The western rock lobster fishery 1972-1973. West. Aust. Dep. Fish. Wildl. Rep. 15, 22 p. Morgan, G. R., B. F. Phillips, and L. M. Joll. 1982. Stock and recruitment relationships in Panulirus cygnus, the commercial rock (spiny) lobster of Western Australia. Fish. Bull, U.S. 80:475-486. Norton, P. 1981. The amateur fishery for the western rock lobster. West. Aust. Dep. Fish. Wildl. Rep. 46, 108 p. 574 SEA SCALLOP FISHING IMPACT ON AMERICAN LOBSTERS IN THE GULF OF ST LAWRENCE G. S. Jamieson' and A. Campbell^ ABSTRACT Damage to American lobsters, Homarus americanus, in Egmont Bay and off Miminegash, Price Edward Island, is minimal from the drags of the seasonal sea scallop, Placopecten magellaniciis, fishery. During May 1981, when commercial sea scallop fishing was occurring, American lobster abundance was low in areas of profitable scallop exploitation. Sea bed substrate in these areas was generally smooth and most lobsters were able to avoid the gear. In the areas with and without commercial scallop fishing, 1.3% and 11.7% of observed lobsters, respectively, were injured or retained by the drag. Lobster abundance in the areas commercially exploited for scallops in May and June was significantly greater in July than in May, but whether this was a result of a natural seasonal movement of lobsters or the cessation of scallop fishing is unclear. Sea scallop, Placopecten magellanicus, and American lobster, Homarus americanus, populations are fully exploited in Northumberland Strait, Gulf of St. Lawrence (Wilder 1947, 1965; Robinson 1979; Jamieson et al. 1981c; Campbell and Mohn 1983). In- dividual fishermen frequently fish both species, com- monly in the same general area, although the fisheries are separated temporally (Jamieson et al. 1981c; Conan and Maynard 1983). Recently, localized low abundance of these important, commercial species has heightened long-held convictions by fishermen of the negative impact of sea scallop fishing on American lobster stocks. Fishermen's con- cern became acute for the Egmont Bay area during 1980, coincident with the discovery and exploitation of new nearshore scallop concentrations near West Point, Prince Edward Island (Fig. 1). Decreased scallop recruitment in recent years (Jamieson et al. 1981b, c) has resulted in a scarcity of scallops in traditional fishing areas, causing increased ex- ploration for commercially exploitable scallop concentrations. The magnitude of scallop gear-lobster interaction is dependent on the spatial and seasonal distribu- tions of scallops and lobsters and the impact of scallop gear on commercial lobster abundance where the distributions of both species overlap. Scallops are widespread in Northumberland Strait (Caddy et al. 1977), but commercial concentrations are found only 'Department of Fisheries and Oceans, Fisheries Research Branch, Halifax, Nova Scotia; present address: Pacific Biological Station, Nanaimo, British Columbia V9R 5K6, Canada. ^Department of Fisheries and Oceans, Atlantic Biological Sta- tion, St. Andrews, New Brunswick, EOG 2X0, Canada. in limited areas. The precise locations of these areas are undocumented, and since they vary with time, they cannot be predicted with any accuracy. However, commercial log data has shown the broad distribu- tion of scallop concentrations in Northumberland Strait during 1979-80 (Jamieson et al. 1981b, c). The seasonal abundance and distribution of commercial-sized lobsters is largely unknown, but their general distribution overlaps that of scallops (Stasko et al. 1977; Conan and Maynard 1983). There have been few field studies conducted on lobsters in this area: Ibmplemann (1933, 1934, 1935, 1936) reported on lobsters and the fishery in Northumber- land Strait; Wilder (1963) and Wilder and Murray (1956) reported on movements and growth of tag- ged lobsters liberated in Egmont Bay. Scallop and Irish moss, Chondrus crispus, drags can damage lobsters, although lobsters exposed on open ground tended to avoid moving drags. Most gear-induced damage has resulted from lobsters in burrows being hit or crushed by rocks disturbed by the drag (Scarratt 1973, 1975; Pringle and Jones 1980). Investigations reported here document 1) the scallop drag/lobster interactions off Miminegash, Prince Edward Island, during August 1978 and in Egmont Bay during May and July 1981; and 2) the relative abundance and movement of lobsters trap- ped and tagged in Egmont Bay prior to (June-July 1981) and during the lobster fishing season (10 August-10 October 1981). Lobster abundance may change relatively rapidly because of their potential high mobility. Therefore it is important to charac- terize lobster microdistribution and assess the con- Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 575-81^ FISHERY BULLETIN: VOL. 83, NO. 4 Sand E Small Rocks ^M Large Rocks 46° 35' 46° 30* 46 25* FIGURE 1.— General substrate type as observed by divers in the areas surveyed in Egmont Bay, Northumberland Strait, Gulf of St. Lawrenca Locations of stations (heavy lines) sampled for lobsters in Egmont Bay, Northumberland Strait: 1 = recently heavily fished scallop ground, 2 = recently lightly fished scallop ground, 3 = no recent scallop fishing (large rocks), 4 = no scallop fishing (deeper water). sequences of scallop fishing on the degree of species overlap. Scallop fishing may directly damage lobsters, or because of the disturbance of the sea bed by the drags, may cause lobsters to avoid or be at- tracted to the overall area. METHODS Scallop Gear-Lobster Interactions 1978 Study The interaction between three types of scallop gear and lobsters was observed by divers between 15 and 30 August 1978. The study area was in 14 m of water about 1 km from shore Oong. 46°52'30"W, lat. 64°14'00"N), and consisted of a sandy bottom with occasional small rocks. The gear used was a two- gang, toothed Gulf rock drag (60 cm buckets) (Fig. 2); a two-gang Digby rock drag (76 cm buckets, no teeth) (MacPhail 1954); and a 152 cm Gulf sweep chain drag. A Gulf sweep chain drag is a smaller, lighter revision of an offshore scallop drag (Bourne 1964). A hood of 38 mm stretch mesh was placed over the drags extending to a height of 81 m above the sea bottom, and one of the buckets (half the chain sweep drag) had a similar mesh hood on the outside of the back of the drag (back cover). The bucket, or portion of the drag, without a back cover had a mesh liner. Two divers hung onto each drag during tows, 576 JAMIESON and CAMPBELL: SCALLOP FISHING IMPACT ON LOBSTERS A. Dors a\ View 274 cm A~ 91 cm cr 81 cm 3 n n n n n nnnr Unlined Drag 60 cm 150 cm V Tow Bar ^^ 4. 130 cm r-i n ri p ri m £-" Teeth L jned Drag 5 I cm .y... . Dumping Ring Spruce Log J ,Drag Hood Rubber Mat B. Lateral View of Bucket 2 Rubber M of Spruce ^S^ Log 81 cm \ 25 cm Figure 2.— Schematic drawings of a four-gang Gulf rocl^ drag: (A) hood and Hner arrangements used with buckets 2 and 3 in 1978 (buckets 1 and 4 were removed). In 1981, 4 unmodified buckets were used. (B) Lateral view of bucket 2 used in 1978. 577 FISHERY BULLETIN: VOL. 83, NO. 4 noting lobster behavior and the physical effect of the drag on lobsters; carapace lengths (CL, back of eye socket to posterior carapace margin) of fished lobsters were measured. Tovv^ velocities, established by engine rpm, were similar to commercial opera- tions and tow duration was 5 min. 1981 Study Dragging was conducted during 14-22 May and 27-31 July 1981. Four general areas (Fig. 3) were surveyed in both periods. Scallops and lobsters were known by fishermen to exist in areas A and A' but scallop fishing had not occurred for several years; five research tows were conducted in May and three tows in July. Areas B and C were reported by fishermen to be prime lobster ground where scallop fishing had occurred recently or was in progress dur- ing the study; 30 research tows were conducted in May and 25 tows in July. Tbw locations were ran- domized within an area and the number per area was arbitrarily assigned according to the apparent distribution of commercial effort in the scallop fishery. Bottom water temperatures averaged 8.8° and 18.4°C in May and July, respectively. A four- gang Gulf rock drag (Fig. 2) with 51 cm buckets was used throughout the study. Scallop rings had 69-75 mm and 80-84 mm inside and outside diameters, respectively. Lead ropes 30 m long were attached to each end of the 2.36 m club stick at the back of the drag to define an area behind the drag to be surveyed by divers. Before the drag was dropped, the lines were let out while the vessel was steaming or drift- ing to establish an unfished control area for survey 64° 20' 64 10 \ \ \ 1 1 .'IVi. .:,._JC^r^«T ■*!= 2^2Easn3 1^ ti yP ^^•^■<»ii<. A^^lS^i \ ..J. \ \ -'' a1 \i / r %s- o\ Egmont B ay ^^ lA- V X!= ^ 3/ "~~ N^ h- /^ TTm . ■* '~^ ~'r\ /4 N -«? / 2 \ / N ">> / \ > \ \ / /A , / L 0 y < tRock Pt. 1 -— I '. > N Z' """"^^ r* / 1 < > / /^ 57\ V A f~ja a \ \- -' '' ^; v. 1 35/ 62 'yi 8 j V- 1 f^^ \ V^ C ^10 cm) (>10 cm) Mean depth (m) 11 13 11 18 No. of traps 50 50 50 50 Trap soak-over (days) (range) 2-5 1-7 2-5 1-5 No. of trap hauls 489 568 500 591 No. of lobsters caught 2,507 1,967 2,568 2,330 Total tags released 500 501 500 501 Total tags returned^ 182 162 234 162 % of total tags released in area^ 36.4 32.3 46.8 32.3 'No scallop fishing. ^Number of tags returned up to 30 October 1981 including tags witfiout recapture locations. 579 FISHERY BULLETIN: VOL. 83, NO. 4 and Flowers (1968) have reported, using a computer program by Campbell et al. (1983). RESULTS Scallop Gear-Lobster Interactions 1978 Study No scallops were present but lobsters were numer- ous and were observed by divers to be frequently foraging in the open. Average carapace length of 22 diver-collected lobsters was 61.3 mm (SD = 26.8). While the microdistribution of substrate type was patchy, tows were of sufficient duration to cover all substrate types. Under the assumption of an average uniform lobster density during tows, lobster catches made by the Gulf sweep-chain drag over sand and rock-sand were highest. Average catches in the lined and unlined portion of the drag were 0.53 and 0.07 lobsters/m of drag width fished per min (m^^min^^), respectively (Ikble 2). No lobsters were retained by the unlined rock drags, but since they were retained in the back cover of the drag, lobsters were entering the drag and passing through the rings. These lobsters did not show any external evidence of damage. The hoods Table 2.— Average number of lobsters caught m~'min'' of drag width In the hoods and linings of the scallop gear used in 1978. Each drag type was hauled over both sand and sand-rock substrates off Miminegash, Prince Edward Island. L = lined gear; UL = unlined gear; H = hood; B = back cover. Drag type No. tows Lobster catch UL H B Gulf sweep chain 5 0.53 0.07 0.15 0.31 Gulf rock drag 5 0.11 0.00 0.06 0.11 Digby rock drag 6 0.02 0.00 0.07 0.04 of all three drag types contained lobsters, indicating that lobsters can escape by swimming over the ad- vancing drag. 1981 Study RELATIVE SCALLOP AND LOBSTER ABUN- DANCE.—Catch results and sightings per tow (Jamieson et al. 1981a) indicated that for each study area, considerable variation existed in abundance of both scallops (Fig. 3) and lobsters (Fig. 4). Substrate type was variable over the distance of a single tow, and this appeared to be a major factor influencing relative scallop and lobster abundance. Scallop and lobster densities in the two areas (B and C) fished for scallops varied significantly on some dates (Tkble 3) from those densities in the non- fished areas (A and A^); fished grounds had a greater number of scallops, but fewer lobsters, than did the nonfished ground. Between the two fished areas, the only significant (P < 0.05) difference was in the scallop drag catch in May, but study area C off Red Head generally yielded more of both species than did study area B off West Point (Figs. 3, 4). There were no significant differences (P > 0.05) in the densities of either scallop or lobster in the non- fished area between the two sampling periods. Lobster sightings per tow in both fished areas were significantly greater (P < 0.05) in July than in May, but the number of sightings averaged less than in the unfished area. Although not always significant, the general seasonal trend of lobster abundance, as indicated by the control sampling procedures, in- creased between May and July in all areas. Scallop catch decreased significantly (P < 0.01) in the ground off Red Head between May and July. On fished ground, average scallop density decreased whereas scallop density on the nonfished ground increased during this time period. Table 3. — Average scallop and lobster catch in a 4-gang Gulf rock drag in 1981, and the abundance per 1,000 m^ In each study area before the scallop drag was towed (control) and in the drag path behind the drag. Values with asterisks are significantly different from the corresponding value in the nonfished scallop location. * = P < 0.05; ** = P < 0.01; n = number of tows. Month Control density Drag Drag path Location n Scallop n Lobster n No. scallop caught n No. lobsters observed n Scallop n Lobster Nonfished area May July 4 2 6.25 12.50 4 2 2.08 16.67 5 3 0.87 0.15 4 3 3.03 4.62 2 3 0.0 4.71 2 3 0.00 4.71 West Point May July 4 3 56.23 22.21 4 3 0.00 2.78* 11 7 7.59* 10.10 11 7 0.48 1.48* 8 5 24.70* 127.08 8 5 1.76 5.65 Red Head May July 9 9 99.03** 58.31 9 9 0.93 5.55 19 18 16.38** 8.34** 19 18 0.87 1.98* 12 10 68.25** 52.24 12 10 1.18 1.41 580 JAMIESON and CAMPBELL: SCALLOP FISHING IMPACT ON LOBSTERS 64° 20' 6 4 10" Figure 4— Number of lobsters observed by divers in May during each tow in Egmont Bay, Northumberland Strait, Gulf of St. Lawrence Substrate type is shown in Figure L Area designations are explained in Figure 3. SCALLOP GEAR: LOBSTER INTERACTION.- No relation was found between the two sample periods and the number of lobsters injured or re- tained during a tow (Ikble 4). The weighted percent- age of lobsters injured or retained was 1L7 and L3, for the nonfished and combined fished areas, respec- tively. Injured lobsters were not found in the drag path, although occasionally lobsters were observed to retreat into burrows in front of a moving drag. Whether they subsequently became damaged or trapped in their burrows is unknown, but the absence of damaged lobsters in the drag path suggests that the frequency of lobster damage is low on commer- cial scallop ground especially where there is a general lack of large rocks and boulders. However, 14 of the 30 locations surveyed had occasional large rocks. Most lobsters encountered were too small (mean CL = 72 mm) (Ikble 5) to be retained by the scallop gear unless the steel rings making the drag were partially blocked by debris. All lobsters <92 mm CL can pass through a 70 mm inside diameter scallop ring (Stasko 1975). Several lobsters were seen by the divers entering the drag and passing through the rings apparently unscathed. In 63 tows, 1 1 lobsters were affected directly by the scallop gear: four were retained by the drag (the ring openings were block- Table 4.— Frequency by month and location of lobster reduction and/or injury during dragging. Nonfished areas Fished areas Details Month West Point Red Head No. lobsters May 28 10 35 observed July 32 24 82 No. lobsters May 3 1 0 injured/retained July 4 1 0 % injured/ May 11 10 0 retained July 13 4 0 581 FISHERY BULLETIN: VOL. 83, NO. 4 Table 5. — Summary of lobster catches during closed fishing season (June-July 1981) and during the fishing season (August-October 1981) in Egmont Bay. Shorts = prerecruits <63.5 mm CL; legals = recruits >63.5 mm CL. No. of No. of Prerecruits Recruits No. per % females No. per % females Week Time lobsters trap % total trap of total trap of total Mean no. period Area sampled hauls lobsters haul shorts haul legals CL (mm) 1 22-26 June 2 2 104 0 0 0 0.02 0 81.0 4 2 96 50.0 0.01 0 0.01 100.0 72.0 2 29 June- 1 3 100 33.3 0.01 100.0 0.02 100.0 71.0 3 July 2 17 102 29.4 0.05 80.0 0.12 33.3 68.8 3 2 100 50.0 0.01 100.0 0.01 100.0 66.0 4 11 100 45.5 0.05 40.0 0.06 50.0 67.3 3 6-10 July 1 141 98 9.2 0.13 53.9 1.31 20.3 71.4 2 114 102 12.3 0.14 35.7 0.98 35.4 71.5 3 99 100 11.1 0.11 72.7 0.88 21.6 70.9 4 35 100 25.0 0.09 22.2 0.27 30.8 72.6 4 13-17 July 1 503 94 17.7 0.95 42.7 4.40 31.9 71.0 2 542 104 10.4 0.59 54.1 5.06 33.8 72.1 3 550 100 12.2 0.67 41.8 4.83 33.3 71.3 4 395 100 9.4 0.38 45.9 3.65 31.6 72.6 5 20-24 July 1 794 99 13.7 1.10 55.0 6.92 33.9 69.9 2 708 104 9.0 0.62 51.6 6.19 36.3 71.6 3 840 100 10.7 0.90 47.8 7.50 36.6 70.4 4 792 98 6.3 0.51 58.0 7.57 33.8 72.4 6 27-30 July 1 1,066 98 15.7 1.70 44.3 9.17 29.1 70.3 2 584 52 9.3 1.04 57.4 10.19 31.0 71.6 3 1,077 100 9.2 0.99 44.4 9.78 29.6 70.9 4 1,095 97 6.2 0.70 39.7 10.59 32.0 73.3 8 10-14 Aug. D 1,582 488 24.5 0.79 44.9 2.40 33.1 71.9 9 17-21 Aug. V) 155 125 47.7 0.59 46.2 0.65 (') {') 10 24-28 Aug. n 501 240 44.7 0.93 48.5 1.13 31.8 73.2 11 13 Aug.- 4 Sept. V) 458 245 42.5 0.80 42.6 1.07 34.4 73.0 13 14-18 Sept. n 496 237 51.6 1.08 50.8 0,99 44.2 71.5 15 28 Sept.- 2 Oct. V) 129 74 7.7 1.18 53.4 0.57 n (') lAII fishing areas du ing fishing season. 2Size and sex not recorded ed with Laminaria longicruris), four passed under the drag and were possibly injured, and three were struck during the course of one tow (in May) resulting in claw loss or a cracked carapace In the latter instance, strong currents impeded the escape of lobsters. Scallop size frequencies were similar in May and July in each of the two commercially fished areas and all scallop age classes were exploited about equally. Lobster Abundance and Distribution Abundance The number of lobsters caught per unit of effort (trap haul; CPUE) increased during the experimen- tal fishing period in all areas (Ikble 5). During the commercial fishing period, CPUE of prerecruits in- creased but CPUE of legal-sized lobsters decreased presumably as a result of fishing mortality. Number of lobsters per trap haul was not significantly dif- ferent between areas observed during the ex- perimental period (Ikble 5). The use of CPUE is unreliable in the quantitative estimation of lobster abundance as many factors affect trapability, in- cluding water temperature, lobster behavior, molting, relative trap and lobster densities, and bait attrac- tiveness (Elner 1980). Lobster Movement Of the 740 tagged lobsters recaptured (37%), 658 had recapture location data (Ikbles 1, 6). In areas 1, 2, and 3, the majority (65-78%) of tagged lobsters were recaptured within 10 km of release, but in area 4, 50% were caught 11-18.5 km from release (Ikble 6). The mean distance moved was not significantly different for tagged lobsters released in areas 1, 2, or 3, but lobsters from area 4 moved a significantly (P < 0.01) greater mean distance than those of the other three areas. Direction statistics (according to Saila and Flowers 1968) summarize the nature of lobster movement 582 JAMIESON and CAMPBELL: SCALLOP FISHING IMPACT ON LOBSTERS Table 6. — Summary statistics of distance travelled and direction moved by tagged lobsters released at four areas in Egmont Bay during June-July 1981, and recaptured up to 30 October 1981. (Direction statistics calculated according to Jones [1959] and Saila and Flowers [1968] are 0 = mean vector angle from true north; V and V = directed movement along north- south and east-west plane, respectively, negative values of V and V indicate net southerly and westerly movement, respectively; R and Z = Rayleigh test statistics for randomness or uniform distribution of points about a circle; * = significant at P < 0.01 indicates a non-uniform distribution.) SD = Standard deviation. Release area 1 2 3 4 Heavily fished Lightly fished Scallop ground Lobster Details scallop ground scallop ground with rocks! ground' % of total recaptures moved 0-10 km 65.2 70.3 78.2 35.3 % of total recaptures moved 11-18.5 km 28.1 16.2 19.8 49.7 % of total recaptures moved >18.5 km 6.7 13.5 2.0 15.0 Total recaptures 164 148 207 139 Mean distance moved, 9.2 10.1 8.3 212.9 km (±1 SD) (0.5) (0.6) (0.4) (0.6) Maximum distance moved, km 25.0 45.0 38.9 49.7 0 247.3 289.3 310.5 344.4 V (km/d) -0.055 0.026 0.062 0.213 V (km/d) -0.132 -0.075 -0.073 -0.059 R 74.9 26.3 64.9 63.5 Z 34.2* 4.7* 20.4* 29.0* 'Ligtit or no scallop fishing. 2Value for mean distance travelled significantly different (P < 0.01 ); other values not significantly different (P > 0.05). (Tkble 6). The Rayleigh test statistic, Z, indicated a non-uniform distribution of tag returns (P < 0.01) for each area of release Results of the Rayleigh test should be treated with caution (Batschalet 1965) since there is some evidence of bimodality. In general, the returns exhibited three main directions of movement for lobsters: south-west for area 1, north-west for areas 2 and 3, and north for area 4 (Tkble 6). V and V, the north-south and east-west coefficients of directed movement, measure the mean daily travel of the group. Lobsters from areas 1-3 showed little dispersion in a north-south direction in contrast to lobsters from area 4, which moved the greatest in a northward direction (0.213 km/d). Lobsters from all areas generally moved west, but lobsters from Area 1, the heavily fished scallop ground, tended to disperse furthest west per day (0.132 km/d). Dispersion to the west is perhaps large- ly the result of the relative proximity of the release areas to the western shore of Prince Edward Island, which restricted lobster movement to the east. Lobster Growth Determinations of molt stage from pleopod ex- aminations indicated that lobsters may have molted as early as 6-12 July (Tkble 7). Trapability of lobsters is affected by molt stage, with late molt stages (eg., D^q; Aiken 1973) being difficult to trap. The high percentage of Dg to D^ animals (stages just before molting) in mid-August indicated that considerable molting was imminent, and this probably affected CPUE at this time Many tagged lobsters (47.8%, N = 46) recaptured during the period 24 August-26 September had molted. DISCUSSION The results of this study are probably area and Table 7. — Percentage of pleopod stages of lobsters sampled from Egmont Bay, July-September 1981. Pleopod stages 3.0-3.5 and 4.0-5.5 predict lobsters molting within 1.9-4.0 wk and 0.3-2.4 wk, respectively, at 15°-19°C (Aiken 1973). Prediction of molting time for pleopod stage 1.0-2.5 is unreliable but usually exceeds 4.0-10.3 wk. % of lobsters Pleopod stage Period 1.0-2.5 3.0-3.5 4.0-5.5 Temper- No. of ature lobsters (°C) 6-10 July 97.1 1.5 1.4 69 15.5 13-17 July 98.2 0.9 0.9 113 16.0 20-24 July 96.5 3.5 0.0 142 16.5 27-30 July 89.9 6.6 3.5 117 19.0 10-14 Aug. 76.3 21.1 2.6 76 18.1 31 Aug.-4 Sept. 91.1 5.9 3.0 101 17.5 583 FISHERY BULLETIN: VOL. 83. NO. 4 time specific Nevertheless the extent of sea scallop gear damage to American lobsters in Egmont Bay was measured, and this permits estimation of the damage to lobsters on similar substrate types in other locations. In the nonfished area there was no significant difference in the lobster abundance between the May and July observations, whereas there were significant differences in relative seasonal lobster abundance in the fished areas. In western Northumberland Strait at the time of this study, scallop fishing occurred primarily between late April and late June, with a minor amount of fishing between mid-October and winter freeze-up (Jamieson et al. 1981c). There was limited scallop fishing during July. If lobsters were displaced by scallop dragging during May and June, normal seasonal lobster densities could be reestablished by late July. It is unclear whether the greater density of lobsters in the fished areas in July was due to nor- mal seasonal migration onto these grounds or to the absence of scallop fishing. Data from tagged lobsters suggested that some immigration may have occurred from the deeper water areas of the Strait, but it also appeared that overall abundance on the scallop grounds may have been reduced by scallop dragging activity. Predators have been reported to be at- tracted to the disturbed substrate in a drag's path (Caddy 1973), but how this relates to lobsters is unclear. The trapability of lobsters is a function of many variables (Elner 1980), making the quantification of lobster abundance difficult in the four trap study areas. There was a lower percentage of short lobsters and a larger mean carapace length in Area 4 than in the other areas (Tkble 5). When large lobsters were trapped there were generally fewer small lobsters in the traps (Tkble 5), but this may have resulted from agonistic behavior (Cobb and Tkmm 1975) rather than relative density. Water temperature increases may also have affected behavior and possibly had major modifying effects on lobster growth and/or movement. McLeese and Wilder (1958) documented an increase in lobster rate of movement with in- creasing temperature, but what effect this had on the average direction of movement during the study period is unclear. The mean movement rate of lobsters in our study (Tkble 7) was similar to that reported by Saila and Flowers (1968) for mature lobsters off Rhode Island. Saila and Flowers (1968) showed that the coefficients of directed movement, V and V, are sensitive to changes in movement pat- terns at various life history stages, and hence are a possible function of lobster maturity level and the sex ratio used in their calculation. These potential influences were not considered here because of limited duration of the study and the relative close proximity of release and recapture areas which were probably not optimal to permit extensive data analysis. There probably was a directed movement of tag- ged lobsters from area 4 (deeper water) into Egmont Bay. This may have contributed to the increased CPUE during July-October. Tfempleman (1936) found there was some movement during the summer with lobsters congregating in the relatively shallow in- shore water areas of Northumberland Strait and that some lobsters moved offshore in the fall. The disproportionate sex ratio of legal-sized lobsters observed in the present study suggested a geographic distributional difference between the sexes of lobsters after maturity during July-August. Lobster trapability, and hence estimated abun- dance, can be influenced by molt stage. Many legal- sized lobsters appeared to have molted between August and September, and while the data are in- sufficient to support the fact that a molt may have occurred prior to or during the experimental fishing period (late June-July), other investigations have presented evidence in the literature that lobsters in this area do molt in late spring (Tfempleman 1934, 1936; Wilder 1963). If, in fact, two molts did occur during the study period, this along with increased water temperature increasing lobster movement (McLeese and Wilder 1958) could partially explain the rapid increase in CPUE during July. However, no soft-shelled lobsters were observed during July, while soft-shelled lobsters were quite frequently en- countered in fishermen's traps during the August- September fishing season. The seasonal nature of the fisheries minimizes the impact of scallop gear on lobsters because lobsters are in low abundance on scallop ground at the time of greatest scallop fishing activity. Commercial con- centrations of scallops and lobsters also appear to be largely separated spatially (Figs. 2, 3). What then is the likely economic impact in Egmont Bay of scallop fishing on lobsters, and how does this com- pare to the value of the exploited scallop resource? No reported commercial scallop fishing was reported in 1980 off Red Head, but vessel logs recorded that 1,509.4 kg of adductor muscle meat were taken near West Point (scallop log areas 77 and 78 combined, Jamieson et al. 1981c). Average CPUE was about 2.4 kg/h • m "\ indicating that 629 h • m of effort was ex- pended. In our study, a total of 8.2 h-m of research effort was expended in May on the fished grounds and 22 lobsters were observed behind the drag in the drag path. Drag velocities over the bottom in 584 JAMIESON and CAMPBELL: SCALLOP FISHING IMPACT ON LOBSTERS both commercial fishing operations and in our study are assumed to be similar. If 2.6% of the lobsters observed are retained or injured, with 50% of these lobsters killed, then the total number of lobsters estimated to be destroyed by commercial scallop fishing in 1980 was 22 lobsters. If each lobster weigh- ed 0.5 kg and was valued at $6.60 kg"\ then the loss would be about $73. In comparison, at $8.27 kg"^ of scallop meat in May, 1980 (Jamieson et al. 1981c), the scallop landings from these two fished areas had an estimated value of $12,483. If lobster abundance was as high as that in the un- fished area, i.a, 3.03 lobsters/1000 m^, then 139 lobsters, with a value of $460, would have been destroyed. In both instances, this loss is negligible in comparison to the values of the lobster and scallop fisheries. These conclusions are in agreement with the observations of Scarratt (1973) and Pringle and Sharp (1980) in their assessments of the impact of Irish moss raking on lobster populations. ACKNOWLEDGMENTS We thank M. Beattie, D. Duggan, S. Hamet, M. Lundy, D. R. Maynard, and N. Witherspoon for diving assistance; Captains E. Wedge and M. Ellis of chartered vessels for their help and advice; D. J. Scarratt and J. D. Pringle for their support and manuscript review, and in particular, R. Chandler, M. Etter, D. R. Maynard, and J. D. Pringle for coor- dinating field studies and/or assistance in data analyses. LITERATURE CITED Aiken, D. E. 1973. Proecdysis, setal development, and molt prediction in the American lobster {Homarus americanus). J. Fish. Res. Board Can. 30:1337-1344. Batschalet, E. 1965. Statistical methods for analysis of problems in arrived orientations and certain biological rhythms. Am. Inst. Biol. Sci., 54 p. Bourne, N. 1964. Scallops and the offshore fishery of the Maritimes. Fish. Res. Board Can. Bull. 145, 60 p. Caddy, J. Y. 1973. Underwater observations on tracks of dredges and trawls and some effects of dredging on a scallop ground. J. Fish. Res. Board Can. 30:173-180. Caddy, J. F., T. Amaratunga, M. J. Dadswell, T. Edelstein, L. E. LiNKLETTER, B. R. McMULLIN, A. B. StASKO, AND W. H. van DePoll. 1977. 1975 Northumberland Stait project, Part I. Benthic fauna, flora, demersal fish and sedimentary data. Can. Fish. Mar. Serv., MS Rep. 1431, 46 p. Campbell, A., S. E. Bellis, G. E. Fawkes, and C. Hastey. 1983. Computer programs for analysis of lobster (Homarus americanus) movements from tag-recapture data. Can. Ms. Rep. Fish. Aquat. Sci. 1705, 22 p. Campbell, A., and R. K. Mohn. 1983. Definition of American lobster stocks for the Canadian Maritimes by analysis of fishery-landing trends. TVans. Am. Fish. Soc 112:744-759. Cobb, J. S., and G. R. Tamm. 1975. Dominance status and molt order in lobsters (Homarus americanus). Mar. Behav. Physiol. 3:119-124. Conan, G. Y., and D. R. Maynard. 1983. Aerial survey of spatial distribution of effort in lobster fishery of southern Gulf of St. Lawrence ICES Shellfish Committee CM 1983/K 13, 14 p. Elner, R. W. 1980. Lobster gear selectivity-A Canadian overview. In V. C. Anthony and J. F. Caddy (editors). Proceedings of the Canada-U.S. workshop on status of assessment science for N. W. Atlantic lobster (Homarus americanus) stocks, St. An- drews, N. B., Oct 24-26, 1978, p. 77-83. Can. Tfech. Rep. Fish. Aquat. Sci. 932. Jamieson, G. S. 1982. A system for the precise determination of tow distance and tow path in offshore resource surveys. Can. Tfech. Rep. Fish. Aquat. Sci. 1035, 34 p. Jamieson, G. S., M. Etter, and R. A. Chandler. 1981a. The effect of scallop fishing on lobsters in the western Northumberland Strait. CAFSAC Res. Doa 81/71, 19 p. Jamieson, G. S., N. B. Witherspoon, and M. J. Lundy 1981b. Assessment of Northumberland Strait scallop stocks - 1979. Can. Ttech. Rep. Fish. Aquat. Sci. 1013, 31 p. 1981a Assessment of Northumberland Strait scallop stocks - 1980. Can. Tfech. Rep. Fish. Aquat. Sci. 1017, 44 p. Jones, R. 1959. A method of analysis of some tagged haddock returns. J. Cons. Perm. Int. Explor. Mer 25:58-72. MacPhail, J. S. 1954. The inshore scallop fishery of the Maritime Provinces. Fish. Res. Board Can., Atl. Biol. Stn. Circ, Gen. Ser. No. 22, 4 p. McLeese, D. W., and D. G. Wilder. ■ 1958. The activity and catchability of the lobster (Homarus americanus) in relation to temperature J. Fish. Res. Board Can. 15:1345-1354. Pringle, J. D., and D. J. Jones. 1980. The interaction of lobster, scallop, and Irish moss fisheries off Borden, Prince Edward Island. Can. Tfech. Rep. Fish. Aquat. Sci. 973, 10 p. Pringle, J. D., and G. J. Sharp. 1980. Multispecies resource management of economically im- portant marine plant communities of eastern Canada. Helgol. wiss. Meeresunters. 33:711-720. Robinson, D. G. 1979. Consideration of the lobster (Homarus americanus) recruitment overfishing hypothesis; with special reference to the Canso Causeway. In F D. McCracken (editor), Canso Marine Environment Workshop Part 3 of 4 Parts Fishery Impacts, p. 77-99. Can. Fish. Mar. Serv., Tfech. Rep. 834. Saila, S. B., and j. M. Flowers. 1968. Movements and behavior of berried female lobsters displaced from offshore areas to Narragansett Bay, Rhode Island. J. Cons. Int. Explor. Mer 31:342-351. Scarratt, D. J. 1968. 1973. The effect of raking Irish moss (Chondrus crisjtus) on lobsters in Prince Edward Island. Helgol. wiss. 585 Meeresunters. 24:415-424. 1975. Observations on lobsters and scallops near Pictou, N.S. Can. Fish. Mar. Serv., Ifech. Rep. 532, 6 p. SCARRATT, D. J., AND P. F. ELSON. 1965. Preliminary trials of a tag for salmon and lobsters. J. Fish. Res. Board Can. 22:421-423. Stasko, H. B. 1975. Modified lobster traps for catching crabs and keeping lobsters out. J. Fish. Res. Board Can. 32:2515-2520. Stasko, A. B., T. Amaratunga, and J. F. Caddy. 1977. Northumberland Strait project, Part II. Commercial shellfish data. Can. Fish. Mar. Serv., MS Rep. 1432, 29 p. Temple MAN, W. 1933. Female lobsters handicapped in growth by spawning. Biol. Board Can., Atl. Coast Stn. Prog. Rep. 6, p. 5-6. 1934. Spring, summer and fall lobster fishing in the southern part of the Gulf of St. Lawrence Biol. Board Can. Bull. XLIII, p. 13. FISHERY BULLETIN: VOL. 83, NO. 4 1935. Lobster tagging in the Gulf of St. Lawrenca J. Biol. Board Can. 1:269-278. 1936. Local differences in the life history of the lobster (Homarus americanus) on the coast of the Maritime pro- vinces of Canada. J. Biol. Board Can. 2:41-88. Wilder, D. G. 1947. The lobster fishery of the southern Gulf of St. Lawrenca Fish. Res. Board Can., Atl. Biol. Stn. Circ, Gen. Ser. No. 24, 16 p. 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. 1965. Lobster conservation in Canada. Rapp. P.-v Reun. Cons. int. Explor. Mer 156:21-29. Wilder, D. G., and R. C. Murray. 1956. Movements and growth of lobsters in Egmont Bay, P.E.I. Fish. Res. Board Can., Prog. Rep. Atl. Coast Stn. 64, p. 3-9. 586 1 AGE, GROWTH, AND DISTRIBUTION OF LARVAL SPOT, LEIOSTOMUS XANTHURUS, OFF NORTH CAROLINA Stanley M. Warlen and Alexander J. Chester^ ABSTRACT Age and growth of the early life history stages of spot, Leiostomus xanthurus, were determined from daily growth increments on otoliths of larval and early juvenile spot collected from Beaufort Inlet, NC, to the continental shelf break during the fall and winter of 1978-79 and 1979-80. Spawning occurred on the mid to outer continental shelf between early November and early March, but appeared to be concen- trated from mid-December through January. Generally, the youngest larvae were found further offshore; ages and lengths increased closer to shore Larvae entered the estuary at an average age of 59 days (range 40-74 days) and an average size of 13.6 mm (range 11.3-15.6 mm). Significantly younger and smaller lar- vae immigrated at the beginning and end of the immigration period. Fish entered the estuary segregated by age as indicated by the small within-sample variation in age A Gompertz growth equation was used to express the relationship between age and standard length for 69 larvae collected in 1978-79 and 557 collected in 1979-80. Spot grew from about 1.6 mm SL at hatching to 17-19 mm SL at 90 days. There were no significant differences in growth parameters between years; age-specific growth rates declined from =5%/day at age 10 days to 8 mm) larvae generally occurred closer to shore within about 40 km of Beaufort Inlet in <25 m of water (stations 10-13, 19), except during February and March 1980 when some larvae were caught 50 km offshore at station 14. Spawning apparently is continuous between late October and late February, since young larvae occurred every month at the three offshore stations (16-18). Most spawning off North Carolina probably occurs between 75 and 95 km offshore, ex- cept for some activity in the mid-shelf area early in the spawning season. Our age-length data provided no evidence that spot spawn near shore Estuarine Immigration Larvae entered the Newport River estuary over a 4-mo period from about mid-December to mid- April (Figs. 2, 5). None were caught prior to December 1979 nor after 17 April 1980. Relative abundance of larvae collected at Pivers Island dur- 589 FISHERY BULLETIN; VOL. 83, NO. 4 Table 1.— Number of larval spot per 100 m^ caught off North Carolina during the seasons of 1978-79 and 1979-80. N.S. = No sample taken. 1978-79 cruise Station number No. Date 1 2 3 4 5 6 7 8 9 10 1 6-7 Dec. 1.9 0 0 0 0 0 0 0 0 0 2 4-5 Jan. 5.0 0 N.S. N.S. 5.6 0 0 19.3 0 0 3 31 Jan.-I Feb. N.S. N.S. N.S. N.S. N.S. N.S. N.S. N.S. 0 0 4 7-8 Mar. 1.8 6.4 N.S. N.S. N.S. 0 0 1.7 2.7 0 5 26-27 IVIar. 0 0 N.S. N.S. N.S. 0 0 0 0 13.3 6 23-24 Apr. 1979-80 0 N.S. 0 0 0 0 0 0 0 0 cruise Stat ion number No. Date Inleti 11 12 13 14 15 16 17 18 7 15-16 Nov. N.S. 0 0 0 N.S. 43.1 3.3 N.S. N.S. 8 3-4 Dec. N.S. 0 0 3.1 1.1 0 4.0 1.1 45.0 9 15-16 Jan. N.S. 2.4 0.4 0.7 108.6 8.8 4.1 N.S. N.S. 10 11-13 Feb. N.S. 0 0 4.1 31.5 17.9 0.6 0.3 2.3 11 19-20 Mar. 65.6 0 0 0.7 4.2 11.4 0.4 0.6 0 ^Same as station 19. ing the winter periods of 1967-70 showed major im- migration peaks in February and March (unpublished data from R. M. Lewis, Beaufort Laboratory; Thayer et al. 1974). The duration of estuarine immigration generally reflected the duration of the spawning season (Fig. 2). Statistically significant monthly variations (ANOVA, P < 0.05) in both age and length of larvae entering the New^port River estuary were observed from December to April (Fig. 5). Mean age at entry increased linearly from December to March and then decreased in April. Thus, larvae spawned at the beginning or end of the season spent relatively less time in the offshore environment than did larvae spawned in the middle of the season. Length follow- ed a similar trend, except during January and early February when it remained about constant, in- dicating a declining rate of growth. As determined from seven samples collected at Pivers Island (Fig. 5) and one at Beaufort Inlet (19 March 1980), spot entering the estuary averaged 59 d-old (range 40-80). In general, larvae entering the estuary together had similar spawning dates. As a rule, 50% of the fish in any Pivers Island sample had been spawned within a period of 5 d and all had been spawned within a period of 14 d (Fig. 2). The one exception was the last sample from Pivers Island in which several larvae were more than a month older than the majority of fish. We infer from the generally small variation in age of fish within a sample that a continuum of cohorts moved past Pivers Island enroute to the upstream parts of the estuary and that early juveniles entered the lower estuary segregated by age. Growth Estimates Average growth of larvae was described by the Laird version (Laird et al. 1965) of the Gompertz growth equation (Zweifel and Lasker 1976) fitted to estimated age and size at time of capture data for 1978-79 and 1979-80 (Fig. 6). Variance about the estimated growth curve was assumed to represent genetic differences in growth potential and the ef- fects of differing environmental conditions over the year (Pennington 1979). To stabilize the variance of length over the observed age interval, we used the log-transformed version of the Gompertz growth equation: In [L.J = In [L,.,] + '(0) 'W (0)J a [1 - e-"'] (2) where L,„ = length at time t, length at i = 0, specific growth rate at ^ = 0, rate of exponential decay of the specific growth rata '(0 ^(0) = The time origin {t = 0) was selected as hatching time (day 0) and values for L q,, A,q,, and a were obtain- ed by nonlinear regression. Age accounted for 96% of the variation in length for one year class (1978-79) and 91% of the variance in length for the other (1979-80) in the log-transformed models. We estimated that spot grew from about 1.6 mm SL at hatching to 17-19 mm at 90 d. The predicted size at hatching agrees well with laboratory observations of Powell and Gordy (1980). Population growth 590 VVARLEN and CHESTER: LARVAL SPOT OFF NORTH CAROLINA Q. < X o < 2 20 10 30 20 10 I- < o UJ oc ^ oc o < 3 >- $ 20 3 OC m u. 10 (21) -cm PI (21) —cm — INLET (12) HXD 1 PI H h (48) I — n C-11 30 20 10 30 OC g 20 g 10 "30 S 20 CO > O 10 (7) I — CEM PI ( (225) 113- (9) KXH PI (9) \-cn — I PI C-10 (84) -cm — I C-9 (9) i-nzH PI U (9) I TIH PI (32) -CD— I C-8 (71) -a- C-7 10 20 I NOVEMBER T I I 10 20 DECEMBER "W I I I I 10 20 JANUARY 30. 10 20 FEBRUARY J L 10 I MARCH SPAWNING DATE Figure 2.— Schematic plots of the spawning times of larval spot caught in the ocean (cruises 7-11 of RV John de Wolf 11) and late-larvae/early-juvenile spot caught in the Newport River estuary at Pivers Island, NC. In each distribution, the vertical line is the median value and 50% of the data points fall within the block. Lines beyond the boxes represent the range of data points. curves were not significantly different between years [Hotelling's T~ test of A . a, and L- Bernard (0)' '(0)' (1981) as modified by Hoenig and Hanumara (1983)]. Age-specific growth rates for both years decHned from =5%/d at age 10 d to « (d •o ^^ UJ O < 60 50 40 30 T I A- length • = age I I I f i i I I I I I I I 1 I (9) (9) (9) (9) (7) (12) (21) 16 15 14 13 12 E E >,^ I I- O z UJ _l o tr < Q Z < co 11 DEC JAN FEB MAR APR 10 IMMIGRATION TIME Figure 5.— Age (mean + 1 standard error) and standard length (mean ± 1 standard error) of late-larval spot entering the Newport River estuary in North Carolina, December 1979-April 1980. The numbers offish measured and aged at each sampling date are in parentheses. both groups was identical (22 December 1979), the variance about the mean was greater for Inlet- caught fish. Consequently, back-calculated lengths also were more variable for Inlet-caught fish {F-test, P < 0.05), but on the average they appeared to be larger at every age (^test corrected for unequal variance, P < 0.05). Significant differences were found for the weight- length relation (Fig. 7) of laboratory-reared larvae <6 mm and those >6 mm (ANCOVA, P < 0.001). We selected 6 mm as the dividing point because basic changes in body form had been observed to occur at around 6 mm (Powell and Gordy 1980). The length exponent for spot <6 mm SL (4.201) was close to the mean value (4.152) reported by Laurence (1979) for larvae of seven marine fishes, while larvae > 6 mm (3.282) approached isometric growth (Ricker 1975). 594 WARLEN and CHESTER: LARVAL SPOT OFF NORTH CAROLINA 20 15- 10 S 5 \- (D Z UJ -I o < a z < I- co 1978-1979 15 10 • •• L(,)=1.686e2«39(l-e-00211' n= 69 1979-1980 L. . = 1 .609e n=557 2.624 l-e -0.0255t 10 20 30 40 50 60 70 80 90 100 AGE (days) Figure 6— Growth of larval and early-juvenile spot collected from oceanic and estuarine waters of North Carolina in the fall-winter, 1978-79 and 1979-80. The Laird-Gompertz growth model was used to describe the data. Estimates of the parameters were obtained by fitting the log-transformed version of the model to the data. "T" is the point of maximum absolute growth (inflection point) in the growth curve 595 FISHERY BULLETIN: VOL. 83, NO. 4 Table 2.— Mean back-calculated standard length of 10 spot from each of two collections of the same cohort (average spawning date 22 December 1979) taken about 2 mo apart in 1980. 15-16 January 19 March Age (d) Stations 15-16i Station 19 (Beaufort Inlet)^ SL (mm) SD N SL (mm) SD N 5 3.1 0.13 10 3.5 0.39 10 10 3.7 0.15 10 4.4 0.49 10 15 4.6 0.32 10 5.1 0.62 10 20 5.1 0.34 7 6.0 0.67 10 25 6.0 0.35 2 7.0 0.80 10 ^Larval mean age, 23 d; mean size, 5.7 mm ^Larval mean age, 83 d; mean size, 15.0 mm DISCUSSION Although spot is a winter spawner, it spawns in relatively warm water. Very young larval spot (<15 d) occurred only in water above 19.3°C, an observa- tion corroborated by experimental evidence in- dicating that spot spawn only between 17.5° and 25°C (Hettler and Powell 1981). In late fall and early winter off North Carolina, such warm temperatures are found only on the outer continental shelf near the Gulf Stream. Newly ripe adults probably emigrate in the fall of the year from the cooling waters of bays and sounds in Virginia (Hildebrand and Schroeder 1928), North Carolina (Roelofs 1951), and South Carolina (Dawson 1958) to spawn in such warm waters. Hildebrand and Schroeder (1928) and Dawson (1958) also suggested that spot spawn along the outer continental shelf. Warm coastal waters in the fall and the influence of warm Gulf Stream waters later in the season may provide a suitable spawning temperature regime over a long period. The extended (4.5 mo) spawning season of spot is typical of the general pattern for Atlantic coast sciaenids (Powles 1981). The spawn- ing season of spot in North Carolina in 1979-80 was similar to that found by Hildebrand and Cable (1930) in North Carohna and by Dawson (1958) in South Carolina. Because most of the larvae caught off North Carolina were spawned in December and January, we conclude that these are the months of peak spawning. This conclusion is supported by the observation that peak estuarine immigration occurs in February and March (unpublished data from R. M. Lewis, Beaufort Laboratory; Thayer et al. 1974) for fish we estimate to have been about 2 mo-old. Hildebrand and Cable (1930) and Lewis and Judy (1983) also inferred, from length-frequency informa- tion, that peak spawning occurs in December and January. The trend of decreasing larval age and size (Figs. 2, 3) with distance from shore supports the idea that spot spend virtually their entire larval period in the ocean. Berrien et al. (1978) and Lewis and Judy (1983) also noted an inverse trend of size with distance from shore in the same area to 79 km off- shore A similar trend may exist in the Gulf of Mex- ico where Fruge (1977) found small larval spot to be most abundant 60-80 km off the Louisiana coast. By the time larvae have been transported to shore and enter estuarine nursery areas, they have reach- ed the late larval or early juvenile stage. Although the mechanism is unclear by which lar- val spot from 74 to 93 km offshore arrive at the estuary in about 60 d, their initial onshore movement is probably a passive transport by water currents in Onslow Bay. A consistent counterclockwise eddy (Stefansson et al. 1971) and a strong indication of bottom drift in a northerly direction on the outer and mid-continental shelf and directly to the coast inshore during January-April (Bumpus 1973) could aid in the transport of larvae. Nelson et al. (1977) considered that zonal Ekman transport was a signifi- cant mechanism for movement of larval Atlantic menhaden, Brevoortia tyrannus, from offshore spawning grounds to inshore nursery grounds in the same study area at about the same season of year. Data from recent years, however, does not lend sup- port for this hypothesis (Schaaf^). A recent analysis by Yoder (1983) suggested that mean Ekman transport does not favor onshore flow in surface waters during winter off the southeastern United States. Rather, cross-shelf transport of larval fishes may depend on highly variable, short-term meteo- rological events which reverse the mean surface flow. Variations in transport rates of larvae in the ocean as well as spawning at variable distances from shore may be responsible for the seasonal differences in age and length at immigration (Fig. 5). In addition, factors affecting growth, such as temperature and the distribution of food organisms, interact with the physical factors of transport to produce the temporal pattern of age and length observed in a given year. Young spot undergo several environmentally related changes in growth during their larval and juvenile stages. Growth in length of larval spot is rapid (initially approaching 7%/d) and coincides with the winter peak of plankton productivity in the relatively warm water of the outer continental shelf (Tbrner et al. 1979; TUrner 1981; Yoder et al. 1981; ^W. E. Schaaf, Southeast Fisheries Center Beaufort Laboratory, National Marine Fisheries Service, NOAA, Beaufort, NC 28516-9722, pers. commun. January 1984. 596 WARLEN and CHESTER: LARVAL SPOT OFF NORTH CAROLINA 100,000 >6 mm fish Weight =1.4516 Length - n=93, r2«0.986 10,000 3 S2 lij > CO a 1,000 3.282 100 10 ^ 6 mm fish Weight= 0.2230 LengtH*-^^'' n=32, r2=0.982 J I I I 4 6 810 J I I I 100 BODY LENGTH (mm) Figure 7— Relationships between dry weight and standard length of spot for length classes <6 mm and >6 mm. Fish were from laboratory spawned and reared stocks. Yoder et al. 1983). By the time larvae enter the cooler (often <10°C) coastal and estuarine waters, growth rate has slowed considerably (<1.5%/d). The asymp- tote of 22.2 mm SL (Fig. 6, 1979-80 data) estimated by our growth model corresponds closely to the size of juvenile spot collected early in their estuarine residency (Weinstein and Walters 1981). Increase in length of newly immigrated spot is relatively slow (=0.5%/d from December to March, estimated from figure 3 of Weinstein and Walters 1981), and it is 597 FISHERY BULLETIN: VOL. 83, NO. 4 not until after the usual peak in plankton abundance (Thayer et al. 1974) and increases in water temper- ature that growth rates accelerate and persist at a high level through the summer (=1.0%/d from April to August, estimated from figure 3 of Weinstein and Walters 1981). Within the same cohort, older fish had statistically larger back-calculated sizes at each age than did younger fish (Tkble 2). One explanation is that size- selective mortality (eg., predation, Bailey 1984) favors survival of faster growing larvae and that the apparent growth rate depends on the size (and age) of larvae on which it is calculated. Alternatively, the two groups may have been spawned in different loca- tions and experienced different environmental con- ditions that could affect growth. ACKNOWLEDGMENTS We thank Mary Boyd for preparing and reading the otoliths and John Merriner, William Nicholson, and William Schaaf for their critical reviews of earlier drafts of this manuscript. We also thank the many participants on the 1 1 cruises of the RV John de Wolf II. This research was supported by a cooperative agreement between the National Marine Fisheries Service, NOAA, and the U.S. Department of Energy. LITERATURE CITED Bailey, K. M. 1984. Comparison of laboratory rates of predation on five species of marine fish larvae by three planktonic inverte- brates: effects of larval size on vulnerability. Mar. Biol. (Berl.) 79:303-309. Bernard, D. R. 1981. Multivariate analysis as a means of comparing growth in fish. Can. J. Fish. Aquat. Sci. 38:233-236. Berrien, P. L., M. P. Fahay, A. W. Kendall, Jr., and W. G. Smith. 1 978. Ichthyoplankton from the RV Dolphin survey of con- tinental shelf waters between Martha's Vineyard, Massa- chusetts and Cape Lookout, North Carolina, 1965-66. U.S. Dep. Commer., NOAA, Northeast Fish. Cent., Sandy Hook Lab., Tfech. Ser. Rep. 15, 152 p. Brothers, E. B., C. P. Mathews, and R. Lasker. 1976. Daily growth increments in otoliths from larval and adult fishes. Fish. Bull., U.S. 74:1-8. Bumpus, D. F. 1973. A description of the circulation on the continental shelf of the east coast of the United States. Prog. Oceanogr. 6:111-157. Chao, L. N., and J. A. MusiCK. 1977. Life history, feeding habits, and functional morphology of juvenile sciaenid fishes in the York River estuary, Virginia. Fish. Bull., U.S. 75:657-702. Dawson, C. E. 1958. A study of the biology and life history of the spot, Leiostomus xanthurtis Lacepede, with special reference to South Carolina. Contrib. Bears Bluff Lab. 28:1-48. Fahay, M. P. 1975. An annotated list of larval and juvenile fishes captured with surface-towed meter net in the south Atlantic Bight dur- ing four RV Dolphin cruises between May 1967 and February 1968. U.S. Dep. Commer., NOAA Tfech. Rep. NMFS SSRF-685, 39 p. Fruge, D. J. 1977. Larval development and distribution of Micropogon un- dulatus and Leiostomus xanthums and larval distribution of Mugil cephalv£ and Bregmaceros atlanticus off the south- eastern Louisiana coast. M.S. Thesis, Louisiana State Univ., Baton Rouge, 75 p. Fruge, D. J., and F M. Truesdale. 1978. Comparative lar\'al development of Micropogon un- dulatvs and Leiostomics xanthurus (Pisces: Sciaenidae) from the northern Gulf of Mexico. Copeia 1978:643-648. GovoNi, J. J., D. E. Hoss, and A. J. Chester. 1983. Comparative feeding of three species of larval fishes in the northern Gulf of Mexico: Brevoortia patronus, Leiostomics xanthurus, and Micropogonias undulatiis. Mar. Ecol. Prog. Ser. 13:189-199. Hettler, W. F. 1979. Modified neuston net for collecting live larval and juvenile fish. Prog. Fish-Cult. 41:32-33. Hettler, W. F, and A. B. Powell. 1981. Egg and larval fish production at the NMFS Beaufort Laboratory, Beaufort, N.C., USA. Rapp. P.-v Reun. Cons. int. Explor. Mer 178:501-503. Hildebrand, S. F., and L. E. Cable. 1930. Development and life history of fourteen teleostean fishes at Beaufort, N.C. Bull. U.S. Bur. Fish. 46:383-488. Hilldebrand, S. F., and W. C. Schroeder. 1928. Fishes of Chesapeake Bay Bull. U.S. Bur. Fish. 43:1-366. HoENiG, N. A., and R. C. Hanumara. 1983. Statistical considerations in fitting seasonal growth models for fishes. Cons. Int. Explor. Mer, C. M. 1983/D:25, 25 p. Laird, A. K., S. A. Tyler, and A. D. Barton. 1965. Dynamics of normal growth. Growth 29:233-248. Laurence, G. C. 1979. Larval length-weight relations for seven species of northwest Atlantic fishes reared in the laboratory. Fish. Bull., U.S. 76:890-895. Lee, R. M. 1920. A review of the methods of age and growth determina- tion of fishes by means of scales. Fish. Invest., Minist. Agric Fish. Food (G.B.) Ser. II, 4(2):l-32. Lewis, R. M., and M. H. Judy. 1983. The occurrence of spot, Leiostomus xanthurus. and Atlantic croaker, Micropogonias undulatus, larvae in Onslow Bay and Newport River estuary, North Carolina Fish. Bull., U.S. 81:405-412. McHUGH, J. L. 1966. Management of estuarine fisheries. In R. F. Smith, A. H. Swartz, and W. H. Massman (editors), A symposium on estuarine fisheries, p. 133-154. Am. Fish. Soc, Spec Publ. 3. Nelson, W. R., M. C. Ingham, and W. E. Schaaf. 1977. Larval transport and year-class strength of Atlantic menhaden, Brevoortia tyrannus. Fish. Bull., U.S. 75:23-41. Pannella, G. 1971. Fish otoliths: Daily growth layers and periodical pat- terns. Science (Wash., D.C.) 173:1124-1127. Pennington, M. R. 1979. Fitting a growth curve to field data. In J. K. Ord, G. P. Patil, and C. Ikillie (editors). Statistical distributions in 598 WARLEN and CHESTER: LARVAL SPOT OFF NORTH CAROLINA ecological work, p. 419-428. Int. Co-op. Publ. House, Fairland, MD. Peters, D. S., J. C. DeVane, Jr., M. T. Boyd, L. C. Clements, and A. B. Powell. 1978. Preliminary observations on feeding, growth and energy budget of larval spot (Leiostomits xanthwrus). In Annual Report of the NMFS, Beaufort Laboratory, Beaufort, N.C., to the U.S. Department of Energy, p. 377-397. Powell, A. B., and H. R. Gordy. 1980. Egg and larval development of the spot, Leiostomvs xan- thurus (Sciaenidae). Fish. Bull., U.S. 78:701-714. Powles, H. 1981. Eggs and larvae of North American sciaenid fishes. In H. Clepper (editor), Marine recreational fisheries 6, p. 99-109. Sport Fish. Inst., Wash., D.C. Powles, H., and B. W. Stender. 1976. Observations on composition, seasonality and distribu- tion of ichthyoplankton from MARMAP cruises in the South Atlantic Bight in 1973. S.C. Mar. Resour. Cent., Tfech. Rep. Ser. 11, 47 p. RiCKER. W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191, 382 p. ROELOFS, E. W. 1951. The edible finfishes of North Carolina. In H. F. Ikylor (editor). Survey of the marine fisheries of North Carolina, p. 109-139. Univ. N.C. Press. Chapel Hill. StefXnsson, U., L. p. Atkinson, and D. F. Bumpus. 1971. Hydrographic properties and circulation of the North Carolina shelf and slope waters. Deep-Sea Res. 18:383-420. Thayer, G. W., D. E. Hoss, M. A. Kjelson, W. F. Hettler, Jr., AND M. W. Lacroix. 1974. Biomass of zooplankton in the Newport River estuary and the influence of postlarval fishes. Chesapeake Sci. 15:9-16. TllRNER, R. E. 1981. Plankton productivity and the distribution of fishes on the southeastern U.S. continental shelf. Science (Wash., DC.) 214:353-354. TliRNER, R. E., S. W Woo, and H. R. Jitts. 1979. Estuarine influences on a continental shelf plankton community. Science (Wash., D.C.) 206:218-220. Warlen, S. M. [1982]. Age and growth of larvae and spawning time of Atlan- tic croaker in North Carolina. Proc Annu. Conf. Southeast. Assoc Fish Wildl. Agencies 34:204-214. Weinstein, M. p., and M. P. Walters. 1981. Growth, survival and production in young-of-year populations of Leiostomits xanthwrus Lacepede residing in tidal creeks. Estuaries 4:185-197. Yoder, J. A. 1983. Statistical analysis of the distribution of fish eggs and larvae on the southeastern U.S. continental shelf wath com- ments on oceanographic processes that may affect larval sur- vival. Estuarine Coastal Shelf Sci. 17:637-650. Yoder, J. A., L. P. Atkinson, J. 0. Blanton, D. R. Deibel, D W Menzel, and G. a. Paffenhofer. 1981. Plankton productivity and the distribution of fishes on the southeastern U.S. Continental Shelf. Science (Wash., D.C.) 214:352-353. YoKER, J. A., L. P. Atkinson, S. S. Bishop, E. E. Hofmann, and T N. Lee. 1983. Effect of upwelling on phytoplankton productivity of the outer southeastern United States continental shelf. Continental Shelf Res. 1:385-404. ZWEIFEL, J. R., and R. LASKER. 1976. Prehatch and posthatch growrth of fishes— a general model. Fish. Bull., U.S. 74:609-621. 599 DIET OF PACIFIC COD, GADUS MACROCEPHALUS, AND PREDATION ON THE NORTHERN PINK SHRIMP, PANDALUS BOREALIS, IN PAVLOF BAY, ALASKA W. D. Albers and p. J. Anderson' ABSTRACT Analysis of 455 Pacific cod, Gadus macrocephalus, stomachs collected in 1980 and 1981 from Pavlof Bay, in the western Gulf of Alaska, showed considerable predation on northern pink shimp, Pandalus borealis. The most frequently occurring prey items were pink shrimp, P. borealis, 63%; euphausids, 41%; walleye pollock, Theragra chalcogramma, 27%; and capelin, Mallotus villosus, 26%. Pandalid shrimp and snow (Tknner) crab occurred more frequently with increasing cod size (30-69 cm fork length). Euphausids decreased in frequency of occurrence with increasing cod size Pink shrimp length distributions from cod stomachs and trawl samples were similar. Estimated consumption of pink shrimp by cod in Pavlof Bay ranged from 142 to 857 t over a 112-day period from late May through mid-September 1981. Cod preda- tion may be one reason for the failure of the pink shrimp stock to rebuild in Pavlof Bay following closure of the commercial fishery in 1979. Cod predation may also play a role in keeping other reduced pink shrimp stocks in the western Gulf of Alaska from rebuilding to former levels. ) Pacific cod, Gadus macrocephalus, predation on northern pink shrimp, Pandalus borealis, in Pavlof Bay (Fig. 1) was studied to determine if it is a fac- tor in keeping the pink shrimp stock from rebuilding thera National Marine Fisheries Service (NMFS) and Alaska Department of Fish and Game (ADF&G) survey data from the late 1970's indicate that when pink shrimp populations in regions of western Alaska began to decrease, cod abundance started to in- crease Pink shrimp has been reported to be an im- portant food item in the diet of Pacific cod in the Gulf of Alaska (Jewett 1978; Hunter 1979). Preda- tion of pink shrimp by cod may have substantial in- fluence on shrimp stock abundance Pavlof Bay was chosen as the study area because it supported a commercial fishery for pandalid shrimp in the 1970's and is suspected to contain a geographically isolated stock of pink shrimp (Ander- son 1981). From 1972 through 1979, 13,641 1 of pink shrimp were commercially harvested from Pavlof Bay (calculated from ADF&G commercial catch data and NMFS survey data). Survey data from Pavlof Bay indicate that in 1977 and 1978 when pink shrimp abundance began decreasing, cod abundance began increasing (Fig. 2). Following the 1979 season the bay was closed to commercial shrimping due to depressed shrimp abundance levels which remain- ed low through 1983. This report presents data which suggest that Pacific cod predation is a factor in keeping shrimp stocks from rebuilding. The summer diet of cod, prey size selectivity, and an estimate of pink shrimp biomass consumed by cod in Pavlof Bay during a 112-d period from late May through mid-September 1981 are discussed. MATERIALS AND METHODS Pacific cod were collected from 31 tows during three trawl surveys. The first collection was done by NMFS on 25-26 August 1980, the second by ADF&G on 23-25 May 1981, and the third by NMFS on 10-11 September 1981. The collecting was done during daylight hours over a period of about 14 h a day. All three surveys used a high-opening shrimp trawl with an 18.6 m headrope and footrope described by Wathne (1977). Mesh size of the trawl is 32 mm and path-width is about 10 m. Each tow was about 1.8 km in length. Randomly selected sampling locations were restricted to depths >55 m since previous surveys showed that neither shrimp nor cod were found in abundance in shallow water. Both shrimp and cod are uniformly distributed at depths >55 m in Pavlof Bay. When possible, five stomachs per 5 cm interval of fork length (FL) were removed from every trawl catch and preserved in 10% Formalin^. In the 'Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake Blvd. E., Seattle, WA 98112. ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 601 FISHERY BULLETIN: VOL. 83, NO. 4 Figure 1.— Location of the study area of pink shrimp and Pacific cod (55 m isobath indicated by dotted/dash-line). laboratory, prey items were sorted to the lowest possible taxon. Frequency of occurrence (number of stomachs containing the food item), number of each prey item, volume by water displacement (nearest 0.1 mL), and wet weight (nearest gram) were record- ed. From these measurements, percentages by fre- quency of occurrence, number, and volume were calculated from non-empty stomachs only. Size composition was recorded for prey species of commercial importance including pink shrimp (carapace length, CL); humpy shrimp, Pandalus goniuriLs, CL; snow (Tknner) crab, Chionoecetes bair- di, carapace width (CW); and walleye pollock, Theragra chalcogramma, fork length (FL). Addi- tionally, size composition was recorded for samples of pink shrimp (CL), humpy shrimp (CL), Pacific cod (FL), and walleye pollock (FL) caught in each tow. lb determine if Pacific cod were feeding on selected sizes of pink shrimp, the Kolmogorov- Smirnov test (Sokal and Rohlf 1969) was used to test 602 ALBERS and ANDERSON: DIET OF PACIFIC COD for a significant difference between pink shrimp length distributions measured from Pacific cod stomachs and those measured from the trawl. Pink shrimp lengths tested were ^16.5 mm CL. Ander- son (1981) reported that shrimp <16.5 mm CL are not fully vulnerable to trawl capture Population biomass estimates for pink shrimp and Pacific cod were calculated using the area swept technique (Alverson and Pereyra 1969). An estimate of pink shrimp biomass consumed by Pacific cod in Pavlof Bay during the 112-d period be- tween the late May and mid-September 1981 surveys was determined through methods described by Minet and Perodou (1978). Undigested weights (W) of pink shrimp were determined from carapace lengths using the weight-length relationship W = 0.000802 (CL)-'^^^ (calculated from Pavlof Bay pink shrimp length-weight data). The mean weight of pink shrimp per stomach for each 5 cm length group of Pacific cod was calculated from stomachs where at least 80% of the pink shrimp were measurable (Ikble 1). A stomach which contained <80% measurable pink shrimp was deemed not suitable for determining the weight of undigested shrimp consumed. These data were then weighted using 768 cod lengths measured during the three surveys. The mean weight of pink shrimp in the stomach of an average-sized cod was then estimated for each survey (Tkble 1). The average rates of elimination (r) of food from Pacific cod stomachs collected during the May and September 1981 surveys were calculated from Jones' (1974) equation for food elimination rates from Atlantic gadoids including Atlantic cod, Gadus morhua. Jones found that the rates of elimination for the three species of gadoids studied were effec- tively the same, adjusting for fish and meal size, and temperature Since Pacific cod are very similar to Atlantic cod, we used Jones' equation in the absence of more relevant information: r (g/h) = j^QO.035 {To - Tc) J^O.46 QJ^li 175 where Tc = 6°C; temperature of the experiment (Jones 1974). To = 5°C (May), 7°C (September); observed temperature in Pavlof Bay from expen- dable bathythermograph data. X = 98.9 g (May), 109.0 g (September); the average weight of food found in one stomach. The average weight of food was estimated for each 5 cm length group of Pacific cod and then weighted using 502 cod lengths measured dur- ing the May and September surveys. Stomachs containing at least 60% of the greatest weight of food encounter- ed for each cod length group were used to determine X. Q = 0.12; the average rate of elimination of 1 g of food from the stomach of a 40 cm gadoid (Jones 1974). L = 52.7 cm (May), 53.0 cm (September); the average length of Pacific cod calcu- lated from 236 fish measured in May and 266 fish measured in September. E C a o o c o T) c D M < 0 0 Qj 800 600 — 400 — r— 600 200 — 400 — 200 o 0 Q. > cr c Q. Q D O 0 :r > to o oo (0 3 < u a E o a> c o Q. E jsi: 0) to gi T3 c 3 c 10 E c g w (0 O I 00 l3 00 0) E B a. (X> (0 3 o> 3 < CD = o^ .5>1o E O) 0} c » T3 £1 O E " C3)J CD en l|i .s't; E ffi 0) ;= £> O E " 0> u) t CD ' o o -^ CO p r^ cvi CO o T^ cO cO CO CO , CO 'T -"t CM 1- CvJ I 1- CO I- 00 CM CD CO c^ CJCMCMCOOOOJ'* CO h^ooaiiri^cvjcvj oi cvioo'd-cDr^oir) lo 00 CD •>- 00 T- 1^ O) CM 5|a ^ 00 Q. .5>to E 0) CD C CD *:g'-SS Q> 3 C 2-5^ COpcOCMppoO 05 cocJjcJ>^^r^Kc^ \p ■.- 1- CM 1- IT) in 1^. CMcn'S-oocO'^co CO 'S- C3> CO 1- CO CM ■«t in CD CM CD in 1- co CO CM O CM ^ O •■- CM CM t^ CM CO r^ 1- in ■* o CM CJ) in d) -o JO o E " 2 cj t ft" CM in 00 1^ ■>- in O CD CO CO h^ 00 CM CO (O CM CM CM CJl h- CO CD ■.- T- -"U" CO in CD CO CvJ I O) C3 ^0>^0>^CT>^0 coco'S-'^inincoco ciincbincbmoin cocO'^'tinincDcD Of 455 Pacific cod stomachs examined, 435 con- tained food. Data obtained from cod stomachs col- lected in August 1980, May 1981, and September 1981 are presented in Ikble 2. Combining these data, the most frequently occurring prey items were pink shrimp (63%), euphausids (41%), walleye pollock (27%), and capelin, Mallottis villosus, (26%). Crusta- ceans, mostly shrimp, crab, and euphausids, were present in 93% of the stomachs containing food and comprised 45% of the volume Ifeleosts, mostly wall- eye pollock and capelin, were present in 60% of the stomachs containing food and comprised 54% of the volume Gastropods, bivalves, sipunculids, and asci- dians together comprised 1% of the volume Crustaceans made up the greatest part of the diet in the May sample and teleosts comprised the greatest part of the diet in the August and September samples (Ikble 2). Shrimp, mostly pink shrimp, made up 67% of the volume in May com- pared with 28%, combining August and September data. Crabs, mainly snow crab, comprised 10% of the volume in May compared with 3% in August- September. Teleosts, primarily walleye pollock and capelin, made up 4% of the volume in May and 65% in August-September. Data on the 10 most frequently occurring food items was categorized by 10 cm length (FL) groups of Pacific cod in a size range of 30 through 69 cm (Tkble 3). Hurtubia's (1973) trophic diversity method was used to assure that enough stomach samples had been analyzed to give representative values for each length group. Pink shrimp occurred most frequent- ly in all but one length group (30-39 cm) and was the dominant food item by volume in all length groups. Pink shrimp and snow crab increased both in frequency of occurrence and volume with in- creasing cod size Conversely, euphausids were the most frequently encountered food item in the 30-39 cm length group, but their frequency of occurrence and volume decreased with increasing cod size No trends were evident for the occurrence of teleosts. Although data from the three surveys were combined for Tkble 3, the above trends were evident in each of the surveys. The size ranges and mean sizes of pink shrimp and humpy shrimp consumed by Pacific cod were similar in general to those found in the trawl (Tkble 4). Size range and mean size of walleye pollock consumed by cod were considerably smaller than those fish cap- tured by the trawl. 604 ALBERS and ANDERSON: DIET OF PACIFIC COD Table 2.— Percent frequency ot occurrence (F), percent by number (N). and percent by volume {V) of food items in Pacific cod stomachs for August 1980 and May and September 1981. Food categories followed by (total) is tfie sum of the food items that fall within that category for percent number and percent volume. Percent frequency of occur- rence is the number of stomachs containing the food category or item divided by the total number of stomachs con- taining food. August 1980 May 1981 September 1981 (n = 202) (n = 63) (n = 170) Food items F N V F N V F N V Crustacea (total) 93.1 92.2 29.4 100.0 99.5 95.0 91.2 57.6 37.6 Amphipoda — — — — — — 1.8 0.3 <0.1 Euphausiacea 53.5 77.0 6.1 81.0 87.3 17.8 12.4 3.3 0.1 Decapoda Natantia (total) 71.3 15.0 23.0 76.2 10.4 67.1 80.6 46.1 32.2 Pandalidae — — — 3.2 0.1 0.1 1.8 0.2 <0.1 Pandalus borealis 56.9 9.3 18.5 65.1 7.6 59.7 70.6 34.9 29.9 Pandalopsis dispar — — — 17.5 0.5 1.0 1.2 0.1 <0.1 Pandalus goniurus 20.3 1.7 2.3 25.4 0.4 1.7 4.1 0.9 0.5 Pandalus hypsinotus 0.5 <0.1 0.1 4.8 0.1 1.1 1.2 0.1 0.3 Crangonidae — — — — — — 4.1 0.8 0.2 Crangon sp. 3.0 0.2 0.1 7.9 0.1 0.2 2.9 0.5 <0.1 Crangon communis 2.5 0.2 0.1 12.7 0.2 0.3 18.2 2.9 0.5 Crangon dalli — — — — — — 3.5 0.5 0.1 Argis sp. — — — 4.8 0.1 0.3 1.2 0.1 <0.1 Argis dentata — — — 4.8 0.1 0.3 1.8 0.2 <0.1 Argis far — — — — — — 4.7 0.7 0.2 Hippolytidae 0.5 <0.1 <0.1 6.3 0.1 0.1 — — — Eualus sp. 0.5 <0.1 <0.1 — — — — — — Eualus macilenta — — — 11.1 0.3 0.2 5.3 0.8 <0.1 Eualus suckleyi 0.5 <0.1 <0.1 1.6 0.1 0.2 2.9 0.4 0.1 Unidentified Natantia 26.7 3.6 1.9 30.2 0.7 1.9 20.0 3.0 0.4 Reptantia (total) 3.5 0.2 0.3 39.7 1.8 10.1 31.2 7.9 5.3 Lithodidae Paralithodes camtschatica 0.5 <0.1 <0.1 — — — — — — Majidae Chionoecetes bairdi 1.5 0.1 0.2 41.2 1.8 10.0 25.3 6.2 4.7 Paguridae 2.0 0.1 0.1 — — — 0.6 0.1 <0.1 Pagurus aleuticus — — — — — — 0.6 0.1 0.2 Pinnotheridae Pinnixa sp. — — — 1.6 <0.1 0.1 8.2 1.5 0.4 Osteichthyes (total) 59.4 8.1 68.8 15.9 0.4 3.7 77.1 38.2 60.6 Ammodytes hexapterus 1.5 0.1 0.1 1.6 <0.1 <0.1 — — — Gadus macrocephalus — — — — — — 1.2 0.1 1.8 Hippoglossoides elassodon 0.5 <0.1 1.3 — — — 4.7 0.7 4.1 Icelus sp. — — — — — — 0.6 0.1 0.3 Lumpenella longirostris — — — 1.6 <0.1 0.1 — — — Lumpenus sp. — — — 1.6 <0.1 0.1 10.0 1.8 1.3 Lumpenus fabricii — — — — — — 0.6 0.1 <0.1 Lumpenus maculatus 0.5 <0.1 <0.1 — — — — — — Lumpenus sagitta — — — 1.6 <0.1 0.1 — — — Lycodes sp. — — — — — — 0.6 0.1 0.6 Lycodes brevipes — — — — — — 1.2 0.1 0.9 Mallotus villosus 28.7 4.6 32.7 4.8 0.1 1.4 29.4 8.9 20.5 Theragra chalcogramma 13.4 .7 24.8 1.6 <0.1 1.2 51.8 17.0 21.9 Trichodon trichodon — — — — — — 5.9 0.7 4.3 Zaprora silenus — — — — — — 0.6 0.1 0.7 Unidentified Osteichthyes 30.7 2.7 9.9 14.3 0.3 0.8 37.1 8.5 4.2 Bivalvia (total) 1.0 0.1 0.1 4.8 0.1 0.1 14.7 3.1 0.1 Clinocardium sp. 0.5 <0.1 <0.1 3.2 <0.1 <0.1 1.8 0.2 <0.1 Macoma sp. — — — — — — 0.6 0.1 <0.1 Yoldia sp. — — — 3.2 <0.1 <0.1 12.4 2.6 <0.1 Unidentified Bivalvia 0.5 <0.1 <0.1 1.6 <0.1 <0.1 1.8 0.2 <0.1 Gastropoda (total) 1.0 0.1 0.1 3.2 0.1 0.1 0.6 0.1 <0.1 Naticidae 1.5 0.1 0.1 1.6 <0.1 <0.1 — — — Neptunidae — — 1.6 <0.1 <0.1 0.6 0.1 <0.1 Trochidae Margarities sp. — — — 1.6 <0.1 <0.1 — — — Unidentified Gastropoda — — — 1.6 <0.1 <0.1 — — — Ascidiacea 0.5 <0.1 0.2 — — — — — — Sipuncula (Phylum) 0.1 <0.1 0.1 1.6 <0.1 0.1 2.4 0.4 0.6 Plant 2.0 0.1 1.4 3.2 <0.1 0.9 4.7 0.6 1.0 Pebbles 2.5 — 0.1 19.0 — 0.4 18.8 — 0.4 605 FISHERY BULLETIN; VOL. 83, NO. 4 Table 3,— The 10 most frequently occurring food items are categorized by 10 cm length (FL) groups of Pacific cod from 30 to 69 cm. Data is presented by percent frequency of occurrence (F), percent by number (A/), and percent by volume (V). 30-39 cm 40-49 cm 50-59 cm 60-69 cm {n = 20) (n = 216) (n = 173) (n = 23) Food items F N V F N V F N V F N V Euphausiacea 78.8 51.3 7.5 54.7 58.3 9.0 48.1 54.8 8.9 27.3 32.7 3.0 Pandalus borealis 36.4 15.1 31.9 61.0 19.0 30.6 69.9 17.7 36.4 81.8 29.7 47.6 Pandalus goniurus 3.0 0.5 0.9 13.8 0.7 1.2 20.2 1.3 1.2 22.8 0.6 0.9 Crangon sp. 5.6 2.4 <0.1 3.5 0.2 0.1 4.0 0.3 0.1 9.0 0.3 0,2 Crangon communis — — — 10.1 0.9 0.4 12.9 1.2 0.3 18.2 1.4 0,5 Eualus macilenta 11.1 0.3 0.3 5.2 0.3 0.1 6.7 0.4 0.1 — — — Chionoecetes bairdi — — — 11.5 1.4 2.5 26.7 3.0 5.1 59.1 5.3 9.0 Lumpenus sp. — — — 4.3 0.9 0.8 3.5 0.5 0.4 4.5 0.2 0.3 Mallotus villosus 14.1 8.7 28.3 19.6 3.5 18.3 24.4 4.7 17.5 22.8 4.7 9.3 Theragra chalcogramma 16.7 9.5 8.5 20.5 4.4 15.2 22,7 6.3 16.7 13.7 6.3 3.9 All sizes of cod examined were feeding on both small and large pink shrimp. Pink shrimp length (CL) distributions measured from cod stomachs and from trawl samples began to overlap at about 12 mm (Fig. 3). Piesults from the Kolmogorov-Smirnov test com- paring pink shrimp length distributions >16.5 mm CL from cod stomachs and trawl samples showed no significant difference {P > 0.10) in August 1980 and May 1981. There was a significant difference {P = 0.009) in the September 1981 sample. No signifi- cant difference between length distributions in- dicates that cod were not feeding on selective sizes of pink shrimp. A significant difference indicates that cod consumed a greater proportion of smaller shrimp than was captured by the trawl. stomach were 24.4 g for May and 11.2 g for September 1981 (Tkble 1). Per day, the average amount of pink shrimp consumed by one cod was 8.1 g and 3.7 g for May and September, respectively. The estimated weight of an average length cod was 1,689.5 g for the May survey and 1,720.1 g for the September survey (W = 0.00000593L3168. Owen and Blackburn 1983). Cod biomass estimates were 1,621 t for May and 591 t September, respectively. Based on the above parameters, estimates of pink shrimp biomass consumed were calculated using the May and September data (Ikble 5). With May data. Pacific cod consumed an estimated 875 t of pink shrimp over the 112-d period, whereas the Table 4.- -Size range and mean size of prey and trawl caught animals. Size data were not collected for snow crab in the trawl. CL = carapace length; FL = fork length; CW = carapace width. Number r neasured Trawl Size range Mean size Prey Prey Trawl Prey Trawl Pink shrimp 1,143 Humpy shrimp 102 Walleye pollock 236 Snow crab 69 7,823 202 2,100 6.5-26.0 mm (CL) 8.0-17.5 mm (CL) 6.0-25.0 cm (FL) 6.0-42.0 mm (CW) 10.0-26.0 mm (CL) 10.5-18.5 mm (CL) 6.0-63.0 cm (FL) 16.2 mm 13.5 mm 10.3 cm 22.2 mm 18.0 mm 13.9 mm 22.8 cm Estimate of Pink Shrimp Biomass Consumed The extent of the Pacific cod predation on pink shrimp in Pavlof Bay was examined by estimating total biomass consumed during a 112-d period from late May through mid-September 1981. Analysis using Jones' (1974) equation indicated that for both 1981 surveys the average amount of food found in a cod stomach was digested in about 3 d. The average weights of undigested pink shrimp found in a cod Table 5. — Calculation of the total pink shrimp biomass consumed by Pacific cod during a 112-d period from late May through mid- September 1981. Two estimates are presented using the May and September data. May 1981 Sept. 1981 Mean weight of pink shrimp consumed daily by one cod (g) 8.1 3.7 112-d consumption by one cod (g) 907.2 414.4 Weight of an average length cod (g) 1,689.5 1,720.1 Proportion of pink shrimp eaten relative to cod weight 0.54 0.24 Cod biomass estimate (t) 1,621 591 Pink shrimp biomass consumed (t) 875 142 606 ALBERS and ANDERSON: DIET OF PACIFIC COD September information suggests that cod consum- ed an estimated 142 t of pink shrimp over the same period. DISCUSSION Cod Diet Pink shrimp was the dominant food item identified by frequency of occurrence (63%) and percent volume (31%). In the Gulf of Alaska near Kodiak Island, Jewett (1978) reported pink shrimp occur- ring in 4% of the Pacific cod examined, and Hunter (1979) found pink shrimp occurring in 24% of the cod, representing 16% of the diet by weight. Hunter also identified pink shrimp as the dominant food item in his study. However, our study exhibited a higher percent frequency of occurrence for that species. Unlike our study, Jewett and Hunter's studies included examinations of cod from offshore areas, which were not regions of high pink shrimp density (Gaffney 19773). The prey size ranges of walleye pollock and snow crab in our study were similar to those found by Hunter (1979). In our study, the maximum length of walleye pollock consumed by cod was 25 cm FL, although the majority were between 6 and 20 cm. Hunter (1979) reported that cod around Kodiak Island were feeding on groundfish (including wall- eye pollock) between 2 and 24 cm. The snow crab consumed by cod in Pavlof Bay ranged from 5 to 45 mm CW, which is similar to the size range of snow crab (1-40 mm) found in cod stomachs by Hunter (1979). Jewett (1978) reported a greater size range of snow crab occurring in cod stomachs (from 1.8 to 70 mm), yet 78% were between 7 and 23 mm. The maximum size of cod examined by Jewett was 92 cm TL (total length) compared with 69 cm FL in our study, and this difference probably accounts for his observation of larger snow crab. We believe that cod were not feeding on selected sizes of shrimp. The size ranges and mean sizes of pink shrimp and humpy shrimp consumed by cod were similar to those found in the trawl (Tkble 4). However, cod did consume small (<10.0 mm CL) pink shrimp that were not captured by the trawl. We believe this is due to trawl bias toward larger shrimp. No significant difference was found between pink shrimp length distributions from cod stomachs and trawl samples in two of the three surveys, indicating that cod were not feeding on selected sizes of shrimp. There was a significant difference (P = 0.009) for the September 1981 survey. In this sample either cod selected slightly for smaller shrimp or the trawl caught slightly larger shrimp. Estimate of Pink Shrimp Biomass Consumed The estimated pink shrimp biomass in Pavlof Bay decreased by 1,501 1 between the May and Septem- ber surveys in 1981. During this period we estimated that Pacific cod consumed between 142 and 875 t of pink shrimp. Since Pavlof Bay is believed to con- tain a geographically isolated stock of pink shrimp (Anderson 1981) and because the bay was closed to shrimp fishing in 1981, cod predation is responsible for at least part of the biomass decline The estimate of pink shrimp biomass consumed over the 112-d period using the May survey data was 733 t more than was estimated using September data. Two of the parameters used to calculate con- sumption estimates were responsible for this dif- ference In May the mean weight of pink shrimp con- sumed daily by one cod was about double the amount in September. Pink shrimp were more abundant and made up a larger percentage of the diet in May than in September. Additionally, cod biomass was estimated to be almost three times higher in May than it was in September causing the consumption estimate to be higher in May (Tkble 5). We belive that consumption of pink shrimp by cod probably lies toward the high end of the calculated range (142-875 t). Biomass estimates were probably conservative for pink shrimp and Pacific cod. Bio- mass was calculated on the assumption that all cod and shrimp were on bottom and all those in the path of the trawl were caught. This is not true for cod or shrimp. For example, Edwards (1968) reported that up to 49% of the gadoids in the path of a trawl avoid capture. Also, an estimate of the catchability of shrimp with the high-opening shrimp trawl was about 56% (Alaska Department of Fish and Game 1982'*). If the cod biomass estimate was conservative, the consumption of pink shrimp by cod would be higher than calculated. Further, if the pink shrimp biomass estimate was conservative, the calculated ^Gaffney, F. G. 1977. Kodiak pandalid shrimp research. Com- mercial Fisheries Research and Development Act, Project No. 5-36-R. Unpubi. manuscr., 76 p. National Marine Fisheries Ser- vice, NOAA, Wash., DC 20235. ■•Westward Region Shellfish Staff, Alaska Department of Fish and Game 1982. Westward Region Shrimp Fishery Management Plan. Unpubi. manuscr., 70 p. Alaska Department of Fish and Game, Kasheruaroff, Mission Road, P.O. Box 686, Kodiak, AK 99615. 607 FISHERY BULLETIN: VOL. 83, NO. 4 15 AUGUST 1980 P e r c e n t F r e q u e n c y 10 — oJ_L 10 ^S Cod n=287 ^^ Trawl n=2685 15 20 Carapace Length(mm) 25 30 15—1 MAY 1881 ^S Cod n=406 ^ Trawl n=2586 r c e n t F r e q u e n c y 10 5 — 15 20 CarapacQ Length(mm) 25 30 608 ALBERS and ANDERSON: DIET OF PACIFIC COD t5-i P e r c e n t F r e q u e n c y ID- S- SEPTEMBER 198 ^ Cod n=440 ^ Trawl n=2552 ^ 10 15 20 Carapace Length(mm) 25 30 Figure 3— Size distribution of pink shrimp from trawl samples and Pacific cod stomachs for August 1980 and May and September 1981. biomass decline of pink shrimp between May and September 1981 would also be greater. Pacific cod are probably feeding on pink shrimp in Pavlof Bay the entire year although the largest concentrations of cod are likely to occur from spring through fall. Trawl survey data from Pavlof Bay in- dicate that cod biomass decreased from 93 t in September 1978 to 20 t in February 1979 and then increased to 371 t in May 1979. Pacific cod are migratory; they move to shallow areas (<90 m) in spring to feed and return to deeper areas (165-247 m) offshore in fall or winter to spawn (Moiseev 1953). The majority of Pavlof Bay is <90 m deep which is not preferred winter habitat. Although we believe that Pacific cod predation has an effect on the present reduced population of pink shrimp, predation probably was not the primary reason for the initial decline of pink shrimp in Pavlof Bay that began in 1977. At that time Pacific cod abundance was just beginning to increase (Fig. 2). Fishing removed about 3,819 t (calculated from ADF&G commercial catch data and NMFS survey data) of pink shrimp between the 1977 and 1978 surveys, which was 30% of the estimated available biomass in 1977. This harvest and the dying out of the strong 1971 year class (Anderson 1981) were probably responsible for most of that initial decrease Cod predation did become a factor, however, once the pink shrimp resource was reduced. This impact on pink shrimp appears substantial despite the reduction of cod in Pavlof Bay (Fig. 2). ADF&G (footnote 4) has reported diminishing pink shrimp stocks in other areas of the western Gulf of Alaska. Some areas that once contained high con- centrations of pink shrimp experienced reductions in abundance at the same time as Pavlof Bay. In most areas, no increase in pink shrimp abundance was observed through 1982, though many areas were closed to fishing. Like Pavlof Bay, these other areas experienced an increase in Pacific cod abundance about the same time as pink shrimp populations were declining. Cod predation may play a role in keeping these reduced pink shrimp stocks from rebuilding to former levels. ACKNOWLEDGMENTS Thanks to Pete Jackson and Dave Jackson, Alaska 609 Department of Fish and Game, for collecting the May 1981 stomach samples. LITERATURE CITED Alverson, D. L., and W. T. Pereyra. 1969. Demersal fish explorations in the northeastern Pacific Ocean - an evaluation of exploratory fishing methods and analytical approaches to stock size and yield forecasts. J. Fish. Res. Board Can. 26:1985-2001. Anderson, P. J. 1981. A technique for estimating growth and total mortality for a population of pink shrimp Pandaliis borealis from the western Gulf of Alaska. In T. Frady (editor), Proceedings of the International Pandalid Shrimp Symposium, February 13-15, 1979, Kodiak, Alaska, p. 331-342. Univ. Alaska, Fair- banks, AK, Sea Grant Program, Sea Grant Rep. 81-3. Edwards, R. L. 1968. Fishery resources of the North Atlantic area, hi D. W. Gilbert (editor), The future of the fishing industry of the United States, p. 52-60. Univ Wash. Publ. Fish., New Sen 4. Hunter, M. A. 1979. Food resource partitioning among demersal fishes in the vicinity of Kodiak Island, Alaska. M.S. Thesis, Univ. Washington, Seattle, 120 p. HURTUBIA, J. 1973. TVophic diversity measurement in sympatric predatory species. Ecology 54:885-890. FISHERY BULLETIN: VOL. 83, NO. 4 Jewett, S. C. 1978. Summer food of the Pacific cod, Gadus macrocephalics, near Kodiak Island, Alaska. Fish. Bull., U.S. 76:700-706. Jones, R. 1974. The rate of elimination of food from the stomachs of haddock, Melanogrammus aeglejinus, cod Gadvs morhua and whiting Merlangivs merlangiis. J. Cons. Int. Explor. Mer 35:225-243. MiNET, J. P., AND J. B. PERODOU. 1978. Predation of cod, Gadus morhua, on capelin, Mallotus villosus, off eastern Newfoundland and in the Gulf of St. Lawrence Int. Comm. Northwest Atl. Fish. Res. Bull. 13: 11-20. MOISEEV, P. A. 1953. Treska i kambaly dalnevostochnykh morei (Cod and flounders of far-eastern seas). [In Russ.] Izv. Tikhookean. Nauchno-Issled Inst. Rybn. Khoz. Okeanogr. 40:1-287. [Engl, transl. by Fish. Res. Board Can., Transl. Ser. No. 119, 576 p.] Owen, D. L., and J. E. Blackburn. 1983. Bottomfish catch and trawl data from an otter trawl survey in northern Shelikof Strait, Chignik area, and Chiniak gully, Alaska, July and August 1981. Alaska Dep. Fish Game, Ifech. Data Rep. 82, 68 p. SOKAL, R. R., and F. J. ROHLF. 1969. Biometry, the principles and practice of statistics in biological research. W H. Freeman and Company, San Fran- cisco, 776 p. Wathne, F. 1977. Performance of trawls used in resource assessment. Mar. Fish. Rev 39(6): 16-23. 610 VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON OFF THE OREGON COAST IN SPRING AND SUMMER MONTHS George W. Boehlert,' Dena M. Gadomski,^ and Bruce C. Mundy^ ABSTRACT Day and night discrete-depth tows were taken off the Oregon coast in spring and summer months to assess the vertical distribution of ichthyoplankton in nearshore waters. Over 1,000 larvae representing 33 taxa of both coastal and offshore ichthyoplankton assemblages were taken; Psettichthys melanostictus was the most abundant coastal species and Lyopsetta exilis the most abundant offshore species. Larvae were generally most abundant at 10-30 m, near the seasonal thermocline in both day and night collec- tions. Larval abundance in July was much higher than in April-May collections. Limited evidence for diel vertical migration suggests that Psettichthys melanostictiis moves to surface waters at night and Gadus macrocephalus moves to deeper water at night. No trends of changes in depth distribution were observed with increasing size Knowledge of the vertical distributions of larval fishes is crucial to full understanding of their biology and to understanding the results of ichthyoplankton surveys (Ahlstrom 1959; Kendall and Naplin 1981). The interaction between vertical distributions and physical processes can have important effects on onshore-offshore distributions of planktonic organisms in upwelling regions such as the coastal northeastern Pacific (Peterson et al. 1979; Parrish etal. 1981; Wroblewski 1982; Rothlisberg et al. 1983). Near-surface distribution, for example, may result in shoreward transport in slicks associated with in- ternal waves (Shanks 1983). In the coastal region off Oregon, the only information on larval fish vertical distribution is a comparison between abundances of Paro'phrys vetultcs and Isopsetta isolepis larvae from neuston and oblique bongo net tows (Laroche and Richardson 1979) and one 24-h study with stratified samples taken by bongo nets without opening-closing devices (Richardson and Pearcy 1977). With the ex- ceptions of the classic study by Ahlstrom (1959) and recent studies by Brewer et al. (1981) and Schlotter- beck and Connally (1982), little else is known about the vertical distribution of larval fishes in north- eastern Pacific coastal waters. In this paper, we pre- sent information on vertical distributions of larval fishes off Oregon. 'College of Oceanography and Marine Science Center, Oregon State University, Newport, OR; present address: Southwest Fisheries Center Honolulu Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 3830, Honolulu, HI 96812. ^College of Oceanography and Marine Science Center, Oregon State University, Newport, OR; present address; Section of Fishes, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, CA 90007. METHODS Six series of samples were collected in 1982, four during daylight (30 April, 14 May, 2 and 13 July) and two during night (2 and 6-7 July). The first two series (30 April, 14 May) were taken at station NHIO, 10 nmi (18.5 km) off Newport, OR, on the Newport hydroline (lat. 44°40'N; Fig. 1). All others were col- lected at NHS (9.2 km offshore). Each sample series consisted of a variable number of tows at discrete depth strata from the surface to within about 4 m of the bottom (Tkble 1). Tows were stepped oblique in five intervals of 3 min each, resulting in a total sampling time of 15 min in each 5 or 10 m depth stratum. The sampler was an opening-closing lUcker trawl (Clarke 1969) with three nets and a double-release mechanism operated by messengers. The nets were 0.505 mm mesh (Nitex^) with a 1 m^ mouth; all tows were at a wire angle of 45° at approximate tow speeds from 0.9 to 1.1 m/s. At this angle, effective mouth area of the net is 0.71 m^. An uncontaminated, discrete depth sample was collected in the second net by lowering the trawl with the first net open, opening the second net for the desired sampling time, and retrieving the trawl with the third net open. Water volumes filtered were estimated with General Oceanics flowmeters mounted in the center of each net. Volumes of water filtered usually ranged be- tween 250 and 450 m^/sampla Temperature and salinity data were collected throughout the water col- umn on each cruise using Niskin bottles to collect ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 611-9l| FISHERY BULLETIN: VOL. 83, NO. 4 124' I I' W 440 45'N 440 30' 440 45' N 44 30 . N 124 II ' • W Figure L— Locations of the sampling conducted during the present study. NH5 and NHIO indicate the stations used. Table 1.— Number of tows per depth stratum with total volumes filtered. (Asterisk indicates that for subsequent analysis the NHIO day sample In the two deepest categories were combined with the 50-60 m stratum.) Sample times were as follows: Day— 4/30, 1018-1547; 5/14, 0808-1600; 7/2, 0608-0907; 7/13, 0554-1048; night— 7/2, 0016-0514, 7/6-7/7, 2330-0410. Day Night NHIO NHS Total volume filtered NH5 Depth filtered 7/6- filtered (m) 4/30 5/14 (m3) 7/2 7/13 (m3) 7/2 7/7 (m3) 0-5 2 2 1,682.7 2 2 1,392.6 2 2 1 ,574.8 5-10 2 2 1,783.7 2 2 1,451.2 2 2 1,575.7 10-20 2 2 1,595.7 2 862.9 2 2 1,292.9 20-30 2 2 1 ,796.7 2 929.3 2 2 1,237.6 30-40 1 2 1,415.8 2 920.5 1 2 1,184.0 40-50 1 2 1,488.1 2 920.9 1 2 1,034.8 50-60 1 2 1,891.2 2 671.3 1 1 808.3 60-70* 1 1 930.0 — — — — — — 70-80* — 1 704.8 — — — — — — 612 BOEHLERT ET AL.: VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON water samples at 5 to 10 m intervals or using a self- contained Applied Microsystems CTD-12. The salinities of water bottle samples were determined in the laboratory using a Guideline 8400 autosali- nometer. The CTD salinities from 13 July were not used because of suspected machine malfunction. The CTD temperatures from 13 July agreed with sur- face bucket temperatures and were used in our study. Plankton samples were preserved at sea in 10% buffered Formalin. Samples were sorted for fish lar- vae in the laboratory using a dissecting microscope. Larvae were identified to the lowest possible taxon, measured (standard length), and stored in 5% buf- fered Formalin. Larval abundances were calculated as number per 1,000 m'* for each tow. The six sampling series were combined into three data sets, spring daytime samples (30 April and 14 May, NHIO), summer day- time (2 and 13 July, NHS), and summer nighttime (2 and 6-7 July NHS). In April and May the salinity profiles closely paralleled each other, with lower salinity at the surface, 30.S-31.0"/„o, than in deeper water, where salinities gradually increased to about 33-34°/oo (Fig. 2a). Water temperatures above 50 m April 30 3iO 33J 34.0 0 5 10 20- 30- ■C 40- o. SO- SO 70- April 30 May 14 Salinity (%o) "~1 1 1 1 1 1 1 1 7.0 e.O 9.0 10.0 1 1.0 12.0 13.0 14.0 15.0 Temperature ('O Q. O 0- 5- 10- 20' 30- 40- 50- 60- 70- July 13 '\ 7 July 2 30^ I 31/) T -r 31.5 32.0 32.5 Salinity (%o) I 33.0 — I 3i5 —I 34.0 o^ 5 10 20 30 ^ 40 Q. 50- 60 70- ~1 1 1 1 1 1 1 1 7.0 8.0 9.0 10.0 II.O I2X) liO 14.0 liO Temperature CO Figure 2.— Salinity and temperature profiles during the collections, a. Salinity profiles during the spring collections, b. Spring temperature profiles, c Summer salinity profiles, d. Summer temperature profiles. 613 FISHERY BULLETIN: VOL. 83, NO. 4 in April and May were about 9°-ll°C (Fig. 2b). No thermocline was present. Thus, hydrographic regimes support grouping April and May samples together. Temperature and salinity profiles in April and May were typical for the winter Oregon hydrographic regime prior to the onset of late spring- early summer upwelling (Huyer 1977). In July the salinity and temperature profiles dif- fered from those in April and May (Figs. 2c, d). Salinities were more uniform in summer than spring throughout the water column, ranging from 32.07oo at the surface to 33.4"/u„ below about 20 m (Fig. 2c). The temperature gradient in July was greater than in April and May due to warmer surface waters (Fig. 2d). Surface temperatures ranged from 12.8° to 14.6°C, decreasing with depth to about 8.2°-9.6°C at 40 m. A thermocline was present at about 10-20 m. Temperature and salinity profiles in July were typical of the summer Oregon hydrographic regime (Huyer 1977). Surface temperatures suggest that samples were not taken during active upwelling. RESULTS In this study, a total volume of 29,145.5 m^ was filtered and 1,007 larvae, representing 33 taxa, were enumerated from 75 discrete depth tows. Larvae were most abundant in summer, with an abundance peak 10-30 m deep during daytime and 20-30 m deep during nighttime (Fig. 3). In spring, larvae were distributed relatively uniformly throughout the water column below 5 m with small abundance peaks at 10-20 and 40-50 m. During daytime in both spring and summer, larvae were least abundant at the sur- face (0-5 m), although abundance at the surface in- creased at night. The depth distribution at night also differed in having a secondary abundance peak near the bottom (50-60 m) and greater overall larval abun- dance than during the day. The larval fish species were categorized as coastal (most abundance 2-28 km from the coast, see Tkbles 2, 3, and 4), or offshore (most abundant 37-111 km from the coast, see Ikbles 5, 6, and 7), according to larval assemblages described by Richardson and Pearcy (1977). Most larvae in this study were of the coastal assemblage because samples were collected at NH5 and NHIO (9.2 and 18.5 km from the coast, respectively). The spawning seasons of the dominant species off Oregon are discussed in Mundy (1984); most of the fall-winter spawning species were not represented in this study. Since many species were not abundant enough to demonstrate trends, only the dominant species will be discussed below. Coastal Assemblage Gadus macrocephalus, Microgadus proximtis, Isopsetta isolepis, and Psettichthys melanostictiis lar- vae were abundant in all three sampling periods (Tkbles 2, 3, 4, Fig. 4). Gadus macrocephalus larvae were most abundant during the day at 20-30 m in both spring and summer, but were very abundant in the deepest stratum (below 50 m) in night samples (Fig. 4). Microgadus proximvs larvae do not show as clear a trend, but were most abundant in deeper water during summer, particularly at nighttime In spring they were distributed throughout the water column. Isopsetta isolepis and P. melanostictus were also most abundant in nighttime samples. More /. isolepis larvae were found at 10-20 m in spring, whereas in summer they were collected throughout the water column, with abundance peaks near the bottom. Psettichthys melanostictus larvae were more abundant in summer than spring samples. During daytime in summer, P. melanostictus were most abundant below 10 m, whereas at nighttime, although found throughtout the water column, they were most abundant in waters shallower than 10 m (Fig. 4). Seasonal abundance changes were observed for 0 -I 5 10 20 -^ 30 E a. O 40 50 4.4 4>--., 3.3,3 • .4 Day, Spring • • Day, Summer •— — • Nig hi, Summer H I — I — I — I — I — I — I — I — I — 0 10 20 30 40 50 60 70 80 90 Mean Abundance (No./IOOOm^) Figure 3.— Overall larval abundance (larvae per 1,000 m^) for all collections. Numbers adjacent to data points indicate the number of samples taken. 614 BOEHLERT ET AL.: VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON Table 2.— Mean abundances (number per 1,000 m^) for coastal larval species from spring (day only) samples. Depth (m) Species 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean Clupeidae Clupea harengus — — 1.76 2.36 1.25 0.75 — 0.87 Osmeridae Undetermined spp. — 1.85 5.02 1.22 1.51 1.16 0.23 1.57 Gadidae Gadus macrocephalus 1.14 0.53 5.69 7.59 2.91 — 0.89 2.68 Microgadus proximus 1.13 2.04 3.87 2.97 0.73 2.95 0.45 2.02 Cottidae Artedius fenestralis — — _____ _ Artedius harringtonl — — — — _ 0.75 0.41 0.17 Artedius meanyi — — — — — — 0.44 0.06 Clinocottus embryum — — — — — — — — Cottus asper — 0.53 _____ o.08 Radulinus asprellus — — — — — _ 2.60 0.37 Agonidae Odontopyxis trispinosa — — — — — _ 0.23 0.03 Cyclopteridae spp. 1 _ _ _ o.74 2.29 3.64 0.45 1.02 Undetermined spp. — — — 0.47 _ _ — 0.07 Bathymasteridae Ronquilus jordani — 0.59 — — — — — 0.08 Ptilichthyidae Ptilichthys goodei — — 0.94 — — — — 0.13 Bothidae Citharichthys stigmaeus _ _ _ _ o.73 — 0.87 0.23 Pleuronectidae Isopsetta isolepis 0.49 1.79 7.78 2.86 — 1.29 0.66 2.12 Lepidopsetta bilineata — 0.59 0.94 _ _ _ _ o.22 Parophrys vetulus — 0.59 1.76 — — — 0.45 0.40 Psettichthys melanostictus 1.48 5.46 5.58 — 0.73 — 0.44 1.96 Table 3. — Mean abundances (number per 1,000 m^) for coastal larval species from summer day samples. Depth (m) Species 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean Clupeidae Clupea harengus — — _____ _ Osmeridae Undetermined spp. — — _ _ _ _ 3 59 0.53 Gadidae Gadus macrocephalus — — 1.38 12.01 1.08 1.09 — 2.22 Microgadus proximus — — 2.01 1.01 — 3.28 3.49 1.40 Cottidae Artedius fenestralis — — 1.00 1.15 — — — 0.31 Artedius harringtonl — 0.68 4.54 3.30 2.18 2.19 3.39 2.33 Artedius meanyi — — 3.38 2.02 1.10 — — 0.93 Clinocottus embryum — — — — — — — — Cottus asper — 0.53 _____ 0.O8 Radulinus asprellus _______ _ Agonidae Odontopyxis trispinosa — — — 1.01 — — — 0.14 Cyclopteridae spp. 1 — — — 2.22 2.16 — — 0.63 Undetermined spp. — — 2.16 4.18 1,08 2.18 6.05 2.24 Bathymasteridae Ronquilus jordani 0.63 0.70 2.01 1.01— — — 0.62 Ptilichthyidae Ptilichthys goodei — — — — — — — — Bothidae Citharichthys stigmaeus — — 1.00 — — — — 0.14 Pleuronectidae Isopsetta isolepis — — 2.07 1 .01 2.20 — 4.82 1 .44 Lepidopsetta bilineata — ______ _ Parophrys vetulus — — — — — — — _ Psettichthys melanostictus 0.98 2.03 22.98 7.48 3.24 3.26 11.18 7.31 615 FISHERY BULLETIN: VOL. 83, NO. 4 Table 4.— Mean abundances (number per 1,000 m^) for coastal larval species from summer night samples. Species Clupeidae Clupea harengus Osmeridae Undetermined spp. Gadidae Gadus macrocephalus MIcrogadus proximus Cottidae Artedius fenestralis Artedius harringtoni Artedius meanyi Clinocottus embryum Cottus asper Radulinus asprellus Agonidae Odontopyxis trispinosa Cyclopteridae spp. 1 Undetermined spp. Bathymasteridae Ronquilus jordani Ptilichthyidae Ptilichthys goodei Botfiidae Citharichthys stigmaeus Pleuronectidae Isopsetta isolepis Lepidopsetta bilineata Parophrys vetulus Psettichthys melanostictus Depth (m) 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean — 4.42 15.15 4.56 — 0.75 1.12 3.71 — — — 0.82 3.16 8.11 22.98 5.01 0.66 0.61 — 4.66 1.95 3.29 6.91 2.58 3.20 — 0.78 1.57 — 0.75 2.60 1.27 1.35 4.41 5.44 11.30 0.73 1.27 1.40 3.70 0.66 4.48 0.71 4.12 0.74 1.01 3.62 2.19 0.66 — — ____ 0.09 — — — — — — 2.50 0.36 — — 0,77 0.79 — — — 0.22 — — — 3.96 0.73 — 4.74 1.35 — — 0.71 1.70 — 1.27 — 0.53 — — — 0.74 0.74 — — 0.21 — 2.57 1.63 6.50 2.64 10.90 1.12 3.62 0.66 — 0.71 — _ _ _ 0.20 24.63 31.37 13.65 18.86 7.78 6.07 6.91 15.61 to -P 20 -I 50 - 40 -1 SO ' Spring- Ooy Summer - Doy Summer- Night 60 0 S 10 f 30 a. e 40 O 50 - 60 Godm moerocepholm 20 I > 10 20 p Pselllchlhvi melano3ticlu> I I 20 40 20 40 0 S - 10 20 - 30 - 40 50 - 60 I Lyopselto eiilis I 1 20 40 20 40 20 Abundonce (No./IOOOm^) 40 Figure 4.— Vertical abundance patterns of the three most abun- dant taxa {Gadus niacrocephaltus, Psettichthys melanostictus, and Lyopsetta exilis) during the three sampling periods. 616 Artedius and cyclopterid species. Artedius fenestralis, A. harringtoni, and A. meanyi were taken almost exclusively in the summer sampling period. All species were most abundant in nighttime samples. During the day, A. harringtoni larvae were distributed relatively uniformly throughout the water column from about 10 to 60 m, whereas at night most larvae were taken from the 5-30 m depth strata. Most cyclopterid larvae collected in spring were the larval type referred to as Cyclopteridae spp. 1 by Richardson and Pearcy (1977). Other cyclopterid species were more abundant in summer. Most cyclopterid larvae were collected below 20 m during both night and day, and in summer were abundant at the deepest sampling depths. Osmerid larvae of undetermined species had a unique distribution pattern. They were abundant during spring, rare during summer daytime samples, and most abundant during summer nighttime samples. During both spring and summer, larvae were most abundant at the 10-20 m depth stratum. Offshore Assemblage Spring to summer differences in abundance pat- terns were more distinct for offshore larval species as compared with coastal species (Tkbles 5, 6, 7). BOEHLERT ET AL.: VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON Table 5.— Mean abundances (number per 1,000 m^) for offshore larval species from spring (day only) samples. Depth (m) Species 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean Offshore Engraulidae Engraulis mordax Bathylagidae Bathylagus ochotensis Bathylagus pacificus Myctophidae Protomyctophum crockeri Protomyctophum thompsoni Stenobrachius leucopsarus Bythitldae Brosmophycis marginata Scorpaenidae Sebastes spp. Bothidae Citharichthys sordidus Pleuronectidae Glyptocephalus zachirus Lyopsetta exilis Coastal-Offshore Pleuronectidae Hippoglossoides elassodon — — 0.56 — 0.78 7.28 0.87 1.36 — — — — — — 0.22 0.03 — — — — — 0.75 — 0.11 — — — — — 1.74 1.19 0.42 0.49 6.79 2.76 0.47 — — — 1.50 — 2.73 1.06 1.49 — — — 0.75 — — 1.06 — — — — — — — 11.27 14.9 1.49 — — 8.56 0.15 4.96 0.21 Table 6.— Mean abundances (number per 1,000 m^) for offshore larval species from summer day samples. Depth (m) Species 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean Offshore Engraulidae Engraulis mordax Bathylagidae Bathylagus ochotensis Bathylagus pacificus Myctophidae Protomyctophum crockeri Protomyctophum thompsoni Stenobrachius leucopsarus Bythitidae Brosmophycis marginata Scorpaenidae Sebastes spp. Bothidae Citharichthys sordidus Pleuronectidae Glyptocephalus zachirus Lyopsetta exilis Coastal-Offshore Pleuronectidae Hippoglossoides elassodon 0.98 2.13 — — — — — 0.44 1.49 0.73 _____ 0.32 — 0.65 3.03 1.08 1.10 — — 0.84 — 0.73 _____ 0.10 — 2.11 19.55 4.52 _ _ _ 3.74 — — 10.03 11.53 4.38 2.13 — 4.01 1.10 — 0.16 Only Sebastes and Lyopsetta exilis larvae were abun- dant in all three sampling periods. Sebastes larvae were most abundant in nighttime samples, when they were mainly collected in shallow water (0-20 m). In both spring and summer daytime samples they were also in relatively shallow waters (5-40 m), although they were not abundant at the shallowest stratum (0-5 m). During day, L. exilis larvae were distributed in deeper water, particularly in spring, when all larvae were collected below 30 m (Fig. 3). At night, most L. exilis were shallower, between 5 and 30 m. Two species, Bathylagus ochotensis and Steno- brachius leucopsarus, were collected only in spring samples. These two species were predominantly col- lected at different depths with B. ochotensis found 617 FISHERY BULLETIN: VOL. 83, NO. 4 Table 7.- -Mean abundances (number per 1,000 m^) samples for offshore larval spec les from summer night Species Depth (m) 0-5 5-10 10-20 20-30 30-40 40-50 >50 Mean Offshore Engraulidae Engraulis mordax Bathylagidae Bathylagus ochotensis Bathylagus pacificus Myctophidae Protomyctophum crockeri Protomyctophum thompsoni Stenobrachius leucopsarus Bythitidae Brosmophycis marginata Scorpaenidae Sebastes spp. Bothidae Citharichthys sordidus Pleuronectidae Glyptocephalus zachirus Lyopsetta exilis Coastal-Offshore Pleuronectidae Hippoglossoides elassodon 4.09 3.76 1.12 3.86 0.68 2.38 0.82 — — 1.12 1.27 — — 0.78 — 0.74 1.27 10.27 1.87 0.66 0.68 0.77 0.91 _ — _ 0.43 — 8.28 15.05 32.99 — 3.53 1.12 8.71 — — 0.71 — — 1.12 0.26 in deeper water (40-50 m) and S. leucopsarus in shallow water (5-20 m). Two species of larval flatfishes and Engraulis mor- dax were collected only in summer samples. Glypto- cephalus zachirus were most abundant during day at 5-30 m, and Citharichthys sordidus at night below 50 m. Engraulis mordax larvae were collected only above 10 m. Engraulis mordax were most abundant at night when more than half were in very shallow waters, <5 m. During the day, more E. mordax were found at 5-10 m than at 0-5 m. A relationship between larval size and depth was not evident for any species. Because of the low abun- dances of larvae, however, this relationship could not be adequately considered for most species. A change in larval size with season was demonstrated for the most abundant species (Tkble 8), with mean larval standard lengths of all species greater in summer than in spring samples. There were no obvious dif- ferences between the size of larvae caught in day and night summer samples. DISCUSSION Peak abundances of all taxa combined occurred at 10-30 m on all sample dates (Fig. 3) and character- ized several individual taxa during the day, including Clupea harengus, Osmeridae, Gadus macrocephalus, Sebastes spp., and Parophrys vetulus, as well as Lyopsetta exilis and Psettichthys mslanostictus in the summer. The 10-30 m depth range bracketed the lower boundary of the seasonal thermocline in July, although no thermocline was present in April-May (Fig. 2). This trend for the peak abundance of fish larvae to be centered near the thermocline is similar to that found in other regions (Ahlstrom 1959; MOler et al. 1963; Kendall and Naplin 1981). The trend for most larvae to be found in midwater was similar to that described by Brewer et al. (1981) for their deepest stations off southern California. We did not find large concentrations of larvae near the bottom as they did, except at night, when gadids, cot- tids, cyclopterids, and pleuronectids were abundant. Our sampling gear was ineffective just above the bot- tom as compared with the roller-equipped gear used by Brewer et al. (1981). Richardson and Pearcy (1977) found larvae to be most abundant at 0-10 m and least abundant at 51-100 m during late May off Oregon. We found lar- vae to be most abundant at 10-30 m. This difference may be due to differences in hydrography and sta- tion locations. Their station was 18 km offshore, closer to the shelf break where the depth of water was over 150 m deep. The faunal composition in each study was also different. Richardson and Pearcy (1977) captured more specimens of several surface- associated taxa than we did, including large Clupea harengus, Stenobrachius leucopsarus, Ronquilus jor- dani, and Ammodytes hexapterus. We captured higher densities of deeper dwelling taxa, including gadids and cottids. Several taxa taken in both studies had different distributions in each, including Sebastes 618 BOEHLERT ET AL: VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON Table 8. — Ranges and mean standard lengths (mm) for dominant fish larvae. N = number of larvae; Min. = minimum; Max. = maximum. Spring: Day Summer: Day Summer: Night Species N Min. Mean Max. N Min. Mean Max. N Min. Mean Max. Coastal Clupea harengus 10 7 8.2 9 0 — — — 0 — — — Osmerid 17 3 7.1 17 3 29 33.1 38 35 23 31.8 40 Gadus macrocephalus 29 8 10.8 20 17 9 13.6 17 28 9 13.3 19 Microgadus proximus 22 6 10.9 19 9 11 21.3 36 18 6 18.4 33 Artedius fenestralis 0 — — — 2 11 12.0 13 8 10 11.6 13 Artedius harringtoni 2 6 6.5 7 15 5 8.2 13 33 5 8.6 14 Artedius meanyl 1 — 4.9 — 6 7 9.9 16 19 7 9.4 18 Radulinus asprellus 10 6 7.6 9 0 — — — 0 — — — Liparid type 1 12 4 5.7 6 4 3 3.1 3 2 3 3.5 4 Liparid unknown 1 — 10.2 — 14 5 16.7 25 10 15 20.4 23 Ronquilus jordani 1 — 6.6 — 5 22 27.1 33 5 24 24.1 25 Isopsetta isolepis 22 4 8.9 21 9 10 20.0 23 28 7 16.1 23 Psettichthys melanostictus — 3 7.0 22 41 9 20.0 27 150 8 20.8 28 Offshore Engraulis mordax 0 — — — 4 12 13.5 15 12 7 13.1 16 Bathylagus ochotensis 16 5 8.1 19 0 — — — 0 — — — Stenobrachius leucopsarus 19 3 5.2 9 0 — — — 0 — — — Sebastes spp. 8 4 4.7 6 6 4 14.6 18 12 12 16.0 18 Citharichthys sordidus 0 — — — 1 — 20.5 — 11 17 37.8 40 Glyptocephalus zachirus 2 13 25 37 22 14 29.9 39 4 22 34.5 50 Lyopsetta exilis 71 4 8.4 15 27 11 13.2 16 79 7 12.2 20 spp., Cyclopteridae spp. 1, and Isopsetta isolepis. These differences indicate the need for more exten- sive sampling before the variability of vertical distributions off Oregon can be understood, par- ticularly as they relate to hydrographic condi- tions. We found Engraulis mordax larvae entirely at 0-10 m. Brewer et al. (1981) found greater concentrations of Engraulis below 10 m, while Ahlstrom (1959) found Engraulis to be concentrated in the upper 23 m with some specimens occurring to 105 m. Off Oregon, Engraulis larvae are found concentrated at 0-20 m (Richardson 1973), in association with the Columbia River plume, a lens of warm, low salinity water usually 20-40 m deep (Richardson 1980). Our limited data suggest that Engraulis mordax larvae occur at depths that would place them within the plume, rather than beneath it or at its boundary. The vertical distribution suggests restriction to the warmest part of the water column (Fig. 2d); north- ern anchovies rarely spawn in waters with surface temperatures below 14°C (Lasker et al. 1981). The seasonal differences in species composition be- tween the April-May and July samples were those that would be expected in samples from winter and summer hydrographic regimes, except that Artedius fenestralis and A. meanyi have been taken in April and May of other years (Mundy 1984). The presence of Clupea harengus, Radulinus asprellus, myctophid, and bathylagid larvae only in April-May, during a winter hydrographic regime, is expected from previous studies (Richardson and Pearcy 1977; Mundy 1984). Studies of day/night differences in the distribution of fish larvae are confounded by daytime avoidance of nets by larvae (Ahlstrom 1959). Daytime avoidance of nets is suggested in our study by the greater numbers of larvae taken during the night than day at all but two depth strata. The lack of length dif- ferences between larvae caught in day and night, however, and the fact that no taxa were taken only in night samples during July suggests that diurnal net avoidance was not related to taxon or size The same comparisons with 70 cm bongo net samples (Richardson and Pearcy 1977) suggest that diurnal avoidance by large larvae was greater for bongo nets than for the Tlicker trawl. Evidence for vertical migration exists for several species in this study (Tkbles 3, 4, 6, 7). Psettichthys melanostictus abundance in surface waters (0-10 m) increased greatly at night (Fig. 4). Engraulis mor- dax were most abundant at 5-10 m than 0-5 m dur- ing the day, but more evenly distributed at night. This could be due either to vertical migration or net avoidance in the shallowest stratum during the day. Ahlstrom (1959), however, presented evidence for negative phototaxis by anchovy larvae, and Hunter and Sanchez (1976) demonstrated nighttime migra- tion to the surface in larvae larger than 10 mm SL. Thus larvae migrate upwards at night, but are con- 619 strained to shallower water in the day as compared with the southern subpopulation. The clearest case of vertical migration was that of Gadus macrocephalus (Ihbles 3, 4), which was most abundant at 20-30 m in the day and deeper than 50 m at night. The migration of this species was primarily responsible for the increased total abun- dance of larvae near the bottom at night (Fig. 3). This pattern of movement is similar to that observed for larval Ammodytes personatics by Yamashita et al. (1985), who suggested that this reverse vertical migration allowed feeding in daytime and avoidance of migrating predators at night. This nocturnal de- scent, not previously reported for Gadus larvae, should be confirmed with further sampling. Gadus morhua larvae 3.8-4.9 mm long move from deeper water in the day to 0-2 m at night, and descend in the water column with growth (Hardy 1978). Lar- vae of another gadid, Melanogrammus aeglefinu^, are most common in the thermocline and their depth of greatest abundance fluctuates as the thermocline depth changes with rotary tidal currents, causing oc- casional descent in the water column at night (Miller et al. 1963). Offshore taxa in Oregon coastal waters should oc- cur in greatest numbers during onshore surface water transport during winter and early spring. This was true in our study for the mesopelagic Myc- tophidae and Bathylagidae, but not for other offshore assemblage taxa (Tkbles 5, 6, 7). Almost all of the bathylagid and myctophid larvae except Steno- brachiv^ leucopsaru^ were found below 30 m. Ahlstrom's (1959) work confirms these general distributions; he found most larvae of the genera taken in this study (Electrona = Protomyctophum; Laynpanyctus = Stenobrachiu^) at depths >56 m, beneath the thermocline, except Stenobrackius. He found Stenobrachius to have the shallowest distribu- tion of all myctophid larvae in his study (0-41 m). Richardson and Pearcy (1977) also found Steno- brachius larvae to be in shallow waters (0-50 m) with many at 0-10 m during the day. The distribution of larval mesopelagic fishes, or other offshore taxa, can- not be related to the depth of onshore transport because virtually nothing is known about the depth of winter onshore transport off Oregon (Peterson et al. 1979; Huyer*). Both deep and surface dwelling larvae of mesopelagic fishes collected in our study appear to be transported onshore, however, sug- gesting that transport occurs over a broad depth range off Oregon. *A. Huyer, Associate Professor, Oregon State University, Cor- vallis, OR 97331, pars, commun. 29 September 1983. FISHERY BULLETIN: VOL. 83, NO. 4 ACKNOWLEDGMENTS This research was supported by NOAA Office of Sea Grant, Department of Commerce, under Grant No. NA81-D-00086. We thank M. Yoklavich, J. Shenker, and the crew of the RV Sacajawea for assistance in sampling. We also thank W. G. Pearcy and H. G. Moser for reviewing the manuscript. LITERATURE CITED Ahlstrom, E. H. 1959. Vertical distribution of pelagic fish eggs and larvae off California and Baja California. U.S. Fish Wildl. Serv., Fish. Bull. 60:107-146. Brewer, G. D., R. J. Lavenberg, and G. E. McGowan. 1981. Abundance and vertical distribution of fish eggs and larvae in the southern California bight: June and October 1978. Rapp. R-v. Reun. Cons. int. Explor. Mer 178:165-167. Clarke, M. R. 1969. A new midwater trawl for sampling discrete depth horizons. J. Mar. Biol. Assoc U.K. 49:945-960. Hardy, J. D., Jr. 1978. Family Gadidae: Gadus morhua Linnaeus, Atlantic cod. In Development of fishes of the Mid-Atlantic Bight, an atlas of egg, larval -i- juvenOe stages. Vol. II, p. 236-259. U.S. Dep. Inter., Fish Wildl. Serv. Hunter, J. R., and C. Sanchez. 1976. Diel changes in swim bladder inflation of the larvae of the northern anchovy, Engraulis mordax. Fish. Bull., U.S. 74:847-855. Huyer, A. 1977. Seasonal variation in temperature, salinity, and densi- ty over the continental shelf off Oregon. Limnol. Oceanogr. 22:442-453. Kendall, A. W., Jr., and N. A. Naplin. 1981. Diel-depth distribution of summer ichthyoplankton in the Middle Atlantic Bight. Fish. Bull., U.S. 79:705-726. Laroche, J. L., AND S. L. Richardson. 1979. Winter-spring abundance of larval English sole, Parophrys vetulus, between the Columbia River and Cape Blanco, Oregon during 1972-1975 with notes on occurrences of three other pleuronectids. Estuarine Coastal Mar. Sci. 8:455-476. Lasker, R., J. PelXez, and R. M. Laurs. 1981. The use of satellite infrared imagery for describing ocean processes in relation to spawning of the northern an- chovy Engraulis mordax. Remote Sens. Environ. 11:439- 453. Miller, D., J. B. Colton, Jr., and R. R. Marak. 1963. A study of the vertical distribution of larval haddock. J. Cons. Perm. Int. Explor. Mer 28:37-49. Mundy, B. C. 1984. Yearly variation in the abundance and distribution of fish larvae in the coastal upwelling zone off Yaquina Head, Oregon from June 1969 to August 1972. MS Thesis, Oregon State Univ., Corvallis, 158 p. Parrish, R. H., C. S. Nelson, and A. Bakun. 1981. Transport mechanisms and reproductive success of fishes in the California current. Biol. Oceanogr. 1:175- 203. Peterson, W. T., C. B. Miller, and A. Hutchinson. 1979. Zonation and maintenance of copepod populations in the Oregon upwelling zone Deep-Sea Res. 26A:467-494. 620 BOEHLERT ET AL.: VERTICAL DISTRIBUTION OF ICHTHYOPLANKTON Richardson, S. L. 1973. Abundance and distribution of larval fishes in waters off Oregon, May-October 1969, with special emphasis on the northern anchovy, Engraulis mordax. Fish. Bull., U.S. 71: 697-711. 1980. Spawning biomass and early life of northern anchovy, Engraulis mordax, in the northern subpopulation off Oregon and Washington. Fish. Bull., U.S. 78:855-876. Richardson, S. L., and W. G. Pearcy. 1977. Coastal and oceanic fish larvae in an area of upwelling off Yaquina Bay, Oregon. Fish. Bull., U.S. 75:125-145. ROTHLISBERG, P. C, J. A. CHURCH, AND A. M. G. FORBES. 1983. Modelling the advection of vertically migrating shrimp larvaa J. Mar. Res. 41:511-538. SCHLOTTERBECK, R. E., AND D. W. CONNALLY. 1982. Vertical stratification of three nearshore southern California larval fishes Engraulis mordax, Genyonemus lineatus, and Seriphus politus. Fish. Bull., U.S. 80:895- 902. Shanks, A. L. 1983. Surface slicks associated with tidally forced internal waves may transport pelagic larvae of benthic invertebrates and fishes shoreward. Mar. Ecol. Prog. Ser. 13:311-315. Wroblewski, J. S. 1982. Interaction of currents and vertical migration in main- taining Calanus marshallae in the Oregon upwelling zone— a simulation. Deep-Sea Res. 29A:665-686. Yamashita, Y., D. Kitagawa, and T. Aoyama. 1985. Diel vertical migration and feeding rhythm of the lar- vae of the Japanese sand-eel, Ammodytes personatus. Bull. Jpn. Sot Sci. Fish. 51:1-5. 621 DOLPHIN HABITATS IN THE EASTERN TROPICAL PACIFIC David W. K. Au and Wayne L. Perryman^ ABSTRACT Research-ship surveys by the Southwest Fisheries Center provided information on the distributions of spotted, spinner, striped, and common dolphins in the eastern tropical Pacific The main surveys were conducted from January to March during 1976, 1977, 1979, and 1980. Two ships were used per survey, and together they overlapped most areas in the eastern Pacific where dolphins and yellowfin tuna are jointly fished by purse seiners. The spatial distribution of sightings and of sighting rate of these species show a complementarity to their patterns, although there is a broad overlap. Spotted and spinner dolphins occurred primarily in tropical waters north of the Equator, but also in the seasonal tropical waters south of the Galapagos Islands. These dolphins were relatively infrequent along the Equator, off Costa Rica, and northern South America. Com- mon and striped dolphins tended to be more frequent in these same areas of less frequent spotted and spinner dolphins. The differences in habitats of these two species pairs can be described in oceanographic terms. Spotted and spinner dolphins are primarily in Tropical Surface Water, centered off southern Mexico and extending westward along lat. 10°N, where thermocline "ridging" and relatively small annual variations in surface temperature are features. Common and striped dolphins appear to perfer equatorial and subtropical waters with relatively large seasonal changes in surface temperature and thermocline depth and with seasonal upwelling. The species composition of various areas in the eastern tropical Pacific supports the contention of two major communities. South of where spotted and spinner dolphin schools predominate (along with Risso's, bottlenose, and rough-toothed dolphins), striped and common dolphins and also pilot whales become in- creasingly important. Observations along the Equator also suggest a fauna different from that of the Tropical Surface Water that is most characterized by spotted and spinner dolphins. A trophic basis to these faunal differences is suggested by the interactions with fish and birds. Assum- ing the birds indicate co-occurring tuna, only the spotted and spinner dolphins are commonly found with fish. The distribution of these dolphins as they co-occur with bird flocks and tuna indicates that this inter- specific association is confined primarily to the TVopical Surface Water and is a characteristic feature of its epipelagic community. The eastern tropical Pacific Ocean supports produc- tive tuna fisheries as well as an abundant and diverse cetacean fauna. Ibna fishermen there take advan- tage of the fact that tuna and dolphins frequently swim together. In the "porpoise-tuna" fishery for yellowfin tuna, Thunnus albacares, spotted and spin- ner dolphins, Stenella attenuata and S. longirostris, are temporarily caught by purse seiners in order to take the associated tuna. Striped and common dolphins, S. coeruleoalba and Delphinus delphis, are caught to a lesser extent for the same reason. These dolphins suffer incidental mortality in the fishery and, because of the resulting concern, the Southwest Fisheries Center has been studying their populations to better advise on their management (Smith 1983). Learning about their habitats is one aspect of these studies. .<■ 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, 8604 La Jolla Shores Drive, La Jolla, CA 92038. Perrin (1975a), using the information collected mainly aboard tuna seiners, first showed that the geographic distributions of spotted, spinner, and striped dolphins in the eastern Pacific are extensive, stretching westward from the American coasts past long. 145°W at about lat. 10°N and also dipping south and southwest of the Galapagos Islands. Evans (1975) showed that the common dolphin occurs off- shore of Central America to about long. 112°W and also along the Equator, westward past the Galapagos Islands. Recent summaries of available information (Au et al. 19792; gcott 1981; Perrin et al. 1983) have shown that the distributions of these dolphins are even more extensive than originally perceived. In- deed the species have been reported from localities across the entire Pacific (Alverson 1981). Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83. NO. 4, 1985. 2Au, D. W. K., W. L. Perryman, and W. F Perrin. 1979. Dolphin distribution and the relationship to environmental features in the eastern tropical Pacific Admin. Rep. LJ-79-43, 59 p. Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 623 FISHERY BULLETIN: VOL. 83, NO. 4 The purpose of this paper is to describe the January-March distributions of schools of spotted, spinner, striped, and common dolphins as determin- ed from research ship surveys. Being independent and distinct from data collected by observers aboard tuna seiners, these survey data enable a separate evaluation of habitats and ecological relationships. The distributions derived will be discussed in terms of habitat features and interspecific associations in- cluding those of other cetaceans, yellovvrfin tima, and certain seabirds. We propose that tv^o major ceta- cean communities can be recognized, centered broadly about the tropical and about the equatorial- subtropical surface water provinces (Fig. 1). METHODS The Southwest Fisheries Center conducted or par- ticipated in 17 research cruises studying cetaceans between 1976 and 1981. Eight major cruises were carried out between January and March of 1976, 1977, 1979, and 1980, with the NOAA ships David Starr Jordan and Tovmsend Cromwell. During these surveys, schools of all cetacean species encountered (at least 23 species in the areas of interest) were ap- proached to allow close observation. The combined cruise tracks of these January-March surveys form- ed an extensive coverage of the eastern tropical Pacific and included areas not frequently searched by fishing vessels (Fig. 2). The latter are the equatorial waters and areas both south of lat. 8°N and west of long. 110°W. The cruise tracks during any particular year were chosen to investigate cer- tain aspects of dolphin distribution, ag., the inten- sive surveys during 1979 off southern Mexico and central America constituted a joint aircraft-ship survey of the nearshore habitat. The remaining other nine cruises were either for special studies or of ships of opportunity. In the latter case, the ships did not usually divert course to inspect cetacean schools that were sighted. Cetacean observations from a research ship were conducted by scanning the waters ahead and to the sides of the ship through 20 x 120 mm USN MK-3 Figure L— Surface water masses and boundaries of the tropical ocean in the eastern Pacific, based on Wyrtki's (1964a, 1967) analysis to long. 140°W. Boundaries are a function of the 24°C surface isotherm and the TC/IO m thermocline gradient. Notice that in October the southern boundary to the eastern tropical ocean lies mostly north of the Equator, while in March this boundary lies far to the south so that southern subtropical waters become tropical. 624 AU and PERRYMAN: DOLPHIN HABITATS or 25 X 150 mm Fuji binoculars, mounted both port and starboard either on or above the flying bridge The sighting distance to the horizon was about 7.0 and 5.5 nmi for the David Starr Jordan and Town- send Cromwell respectively. Both ships cruised at about 10 kn (18.5 km/h). The cetacean search-day generally started at 0600 and ended 1800, with two teams, or watches, alternating. In 1976 and 1977 the watches were each 2 or 3 h long, depending on the ship. In 1979 and 1980 they were all standardized to 3 h. There were two experienced persons on each watch, who alternated each hour between the port and starboard binoculars. A third person sometimes helped in record keeping, obtaining supplementary data, and temporarily relieving the other watch members. The peripheral data collected included bathythermograms, surface temperature and salin- ity, weather observations, sighting condition measurements, sighting effort, and observations on fish and birds. When a cetacean school was sighted, its angle relative to the ship's heading was measured and its distance from the ship estimated. The ship then ap- proached the school for closer observation. School size was estimated and the species identified. The reader is cautioned that the distributions to be described are based upon relatively few sightings of schools (Tkble 1) and that they pertain to the January-March season specifically. Though the pat- tern of sampling was widespread, the actual area surveyed was a very small fraction of the huge area Table 1.— Summary of school sizes for spotted, spin- ner, striped, and common doipfiins from the January- March research cruises of 1976, 1977, 1979, and 1980. Arithmetic Geom( X 3tric Species n X s S1 Spotted2 157 148.52 300.93 65.55 3.56 Spotted and spinner 79 357.42 444.50 211.67 2.93 Spinner^ 44 228.39 291.54 107.05 3.88 Striped 187 60.84 69.78 40.74 2.56 Common 98 261.16 484.64 108.47 3.72 Totai^ 565 'Standard deviation factor for the geometric mean. zjotal spotted dolphin schools is 157 + 79 = 236. ^Total spinner dolphin schools is 44 + 79 = 123. "Additionally, there were 838 school sightings of unidentified and other species of dolphins (Delphinidae) and 543 school sightings of whales (Balaenopteridae, Ziphiidae, and Physeteridae). 170 30' 70" 30" Hawaiian Islands JANUARY-MARCH R/V CRUISE TRACKS FOR 1976, 1977, 1979, 1980 Christinas Island * 30' 170 J Figure 2.— Cruise tracks of the January-March cruises by RV David Starr Jordan and RV Townsend Cromwell. These cruises took place in 1976, 1977, 1979, and 1980. 625 FISHERY BULLETIN: VOL. 83, NO. 4 of the eastern Pacific The patterns of species distribution we describe will not always be obvious. Nevertheless there is evidence from other distribu- tion studies, using data collected from the fishery, that these patterns are real (Perrin et al. 1983). The results described in this paper are based upon our personal experiences at sea where the procedure of investigating all cetacean schools and continuously monitoring the physical and biotic environment enabled the development of an ecological under- standing of these dolphins. Continuity in these studies was provided by the fact that on each ship there was at least one experienced observer who par- ticipated during all cruise years. We present our results as an independent, research-ship based, assessment of species distributions and habitat areas. RESULTS Distribution and Relative Abundance of Dolphin Schools Plots of the sighting localities of schools of spot- ted, spinner, striped, and common dolphins, obtain- ed during the January-March research cruises, show the geographical distributions of these species dur- ing the northern winter season. Table 1 summarizes the numbers and sizes of the schools which were identified and studied. Geographic locations referred to in the text can be found in Figure 1. Spotted and spinner dolphins are often in mixed schools (33.5% of spotted dolphins schools also con- tained spinner dolphins. Table 1), and so the sight- ings of spotted or of mixed spotted plus spinner dolphin schools are shown together (Fig. 3). Most of the mixed schools were encountered off southern Mexico, where the eastern form of spinner dolphin (Perrin 1975a, b; Perrin et al. 1977) usually accom- panied spotted dolphins (cf. Figs. 3, 4). Mixed schools were uncommon along the Equator, as was the spot- ted dolphin itself, especially west of long. 110°W. Large schools (>300 individuals) of spotted dolphin were widely scattered, but tended to be more com- mon off southern Mexico. Westward extensions of distribution appeared to occur as three main lobes: about lat. 10°n, between lat. 0° and 5°N, and be- tween lat. 2°S and 5°S. 170° 30° 140° — r 130° — r- 120° — r 70° 30° Hawaiian Islands \ O o° Christmas Island * o o >4 . i^ A. Marquasas Islands O SPOTTED DOLPHIN • SPOTTED a SPINNER DOLPHIN o • o °0° Figure 3.— Distribution of spotted and spotted plus spinner dolphin schools seen during the January-March research cruises. Larger circles indicate schools of >300 animals. 626 AU and FERRYMAN: DOLPHIN HABITATS The spinner dolphin appears to be distributed like the spotted dolphin, with westerly extending lobes of distribution in similar latitudes (Fig. 4). Schools were infrequent along the Equator, more so than were those of spotted dolphin. The eastern spinner was encountered frequently off southern Mexico, where its main population center appeared to be relatively localized. However, eastern spinner dolphins were seen as far offshore as long. 126°W near lat. 10°N and also nearly to the Equator within a broad, 600 nmi coastal belt off Central America. The whitebelly form of spinner dolphin (Perrin 1975a, b; Perrin et al. 1977) occurred in a broad range north and south of the Equator and at the southern and western portions of the total spinner dolphin range Large schools (>300) were seen most- ly off southern Mexico (eastern spinner) and south of the Equator (whitebelly spinner). Striped and common dolphins seem to have a distribution pattern qualitatively different from that of spotted and spinner dolphins (Fig. 5). A distinct distributional lobe, consisting primarily of common dolphins, occurred off Baja California, with exten- sions around the Revilla Gigedo Islands (ca. lat. 19°N, long. 111°W). Between lat. 5° and 12°N, off Central America, there was a second lobe made up of both species. This lobe attenuated to the west along lat. 10°N, but appeared to re-intensify past long. 130°W. Only striped dolphins were seen in the intermediate interval between long. 100°W and 130°W along this lobe A third lobe appeared to originate off Peru. It merged somewhat with the outer portions of the Central American lobe, and then extended westward along the Equator. Equatorial waters were frequented by both species out to about long. 100°W, beyond which striped dolphins apparently predominate. The striped dolphin is the more widespread of the two species; however, its school sizes are small (Tkble 1). Large schools (>300) of the common dolphin occurred within each of the three main centers of its distribution. The above sighting data were adjusted to show the distributions of relative abundance of schools by cor- recting for searching effort. Effort is calculated as miles searched = time searched x ship speed. Relative abundance of schools was expressed as schools per 100 nmi searched (approximately the 170° 30° ISO* — T" 120» — r 90- 80" 70° 30" Hawaiian Itlandt Christmas Island * Marquasa* Islands I I • EASTERN SPINNER O WHITEBELLY SPINNER O CD O Figure 4.— Distribution of spinner doipliin schools, eastern and whitebelly forms, seen during the January-March research cruises. Larger circles indicate schools >300 animals. 627 FISHERY BULLETIN: VOL. 83, NO. 4 170° 30° Hawaiian Islands Christmas Island * o ^o-^o' 10' - Marqussas Islands Figure 5.— Distribution of striped and common dolphin schools seen during the January-March research cruises. Larger circles indicate schools >300 animals. distance searched during 1 d). Data for days during which the Beaufort wind force was equal to 4 (11-16 kn) for more than 50% of the time, or averaged more than 4, were not used. Only schools sighted at distances not more than 3 perpendicular nmi from the ship's track and not more than 5 radial nmi dis- tant were considered, to reduce the effects due to distance on sightability. The latter criterion was not applied to the 1976 cruises, because the radial distances then were frequently overestimated. An areal smoothing procedure that consisted of calculating the average number of schools per 100 mi searched within sequential, overlapping 5° squares was employed. Sequential squares were off- set 2.5° in latitude and longitude, so that each was wholly overlapped by 25% of the area of each of four adjacent squares. A day's sighting rate was assign- ed to a particular square if more than 50% of the search effort occurred therein, and the value was plotted at the center point of that square If the search effort fell equally in two squares, the day's results were assigned to both squares. These moving, areal means of sightings per 100 nmi searched were plotted and contoured. Contouring (and interpola- tion) constituted a second level of areal smoothing. The maps so generated describe the distribution of relative abundance of species schools as surveyed during January to March. These relative abundance data were combined for spotted and spinner dolphins and for striped and common dolphins, both because these species pairs had similar distributions and because pooling gave desirable sample sizes. The combined spotted and spinner dolphin map (Fig. 6) shows some patterns already noted from the school distribution. These schools appeared relatively more abundant off southern Mexico (mostly spotted and eastern spinner dolphins) and again along an east-west band just north of the Equator, especially west of long. 105°W. Another band of greater abun- dance occurred south of the Equator. Spotted and spinner dolphins appeared less abundant just west of Costa Rica, off the coast of northern South America, and along the Equator. A weak lobe of higher relative abundance extended west of long. 120°W broadly about lat. 10°N. Striped and common dolphins show a relative abundance pattern in which areas of higher density tend to be complementary to that of spotted and 628 AU and FERRYMAN: DOLPHIN HABITATS MO' 30' 150' — T" 70' 30' Hawaiian lalandt SPOTTED AND SPINNER DOLPHIN SCHOOLS PER 100 MILES SEARCHED RESEARCH CRUISES JAN.-MAR. 1976. 1977, 1979, 1980 30" I O.S Chrittmaa Island * i Figure 6— Distribution of relative abundance of spotted and spinner dolphins, inferred from data of the January-March research cruises. Hawaiian and Marquesan sightings are not considered. spinner dolphins (Fig. 7). Conspicuous lobes extend- ed from off Baja California and also broadly from the coasts of Central America and northern South America out to and along the Equator. Within the latter lobe, centers of higher relative abundance oc- curred west of Nicaragua and Costa Rica and west of the Galapagos Islands, all areas where reduced abundance of spotted and spinner dolphins occurred. The Galapagos area and the lobe off Baja Califor- nia were dominated by the common dolphin (cf. Fig. 5). Dolphins of Tropical Water and Upwelling-Modified Water Habitats The areas of greater frequency of spotted and spin- ner dolphins during January-March are the typical tropical waters of the eastern tropical Pacific. These waters are underlain by a sharp thermocline, generally >2°C/10 m, at depths usually much <50 m. The surface temperatures are >25°C, and the salinities <34''/oo. Such tropical waters are defined by Wyrtki (1966, 1967) as Tropical Surface Water (see Fig. 1). In particular the warm "Inner Tropical" Waters (Wyrtki 1964a) lying primarily north of the Equator comprise the major habitat of spotted and spinner dolphins (Fig. 6). The waters south of the Equator, where the relative abundance of these dolphins is also higher, are seasonally tropical, and are therefore called southern Subtropical Surface Water (Wyrtki 1966, 1967). These waters, occurring approximately south of lat. 2.5° S, have surface salinities >357oo, and during January-March (southern summer), warm to more than 26°C over a shallow, sharp thermocline Since spotted and spin- ner dolphins occur there frequently, at least during January-March, it appears that spotted and spinner dolphins prefer all waters whose characteristics are, or become, tropical in the eastern Pacific However the primary habitat appears to be the "Inner" Tropical Waters north of the Equator. In contrast striped and common dolphins appear to prefer waters with more variable conditions dur- ing January-March. Their most important habitat ap- pears to be broadly centered about equatorial waters (Fig. 7). This band of distribution extends into the central Pacific along the Equator. In the east it broadens widely to include Tropical Water off Cen- 629 FISHERY BULLETIN: VOL. 83. NO. 4 tral America and Subtropical Water off Peru. The equatorial distribution is in Equatorial Surface Water (Wyrtki 1966, 1967), a transitional water mass straddling the Equator and characterized by salinities between 34 and 35"/oo, upwelling, and a relatively weak thermoclina These waters are markedly cooled from June to December (southern winter-spring) by increased upwelling and by advec- tion from the Peru Current. In the Subtropical Water habitats of striped and common dolphins, both off Peru and Baja California, there are also large seasonal changes in temperature structure and ef- fects from upwelling. Finally the Tropical Water habitat in the Central American Bight is notably variable (below). The waters we call the "Central American Bight" (roughly, the near coastal waters from Guatemala to Ecuador) constitute the most important area of overlap for spotted, spinner, striped, and common dolphins, but this overlap is not balanced among the species. These waters are tropical, but they are the most variable within the Tropical Surface Water pro- vince The Equatorial Countercurrent, flowing eastward between lat. 4°N and 10°N, terminates and turns there, creating a complex circulation. The an- nual north-south migration in these latitudes of the Intertropical Convergence Zone, where north and south trade winds meet, bring southerly winds, rain, reduced salinity, and an intensified Countercurrent during the second half of the year (Bennett 1966; Wyrtki 1967, 1974; Forsebergh 1969). Later during the northern winter (January-March), northeaster- ly winds blow across Central America from the Atlantic, producing coastal upwelling, wind stir- ring, and more complex temperature patterns. The Costa Rica Dome, a localized, offshore upwelling at about lat. 8°N, long. 90°W (Wyrtki 1964b), also may be seasonally intensified (Hofmann et al. 1981). These variable Central American Bight waters ap- pear to have more abundant schools of striped and common dolphins than of spotted and spinner dolphins (cf. Figs. 3-7). It seems that all areas with greater concentration of striped and common dolphins have highly variable oceanographic features that are "upwelling-modified". In spite of the rather strong overlap in distribu- tion among the four dolphin species in the Central America Bight, the biogeographic distinction, in- cluding the relationships to environment, between 170' 30" 160* — r- 1«0' — r 130* — I— 70' 30' 20- Chriatmat Island * COMMON AND STRIPED DOLPHIN SCHOOLS PER 100 MILES SEARCHED RESEARCH CRUISES JAN. -MAR. 1976, 1977, 1979, 1980 / Figure 7.— Distribution of relative abundance of striped and common dolphins, inferred from data of the January-March research cruises. 630 AU and FERRYMAN: DOLPHIN HABITATS the spotted and spinner dolphins of the Tropical Water and the striped and common dolphins of the Upwelling-Modified Water may be quite apparent. This was the case during the intensive winter surveys of 1979 in the Central American Bight, shown in Figure 8. Superimposed in the figure are contours of the 20°C isotherm depth (essentially the ther- mocline depth) which were obtained from expendable bathythermograph probes dropped at 30-60 mi in- tervals (55.6-111.1 km). Notice that spotted and spin- ner dolphins were encountered mainly off southern Mexico, where the deeper 20°C isotherms indicated the occurrence of a large surface lens of warm water. The warmest surface waters in the eastern tropical Pacific normally occur in this area; the thermocline gradient is weaker, and the annual variation in sur- face temperature is relatively small (Wyrtki 1964a). In the more variable tropical waters of the Central American Bight, where the thermocline had shoal- ed or ridged to <60 m, both these species were seen too. However, striped and common dolphin schools predominated, especially near the shallower iso- therms that mark the location of the Costa Rica Dome Finally the equatorial distribution of primar- ily striped and common dolphins was evident. The sampling suggested that Subtropical Waters south of the Galapagos Islands were probably also impor- tant to these latter two species. A 1977 aerial survey of cetaceans in these waters off Central America (SWFC 1977'^) obtained results similar to those just described. Though there appear to be large-scale geographic differences in the habitats of spotted and spinner dolphins and of striped and common dolphins, there was no evidence of negative association among these species. The frequency of days in the Central American Bight, both in 1979 and 1980, with dif- ferent combinations of these species encountered, are summarized in Tkble 2. There was no evidence, using chi-square contingency tests for association among spotted and/or spinner dolphins and striped or common dolphins, that the species were not oc- curring independently on any particular day. Our contention that spotted and spinner dolphins and striped and common dolphins differentially in- habit waters of different oceanographic characteris- ^Southwest Fisheries Center (SWFC). 1977. Aerial survey trip report January - June 1977. Admin. Rep. No. LJ-78-01, 73 p. Soutliwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 120 30° 110" 100° 90° 80° 20< 10° o SPOTTED OB SPINNER DOLPHIN • STRIPED OR COMMON DOLPHIN — DEPTH OF 20° C ISOTHERM (M) 10° 120° 70° -130° 20' - 10° SOUTH AMERICA 10' 110° 100° 90° 80° 70° Figure 8.— Distribution of dolphin schools by species type in the Central American Bight, with reference to the depth of the 20°C isotherm. Data are from the January-March cruises of 1979. 631 FISHERY BULLETIN: VOL. 83, No. 4 Table 2.— Number of days according to combinations of four species of dolphins that were encountered in the Central American Bight. Spotted and/or spinner Striped Common With With With With None of Year Area' Alone striped common striped + common Alone common Alone these spp. 1979 N 17 2 0 0 1 1 1 6 S 6 7 0 4 6 1 6 8 1980 N 9 2 0 0 0 0 0 0 S 5 5 1 4 6 3 3 7 'Northern (N) and southern (S) areas are partitioned to illustrate species differences by area; the dividing line is perpen- dicular to the coast at lat. 16°N. tics is supported by physical environmental measure- ments. The data indicated a species differential with respect to waters with temperature >25°C and salinities <34.5''/oo, and waters of <25°C and >34.5%o. The former is primarily Tropical Surface Water and the latter primarily Equatorial- Subtropical Surface Water. Of 217 spotted and/or spinner dolphin schools encountered during the January-March cruises, and having associated temperature and salinity measurements, 80.6% were in this TVopical Surface Water, while only 19.4% were in the Equatorial- Subtropical Surface Water. In com- parison 53.7% of 229 similarly observed schools of striped and/or common dolphins were in the Ti"opical Surface Water and 46.3% were in the Equatorial- Subtropical Water. The differential in percentages by species pair reflects the more uniform "Inner Tropical" Surface Water habitat of spotted and spin- ner dolphins and the importance to striped and com- mon dolphins of both the variable Tropical Water off Central America and the variable waters along the Equator, west of Peru, and southwest of Baja California. Because the school sighting data are represented by many combinations of temperature and salinity from various areas, it is useful to summarize these results in terms of an integrated measure of temper- ature and salinity, i.e, sigma-^, a measure of water density (see Sverdrup et al. 1942). Table 3 lists the numbers and percentages of spotted and spinner and of striped and common dolphins according to their occurrence at different intervals of sigma-^. The percentages of striped and common dolphin schools were higher than that of spotted and spinner dolphin schools for sigma-t ^22.5 (primarily Equatorial- Subtropical Water). The opposite was true for sigma- t <22.5 (primarily Tropical Water). The difference in percentage distribution by species pair is signifi- cant (P < 0.01, Kolmogorov-Smirnov test). Seasonal Features of Habitats It is not clear how dolphin populations are affected by seasonal changes in the environment, although the available data suggest what may happen. The far offshore habitat of spotted and spinner dolphins, be- tween long. 120°W and 140°W at about lat. 10°N, is an important "porpoise-tuna" fishing area during Table 3. — Percent distribution of dolphin schools^ according to surface water density (sigma-f). sigma-f Spotted Spinner^ Total o/o Striped Common Total % + 18.5-19.0 2 2 0.9 + 19.0-19.5 2 1 3 1.4 5 5 2.2 + 19.5-20.0 6 6 2.8 8 8 3.5 + 20.0-20.5 0.0 2 1 3 1.3 + 20.5-21.0 5 2 7 3.2 6 2 8 3.5 + 21.0-21.5 16 12 28 12.9 11 2 13 5.7 + 21.5-22.0 43 28 71 32.7 27 6 33 14.4 + 22.0-22.5 12 32 54 24.9 26 19 45 19.7 + 22.5-23.0 15 6 21 9.7 22 8 30 13.1 + 23.0-23.5 8 11 19 8.8 39 15 54 23.6 + 23.5-24.0 2 4 6 2.8 10 3 13 5.7 + 24.0-24.5 2 2 0.9 6 6 12 5.2 + 24.5-25.0 1 2 3 1.3 + 25.0-25.5 Total 111 96 217 100 165 64 229 100 'January-March, research cruises, 1976, 1977, 1979, 1980. ^Includes mixed spinner + spotted dolphin schools. 632 AU and FERRYMAN: DOLPHIN HABITATS the spring and summer months (Calkins 1975). However there is little information from that area during the winter months (when there is little fishing), except for results from the January-March research cruises, which indicated that population densities there were not high (Fig. 6). This suggests a summer buildup in the concentration of these dolphins. This is likely because the offshore habitat is centered close to or along the divergence zone at the northern boundary of the North Equatorial Countercurrent, where the thermocline ridges and biological production is increased (Cromwell 1958). During the northern summer, the trade winds over the offshore habitat abate, ridging intensifies (Wyrt- ki 1964a, 1974), and porpoise-tuna fishing expands west of Clipperton Island (at ca. lat. 10°N, long. 109°W). The increase in fishing may be due to bet- ter weather, but possibly also to an increased abun- dance of dolphins and tuna. The seasonal change in environment of the eastern tropical Pacific most likely to affect the distribution of dolphins is that due to the movement of the southern border of the tropical waters. During the northern winter, tropical conditions usually extend to about 15° south of the Galapagos Islands, when a shallow surface layer of warm water develops over what is actually Subtropical Water. During the north- ern summer, the cool Peru and South Equatorial Current strengthen, and a conspicuous thermal front, marking the southern boundary of Tropical Surface Water, usually develops. This Equatorial Front (Wyrtki 1966) is located a few degrees north of the Equator except for a short section east of the Galapagos Islands. We studied the relationship of the Equatorial Front to dolphin populations during an October- November 1977 cruise of the David Starr Jordan. The Equatorial Front was conspicuous, as were the effects of the cool Peru and South Equatorial Cur- rents (Fig. 9). Only 4 of 27 sightings of spotted and spinner dolphins schools in the equatorial region oc- curred in the cool, Equatorial and Subtropical Waters south of the front. The majority of the re- maining 23 sightings were along the Equatorial Front, at the southern border to the warm Tropical Surface Water. These same species had occurred throughout these southern waters during January to March (Figs. 3, 4), when sea surface temperatures of 25°C or more prevailed over this entire area. The apparent redistribution of dolphin schools along the warm edge of the Peru Current and Equatorial Front appeared to be restricted to the spotted and spin- ner species, suggesting their seasonal movements away from cool southern waters. By implication these same dolphin species may migrate into southern waters during the warm season. Seasonal move- ments of dolphins (unidentified) in the southern waters is also suggested by results from the 1967 and 1968 EASTROPAC cruises (Love 1971, 1972). On the other hand, we did see during the October- November cruise four schools of spotted and spin- ner dolphins in cool Subtropical Water, and observers aboard tuna seiners have also reported these same species there during the cool season. I30» IQo — c - I0« - Figure 9.— Distribution of dolphin schools in the equatorial region during October-November 1977, relative to surface isotherms (°C). Notice the Equatorial Front or zone of rapid temperature change just north of the Equator. Clumping of sightings along the track is due to nighttime travel by ship. "Other" dolphins are primarily pilot whales. 633 FISHERY BULLETIN: VOL. 83, NO. 4 Dolphin Communities The habitat differences discussed also apply to other delphinid species, so that there seem to be dif- ferent communities of dolphins in the eastern Pacific To show how the cetacean communities differ be- tween the habitats dominated by spotted and spin- ner and by striped and common dolphins, the eastern Pacific was divided into Areas I and II (Fig. 10, in- set) that separate the main habitat areas of these two species pairs. Area I is primarily Tropical Sur- face Water and includes most areas where thermo- cline ridging is a dominant physical feature Area II is primarily Equatorial-southern Subtropical Sur- face Water, but also includes the wedge-shaped area of variable Tropical Water in the Central American Bight. Area II comprises most of the waters we have called Upwelling-Modified. In each of these two areas, only schools sighted at <1.0 nmi perpendicular to the ship's tracks were listed. This requirement was imposed so that per- cent species composition could be based on species schools, that to the largest practical degree, could all be sighted with equal probability, if present. The change in "sightability" with distance is different for each species because of differences in behavior, coloration, size, etc Our data indicate that the species composition of delphinids is different in these two areas. Percent- age composition was determined for spotted; spin- ner; striped; common; bottlenose, Tursiops trun- catus; rough-toothed, Steno bredanensis; and Risso's, Grampus griseus, dolphins, and for "blackfish", Peponocephala electra/Feresa attenuata; pilot whales, Globicephala macrorhynchus; and others (Tkble 4). Among 8 of 10 species-groups specifically identified in Tkble 4, there were significantly higher percent- ages of spotted, spinner, and rough-toothed dolphin schools in Area I than in Area II (Fig. 10). Risso's and bottlenose dolphins were important species in both areas, and their percentage values did not dif- fer significantly between the areas. The percentages due to striped and common dolphins and pilot whales increased in Area II relative to Area I. Though reduced, the spotted dolphin remained important in the Area II dolphin community. The increase in per- cent composition of the common dolphin in Area II was not quite significant, reflecting the inclusion in Area I of that species' distributional lobe off Baja California. Overall, the species composition differed significantly between the two areas, as determined by chi-square contingency test of the frequency of species other than spotted, spinner, striped, and com- mon, i.e, the species not initially considered when delimiting Areas I and II (x^ = 74.4, df = 5, P < 0.005). Additional evidence for the distinctiveness of the equatorial and subtropical portions of the Area II community is provided by observations along equatorial transects and transects south and south- 30 r 25 (0 §< W< 20 •^ lU 11. H O Z UJ Q. 15 10 27.5* Area I [ 1 Area II 120' 110° 100" 90* Spotted Bottlenose Spinner Risso's Rough- toothed Striped Common DELPHINID SPECIES Pilot Whale Figure 10.— Percent species composition of some important cetaceans, by Area I and II. See Ikble 4. 634 AU and FERRYMAN: DOLPHIN HABITATS Table 4. — Percent composition of species as encountered in two areas' during the January- March research cruises. Area 1 Area II 95%3 95%3 Delphinid spp.^ Schools % C.I. Schools % C.I. Spotted, Stenella attenuata 67 24.7 19.6-29.8 55 13.2 10.0-16.4 Spinner, Stenella longirostris 41 15.1 10.8-19.4 19 4.6 2.6-6.6 Striped, Stenella coeruleoalba 22 8.1 4.9-11.3 86 20.6 16.7-24.5 Common, Delphinus delphis 10 3.7 1.5-5.9 34 8.2 5.6-10.8 Pilot whale. Globicephala macrorhynchus 5 1.8 0.2-3.4 75 18.0 14.3-21.7 Risso's, Grampus griseus 31 11.4 7.6-15.2 65 15.6 12.1-19.5 Bottlenose, Tursiops truncatus 50 18.5 13.9-23.1 58 13.9 10.6-17.2 Rough-toothed, Steno bredanensis 31 11.4 7.6-15.2 10 2.4 0.9-3.9 "Blackfish", Peponocephala electra or Feresa attenuata 8 3.0 1.0-5.0 1 0.2 0-0.6 Other 6 2.2 0.2-4.2 14 3.4 1.7-5.1 Total 271 417 'Areas are shown in Figure 10. ^Species in mixed schools were tabulated separately. ^Normal approximation to binomial distribution. east of the Galapagos Islands, i.a, off Peru. Unlike during the January-March cruises, physical ocean- ography was the primary task on most of these transects, hence the ships did not usually divert course toward the schools, and many schools could not be identified. Nevertheless some idea of the species compositions can be obtained. The observa- tions (Ihble 5) showed that pilot whales and Risso's and bottlenose dolphins were frequently encountered species during October- December off Peru, and com- mon dolphins were often seen near the coast. On the equatorial transects, between long. 85°W and 110°W, striped and common dolphins were the characteristic species. The common dolphin was seen most often near the Galapagos Islands. Pilot whales were relatively abundant during May- July 1981 in this equatorial section. West of long. 110°W along the Equator, pilot whales again were the most fre- quently encountered species. Interestingly, sightings of Eraser's dolphins, Lagenodelphis hosei, and "blackfish" (probably Peponocephala electra) were also relatively frequent, especially between long. 110°W and 145°W. These two species often school together and appear to prefer equatorial waters (Perryman et al.'*). In the next section, another distinctive feature of equatorial waters will be brought out. Dolphins, Birds, and Tuna A conspicuous feature distinguishing the dolphin communities is the difference in the species-specific association with tunas. In the eastern tropical Pacific, spinner dolphins and especially spotted dolphins are found associated with "surface" yellowfin tuna. It is these two species, therefore, that are mainly affected by the porpoise-tuna fishery (Smith 1983). Surface tunas occur at the sea surface and can be caught by purse seine, trolling, and pole- and-line gear. "Deep tunas" of the same species are caught by longline gear, generally in and below the thermocline Since these surface tunas drive food to the surface, making it available to certain seabirds (Ashmole and Ashmole 1957; Murphy and Shomura 1972), a reliable indication that tuna are accompany- ing a dolphin school is the presence of a bird flock. Birds are the most important cue used by fishermen to locate dolphin-tuna schools. Birds do not occur equally among the different dolphin species. During the 1977, 1979, and 1980 January-March cruises (when the best bird observa- ■•Perryman, W. F., D. W. K. Au, and S. Leatherwood. Manuscr. prep. Melon-headed whale, Peponocephala electra (Gray, 1946) (with notes on the pygmy killer whale Feresa attenuata). South- west Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 635 FISHERY BULLETIN: VOL. 83, NO. 4 05 o CVI § 2 o o O C\J 00 L.' o o in CB Tt 3 a- LU o o 0} a. CO JO o o .c o CO E 3 in UJ m CO 0 5 o o in CM ^ o o o « 3 T~ (T ai CO o O o ^' in c» CO 3 I 0 CO 00 ^ ° K CD .-■ (35 ^ Cj CD Q 00 •<* Tj- CM T- 1- CO CM •* CO CO •>- •* CO m ■^ CM •>- 1- CD in C3> ^1- CO O ■^ CO '- •>- 1- CM CM C3. Q. CO c ce *- j= CD Q. 3 O O 1 ■a CD .* CD CO e c: CD CD O CD 5 CU iQSSlcofcoE C Q. Z) CO CO CO .CO CO " .C CO -c o Ci c m CO CD O c: CO a .o io 5 o' I o CD E .CD CD -C & o 5 "> o o rri "^ Z -5 « CD 3 -C C Q. m 0) fc O CD C5 n. r- CD o Q. SI CQ CO CO C I 0) CO s (D . t: ^ p ® o O Q CO CO = -Q -C 3 s ■- e 636 AU and FERRYMAN: DOLPHIN HABITATS tions were taken), 70.7% of 167 spotted, spinner, or mixed spotted and spinner dolphin schools, sighted between lat. 5°N and 30°N (where most dolphin-tuna associations occur), had 10 or more associated birds (Tkble 6). We assume this minimum flock size in- dicates associated tuna, although we did not often actually see the tuna. For the same period and area, 30.6% of 49 common dolphin schools and only 1.6% of 64 striped dolphin schools were with 10 or more birds. The different percent occurrences indicate that tuna are most frequently associated with spot- ted and spinner dolphins and very seldom with the striped dolphin. The bird species most numerous in offshore waters with dolphin schools are boobies {Sula spp.); wedge- tailed shearwaters, Puffinus pacificus; and sooty terns. Sterna fuscata. Frigate birds {Fregata spp.) are also closely associated with these dolphins, though their average flock size is only seven (Au and Pitman^). These bird species are all strongly depen- dent upon tunas in their feeding. Our observations are that the birds feed primarly in association with the fish, not the dolphins. The dolphin-tuna-bird association appears to be area- as well as species-specific Assuming flocks of ^10 birds indicate the presence of yellowfin tuna, this association seems to occur in all areas with higher relative abundance of spotted and spinner dolphins (Fig. 11). The association seldom occurs along the Equator, or in areas outside the traditional porpoise-tuna fishing grounds (roughly these are waters within the triangular-shaped area whose base is formed by the American coasts between lat. 25°N and 15°S, and whose apex is at lat. 10°N, long. 150°W; see Calkins 1975 and lATTC 1979-81), even though the required species of dolphins, tuna, and ^Au, D. W. K., and R. L. Pitman. Manuscr. prep. Seabird interactions with dolphins and tuna in the eastern tropical Pacific Southwest Fisheries Center La Jolla Laboratory, Na- tional Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. birds may be present there Relatively few of the spotted or spinner dolphin schools sighted near the Equator were with bird flocks. South of the Equator and in the Central Pacific, there are abundant flocks of sooty terns with fish schools, but dolphins do not usually co-occur (Au and Pitman fn. 5). The most im- portant area of the dolphin-tuna-bird association is centered about the divergence zone near lat. 10°N, an important porpoise-tuna fishing ground for yellowfin tuna (primarily Area I of Figure 10). These areal changes in the interactions among species are usually clearly apparent during north- south transects across the eastern Pacific during any season. An example is the July-September 1980, NOAA ship Researcher transits between Manzanillo, Mexico, and lat. 3°S, long. 100°W, via Clipperton Island. At that time Tropical Surface Water (T > 25°C) extended to the Equator with transition toward Equatorial Water marked by temperature and salinity fronts at lat. 5°N-6°N and at the Equator (Fig. 12). In the tropical waters north of the front at lat. 5°N there was a conspicuous increase in abundance of flocks of sooty terns, boobies, and shearwaters {Puffinus spp.), all broadly centered about the thermocline ridge at lat. 10°N. The larger gadfly petrels (Pterodroma spp.) were also abundant. All of these birds frequently flock over fish and dolphins. South of this boundary was another avian group, with Pterodroma leucoptera, a small petrel from the Southern Hemisphere, and planktivorous storm petrels (Oceanodroma spp.) predominating, and peak abundances at the Equator. These latter birds usually feed independently of fish and seldom flock over fish and dolphins. Correlations among water masses, seabirds, and the different cetaceans along the transect were dif- ficult to quantify because of the small sample size of the latter. However, the observations are suppor- tive of such relationships. There were 39 dolphin schools, of which 15 were unidentified, and 21 whale schools along this transect. Spotted and spinner Table 6. — Dolphin schools associated with seabirds. 5°N-30°N <5<'N with with Dolphin spp. Schools >10 birds o/o Schools >10 birds % Spotted 95 56 58.9 40 8 20.0 Spotted and spinner 55 53 96.4 21 8 38.1 Spinner 17 9 52.9 19 6 31.6 Common 49 15 30.6 10 1 10.0 Striped 64 1 1.6 76 1 1.3 Rough-toothed 40 5 12.5 13 1 7.7 Other/unidentified 388 17 4.4 189 6 3.2 637 FISHERY BULLETIN: VOL. 83, NO. 4 170" 160* ISO* 140* 130* 120* no* 100* 90* 80* 70* 1 30° 1 1 1 1 1 NA\ ( SPOTTED AND SPINNER )\ \ uFYirn i DOLPHIN SCHOOLS AND «^x MCAH-u * ASSOCIATED BIRD FLOCKS 30- Hawaiian Islands • o o \ ^ *■§. \ \ r^ 20" i> 0 ° 0 °° •^'S^ ^X-^^TJ U '"®- '^^^' '3^3' '880 ~ • jfVo^capuIco ]' RESEARCH VESSEL CRUISES 20* 10* _ o* O o • o • o o oo • ,o o. • 10* 0 •• o o • oo 0* ~ Christmas Island ° o °."-o • o • o §°ef ° °°°0 0 J3 . ** • / Oo " o o [Guayaquil ( SOUTH o o, o°OoO 10 BIRDS WITH SCHOOL 1 30' 1 170* 1 160* 1 ._. ISO* 1 140* 1 130* 1 120* 1 1 1 1 U30* 110* 100* 90* 80* 70* Figure 11.— Distribution of bird-associated spotted and spinner dolphin schools during the January-March research cruises. Schools with 10 or more birds are assumed to be with tuna. dolphns did occur coincidentally with bird flocks (and probably with tuna) in the warmer tropical waters centered about lat. 10°N, but also were seen at the Equator, though without associated bird flocks. Striped and common dolphins occurred in coastal and equatorial waters, without bird flocks. These lat- ter dolphins were the characteristic species on the equatorial transect of this same cruise (see Ikble 5, col. 4), where only one of the schools seen, a mixed spotted and spinner dolphin school, had an associated bird flock. The distinct change in the bird fauna south of lat. 5°N to species that do not forage commensally with fish suggests there are changes with water masses in the nature of epipelagic prey and how the top predators forag& DISCUSSION It should not be surprising that the two major divergence zones of the eastern Pacific, near lat. 10°N and along the Equator, are important features of the oceanic habitats of eastern Pacific cetaceans. Enriched by the effects of wind and the major zonal currents (Brandhorst 1958; Cromwell 1958; Reid 1962; Wyrtki 1966), the zones are evident areas of enhanced biological production (Blackburn 1966, 1976; Blackburn et al. 1970; King 1974; Parsons et al. 1977; Brinton 1979). that are important to tunas (Calkins 1975; Blackburn 1965; Blackburn and Laurs 1972; Blackburn and Williams 1975; Sund 1981) and cetaceans as discussed above These two zones are not qualitatively the same, however; the strong, shallow thermoclines that have been related to suc- cessful porpoise-tuna fishing (Green 1967; Miller et al.^) and to aggregations of dolphins and baleen whales (Rovnin 1969; Volkov and Moroz 1976) are characteristic of the lat. 10°N zone, but not of the Equator. Similarly the oxygen minimum layer, noted by Perrin et al. (1976) to be correlated with the distribution of the spotted dolphin, occurs only north and south of the Equator in the eastern Pacific Equatorial waters are characterized by shallow, weak (<2°C/10 m) thermoclines, due to upwelling and the Equatorial Undercurrent, and cool surface temper- «Miller, F. R., C. J. Orange, R. H. Evans, and K. A. Bliss. Manuscr. prep. Analysis of environment related to tuna fishing in ETR Inter-American Tropical TUna Commission, La Jolla, CA 92038. 638 AU and FERRYMAN: DOLPHIN HABITATS UJ o o CO _J o o X o o < h- o o o 2 - □ Whales ^ Spotted / Spinner Dolphin 0 Striped / Common Dolphin □ Other Dolphin July - Sept. 1980 i p i i i f i 30 28 26 24 22 20° - SURFACE "~^^v ^ SALINITY %o -^ \ ^^^,^^'?~^ _ /T SURFACE TEMPERATURE "C 1 1 1 1 1 1 1 1 1 1 1 1 35.0 34.5 34.0 o 335 -^ 330 200 100 Boobies, Shearwaters, Petrels White Winged Petrel Terns Storm Petrel I LATITUDE Figure 12.— Transect along long. 110°W, July-September 1980, showing changes in cetaceans, birds, and temperature and salinity. The relative abundance of dolphin schools include unidentified schools which were prorated according to species ratios of identified schools. 639 FISHERY BULLETIN: VOL. 83, NO. 4 ature from June through December. These waters support a different cetacean community (Ikble 5), though school densities there can be as high as in areas off Mexico and Central America (Au et al. 1980). Other relationships between distribution and movements of dolphins and water masses, conver- gences, and thermal conditions have been described by Fraser (1934), Gaskin (1968), Kasuya (1971), Nishiwaki (1975), Evans (1975), and Miyazaki and Nishiwaki (1978). Hui (1979) found that common dolphins off California tended to occur over promi- nent features of bottom topography. The deep depths of such areas suggest that surface eddies and conver- gences caused by topography-induced accelerations to deep reaching currents (Sverdrup et al. 1942; Neumann 1960) may have concentrated food and at- tracted the dolphins. The distributions of dolphin species as seen from the January-March cruises (Figs. 3-7) are similar to the all-season school distributions derived from data of scientific observers aboard tuna seiners. These data, consisting of thousands of sightings per species, were recently summarized by Scott (1981) and Perrin et al. (1983). The same major distribu- tional patterns as presented here for the January- March cruises were apparent, including, for spotted and spinner dolphins, the relative unimportance of equatorial latitudes and the secondary band of in- creased concentration of schools 2°5° north of the Equator. The latter may be related to the Equatorial Front and increased food concentration and possibly production in the convergence zone there (Sette 1955; King and Iverson 1962; Blackburn and Laurs 1972; Murphy and Shomura 1972; Pak and Zaneveld 1974; Blackburn and Williams 1975; Greenblatt 1979). Increased abundance of micronekton occurs at least sometime in this zone (Love 1971, 1972; Blackburn and Laurs 1972). The purse seiner data, like those of this paper also showed spotted and spin- ner dolphins more concentrated in the tropical waters off Mexico and along lat. 10°N, while striped and common dolphins tended to be found in the Cen- tral American Bight and along the Equator. This complementary type of distribution was less ap- parent with the more broadly distributed striped dolphin. It should be noted that our southern distributional lobe for spotted and spinner dolphins, at ca. lat. 5°S, may in part be due to the sampling pattern. However, the density of these dolphins along the Equator is definitely reduced, and we know of no information that does not suggest a rapid decline in density south of our lobe Our January-March data differs from the all- season data in indicating fewer schools in the area around the Revilla Gigedo Islands (at ca. lat. 19°N, long. 111°W) and between long. 90 °W and 100°W along lat. 10°N for spotted and spinner dolphins. Also our data suggested that striped and common dolphins had a more localized distribution near the region of the Costa Rica Dome, and were relatively infrequent between long. 105°W and 120°W, along lat. 10°N. These differences may be due to seasonal changes in distribution. The relative densities of these dolphins, as school encounter rates in the tuna purse seine fishery, were recently calculated by Polacheck (1983). The patterns he derived were fragmentary, but not inconsistent with those of this paper. He showed, for example, higher densities of spotted and spinner dolphins ex- tending to the southwest from off southern Mexico and reduced densities in the Central American Bight. For striped and common dolphins, he also described a three-lobed distribution pattern as in this paper. However his equatorial lobe was centered just south of the Equator. It seems likely that the dolphin community of the Upwelling-Modified Water differs from the Tropical Water community because of water-mass specific dif- ferences in the distribution and availability of food. This is supported by the different biotic features of Equatorial and Subtropical Waters relative to Tropical Waters. The distinction is clearly shown by the surface distribution of nutrients and primary production in these waters as measured during the EASTROPAC cruises (Love 1971, 1972). The equatorial waters of the eastern Pacific in particular are different. They support abundant plankton- feeding storm petrels rather than fish and cephalopod-feeding flocking birds that are usually abundant both north and south of the Equator (see also Love 1971, 1972 and King 1974). Dolphin species along the Equator tend not to be with fish or birds (Figs. 11, 12), and the species composition of the cetacean community appears to be distinct (Tkble 5; Au and Pitman 1981). Of course it has previously been known that equatorial waters are notable in being important sperm whale grounds (Ibwnsend 1935) and have a zooplankton community distinct from other parts of the eastern tropical Pacific (McGowan 1972). Finally the fact that the common dolphin, a species characteristic of coastal upwell- ing waters from California to Peru, occurs with greater frequency in the equatorial waters and near upwelling areas in the Central American Bight, sug- gests that the shorter and different food chains of the upwelling environments (Parsons et al. 1977) may be the basis of the community difference 640 AU and FERRYMAN: DOLPHIN HABITATS The dolphin-tuna-bird association is one manifesta- tion of community difference that is both striking and of ecological interest. The distribution of this association is notable in that it seems coincident with both the main habitats of spotted and spinner dolphins and the distribution of "surface" yellowfin tuna in the eastern Pacific (see Figure 11, Shingu et al. 1974, and Suzuki et al. 1978). Since it is primarily these dolphins that are associated with birds and with yellowfin tuna, the geography of the dolphin-bird association also defines that of the dolphin-yellowfin tuna association. This association of birds and fish with dolphins occurs in all tropical waters, including the southern Subtropical Water during the southern summer. It is apparently rare in equatorial waters of the eastern Pacific, in the cen- tral and western Pacific (Myazaki and Wada 1978; Au et al. fn. 2), and in the eastern tropical Atlantic (Levenez et al. 1980). In the central Pacific the same bird species found with spotted and spinner dolphins in the eastern tropical Pacific can be abundant (Gould 1974), and sooty terns especially, are fre- quently associated with small tunas, but these are most likely skipjack tuna (Murphy and Ikehara 1955; Waldrom 1964; Hida 1970; Blackburn and William 1975). Apparently those flocks seldom accompany dolphins or schools of larger yellowfin tuna. Yellowfin tuna and dolphins seldom seem to associate outside the eastern tropical Pacific The obvious feeding activity often seen in these joint aggregations of birds, spotted and spinner dolphins, and tuna suggests that these species have similar food and foraging requirements. Our obser- vations indicate that the mammals and fish are not tightly associated in the aggregations (see Au and Perryman 1982) and probably feed independently (see also Norris and Dohl 1980a). However the tuna, birds, and spotted dolphin (at least) do appear to be feeding at the same time. Both the tuna and spot- ted dolphins feed on epipelagic fish and on squids (Perrin et al. 1973; Olson 1982), but the spinner dolphins feed differently and may forage more at night (Perrin et al. 1973; Norris and Dohl 1980b); though they are active in these feeding aggregations, they may not be directly associated with the tuna. Judging from the associations of bird and dolphin species, only spotted and spinner dolphins frequently find it advantageous to feed with yellowfin tuna. Fur- thermore the distribution of this association suggests that the necessary kind and behavior of prey that is likely the basis of the association appears charac- teristic of tropical, but not equatorial, waters. It oc- curs especially where a shallow thermocline may con- strain the yellowfin tuna to the surface layer with the dolphins, a complex interaction between environ- ment and physiology (Sharp 1978) that may cause the phenomenon known as "surface tuna". Finally the distribution of the bird-dolphin association in- dicates that the dolphin-tuna association is characteristic of areas of higher school density of spotted and spinner dolphins. High population den- sities of both dolphins and yellowfin tuna, and suitable prey are therefore likely requisites for joint dolphin-tuna schools. The dolphin-tuna association is a feature of the most productive tuna fishing zones of these tropical seas. In such rich areas, feeding tac- tics to exploit clumped prey could lead to multi- species aggregations of predators, as explained by Schoener (1982). ACKNOWLEDGMENTS This paper was reviewed by the following persons whose help is much appreciated: David E. Gaskin, Kenneth S. Norris, Carleton Ray, Gunter R. Seckel, Paul N. Sund, James G. Mead, Frank G. Alverson, Richard Pimentel, John A. McGowan, Gary T. Sakagawa, Jay Barlow, Andy Dizon, Albert C. Myrick, William Perrin, Steve Reilly Tim Smith, Paul E. Smith, Rennie S. Holt, Eric D. Forsebergh, and Michael D. Scott. We thank the officers and crews of the NOAA ships involved in this study, especially those of the Townsend Cromwell and David Starr Jordan, for their cooperation and help. We particularly thank the biological technicians who manned the watches through high-powered binoculars, for their keen observations and provoking discussions at sea. Among them we especially appreciate the work of Gary Friedrichsen, Phillip Unitt, James Lambert, Dale Powers, Robert Pitman, James Cotton, and Scott Sinclair. We sincerely thank Lorraine Prescott and her staff for careful typing of this manuscript and Ken Raymond, Roy Allen, and Henry Orr for drafting the illustrations. LITERATURE CITED Alverson, F. G. 1981. Comments on the distribution of spotted, spinner, com- mon and striped dolphin in the tropical Pacific Ocean. In P. S. Hammond (editor), Report on the Workshop on lUna- Dolphin Interactions, p. 109-124. Inter-Am. Trop. Ibna Comm., Spec Rep. 4, App. 5. ASHMOLE, N. P., AND M. J. ASHMOLE. 1957. Comparative feeding ecology of seabirds of a tropical oceanic island. Yale Univ. Peabody Mus. Nat. Hist. Bull. 24:1-131. 641 FISHERY BULLETIN: VOL. 83, NO. 4 Au, D. W., W. L. Ferryman, R. L. Pitman, and M. S. Sinclair. 1980. Cetacean populations and equatorial oceanography. Trop. Ocean Atmos. Newsl., Univ. Wash., 4:6-10. Au, D. W. K., and R. L. Pitman. 1981. The equatorial cetacean community of the eastern Pacific. (Abstr.) Proceedings of the Fourth Biennial Con- ference on the Biology of Marine Mammals, Dec 14-18, 1981, p. 95. Au, D., AND W. Perryman. 1982. Movement and speed of dolphin schools responding to an approaching ship. Fish. Bull., U.S. 80:371-379. Bennett, E. B. 1966. Monthly charts of surface salinity in the eastern tropical Pacific Ocean. [In Engl, and Span.] Inter-Am. Trop. lUna Comm., Bull. 11:3-44. Blackburn, M. 1965. Oceanography and the ecology of tunas. Oceanogr. Mar. Biol. Annu. Rev 3:299-322. 1966. Biological oceanography of the eastern tropical Pacific: summary of existing information. U.S. Fish. Wildl. Serv, Spea Sci. Rep. Fish. 540, 18 p. 1976. Review of existing information on fishes in the deep ocean mining environment study (DOMES) area of the tropical Pacific Inst. Mar. Res., Univ Calif., IMR NO. 76-1, 77 p. Blackburn, M., and R. M. Laurs. 1972. Distribution of forage of skipjack tuna (Euthynnus pelamis) in the eastern tropical Pacific U.S. Dep. Commer., NOAA Ifech. Rep. NMFS SSRF-649, 16 p. Blackburn, M., R. M. Laurs, R. S. Owen, and B. Zeitschel. 1970. Seasonal and areal changes in standing stocks of phyto- plankton, zooplankton, and micronekton in the eastern tropical Pacific Mar. Biol. (Berl.) 7:14-31. Blackburn, M., and F. Williams. 1975. Distribution and ecology of skipjack tuna, Katsuwonics pelamis, in an offshore area of the eastern tropical Pacific Ocean. Fish. Bull., U.S. 73:382-411. Brandhorst, W. 1958. Thermocline topography, zooplankton standing crop, and mechanisms of fertilization in the eastern tropical Pacific J. Cons. Int. Explor. Mer 24:16-31. Brinton, E. 1979. Parameters relating to the distribution of planktonic organisms, especially euphausiids in the eastern tropical Pacific Prog. Oceanogr. 8:125-189. Calkins, T R 1975. Geographical distribution of yellowfin and skipjack tuna catches in the eastern Pacific Ocean, and fleet and total catch statistics, 1971-1974. [In Engl, and Span.] Inter-Am. Trop. llina Comm., Bull. 17:3-116. Cromwell, T. 1958. Thermocline topography, horizontal currents and "ridging" in the eastern tropical Pacific [In Engl, and Span.] Inter-Am. Trop. Ibna Comm., Bull. 3:135-164. Evans, W. E. 1975. Distribution, differentiation of populations, and other aspects of the natural history of Delphiniis delphis Linnaeus in the northeastern Pacific Ph.D. Thesis, Univ. California, Los Angeles, 164 p. FORSBERGH, E. D. 1969. On the climatology, oceanography, and fisheries of the Panama Bight. [In Engl, and Span.] Inter-Am. Trop. Ibna Comm., Bull. 14:45-385. Eraser, F C. 1934. Report on cetacea stranded in British coasts from 1927 to 1932. Br. Mus. (Nat. Hist.) No. 11, 41 p. Gaskin, D. E. 1968. Distribution of delphinidae (Cetacea) in relation to sea surface temperature off eastern and southern New Zealand. N.Z. J. Mar. Freshw. Res. 2:527-534. Gould, P. J. 1974. Sooty tern {Sterna fuscata). In W. B. King (editor). Pelagic studies of seabirds in the central and eastern Pacific Ocean, p. 6-52. Smithson. Contrib. Zool. No. 158. Green, R. E. 1967. Relationship of the thermocline to success of purse sein- ing for tuna. Trans. Am. Fish. Soc 96:126-130. Greenblatt, P. R. 1979. Associations of tuna with flotsam in the eastern tropical Pacific Fish. Bull., U.S. 77:147-155. HiDA, T S. 1970. Surface tuna schools located and fished in equatorial eastern Pacific Commer. Fish. Rev 32(4):34-37. Hofmann, E. E., a. J. Busalacchi, and J. J. O'Brien. 1981. Wind generation of the Costa Rica Dome Science (Wash., D.C.) 214:552-554. Hui, C. A. 1979. Undersea topography and distribution of dolphins of the genus Delphinus in the southern California Bight. J. Mam- mal. 60:521-527. Inter-American Tropical Hina Commission. 1979. Annual report of the Inter-American Tropical Ibna Commission. [In Engl, and Span.] Inter-Am. Trop. Thna Comm., La Jolla, CA. 1980. Annual report of the Inter-American Tropical Ibna Commission. [In Engl, and Span.] Inter-Am. Trop. Ibna Comm., La Jolla, CA. 1981. Annual report of the Inter-American Tropical Ibna Commission. [In Engl, and Span.] Inter-Am. Trop. Ibna Comm., La Jolla, CA. Kasuya, T 1971. Consideration of distribution and migration of toothed whales of the Pacific coast of Japan based upon aerial sighting record. Sci. Rep. Whales Res. Inst. 23:37-60. King, J. E., and R. T. B. Iverson. 1962. Midwater trawling for forage organisms in the central Pacific 1951-1956. U.S. Fish Wildl. Serv, Fish. Bull. 62: 271-321. King, W. B. (editor). 1974. Pelagic studies of seabirds in the central and eastern Pacific Ocean. Smithson. Contrib. Zool. No. 158, 277 p. Levenez, J., A. Fonteneau, and R. Regalado. 1980. Resultats d'une enquete sur I'importance des dauphins dans la pecherie thoniere FISM. Int. Comm. Conserv. Atl. Ibnas, Coll. Vol. Sci. Pap. IX(1):176-179. Love, C. M. (editor). 1971. Biological and nutrient chemistry data from principal participating ships, first survey cruise, February-March 1967. EASTROPAC Atlas, Vol. 2. Natl. Mar. Fish. Serv., Circ 330, 87 p. 1972. Biological and nutrient chemistry data from principal participating ships, second survey cruise, August-September 1967. EASTROPAC Atlas, Vol. 6. Natl. Mar. Fish. Serv., Circ 330, 80 p. McGowan, J. A. 1972. The nature of oceanic ecosystems. In C. B. Miller (editor), The biology of the oceanic Pacific, p. 9-28. Proc 33d Annu. Biol. Colloq., Oreg. State Univ. Press, Corvallis, OR. MiYAZAKI, N., AND M. NiSHIWAKI. 1978. School structure of the striped dolphin off the Pacific coast of Japan. Sci. Rep. Whales Res. Inst. 30:65-115. 642 AU and PKRRYMAN: DOLPHIN HABITATS MiYAZAKI, N.. AND S. WADA. 1978. Observations of cetacea during whale marking cruise in the western tropical Pacific 1976. Sci. Rep. Whales Res. Inst. 30:179-195. Murphy, G. I., and I. I. Ikehara. 1955. A summary of sightings of fish schools and bird flocks and of trolling in the central Pacific U.S. Fish Wildl. Serv., Spec Sci. Rep. Fish. 154, 19 p. Murphy, G. I., and R. S. Shomura. 1972. Pre-exploitation abundance of tunas in the equatorial central Pacific Fish. Bull., U.S. 70:875-913. Neumann, G. 1960. On the effect of bottom topography on ocean cur- rents. Dtsch. Hydrogr. Z., 13:132-141. NiSHIWAKI, M. 1975. Ecological aspects of smaller cetaceans, with emphasis on the striped dolphin (Stenella coeruleoalba). J. Fish. Res. Board Can. 32:1069-1072. Norris, K. S., and T. D. Dohl. 1980a. The structure and functions of cetacean schools. In L. M. Herman (editor). Cetacean behavior, mechanisms and functions, p. 211-261. John Wiley and Sons, N.Y. 1980b. Behavior of the Hawaiian spinner dolphin, Stenella longirostris. Fish. Bull, U.S. 77:821-849. Olson, R. J. 1982. Feeding and energetic studies of yellowfin tuna, food for ecological thought. Coll. Vol. Sci. Pap., Int. Comm. Con- serv Atl. Ibnas XVII(2):444-457. Parsons, T. R., M. Takahashi, and M. Hargrave. 1977. Biological oceanographic processes. Pergamon Press, N.Y., 332 p. Pak, H., and J. R. V. Zaneveld. 1974. Equatorial front in the eastern Pacific Ocean. J. Phys. Oceanogr. 4:570-578. Perrin, W E 1975a. Distribution and differentiation of populations of dolphins of the genus Stenella in the eastern tropical Pacific J. Fish. Res. Board Can. 32:1059-1067. 1975b. Variation of the spotted and spinner porpoise (genus Stenella) in the eastern tropical Pacific and Hawaii. Bull. Scripps Inst. Oceanogr. 21, 206 p. Perrin, W. E, J. M. Coe, and J. R. Zweifel. 1976. Growth and reproduction of the spotted porpoise, Stenella attenuata, in the offshore eastern tropical Pacific Fish. Bull.. U.S. 74:229-269. Perrin, W. F, D. B. Holts, and R. B. Miller. 1977. Growth and reproduction of the eastern spinner dolphin, a geographical form of Stenella longirostris in the eastern tropical Pacific Fish. Bull., U.S. 75:725-750. Perrin, W. F, G. J. Walker, F M. Ralston, and D. W. K. Au. 1983. Distribution of four dolphins (Stenella spp. and Delphinus delphis) in the eastern tropical Pacific with an an- notated catalog of data sources. U.S. Dep. Commer., NOAA Tfech. Memo. NMFS-SWEC-38, 65 p. Perrin, W. E, R. R. Warner, C. H. Fiscus, and D. B. Holts. 1973. Stomach contents of porpoise, Stenella spp., and yellowfin tuna, Tkunnus albacares, in mixed-species aggre- gations. Fish. Bull., U.S. 71:1077-1092. Polacheck, T. 1983. The relative abundance of dolphins in the eastern tropical Pacific based on encounter rates with tuna purse seiners. Ph.D. Thesis, Univ. Oregon, Eugene, 475 p. Reid, J. L., Jr. 1962. On circulation, phosphate-phosphorus content, and zoo- plankton volumes in the upper part of the Pacific Ocean. Limnol. Oceanogr. 7:287-306. ROVNIN, A. A. 1969. Distribution of true whales in the tropical part of the Pacific In V. A. Arseniev, B. A. Zenkovich, K. K. Chapsky (editors). Transactions of the Third AU-Union Conference on Marine Mammals, p. 144-149. Nauka, Moskva, U.S.S.R. Schoener, T. W. 1982. The controversy over interspecific competition. Am. Sci. 70:586-595. Scott, M. D. 1981. Dolphin stocks in the eastern tropical Pacific In P. S. Hammond (editor). Report on the workshop on tuna-dolphin interactions, p. 98-107. Inter-Am. Trop. llina Comm. Spec Rep. No. 4, App. 4. Sette, 0. E. 1955. Consideration of midocean fish production as related to oceanic circulatory systems. J. Mar. Res. 14:398-414. Sharp G. D. 1978. Behavioral and physiological properties of tunas and their effects on vulnerability to fishing gear. In G. D. Sharp and A. E. Dizon (editors). The physiological ecology of tunas, p. 397-449. Acad. Press, N.Y. Shingu, C, p. K. TbMLiNSON, AND C. L. Peterson. 1974. A review of the Japanese longline fishery for tunas and billfishes in the eastern Pacific Ocean, 1967-1970. [In Engl, and Span.] Bull. Inter-Am. Trop. TUna Comm. 16:67-230. Smith, T D. 1983. Changes in size of three dolphin (Stenella spp.) popula- tions in the eastern tropical Pacific Fish. Bull., U.S. 81:1-13. SuND, P. N., M. Blackburn, and F. Williams. 1981. Ibnas and their environment in the Pacific Ocean: a review. Oceanogr. Mar. Biol. Annu. Rev. 19:443-512. Suzuki, Z., P. K. IbMLiNSON, and M. Honma. 1978. Population structure of Pacific yellowfin tuna. [In Engl, and Span.] Inter-Am. Trop. TUna Comm. Bull., 17:277-358. SvERDRUP, H. U., M. W. Johnson, and R. H. Fleming. 1942. The oceans, their physics, chemistry, and general biology. Prentice-Hall, Inc, N.Y., 1087 p. TbWNSEND, C. H. 1935. The distribution of certain whales as shown by logbook records of American whaleships. Zoologica 19:1-50. Volkov, a. F, and I. F. Moroz. 1976. Oceanological conditions of the distribution of cetacea in the eastern tropical part of the Pacific Ocean. Int. Whal- ing Comm. Rep. 27:186-188. Waldron, K. D. 1964. Fish schools and bird flocks in the central Pacific Ocean, 1950-1961. U.S. Fish Wildl. Serv., Spec Sci. Rep. Fish. 464, 20 p. Wyrtki, K. 1964a. The thermal structure of the eastern Pacific Ocean. Erganzungsh. Dtsch. Hydrogr. Z. Reiche A 8(6), 84 p. 1964b. Upwelling in the Costa Rica Dome U.S. Fish Wildl. Serv., Fish. Bull., 63:355-372. 1966. Oceanography of the eastern equatorial Pacific Ocean. Oceanogr. Mar. Biol. Annu. Rev. 4:36-68. 1967. Circulation and water masses in the eastern equatorial Pacific Ocean. Int. J. Oceanol. Limnol. 1:117-147. 1974. Equatorial currents in the Pacific 1950 to 1970 and their relations to the trade winds. J. Phys. Oceanogr. 4:372- 380. 643 ASPECTS OF THE LIFE HISTORY OF THE FLUFFY SCULPIN, OLIGOCOTTUS SNYDERI Mary C. Freeman,^ Nate Neally,^ and Gary D. Grossman^ ABSTRACT We examined age structure, growth rates, and diets of male and female Oligocottus snyderi Greeley at Dillon Beach, CA, where this sculpin numerically dominates the mid- and lower intertidal fish assemblage Two age classes, 0+ and 1 + , were present; maximum lifespan was about 1.5 years. Instantaneous popula- tion growth rates were highest for the O-t- age class, and most individuals attained spawning size during the first year of life Growth rates for both age classes were highest during the high productivity Upwell- ing period and minimal during the low productivity, Oceanic-Davidson Current period. Males and females primarily consumed gammarid amphipods and polychaetes. Larger individuals (>50 mm SL) of both sexes consumed a wider variety of prey, including shrimps, crabs, and isopods. Among year and seasonal dietary changes were minimal. Females consumed two times more gammarids by weight than males during the low productivity Oceanic-Davidson Current period, when ovarian recrudescence occurs. Females may increase food consumption to meet the increased energetic demands of egg production. Rapid sexual maturation and growth and the occurrence of recruitment during upwelling probably are adaptations to the pronounced annual cycle of productivity. These adaptations, together with intense utiliza- tion of an abundant prey (gammarids) not widely consumed by other assemblage members, probably con- tribute to 0. snyderi's numerical dominance in the rocky intertidal of central California. The fluffy sculpin, Oligocottus snyderi Greeley, is a common species which inhabits the rocky intertidal from Baja California to Sitka, AK (Miller and Lea 1972). Between central California and British Col- umbia, 0. snyderi frequently is very abundant (Green 1971; Cross 1981; Yoshiyama 1981; Grossman 1982). This species occurs primarily in mid- and lower in- tertidal areas (Green 1971; Yoshiyama 1981), and often is associated with surfgrass (Green 1971; Nakamura 1976a). The general absence of this species from the high intertidal is probably due to its inability to tolerate higher temperatures which frequently occur in high intertidal pools (Nakamura 1976b). Life histories of intertidal fishes, particularly cot- tids, are poorly known (Gibson 1969, 1982). Publish- ed information on the demography of 0. snyderi is restricted to the work of Moring (1981), who ex- amined age structure of a northern California 0. snyderi population, and Grossman and deVlaming (1984), who described the species' reproductive ecology. This paper presents data on age structure, growth rates, and dietary habits of an 0. snyderi population at Dillon Beach, CA, a site that is sub- ject to pronounced annual cycles of oceanic produc- tivity (Parrish et al. 1981). Oligocottus snyderi numerically dominates the intertidal fish assemblage at Dillon Beach (Grossman 1982); the present study explores demographic and ecological characteristics which may account for this species' ecological suc- cess in the rocky intertidal. MATERIALS AND METHODS Collections Oligocottus snyderi were collected from a series of mid- and lower intertidal pools at Dillon Beach, CA, on 15 dates from January 1979 to July 1981 (see Grossman in press a for sampling dates). Repeated collecting did not affect assemblage structure (Gross- man 1982, in press a). Fish were obtained by spreading a 10% solution of quinaldine in isopropanol through the pools and then collecting individuals after anesthetization. Over 1,400 0. snyderi were collected. Specimens were preserved in buffered Formalin^ and were later washed and transferred to 45% isopropanol. Individuals were measured to the nearest millimeter standard length (SL) and weighed to the nearest 0.1 g. Sexes of all 1 School of Forest Resources, University of Georgia, Athens, GA 30602. ^Department of Wildlife and Fisheries Biology, University of California, Davis, CA 95616. ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 645 FISHERY BULLETIN: VOL. 83, NO. 4 individuals collected during 1979 and of specimens used for gut analysis also were recorded. A complete description of collecting methods and the study site are presented in Grossman (1982, in press b). Collection dates were assigned either to the Up- welling or the combined Oceanic-Davidson Current (ODC) oceanographic seasons, based on water tem- peratui'e and a calculated upwelling index (see Gross- man in press b). The Upwelling period is character- ized by frequent upwelling of nutrient-rich waters and high productivity (Bolin and Abbott 1963; Gar- rison 1979). In contrast, the ODC period exhibits minimal upwelling, frequent downwelling, and low productivity (Bolin and Abbott 1963; Garrison 1979). Data from individuals collected during the latter seasons were pooled because these periods were not readily distinguishable (Grossman in press b). Con- sequently, comparisons of 0. snyderi food habits and length-weight relationships represent contrasts between a high productivity period (Upwelling, all months from April to August, plus March 1980) and the low productivity ODC period (October-February, plus March 1979). Monthly mean water temperatures ranged from 9.7° to 11.9°C during the Upwelling period from 11.2° to 14.2° C during the ODC period (Grossman in press b). Length-Weight and Growth Calculations Regressions of individual weight versus standard length were calculated separately for males and females collected in the Upwelling and ODC periods. Regression equations were of the form W = aL'', fitted as In M^ = In a + 6 In L. Slopes (b) of regres- sion equations for males and females were compared for each period by analysis of covariance (ANCOVA). Seasonal differences between slopes also were ex- amined for each sex. Regression slopes for the two sexes and periods were divided by their correlation coefficients to estimate slopes for geometric mean functional regressions of weight versus length (Ricker 1973). Functional regressions are recommended for describing relations between weight and length because both variables are subject to natural variability (Ricker 1973). Functional regression slopes were used to derive ^/-intercepts from the equation y-int. = y - {b/r)x (Ricker 1973). These slope estimates were employed to calculate instan- taneous rates of growth in weight (G) as G = (blr) (In L2 - In Lj), (Ricker 1975), where b/r = slope of the functional length-weight 646 regression; Lj = mean length at the time t; and L2 = mean length at time ^ -1- 1. Growth rates were calculated for individual year classes which were separated using length-frequency distributions. lb compare growth rates for males and females, ^tests for paired samples were used. Sex ratios for the 1979 samples were tested for deviations from unity using chi-square tests cor- rected for continuity. Dietary Analyses Stomach contents were removed from a subsam- ple (597 total) of the 0. snyderi collected between January 1979 and July 1981. Ninety-seven percent of stomachs examined contained food. Prey were identified to the lowest possible taxon and weighed (wet) to the nearest milligram. The weights of each prey type were pooled separately for males and females from each collection. Diets were then calcu- lated as the percent of the total stomach content weight attributable to each prey typa The number of fish containing each prey type (i.a, prey frequen- cy of occurrence, henceforth FO) also was recorded. Dietary data were pooled across collections wdthin the Upwelling and ODC periods for each sex, to com- pare feeding habits between sexes and seasons. Because the study period spanned 3 yr, it also was possible to quantify among year variability in seasonal and sex-specific feeding habits. Sex-linked ontogenetic dietary changes were examined by separating males and females into 20 mm size classes and comparing prey consumed by each size class in different seasons and years. Niche breadth was calculated as I l/p,;^ (Levins 1968), where p, is the proportion of the total prey weight comprised of the ith prey type. Only prey types constituting at least 1% of the total prey weight were included in niche breadth calculations. Niche breadth was compared among size classes by using the Wilcoxon signed- rank test. RESULTS Length-Weight Relationships The slope for the male length-weight regression for the Upwelling period was significantly greater than that for females (ANCOVA, F = 12.875, P < 0.001; T^ble 1). Slopes of length-weight regressions for males and females collected during the ODC period were not significantly different (ANCOVA, FREEMAN ET AL.: LIFE HISTORY OF FLUFFY SCULPIN Table 1.— Length-weight relationships, described as W^ = a/.'', for Oligocottus snyderi. Coefficients for the least-squares (In W = \r\ a + bin L) and functional (In IV = In a' -h (bir) In L) regressions are presented, as well as the correlation coefficients for the least- squares regressions (r) and the number of individuals used in each regression (A/). ODC = Oceanic-Davidson Current. Least-squares regression Functional regression N r a b a' bIr Males Upwelling 150 0.984 0.0000178 3.097 0.0000146 3.147 ODC 112 0.973 0.0000253 2.991 0.0000182 3.075 Females Upwelling 164 0.972 0.0000472 2.835 0.0000346 2.917 ODC 131 0.976 0.0000202 3.069 0.0000151 3.143 F = 0.736, P > 0.25). Seasonal comparisons for each sex showed that tlie male Upwelling and ODC slopes were not significantly different (ANCOVA, F = 1.483, P > 0.10), whereas the females had a significantly higher slope during the ODC period (ANCOVA, F = 6.147, P < 0.025). Growth Rates Length-frequency histograms for 0. snyderi at Dillon Beach indicated that two year classes were present on most dates (Fig. 1). Recruitment began in spring and peaked during summer. The onset of recruitment differed between years. Age O-i- fish were first taken in May of 1979; however, in 1980 recruitment began in March. Age O-i- individuals of the 1979 cohort grew 20 mm in length (from 20-25 to 40-55 mm SL) by December of their first year. Members of the 1978 cohort (= age l-i- fish) in- creased in length from 40-55 to 60-70 mm SL dur- ing spring, summer, and fall 1979. Individuals did not appear to survive a second winter during 1980, although a few males recruited in 1977 may have sur- vived until spring of 1979 (Fig. 1). Sex ratios for the O-i- age class were significantly different from unity in the August 1979 collection in which there were significantly more males than females (1.7:1; x^ = 7.32, P < 0.01). Age 1+ females significantly outnumbered males in January 1979 (4:1; x^ = 6.05, P < 0.025), July 1979 (1.8:1; x^ = 4.38, P < 0.05), and December 1979 (3:1; x^ = 5.04, P < 0.025). Instantaneous growth rates were calculated for males and females from the two year classes pres- ent in 1979 (Tkble 2). Prolonged recruitment (lasting from May through August) resulted in apparent depressed spring and summer growth rates for the age O-I- class when all individuals were included in the calculations of mean lengths (Tkble 2). Therefore, fish which appeared to have been recruited later than the majority of the class were excluded from con- sideration in the June, July, and August calculations, as shown in Figure 1. Only the very large males col- lected in January 1979 were excluded from the age 1 + mean length calculations, as these individuals ap- parently were survivors from the 1977 year class and probably died shortly thereafter. The highest monthly instantaneous growth rates were obtained during May and June, averaging 0.860 and 0.655 for age O-I- males and females, and 0.209 and 0.188 for age 1 -I- males and females (Tkble 2, Fig. 2). Growth rates markedly decreased after August, averaging 0.065 and 0.169 for age O-i- males and females, and 0 and 0.061 for age l-i- males and females during October and November (Ikble 2, Fig. 2). Males and females attained nearly equal lengths in the first season of growth; however, age 1 -i- males displayed significantly greater mean lengths (^tests, P « 0.05) than age 1 + females from June through November 1979 (Fig. 2). Growth rates were not significantly higher for males than females in either size class when compared from January to December 1979 (age 0 + :^ = 1.002, P> 0.1; age 1 + : t = 1.232, P > 0.1); however, if only the data from January to October are included for the age 1 + class, males did have significantly higher growth rates (t = 2.255, P < 0.05). This discrepancy is caused by the low numbers of age 1 -i- fish collected in November (14) and December (24). Length-frequency histo- grams (Fig. 1) also show that age l-i- males are larger than females, whereas length distributions are Table 2. — Instantaneous growth rates of Oligocottus snyderi males and females in their first (age O-i-) and second (age 1-t-) years of growth. Rates were calculated for the intervals between the 1979 collections. Values in parentheses were calculated with all in- dividuals included in the age O-i- cohort (see text). Instantaneous growth rate Collection dates Age O-I- Age 1 + Males Females Males Females 1-27-79 — — 0.257 0.079 3-26-79 — — 0.295 0.171 5-16-79 1.059(1.059) 0.718 (0.562) 0.318 0.204 6-13-79 0.661 (0.429) 0.591 (0.358) 0.099 0.171 7-12-79 0.352 (0.306) 0.295 (0.390) 0.111 0.074 8-09-79 0.303 (0.582) 0.124 (0.419) 0.164 0.063 10-07-79 0.129 0.149 0 0.122 11-04-79 — 0.188 — — 12-04-79 647 FISHERY BULLETIN: VOL. 83, NO. 4 o -a cz E Figure L— Length-frequency distributions for Oligocottus snyderi. Males and females are separated in the 1979 collections, with females shown below the axis. Arrows indicate divisions between year classes 0 + and 1 + , and broken lines designate limits for inclusion in cohort mean SL calculations (January, June, July, and August 1979). The 1980 and 1981 collections are included for comparison, but the sexes are not shown separately. Jan. 79 n=23. ^ ^ Mar. 79 2- n=33 . 2- * May '79 21- 2 J^ I i t 6 - n=49 t fi Mar.'SO ^•^ n=32 ^ Apr. 80 6- n="" 2 - 30 I ■ I Hi ■■ 6r Jun.79 Jun.'SO n=98 Jul. 79 n=l45. ,0 Nov 80 2 Aug. 79 n=232 Jul.'8l ^ Nov. 79 2r n=3l 10 30 50 70 2- •^rr l^yp HI Dec. 79 I 1 1 1 1 1 r 10 30 50 70 Standard Length (nnnn) 648 FREEMAN ET AL.: LIFE HISTORY OF FLUFFY SCULPIN 70 £ 60 01 c 50 - 40 - o -o 30 c o tr> 20 I I I I _LJ I ^6^ ^51 ^ "b <^. 0.1; weight: t = 1.036, P > 0.1), females had significantly greater amounts of food in their stomachs (t = 4.857, P = 0.0001, Ikble 4). This dif- ference is attributable to the weight of ingested gam- marids because 131 (FO) females contained a total of 8.4 g of gammarids, whereas 132 males contain- ed only 4.3 g (total weight). Examination of food habits across size classes shows that larger individuals generally possessed a more diverse diet than smaller individuals (Figs. 3, 4). Dietary breadth was significantly higher (Wilcox- on signed-rank test, n = 12, T = 69, P < 0.01) for 50-69 mm SL fish than for 30-49 mm SL individuals when compared across both sexes for all years. In- dividuals <30 mm SL consumed mainly gammarids and isopods; polychaetes and shrimp were major prey only for larger fish. Pagurus spp. and crabs, in- cluding Cancer spp. and Pachycheles spp., were found only in individuals >40 mm SL. The general observation of decreased gammarid consumption by larger 0. snyderi, concomitant with increased polychaete and shrimp consumption, ap- plies to collections made throughout the 3-yr study period. The majority of among year variation in food habits resulted from the presence of large, rare prey items in single individuals (Figs. 3, 4). An exception was the high consumption of polychaetes by males collected in July 1981 (Fig. 3). In this case, poly- chaetes occurred in all of the 50-69 mm fish and in 56% of the 30-49 mm fish, and composed a con- siderably higher proportion of the diet than during previous Upwelling collections. DISCUSSION Age-Structure and Growth At Dillon Beach, a habitat displaying marked seasonal fluctuations in productivity (Grossman 1982, in press b; Grossman and deVlaming 1984), 0. snyderi is short lived, with a maximum lifespan of about 1.5 yr and a maximum length of about 70 mm SL. Whereas two age classes were observed in this study, Moring (1981) observed as many as three age classes in Trinidad Bay. Standard lengths attained 649 FISHERY BULLETIN: VOL. 83, NO. 4 Table 3.— Food habits of male and female Oligocottus snyderi collected during the Upwelling period. Values shown for each prey type are total weight ingested, percent of total prey weight, and frequency of occurrence (FO). Females, n = 164 Males, n = 150 y 80 mm SL at Port Renfrew, British Colum- bia. These results indicate that 0. snyderi may have a longer lifespan at more northerly locations. The scarcity of individuals >70 mm SL at Dillon Beach probably did not result from repeated sampling because such large individuals were absent from the initial samples and from collections made in previously unsampled pools near the study site in August 1979, December 1979, and April 1980. There was no discernable pattern in sex-ratio deviations from unity observed in four 1979 collections, which 650 FREEMAN ET AL.: LIFE HISTORY OF FLUFFY SCULPIN < c o © 20-29mm n=7 Females Upwelling Period 30-49mm n=47 50-69mm n=59 cj> < -3 I 20-29mm o n=3 Males 30-49mm n=2l 50-69mm n=68 O 00 d 3 ■^ wT Q. < 20-29mm n = 4 30-49mm n=l6 50-69mm n = ll O _oo d 3 20-29mm n=5 30-49mm n=22 50-69mm n = ll -» io« 00 3 -3 IO-29mm 30-39mm n=6 n=5 50-69mm n = 9 CO 3 -3 30-49nnm n=9 SnxL-PRS 50-69mm n = ll Figure 3.— Sex-specific ontogenetic and among year variation in Oligocottus snyderi feeding habits during the 1979, 1980, and 1981 Upwell- ing periods. Prey are represented as percent total prey weight, and only prey composing >1% of the total prey weight are shown. Asterisks indicate prey which occurred in only one individual. Prey abbreviations are G (gammarids); PR (hard-bodied prey remains); Po (polychaetes); Is (isopods); S (shrimps, all spp.); Ca (caprellids); Sn (snails); Id (Idotea spp.); Pa (Pagurus spp.); Cr (crabs, all spp.); Al (algae); and BC (barnacle cirri). suggests that males and females have about equal lifespans at Dillon Beach. Oligocottus maculosus Girard, another abundant intertidal cottid (Nakamura 1976a, b; Moring 1981), was estimated by Chadwick (1976) to survive as many as five growing seasons, based on counts of vertebral rings, at Port Renfrew, B.C., and Bruels Point, CA. Annuli were not validated in this study, however, and it is questionable if the rings observed were actually formed once yearly. In addition, Chad- wick's (1976) growth rate estimates are extremely low, suggesting that 0. maculosiis grew no more than 7 mm in any season after the first. Moreover, all previous studies of both 0. snyderi (Moring 1981) and 0. maculosus (Atkinson 1939; Green 1971; Mor- ing 1979) showed that populations of these species were composed primarily of age 0-t- and l-i- in- dividuals, as was 0. snyderi at Dillon Beach. Conse- quently, it appears unlikely that 0. maculosus sur- vives to age 5-f as suggested by Chadwick (1976), and in any case such longevity was not observed for 0. snyderi at Dillon Beach. Intertidal fishes appear to display several distinct life history patterns. Short lifespan (1-3 yr), early maturation, and high reproductive effort have been reported for several intertidal gobies in northern temperate habitats (Gibson 1969, 1982; Grossman 1979; Miller 1979). In contrast, lifespans of 4 to over 10 yr, accompanied by delayed maturation, have been observed in many intertidal species, including gobiids and blenniids (Gibson 1969, 1982; Stephens et al. 1970; Grossman 1979; Miller 1979). Little informa- tion is available for intertidal cottids. The largest North American cottid, Scorpaenichthys mar- moratus Ayers, may live 13 yr, but only inhabits tide- pools during its first or second year of life (O'Con- nell 1953; Grossman and Freeman unpubl. data). Another large sculpin, Leptocottus armatvs Girard, common in the Pacific coast bays and estuaries, is known to live to age 3 and reaches sexual maturity 651 FISHERY BULLETIN: VOL. 83, NO. 4 (7> c a cn Q >■ o _co o" _CD O o O Oceanic-Davidson Females 30-49 mm n = 2l PRfls 30-49mm n=20 30-49mm n=27 Id D^» r« 50-59mm n=l3 Current Periods Males 40-49mm n=6 50-69mm n=45 50-69mm n=5 50-69mm n=6 o 30-49mm ^ n=24 50-69mm n=38 ^ 30 -49 mm S n=40 50-69mm n=20 70-79mm n=4 Figure 4.— Sex-specific ontogenetic and among year variation in Oligocottus snyderi feeding habits during the three ODC (Oceanic-Davidson Current) periods encompassed by this study. Prey are represented as percent total prey weight; only prey composing >1% total prey weight are shown. Asterisks indicate prey occurring in only one individual. Abbreviations are as in Figure 3, with the addition of E (eggs); An (sea anemones); and Bi (bivalves). after 1 yr (Jones 1962; Tksto 1975). Oligocottiis snyderi and 0. maculosus apparently are best characterized by the short hfespan, early matura- tion, life history pattern. This conclusion is based on growth rate data coupled with the scarcity of in- dividuals older than 1.5 yr, and data showing early maturation and high reproductive effort for 0. snyderi (deVlaming et al. 1982). The majority of age 0+ 0. snyderi at Dillon Beach attained sufficient size to spawn during their first year; it is less certain what proportion of these in- dividuals survive to spawn in their second year. Grossman and deVlaming (1984) indicated that, at Dillon Beach, females mature at about 40 mm SL and contain vitellogenic oocytes from October through May. These females also probably spawn more than once, primarily during winter and spring (Grossman and deVlaming 1984). Length-frequency distributions from March and May 1979 and March and April 1980 show that a single age class dominated the population. Hence, assuming that recruitment primarily was derived from this popula- tion, age 1+ individuals (recruited the previous spring and summer) must have been responsible for nearly all spring spawning. The paucity of large in- dividuals (i.e., >60 mm SL) in winter or spring samples indicates that if individuals spawned in their second year, this reproduction must have occurred prior to January. Substantial numbers of age 1 -i- in- dividuals were captured from October to December 1979 and in November 1980. Although the earliest recruitment observed at Dillon Beach occurred in March 1980, Moring (1981) captured newly recruited individuals as early as January. This suggests that fall spawning may occur in more northerly popula- tions of 0. snyderi. Reproduction apparently is timed to insure that larvae metamorphose during upwelling. Grossman (1982) found a significant correlation between up- welling activity and the number of resident species with young-of-the-year present at Dillon Beach. Grossman and deVlaming (1984) also observed that 0. snyderi recruitment was strongly correlated with productivity; the authors pointed out that the early 652 FREEMAN ET AL.: LIFE HISTORY OF FLUFFY SCULPIN recruitment observed in March 1980 coincided with the onset of upwelling, whereas both recruitment and upwelHng were delayed until May 1979. If some individuals do survive to reproduce dur- ing their second year, the larger sizes attained by males during their second growing season may be advantageous during spawning. This is because male 0. snyderi have a lengthened and prehensile first anal ray with which they clasp females during copulation, and the larger a male is in relation to a female, the more efficiently he will be able to clasp her (Morris 1956). Dietary Habits Oligocotttcs snyderi at Dillon Beach consumed primarily gammarid amphipods and polychaetcs. These data are consistent with previous observations (Johnston 1954; Nakamura 1971; Yoshiyama 1980), although they differ somewhat from results obtain- ed by Cross (1981) from two sites in northern Wash- ington. Cross (1981) reported that harpacticoid copepods were a major prey for 0. snyderi at one site; similarly, polychaetes also were consumed only at one site However, Cross (1981) also observed high gammarid consumption by 0. snyderi, as well as by most other intertidal species he studied. In addition, gammarids have been cited as a major prey in other intertidal fish assemblages (Mitchell 1953; Johnston 1954; Zander 1979, 1982; Grossman in press b). At Dillon Beach, gammarids frequently were consum- ed by resident and seasonal intertidal fishes, although only one other resident (Apodichthys flavidus Girard) possessed a diet dominated by gam- marids (Grossman in press b). Among year, seasonal comparisons, however, show that a variety of minor prey types also are consumed by 0. snyderi. Combinations of shrimps, crabs, hermit crabs, Idotea, and other irregularly consumed prey con- stituted a considerable proportion of 0. snyderi's diet throughout the year. This was particularly evident in larger fish, which suggests that capture of these prey is either more difficult for small fish due to mor- phological, physiological, or behavioral constraints, or involves increased predation risk. Similarly, a shift from gammarids to larger prey concomitant with in- creasing length was observed for the majority of intertidal fishes at Dillon Beach (Grossman in press b), and for some species studied by Cross (1981). In contrast, Yoshiyama (1980) was unable to detect dietary differences between small and large 0. snyderi, or two other intertidal cottids. Yoshiyama pooled small samples collected throughout a year, however, which may have obscured seasonal changes in prey consumption. Because his samples were small, Yoshiyama also may have underestimated con- sumption of rare prey. Although seasonal dietary changes were minor, there is evidence that females possessed higher in- gestion rates than males during the ODC period. During this season of lowered productivity, females consumed two times more gammarids (by weight) than males, although mean fish length and gam- marid frequency of occurrence were nearly identical between sexes. This difference between consumption rates may not be artifactual. Females develop and carry vitellogenic eggs during this period (Grossman and deVlaming 1984), and consequently have high energy demands. There was no evidence that inter- sexual or ontogenetic dietary differences resulted from differential distribution or collection disturbance The high prey weight observed in males and females collected during the ODC period is of in- terest in light of the low growth rates observed dur- ing winter months. Moring (1979, 1981) also ob- served cessation of growth during winter for both 0. snyderi and 0. maculosus in northern California. He suggested that reduced foraging activity caused by increased wave action during winter might par- tially explain this growth reduction. Dietary data from Dillon Beach do not support this conclusion because ODC specimens contained a greater total weight of prey than individuals collected during Upwelling, even though a greater number of in- dividuals were examined during the latter period. The Dillon Beach study site probably is more sheltered from winter storm activity, however, than the Trinidad Bay sites observed by Moring (Gross- man pers. obs.). In the absence of ingestion rate data for both seaons, results based on gut content weight alone are equivocal. Moring (1981) also suggested that gonadal development during the winter months might be responsible for reduced growth; this could also apply to 0. snyderi at Dillon Beach. In conclusion, 0. snyderi the most abundant inter- tidal cottid at Dillon Beach, possesses a suite of characteristics which suggest that productivity has influenced the biology and behavior of this species. For example, 0. snyderi reaches sexual maturity dur- ing its first year of life, and spawns at a time which enables recruitment to take place during the season of highest productivity (i.e, Upwelling (Grossman and deVlaming 1984)). Grossman (1982, in press b) presented strong evidence that productivity cycles also affect many other species at Dillon Beach, because this assemblage appears to be organized through interspecific exploitative competition for 653 FISHERY BULLETIN: VOL. 83, NO. 4 food. Recuitment and numerical abundances within this assemblage also were strongly correlated with productivity (Grossman 1982). Oligocottus snyderi possesses a variety of adaptations (ag., rapid matura- tion, high female reproductive effort, utilization of an abundant prey not widely consumed by other assemblage members) which probably are responsi- ble for its numerical dominance in a fluctuating environment. ACKNOWLEDGMENTS We thank the many friends and colleagues who aided in the collection and analysis of data. Jeff Bar- rett, Dan Erickson, Doug Facey, Margi Flood, Steve Floyd, and Joe Hightower reviewed the manuscript and their comments are greatly appreciated. We also appreciate the help of Bonnie Fancher who typed various drafts of this manuscript. LITERATURE CITED Atkinson, C. E. 1939. Notes on the life history of the tidepool johnny (Oligocot- tus maculosus). Copeia 1939:23-30. BoLiN, R. L., AND D. P. Abbott. 1963. Studies on the marine chmate and phytoplanliton of the central coastal area of California, 1954-1960. Calif. Coop. Oceanic Fish. Invest. Rep. 9:23-45. Chadwick, E. M. R 1976. A comparison of growth and abundance for tidal pool fishes in California and British Columbia. J. Fish Biol. 8:27-34. Cross, J. N. 1981. Structure of rocky intertidal fish assemblage Ph.D. Thesis, Univ. Washington, Seattle, 259 p. deVlaming, v., G. D. Grossman, and F. Chapman. 1982. On the use of the gonosomatic index. Comp. Biochem. Physiol. 73A:31-39. Garrison, D. L. 1979. Monterey Bay phytoplankton. I. Seasonal cycles of phytoplankton assemblages. J. Plankton Res. 1:241-265. Gibson, R. N. 1969. The biology and behavior of littoral fish. Oceanogr. Mar. Biol. Annu. Rev 7:367-410. 1982. Recent studies on the biology of intertidal fishes. Oceanogr. Mar. Biol. Annu. Rev 20:363-414. Green, J. M. 1971. Local distribution of Oligocottus maculosus Girard and other tidepool cottids of the west coast of Vancouver Island, British Columbia. Can. J. Zool. 49:1111-1128. Grossman, G. D. 1979. Demographic characteristics of an intertidal bay goby (Lepidogobius lepidus). Environ. Biol. Fishes 4:207-218. 1982. Dynamics and organization of a rocky intertidal fish assemblage: the persistence and resilience of taxocene struc- ture Am. Nat. 119:611-637. In press a. Long term persistence in a rocky intertidal fish assemblage Environ. Biol. Fishes. In press b. Division of food resources among fish of the rocky intertidal zone J. Zool. (Lond.). Grossman, G. D., and V. deVlaming. 1984. Reproductive ecology of female Oligocottus snyderi Greeley: a North American intertidal sculpin. J. Fish Biol. 25:231-240. Johnston, R. F. 1954. The summer food of some intertidal fishes of Monterey County California. Calif. Fish Game 40:65-68. Jones, A. C. 1962. The biology of the euryhaline fish Leptocottvs anmatus armatus Girard (Cottidae). Univ. Calif, Publ. Zool. 67: 321-367. Levins, R. 1968. Evolution in changing environments. Princeton Univ. Press, Princeton, 120 p. Miller, D. J., and R. N. Lea. 1972. Guide to the coastal marine fishes of California. Calif. Dep. Fish Game, Fish. Bull. 157, 235 p. Miller, P. J. 1979. Adaptiveness and implications of small size in teleosts. In P. J. Miller (editor), Fish phenology: anabolic adaptiveness in teleosts, p. 263-306. Zool. Soe Lond. Symp. 44. Mitchell, D. F. 1953. An analysis of stomach contents of California tidepool fishes. Am. Midi. Nat. 49:862-871. Moring, J. R. 1979. Age structure of a tidepool sculpin, Oligocottus macu- losus, population in northern California. Calif. Fish Game 65:111-113. 1981. Seasonal changes in a population of the fluffy sculpin, Oligocottus snyderi, from Trinidad Bay, California. Calif Fish Game 67:250-253. Morris, R. W. 1956. Clasping mechnisms of the cottid fish Oligocottus snyderi Greeley Pac Sci. 10:314-317. Nakamura, R. 1971. Food of two cohabiting tide-pool Cottidae J. Fish. Res. Board Can. 28:928-932. 1976a. Experimental assessment of factors influencing micro- habitat selection by two tidepool fishes Oligocottus maculosus and 0. snyderi. Mar. Biol. (Berl.) 37:97-104. 1976b. Tfemperature and the vertical distribution of two tide- pool fishes (Oligocottus maculosus, 0. snyderi). Copeia 1976: 143-152. O'CONNELL, C. R 1953. Life history of the cabezon Scorpaenichthys marmora- tus (Ayres). Calif Dep. Fish Game, Fish. Bull. 93, 76 p. Parrish, R. H., C. S. Nelson, and A. Bakun. 1981. Transport mechanisms and reproductive success of fishes in the California Current. Biol. Oceanogr. 1:175-203. Ricker, W. E. 1973. Linear regressions in fishery research. J. Fish. Res. Board Can. 30:409-434. 1975. Computation and interpretation of biological statistics of fish populations. Fish. Res. Board Can., Bull. 191, 382 p. Stephens, J. S., Jr., R. K. Johnson, G. S. Key, and J. E. McCOSKER. 1970. The comparative ecology of three sympatric species of California blennies of the genus Hypsoblen nius Gill (Tteleo- stomi, Blenniidae). Ecol. Monogr. 40:213-233. Tasto, R. N. 1976. Aspects of the biology of the Pacific staghorn sculpin, Leptocottus armatus Girard, in Anaheim Bay. Calif Dep. Fish Game, Fish. Bull. 165:123-135. YOSHIYAMA, R. M. 1980. Food habits of three species of rocky intertidal sculpins (Cottidae) in central California. Copeia 1980:515-525. 654 FREEMAN ET AL.: LIFE HISTORY OF FLUFFY SCULPIN 1981. Distribution and abundance patterns of rocky intertidal stony ground off M0n, Denmark. Ophelia 18:179-190. fishes in central California. Environ. Biol. Fishes 6:315-332. 1982. Feeding ecology of littoral gobiid and blennioid fish of Zander, C. D. the Banyuls area (Mediterranean Sea). I. Main food and 1979. On the biology and food of small-sized fish from the trophic dimension of niche and ecotopa Vie Milieu 32:1-10. North and Baltic Sea areas. II. Investigations of a shallow 655 VARIABILITY, TRENDS, AND BIASES IN REPRODUCTIVE RATES OF SPOTTED DOLPHINS, STENELLA ATTENUATA Jay Barlowi ABSTRACT Tfemporal changes were examined in three parameters that affect reproduction of spotted dolphin popula- tions in the eastern Pacific Of mature females, percent pregnant decreased markedly from the period 1971-73 to the period 1974-83. Within the period 1974-83, percent pregnant remained relatively constant. Of pregnant females, percent lactating increased during the period 1971-83. The percentage of sexually mature females did not change Potential biases in the measurement of the three parameters were iden- tified by examining the effects of sampling conditions. The percentage of mature females that are preg- nant and the percentage of pregnant females that are lactating were found to be robust to sampling con- ditions. The percentage of mature females in a sample was found to depend significantly on the number of dolphins killed per set, and annual variability was too large to be explained by random sampling error. Comparisons between two populations show that the more exploited population has a lower percent preg- nant, although the opposite might be expected from density compensatory effects. Percent lactating and percent immature were higher in the more exploited population. Changes in the reproductive parameters of cetacean populations can be used to make inferences about the status or general "health" of a population. For instance, increases in pregnancy rates and decreases in the age at attainment of sexual maturity were link- ed to reductions in Antarctic whale populations (Gambell 1975). Re-analysis of these data, however, revealed unsuspected biases, and Gambell's results are now being questioned (Mizroch 1983). The pur- pose of this paper is to examine potential biases in measuring reproductive rates of spotted dolphins, Stenella attenuata. This species is taken incidentally in the tuna purse seine fishery in the eastern tropical Pacific (Smith 1983). The intent is to determine whether reproductive rates can be measured with sufficient precision to monitor intrapopulation changes or to make interpopulation comparisons. Previous studies of female reproduction in spot- ted dolphins of the eastern Pacific have shown an apparent decrease in pregnancy rates from 1973 to 1975 (Perrin et al. 1977), from 1973 to 1978,^ and from 1971 to 1978 (Hester 1984). Hester (1984) sug- gested that this decline in pregnancy rates is related to the decline in fishing-related dolphin mortality during the same time period. 'Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, La Jolla, CA 92038. ^Henderson, J. R., W. F Perrin, and R. B. Miller. 1980. Rate of gross annual reproduction in dolphin populations {Stenella spp. and Delphinics delphis) in the eastern tropical Pacific, 1973-78. Admin. Rep. LJ-80-02, 51 p.; available from Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. Three indices of the reproductive status of female spotted dolphins are examined in the present paper: 1) the fraction of sexually mature individuals that are pregnant, 2) the fraction of pregnant females that are lactating, and 3) the fraction of females that are sexually mature These measures have been used previously in calculating what has been termed the gross annual reproductive rate (GARR) of spotted dolphins (Perrin et al. 1976). This paper reexamines data from 1971 to 1978 plus additional data from 1979 to 1983 to determine whether the previously noted trends in reproductive rates are real, and if so, whether they are continuing. Also, factors are ex- amined which may be biasing estimates of reproduc- tive rates and which could be causing spurious changes in apparent pregnancy rates and GARR. Finally, differences in these reproductive indices be- tween two geographic stocks of spotted dolphin are discussed in view of their different histories of in- cidental fishing mortality. MATERIALS Reproductive data were collected from a sample of the dolphins killed in tuna purse-seining opera- tions in the eastern tropical Pacific (ETP). Three stocks of spotted dolphins are recognized in this area based on morphological differences (Perrin et al. 1979). Samples considered here include two of these: the northern offshore stock which has been subject to tuna fishing since 1959 and the southern offshore stock which has been subject to exploitation since Manuscript accepted January 1985. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 657 FISHERY BULLETIN: VOL. 83, NO. 4 the early 1970's. In 1971 and 1972, field technicians collected samples predominantly from females with the "adult" or fused color pattern (Perrin 1970). Beginning in 1973, field technicians were instructed to collect samples nonselectively with respect to size and sex. Operationally, this meant working-up speci- mens in the order in which they appeared on the deck of the tuna vessel. Sampling methods and laboratory procedures are described in detail by Per- rin et al. (1976). Sample locations are shown in Figure 1. Reproductive tracts of mature and nearly mature females were preserved in the field for laboratory examination. In 1971 and 1972, the definition of "mature and nearly mature" was not explicit. In 1973, "mature and nearly mature" was defined as individuals with "mottled" or "fused" developmen- tal color phases (Perrin 1970). Because females in the younger "speckled" color phase occasionally were found to be pregnant, "mature and nearly mature" was redefined operationally (beginning in 1974) as specimens >150 cm total length (TL, measured from tip of rostrum to fluke notch). Laboratory examina- tion of preserved ovaries was used to determine the presence of corpora from past ovulations. Pregnancy was determined by visual examination of the uterus (in later years, fetuses >30 cm TL were removed and measured in the field). Mammary glands were slit and checked in the field for the presence of milk. In addition to the above life history information, field technicians collected data pertaining to condi- tions under which the samples were taken. Informa- tion used in this report includes the observer's estimate of the size of the school from which the sample was taken, the duration of the chase before the net was set, the number of dolphins known to be killed during fishing operations, and the geographic location at which the sample was taken. METHODS Three indices of female reproduction are con- sidered in this paper: the percent pregnant, the per- cent lactating, and the percent mature. Tbmporal trends in these three indices were examined by regressing annual means against year (weighting by the inverse of binomial variances). In calculating the percentages of mature females that were pregnant and that were lactating, specimens were used only if both ovaries were col- lected and if at least one corpus of ovulation (cor- pus albicans or corpus luteum) was present. Previously, 1971 and 1972 samples were excluded from calculation of percent pregnant because of undersampling of younger females with a mottled color pattern (Perrin et al. 1977). This was not deem- ed necessary in this study, because in 1973-83 samples the percent pregnant for mature mottled females (31.1%) was essentially the same as that for mature females with a fused color pattern (31.4%). In calculating the percentage of females that were sexually mature, two different criteria were used for determining maturity. In the majority of cases both ovaries were examined, and the presence of one (or more) corpus of ovulation was taken as evidence of sexual maturity. If ovaries were not collected or ex- amined (which was true for about 30% of females over 150 cm TL), a length criterion was used for maturity. Samples from 1971 and 1972 were ex- cluded from these analyses because sampling was not random in those years. i "Length at attainment of sexual maturity" was determined by the method used by Perrin et al. (1977). Based on the sample for which ovaries were examined, this length was estimated as the length at which the number of longer immature individuals equals the number of shorter mature individuals. For the northern stock, the length at the onset of sex- ual maturity was determined independently for each year 1974-83 (176.5, 177.5, 177.0, 177.0, 178.0, 177.5, 179.0, 178.5, 180.0, and 182.0 cm, respectively). In 1973 the decision to collect ovaries was not based on specimen length. The apparent trend in these data yields a significant regression (P = 0.0008); hence, regression estimates were substituted for an- nual estimates for 1974-83, with an extrapolation to 1973. These values were 175.6, 176.1, 176.6, 177.1, 177.6, 178.0, 178.5, 179.0, 179.5, 180.0, and 180.5 cm, respectively, for 1973-83. For the southern stock, in- sufficient data exist to calculate a length at attain- ment of sexual maturity for individual years, hence the collective value was used for all years (175.0 cm). Six factors were examined to determine whether annual changes in the above percentages of pregnant females were caused or affected by changing biases in the sampling methods. These factors include 1) geographical provenance, based on two strata (Fig. 1) which roughly correspond to the historical tuna fishing grounds (inside the Commission Yellowfin Regulatory Area, CYRA) and a more recently ex- ploited area (outside the CYRA); 2) the quarter of the year; 3) the length of chase, or the time between sighting the dolphin school and capture (net set); 4) the observer's estimate of the dolphin school size (only available since 1973); 5) the number of dolphins known to have been killed in the set; and 6) the total number of tons of tuna caught in the set. The selection of these six factors was guided to 658 BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS N „S£ N ,0£ N „S2 N „02 N „SI N „0I N „S S ? S „0I S „SI S n2 3aniiiun 3 his] ter, in rs. ce M 2 (U 3 -5 -^ O 00 ^.22 '^ 1^ T3 * c § 2 *l o Ct. s LT) a o OO 03 M S r 1^ Z to o a 35^ S .S 2 JS "" a. (u 3 o ^ -O at IT) C35 "^ 'i a " P (M :z -b 9 o O 00 O S '-^ O 1-^ 3 ^'^ .a c in o tl &,< o «J n^ 2 a,B^ o O o|§ 1 £< — • S -2 <^ 3 cS S .f t^ aj s i2 ^ 00 t3 . — E-. 2.S^ 3 C£) o 2 o O f^ _J 3 ■K i'^ o ;sc2s^ •-* -73 ^ =« a) o , - 3 ^ ;S :S O to u 5 ^ ^ «. 13 c ^ 3 ^ fe d" ^ Sfi 'E g ^ a) ^ -2 S n! j5 o 3 o in x: a> ^ •— « c« 13 2^ 3 22 "o 3 3 § » .& LD m '^ rA^ o <^ O 00 ^ 3 ►J 00 <; 3 o 1. ^g i-H i-H >-l S "5£. as 40 31.1 747 4) School size <500 30.2 443 500-1 ,500 29.8 741 0.49 >1,500 27.4 693 5) Number killed 1-10 29.8 1,104 11-30 30.1 917 <0.001 >30 37.4 1,301 6) Tuna caught (tons) 0-5 29.0 428 6-15 33.0 848 0.02 16-30 30.4 760 >30 35.6 1,280 tested separately for these two time periods. When the years 1971-73 were excluded (Tkble 4), the inter- actions between percent pregnant and season, dolphin kill, and tuna catch are no longer significant. When tests are performed on data from 1971 to 1973 alone (Ikble 4), season and tuna catch are still significantly related to pregnancy rate Percent Lactating Annual trends in percent lactating for the north- ern stock of spotted dolphins are illustrated in Figure 3. Two cases are considered: 1) the percen- tage of all mature females that are lactating and 2) the percentage of pregnant females that are lac- tating. For both cases, a weighted regression shows a significant increase in the fraction of lactating females through time (P < 0.05). In the former case, the regression again appears to be driven by anomalous values in 1971-73. Percentages and sam- ple sizes for the latter case are presented in Ikble Table 3. — Multiway tests of factors affecting per- cent pregnant. Log-likelihood chi-square was used to calculate the probability that percent pregnant is independent of the stated factor(s) using the log-linear model. Pregnancy state (pregnant/not pregnant) is implicit as the first factor in each comparison. Tests Probability 3-way a) Year <0.0001 Season 0.70 Year x season <0.0001 b) Year <0.0001 Kill 0.09 Year x kill 0.31 c) Year <0.00^ Catch 0.14 Year x catch 0.0004 4-way Year <0.0001 Season 0.90 Catch 0.15 Year x season <0.0001 Year x catch 0.0002 Season x catch 0.003 Year x season x catch 0.07 Table 4. — Percentage of sexually mature females that were pregnant, grouped by season, dolphin kill, and tuna catch. The years 1971-73 and 1974-83 are grouped separately. Probabilities are based on chi-square contingency tests. Only samples from the northern offshore stock of spotted dolphins are included. Pregnant Prob- Year (%) N ability 1971-73 overall 41.9 1,101 Season 1st quarter 44.0 722 2d quarter 47.6 210 <0.001 3d quarter — 0 4th quarter 25.4 169 Number killed 1-10 38.0 171 11-30 39.5 248 0.23 >30 44.1 651 Tuna caught (tons) 0-5 37.6 101 6-15 44.5 247 0.04 16-30 35.0 226 >30 45.5 490 1974-82 overall 29.0 2,566 Season 1st quarter 28.4 1,077 2d quarter 27.9 531 0.39 3d quarter 28.9 584 4th quarter 32.7 373 Number killed 1-10 28.3 933 11-30 26.6 669 0.27 >30 30.6 650 Tuna caught (tons) 0-5 26.3 327 6-15 28.3 601 0.76 16-30 28.5 534 >30 29.5 790 662 BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS (J Z < I- O < -J I- z Hi o • All mature females o Only pregnant females (479) (473) (358) (203) <'»44> (558) 80 70 60 50 40 30 20 10 0 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 YEAR Figure 3.— Percentages of mature females that were lactating and percentages of pregnant females that were lac- tating for the northern offshore stock of spotted dolphins from 1971 to 1983. Solid lines represents weighted regres- sions. Sample sizes are in parentheses. (14) 1 for both the northern and southern stocks. Again sample size for the southern stock is too small to examine individual years; however, the overall per- cent of pregnant females that were lactating in 1973-83 samples shows significant differences be- tween stocks ixl = 6.50, P = 0.01). Annual variability for the northern stock is greater than expected from random sampling of a population with a constant percent lactating (xfg = 63.5, P < 0.001). When the sample of pregnant females from the northern stock is stratified by the six sampling fac- tors, lactation state was significantly related to sam- pling season, dolphin kill, and tuna catch (Tkble 5). Again, 3-way tests showed that the first order effect of these factors was not significant when year was included as the third factor (Ikble 6). In each of these cases, the first order effect of year was important. In one case, dolphin kill, a second order interaction between kill and year was also significant. Percent Mature The fractions of females that were sexually mature are given in Ikble 1 by stock and by year (1973-83). Again the sample sizes are sufficient in all years for the northern stock but are inadequate in some years for the southern stock. The southern stock is significantly different from the northern stock in percent mature (xj = 31.2, P < 0.001), and (given its small sample size) the southern stock was excluded in subsequent stratifications. The percentage of all females that are mature from 1973 to 1983 is illustrated in Figure 4 for the north- ern stock. In this case, the weighted regression is not significant. Using chi-square tests, the level of annual variability in percent mature is larger than would be expected from randomly sampling a population with a constant fraction of mature females (P < 0.001). For long-lived animals such as dolphins, annual variability in percent mature should be small and changes in this population parameter should be gradual. Since the annual variability observed in the data is larger than would be expected from random sampling error, year-to-year changes in sampling biases are likely. Percent mature was found to be significantly related to three of the six sampling fac- tors examined: sampling season, dolphin kill-per-set, and tuna catch-per-set (Ikble 7). Each of these three significant factors was tested with maturity state and year using 3-way tests (Ikble 8). For each of these factors, year was a significant factor and all other first order effects were not significant. Only dolphin kill showed a significant sec- ond order interaction with year. 663 FISHERY BULLETIN: VOL. 83, NO. 4 Table 5. — Percentage of pregnant dolphins that were lac- tating (1971-83) grouped by 1) the area in which the speci- mens were collected; 2) the quarter of the calendar year; 3) the length of time between sighting the school and release of the net; 4) the observer's estimate of the total school size; 5) the number of dolphins known to be killed in the set; and 6) the tons of tuna caught in the set. Note that total sample size varies with the availability of data on the stratifying variable. Probabilities are based on chi- square contingency tests. Only samples from the northern offshore stock of spotted dolphins are included. CYRA = Commission Yellowfin Regulatory Area. Lactating Prob- (%) N ability 1) Geographic area Inside CYRA 15.7 988 0.30 Outside CYRA 18.8 181 2) Season 1st quarter 11.9 611 2d quarter 17.2 239 <0.001 3d quarter 26.7 161 4th quarter 20.3 158 3) Chase time (min) <20 14.8 331 20-40 14.0 479 0.27 >40 18.6 226 4) School size <500 20.0 130 500-1,000 18.0 211 0.87 >1,500 19.7 183 5) Number killed 1-10 19.1 319 11-30 16.5 267 0.03 >30 12.3 471 6) Tuna caught (tons) 0-5 20.5 117 6-15 16.4 275 0.17 16-30 16.7 221 >30 12.8 444 DISCUSSION Changes in the reproductive status of the female segment of a population can be monitored using a variety of reproductive indices: 1) mean age at sexual maturation, 2) mean length (or weight) at sexual maturation, 3) annual pregnancy rates, 4) calving interval, 5) percentage of mature females that are pregnant, 6) percentage of females that are lactating, and 7) percentage of females that are sexually mature Changes in each of these are examined below. Changes in Maturation Parameters Myrick et al. (1984) have found no significant dif- ference in the age at sexual maturation (ASM) between a sample from 1973 to 1978 and another sample from 1981. In the present study, length at attainment of sexual maturity is estimated to have increased 4.4 cm from 1974 to 1983. If these results hold, dolphins must be growing faster in recent Table 6. — Multiway tests of factors affecting the percentage of pregnant females that are lactating. Log-likelihood chi-square was used to calculate the probability that per- cent lactating is independent of the stated factor(s) using the log-linear model. Lacta- tion state (lactating/not lactating) is implicit as the first factor in each comparison. 3-way tests Probability a) Year <0.0001 Season 0.29 Year x season k 0.007 b) Year <0.0001 Kill 0.79 Year x kill 0.51 Table 7. — Percentage of female dolphins that were sexual- ly mature (1973-83) grouped by 1) the area in which the specimens were collected; 2) the quarter of the calendar year; 3) the length of time between sighting the school and release of the net; 4) the observer's estimate of the total school size; 5) the number of dolphins known to be killed in the set; and 6) the tons of tuna caught in the set. Note that total sample size varies with the availability of data on the stratifying variable. Probabilities are based on chi- square contingency tests. Only samples from the northern offshore stock of spotted dolphins are included. CYRA = Commission Yellowfin Regulatory Area. Mature (%) N Prob- ability 1) Geographic area Inside CYRA 55.7 6,329 0.19 Outside CYRA 57.5 1 ,625 2) Season 1st quarter 54.2 3,495 2d quarter 57.2 1 ,738 0.02 3d quarter 58.2 1,580 4th quarter 57.6 1,155 3) Chase time (min) <20 54.4 2,067 20-40 56.1 3,084 0.36 >40 56.5 1 ,689 4) School size <500 53.5 1,183 500-1,500 56.8 1,970 0.19 >1,500 56.0 1,753 5) Number killed 1-10 57.9 2,465 11-30 54.9 2,068 0.02 >30 54.0 2,321 6) Tuna caught (tons) 0-5 53.3 920 6-15 55.1 1,779 0.05 16-30 54.5 1,668 >30 57.7 2,482 years. Given small sample sizes of aged individuals, significant changes in ASM may be difficult to detect. Previous studies have shown that the age at sexual maturation is quite responsive to population changes in marine mammals (Fowler 1984), while length at maturation tends to show little change For fin whales, Balaenoptera physalus, Lockyer (1972) 664 BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS 80 t- UJ H < 2 UJ o tu 0. 70 - 60 50 40 30 (995) (1149) (1013) (1215) (593) (564) . (509) (465) (579) (751) (137) ± 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 YEAR Figure 4.— Percentages of all females that were mature for the northern offshore stock of spotted dolphins from 1971 to 1983. Sample sizes are in parentheses. Table 8.— Multiway tests of factors affecting percent mature. Log-likelihood chi-square was used to calculate the probability that percent mature is independent of the stated factor(s) using the log-linear model. I\/latura- tion state (mature/not mature) is implicit as the first factor in each comparison. 3-way tests Probability a) Year 0.003 Season 0.49 Year x season 0.08 b) Year 0.003 Kill 0.16 Year x kill 0.002 c) Year 0.0009 Catch 0.09 Year x catch 0.42 showed a decrease in ASM without any change in the length at which maturity is attained. Laws (1956) predicted an inverse relationship between ASM and early growth rates for marine mammals. Spotted dolphins appear to show an increase in length at maturation with no change in ASM, and thus do not follow predicted patterns. Trends in Percent Pregnant Annual pregnancy rates, calving interval, and per- cent pregnant all measure essentially the same thing. Annual pregnancy rate and calving interval require knowledge of gestation times. Because density com- pensatory responses have not been shown in ceta- cean gestation times, it is more straightforward to deal directly with percent pregnant. What appeared to be a rapid decline in dolphin pregnancy rates from 1973 to 1978 (Henderson et al. fn. 2), now appears as two eras with distinctly different pregnancy rates. The fraction of pregnant females in the 1971-73 samples was quite high. The 10 years since 1973 show a lower and relatively con- stant fraction of pregnant females. This difference in results is due largely to use of a larger sample size and a longer time series. There is no evidence of the sort of slow, long-term trends in pregnancy rates that might be associated with changes in population sizes. The reason for the dramatic change in pregnancy rates between 1973 and 1974 is not known. At least three hypotheses could be used to explain this change: 1) it was the result of a naturally high pregnancy rate in 1971-73; 2) it was the result of heavy fishing-related morta- lity of nursing calves prior to 1974 that resulted in artificially high pregnancy rates; or 3) it was the result of a bias in the sampling by tuna vessels. The first hypothesis suggests that changing en- vironmental conditions result in annual changes in pregnancy rates. In the ETP the largest environ- mental perturbations are associated with "El Nino" events which occur on the time scale of from 5 to 10 yr (Rasmusson and Carpenter 1982). El Nifio con- ditions prevailed in 1972 (moderate), 1975-76 (weak), and 1982-83 (very strong). These dates do not help explain the change in pregnancy rates that occurred between 1973 and 1974. The second hypothesis is that heavy dolphin mor- tality in the 1960's and early 1970's may have some- how affected dolphin pregnancy rates. Large reduc- 665 FISHERY BULLETIN: VOL. 83, NO. 4 tions in dolphin mortality occurred following the passage of the Marine Mammal Protection Act of 1972 (Tkble 9). If mortality rates were higher for nursing calves, calving interval might have been shortened. This would result in higher pregnancy rates and lower lactation rates, both of which were observed in 1971-73. Analyses have indicated that very young calves are more susceptible to tuna-net mortality (Powers and Barlow fn. 3; Stuntz^). In- directly, high calf mortality may also result from the separation of a calf from its mother during long chases. It is not known if the magnitude of these ef- fects could have resulted in the observed changes in pregnancy or lactation rates. The third hypothesis is that sampling methods were somehow different between 1971-73 and 1974-83. The only difference in the sampling design was that in 1971-73, scientific technicians were placed only on tuna vessels that agreed to cooperate Beginning in 1974, the selection of vessels was ran- dom. It is difficult, however, to see how this change would affect the percent pregnant in the samples. As was noted above, percent pregnant was signifi- cantly correlated with sampling season, dolphin kill- per-set, and tuna catch-per-set during the years 1971-73, but not during the years 1974-83. The reason for this difference is not known, but this would seem to be evidence that sampling was more random in the latter period. The observed change in percent pregnant from 1971-73 to 1974-83 cannot be explained with certain- ty. The high pregnancy rates in 1971-73 can be logically explained by direct or indirect effects of the fishery or by sampling biases in those years (Hypo- theses 2 and 3). Determining whether either (or both) hypothesis is true may not be possible with existing data. Trends in Percent Lactating Changes were also found in the percentage of lac- tating females. For mature females, the fraction lac- tating shows low values in 1971-73 and high values in 1974-83, which is opposite the pattern seen for fraction pregnant. This inverse correlation would be expected given that pregnancy state and lactation state are physiologically linked (i.&, cessation of lac- tation leads to ovulation and pregnancy). Perhaps more meaningful is the increase in the fraction of Table 9. — Estimates of numbers of spotted dolphins killed by all purse seine vessels in the eastern tropical Pacific, 1968-78 (data from Smith 1983). Spotted dolphins Year killed 1968 178,000 1969 365,000 1970 355,000 1971 176,000 1972 288,000 1973 131,000 1974 95,000 1975 105,000 1976 47,000 1977 22,000 1978 19,000 *Stuntz, W. E. 1980. Variation in age structure of the inciden- tal kill of spotted dolphins, Stenella attenuata, in the U.S. tropical purse-seine fishery. Admin. Rep. LJ-80-06, 29 p.; available from Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. pregnant females that were lactating. Because being simultaneously pregnant and lactating represent the greatest energy drain on female dolphins, this quan- tity is likely to be very sensitive to changes in en- vironmental conditions. Because nonpregnant females are excluded, this quantity should also be insensitive to sample biases that are related solely to pregnancy state Given that no trends were seen in the percent pregnant from 1974 to 1983, we can infer that the calving interval, or the mean period between births for a mature female, also did not change during that time If calving interval were constant, the increase in the fraction of pregnant females that were lac- tating indicates that females may be nursing their calves for a longer period of time, hence a longer lac- tation period. This increase in the lactation period may have resulted from a decrease in fishery related calf mortality during the 1971-83 period. Because calves may be more susceptible to death or separa- tion from their mothers during the chase, capture, and release of a dolphin school, mean lactation periods may have been abbreviated during the earlier years (Hypothesis 2 above). Trends in Percent Mature No significant trends in the percentage of females that were sexually mature during 1971-83 are evi- dent for the northern stock of spotted dolphins. An- nual variability was far too great to be explained by random sampling error. This parameter showed a significant correlation with dolphin kill-per-set. Therefore, unless sampling conditions remain con- stant (which they have not), percent mature is not a useful index for monitoring reproductive capabili- ty of the spotted dolphin populations. 666 BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS Variability in percent mature with sampling con- ditions may result from several interacting factors. Preliminary data have indicated that spotted dolphins in the ETP may segregate on the basis of reproductive maturity (A. A. Hohn fn. 4 and M. D. Scott fn. 5). Schools that consist principally of im- mature dolphins may have a smaller characteristic school size, may be less likely to have large numbers of tuna associated with them, and may be more vulnerable to high kills-per-set due to the inex- perience of younger dolphins. Also, the consistent underrepresentation of immature age classes in the spotted dolphin age distribution (Barlow and Hohn 1984) indicates that a very significant bias may oc- cur in the sampling of immature animals. These are largely speculations, and until a well-supported ex- planation for sampling variability is presented and until some method of removing this bias is found, percent mature should not be used as an index of changes in spotted dolphin reproduction. Gross Annual Reproductive Rate Changes in GARR have been used as a measure of changes in the net rate of growth for a popula- tion (Smith 1983). This approach has been faulted on the basis that it does not consider age structure effects (Polacheck 1982), and more critically on the basis that such an approach is theoretically unsound (Goodman^). These criticisms do not, however, detract from the usefulness of GARR as an index of gross per capita reproduction for a popula- tion. If GARR were robust to sampling conditions, it could be one of the most useful indices of popula- tion reproduction. One advantage is that a GARR index considers percent pregnant and percent mature simultaneously, and hence compensatory changes in these two do not affect the index. Sim- ply stated, GARR is calculated as (the fraction of females in a population) x (the fraction of females that are mature) x (the fraction of mature females that are pregnant)/(gestation time). Unfortunately, percent mature is a major component in these calculations, and this parameter has been found to be dependent on sampling conditions. Until sampling problems associated with estimating percent mature are resolved, GARR is not likely to be a useful in- dex of change in reproductive rates. Between-Population Comparisons The northern and southern stocks of spotted dolphins have been subjected to very different levels of fishing-related mortality. Smith* has estimated the northern stock to be at 38-55% of its 1959 level and the southern stock to be at 93-98% of its historical level. Density dependent increases in reproductive rates might be predicted for the northern stock relative to the southern stock. The percentage of mature females that were preg- nant differs significantly between the northern and southern stocks. Surprisingly, however, the southern stock was found to have the higher percent pregnant (36% vs. 33%). Another exploited population of spot- ted dolphins in the western Pacific was found to have an annual pregnancy rate of 0.254 (Kasuya 1976), which (with a gestation time of 11.2 mo) would give an average percent pregnant of about 24%. Con- siderable variability in percent pregnant can thus exist between spotted dolphin populations, none of which is obviously related to density compensatory effects. Sampling of the southern population has, however, been sporadic, and if annual variability in pregnancy rates is greater for that stock, a few years' data may not be sufficient to accurately estimate a long-term mean. Nonetheless, the tendency for a more exploited stock to have lower reproductive rates is worrisome, and future life history comparisons between the northern and southern stocks would probably be useful. Evidence for density compensatory changes in pregnancy rates were also lacking when two spin- ner dolphin, 5. longirostris, populations were com- pared (Perrin and Henderson 1984). They found similarly that the more heavily exploited stock (eastern spinners) had a lower percent pregnant than the less heavily exploited stock (whitebelly spinners). The opposite would be predicted based purely on density compensatory effects. The overall percentage of pregnant females that are lactating is significantly higher for the north- ern spotted dolphins than for the southern stock. The biological significance of this result is ques- tionable given the year-to-year variability in this parameter. Between-population comparisons of this percentage are not likely to be meaningful until the cause of this large annual variability is identified. As was noted above, the percentage of females that 'Goodman, D. 1984. Uses of the gross annual reproduction rate calculation in the dolphin assessment. Admin. Rep. LJ-84-22C, 17 p.; available from Southwest Fisheries Center La JoUa Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. *Smith, T. D. (editor). 1979. Report of the status of porpoise stocks workshop (August 27-31, 1979, La Jolla, California). Ad- min. Rep. LJ-79-41, 120 p.; available from Southwest Fisheries Center La Jolla Laboratory, National Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 92038. 667 are mature also appears to be higher in the southern stock than in the northern stock. Given the depen- dence of this parameter on sampling conditions, lit- tle confidence should be placed on this result. The direction of the difference (more immatures in the northern population) is consistent with a higher population growth rate in the north. Work is in pro- gress to determine whether this difference could be due to differences in the age at sexual maturation (Myrick^). No data exist yet on the mean age at sexual maturation for females in the southern stock. As mentioned in the Methods section, data do exist on the "length at sexual maturation" for both the north- ern and southern stocks. For the northern stock, this length appears to have increased from 176 cm in 1974 to 181 cm in 1983. For the southern stock this length was estimated as 175 cm from the pooled 1974-83 data. The length at sexual maturation is greater for the northern stock, which is consistent with the greater mean asymptotic length of the northern specimens (Perrin et al. 1979). CONCLUSION My intent in writing this paper was to identify in- dices that may be of value in monitoring the repro- ductive health of spotted dolphins in the eastern Pacific. Two of the indices that were examined (the percentage of mature females that are pregnant and the percentage of pregnant females that are lac- tating) are likely to be useful for this purpose Both are relatively insensitive to sampling biases, and both measure important aspects of the female reproduc- tive cycla Problems exist in measuring the fraction of females that are mature This parameter is also an important index of net reproduction in a popula- tion. It is possible that a stable index of percent mature could be obtained using some stratification scheme A first approach might be to examine finer scale geographic differences in percent mature Addi- tional work is necessary before significance can be ascribed to between-population differences in per- cent mature ACKNOWLEDGMENTS Much credit is deserved by the many technicians who gathered the data upon which this report is based. Their labors were often under very difficult working conditions. I also wish to thank the data ^A. C. Myrick, NOAA, National Marine Fisheries Service, P.O. Box 271, La Jolla, CA 92038, pers. commun. December 1984. FISHERY BULLETIN: VOL. 83, NO. 4 editing and managing group and the graphics department, SWFC. Ideas presented in this manu- script were largely generated in discussions with A. Hohn and W. Perrin. The current draft of this manu- script benefited greatly from critical reviews by J. 4Bengtson, D. Chapman, D. DeMaster, F. Hester, A. Hohn, J. Mean, A. Myrick, W. Perrin, S. Reilly and G. Sakagawa. LITERATURE CITED Barlow, J., and A. A. Hohn. 1984. Interpreting spotted dolphin age distributions. U.S. Dep. Commer., NOAA Ifech. Mema NOAA-NMFS-SWFC-48, 22 p. Dixon, W. J. (editor). 1981. BMDP statistical software. Univ. Calif. Press, Berkeley, 725 p. Fowler, C. W. 1984. Density dependence in cetacean populations. Rep. Int. Whaling Comm., Spec Issue 6:373-379. Gambel, R. 1975. Variations in reproduction parameters associated with whale stock sizes. Rep. Int. Whaling Comm. 25:182-189. Hester, R. 1984. Possible biases in the estimates of rates of reproduc- tion in the spotted dolphins, Stenella attenuata. Rep. Int. Whaling Comm., Spec. Issue 6:337-341. Kasu\a, T. 1976. Reconsideration of life history parameters of the spot- ted and striped dolphins based on cemental layers. Sci. Rep. Whales Res. Inst. 28:73-106. Laws, R. M. 1956. Growth and sexual maturity in aquatic mammals. Nature (Lond.) 178:193-194. Lockyer, C. H. 1972. The age of sexual maturity of the southern fin whale (Balaenoptera physalus) using annual layer counts in the ear plug. J. Cons. Int. Explor. Mer 34:276-294. MiZROCH, S. A. 1983. Reproductive rates in Southern Hemisphere baleen whales. MS Thesis, Univ. Washington, Seattle, 103 p. Myrick, A. C, Jr., A. A. Hohn, J. Barlow, and P. A. Sloan. In press. Reproduction in the female spotted dolphin, Stenella attenuata. Fish. Bull., U.S. Perrin, W. F. 1970. Color patterns of the eastern Pacific spotted porpoise Stenella graffmani Lonnberg (Cetacea, Delphinidae). Zoo- logica (N.Y.) 54:135-149. Perrin, W. F, J. M. Coe, and J. R. Zweifel. 1976. Growth and reproduction of the spotted porpoise, Stenella attenuata, in the offshore eastern tropical Pacific Fish. Bull., U.S. 74:229-269. Perrin, W. F., and J. R. Henderson. 1984. Growth and reproductive rates in two populations of spinner dolphins, Stenella longirostris, with different histories of exploitations. Rep. Int. Whaling Comm., Spec Issue 6:417-430. Perrin, W. F, R. B. Miller, and P. A. Sloan. 1977. Reproductive parameters of the offshore spotted dolphin, a geographic form of Stenella attenuata, in the eastern tropical Pacific, 1973-75. Fish. Bull., U.S. 75:629- 633. 668 BARLOW: REPRODUCTIVE RATES OF SPOTTED DOLPHINS Perrin, W. F., p. a. Sloan, and J. R. Henderson. NOAA-TM-NMFS-SWFC-19, 9 p. 1979. Tkxonomic status of the "southwestern stocks" of spin- Rasmusson, E. M., and T. H. Carpenter. ner dolphins, Stenella longiroatris, and spotted dolphins, S. 1982. Variations in tropical sea surface temperature and sur- attenuata. Rep. Int. Whaling Comm. 29:175-184. face wind fields associated with the Southern Oscillation/ POLACHECK, T. El Nino. Mon. Weather Rev. 110:354-384. 1982. The relationship between changes in gross reproductive Smith, T. D. rate and the current rate of increase for some simple age 1983. Changes in size of three dolphin (Stenella spp.) popula- structured models. U.S. Dep. Commer., NOAA Ifech. Memo. tions in the eatern tropical Pacific Fish. Bull., U.S. 81:1-13. 669 NOTES ANNUAL BAND DEPOSITION WITHIN SHELLS OF THE HARD CLAM, MERCENARIA MERCENARIA: CONSISTENCY ACROSS HABITAT NEAR CAPE LOOKOUT, NORTH CAROLINA The presence of periodically repeating features in the preservable hard parts of various organisms allows scientists in several disciplines to make im- portant inferences about the rates and timing of past events (Jones 1980; Rhoads and Lutz 1980). Analysis of growth lines deposited in shells of bivalve molluscs, for example, finds powerful application in the fields of paleontology (Rosenberg and Runcorn 1975), an- thropology (Clark 1979), population ecology (Ken- nish 1980), and fisheries biology (Peterson et al. 1983). Possession of a reliable age marker in a bivalve shell enables fisheries biologists 1) to construct age- frequency distributions for various populations, which reflect the age-specific mortality rates and help permit estimates of sustainable yield, 2) to calculate individual growth rates and their variability among habitats, and 3) to understand age-specific reproductive schedules in exploited populations. Unfortunately, the potential rewards in applying this aging technique have encouraged widespread use of growth line analysis prior to performing the necessary controls to test the annual periodicity of line deposition (Clark 1974; Gould 1979; Jones 1981). Because of the tremendous potential utility of this aging technique, we carried out mark-recapture tests of the annual nature of growth band deposition in shells of the commercially important hard clam, Mercenaria mercenaria, in a North Carolina sound (Peterson et al. 1983). Although these experiments provided convincing evidence that M. mercenaria deposits a reliable annual marker in the form of an internal summer growth band in its shell, this study was carried out in only a single locality in Back Sound, NC. Patterns of growth band deposition in bivalve molluscs may vary with environment on several scales: 1) over a broad geographic scale, M. mercenaria deposits summer bands in Back Sound, NC, and in Chesapeake Bay, but winter bands in all localities in northeastern states (Pannella and MacClintock 1968; Rhoads and Pannella 1970; Ken- nish and Olsson 1975; Clark 1979; Clark and Lutz 1982; Fritz and Haven 1983; Peterson et al. 1983); 2) among habitats within estuaries, Protothaca staminea appears to deposit unambiguous annual bands in muddy sand but not in a clean-sand habitat in Mugu Lagoon, CA (Peterson and Ambrose 1985); and 3) among nearby individuals within a single habitat, both Chione fluctifraga and Protothaca staminea from within the same restricted sample at Mugu Lagoon exhibit radically different patterns of daily line deposition (Hughes and Clausen 1980). We present here results of additional tests of the an- nual nature of internal growth band deposition in shells of M. mercenaria placed for 2 yr in several different field localities and estuarine habitats, in order to test whether our earlier (Peterson et al. 1983) demonstration of annual banding in North Carolina's M. mercenaria is robust to change in local habitat. Materials and Methods lb extend the generality and power of our previous results, we designed a mark-recapture experiment to examine the frequency and clarity of band deposi- tion in M. mercenaria at 5 additional sites (Fig. 1) within Carteret County, NC, near Cape Lookout. These sites were chosen to represent a wide geo- graphic spread among several local water bodies, to permit contrasts between vegetated and unvegetated habitats, and to include more sandy (coarse) sub- strate than that in our original site One site was selected on a fine sand flat in the North River about 12 km from our earlier Middle Marsh study site in Back Sound. Two sites were chosen about 38 km from Middle Marsh near the western end of Bogue Sound by the town of Cape Carteret: one on a fine sand flat and the other in a seagrass bed with mix- ed stands of Zostera marina and Halodule wrightii. The other two sites were situated in Core Sound about 6 km from Cedar Island Point and about 47 km from our initial Middle Marsh study site: one on a sand flat and the other in a Halodule wrightii meadow. All sites were on shallow subtidal bottom, accessible by wading and amenable to recovery of marked animals. Tkble 1 summarizes the results of particle-size analyses done on duphcate surface (0-5 cm) sediment cores taken in August-September 1981 at each site to permit comparisons among the five new and one previous study sites. The five new sites are clearly characterized by having much coarser sediments than the previous study site but differ among them- selves in sediment grade (Tkble 1). Contemporaneous FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 671 Figure 1.— Geographic locations of study sites (marked by dots near circled numbers) within North Carolina near Cape Lookout (marked by the arrow on the NC map): the previous site at Middle Marsh (1); new sites at North River (2); Bogue Sound sand (3) and seagrass (4); Core Sound sand (5) and seagrass (6). 672 Table 1. — Locations and sedimentary characteristics of the one previous (IVIiddie Marsh) and five new study sites of Mercenaria mercenaria. Sedimentary data came from seiving and pipetting (Folk 1974) duplicate surface (0-5 cm) sediment cores taken in sum- mer 1981 at each site. Percent silt-clay is percent of total sediment dry weight in fine (>40) size classes. Mean sediment parameters Location (±SD) Graphic Percent Sorting Site coordinates mean silt-clay coefficient Middle Marsh 34"'41'28"N 5.07 47.22 3.00 in Back Sound 76°37'03"W (0.28) (12.24) (0.51) North River 34°48'22"N 2.35 2.97 0.61 76°36'48"W (0.02) (1.72) (0.01) Bogue Sound 34°41'29"N 2.83 2.15 0.47 Sand 76°59'06"W (0.09) (0.05) (0.02) Bogue Sound 34°41'36"N 3.05 9.31 0.95 Seagrass 76°59'05"W (0.02) (2.22) (0.17) Core Sound 34°57'03"N 2.72 8.81 0.91 Sand 76°12'44"W (0.07) (2.03) (0.16) Core Sound 34°56'59"N 2.40 3.10 0.60 Seagrass 76°12'43"W (0.00) (1.33) (0.06 water temperature and salinity data are not available for all sites, but records from a variety of sources (Brett 1963; Thayer 1971; Williams et al. 1973; Sutherland and Karlson 1977; H. J. Porter, Univer- sity of North Carolina, Chapel Hill, unpubl. data; W. Kirby-Smith, Duke University, unpubl. data) suggest that 1) water temperature patterns probably do not differ greatly across sites, with monthly averages ranging from winter minima of 2°-4°C to summer maxima of 29°-30°C, and 2) that salinities are slightly more variable across sites. Localities close to Atlantic Ocean inlets (Bogue Sound sand and seagrass sites and the previous Back Sound site at Middle Marsh) experience uniformly high salinities (30-36°/oo), ex- cept after severe rainstorms (Brett 1963; H. J. Porter, unpubl. data). Salinities in the upper portion of North River are only slightly lower because there is little freshwater inflow into that system (Thayer 1971). The lowest (22-28o/oo) and probably most variable salinities on a week-to-week scale occur at the two Core Sound sites, where exchange with the ocean is reduced and where any persistent north winds bring intrusions of low-salinity waters from Pamlico Sound (Williams et al. 1973). At each of the five new study sites, we placed groups of 80 Mercenaria mercenaria in 1 m^ field plots in late summer 1980, excavated them by hand in late summer 1981 to estimate growth and mor- tality and to replace missing and dead clams, and then finally recovered all living clams present in late summer 1982 (Ikble 2). All M. mercenaria used in these experiments were individually marked on the external shell surface with color-coded dots of Mark- Ibx Corporation paints and measured initially and at both yearly samplings by calipers to the nearest 0.1 mm in each of three mutually perpendicular dimensions (length, height, thickness). Clams used in these mark-recovery experiments were chosen to reflect a size range from 1 to 10 cm in length. Before placing the marked and measured clams into the field plots, we first installed fences of 6 mm mesh plastic (VEXARi) around the 1 m^ plots. These fences were identical to those used and described previously (Peterson et al. 1983) and were designed to inhibit emigration and to mark off bottom plots to improve our ability to recover the marked clams. At the three unvegetated sites, we removed all ini- tially present M. mercenaria and other large macro- fauna before adding marked clams by first using fingers to plow systematically the top 10 cm of sediments and then twice systematically sieving in situ through 6 mm mesh the entire 1 m^ surface to that same 10 cm depth. This procedure was not used at the initiation of the experiment at the two sea- grass sites because it would have removed the sea- grass itself. This same procedure was employed, although using a 3.2 mm mesh, at both yearly sam- plings to recover all marked clams from all 1 m^ plots at each unvegetated site. At the two seagrass sites, marked clams were recovered by using a hydraulic suction dredge and collecting the contents of the top 15 cm on a 3 mm nylon mesh bag (see Peterson et al. 1983 for data on sampling efficienqr of this device). Because of the removal of seagrasses, the locations of all seagrass plots were then shifted slightly (<3 m) to new, undisturbed positions for the second and final year. 'Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. Table 2. — Dates of experiment initiation and subsequent sampling of Mercenaria mercenaria for each of the five study sites. Measuring dates 1980 1981 1982 Sitei initiation sampling termination North River 6 August 30 July 2 August Bogue Sound Sand 15 August 2 September 24 August Bogue Sound Seagrass 15 August 24 August 24 August Core Sound Sand 3 September 9 September 3 September Core Sound Seagrass 3 September 11 September 3 September 'Each site held 1 m ~2 enclosures of Mercenaria mercenaria at 80 m ~^: one enclosure contained clams derived from a common Back Sound site (sup- plemented by 10-20 mm clams from other Back Sound origins), and at least one other enclosure contained only locally derived clams. 673 At each site, all M. mercenaria in these ex- periments >20 mm in initial length came from one of two different sources. One enclosure at each site held clams planted at the constant density ^f 80 m"^ and collected initially from a single source, a seagrass bed along the southern shore of Back Sound (described in Peterson et al. 1984). One or two additional enclosures held clams collected locally at the specific study site, but again kept at the same 80 m"^ density. Low availability of clams in the 10-20 mm size class from the Back Sound source forced us to add hatchery-reared clams (spawned from Back Sound individuals) and wild-caught clams from Middle Marsh also in Back Sound to represent the 10-20 mm sizes in the "common source" enclosure at each sita This smallest size class was available locally at each site to complete the "local-origin" enclosures. This design was chosen to enable us to test whether there were any effects of clam origin (the single Back Sound source vs. local clams) at each of the five study sites. Upon termination of the experiment almost exact- ly 2 yr after initial planting of the marked clams (Tkble 2), subsets of those clams that had survived the complete 2-yr period were selected from each origin treatment at each study site in as wide a range of sizes as possible These clams were returned to the laboratory and killed by steaming. One valve from each of these 2-yr survivors was then section- ed, using a diamond blade on a slow-speed Buehler ISOMET saw, from umbo to ventral margin along the axis of greatest growth. The shell sections were sanded and polished when necessary to enhance the clarity of natural banding patterns. Because the ini- tial size in August- September 1980 and the inter- mediate size 1 yr later were known for each of the marked individuals and could be marked on the shell surface, and because the marking and measuring process itself almost invaribly induces deposition of a disturbance check which serves as a very precise internal shell marker (Peterson et al. 1983), we were then able to count with the unaided eye the number of additional growth bands deposited in the internal shell matrix of each clam during its final 2 yr of life We also observed where these bands were deposited relative to the known sizes at the initial, interme- diate, and final measuring dates. These observations permit a test of whether the reliability of using sum- mer growth bands to age North Carolina's M. mercenaria varies with site (habitat) or clam origin in the vicinity of Cape Lookout. Results We sectioned shells from a total of 89 M. mercenaria collected alive in August-September 1982 and known by their paint codes to have been present in the field since the experiment's initiation 24 mo before (Table 2). Of these 89 individuals, 17 either exhibited insufficient growth to permit an ac- curate determination of the precise shell size at the experiment's initiation or else lacked a disturbance check to mark the precise size at initiation. Of the remaining 72 individuals, all but 2 deposited exact- ly 2 additional dark growth bands in the final 24 mo of life (Tkble 3). This pattern was consistent across all five study sites and did not change as a function of clam origin (Ikble 3). The appearance of the dark Table 3.— For each of five new study sites: 1) numbers of hard clams cut for growth analysis from each origin treatment, 2) numbers of those with insufficient growth to assess band deposition accurately, 3) range of ini- tial clam lengths for those clams with sufficient growth, and 4) average number of bands deposited in the 2-yr experimental period. Clam origin Back Sound Local site No. Study site cut North River 10 Bogue Sound Sand 10 Bogue Sound Seagrass 3 Core Sound Sand 10 Core Sound Seagrass 10 Range in Av. no. of No. with initial annual bands No. with insufficient length added in 2 yr No. insufficient growth (mm) (±1 SE) cut growth Range in Av. no. of initial annual bands length added in 2 yr (mm) (±1 SE) 1 19-74 2(±0) 17 1 43-80 1.£ M±o. 1 14-57 2(±0) 11 2 21-72 2 (±0) 1 39-69 2(±0) 3 1 44-48 2 (±0) 0 18-75 2(±0) 5 3 63-70 2 (±0) 1 15-72 2(±0) 10 6 46-56 2 (±0) 674 band in cross-section was identical to that previous- ly described and illustrated by photograph (Peter- son et al. 1983) of clams harvested from the Middle Marsh locality. The pattern of band deposition relative to times of initial planting, first measurement (12 mo), and collection (24 mo) was also extremely consistent across all data sets. Initial planting in 1980 occurred during the period of annual band deposition for 70 of the 72 clams. (In one clam, the 1980 annual band was just completed and in another the 1980 annual band was just about to begin at the time of initial planting.) The disturbance check caused by the 12-mo measurement fell near the end of the growth band for 70 clams and just after the band for the two others. The time of collection in 1982 fell dur- ing or just immediately after the deposition of the 1982 annual band for all clams except those from local origin at North River. Of the 16 cut clams in that data set with sufficient growth for band resolu- tion, 12 were just beginning to deposit their 1982 band at the time of collection (2 August, 3-4 wk earlier than the other sites— Ikble 2). Two of the 16 lacked the terminal band, whereas the remaining two had already deposited a substantial amount of the 1982 band. This North River local data set was the only one that contained any clams (only three) which had bands sufficiently faint to cause any doubt about recording them. By counting all presumptive annual bands over the complete growth record of each clam, we also estimated the age of each of the 89 M. mercenaria used in this experiment. The estimated age at the experiment's initiation for the 17 clams excluded from our 2-yr tests ranged from 6 to 29 yr and averaged 15.5 yr (±1.7 SE). For the 72 clams that grew sufficiently and included a sufficient shell marker at initiation to be used in our 2-yr tests, age at experimental initiation ranged from 0 to 17 yr and averaged 3.9 yr (±0.3 SE). Thus, the average age of the clams that could not be used for our tests was significantly (P < 0.01 in a t-test) higher than that of the 72 clams that were used. Most (16 of 17) of the excluded clams lacked both sufficient growth and an obvious disturbance check at initiation. Only one clam was excluded with sufficient growth but with- out an adequate disturbance check. Although 16 clams lacked sufficient growth to determine ac- curately the shell size at the initiation of the experi- ment and were therefore excluded from our tests, all of these clams possessed discrete bands in their shells that could be counted separately. They were, however, close together at the terminal margin of the shell where separating them was not always possible and caused some uncertainty in their age estimates. Discussion Our banding data from recovery of marked and measured M. jnercenaria after virtually 24 mo of terminal growth provide a compelling case for the reliability across different habitats of using major growth bands in sectioned shells to age hard clams in the Cape Lookout region of North Carolina. Our previous test of the annual periodicity of banding in North Carolina's M. mercenaria (Peterson et al. 1983) was carried out in only single locality, a Back Sound seagrass bed, characterized by almost equal proportions by weight of sands and muds in its sur- face sediments (Tkble 1). Through this study, we ex- tend our tests of the reliability of annual band deposi- tion in M. mercenaria to several additional sites, located in different bodies of water and characterized by much sandier sediments (Ikble 1). Of the six sites that we used for these tests, three were vegetated by seagrasses and three lacked macrophytic cover (Tkble 1). (Although our initial experiment in Mid- dle Marsh was situated inside a seagrass bed, sea- grasses were removed from the experimental plots during each sampling.) Despite these differences in local geographic location (and probably salinity), sediment grade, and seagrass presence, banding pat- terns were consistent and bands were deposited annually. By using relatively high densities of 80 m"^ (over 10 times the average natural density found in a Bogue Sound seagrass bed by Peterson 1982, in North River, Bogue, Back, and Core Sounds by Beal 1983, and in Back Sound seagrass beds and sand flats by Peterson et al. 1984), we ran the risk of causing inhibition of growth. In fact, we were unable to analyze growth band deposition in 17 of our 89 clams largely because of insufficient growth in the terminal 2-yr increment. This problem may have been induced by our choice of relatively high densi- ty in these experiments, but it does have a natural analog. In areas with relatively slow growth and in older age classes where growth rate slows, aging North Carolina's M. mercenaria by counting annual bands in shell cross-section may be more difficult and lead to greater error than the consistency of band- ing results on our other 72 clams implies (Tkble 3). Nevertheless, banding even in these generally older clams that were excluded from our analyses was discrete and sufficient to permit us to estimate their ages. Aging does not appear to imply cessation of annual band deposition but only an increased dif- 675 ficulty in distinguishing one band from another. Observations on the timing of annual band deposi- tion in this study agree well with our previouswesults (Peterson et al. 1983). The 6 August-3 September period in 1980 consistently fell within the season of annual band deposition and near the end of the band at all five new study sites. P\irthermore, the annual band was still being formed or had just been com- pleted in all clams collected 24 August-3 September 1982. The banding of M. mercenaria in North Carolina appears to be a summertime event in con- trast to the winter banding in northeastern popula- tions (Barker 1964; Pannella and MacClintock 1968; Rhoads and Pannella 1970; Clark and Lutz 1982). The only clams that failed to deposit two additional annual bands in the 24 mo of this study were taken from the North River locals. This is also the only group that deviated in the timing of final band deposition relative to the 1982 collection date Most of these clams had just initiated their 1982 bands at the time of collection in contrast to those from all other sites where 1982 band depositon was either far advanced or even terminated. This difference be- tween sites is probably a consequence of the 3-4 wk earlier date of collection at North River (Tkble 2). Despite an identical, early collection date, the Back Sound clams transplanted to North River exhibited a pattern of band deposition in 1982 that more close- ly resembled the other four sites than did the North River local clams. This difference provides our only suggestion of an effect of clam origin, but we have no explanation for the possible effect and do not con- sider it a serious cause to doubt the consistency of annual band deposition in North Carolina's M. mercenaria. The tests of consistency of annual band deposition across habitats in a local estuarine system provide an additional source of confidence in the accuracy of using internal banding patterns to age M. mercenaria in the Cape Lookout region of North Carolina. Concern over the lack of such controlled tests had earlier prompted Clark (1974), Gould (1979), and Jones (1981) to question the widespread assumption of regular periodicity in repeating shell features. Our demonstration of consistency in annual banding across local habitats should remove any doubts about the general applicability of using an- nual bands to age M. mercenaria in the Cape Look- out region of North Carolina. The variation in line deposition patterns which has been shown across habitats for Protothaca staminea (Peterson and Am- brose 1985) and among individuals within habitat for P. staminea and Chione fluctifraga (Hughes and Clausen 1980) does not exist for M mercenaria near Cape Lookout. Our results will not only enable inver- tebrate fisheries biologists to use growth bands with confidence to age North Carolina's M. mercenaria but also should stimulate further research on under- standing the environmental causes of variation in bivalve shell deposition patterns. Acknowledgments K. Bowers, M. E. Colby S. R. Fegley C. Furman, C. Groat, S. A. Hughes, K. C. Pierce, G. W. Safrit, Jr., S. Smith, N. T Sterman, and J. Tbcker provided field and laboratory assistance V. Page drafted and H. Page photographed Figure 1, adapted from Beal (1983). Reviews by S. R. Fegley W. Sutherland, and M. C. Watzin improved the paper. This study was sponsored by the Office of Sea Grant, NOAA, U.S. Department of Commerce, under grant No. NA81AA-D-00026, North Carolina Department of Administration. Literature Cited Barker, R. M. 1964. Microtextural variation in pelecypod shells. Malaco- logia 2:69-86. Beal, B. F. 1983. Effects of environment, intraspecific density, predation by snapping shrimp and other consumers on the population biology of Mercenaria mercenaria near Beaufort, North Carolina. M.S. Thesis, Univ. North Carolina, Chapel Hill, 180 P- Brett, C. E. 1963. Relationships between marine invertebrate infauna distribution and sediment type distribution in Bogue Sound, North Carolina. Ph.D. Thesis, Univ. North Carolina, Chapel Hill, 212 p. Clark, G. R., H. 1974. Growth lines in invertebrate skeletons. Ann. Rev. Earth Planet. Sci. 2:77-99. 1979. Seasonal grow^th variations in the shells of recent and prehistoric specimens of Mercenaria mercenaria from St. Catherine's Island, Georgia. Am. Mus. Nat. Hist. Anthropol. Pap. 56:161-179. Clark, G. R., II, and R. A. Lutz. 1982. Seasonal patterns in shell microstructure oi Mercenaria mercenaria along the U.S. Atlantic coast. Geol. Soc Am. Abstr with Programs 14:464. Folk, R. L. 1974. Petrology of sedimentary rocks. Hemphill Publ. Co., Austin, TX, 182 p. Fritz, L. W., and D. S. Haven. 1983. Hard clam, Mercenaria mercenaria: Shell growth pat- terns in Chesapeake Bay Fish. Bull., U.S. 81:697-708. Gould, S. J. 1979. Time's vastness. Nat. Hist. 88(4):18-27. Hughes, W. W., and C. D. Clausen. 1980. Variability in the formation and detection of growth increments in bivalve shells. Paleobiology 6:503-511. Jones, D. S. 1980. Annual cycle of shell growth increment formation in two 676 continental shelf bivalves and its paleoecologic significance Paleobiology 6:331-340. 1981. Repeating layers in the moUuscan shell are not always periodic J. Paleontol. 55:1076-1082. Kennish, M. J. 1980. Shell microgrowth analysis. Mercenaria mercenaria as a type example for research in population dynamics. In D. C. Rhoads and R. A. Lutz (editors), Skeletal growth of aquatic organisms: Biological records of environmental change, p. 255-294. Plenum Press, N.Y. Kennish, M. J., and R. K. Olsson. 1975. Effects of thermal discharges on the microstructural growth of Mercenaria merce7iaria. Environ. Geol. 1:41-64. Pannella, G., and C. MacClintock. 1968. Biological and environmental rhythms reflected in molluscan shell growth. Paleontol. Soc. Mem. 2:64-80. [J. Paleontol. 42 (Suppl. to No. 5)]. Peterson, C. H. 1982. Clam predation by whelks (Biisycon spp.): experimen- tal tests of the importance of prey size, prey density, and sea- grass cover. Mar. Biol. (Berl.) 66:159-170. Peterson, C. H., and W. G. Ambrose, Jr. 1985. Potential habitat dependence in deposition rate of presumptive annual lines in shells of the bivalve Protothaca staminea. Lethaia 18:257-260. Peterson, C. H.. P. B. Duncan, H. C. Summerson, and G. W. Safrit, Jr. 1983. A mark-recapture test of annual periodicity of internal growth band deposition in shells of hard clams, Mercenaria mercenaria, from a population along the southeastern United States. Fish. Bull., U.S. 81:765-779. Peterson, C. H., H. C. Summerson, and P. B. Duncan. 1984. The influence of seagrass cover on population structure and individual growth rate of a suspension-feeding bivalve, Mercenaria mercenaria. J. Mar. Res. 42:123-138. Rhoads, D. C., and R. A. Lutz (editors). 1980. Skeletal growth of aquatic organisms: Biological records of environmental change Plenum Press, N.Y., 750 p. Rhoads, D. C., and G. Pannella. 1970. The use of molluscan shell growth patterns in ecology and paleoecology. Lethaia 3:143-161. Rosenberg, G. D., and S. K. Runcorn (editors). 1975. Growth rhythms and the history of the earth's rotation. John Wiley and Sons, Lond., 559 p. Sutherland, J. P., and R. H. Karlson. 1977. Development and stability of the fouling community at Beaufort, North Carolina. Ecol. Monogr. 47:425-446. Thayer, G. W. 1971. Phytoplankton production and the distribution of nutrients in a shallow unstratified estuarine system near Beaufort, N.C. Ches. Sci. 12:240-253. Williams, A. B., G. S. Posner, W. J. Woods, and E. E. Deubler, Jr. 1973. A hydrographic atlas of large North Carolina sounds. Univ. North Carolina Sea Grant Publ. UNC-SG-73-02, 130 p. (U.S. Fish Wildl. Serv., Data Rep. 20, 130 p.) Charles H. Peterson P. Bruce Duncan Henry C. Summerson Brian F. Real Institute of Marine Sciences University of North Carolina at Chapel Hill Morehead City, NC 28557 STANDING STOCK OF JUVENILE BROWN SHRIMP, PENAEUS AZTECUS, IN TEXAS COASTAL PONDS The increased demand for timely information con- cerning management of shrimp stocks has renewed interest in developing reliable methods of predicting brown shrimp, Penaeus aztecus, crop size for the off- shore Gulf of Mexico fishery. Advance information regarding shrimp abundance would also enable elements of the shrimp industry to prepare for a potentially good or poor harvest. Studies exploring the feasibility of predicting the annual abundance of brown shrimp off the Tfexas coast, initiated in 1960 (Baxter 1963), are based on the premise that post- larval and juvenile shrimp abundances are propor- tionally related to the subsequent commercial harvest (Berry and Baxter 1969). A "good" predictor is one that is precise, timely, and cost effective The abundance of postlarval shrimp as they move from the Gulf of Mexico into coastal bays is determined from semiweekly collec- tions made by the National Marine Fisheries Ser- vice, Galveston, at the entrance to Galveston Bay be- tween late February and early May (Baxter 1963). The postlarval shrimp index gives the earliest but least reliable indication of potential harvest. A more accurate but less timely prediction is derived from landings of the bait shrimp fishery. Statistics for bait shrimp landings since 1960 provide the basis for a predictive model developed by K. N. Baxter (Klima et al. 1982) defining the relationship between the bait abundance index and subsequent offshore catch. However, this prediction is not available until mid- June, just prior to the seasonal opening, because recruitment of brown shrimp into the bait fishery does not begin until May (Chin 1960). A third possi- ble indicator is the standing stock size of juvenile shrimp in estuarine nursery areas measured before shrimp emigrate and become vulnerable to the bait fishery. This would provide an estimate earlier in the season than the bait index and may be a more ac- curate predictor than the postlarval abundance Predictive capability increases with each successive life stage because of the decreased time span be- tween the estimate and subsequent commercial harvest. Tb examine the relationship between juvenile brown shrimp standing stock and offshore harvest, and to determine the suitability of juvenile brown shrimp abundance as a predictor, we conducted a mark-recapture study in Galveston Bay, TX, during the first week of June 1983. In this report we sum- marize the results of our study, compare estimates FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 677 obtained by mark recapture and an alternative drop sampler method, and discuss previously unreported results of 1970-71 studies (Welker and Baxter^). Methods Sydnor Bayou is a shallow coastal tidal pond in Galveston Bay (Fig. 1). The site was chosen because the single narrow entrance could be blocked easily with netting, thus preventing immigration and emigration of shrimp during the experiment, and because Sydnor Bayou was the site of a similar study in 1970. The pond covers 32.4 to 36.4 ha, depending on the tide, is about 0.9 km long and 0.2 km at its widest point, narrowing to 6 m at the mouth. Maximum depth is about 1.5 m at high tide, with a 0.25 m tidal difference Average salinity during the marking was 20.5 ppt and mean surface temperature was 28°C. Weekly sampling of Sydnor Bayou with a 3 m otter trawl (25 mm stretched mesh) began 25 April 1983 to monitor the size of the juvenile shrimp. By 23 May 1983, most shrimp caught in the trawl were larger than the 40 mm TL (total length) minimum needed for tagging, and we decided to begin the mark- recapture experiment the next week. Sydnor Bayou was blocked at dawn on 31 May 1983 across bridge B-1 (Fig. 1) with a 45.7 m net having a 6 mm mesh. The net was anchored to the bottom and remained in place for the duration of the experiment. A 1.8 m diameter, 0.8 m deep round tank with con- tinuous water flow and two 147 L aerated ice chests were set up on shore to hold shrimp during the mark- ing process. Shrimp were caught in 49 5-min trawl hauls and transported to the marking site in aerated 45 L ice chests. lb minimize marking mortality, only shrimp 40 mm TL and larger were marked and held in the large tank. Marking was accomplished by in- jection with pink fluorescent pigment as described by Klima (1965). Representative length -frequency, species-composition, and sex ratio information was obtained from shrimp captured in one trawl haul. Marked shrimp were released within the hour after the target number (4,100) had been marked. Shrimp were scattered along the shallow grassy shoreline from moving skiffs. No dead or moribund shrimp were released, and release operations ceased whenever shore birds were attracted. Four 61 cm x 61 cm x 20 cm wire cages, each SYDNOR BAYOU 36.4 HA CORPUS CHRIS COW TRAP 1 AND 2 27.5 AND 2.8 HA MUD LAKE 6.4 HA Figure 1.— Tfexas ponds selected for brown shrimp mark-recapture studies: Sydnor Bayou (1970 and 1983); Cowtrap, Mushroom, Caran- cahua, and Mud Lake (1971). containing 25 marked and 25 unmarked shrimp, were set out in the pond and remained submerged through all tidal stages. After 24 h, cages were raised and all shrimp, dead and alive, were counted, measured, and recorded for an estimate of marking mortality. Recapture trawling began 18 h later, allowing marked shrimp time to distribute themselves in the unmarked population. For three consecutive days, all trawlable bottom was sampled by 5-min trawl hauls. Shrimp were returned to the laboratory where marked shrimp were identified under ultraviolet light. All marked and up to 100 unmarked recoveries were measured per tow. Length-frequency distribu- tions for releases, marked recoveries, and unmark- ed recoveries are shown in Figure 2. We estimated an initial population of juvenile brown shrimp using Bailey's (1951) modification of the Petersen formula ^Welker, W., and K. N. Baxter. Juvenile brown shrimp popula- tion estimates in Itexas tidal marsh ponds. Unpubl. manuscr., 8 p. Southeast Fisheries Center Galveston Laboratory, National Marine Fisheries Service, NOAA, 4700 Avenue U, Galveston, TX 77550. 678 N = M (C + 1) R + 1 N=122 where M = number of marked shrimp released, corrected for marking mortahty, C = number of shrimp examined for marks, R = number of recaptured marked shrimp in the sample The 95% confidence limits for the true population were estimated using the standard error of the large sample variance formula (Bailey 1951) V(N) = M^jC + 1){C - R) (R + 1)2 (R + 2) Application of this method is justified under the following conditions (Ricker 1975): 1) Marked shrimp suffer the same natural mor- tality as unmarked. 2) Marked and unmarked shrimp are equally vulnerable to fishing. 3) Marked shrimp do not lose their mark. 4) Marked shrim.p become randomly distributed among unmarked. 5) All marks are recognized and reported on recovery. 6) There is not emigration or immigration occur- ring in the catchable population. Results and Discussion Overall marking mortality was 9%. One cage had unusually high mortality. Nineteen of 25 marked shrimp were alive at the end of 24 h, and the only evidence of the other 6 marked shrimp was pieces of exoskeleton. They apparently molted and were cannibalized. Holt (1982) suggested that the condi- tion of shrimp prior to tagging dictates the survival of the tagged animals. When stressed animals were tagged, mortality more than doubled. Howe and Hoyt (1982) hypothesized that tags and marks may indirectly cause mortality by attracting predators. Farmer and Al-Attar (1979) found shrimp marked with subcutaneous pigment suffered high mortality (compared with controls) only when held with un- marked conspecifics. Clark and Caillouet (1973), however, found negligible marking mortality in a mark-recapture experiment with white shrimp, R setiferus, when 50 marked and 50 unmarked con- trol shrimp were held in a large pen in a pond rather than in several small cages. Costello and Allen (1962) stated that stained shrimp may be expected to sur- 1 20 40 60 80 100 TOTAL LENGTH (MM) 120 Figure 2.— Length-frequency distribution of brown shrimp in Syd- nor Bayou, June 1983: A) representative sample of shrimp col- lected during marking; B) unmarked shrimp caught during recap- ture operations; and C) marked shrimp caught during recapture operations. vive at essentially the same rate as unmarked shrimp, regardless of presence of predators, lb avoid overestimating marking mortality, we did not include the counts in the high cage in the calculation. The resulting 4% (3 dead marked shrimp out of 75) was similar to the marking mortalities of past studies in Sydnor Bayou, Mud Lake, and Mushroom (Welker and Baxter fn. 1). 1983 Population Estimate A total of 223 marked shrimp were among 12,304 shrimp caught in 94 recapture tows. Tides during the recovery period were low in the morning, ap- proaching high tide in the afternoon. Areas along the shore and the south end of the bayou were shallow for trawling in the mornings, but could be adequately sampled in the afternoon. Distribution 679 of marked shrimp was random (one-sample runs test, P = 0.960; Siegel 1956). ^ The population estimate of 207,786 shrimp deter- mined from mark-recapture data compared favorably with the results of a concurrent drop sampler ex- periment (Ikble 1). Shrimp densities were obtained using a 2.8 m^ drop sampler at high tida Detailed methodology has been described by Zimmerman et al. (1984). Drop samples were taken in two sets, four pairs each, in vegetated and nonvegetated areas, divided between the south and north ends of the bayou. Vegetated habitat was sampled along the bayou margins, while nonvegetated area sampling was in open waters of the bayou. Numbers of shrimp within the sampler were extrapolated to represent the shrimp population in the vegetated, nonvege- tated, and total areas of Sydnor Bayou. Confidence intervals for the drop sampler were much wider than those for Petersen estimate because drop sampler estimates were based on a small number of samples. The drop sampler estimate for 36 ha was higher by about 92,000 shrimp. One reason for this difference is that the mark-recapture estimate reflects only that part of the population >40 mm TL, while the drop sampler measures density of small (<40 mm) shrimp more effectively, and these small shrimp are included in the estimate (Ikble 2). We calculated the drop sam- pler population estimate using only shrimp larger than 40 mm TL (Tkble 1). A chi-square test shows a significant difference between the drop sampler and mark- re capture size-frequency samples, cate- gories 41-50 mm and higher (x^ = 109.45, df = 6, P very small). The high chi-square value is due main- ly to the greater number of 41-50 mm shrimp and the lower number of larger shrimp (81-90 mm), which may avoid the sampler, in the drop sampla Length- frequency composition of the drop sampler catch in- dicates that 23% of the 103 shrimp taken were smaller than 40 mm TL, while no shrimp smaller than 40 mm were captured by the otter trawl. 1970-71 Population Estimates Our methodology for conducting a Petersen single census mark-recapture experiment with juvenile brown shrimp was developed during June and July 1971 studies of five Tbxas coastal ponds (Fig. 1). All ponds ranged from 0.3 to 0.9 m in depth during a normal summer tidal (ycle Cow Trap 1 and 2 had considerable emergent vegetation along their shore- lines and were part of a large marsh complex. Ex- tensive flooding of the marsh surrounding these ponds at flood tide greatly increased the area ac- cessible to shrimp, but this shallow, vegetated area Table 1. — Sydnor Bayou brown shrimp population estimates determined by mark-recapture and drop sampler methods, June 1983. Estimated Method population 95% C.I. Mark-recapture^ 32.4-36.4 ha 207,786 180,884-234,688 Drop sampler 32.4 nonvegetated ha 185,000 41 ,900-479,000 4.0 vegetated ha 115,000 49,000-248,000 36.4 total ha 300,000 90,800-727,000 Drop sampler' 32.4 nonvegetated ha 157,000 113,000-423,000 4.0 vegetated ha 88,000 53,500-183,000 36.4 total ha 245,000 166,000-606,000 'Estimate of shrimp population >40 mm TL. Table 2. — Length-frequency composition of Sydnor Bayou brown shrimp samples taken with the otter trawl (N = 8,197) and drop sampler (W = 83), 1-3 June 1983. Length Otter trawl Drop sampler (mm) (0/0) (0/0) <20 0.0 9.7 21-30 0.0 1.9 31-40 0.0 11.7 41-50 3.8 19.4 51-60 17.3 19.4 61-70 32.2 22.3 71-80 25.7 12.6 81-90 16.0 1.9 91-100 4.2 0.0 >100 0.6 0.8 could not be sampled. Shrimp could move from pond to pond via flooded marsh and ditches, rendering block nets ineffective Evidence of this movement was the netting of marked shrimp released in Cow Trap 1 and recaptured in Cow Trap 2. These problems precluded reasonable population estimates for the Cow Trap ponds, and large marsh complexes were avoided for future studies of this typa Mud Lake, Carancahua, and Mushroom had generally well-defined shorelines, even during flood tide, and were not contiguous with other ponds or ditches. Mark-recapture methods were essentially the same as described for the 1983 study. Marking and holding operations were conducted on a portable barge rather than from shore (Emiliani and Marullo 1973). Population estimates determined by Bailey's (1951) formula ranged from 7,490 to 17,119 brown shrimp per hectare (Ikble 3). The lowest estimate was recorded in Mud Lake, where the highest percentage of total catch was <40 mm TL, while the highest estimate was for Carancahua. The density in Mushroom was close to that in Carancahua. Although marking methods differed, a 1970 mark- 680 recapture study in Sydnor Bayou provided a popula- tion estimate for comparison. Marking was accom- plished by spraying shrimp >40 mm TL with granular fluorescent pigment (Benton and Lightner 1972). Data analysis was as described for the 1983 Sydnor Bayou study. The average density of shrimp in Sydnor Bayou during the 1983 study was 37% of the May 1970 density and was the lowest per hec- tare estimate of any pond previously sampled (Tkble 3). We believe that juvenile brown shrimp population density, determined by the mark-recapture method, may prove to be a good predictor of offshore pro- duction as we compile a longer term data base Although the drop sampler (area-density method) may measure shrimp density more accurately, the Peterson mark-recapture method gives a more precise (having less variance) population estimate Acknowledgments We thank the many people who helped us in the field and processing shrimp; also, Roger Zimmer- man, for supplying drop sampler data. We are especially grateful to Don Hanson for allowing us the use of his property along Sydnor Bayou during the study. Literature Cited Bailey, N. J. 1951. On estimating the size of mobile populations from recap- ture data. Biometrika 38:293-306. Baxter, K. N. 1963. Abundance of postlarval shrimp - one index of future shrimping success. Proa Gulf Caribb. Fish. Inst. 15:79-87. Benton, R. C, and D. Lightner. 1972. Spray marking juvenile shrimp with granular fluo- rescent pigment. Contrib. Mar. Sci. 16:65-69. Berry, R. J., and K. N. Baxter. 1969. Predicting brown shrimp abundance in the north- western Gulf of Mexico. FAO Fish. Rep. 57, 3:775-798. Chin, E. 1960. The bait shrimp fishery of Galveston Bay, Ifexas. TVans. Am. Fish. Soc 89:135-141. Clark, S. H., and C. W. Caillouet, Jr. 1973. White shrimp {Penaeus setiferus) population trends in a tidal marsh pond. Mar. Fish. Rev. 35(3-4):27-29. Costello, T. J., and D. M. Allen. 1962. Survival of stained, tagged, and unmarked shrimp in the presence of predators. Proa Gulf Caribb. Fish. Inst. 14:16-19. Emiliani, D. a., and F. Marullo. 1973. Portable barge for estuarine research. Mar. Fish. Rev. 35(l-2):27-29. Farmer, A. S. D., and M. H. Al-Attar. 1979. Results of shrimp marking programmes in Kuwait. Kuwait Bull. Mar. Sci. 1:1-32. Holt, B. 1982. Short term mortality of tagged shrimp during field tag- ging experiments. U.S. Dep. Commer., NOAA Tfech. Memo. NMFS-SEFC-97, 9 p. Howe, N. R., and P. R. Hoyt. 1982. Mortality of juvenile brown shrimp Penaeus aztecus associated with streamer tags. Trans. Am. Fish Soa 111: 317-325. Klima, E. F. 1965. Evaluation of biological stains, inks, and fluorescent pigments as marks for shrimp. U.S. Fish WOdl. Serv., Spea Sci. Rep. 511, 8 p. Klima, E. F, K. N. Baxter, and F. J. Patella, Jr. 1982. A review of the offshore shrimp fishery and the 1981 Ifexas Closure Mar. Fish. Rev. 44(9-10):16-30. RiCKER, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191, 382 p. SlEGEL, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, N.Y., p. 52-60. Zimmerman, R. J., T. J. Minello, and G. Zamora. 1984. Selection of vegetated habitat by brown shrimp, Penaeus aztecus, in a Galveston Bay salt Marsh. Fish. Bull., U.S. 82:325-336. Loretta F Sullivan Table 3.— Summary of juvenile brown shrimp population studies in Texas coastal ponds. Number Percent Number Number marked released Percent 40+ mm 95% Start 5-min shrimp Percent and and marking population confidence Location date tows caught <40 mm released recovered mortality per hectare intervaP Sydnor Bayou 32.4 ha 5/31/83 49 5,188 0.3 3,994 5.9 4.0 6,412 5,583-7,244 36.4 ha 5,709 4,970-6,448 Sydnor Bayou 32.4 ha 5/21/70 32 8,045 — 7,718 1.7 4.0 17,933 14,198-20,042 36.4 ha 15,238 12,637-17,839 Mud Lake 6.4 ha 6/3/71 27 6,750 20.0 6,120 10.8 4.0 7,490 6,956-8,025 Carancahua 3.5 ha 6/7/71 26 6,301 7.0 4,574 9.8 8.0 15,697 11,815-17,087 Mushroom 1.8 ha 7/2/71 24 8,348 6.0 4,142 28.8 4.0 14,375 13,628-15,120 'Bailey (1951) large sample variance. 681 Southeast Fisheries Center Galveston Laboramry National Marine Fisheries Service, NOAA Jt700 Avenue U Galveston, TX 77550 Present address: 1 78 Plaza Circle Danville, CA H526 Dennis A. Emiliani K. Neal Baxter Southeast Fisheries Center Galveston Laboratory National Marine Fisheries Service, NOAA 4700 Avenue U Galveston, TX 77550 A POSSIBLE LINK BETWEEN COHO (SILVER) SALMON ENHANCEMENT AND A DECLINE IN CENTRAL CALIFORNIA DUNGENESS CRAB ABUNDANCE Dungeness crab, Cancer magister, are taken com- mercially along the west coast of the contiguous United States from Avila, CA, to Destruction Island, WA (Fig. 1). During the early years of the Califor- nia Dungeness crab fishery, effort was concentrated on the central California population which produced most of the state's landings (Fig. 2). The northern population subsequently became the major con- tributor to California's landings after an expansion of the fishery there during the 1940's. Northern California landings (Fig. 2) generally have followed a fluctuating pattern similar to one ex- pressed in Oregon and Washington; however, land- ings from the relatively isolated central California population failed to recover from a coastwide low during the early 1960's. The lower landings reflect a long-term reduction in abundance which has been variously attributed to egg predation by a nemer- tean worm Carcinonemertes errens (Wickham 1979) and to the effects of a long-term change in oceanic conditions (Wild et al. 1983). The failure of the central California population to recover from the coastwide period of low abundance also occurred about the time coho salmon, Oncorhyn- chus kisutch, reared in Oregon and Washington hatcheries began to make a significant contribution to the west coast salmon fishery (Oregon Depart- ment of Fish and Wildlife 1982). The effect of salmonid predation on commercially important marine crustaceans has received little attention, although it is suspected that predation by salmonids introduced into a number of both small and large freshwater lakes (Nilsson 1972; Morgan et al. 1978) has substantially altered the abundance and species composition of their planktonic crustacean com- munities. Since numerous salmonid food habit studies (Heg and Hyning 1951; Petrovich 1970; Reilly 1983a) show that planktonic Dungeness crab megalops are a major component of the coho salmon diet, it is conceivable that an increase in the coho predation rate associated with an influx of hatchery coho into the central California region is at least par- tially responsible for the prolonged decline in Dungeness crab landings. In this paper I first present evidence showing that a large portion of the coho salmon ultimately caught each summer off the west coast are in California waters during spring, the period Dungeness crab megalops are most abundant. I then compare and contrast survival indices to determine if the temporal variation in survival of both species is consistent with the predator-prey hypothesis. Oregon Production Index Area Coho Each spring and summer, a single coho salmon brood (year class) is recruited to the commercial salmon fishery off California, Oregon, and southern Washington, an area collectively referred to as the Oregon Production Index area or O.P.I, area (Oregon Department of Fish and Wildlife 1982). These fish entered the ocean to feed in May and June of the previous year, after having spent about 18 months in freshwater. Coho caught in the O.P.I, area before 1961 (Fig. 3) were predominately wild stocks. These stocks had declined to extremely low levels by 1960; however, the successful introduction of Oregon and Washington hatchery-reared coho resulted in a return to historical landing levels during the 1960's and 1970's. Much of the hatchery fish responsible for the increased landings are derived from early return Tbutle River coho, which tend to enter fisheries south of their stream of origin (Hopley 1978). Coho salmon made up only 10% or less of Califor- nia's ocean salmon catch prior to the development of Oregon and Washington enhancement programs (FVy 1973). Most of these wild coho originated in the streams and rivers of Oregon and Washington (Allen 1965) and were landed primarily in the northern California ports of Crescent City and Eureka. The recruitment of hatchery fish increased the average annual coho contribution to 25% of the total ocean salmon catch, with the central California ports of San Francisco and Fort Bragg accounting for a con- siderably larger portion of the total coho catch. 682 FISHERY BULLETIN: VOL. 83, NO. 4, 1985. Figure 1.— Commercial fishing areas for Dungeness crab off Washington, Oregon, and Cahfornia. (Pacific Fishery Management Council (1979).) Before 1973, the California salmon season (coho and Chinook) opened on 15 April, although few coho were landed before June because of a minimum size restriction. As Oregon and Washington hatchery coho became available, a substantial increase in the hook and release of sublegal ("shaker") fish developed during the latter half of April to the middle of Juna In an attempt to reduce the shaker problem, the season opening for coho was delayed until 15 May and the minimum size was reduced in 1973 (O'Brien and Lesh 1975). California coho catches generally peak in July, then drop sharply in August, 2 mo before the salmon season closure The midsummer decline is attributed to the northward exodus of fish returning to their natal stream to spawn (Fry 1973). It is however unclear when and by what route these fish entered California waters. A general migration model (Loeffel and Forster 1970; Godfrey et al. 1975; Hartt 1980) proposes that newly emigrated west coast coho move immediate- ly northward into the Gulf of Alaska, then during winter, undertake a southeasterly migration which brings them back into California, Oregon, and Wash- 683 10 < Hi o r; t/J o f ) to a «i 2 i^ ti, ^ c 00 m CO r^ 00 02 >, T— I -a m 3 ho 0) O 5 PQ T— I lO cS o ?. o "~^ m n! C U B m c i3 £ -a c ^ (fl r/l -o hn i-; S' S B3 CO 3 m u c fc " bo c O CO O 0) U C 3 a .3 -3 .2 % ^ i !- . . 2 I gffi gOI. X SNOl Diyi3W 684 m r>. o (]} e o r^ •s 01 § 3 O in J3 (O C ,< ■~ (M _- « •-H OS 00 ^ Oi — ^ £ o ^ 3 u> bD-a 3 :73 ti-S: J3 -w -a §§ in 2-S c^-i^ O 4> •*3 Q o _ 3 C -o P o bo in ^ O § F t* o 0) k. i, «« O^ o ^ •>-' a .£"§ be ^ .£ a in c £ CO ^ CQ C C o O E ^ 13 at o en % 1 CO a » in p (M o .OL X SNOl Diai3W 685 ington coastal waters each spring. Coho returning to the coast south of their natal stream would subse- quently undergo the observed northward spawning migration. Each of these authors conceded however, that certain stocks or different portions of the same stocks may follow entirely different migratory routes. Scarnecchia (1981) felt that many coho pro- duced along the west coast may either remain in ad- jacent coastal waters or move directly south after emigration from freshwater. The dispersal pattern of 437 coho salmon tagged off northern California early in the 1971 and 1972 seasons clearly showed the northward movement of adult coho out of California (O'Brien 1973). Nearly all of the tagged fish recaptured in California were caught in May and June of each year while tagging was still being conducted. Oregon recoveries peak- ed in July and the first half of August, while most Washington recoveries occurred during the latter half of August through September. California's share of the recaptured coho tagged off northern California (O'Brien 1973) was 9.3% in 1971 and 8.8% in 1972. These percent returns are very similiar to California's 13.0% and 8.3% share of the O.P.I, area catch in 1971 and 1972 respective- ly. Because, for practical purposes, one can assume that all of the coho caught in California originate to the north, the similarity between California's catch and tag returns would indicate that nearly all O.P.I, area coho stocks were off California during the tag- ging period. This supposition is extreme, but the results do suggest that a major portion of the coho ultimately caught in the O.P.I, area each year are in California waters during spring. The northward migration of large numbers of coho is further sup- ported by the northward progression of peak monthly catches within the O.P.I, area (Pacific Fishery Management Council 1983), and the monthly catch distribution of hatchery marked coho (Hopley 1978). Survival Indices Comparison and Discussion Dungeness crab, unlike coho salmon, do not move any appreciable distance, therefore local landings are considered to be a good indicator of local abundance In California seasonal landings are composed of at least three year classes, however northern Califor- nia landings are generally dominated by 4-yr-old crab (Warner 1985), while central California landings, because of a faster growth rate, are dominated by 3-yr-old crab (Collier^). An alignment of Dungeness crab seasonal landings with their dominant or "primary" year classes (Fig. 4) generates reasonably representive year class in- dices, if it is taken into consideration that extreme- ly abundant year classes, such as the 1966 and 1972 year classes in northern California, probably dominate landings for more than 1 yr (Methot and Botsford 1982). The Dungeness crab year class in- dices (Fig. 4) suggest that a period of poor landings in both central and northern California during the early 1960's (Fig. 2) reflects poor survival of the 1958-60 year classes. As mentioned earlier, northern California landings have been characterized by large seemingly cyclic fluctuations, the cause of which has been the sub- ject of considerable research and debate (see Methot and Botsford 1982; Botsford 1984, for a review of this work). Of the hypotheses generated by these in- vestigations. Wild et al. (1983) presented, in my opin- ion, the most tenable explanation for this particular period of low survival. They attributed the drop dur- ing this period to a reproductive failure caused by an unprecedented warming of coastal waters associated with the 1957 El Nino (the "warm water years" 1957-59, Radovich 1961). The apparent recruitment of Dungeness crab to the northern California population of a "normal" year class in 1961 (Fig. 4), with the return of "nor- mal" environmental conditions, ushered in several years of good survival. This recovery was not duplicated in the central California population, where a drop in the strength of the 1961 year class anteceded an extended period of poor survival. Wild et al. (1983) further proposed that a major change in the oceanic regime off central California is the primary cause of the continued poor survival there, although they do concede that ocean temperatures in certain years appear to have been favorable to Dungeness crab survival. Wickham (1979), on the other hand, suggested that the central California population has reached a new equilibrium, with worm predation now being the dominant biological control. It has yet to be proven which, if either, of these mechanisms is the primary cause of the con- tinued poor survival in central California. Alternatively, a direct comparison of O.P.I, catches with central California Dungeness crab year class indices (Fig. 5) illustrates a long-term inverse rela- tionship which developed with the first recruitment of hatchery-reared coho salmon stocks in 1961. O.P.I, area landings are used to express the annual sur- vival of coho potentially impacting central Califor- 'P. Collier, California Department of Fish and Game, 619 Second St., Eureka, CA, 95501, pers. commun. November 1984. 686 < _l o q: < UJ >- >- < Q. T3 C OS ►J B o a? :S to bo .S ca J I cS -^ c cS O Q CO cS C CO a> . — . u o >, °9 ^ -* -o OS C »—* c^ to S en r C3 O m _ m a a) .— C s_ C 00 - ho >^ ^ ^ CQ CS — Q 0) 1 O ,01 X SNOi OiaiBW 687 CO < a: O NORTHERN CALIFORNIA j:;::: central California i: — o ID 9 1 o X o o >■ a u z o »- < 0 u X « Q to in ■ z u. tM Ifl -J c u 3 00 1 CM 1 1 1 O 00 (0 * CM CM »t I CO 1 00 O CM I Tt 0) & 13 r~ t- t^ Tf Oi f— 1 cd o Li o (fl on >vOO !^ Cfl £ be a. c CD -a -H c o m H cfl -H o ed m y 0) 1 C c 3 o Q 1^ oi 3 1 -a i •a u "(3 c 0) t) TS Gl3 g 0) .c t: o 1 U) m a D O ,01 X SNOi Oiai3W 688 nia Dungeness crab because of the evidence that California catch statistics underestimate the number of fish actually in the state during spring. These land- ings provide a straightforward measure of brood sur- vival that is independent of distribution and local catchability. Even though the relatively low O.P.I, area landings in 1961, 1962, 1963, 1977, and 1980 are comparable with the predecline era, the general pattern of cor- respondence in Figure 5 is consistent with an in- crease in the coho salmon predation rate on Dunge- ness crab megalops. Within the framework of the predator-prey hypothesis, the association of low O.P.I, area coho catches during the early years of the hatch- ery era with reduced Dungeness crab survival would indicate that a relatively small number of hatchery coho can effectively surpress megalops survival. This is particularly apparent when it is considered that hatchery production was at a minimum during the 1961-63 period and wild fish still dominated the catch (Oregon Department of Fish and Wildlife 1982). The proposed impact of hatchery coho salmon on the Dungeness crab resource is best explained by the differences in the "functional response"^ of wild and hatchery coho salmon. In controlled behaviorial ex- periments, Glova (1978) found that hatchery fry (43-88 mm) were largely nonterritorial, exhibiting a stronger tendency to aggregate than the wild fry. This behavioral pattern is believed to be the direct result of the unnaturally high densities found in hatchery operations. If adult hatchery coho retain this behavior, the tendency for Dungeness crab megalops to aggregate or "swarm" in coastal sur- face waters (Lough 1976) would theoretically make them more susceptible to predation (Eggers 1976). Also a reduction in the number of "search images" available to hatchery fish is believed to result in a more homogenous diet (Sosiak et al. 1979). Under these circumstances Dungeness crab megalops may become a more important component of the hatch- ery coho salmon diet. The apparent good survival of the 1961-66 year classes in northern California (Fig. 5) suggests that the majority of the hatchery coho salmon produced during those years concentrated to the south of that population during the period when Dungeness crab megalops are most abundant. This supposition, together with recently acquired evidence that the central California Dungeness crab population is at least partially dependent on the recruitment of southward drifting megalops (Hatfield 1983; Reilly 1983b), further suggest that the theoretical predation zone critical to the central California population lies somewhere in the region of strong upwelling and high productivity between the two populations (Fig. 1). Not surprisingly, commercial fishermen have found coho salmon concentrated either before or early in the season in this region. The coho salmon stocks initally released during the early 1960's may possess an inate affinity for these waters. Northern California landings of the Dungeness crab declined again during the 1970-71 season (Fig. 2). This period of low landings is apparently due to poor survival of the 1967-71 year classes (Fig. 5), which cannot be readily explained by an extended period of warmer than normal water. The various hypotheses to explain the northern California fluc- tuations notwithstanding, it is possible that hatchery- reared coho salmon began to limit Dungeness crab survival in northern as well as central California, concomitant with increased hatchery production^ and/or environmental caused changes in distribution. There is some evidence from coho tagging that sup- ports this supposition. O'Brien (1973) reported that 17.3% of his return- ed tags were found in Oregon and Washington hatch- eries during the 1971 season, whereas in 1972 only 3% were found in the hatcheries. An exceedingly strong 1972 Dungeness crab year class in northern California (Warner 1984) is in direct contrast with the very weak 1971 year class (Fig. 5) and is inversely related to the small number of tags found in hatch- eries during the 1972 season. The small percentage of hatchery returns in 1972 suggest that there were fewer hatchery coho available for tagging in the northern California area during the 1972 season, and this could indicate relatively poor survival of hatch- ery fish throughout the O.P.I, area. It should be remembered that hatchery-reared coho theoretical- ly have a much larger effect on Dungeness crab sur- vival than wild fish. Between 1972 and 1977 (Fig. 5), O.PI. area coho survival and northern California Dungeness crab survival became more erratic. The association of relatively good Dungeness crab survival with good coho landings in 1974 and 1976 may, however, only indicate that coho were farther south than usual. McLain and Thomas (1983) showed that both 1973 ^In predator-prey theory "functional response" is defined as the relationship between the rate at which individual predators con- sume prey and the density of that prey (Holling 1959). ^The number of hatchery-reared coho salmon released in the O.P.I, area increased from 7.5 million fish in 1960 to 60.8 million fish in 1981 (Oregon Department of Fish and Wildlife 1982). 689 and 1975 were years with an unusually weak Califor- nia Countercurrent, or conversely, stronger than nor- mal southward flow and cooler than normal coastal waters. If yearling coho do move directly into Califor- nia waters after emigration from freshwater, then these anomalous conditions may have caused these fish to move farther south than usual, with the result that adult coho would have been south of the preda- tion zone critical to the northern California Dunge- ness crab population in the spring of 1974 and 1976. Since 1976 O.P.I, area coho landings have under- gone an inexplicable decline in spite of increasing hatchery production. Theoretically, an increase in Dungeness crab survival should have accompanied this drop in coho survival. The drop in Dungeness crab survival, evident in Figure 5, is obviously in- consistent with the general theory but can be ex- plained in two ways. First, it should be considered that the earlier O.P.I, area coho landings contained far fewer hatchery fish than those during the later years. It has been estimated that hatchery fish com- prised 75% of the west coast coho catch by 1977 (Scarnecchia and Wagner 1980). Secondly, a coastal warming trend that began in 1976 (McLain 1983) may have resulted in a northward shift in coho distribution with a concomitant reduction in Dunge- ness crab megalops survival. If coho have become the major limiter of Dunge- ness crab megalops survival within California, then the observed survival patterns suggest that a group of coho, possibly representing the original hatchery stocks, still experience consistently good survival and continue to move into the predation zone critical to the central California population. On the other hand, the more irregular Dungeness crab survival observed in northern California suggest that megalops survival there is more dependent on the vagaries of hatchery-reared coho distribution associated with environmental nuances. Admittedly, most of the evidence used to support the predator-prey hypothesis is circumstantial. Nevertheless, three of the considerations presented —1) the fact that coho feed heavily on Dungeness crab megalops, 2) the evidence showing that many, if not most, Oregon and Washington hatchery coho are in California during the period megalops are most abundant, and 3) the coincidence of the ex- tended central California Dungeness crab decline with a large increase in the number of hatchery coho within the O.P.I, area— suggest a possible relation- ship that deserves attention. The capricious nature of predation on the early life stages of commercially important invertebrates un- doubtedly contributes to the difficulties encountered when attempting to manage these relatively short- lived species on a sustained yield basis. If the hypo- thesized relationship between coho salmon and Dungeness crab eventually proves to be correct, then the salmonid enhancement process itself can be con- sidered an experiment, offering insight into the role predators play in controlling the commercial abun- dance of many marine species. Acknowledgments I would like to thank Bob Tksto, John Geibel, Ron Warner, and two anonymous reviewers for review- ing the manuscript. Special thanks are also due Bob Tksto, who offered considerable encouragement, and Anita Thomas for drafting the figures. Literature Cited Allen, G. 1965. Estimating error associated with ocean recoveries of fin-marked coho salmon. Trans. Am. Fish. Soc 94:319-326. BOTSFORD, L. W. 1984. Effect of individual growth rates on expected behavior of the northern California Dungeness crab {Cancer magister) fishery. Can. J. Fish. Aquat. Sci. 41:99-107. Eggers, D. M. 1976. Theoretical effect of schooling by planktivorous fish predators on rate of prey consumption. J. Can. Fish. Res. Board 33:1964-1971. Fry, D. H., Jr. 1973. Anadromous fishes of California. Calif. Dep. Fish Game, Resourc Agency, 111 p. Glova, G. J. 1978. Behavioral differences between wild and hatchery- produced coho salmon juveniles and their management im- plications. In B. G. Shepherd and R. M. J. Ginetz (Rap- porteurs), Proceedings of the 1977 Northeast Pacific Chinook and Coho Salmon Workshop, p. 84-88. Can. Fish. Mar. Serv, Tfech. Rep. 759. Godfrey, H., K. A. Henry, and S. Machidorl 1975. Distribution and abundance of coho salmon in offshore waters of the North Pacific Ocean. Int. North Pac Fish. Comm., Bull. 31:1-80. Hartt, a. C. 1980. Juvenile salmonids in the oceanic ecosystem— the critical first summer. In W. J. McNeil and D. C. Himsworth (editors), Salmonid ecosystems of the North Pacific, p. 25-27. Oreg. State Univ. Press. Hatfield, S. E. 1983. Distribution of zooplankton in association with Dunge- ness crab, Cancer magister, larvae in California. In P. W. Wild and R. N. Tksto (editors). Life history, environment, and mariculture studies of the Dungeness crab. Cancer magister, with emphasis on the central California fishery resource, p. 97-123. Calif. Dep. Fish Game, Fish Bull. 172. Heg, R., and J. Van Hyning. 1951. Food of the chinook and silver salmon taken off the Oregon coast. Oreg. Fish. Comm., Res. Briefs 3(2):32-40. HOLLING, C. S. 1959. The components of predation as revealed by a study of 690 small mammal predation of the European pine sawfly. Can. Entomol. 91:293-320. HOPLEY, B. B. 1978. Genetic effects on Ibutle-Cowlitz coho studies on the Columbia River. In B. G. Shepherd and R. M. J. Ginetz (Rap- parteurs), Proceedings of the 1977 Northeast Pacific Chinook and Coho Salmon Workshop, p. 103-121. Can. Fish. Mar. Serv. Tfech. Rep. 759. LOEFFEL, R. E., AND W. 0. FORSTER. 1970. Determination of movement and identity of stocks of coho salmon in the ocean using the radionuclide zinc-65. Oreg. Fish. Comm., Res. Rep. 2(l):15-27. Lough, R. G. 1976. Larval dynamics of the Dungeness crab, Cancer magister, off the central Oregon coast, 1970-1971. Fish. Bull, U.S. 74:353-376. McLain, D. R. 1983. Coastal ocean warming in the northeast Pacific, 1976- 1983. In W. G. Pearcy (editor). The influence of ocean con- ditions on the production of salmonids in the North Pacific - A workshop, p. 61-86. Sponsored by the Cooperative In- stitute for Marine Resources Studies, Oregon State Univer- sity Sea Grant College Program, ORESU-W-83-001, Newport, OR. McLain, D. R., and D. H. Thomas. 1983. Year-to-year fluctuations of the California Counter- current and effects on marine organisms. Calif. Coop. Oceanic Fish. Invest. Rep. 24:165-181. Methot, R. D., Jr., and L. W. Botsford. 1983. Estimated preseason abundance in the California Dungeness crab {Cancer magister) fisheries. Can. J. Fish. Aquat. Sci. 39:1077-1083. Morgan, M. D., S. T. Threekeld, and C. R. Goldman. 1978. Impact of the introduction of kokanee (Oncorhynchus nerka) and opossum shrimp (Mysis relicta) on a subalpine lake J. Fish. Res. Board Can. 35:1572-1579. Nilsson, N.-A. 1972. Effects of introductions of salmonids into barren lakes. J. Fish. Res. Board Can. 29:693-697. O'Brien, R 1973. An evaluation of the California troll silver salmon regu- lations. Calif. Dep. Fish Game, p. 1-12. (Mimeogr.) O'Brien, P., and E. W. Lesh. 1975. Preliminary evaluations of the effects of California's new silver salmon troll regulations on the 1973 and 1974 troll seasons. Appendix 2 - Special report. In Pacific Marine Fisheries Commission, 27th Annual Report for the Year 1974, p. 49-53. Oregon Department of Fish and Wildlife. 1982. Comprehensive plan for production and management of Oregon's anadromous salmon and trout. Part 2, Coho salmon plan. Oreg. Dep. Fish Wildl., var. pag. Pacific Fishery Management Council. 1979. Draft management plan for the Dungeness crab fishery of Washington, Oregon, and California. Portland, OR, 93 P- 1983. Proposed plan for managing the 1983 salmon fisheries off the coasts of California, Oregon, and Washington. An amendment and supplemental environmental impact state- ment to the "Fishery Management Plan for Commercial and Recreational Salmon Fisheries off the Coasts of Washington, Oregon, and California Commencing in 1978." Pacific Fisheries Management Council (526 S.W. Mill Street, Portland, OR), var. pag. Petrovich, A. A., Jr. 1970. Biota of nearshore waters off Humboldt Bay and Trinidad Head 1960-1964, as shown by the diet of Pacific salmon. M.S. Thesis, Humboldt State Univ, Areata, CA, 69 P- Radovich, J. 1961. Relationships of some marine organisms of the north- east Pacific to water temperature Calif. Dep. Fish Game, Fish Bull. 112:1-62. Reilly, P N. 1983a. Predation on Dungeness crabs. Cancer magister, in central California. In P. W. Wild and R. N. Tksto (editors). Life history, environment, and mariculture studies of the Dungeness crab. Cancer magister, with emphasis on the cen- tral California fishery resource, p. 155-164. Calif. Dep. Fish Game, Fish Bull. 172. 1983b. Dynamics of Dungeness crab. Cancer magister, larvae off central and northern California. In P. W Wild and R. N. Iksto (editors), Life history, environment, and mariculture studies of the Dungeness crab. Cancer magister, with em- phasis on the central California fishery resource, p. 57-84. Calif. Dep. Fish Game, Fish Bull. 172. Scarnecchia, D. L. 1981. Effects of streamflow and upwelling on yield of wild coho salmon {Oncorhynchus kisutch) in Oregon. Can. J. Fish. Aquat. Sci. 38:471-475. Scarnecchia, D. L., and H. H. Wagner. 1 980. Contribution of wdld and hatchery-reared coho salmon, Oncorhynchus kisutch, to the Oregon ocean sport fishery. Fish. Bull., U.S. 77:617-623. Sosiak, a. J., R. G. Randall, and J. A. McKenzie. 1979. Feeding by hatchery- reared and wild Atlantic salmon (Salmo salar) parr in streams. J. Fish. Res. Board Can. 36:1408-1412. Warner, R. W. 1985. Age and growth of male Dungeness crabs. Cancer mngister, in northern California. In Proceedings of the Sym- posium on Dungeness Crab Biology and Management, p. 185-187. Lowell Wakefield Fish. Symp. Ser., Univ. Alaska, Sea Grant Rep. 85-3. Wickham, D. E. 1979. Carcinonemertes errans and the fouling and mortality of eggs of the Dungeness crab, Cancer magister. J. Fish. Res. Board Can. 36:1319-1324. Wild, P W, P M. W. Lav^^, and D. R. McLain. 1983. Variations in ocean climate and the Dungeness crab fishery in California. In P. W. Wild and R. N. Tksto (editors), Life history, environment, and mariculture studies of the Dungeness crab. Cancer magister, with emphasis on the cen- tral California resource, p. 175-188. Calif. Dep. Fish Game, Fish Bull. 172. David H. ThoMAS Marine Resources Region California Department of Fish and Game If 11 Burgess Drive Menlo Park, CA H025 691 THE EFFECTS OF NET ENTANGLEMENT ON THE DRAG AND POWER OUTPUT OF A CALIFORNIA SEA LION, ZALOPHUS CALIFORNIANUS Interactions between pinnipeds and fisheries can be broadly divided into two categories: the role of pin- nipeds on the mortality of commercially important fish species and the effect of commercial fisheries on the dynamics of pinniped populations. Although the former has received considerable attention (Hirose 1977; Fiscus 1979, 1980; Matkin and Fay 1980; DeMaster et al. 1982), the importance of the latter has been addressed only recently (Shaugh- nessy 1980; Fowler 1982; Scordino and Fisher 19831). Fishery interactions may affect pinniped stocks through changes in prey abundance, incidental takes, or entanglement in discarded fishing gear and plastic packing bands. Scordino and Fisher (fn. 1) have shown that the number of entangled northern fur seals, Callorhinus ursiniLS, on the Pribilof Islands, AK, has recently increased, and now comprises 0.4% of the harvested animals. Fowler (1982^) reviewed existing data concerning the accumulation of plastic litter on beaches of several Alaskan islands. Using the number of net fragments found on shore as a rough estimate of the size distribution of material adrift at sea, he concluded that at least 60% are larger than those measured on fur seals. Because most nets found on these animals weigh <600 g, a significant mortality undoubtedly occurs at sea from entanglement in larger fragments. This paper evaluates the hydrodynamic effect of net entanglement and documents the behavior of an entangled animal. A California sea lion was trained to allow itself to become entangled in a twine trawl net fragment and the subsequent rise in drag was measured. Increased energy consumption and swim- ming power requirements associated with dragging net fragments were calculated from these measure- ments. The results provide an initial basis for assess- ^Scordino, J., and R. Fisher. 1983. Investigations on fur seal entanglement in net fragments, plastic bands and other debris in 1981 and 1982, St. Paul Island, Alaska. Background paper sub- mitted to the 26th Annual Meeting of the Standing Scientific Com- mittee, North Pacific Fur Seal Commission, 33 p. ^Fowler, C. W. 1982. Entanglement as an explanation for the decline in Northern fur seals of the Pribilof Islands. Background paper submitted to the 25th Annual Meeting of the Standing Scien- tific Committee, North Pacific Fur Seal Commission, 24 p. y C Figure 1.— Instruments and cart used in the drag experiments. The sea lion was towed passively underwater and the resultant force recorded. See text for further details. 692 FISHERY BULLETIN: VOL. 83, NO. 4, 1985. ing the possible role of net entanglement on the mor- tality of pinnipeds at sea. Materials and Methods A female California sea lion, Zalophus califor- nianus, was used in this work. The animal was kept in large seawater holding tanks at Scripps Institu- tion of Oceanography. Its weight (45 ± 0.5 kg) re- mained constant throughout the course of this study, conducted during April 1983. lb measure drag, the sea lion was trained to bite onto a neoprene mouthpiece and be towed through the water behind a moving cart (Fig. 1). The cart, powered by a variable speed electric motor, travel- led around a circular "ring" tank which had a depth of 3.5 m and inner and outer diameters of 14.5 and 21 m, respectively. A line was connected to the mouthpiece and the other end secured to a load cell (Western Scale Co.) which produced a voltage out- put proportional to the amount of tension on the lina The tow line extended down from the load cell, through a streamlined strut and around a teflon pulley attached to the end of the strut (Fig. 1). The pulley, enclosed by a streamlined fiberglass housing, was set at a depth of 1 m (>3 body diameters) to eliminate surface wave effects on drag (Hoerner 1959). Drag was measured by continuously recording the signal output from the load cell during each towing session. The signal was amplified and recorded on a Brush^ 220 strip chart recorder (Gould In- struments). At the end of each session, the load cell was calibrated using a hand-held dynamometer. A tachometer, attached to one of the outer cart wheels, was used to determine cart velocity. This was simultaneously recorded on the strip chart. The sea lion's velocity, while it was being towed down the mid- dle of the tank, was computed using the speed of the outer wheel and the tank's circumference After each experiment the data were smoothed by eye and drag and velocity determined. Only steady traces which varied less than ±3% were analyzed. Drag was then converted to newtons by multiplying the kilogram force reading of the load cell by the acceleration of gravity. Once the sea lion's drag without a net was measured, the animal was trained to place its head through an opening cut in the mesh of a 1/8-in (3.2 mm) nylon twine trawl net. The opening was near the center of the net which measured 1.4 m x 5 m, with a stretched mesh size of 19 cm. The net had a dry weight of 580 g. After several trials, the sea lion became accustomed to the procedure and would allow itself to be towed with the net trailing from its neck. The net was removed after each session. Results Drag on the sea lion, both with and without the net, increased with velocity (Fig. 2). This rise, however, was significantly greater when the animal was entangled, with the difference between the two curves increasing throughout the range of speeds. At the highest velocity of 3.5 m/s, the entangled drag was 111 N greater than that of the free animal (Tkble 1). Therefore, to maintain a cruising speed of 2.0 m/s an animal of this size, entangled in a net with similar hydrodynamic characteristics, would experience the equivalent drag of a free animal swimming at speeds above 4 m/s. Power that the sea lion must expend for swimming can also be calculated from these measurements. Since drag is a force, power output (in watts) is a product of drag times swimming velocity (Webb 1975): Pq = drag x velocity. Tkble 1 shows the results of such calculations and the effect of the net on the sea lion's required output. Power output is a measure of the mean rate of energy expended by the swimming muscles at a given velocity (Webb 1975). It does not, however, reveal the total energetic requirements of the sea 150t CO §100 LU o 50 < tr o "t WITH NET / . _--v ^^-^» WITHOUT ,y-rfr' NET ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. 0 12 3 4 VELOCITY (M/S) Figure 2.— Drag of a 45 kg sea lion with and withoug a net trail- ing from its neck. In both cases drag increased geometrically with speed. The regression equation with the net was 17.19 vel.'^^, SEE (standard error of estimate) = 0.052. The equation for drag without the net was 2.93 vel.^•"^ SEE = 0.118. 693 Table 1.— The increase in drag and power output, and the estimated power input of the entangled sea lion. Drag with and without the net was calculated from the best fit regressions determined by the experiments. Power output was calculated by multiplying drag and the appropriate velocity. Weight specific power input is based on an efficiency (power out/power in) of 10%. Velocity (m/s) 1.0 1.5 2.0 2.5 3.0 .3.5 Drag (N) Without net 2.93 6.75 12.21 19.33 28.13 38.64 With net 17.19 34.66 57.01 83.87 114.97 150.11 Power output (W) Without net 2.93 10.13 24.42 38.65 84.40 135.24 With net 17.19 51.99 114.02 167.74 344.91 525.39 Power input (W/kg) Without net 0.7 2.3 5.4 8.6 18.8 30.0 With net 3.8 11.6 25.3 37.3 76.6 116.8 lion. Animals are not 100% efficient in converting metabolic energy to mechanical power needed for locomotion (Hicker 1975). Studies of penguins and fish which swim with their pectoral fins show that efficiency (power output/power input) varies between 5 and 15% (Webb 1973; Hui 1983). Female north- ern fur seals consume about 8 W/kg while at sea (Costa and Gentry in press). Using this number, the power output values estimated for the sea lion, and an assumed cruising speed of 2.5 m/s, an efficiency for fur seals of roughly 10% is obtained. Tkble 1 shows estimated energy requirements based on this efficiency for the sea lion with and without the net. Again it can be seen that to maintain a swimming speed of 2.5 m/s the sea lion would need to increase its metabolic expenditure by 50-fold, an impossibly high figure (Bartholomew 1977). Discussion There is little doubt that for an animal of this size, entanglement in a 600 g net will substantially in- crease its chance of mortality at sea. Drag, and hence the power required for swimming, is increased by the presence of a net. As a result, these animals will swim slower, at a greater energetic cost than free animals. Drag of the net, which rises geometrically with velocity, will prevent activities requiring high speeds as would be the case if such animals engaged in the pursuit of rapidly moving or evasive prey species. Additionally, migration or travel to and from the rookery will be energetically more costly. It is likely, then, that once an animal becomes entangled in net fragments of this size or larger, it enters a state of negative energy balance The animal's size as well as the size of the net plays an important role in the amount of drag experienced. A larger net will present more surface area to the water. Since drag is dependent on surface area (Vogel 1981), larger net fragments will result in greater drag. Similarly, if two different-sized animals are entangled in nets of the same dimensions, the smaller will experience a larger relative increase in its drag and power requirements. If animals of dif- ferent age classes encounter net fragments with equal probability, it is expected that the younger age classes will suffer a proportionally higher mortality. Although starvation is undoubtedly the long-term result of net entanglement, other factors may have a more immediate effect. This was particularly evi- dent during an observation of an actual entangle- ment. In the initial training phase of the sea lion, a net with a larger mesh size was used. At one point, while the net was floating in the water, the sea lion swam up from below and inserted its head through one of the mesh openings. Upon sensing the net around its neck it gave a strong backwards stroke, trying to retract its head. The backward movement brought some of the trailing net in front of it and when the animal then swam forward and dived underwater, another strand slipped onto its neck. This caused a violent reaction with the sea lion twisting and thrashing wildly. The twisting further entangled the animal and tightened the net. Within IV2 to 2 min the animal was completely entangled with three or four loops of mesh tight around its neck. The net was so tight that an observer on the side of the tank was unable to pull it from the sea lion's head, and it was necessary to drain the holding tank. During this time, the animal swam around the tank barking and often thrashing about while trying to bite the net. This appeared to further tighten the net which, when finally removed, was so tight that a finger could not be slipped between the net and the animal's neck. 694 If the behavior of otariids at sea is similar to that exhibited by the entangled sea lion, then drowning may be another more immediate cause of mortality. Twisting and rolling could foul the foreflippers and prevent the animal from swimming. This seems par- ticularly likely if it became caught in a larger net. Additionally, because the net was so tightly wrap- ped around the sea lion's neck, necrosis of the skin tissue and an open wound may have occurred within a matter of hours to a few days. Constant swimming could continue to tighten the net. Although several authors (Scordino and Fisher fn. 1) have speculated that neck wounds indicate a period of entanglement longer than 4 mo, these observations suggest that beached animals with open wounds may have become tangled only a few days prior to sighting. Acknowledgments The sea lion was made available for this research by L. H. Cornell, Sea World of San Diego. The assistance provided by Phil Thorson and Tferrie Williams in feeding and training the sea lion is grate- fully acknowledged and appreciated. C. Fowler, J. Graham, G. Kooyman, and T Williams provided useful comments on the manuscript. This study was supported by NOAA grant 82ABC-02743 from the National Marine Mammal Laboratory to G. L. Kooyman. Literature Cited Bartholomew, G. A. 1977. Energy metabolism. In M. S. Gordon (editor), Animal physiology: Principles and adaptations, p. 57-110. Macmillan Publ. Co., N.Y. Costa, D. P., and R. L. Gentry. In press. Free ranging energetics of northern fur seals. In R. L. Gentry and G. L. Kooyman (editors), Fur seals: Mater- nal strategies on land and at sea. Princeton University Press. DeMaster, D. P., D. J. Miller, D. Goodman, R. L. DeLong, and B. S. Stewart. 1982. Assessment of California sea lion fishery interactions. Trans. N. Am. Wildl. Nat. Resour. Conf. 47:253-264. Fiscus, C. H. 1979. Interactions of marine mammals and Pacific haka Mar. Fish. Rev. 41(10):l-9. 1980. Marine mammal-salmonid interactions: a review. In W. J. McNeil and D. C. Himsworth (editors), Salmonid eco- systems of the North Pacific Ocean, p. 121-131. Oregon State Univ. Press, Corvallis. Fowler, C. W. 1982. Interactions of northern fur seals and commercial fisheries. Trans. N. Am. Wildl. Nat. Resour. Conf. 47:278- 292. Hirose, p. 1977. Incidence of seal-damaged salmonids sampled from the lower Columbia river gillnet fishery, 1972-1976. Oreg. Dep. Fish Wildl, Inf. Rep. 77-4, 6 p. Hoerner, S. F. 1959. Fluid-dynamic drag. Practical information on aero- dynamic drag and hydrodynamic resistance [Published by the author.] Midland Park, N.J. 416 p. Hui, C. A. 1983. Swimming in penguins. Ph.D. Thesis, Univ. California, Los Angeles, 185 p. Matkin, C. 0., and F H. Fay. 1980. Marine mammal-fishery interactions on the Copper river and in Prince William Sound, Alaska, 1978. Final Report to U.S. Marine Mammal Commission for contract MM8AC-013, 71 p. Publ. No. PB80-159536, Natl. Tfech. Inf. Serv., Springfield, VA. Shaughnessy, p. D. 1980. Entanglement of Cape fur seals with man-made objects. Mar. Pollut. Bull. 11:332-336. Hjcker, V. A. 1975. The energetic cost of moving about. Am. Sci. 63:413- 419. VOGEL, S. 1981. Live in moving fluids: The physical biology of flow. Willard Grant Press, Boston, MA, 352 p. Webb, R W. 1973. Efficiency of pectoral-fin propulsion of Cymatogaster aggregata. In T. Y. -T. Wu, C. J. Brokaw, and C. Brennen (editors). Swimming and flying in nature. Vol. 2, p. 423-1005. Plenum Press, N.Y. 1975. Hydrodynamics and energetics of fish propulsion. Fish. Res. Board Can. Bull. 190, 158 p. Steven D. Feldkamp Scripps Institution of Oceanography Physiological Research Laboratory A-004 La Jolla, CA 92093 Present address: Long Marine Laboratory University of California at Santa Cruz Santa Cruz, CA 9506Jt NOTES ON THE LIFE HISTORY OF THE CATSHARK, SCYLIORHINUS MEADI The catshark, Scyliorhinus meadi (family Scylio- rhinidae) is a rare, poorly known species, easily iden- tified by the eight dark saddle-like blotches along the dorsal surface Springer (1966) first described S. meadi and Springer and Sadowski (1970) assigned it to subspecies status of S. retifer. In Springer's (1979) revision of the family, it was again given species status. At present only 10 immature specimens of 5. meadi have been collected, seven males (180-490 mm in length), two females (235 and 385 mm in length), and one 190 mm specimen of unknown sex. This paper reports on the collection of an additional specimen of S. meadi and provides valuable life history information. During a cruise aboard the RV Delaware II on 5 FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 695 May 1984, a specimen of S. meadi was collected at a depth of 412 m using a 17 m otter trawl at lat. 28°59.85'N, long. 79°55.98'W. The shark measured 430 mm in length and weighed 0.4 kg. This shark, together with its stomach contents, is deposited at the Ichthyological Museum of the Florida Depart- ment of Natural Resources, St. Petersburg (FSBC 16208). Examination of the reproductive system revealed a developing right ovary measuring 4.4 cm long and 0.6 cm wida The left ovary was atrophied. Follicles teased from the right ovary measured 0.75 mm in diameter. The nidamental glands were 3.0 mm wide and 8.0 mm long. The oviduct, from nidamen- tal gland to vagina, was 82 mm long and 1.0 mm wide Gut content examination revealed several cephalopod beaks and tentacles, shrimp remains, and the articulated caudal skeleton from a relatively large, unidentified teleost. This specimen of S. meadi is the third and largest female collected to date. Burgess et al. (1979) reported on the collection of a 385 mm total length female in which the right ovary measured 33.2 mm and the left 8.9 mm in length. This represents about 8.6% of the total length of the shark. The right ovary of the specimen reported in this paper represents about 10.2% of the shark's total length. This specimen is immature, but the allometric increase in ovary length, and the fact that the left ovary was completely atrophied suggests that the specimen might be considered a subadult (maturation begun). The small size of the Florida Bahama specimens compared with the North Carolina examples led Burgess et al. (1979) to suggest geographical size segregation. This large specimen from off the cen- tral Florida east coast does not support this segrega- tion. Discovery, in gut content examination, of the large, teleost caudal skeleton suggests an unexpected ability to prey on relatively large fishes. Acknowledgments I would like to thank David Camp, Mark Leiby, Bill Lyons, Mike Murphy Jim Quinn, Steve Ealsh, and the captain and crew of the RV Delaware II. Literature Cited Burgess, G. H., G. W. Link, Jr., and S. W. Ross. 1979. Additional marine fishes new or rare to Carolina waters. Northeast Gulf Sci. 3(2):74-87. Springer, S. 1966. A review of western Atlantic cat sharks, Scyliorhinidae, with descriptions of a new genus and five new species. U.S. Fish. Wildl. Serv., Fish. Bull. 65:581-624. 1979. A revision of the catsharks, Family Scyliorhinidae U.S. Dep. Commer., NOAA Ttech. Rep. NMFS Cir. 422, 152 p. Springer, S., and V. Sadowsky. 1970. Subspecies of the western Atlantic cat shark, Scyliorhinus retifer. Proa Biol. Soc Wash. 83:83-98. Glenn R. Parsons Department of Marine Sciences University of South Florida HO Seventh Avenue South St. Petersburg, FL 33701 A COMPARISON OF SCALE AND OTOLITH AGING METHODS FOR THE ALEWIFE, ALOSA PSEUDOHARENGUS ' Beginning in 1971, the Maine Department of Marine Resources monitored the harvests of anadromous alewives ascending the Damariscotta River (Libby 1982). Part of this monitoring assessed changes in age composition within and between years. Aging was done by interpreting the number of scale annuli in terms of fish age as has been done in earlier in- vestigations (Havey 1961; Rothschild 1963; Marcy 1969). However, scale annuli were sometimes difficult to interpret, so in 1979, methods for removing and reading alewife otoliths were studied. A relatively fast and efficient method was developed for remov- ing otoliths. The ease with which the otoliths were processed to age fish prompted an analysis of which method (scales or otoliths) was best for determin- ing the age of an alewife This paper compares the precision of reproducibility and accuracy between the scale and otolith methods. Materials and Methods Alewives, Alosa pseudoharengus, were taken daily from the commercial harvest throughout the fishing period for their otoliths and scales. The fish were taken to the laboratory sexed, and measured for length and weight. About 10 scales were removed from the left side above the lateral line just posterior to the dorsal fin. The scales were cleaned and put into envelopes labeled with the length and sex of the alewife In 1963, Rothschild described the alewife scale and characteristics of the annuli. Otoliths were collected and stored as follows: A "This study was conducted in cooperation with the U.S. Depart- ment of Commerce, National Marine Fisheries Service, under Public Law 89-304, as amended, Commercial Fisheries Research and Development Act, Project AFC-21-1. 696 FISHERY BULLETIN; VOL. 83, NO. 4, 1985. transverse cut was made at the point of attachment of the operculum, which severed the head leaving some attachment of skin to the body. The head was then pulled away from the body removing the gills, leaving the skull clear of the gills and viseral blood. When time did not permit for further processing, all the heads from a day's sample were placed onto a tray and frozen for later analysis. For otolith removal, the head was held ventral side up and a transverse cut was made into the skull at the point of dorsal musculature attachment (Fig. la). A cor- rect cut was sliced through the ends of the semicir- cular canals containing the sagitta otolith (Fig. lb). Kornegay, in 1978, described the sagitta otolith that is used for age determination of the alewifa Each otolith was extracted with microforceps, placed on absorbent paper, and rubbed lightly to remove any adhering tissue. After drying, otoliths were placed in depressions in black Plexiglas^ trays and covered with Permount (see Libby 1982). Williams and Bedford (1974) described the growth and collection of otoliths and interpretation of otolith annuli in general. The Atlantic herring, Clupea harengus, otolith which is similar to the alewife otolith, was described by Watson (1964). He reported on the high validity of its use for aging and the high reproducibility of readings between readers. Scales and otoliths were collected from 536 fish. Thirty-one fish were discarded because the otoliths or scales were unreadable or the shape of the otolith revealed that the fish was a blueback, A. aestivalis. Price (1978) explained the difference in otolith mor- phology between the two species. A final count of ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service^ NOAA. Figure 1.— a) Ventral side of head of Alosa pseudoharengtis. Dark line shows muscula- ture attachment to the skull where transverse cut is made b) Section of skull cut away showing posterior ends of the semicircular canals containing the sagitta otoliths. 697 505 alewives was used for the analysis. Otoliths and scales were read using a binocular dissecting scope at 30x-60x. Each otolith and scale for an age shown in Figure 2 was collected from the same fish. The otoliths were taken from the left side of the head and each scale was chosen for the best annuli appearance of all scales from that fish. The scale annuli shown are at the anterior portion of the scale that is normally covered by surrounding scales. The otoliths were photographed with a 35 mm camera mounted on a microscope at 11 x. Aging was done independently by two people (readers; subsequently referred to as R^ and Rg). Ages derived from scales and otoliths by each reader were referred to as an age set. Scales and otoliths were read without knowledge of fish length or sex. A true age was established for each fish by reexamin- ing scales and otoliths together with length and sex. The five age sets were compared to show 1) the preci- sion of aging reproducibility (measurement of how close the ages are for two or more readings); and 2) the aging accuracy (age determinations compared with the true age). Three analyses were used to evaluate precision and accuracy. Percent agreement (PA) compared two to three age sets to reveal what portion of the fish were aged alike between age sets. An index of average per- cent error (APE) developed by Beamish and Four- nier (1981) was used to show the degree of varia- tion between age sets. APE is expressed as 1 N I 1 R I l^y X, where N is the number of fish aged; R is the number of ways each is aged; X^ is the ith determination of 3yr. 259mm (f 4yr. 2 8 I mm (f 5yr. 293mm 5°C) 100 80 . (1). 'Vi - (14^ V, (1) (18) • (11L_ ,^* '(lO) '•(5) 60 40 20 . .•(54) -'•(36) c cr (3)'^5E^ y,(22) >^46) ,(17) 100 80 60 <24)^ . X^) • 9 (19)(29) (13) (23) 9 (5) (30) (14) ''^<'"""23) ^l"'* y^*(36) 40 - / 20 ^^/,(36) 1 1 1 t . . 1 (15)''?i-»^.(27) (5)1 .--• (38) • 1 2 3 4 5 6 7 8 0 ^^»?3t?»'^«^..3U'?D'5>-j Age Fork Length (Midpoint o( 50mm increments) Figure 3.— Percent mature walleye by age and length and by sex for specimens collected in the John Day pool of the Columbia River, April- September 1980-81. Curves were drawn by eye (Sample size in parentheses.) 800 700 -g 600 E £ 500 c « -J 400 a t- 300 200 100 - 0,(2730) . o ^ ^S/-" o » O o (1178)7 o -o — o ^°^ ^(3900) - oX .y /°^ ..o-' O o -O ft^ A-^-— ^"^ "^ o-;:::^(i9oi) o - \/ D*^ ^8^ O // - ::o^°'^ o — o o 0-John Day Pool o A-Norrls Reservlor n-Lake Gogebic o-Lac la Ronge . ° o O-Hlghest Values Reported o OLowest Values Reported 6 7 8 Age (Years) 10 11 12 13 Figure 4.— Comparison of length-at-age for walleye from the John Day pool, Columbia River; Norris Reservoir, Tfennessee (Stroud 1949); Lake Gogebic, Michigan (Eschmeyer 1950); Lac la Ronge, Saskatchewan (Rawson 1957) and the composite high and low values reported by Colby et al. (1979). Numbers in parentheses are the mean growing degree-days above 5°C, John Day value is from Anonymous (1969), all others are from Colby and Nepszy (1981). 704 (Colby and Nepszy 1981) for each area are included in Figure 4 as a measure of solar energy input to the system. Colby and Nepszy (1981) found that walleye growth was directly correlated to ODD >5°C and that the optimum range was from 2,500 to 4,000 GDD >5°C. While the GDD >5°C for the John Day pool is within this range, the walleye growth reported here is greater than would be predicted using this variable Water temperature may be the most important factor governing the growth of fishes (Brett 1979). Kitchell et al. (1977b) presented a bioenergetics model for walleye growth and indicated that ther- mal optima and maxima for weight specific con- sumption are 22°C and 27°C, respectively and 27°C and 32 °C, respectively, for weight specific respira- tion. Water temperatures in the John Day pool during the growing season remain at or near the thermal optimum for consumption and, perhaps more importantly, do not approach the thermal max- ima for consumption or respiration (Tkble 1). Many northern lakes may not reach the thermal optima (Rawson 1957; Swenson 1977) and the southern lakes or lakes which stratify in the summer may exceed the thermal maxima (MacLean and Magnuson 1977) not only reducing consumption but increasing respiration. Dendy (1948) reported that in June 1944 the surface temperature of Norris Reservoir was about 30 °C and that walleye appeared to prefer water temperature of about 24 °C, even though these areas had oxygen concentrations <3.0 mg/L. Con- versely, water temperature of Lac la Ronge did not exceed 20°C (Rawson 1957), well below the thermal optima. Exceptions to the north-south trend in high wall- eye growth occur in systems of high exploitation (Forney 1965) and/or where there have been decreases in interspecific competition (Wolfert 1969; Forney 1977) which results in density dependent in- creases in growth rates. The quantity and quality of food are important factors in walleye growth (Kelso 1972; Kerr and Ryder 1977; Kitchell et al. 1977b) and fecundity (Colby and Nepszy 1981). Schupp (1978) looked at the growth of walleye from several areas within Leech Lake, MN, and found food of walleye from areas of highest average growth was almost totally young-of-the-year yellow perch, whereas small walleyes from slow growth areas had eaten mostly invertebrates and small minnows. We have found (Maule and Horton 1984) that about 99% by volume of Columbia River walleye stomach con- tents were fish (ag., sculpins, suckers, cyprinids) and that 61% of walleye sampled contained food. Eschmeyer (1950) reported that 89% of the volume of stomach contents from Lake Gogebic walleye was fish, but he did not report percent empty stomachs. Dendy (1946) reported that Norris Reservoir wall- eye stomachs contained 99% fish by volume, but only 45% of the walleye examined contained food. Rawson (1957) studied Lac la Ronge walleye and reported that fish comprised 97% of the volume of stomach contents and that 39% of the walleye stomachs con- tained food. Colby and Nepszy (1981) stated that age to matu- rity is indirectly correlated to growth, but that fecun- dity is probably a function of population density and food availability. They further suggested that the wide variability in walleye fecundities is a mechanism by which walleye can adjust production in response to environmental conditions. Ikble 3 includes fecun- dity data from Norris Reservoir (Smith 1941), Lake Gogebic (Eschmeyer 1950), and western Lake Erie (Wolfert 1969). Based on a comparison of growth, stomach content analysis, and fecundity the mid-Columbia River walleye have a more favor- able food supply than the other areas considered hera Hackney and Holbrook (1978) suggested that there is a southern race of walleye that is characterized by rapid, large growth and short life span, and a northern race characterized by slow growth and long life span. They suggested that the pattern of rapid walleye growth seen after the impoundment of southern waters, followed by decreased growth rates some years later is due to a shift from the southern race to the northern race as the result of walleye stocking programs. The movements of young-of-the- year walleye downstream past Columbia River dams has been documented (Brege 1981). Assuming that this is a means by which walleye have colonized the Columbia River, it is biologically similar to impound- ing waters already containing walleye populations, in that new habitat is available for population growth. Although we cannot discount the possibility that the extreme life history characteristics reported here are the result of genetic stock differences, we suggest that they can more reasonably be explained by a favorable temperature regimen and an abundant, high quality food supply. Acknowledgments We thank Hiram Li and Carl Bond for their reviews of the manuscript. Funding was provided by the U.S. Army Corps of Engineers contract DACW 57-79-C-0067, the Oregon Agricultural Experiment Station, and the Milne Computer Center, Oregon State University, Corvallis, OR. 705 Literature Cited Anonymous. 1969. Climatological handbook. Columbia Basin states. Ttem- perature Volume 1, part A. Meteorology Committee, Pacific Northwest River Basins Commission, Vancouver, WA, 268 p. Brege, D. A. 1981. Growth characteristics of young-of-the-year walleye, Stizostedion vitreum vitreum, in John Day Reservoir on the Columbia River, 1979. Fish. Bull., U.S. 79:567-569. Brett, J. R. 1979. Environmental factors and growth. In W. G. Hoar, D. J. Randall, and J. R. Brett (editors), Fish physiology. Volume VIII. Bioenergetics and growth, p. 599-676. Acad. Press, N.Y. Campbell, J. S., and J. A. Babaluk. 1979. Age determination of walleye, Stizostedion vitreum vitreum, (Mitchill), based on the examination of eight dif- ferent structures. Can. Fish. Mar Serv., Ttech. Rep. 849, 23 p. Carlander, K. D. 1982. Standard intercepts for calculating lengths from scale measurements for some centrarchid and percid fishes. Trans. Am. Fish. Soa 111:332-336. Carlander, K. D., and R. R. Whitney. 1961. Age and growth of walleyes in Clear Lake, Iowa, 1935- 1957. Trans. Am. Fish. Soa 90:130-138. Christensen, J. M. 1964. Burning of otoliths, a technique for age determination of soles and other fish. J. Int. Counc. Explor. Sea 29:73-81. Colby, P. J., R. E. McNicol, and R. A. Ryder. 1979. Synopsis of biological data on the walleye, Stizostedion vitreum vitreum (Mitchill 1818). FAO Fish. Synop. 119, 139 P- Colby, P. J., and S. J. Nepszy. 1981. Variation among stocks of walleye (Stizostedion vitreum vitreum): management implications. Can. J. Fish. Aquatic Sci. 38:1814-1831. Dendy, J. S. 1946. Food of several species of fish, Norris Reservoir, Tfen- nessee J. Ifenn. Acad. Sci. 21(1):105-127. 1948. Predicting depth distribution of fish in three TVA storage-type reservoirs. Trans. Am. Fish. Soc 75:65-71. ESCHMEYER, P. H. 1950. The life history of the walleye, (Stizostedion vitreum^ vitreum (Mitchill)), in Michigan. Mich. Dep. Conserv., Bull. Inst. Fish. Sci. 3. Forney, J. L. 1965. Factors affecting growth and maturity in a walleye population. N.Y. Fish Game J. 12:217-232. 1977. Evidence of inter- and intraspecific competition as fac- tors regulating walleye (Stizostedion vitreum vitreum) biomass in Oneida Lake^ New York. J. Fish. Res. Board Can. 34:1812-1820. Hackney, P. A., and J. A. Holbrook II. 1978. Sauger, walleye, and yellow perch in the southwestern United States. In R. L. Kendall (editor). Selected coolwater fishes of North America, p. 74-81. Am. Fish. Soc Spec Publ. 11. Hjort, R. C, B. C. Mundy, and P. L. Hulett. 1981. Habitat requirements for resident fishes in the reser- voirs of the lower Columbia River Final report. U.S. Army Corps Eng. Contract No. DACW57-79-C-0067. Portland, OR. Kelso, J. R. M. 1972. Conversion, maintenance and assimilation for walleye, Stizostedion vitreum vitreum, as affected by size, diet and temperature J. Fish Res. Board Can. 29:1181-1192. Kerr, S. R., and R. A. Ryder. 1977. Niche theory and percid community structure J. Fish. Res. Board Can. 34:1952-1958. KiTCHELL, J. F, M. G. Johnson, C. K. Minns, K. H. Loftus, L. G. Greig, and C. H. Olver. 1977a. Percid habitat: The river analogy. J. Fish. Res. Board Can. 34:1936-1940. KiTCHELL, J. F., D. J. Stewart, and D. Weininger. 1977b. Applications of bioenergetics model to yellow perch (Percajlavescens) and walleye (Stizostedion vitreum vitreum). J. Fish. Res. Board Can. 34:1922-1935. MacLean, J., and J. J. Magnuson. 1977. Species interactions in percid communitltes. J. Fish. Res. Board Can. 34:1941-1951. Maule, a. G., and H. F. Horton. 1984. Feeding ecology of walleye, (Stizostedion vitreum vitreum) in the mid-Columbia River, with emphasis on the interactions between walleye and juvenile anadromous fishes. Fish. Bull., U.S. 82:411-418. Rawson, D. S. 1957. The life history and feeding ecology of the yellow wall- eye, Stizostedion vitreum, in Lac la Ronge; Saskatchewan. Trans. Am. Fish. Soc 86:15-37. ScHUPP, D. H. 1978. Walleye abundance, growth, movement, and yield in disparate environments within a Minnesota Lake In R. L. Kendall (editor), Selected coolwater fishes of North America, p. 58-65. Am. Fish. Soc Spec Publ. 11. Smith, C. G. 1941. Egg production of walleyed pike and sauger Norris Reservoir fish differ from same species in other localities. Prog. Fish-Cult. 54:32-34. Stroud, R. H. 1949. Growth of Norris Reservoir walleye during the first twelve years of impoundment. J. Wildl. Manage 13:157-177. Swenson, W. a. 1977. Food consumption of walleye (Stizostedion vitreum vitreum) and sauger (S. canadense) in relation to food availability and physical conditions in Lake of the Woods, Minnesota, Shagawa Lake, and western Lake Superior J. Fish. Res. Board Can. 34:1643-1654. WOLFERT, D. R. 1969. Maturity and fecundity of walleyes from the eastern and western basins of Lake Erie J. Fish. Res. Board Can. 26: 1877-1888. Alec G. Maule Oregon Cooperative Fisheries Research Unit Department of Fisheries and Wildlife Oregon State University Corvallis, OR 97331 Howard F. Horton Department of Fisheries and Wildlife Oregon State University Corvallis, OR 97331 706 BIOLOGICAL ASPECTS OF THE SPRING BREEDING MIGRATION OF SNOW CRABS, CHIONOECETES OPILIO, IN BONNE BAY, NEWFOUNDLAND (CANADA) The occurrence of an annual (April-May) deep- to shallow-water breeding migration of snow crabs, Chionoecetes opilio, in Bonne Bay, on the west coast of Newfoundland, has been documented by Hooper (in press). In addition to being the first record of this phenomenon in this species, his observations con- tradict some generally accepted conclusions regard- ing the species' reproductive biology. The most significant of these are that females undergo a ter- minal molt to maturity and do not mate in the hard shell condition (Ito 1967; Watson 1972; Tkkeshita and Matsuura 1980). Little morphometric sampling data are included in Hooper's general description of the breeding migration. The purpose of this paper is to provide a more detailed description of various biological aspects of the phenomenon, such as size difference between paired males and females, and condition of the external egg masses, ovaries, and spermathecae during the breeding period. Materials and Methods Three hundred and three sexually paired snow crabs were collected during three field trips to Bonne Bay from 24 April to 29 May 1984 by scuba diving (10-30 m depth). Each pair was kept in a separate mesh bag. At the surface, each crab was measured to the nearest millimeter (maximum carapace width (CW)) and its shell condition (soft, new/hard, or old/hard) determined. The eggs of females were ex- amined to determine their stage of development. Following this, males were tagged with Floy vinyl "T-bar" tags (Tkylor 1982) and released, and females were either tagged and released, or retained for later examination of their ovaries and spermathecae in the laboratory. Results Size Distribution Size distributions were unimodal for each sex but with no overlap in their carapace widths (Fig. 1). Males ranged from 89 to 140 mm (x = 116.4 mm) CW and females from 55 to 86 mm (x = 67.8 mm). Other than the fact that males were invariably larger than females, there was no discernible relationship between size of the male and size of the female with which it was paired (Fig. 2). Mean sizes of females paired with small, medium, and large size males were the same (P < 0.005, Bartlett's test of homogeneity of variance). Male CW (mm) Female CW (mm) Range Range Mean N 89-109 (small) 55-86 69.2 59 110-120 (medium) 59-86 69.6 136 121-120 (large) 59-84 70.9 108 Ibtal 303 The mean difference in carapace width between paired males and females increased from 21 mm at 89 mm male CW to 70 mm at 140 mm. Only 3 males in 303 pairs were smaller than 95 mm, the legal size limit. Female Reproductive Condition During the 24-27 April sampling period, 92% of the females carried full clutches of eyed eggs and the remainder had liberated all or most of the lar- vae (Table 1). By 7-11 May, 59% had empty brood pouches indicating that hatching was well advanced. However, during 22-25 May, 53% of the females were carrying full clutches of eyed eggs and only 39% had empty brood pouches. This increase in relative abundance of females with eyed eggs could have resulted from a return to deeper water of females that had liberated larvae or an influx of new animals from deeper water. Dead eggs were carried by 1.4% of females examined. All females dissected (77) had ripe (extrusion imminent) ovaries (Table 2); however, only two with partially extruded clutches Table 1. — Summary of observations on external egg masses of female Chionoecetes opilio col- lected in Bonne Bay, Newfoundland, April-May 1984. Sampling period Larvae Larvae Eyed liberating liberated (%) (%) (%) N 24-27 April 7-11 May 22-25 May 92 0 8 128 9 32 59 81 53 8 39 87 Table 2.— Summary cefes opilio collected i of internal observations on female Chionoe- n Bonne Bay, Newfoundland, April-May 1984. Ripe Sampling ovaries period (%) Spermatophores (%) Old only Old and new New only N 24-27 April 100 22-25 May 100 45.7 48.6 5.7 35 7.1 92.9 0 42 FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 707 55-59 65-69 75-79 85-89 95-99 105-109 115-119 125-129 135-139 60-64 70-74 80-84 90-94 100-104 110-114 120-124 130-134 140-144 CARAPACE WIDTH (5 mm GROUPS) Figure 1.— Size-frequency distributions of male and female Chionoecetes opilio collected as pairs in Bonne Bay, Newfoundland, during April-May 1984. of new eggs were found over the entire sampling period. Females were observed with spermathecae con- taining both old and new spermatophores. In these, spermathecae were engorged with a very white glutinous material containing new spermatophores for three-fourths of their length, while the remain- ing one-fourth at the dorsal end of the organ was shrunken and contained a yellowish brown substance of a "waxy" consistency. Females which did not have new spermatophores had very small spermathecae which were entirely yellowish brown in color. This is very similar to that described for Chionoecetes bairdi by Paul (1982). While 97% of all females examined contained old spermatophores, those containing new spermatophores as well in- creased to 92.9% from 48.6% between 24 April and 25 May (Table 2). Two specimens contained new spermatophores only and all had old epizooite- encrusted shells. Thirty-six percent of the females with new spermatophores carried full clutches of eyed eggs. Diving during 28-31 May revealed that all crabs had left the sampling area. Discussion Small numbers of grasping, male/female pairs of C. opilio and C. bairdi have been observed in shallow water elsewhere. Ennis (Unpubl. data) found five pairs and Hooper (Unpubl. data) found three pairs of C. opilio in Bonavista Bay and Placentia Bay, Newfoundland, respectively. Donaldson (1975) reported two pairs of C. bairdi in Alaska. However, nothing comparable with the magnitude of the breeding migration of C. opilio, observed in Bonne Bay, Newfoundland, has been reported for other areas. There is considerable scope for speculation on the ecological significance of this migration. Although about half the females examined just prior 708 to their departure from the shallow (<35 m) sam- pling depths in 1984 still had full clutches of eyed eggs, liberation of a large proportion of larvae in shallow water may enhance chances for larval sur- vival overall. At the time of the migration, bottom temperatures in Bonne Bay at depths beyond 35 m are probably 0°C or lower [deep water temperatures are not available for Bonne Bay but Squires et al. (1971) reported temperatures <0°C at depths beyond 30 m in early June in North Arm, Bay of Islands, about 40 km to the south]. Release of lar- vae in shallow, warmer water (temperature was 3°C at 30 m during 7-11 May) would considerably reduce the degree of thermal shock associated with larvae swimming to the surface. The rate of embryonic development would likely be increased also, result- ing in earlier larval release. In the development of a management strategy for C. opilio stocks on the Atlantic coast of Canada, a key assumption has been that, despite high levels of exploitation, reproductive potential in a stock re- mains at prefishery levels. The basis for the assump- tion is that females are protected from exploitation by the 95 mm CW minimum legal size because they do not grow to that size and also that males mature at sizes much smaller than 95 mm CW. In a recent review, following more than 15 yr of heavy fishing in some areas, there was no evidence to indicate that the assumption was invalid (Elner and Robichaud 1983). However, the observations presented here suggest that a large size differential between the male and female of a pair is an important element of behavioral interactions during breeding activity. It is possible that males smaller than 89 mm CW (the smallest male observed paired with a female), even though physiologically mature, may be less like- ly to mate successfully in competition with large males. Males and females appear to be segregated over most of the year (Hooper in press). Observations on the east coast of Newfoundland indicate that large males occur mainly on muddy bottom in deep water whereas females and small males occur on sand- gravel or rocky bottom somewhat shallower (Miller and O'Keefe 1981). In the breeding migration which occurs in Bonne Bay, Hooper (in press) suggested that males leave the deeper water area after select- ing a mature female which is carried to the shallow water breeding area. Males retain possession of in- dividual females for extended periods (Hooper in 86 84 82 80 1 78 =: 76 t— o » 74 2 72 4 CC S 70 UJ ^ 68 UJ u. 66 64 62 60 58 _l ' 1 ' 1 ' 1 . 1 1 1 1 1 ' 1 ' 1 • ■Ml ' 1 ' 1 ' 1 ' • I ' 1 ' 1 ' • 1 ' 1 ' 1 ' 1 < 1 MM ' l_ '_ Y = 00919 X +59-3021 ^ J - R^= 0018? - N ■- 303 • • • • • • • • • • • • • • • • • - - .2 • . • • - — • • • — - • • .2 . • • - — • • • .2. - - • • .2 . . - — • • • . . . . - - • • • • .2. • • .2 • • • • • • .3 Ji. • • . .2. .2.3 • • .2. ... .2 ... .2 j2j_j2_j3_j__^ • . . .4 . . . .2 .2 . . .2 • • ■ •— - • • • - • • • • • • • •2 • .2. —72 T ^^. . .2 . . .2. .2 .2 • • • • • • • • .2 .2. • • .2 • .3 .2 .2 • •3 •& • • .3. .3. . .2 • • • • • • - • • • .2 . • .2 .2 • • • .3 .2 .2 .2 — — • • • • • • — - • • • • ... - " 1 , 1 , 1 , 1 1 1 1 1 1 1 I 1 , 1 • ill! • • 1 1 • 1 1 1 1 , • .2 1 1 1 1 1 1 1 1 1 1 1 . 1 . 1 1 1 . , 1 ' 92 94 96 98 100 102 104 106 108 110 112 114 116 118 120 122 124 126 128 130 132 134 136 138 140 MALE CARAPACE WIDTH (mm) Figure 2— Recession of female carapace width on male carapace width for pairs of Chionoecetes opilio collected in Bonne Bay, New- foundland, during April-May 1984. Numbers adjacent to points indicate more than one observation. Slope of the regression is not signifi- cant (P = 0.017). 709 press) (possibly up to 2 mo) during which time the female is held by and dependent on the male. In laboratory studies on C. bairdi, Paul and Adams (1984) demonstrated that muciparous females are receptive to mating for periods ranging from <1 to 7 d after all their eggs have hatched. In fact, they reported that only one ovigerous female mated suc- cessfully during their study. In the Gulf of St. Lawrence, male snow crabs mature within the 50-65 mm CW size range (Powles 1968; Watson 1970); however, in the sampling reported here, only 3 males from the 303 pairs ex- amined were smaller than the 95 mm CW minimum legal size, the smallest being 89 mm. Except for these, even solitary males of this size and smaller were absent from the area indicating that competi- tion for females had occurred in deeper water. This snow crab population appears to be small and is isolated from populations elsewhere in the Gulf of St Lawrence by the 35 m deep sill at the mouth of Bonne Bay. This area has not been fished commer- cially and at present the population is considered to be in the virgin state. Hooper's (in press) observa- tions indicated there is keen competition between single males and males already paired with a female for possession of the female. Under prefishery con- ditions this competition can be expected to eliminate small males from participating in breeding activi- ty. Adams (1982) demonstrated that muciparous female C. bairdi resisted mating attempts by small males, and when males of significantly different sizes competed for the same female, the larger male was invariably successful. Small numbers of the largest of the sublegal (<95 mm CW) male C. opilio appear to be capable of competing and mating suc- cessfully. However, it is presently unknown whether males smaller than those observed are capable of successful mating with multiparous females in the absence of competition from large males and, if they are not, whether there are sufficient numbers of large sublegal males to maintain full reproductive potential in a heavily fished population. Acknowledgments We are indebted to P. O'Keefe, G. Badcock, and D. Keats for assistance with field work, particular- ly diving, and to P. Collins, as well as P. O'Keefe and H. Mullett for assistance with data analysis and drafting. Literature Cited Adams, A. E. 1982. The mating behavior of Chionoecetes bairdi. In B. Melteff (editor), Proceedings of the International Symposium on the genus Chionoecetes. Lowell Wakefield Fisheries Sym- posia Series, p. 273-281. Univ. Alaska, Fairbanks, Sea Grant Rep. 82-10. Donaldson, W. E. 1975. Kodiak Tanner Crab Research. Technical Report for period July 1, 1974 to June 30, 1975. NOAA, NMFS, 69 p. Elner, R. W., and D. a. Robichaud. 1983. Observations on the efficacy of the minimum legal size for Atlantic snow crab, Chionoecetes opilio. CAFSAC Res. Doc. 83/63, 26 p. Hooper, R. G. In press. A spring breeding migration of the snow crab (Chionoecetes opilio) into shallow water in Newfoundland. Crustaceana. ITO, K. 1967. Ecological studies on the edible crab, Chionoecetes opilio 0. Fabricius in the Japan Sea. I. When do female crabs first spawn and how do they advance into the following reproductive stage? Bull. Jpn. Sea Reg. Fish Res. Lab. 17:67-84. [Engl, transl. from Jpn. by Fish Res. Board Can. Trans. Ser. No. 1103.] Miller, R. J., and P. G. O'Keefe. 1981. Seasonal and depth distribution, size, and molt cycle of the spider crabs, Chionoecetes opilio, Hyas araneus and Hyas coarctatus in a Newfoundland bay. Can. Tech. Rep. Fish. Aquat. Sci. 1003, 18 p. Paul. A. J. 1982. Mating frequency and sperm storage as factors affect- ing egg production in multiparous Chionoecetes bairdi. In B. Melteff (editor). Proceedings of the International Sym- posium on the genus Chionoecetes. Lowell Wakefield Fish- eries Symposia Series, p. 273-281. Univ. Alaska, Fairbanks, Sea Grant Rep. 82-10. Paul, A. J., and A. E. Adams. 1984. Breeding and fertile period for female Chionoecetes bairdi (Decapoda, Majidae). J. Crustacean Biol. 4:589-594. Powles, H. W. 1968. Distribution and biology of the spider crab Chionoecetes opilio in the Magdalen Shallows, Gulf of St. Lawrence. Fish. Res. Board Can. MS Rep. 997, 106 p. Squires, H. J., G. E. Tucker, and G. P. Ennis. 1971. Lobsters (Homarus americanus) in Bay of Islands, Newfoundland, 1963-65. Manuscr. Rep. Ser. (Biol.) No. 1151, 58 p. Taylor, D. M. 1 982. A recent development in tagging studies on snow crab, Chionoecetes opilio in Newfoundland - Retention of tags through ecdysis. In B. Melteff (editor). Proceedings of the International Symposium on the genus Chionoecetes. Lowell Wakefield Fisheries Symposia Series, p. 405-417. Univ. Alaska, Fairbanks, Sea Grant Rep. 82-10. Takeshita, K., and S. Matsuura. 1980. Mating and egg-laying in Tanner crabs. Bull. Far Seas Fish. Res. Lab. Watson, J. 1970. Maturity, mating and egg-laying in the spider crab, Chionoecetes opilio. J. Fish. Res. Board Can. 27:1607- 1616. 1972. Mating behavior in the spider crab, Chionoecetes opilio, J. Fish. Res. Board Can. 29:447-449. D. M. Taylor Fisheries Research Branch Department of Fisheries and Oceans 710 P.O. Box 5667 St. John's, Newfoundland AlC 5X1, Canada NICOS/Biology Department Memorial University of Newfoundland St. John's, Newfoundland A IB 3X9, Canada Fisheries Research Branch Department of Fisheries and Oceans P.O. Box 5667 St. John's, Newfoundland, AlC 5X1, Canada R. G. Hooper G. P. Ennis FEEDING, DIET, AND REPEAT SPAWNING OF BLUEBACK HERRING, ALOSA AESTIVALIS, FROM THE CHOWAN RIVER, NORTH CAROLINA Current knowledge of the frequency of feeding among spawning blueback herring, Alosa aestivalis Mitchill, is limited. Other aspects of the blueback herring's life history have been more extensively studied: feeding of juveniles (Davis and Cheek 1966; Nichols 1966; Burbidge 1974; Domermuth and Reed 1980; Crecco and Blake 1983), distribution at sea (Hildebrand 1963; Holland and Yelverton 1973^; Neves 1981), and spawning range (Bigelow and Schroeder 1953; Hildebrand 1963; Scott and Cross- man 1973). However, determination of the occur- rence of feeding by adults in freshwater has received little attention despite the fact that spawning bluebacks are common in rivers from southern New England (Bigelow and Schroeder 1953) to the St. Johns River, FL (Hildebrand 1963). Throughout this extensive range only Frankensteen (1976) has studied feeding among adult bluebacks in fresh- water. Furthermore, no attempt has been made to correlate feeding with length, weight, and sex of in- dividual fish, distance upstream, or the number of seasons a blueback has spawned. The objective of this study is to enhance our knowledge of the freshwater feeding of blueback herring. In this paper I describe the occurrence of feeding, diet, and percentage of repeat spawning among adults collected in the Chowan River, NC. I also examined, by multiple regression analysis, the relation between feeding activity in freshwater and length, weight, sex, the number of repeat spawnings, and the distance travelled upstream. Materials and Methods Collection of Data Bluebacks were collected at two sites in the lower Chowan River system during April 1980 and 1981. Williams' Fishery, where five collections were made in 1980, is located on the lower Meherrin River near its junction with the Chowan River, 90 km upstream from the Chowan River's mouth. Rocky Hock Creek, where bluebacks were sampled twice in 1980 and once in 1981, is roughly 20 km from the mouth of the Chowan River. Bluebacks at Williams' Fishery were still migrating upstream while those at Rocky Hock Creek, a known spawning ground,^ were preparing to spawn. Bluebacks were caught in chicken-wire dip nets and fixed gill nets with 58 mm stretched mesh at Rocky Hock Creek. A drift gill net of similar mesh size and a haul seine were used at Williams' Fishery. None of the fish collected had spawned yet. Bluebacks were measured, weighed, and sexed, and scales were removed for aging. The foregut and midgut regions of the stomach anterior to the pyloric caeca were removed and placed in 15% Formalin^ within 10-15 min of capture Stomach contents were examined in the laboratory under a dissecting scopa First, fullness of the foregut and midgut, which are separate sections, was estimated visually following Hynes (1950) and Yoshiyama (1980). Five levels of fullness were used: half full (1/2), full (1), and distended with food (2) (as in Yoshiyama 1980), plus one quarter full (1/4), and empty or with traces of food (0). Contents of each section were then placed in a petri dish, iden- tified, and counted. Also, the presence or absence of prey items was noted. Scales were viewed at 50 x through an EPO LP-2 Profile Projector and marks were interpreted follow- ing Marcy (1969). iRolland, B. F., Jr., and G. F. Yelverton. 1973. Offshore anadromous fish exploratory fishing program. Completion report, Project AFC-5, 123 p. North Carolina Department of Natural and Economic Resources, Division of Commercial and Sports Fisheries, Raleigh, NC 27611. ^S. Winslow, North Carolina Division of Marine Fisheries, Elizabeth City, NC 27909, pers. commun. February 1980. S. Winslow had determined the previous year (1979) that blueback herring collected at this site on Rocky Hock Creek were spawn- ing. Also, a dam upstream prevented blueback herring from moving any further than 150 m above my collection sita ^Reference to trade names does not imply endorsement by the National Marine Fisheries Service, NOAA. FISHERY BULLETIN: VOL. 83, NO. 4, 1985. 711 Statistical Analysis lb obtain a single index of stomach fullness, values for the foreguts and midguts were combined by using a weighted average In calculating mean volumes each gut section was assumed to be approx- imately cylindrical. The foregut to midgut volume ratio, determined from five randomly chosen stomachs, was 3.16:1. The following equation was used to calculate the overall gut fullness: F X 3.16 + M 4.16 with F and M representing the foregut and midgut values, respectively. A multiple regression (General Linear Model- Statistical Analysis Systems) was initially employed to determine which of the variables collected for each fish (i.e, distance upstream, length, weight, sex, and number of repeat spawnings) was most strongly cor- related with stomach fullness, the dependent variable Significant variables identified through multiple regression analysis were further analyzed with chi-square and F-tests. Results Presence of Food Nearly all (91 of 103 or 88%) fish sampled in April 1980 contained food (Tkble 1) as did all 15 fish col- lected in April 1981. High percentages of the fish collected on each date in 1980 had food in their stomachs (Tkble 1). About 53% (48 of 91) of the blue- backs in 1980 (Tkble 1) and 73% (11 of 15) of the blue- backs in 1981 had either foregut and/or midgut fullnesses of 1/4 or greater. Approximately half of the fish from each date in 1980, with the exception of 13 April, had either foregut and/or midgut full- nesses of 1/4 or greater (Tkble 1). Diet The diet of the bluebacks collected in 1980 at both sites was composed of zooplankters, benthos, and terrestrial insects (Tkble 2). Chydorid cladocerans were the only zooplankters consumed in large numbers at either location (Figs. 1, 2). Insects, which accounted for 8.1% of the organisms consumed, oc- curred in about half of the fish. Ephemeroptera (Baetis), Coleoptera (Dytiscidae), and Heleidae lar- vae, as well as chironomid larvae and pupae, were the most conspicuous of the identifiable benthic in- sects. Most insects, benthic and terrestial, were unidentifiable. Chironomids occurred more frequent- ly than other insect groups, but they accounted for only 2.7% of the total prey items. Several terrestrial insects were found in stomachs of bluebacks, par- ticularly at Williams' Fishery. Insects, both benthic and terrestrial, increased in importance with time at Williams' Fishery, reaching about 22% during later collections (Fig. 1). Insects represented a smaller proportion of the diet at Rocky Hock Creek (Fig. 2). Fish eggs (probably from alewives or blue- backs), which occurred in the stomachs of several bluebacks in 1980 (Figs. 1, 2), were the most abun- dant food item in that year although their impor- tance decreased with time Varying amounts of sand and detritus occurred in many stomachs. In 1981 the diet of bluebacks from Rocky Hock Creek was much less diverse (Tkble 3). Cladocerans, the predominant prey items, comprised 84.1% of the diet. Almost half of the prey items were daphnid Table 1.— Incidence of feeding and stomach fullness in male and female blueback herring collected at Williams' Fishery (WF) and Rocky Hock Creek (RH) during April 1980. No. fish collected per station N No. males (M) and females (F) M F No. fish (n) with animal matter In stomach n % F No. of fish with >1/4 fullness for the foregut, midgut or both Station and date of collection Males Females Male Fem Total JS + ales Fore Mid Both Fore Mid i 3oth % N WF-4-5-80 16 8 8 12 75.0 0 1 0 1 2 3 7 43.6 RH -4-6-80 9 3 6 9 100.0 0 0 0 2 1 2 5 55.6 WF-4-7-80 17 8 9 15 88.2 1 3 0 0 2 3 9 52.9 WF-4-11-80 17 10 7 15 88.2 2 2 0 1 2 1 8 47.1 RH -4-1 2-80 7 6 1 7 100.0 1 1 0 0 0 1 3 42.9 WF-4-13-80 23 9 14 20 87.0 2 1 0 0 4 1 8 34.8 WF-4-19-80 14 10 4 13 92.9 0 5 0 0 2 1 8 57.1 Total 103 54 49 91 88.4 6 13 0 4 13 12 48 712 cladocerans (Ikble 3). Few copepods were consum- ed, although they occurred in about half the fish. Ostracods were important numerically and in occur- Table 2.— Diets of 103 blueback herring collected at Williams' Fishery and Rocky Hock Creek during April 1980. Frequencv 'Of Proportions of occurrence prey items in diet No. of fish Prey taxon observed in % No. of prey % Copepoda Calanoida 1 1.0 1 V) Harpacticoida 5 4.9 12 0.4 Cyclopoida 35 33.9 80 2.8 Unidentifiable 7 6.8 21 0.7 Cladocera Chydoridae 61 59.2 839 29.6 Other families 14 13.6 26 0.9 Ostracoda 15 14.6 58 2.0 Insecta Miscellaneous 54 52.4 133 4.7 Diptera Chironomidae larvae 25 24.3 67 2.4 Chironomidae pupae 5 4.9 7 0.3 Other families 12 11.7 20 0.7 Oligochaeta 1 1.0 1 V) Mollusca Sphaerildae 7 6.8 21 0.7 Fish eggs 43 41.8 1 ,273 44.9 Bryozoan statoblasts 20 19.4 178 6.3 Colonial hydrozoan pieces 16 15.5 96 3.4 Hydracarina 5 4.9 5 0.2 Total 2,838 100.0 '<0.1%. Table 3.— Diets of 15 blueback herring collected at Rocky Hock Creek on 18 April 1981. Frequency of Proportions of occurrence prey items in diet No. of fish Prey taxon observed in % No. of prey % Copepoda Calanoida 5 33.3 8 0.1 Harpacticoida 7 46.6 7 0.1 Cyclopoida 7 46.6 23 0.4 Cladocera Daphnidae 15 100.0 3,191 48.7 Daphnidae ephippia 12 80.0 1,825 27.9 Chydoridae 13 86.6 493 7.5 Bosminidae 2 13.3 5 V) Ostracoda 13 86.6 929 14.2 Insecta Miscellaneous 5 33.3 6 0.1 Diptera Chironomidae larvae 5 33.3 8 0.1 Fish eggs 1 6.6 7 0.1 Bryozoan statoblasts 11 73.3 42 0.6 Hydracarina 3 20.0 3 D Fish larvae 1 6.6 1 V) Total 6,548 99.8 rence, as were daphnid ephippia. Benthic prey, ter- restrial insects, fish eggs, detritus, and sand were rare Number of Repeat Spawnings About 85% (87 of 103) of the fish in 1980 and 71% (10 of 14) in 1981 had spawned before (Ikble 4). Although some bluebacks in 1980 had spawned as many as six times, most (72%) had spawned only once or twice before (Tkble 4). Almost equal numbers of males (44) and females (43) were repeat spawners in 1980. Table 4.— Number of previous years that 103 blueback herring collected at Williams' Fishery and Rocky Hock Creek during April 1980 had spawned. No. of years spawned 1 (first time) 2 3 4 5 6 Male Female Total 10 25 14 3 1 6 22 13 6 2 16 47 27 9 3 1 0 1 Relation Between Feeding Activity and Length, Weight, Sex, Number of Repeat Spawnings, and Distance Upstream Sex was the only independent variable that con- tributed significantly to variance in feeding rate (Tkble 5). Female bluebacks fed more actively than males. The full model explained only 21% of the variance in gut fullness but this was significant {F = 2.87, P>F = 0.005, 102 df). The near significant contributions of length and weight to the reduction of variance in the model is believed to have resulted primarily from sex related differences in mean size (x length females = 305.1 mm, x length males = 289.4 mm; x weight females = 258.0 g, x weight males = 212.5 g). F^irther analysis of the feeding activity of male and female bluebacks with a chi- Table 5.— Summary of the contribution of each of the independent variables to the multiple regression model for data from 103 blueback herring collected in 1980. Type IV F values and probability levels are shown. P < 0.05 significance level used. KO.1%. Independent variable F value Probability Site 2.64 0.1078 Sex 7.14 0.0089 No. of repeat spawnings 1.65 0.1818 Length 3.85 0.0527 Weight 3.81 0.0540 718 100 UJ Q 90- INSECTS CHYCXDRIDS 80- 70- ^ 60-] m u Ll O 40H y 3o^ on u ^ 20 10- 0 12/16 OTHERS HYDRO- ZOANS FISH EGGS 4/5 15/17 OTHERS (OSTRAQODS STATO- BLASTS HYDRO - ZOANS CHIRON- OMIDS CYCLO- POIDS INSECTS CHYDORIDS Fl SH EGGS 15/17 OTHERS STATO- B LASTS HYDRO- 20ANS CHIRON - OMIDS CYCLO- POIDS INSECTS CHYDORIDS FISH EGGS 20/23 OTHERS HYDRO- ZOANS CHIRON- OMIDS ■Y CYCLO- POIDS INSECTS CHYDORIDS FISH EGGS 13/14 OTHERS STATO- B LASTS CHIRON- OMIDS ■Y CYCLO- POIDS INSECTS ICHYDORIDS ^ FISH EGGS ^ 4/7 4/11 4/13 COLLECTION DATES (1980) 4/19 Figure 1— Changes in the composition of the combined diet (both males and females) over the five collection dates at Williams' Fishery in April 1980. Numbers above each bar graph indicate number of stomachs with food/total number of stomachs examined. square 2x2 contingency table also found signif- icantly greater levels of feeding activity among females (x^ value = 5.86, P < 0.025). The difference in 1980, however, may depend on site. More females had stomach fullnesses >l/4 than males at Rocky Hock (x^ value 6.349, P < 0.025), but the difference was not significant at Williams' Fishery. For the three most abundant food items females consumed significantly greater numbers of chydorid cladocerans (F = 6.02, P>F = 0.0001), insects (F = 7.64, P>F = 0.0001), and fish eggs (F = 90.15, P>F= 0.0001) than males in 1980. Discussion Blueback herring spawning in the Chowan River do not stop feeding during their freshwater migra- tions. Williams' Fishery and Rocky Hock Creek are 714 100 90 9/9 7/7 UJ Q Q LlI CD §50 O 40 OSTR/COCS 80- 70- CHYDORiDS 60- LlI Q_ 20 H 10- 0 OTHERS CHIRON- CYCLO- POIDS INSECTS FISH EGGS OSTRACOCE CHYDORIDS OTHERS FISH EGGS 4/6 4/12 COLLECTION DATES (1980) Figure 2.— Changes in the composition of the combined diet (both males and females) over the two collection dates at Rocky Hock Creek in April 1980. Numbers above each bar graph indicate number of stomachs with food/total number of stomachs examined. too far from the estuary for bluebacks to travel to saltwater for daily feeding. It is important to note, also, that the interface between freshwater and salt- water often extends far out into Albemarle Sound due to the spring discharge of both the Chowan and Roanoke Rivers. This was the case in the spring of 1980.'* Moreover, the prey were exclusively of fresh- water origin. There is, therefore, little doubt that these bluebacks were feeding in freshwater. The wide diversity of food items consumed was unexpected since bluebacks have previously been reported to be primarily planktivorous (Bigelow and Schroeder 1953; Hildebrand 1963). The limited 1981 data suggest that prey other than zooplankters are consumed infrequently if sufficient zooplankters (or large zooplankters such as Daphnia) are present. However bluebacks are also capable, as the 1980 data demonstrate, of foraging opportunistically on other riverine fauna and terrestrial insects, which could also explain Frankensteen's (1976) unusual finding that chironomids were the dominant prey of blueback herring in the Ikr River. Consumption of benthic prey probably accounts for the presence of detritus and sand in the guts. My data show a difference between male and female feeding activity. There are two possible ex- planations for this difference First, females may re- quire more energy than males during the spawning migration thus they consume more prey. Neither this study nor other studies of bluebacks have produced data to either support or refute this idea. However, moderate to severe weight loss is common among other spawning anadromous fishes (eg., Atlantic salmon, American shad) with females suffering greater weight loss than males (Belding 1934; Chit- tenden 1976; Glebe and Leggett 1981). Glebe and Leggett (1981) found that development of ovaries in female shad required more energy and time than the male shads' testes. Consequently, female shad enter- ing freshwater, particularly southern rivers, often do not have fully developed ovaries. Thus, not only must females expend energy for swimming but for gonad development as well. The same difference in gonad development may exist between male and female bluebacks and could explain the different levels of feeding activity observed in this study. The second explanation for the difference in feeding activity is that all bluebacks, regardless of sex, stop feeding while spawning. However, males might remain on the spawning grounds longer than females. Thus, if females leave the area immediate- ly after they spawn and are replaced by newly ar- rived females with relatively full guts, this could cause the gut samples to be biased. This explana- tion appears to be ruled out by the 1981 data, however, since half the fish with stomachs >1/A full collected at Rocky Hock were males. While previous researchers have found food in bluebacks' stomachs (Williams et al. 1975 as cited ■•R. Holmes, Department of Natural Resources and Community Development, Division of Environmental Management, Raleigh, NC 27611, pers. commun. March 1984. 715 in Rulifson et al. 1982; Frankensteen 1976), my find- ing of a regular occurrence of significant volunnes of food in blueback stomachs is unprecedented. Fur- ther research is needed to determine the extent to which feeding in freshwater is common among spawning bluebacks in other river systems, and possibly other anadromous species, and to determine if a relationship exists between freshwater feeding and spawning energetics. Acknowledgments I am grateful to Seth Reice and Edward Kuenzler for providing assistance, materials, and workspace for this study. I also wish to thank Jerry Diamond, Seth Reice, William Leggett, Angela Arthington, Rob Edwards, Brad Foster, and three anonymous reviewers for their careful readings of the manu- script. Assistance from Mr. and Mrs. Williams and the staff of Williams' Fishery, Sara Winslow, Bar- rel Johnson, and Michael Street of the North Carolina Division of Marine Fisheries, David Ham- mer, Chris Nations, Ernie Patterson, and Diana Hyland, as well as thought-provoking discussions with many other people, is greatly appreciated. Literature Cited Belding, D. L. 1934. The cause of the high mortahty in the Atlantic salmon after spawning. Trans. Am. Fish. Soc 64:219-224. BiGELOW, H. B., AND W. C. SCHROEDER. 1953. Fishes of the Gulf of Maine U.S. Fish Wildl. Serv., Fish. Bull. 53, 577 p. BURBIDGE, R. G. 1974. Distribution, growth, selective feeding, and energy transformations of young-of-the-year blueback herring, Alosa aestivalis (Mitchill), in the James River, Virginia. TVans. Am. Fish. Soc 103:297-311. Chittenden, M. E., Jr. 1976. Weight loss, mortality, feeding, and duration of residence of adult American shad, Alosa sapidissima, in fresh water. Fish. Bull., U.S. 74:151-157. Crecco, V. A., AND M. M. Blake. 1983. Feeding ecology of coexisting larvae of American shad and blueback herring in the Connecticut River. Trans. Am. Fish. Soc 112:498-507. Davis, J. R., and R. R Cheek. 1966. Distribution, food habits, and growth of young clupeids, Cape Fear River system. North Carolina. Proc 20th Annu. Conf., SE Assoc Game Fish. Comm., p. 250-259. DOMERMUTH, R. B., AND R. J. REED. 1980. Food of juvenile American shad, Alosa sapidissima, juvenile blueback herring, Alosa aestivalis, and pumpkinseed, Lepomis gihhosus, in the Connecticut River below Holyoke Dam, Massachusetts. Estuaries 3:65-68. Frankensteen, E. D. 1976. Genus Alosa in a channelized and an unchannelized creek of the Tkr River basin. North Carolina. M.A. Thesis, East Carolina University, Greenville, NC, 123 p. Glebe, B. D., and W. C. Leggett. 1981. Latitudinal differences in energy allocation and use dur- ing the freshwater migrations of American shad (Alosa sapidissima) and their life history consequences. Can. J. Fish. Aquat. Sci. 38:806-820. Hildebrand, S. F. 1963. Family Clupeidae In H. B. Bigelow (editor). Fishes of the Western North Atlantic, Part 3, p. 257-454. Sears Found. Mar. Res., Yale University, New Haven, CT. Hynes, H. B. N. 1950. The food of fresh-vrater sticklebacks (Gasterostevs aculeat-us and Pygosteus pungitivs), with a review of methods used in the studies of the food of fishes. J. Anim. Ecol. 19:36-58. Marcy, B. C, Jr. 1969. Age determinations from scales of Alosa pseudoharen- giis (Wilson) and Alosa aestivalis (Mitchill) in Connecticut waters. Trans. Am. Fish. Soc 98:622-630. Neves, R. J. 1981. Offshore distribution of alewif e, A losa pseudoharengus, and blueback herring, Alosa aestivalis, along the Atlantic coast. Fish. Bull, U.S. 79:473-485. Nichols, P. R. 1966. Anadromous Fishes Program. In Annual Report of the Bureau of Commercial Fisheries Biological Laboratory, Beau- fort, N.C., for the fiscal year ending June 30, 1966, p. 4-8. U.S. Fish Wildl. Serv. Circ 264. Rulifson, R. A., M. T. Huish, and R. W. Thoesen. 1982. Anadromous fish in the southeastern United States and recommendations for development of a management plan. U.S. Fish Wildl. Serv., Fish. Res., Reg. 4, Atlanta, GA, 525 p. Scott, W. B., and E. J. Grossman. 1973. Freshwater fishes of Canada. Fish. Res. Board Can. Bull. 184, 966 p. Yoshiyama, R. M. 1980. Food habits of three species of rocky intertidal sculpins (Cottidae) in central California. Copeia 1980:515-525. Robert P. Creed, Jr. Department of Biology University of North Carolina Chapel Hill, NC 275U Present address: Department of Zoology Michigan State University East Lansing, MI Jt882jlt 716 INDEX Fishery Bulletin Vol. 83, Nos. 1-4 Age-composition anchovy, northern 483 Age determination alewife 696 fishes 103 multiple regression models 108 scale and otolith methods 696 "Age growth and distribution of larval spot Leiostomus xanthurus" by Stanley M. Warlen and Alexander J. Chester 587 ALBERS, W. D, and P. J. ANDERSON, "Diet of Pacific cod, Gadvs macrocephalus, and predation on the northern pink shrimp, Pandalus horealis, in Pavlof Bay, Alaska". . . 601 ALLEN, LARRY G.-see DeMARTINI et al. Alosa aestivalis— see Herring, blueback Alosa pseudoharengus—see Alewife Ampelisca agassizi—see Amphipods, benthic Amphipods, benthic parasites of 497 Anchovy, northern egg production and mortality rate 137 growth and age composition 483 rates of ovarian atresia 119 schooling behavior 235 ANDERSON, P J.-see ALBERS and ANDERSON ANKENBRANDT, LISA, "Food habits of bait-caught skip- jack tuna, Katsuwomis pelamis, from the southwestern Atlantic Ocean" 379 "Annual band deposition within the shells of the hard clam, Mercenaria mercenaria: consistency across habitat near Cape Lookout, North Carolina," by Charles H. Peterson, P. Bruce Duncan, Henry C. Summerson, and Brian F. Beal 671 Anoplopoma fimbria— see Sablefish "An approach to estimating an ecosystem box model," by Jeffrey J. Polovina and Mark D. Ow 457 "Aspects of the life history of the fluffy sculpin, Oligocot- tu£ snyderi" by Mary C. Freeman, Nate Neally, and Gary D. Grossman 645 AU, DAVID W. K., and WAYNE L. PERRYMAN, "Dolphin habitats in the eastern tropical Pacific" 623 Balaena misticetvs—see Whales, bowhead BARLOW, JAY-see GERRODETTE et al. BARLOW, JAY, "Variability, trends, and biases in reproduc- tive rates of spotted dolphins, Stenella attenuata" 657 BARNES, A.-see LESTER et al. BEAL, BRIAN F.-see PETERSON et al. "Behavior of bowhead whales, Balaena mysticetus, sum- mering in the Beaufort Sea: a description," by Bernd Wiir- sig, Eleanor M. Dorsey, Mark A. Fraker, Roger S. Payne, and W John Richardson 357 Behavior studies anchovy, northern 235 dolphins 187 whales, bowhead 357 BERMINGHAM, E. L.-see McFARLAND et al. "Biological aspects of the spring breeding migration of snow crab, Chionoecetes opilio, in Bonne Bay, Newfound- land (Canada)," by D. M. Ikylor, R. G. Hooper, and G P Ennis 707 Bocaccio seasonal changes in fat and gonad volume 299 BOEHLERT, GEORGE W, "Using objective criteria and multiple regression models for age determination in fishes" 103 BOEHLERT, GEORGE W, and MARY M. YOKLAVICH, "Larval and juvenile growth of sablefish, Anoplopoma fim- bria, as determined from otolith increments 475 BOEHLERT GEORGE W, DENA M. GADOMSKI, and BRUCE C. MUNDY, "Vertical distribution of ichthyo- plankton off the Oregon coast in spring and summer months" 611 BROOKS, E. R.-see MULLIN et al. BROTHERS, E. B.-see McFARLAND et al. BROWN, R. S., and N. CAPUTI, "Factors affecting the growth of undersize western rock lobster, Panulirvs cygnus George, returned by fishermen to the sea" 567 CAMPBELL, A.-see JAMIESON and CAMPBELL Cancer magister—see Crab, Dungeness 717 CAPUTI, N.-see BROWN and CAPUTI Carcharhinus plumbevs—see Sharks, sandbar CASEY, JOHN J.-see MEDVED et al. Catshark life history notes 695 CHAN. BRIAN-see NEILSON et al. CHESTER, ALEXANDER J.-see WARLEN and CHESTER Chilipepper seasonal changes in fat and gonad volume 299 Chionoecetes opilio—see Crab, snow Clam, hard growth band deposition 671 Clam, soft-shell spawning cycle in San Francisco Bay 403 Clupea harengus harengus— see Herring, Atlantic Cod, Pacific diet and predation in Pavlof Bay, Alaska 601 COLLINS, ROBSON A.-see LOVE et al. "A comparison of scale and otolith aging methods for the alewife, Alosa pseudoharengus" by David A. Libby .... 696 "Confidence limits for population projections when vital rates vary randomly" by Tim Gerrodette, Daniel Goodman, and Jay Barlow 207 CONOVER, DAVID 0., "Field and laboratory assessment of patterns in the fecundity of a multiple spawning fish: the Atlantic silverside" 331 Crab, Dungeness salmonid predation 683 Crab, snow spring breeding migration 707 Crangonidae— see Shrimp CREED, ROBERT R, Jr., "Feeding, diet, and repeat spawn- ing of blueback herring, Alosa aestivalis, from the Chowan River, North Carolina" 711 CROSS, JEFFREY N., "Fin erosion among fishes collected near a southern California municipal wastewater outfall (1971-1982)" 195 Delphinus delphis—see Dolphins, common DeMARTINI, EDWARD E., LARRY G. ALLEN, ROBERT K. FOUNTAIN, and DALE ROBERTS, "Diel and depth variations in the sex-specific abundance, size composi- tion, and food habits of queenfish, Seriphus politus (Sciaenidae)" 171 "The development and occurrence of larvae of the longfin Irish lord, Hemilepidotiis zapus (Cottidae)," by Ann C. Matarese and Beverly M. Vinter 447 Dichelopandalus leptocems—see Shrimp "Diel and depth variations in the sex-specific abundance, size composition, and food habits of queenfish, Seriphus politiis (Sciaenidae)" by Edward E. DeMartini, Larry G. Allen, Robert K. Fountain, and Dale Roberts 171 Diet— see Food habits "Diet of Pacific cod, Gadiis macrocephalus, and predation on the northern pink shrimp, Pandalus borealis, in Pavlof Bay, Alaska," by W. D. Albers and P. J. Anderson 601 "Distributional patterns of fishes captured aboard commer- cial passenger fishing vessels along the northern Channel Islands," by Milton S. Love, William Westphal, and Robson A. Collins 243 "Dolphin habitats in the eastern tropical Pacific," by David W. K. Au and Wayne L. Perryman 623 Dolphins habitats in the eastern tropical Pacific 623 incidental mortality 521 reactions to population survey vessels 187 Dolphins, common undersea topography and distribution 472 Dolphins, spotted growth rates 553 reproductive rates 657 DORSEY, ELEANOR-see WURSIG et al. DUNCAN, BRUCE P.-see PETERSON et al. "Early postnatal growth of the spotted dolphin, Stenella attenuata" by Aleta A. Hohn and P. S. Hammond 553 Economic studies rock shrimp 1 Ecosystems the ECOPATH model 457 estimating a box model 457 "Effects of feeding regimes and diel temperature cycles on otolith increment formation in juvenile chinook salmon, On- corhynchvs tshawytscha" by John D. Neilson and Glen H. Geen 91 "The effects of net entanglement on the drag and power output of a California sea lion, Zalophus califomianus" by Steven D. Feldkamp 692 "Egg production of the central stock of northern anchovy, Engraulis mordax" by Nancy C. H. Lo 137 718 Eggs— see also Embryos Embryos salmonid 81 Engraulis mordax—see Anchovy, northern ENNIS, G. P.-see TAYLOR et al. Environmental effects anchovy, northern 483 dolphin habitats 623 food web 151 grunts, French 413 larval fish 313 mummichog 467 porpoise, harbor 427 tilefish 443 "Factors affecting the growth of undersize western rock lobster, PanuliTus cygnus George, returned by fishermen to the sea," by R. S. Brown and N. Caputi 567 "Feeding, diet, and repeat spawning of blueback herring, Alosa aestivalis, from the Chowan River, North Carolina," by Robert R Creed, Jr. 711 FELDKAMP, STEVEN D, "The effects of net entangle- ment on the drag and power output of a California sea lion, Zalophus califomianus" 692 "Field and laboratory assessment of patterns in fecundity of a multiple spawning fish: the Atlantic silverside," by David 0. Conover 331 "Fin erosion among fishes collected near a southern Califor- nia municipal wastewater outfall," by Jeffrey N. Cross. . . 195 Fish distributional patterns in the Channel Islands 243 Fish assemblages, demersal estimates of marine populations 508 temporal and spatial patterns 507 Fish, larval distribution and abundance in the northeastern U.S. ... 313 Fishery, eastern Pacific shrimp 1 Fishery, Gulf of Mexico juvenile brown shrimp as abundance predictors 677 Fishery, tima parasite use for stock management 343 Fishery, western Australian lobster, western rock 567 FLIERL, G. R., and J. S. WROBLEWSKI, "The possible influence of warm core Gulf Stream rings upon shelf water larval fish distribution" 313 "Food and feeding of the tomtate, Haemulon aurolineatum (Pisces, Haemulidae) in the South Atlantic Bight," by George R. Sedberry 461 "Food habits of bait-caught skipjack tuna, Katsuwonus pelamis, from the southwestern Atlantic Ocean," by Lisa Ankenbrandt 379 "Food habits of juvenile rockfishes (Sebastes) in a central California kelp forest," by Michael M. Singer 531 Food habits queenfish 171 rockfish 531 sharks, sandbar 395 tomtate 461 tuna, skipjack 379 FOUNTAIN, ROBERT K.-see DeMARTINI et al. FRAKER, MARK A.-see WURSIG et al. FREEMAN, MARY C, NATE NEALLY, and GARY D. GROSSMAN, "Aspects of the life history of the fluffy scizlpin Oligocottus snyderi 645 Fundulus heteroclittis—see Mummichog GADOMSKI, DENA M.-see BOEHLERT et al. Gadus mojcrocephalus—see Cod, Pacific GASKIN, DAVID E.-see READ and GASKIN GASKIN, DAVID E., and ALAN P WATSON, "The harbor porpoise, Phocoena phocoena, in Fish Harbour, New Brunswick, Canada: occupancy, distribution, and move- ments" 427 GERRODETTE, TIM, DANIEL GOODMAN, and JAY BARLOW, "Confidence limits for population projections when vital rates vary randomly" 207 GEEN, GLEN H.-see NEILSON and GEEN GEEN, GLEN H.-see NEILSON et al. Globicephala macrorhynchus—see Whales, Pacific pilot GOODMAN, DANIEL-see GERRODETTE et al. GROSSMAN, GARY D.-see FREEMAN et al. GROSSMAN, GARY D., MICHAEL J. HARRIS, and JOSEPH E. HIGHTOWER, "The relationship between tilefish, Lopholatilus chamaeleonticeps, abundance and sedi- ment composition off Georgia" 443 Growth rates dolphin, spotted 553 herring, Atlantic 289 lobster, rock 567 Grunts, French recruitment patterns in Hague Bay, Virgin Islands ... 413 719 GUILLEMOT, PATRICK J., RALPH J. LARSON, and WILLIAM H. LENARZ, "Seasonal cycles of fat and gonad volume in five species of northern California rockfish". . . 299 HABIB, G.-see LESTER et al. Haemulon aurolineatum—see Ibmtate Haemulon Jlavolineatum—see Grunts, French HAMMOND, R S.-see HOHN and HAMMOND "The harbor porpoise^ Phocoena phocoena, in Fish Harbour, New Brunswick, Canada: occupancy, distribution, and move- ments," by David E. Gaskin and Alan P. Watson 427 HARRIS, MICHAEL J.-see GROSSMAN et al. HAYNES, EVAN B., "Morphological development, iden- tification, and biology of larvae of Pandalidae; Hippolytidaei and Crangonidae (Crustacea, Decapoda) of the northern North Pacific Ocean" 253 Hemilepidotus zapus—see Lord, longfin Irish Herring, Atlantic growth comparison studies 289 Herring, blueback diet and spawning in the Chowan River, North Caro- lina 711 HEWITT, ROGER R, "Reactions of dolphins to a survey vessel: effects on census data 187 HIGHTOWER, JOSEPH E.-see GROSSMAN et al. HINES, ANSON H., KENRIC E. OSGOOD, and JOSEPH J. MIKLAS, "Semilunar reproductive cycles in Fundulus heteroclitus (Pisces: Cyprinodontidae) in an area without lunar tidal cycles" 467 Hippolytidae— see Shrimp HOHN, ALETA A, and R S. HAMMOND, "Early postnatal growth of the spotted dolphin, Stenella attenuata, in the offshore eastern tropical Pacific" 553 Homarus americanvs—see Lobster, American HOOPER, R. G.-see TAYLOR et al. HORTON, HOWARD F.-see MAULE and HORTON HUI, CLIFFORD A., "Undersea topography and com- parative distributions of two pelagic cetaceans 472 HUNTER, J. ROE, and BEVERLY J. MACEWICZ, "Rates of atresia in the ovary of captive and wild northern anchovy, Engraulis mordax" 119 HUNTER, JOHN, and RAGAN NICHOLL, "Visual threshold for schooling in northern anchovy, Engraulis mordax" 235 720 Ichthyoplankton vertical distribution off the Oregon coast 611 JAMIESON, G. S., and A. CAMPBELL, "Sea scallop fishing impact on American lobsters in the Gulf of St. Lawrence" 575 JOHNSON, PHYLLIS T, "Parasites of benthic amphipods: microsporidans of Ampelisca agassizi (Judd) and some other gammarideans" 497 JONES, CYNTHIA, "Within-season differences in growth of larval Atlantic herring, Clupea harengiis harengus . . . 289 Katsuwonits pelamis—see Tbna, skipjack KOTCHIAN-PRENTISS, N. M.-see McFARLAND et al. LARSON, RALPH J.-see GUILLEMOT et al. Larvae fish, environmental effects 313 food web off southern California coast 151 herring, Atlantic, growth studies 289 lord, longfin Irish, development in Bering Sea 447 sable fish grovrth 475 shrimp 253 spot 587 "Larval and juvenile growth of sablefish, Anoplopoma fim- bria, as determined from otolith increments," by George W Boehlert and Mary M. Yoklavich 475 Leiostomus xanthurvs—see Spot LENARZ, WILLIAM H.-see GUILLEMOT et al. LESTER, R. J. G., A. BARNES, and G. HABIB, "Para- sites of skipjack tuna, Katsuwonus pelamis: fishery implications" 343 LIBBY, DAVID A., "A comparison of scale and otolith aging methods for the alewife, Alosa pseudoharengus" 696 "Life history characteristics of Pandalus montagui and Dichelopandalus leptocerus in Penobscot Bay, Maine," by David K. Stevenson and Fran Pierce 219 LO, NANCY C. H., "Egg production of the central stock of northern anchovy, Engraulis mordax" 137 Lobster, American damage from scallop drags 575 Lobster, rock factors affecting growth 567 "Long-term responses of the demersal fish assemblages of Georges Bank," by William J. Overholtz and Albert V. Tyler 507 Lopholatilus chamaeleonticeps—see Tilefish Lord, longfin Irish development of larvae 447 LOVE, MILTON S., WILLIAM WESTPHAL, and ROB- SON A. COLLINS, "Distributional patterns of fishes cap- tured aboard commercial passenger fishing vessels along the northern Channel Islands" 243 Lyopsetta exilis—see Ichthyoplankton plankton off southern California: a storm and a larval fish food web" 151 Mummichog semilunar reproductive cycles . MUNDY, BRUCE C.-see BOEHLERT et al. Mya arenaria—see Clam, soft-shell 467 MACEWICZ, BEVERLY J.-see HUNTER and MACEWICZ MAIS, K. F.-see PARRISH et al. MALLICOATE, D. L.-see PARRISH et al. MATARESE, ANN C, and BEVERLY M. VINTER, "The development and occurrence of larvae of the longfin Irish lord, Hemilepidotus zajms (Cottidae)" 447 MAULE, ALEC G., and HOWARD F. HORTON, "Prob- able causes of the rapid growth and high fecundity of walleye, Stizostedion vitreum vitreum in the mid-Columbia River" 701 McFARLAND, W. N., E. B. BROTHERS, J. C. OGDEN, M. J. SHULMAN, E. L. BERMINGHAM, and N. M. KOTCHIAN-PRENTISS, "Recruitment patterns in young French grunts, Haemulon flavolineatum (family Haemulidae) at St. Croix, Virgin Islands" 413 MEDVED, ROBERT J., CHARLES E. STILLWELL, and JOHN J. CASEY, "Stomach contents of young sandbar sharks, Carcharhinus plumbeus, in Chincoteague Bay, Virginia" 395 Menidia menidia—see Silverside, Atlantic Mercenaria mercenaria—see Clam, hard Migration crab, snow 707 ichthyoplankton 611 lobster, American 575 porpoise, harbor 543 queenfish 171 salmon, coho 682 MIKLAS, JOSEPH J.-see HINES et al. Morphology lord, longfin Irish, larvae 447 shrimp, larvae 253 shrimp, rock 1 "Morphological development, identification, and biology of larvae of Pandalidae, Hippolytidae, and Crangonidae (Crustacea, Decapoda) of the northern North Pacific Ocean," by Evan B. Haynes 253 Mullin, M. M., E. R. BROOKS, F M. H. RE ID, J. NAPP, and E. R. STEWART, "Vertical structure of nearshore NAPP, J.-see MULLIN et al. NEALLY, NATE-see FREEMAN et al. NEILSON, JOHN D, GLEN H. GEEN, and BRIAN CHAN, "Variability in dimensions of salmonid otolith nuclei: implications for stock identification and microstruc- ture interpretation" 81 NEILSON, JOHN D., and GLEN H. GEEN, "Effects of feeding regimes and diel temperature cycles on otolith in- crement formation in juvenile chinook salmon, Oneorhyn- chus tshawytscha" 91 NIESEN, THOMAS M.-see ROSENBLUM and NIESEN "Notes on the life history of the catshark Scyliorhiniis meadi" by Glenn R. Parsons 695 "Observer effect on incidental dolphin mortality in the eastern tropical Pacific tuna fishery," by Bruce E. Wahlen and Tim D. Smith 521 OGDEN, T C.-see McFARLAND et al. Oligocottus snyderi—see Sculpin, fluffy Oncorhynchus tshawytscha— see Salmon, chinook OSGOOD, KENRIC E.-see HINES et al. Otoliths alewives 696 herring, Atlantic 289 multiple regression models 103 rockfish 103 sablefish 475 salmon, chinook 81, 91 trout, rainbow 81 OVERHOLTZ, WILLIAM J., and ALBERT V. TYLER, "Long-term responses of the demersal fish assemblages of Georges Bank" 507 OW MARK D.-see POLOVINA and OW Pandalus borealis—see Shrimp, northern pink Pandalus montagui—see Shrimp Pandalidae— see Shrimp 721 Panulims q^gnus—see Lobster, rock "Parasites of benthic amphipods: microsporidans of Am- pelisca agassizi (Judd) and some other gammarideans," by Phyllis T. Johnson 497 "Parasites of skipjack tuna, Katsuwonus pelamis: fish- ery implications," by R. J. G. Lester, A. Barnes, and G. Habib 343 Parasite studies amphipods 497 tuna, skipjack 343 PARRISH, R. H., D. L. MALLICOATE, and K. F. MAIS, "Regional variations in the growth and age composition of northern anchovy, Engraulis mordax" 483 PARSONS, GLENN R., "Notes on the life history of the catsharks Scyliorhinus meadi" 695 PAYNE, ROGER S.-see WtJRSIG et al. Penaeus aztecus—see Shrimp, brown PEREZ FARFANTE, ISABEL, "The rock shrimp genus Sicyonia (Crustacea: Decapoda: Penaeoidea) in the eastern Pacific 1 FERRYMAN, WAYNE L.-see AU and FERRYMAN PETERSON, CHARLES H., R BRUCE DUNCAN, HENRY C. SUMMERSON, and BRIAN F BEAL, "Annual band deposition within shells of the hard clam, Mercenaria mercenaria: consistency across habitat near Cape Lookout, North Carolina" 671 Phocoena phocoena—see Porpoises, harbor Phytoplankton vertical structure off southern California 151 PIERCE, FRAN-see STEVENSON and PIERCE Plankton ichthyoplankton off the Oregon coast 611 vertical structure off southern California 151 POLOVINA, JEFFREY J., and MARK D. OW, "An ap- proach to estimating an ecosystem box model" 457 Population studies confidence limits for projections 207 dolphin reactions to survey vessels 187 estimates using juvenile shrimp 677 Porpoises, harbor distribution and movements in Fish Harbour 427 movements and activities 543 "The possible influence of warm core Gulf Stream rings upon shelf water larval fish distribution," by G. R. Flierl and J. S. Wroblewski 313 "A possible link between coho (silver) salmon enhancement 722 and a decline in central California Dungeness crab abun- dance," by David H. Thomas 682 "Probable causes of the rapid growth and high fecundity of walleye, Stizostedion vitreum vitreum, in the mid- Columbia River," by Alec G. Maule and Howard F Horton 701 Psettichthys melanostictus—see Ichthyoplankton Queenfish food habits, migration, and abundance 171 "Radio tracking the movements and activities of harbor por- poises, Phocoena phocoena (L.), in the Bay of Fundy, Canada," by Andrew J. Read and David E. Gaskin 543 "Rates of atresia in the ovary of captive and wild northern anchovy, Engraulis mordax" by J. Roe Hunter and Beverly J. Macewicz 119 "Reaction of dolphins to a survey vessel: effects on census data," by Roger R Hewitt 187 READ, ANDREW J., and DAVID E. GASKIN,' "Radio tracking the movements and activities of harbor porpoises, Phocoena phocoena (L.), in the Bay of F\indy, Canada". . . 543 "Recruitment patterns in young French grunts, Haemulon flavolineatum (family Haemulidae), at St. Croix, Virgin Islands," by W N. McFarland, E. B. Brothers, J. C. Ogden, M. J. Shulman, E. L. Bermingham, and N. M. Kotchian- Prentiss 413 "Regional variations in the growth and composition of northern anchovy, Engraulis mordax" by R. H. Parrish, D. L. Mallicoate, and K. F. Mais 483 REID, F. M. H.-see MULLIN et al. "The relationship between tilefish, Lopholatilus chamae- leonticeps, abundance and sediment composition off Georgia," by Gary D. Grossman, Michael J. Harris, and Joseph E. Hightower 443 Reproductive biology clam, soft-shell 403 dolphins, spotted 657 grunts, French 413 mummichog 467 silverside, Atlantic 331 RICHARDSON, W. JOHN-see WtJRSIG et al. ROBERTS, DALE-see DeMARTINI et al. "The rock shrimp genus Sicyonia (Crustacea: Deca- poda: Panaeoidea) in the eastern Pacific," by Isabel P6rez Farfante 1 Rockfish age determination 103 food habits 531 Rockfish, calico fin erosion 195 Rockfish, canary seasonal changes in fat and gonad volume 299 Rockfish, widow seasonal changes in fat and gonad volume 299 Rockfish, yellowtail seasonal changes in fat and gonad volume 299 ROSENBLUM, SHELLY E., and THOMAS M. NIESEN, "The spawning cycle of soft-shell clam, Mya arenaria, in the San Francisco Bay" 403 Sablefish growth 475 Salmo gairdneri—see TVout, rainbow Salmon, chinook otoliths 81, 91 Salmon, coho predation on Dungeness crab 682 Salmon, silver— see Salmon, coho Scallops, sea abundance 580 fishery damage to American lobsters 575 Sculpin, fluffy life history aspects 645 Scyliorhinus meadi—see Catshark Sea-lion, California entanglement studies 692 "Sea scallop fishing impact on American lobster in the Gulf of St. Lawrence," by G. S. Jamieson and A. Campbell. . . 575 "Seasonal cycles of fat and gonad volume in five species of northern California rockfish," by Patrick J. Guillemot, Ralph J. Larsen, and William H. Lenarz 299 5e6astes— see Rockfish Sebastes entomelas—see Rockfish, widow Sebastes flavidns—see Rockfish, yellowtail Sebastes goodei—see Chilipepper Sebastes paucispinis—see Bocaccio Sebastes pinniger—see Rockfish, canary SEDBERRY, GEORGE R., "Food and feeding of the tom- tate, Haemulon aurolineatum (Pisces, Haemulidae) in the South Atlantic Bight" ■ 461 Seriphiis politiis—see Queenfish "Semilunar reproductive cycles in Fundulus heteroclitus (Pisces: Cyprinodontidae) in an area without lunar tidal cycles," by Anson H. Hines, Kenric E. Osgood, and Joseph J. Miklas 467 Sharks, sandbar food habits in Chincoteague Bay, Virginia 395 Shrimp abundance 223 identification and development 253 length-frequency data 222 life history aspects 219 Pandalidae, Hippolytidae, Crangonidae larvae 253 sex transition 225 Shrimp, brown population estimates using juveniles 677 Shrimp, northern pink Pacific cod diet in Pavlof Bay, Alaska 601 Shrimp, rock description and taxonomy in the eastern Pacific 1 SHULMAN, M. J.-see McFARLAND et al. Silverside, Atlantic patterns in fecundity 331 SINGER, MICHAEL M., "Food habits of juvenile rock- fishes (Sebastes) in a central California kelp forest" .... 531 Size-composition queenfish 172 SMITH, TIM D.-see WAHLEN and SMITH Sole, Dover fin erosion 195 Sole, rex fin erosion 195 "The spawning cycle of soft-shell clam, Mya arenaria in San Francisco Bay," by Shelly E. Rosenblum and Thomas M. Niesen 403 Spawming— see Reproductive biology Spot age, growth and distribution of larvae in North Caro- lina coastal waters 587 "Standing stock of juvenile brown shrimp, Penaeus aztecus, in Tfexas coastal ponds," by Loretta F Sullivan 677 Stenella attenuata—see Dolphins, spotted STEVENSON, DAVID K., and FRAN PIERCE, "Life history characteristics oiPandalus montagui and Dichelo- pandalus leptocerus in Penobscot Bay, Maine" 219 723 STEWART, E. F.-see MULLIN et al. STILLWELL, CHARLES E.-see MEDVED et al. Stizostedion vitreum vitreum—see Walleye Stock identification salmonid "Stomach contents of young sandbar sharks, Carcharhinvs plumbeics, in Chincoteague Bay, Virginia," by Robert J. Medved, Charles E. Still well, and John J. Casey SULLIVAN, LORETTA F, "Standing stock of juvenile brown shrimp, Penaeus aztecus, in Ttexas coastal ponds". . . SUMMERSON, C.-see PETERSON et al. 81 395 677 Tkxonomy shrimp, rock 1 TAYLOR, D. M., R. G. HOOPER, AND G. R ENNIS, "Biological aspects of the spring breeding migration of snow crab Chionecetes opilio, in Bonne Bay, Newfoundland (Canada)" 707 THOMAS, DAVID H., "A possible link between coho (silver) salmon enhancement and a decline in central California Dungeness crab abundance" 682 Tilefish abundance and sediment composition off Georgia .... 443 Tbmtate feeding habits in the South Atlantic Bight 461 TVout, rainbow otoliths 81 "Ulna fishery incidental dolphin mortality 521 "Ulna, skipjack foot habits in the southwestern Atlantic 379 parasite use and fishery implications 343 Ibna, yellowfin related to dolphin habitats in the Pacific 623 TYLER, ALBERT V.-see OVERHOLTZ and TYLER "Undersea topography and comparative distributions of two pelagic cetaceans," by Clifford A. Hui 472 "Using objective criteria and multiple regression models for age determination in fishes," by George W. Boehlert .... 103 "Variability in dimensions of salmonid otolith nuclei: im- plications for stock identification and microstructure inter- pretation," by John D. Neilson, Glen H. Geen, and Brian Chan 81 "VariabOity, trends, and biases in reproductive rates of spot- ted dolphins, Stenella attenuata" by Jay Barlow 657 "Vertical distribution of ichthyoplankton off the Oregon coast in spring and summer months," by George W Boeh- lert, Dena M. Gadomski, and Bruce C. Mundy 611 "Vertical structure of nearshore plankton off southern California: a storm and a larval fish food web," by M. M. Mullin, E. R. Brooks, F. M. H. Reid, J. Napp, and E. F. Stewart 151 VINTER, BEVERLY M.-see MATARESE and VINTER "Visual threshold for schooling in northern anchovy, En- graulis mordax" by John Hunter and Ragan Nicholl .... 235 WAHLEN, BRUCE E., and TIM D. SMITH, "Observer ef- fect on incidental dolphin mortality in the eastern tropical Pacific tuna fishery" 521 Walleye growth and fecundity in the Columbia River 701 WARLEN, STANLEY M., and ALEXANDER J. CHESTER, "Age, growth and distribution of larval spot, Leiostomus xanthurvs, off North Carolina 587 WATSON, ALAN P.-see GASKIN and WATSON WESTPHAL, WILLIAM-see LOVE et al. Whales, bowhead behavior in the Beaufort Sea 357 Whales, Pacific pilot undersea topography and distribution 472 "Within-season differences in growth of larval Atlantic herring, Clupea harengus harengus" by Cynthia Jones . . . 289 WROBLEWSKI, J. S.-see FLIERL and WROBLEWSKI WURSIG, BERND, ELEANOR M. DORSEY, MARK A. FRAKER, ROGER S. PAYNE, and W JOHN RICHARD- SON, "Behavior of bowhead whales, Balaena mysticetus, summering in the Beaufort Sea: a description" 357 YOKLAVICH, MARY M.-see BOEHLERT and YOKLAVICH Zalophiis californianus—see Sea-lion, California Zooplankton vertical structure off southern California . 151 724 NOTICES NOAA Tfechnical Reports NMFS published during first 6 months of 1985 Technical Report NMFS 19. Synopsis of biological data on the spottail pinfish, Diplodus holbrooki (Pisces: Sparidae). By George H. Darcy. January 1985, iv + 1 1 p., 8 figs. 20. Ichthyoplankton of the continental shelf near Kodiak Island, Alaska. By Arthur W. Kendall, Jr., and Jean R. Dunn. January 1985, iii + 89 p., 5 figs., 7 tables. 21. Annotated bibliography on hypoxia and its effects on marine life, with emphasis on the Gulf of Mexico By Maurice L. Renaud. February 1985, iii + 9 p. 22. Congrid eels of the eastern Pacific and key to their leptocephali. By Solomon N. Raju. February 1985, iii + 19 p., 12 figs., 2 tables. 23. Synopsis of biological data on the pinfish, Lagodon rhomboides (Pisces: Sparidae). By George H. Darcy. February 1985, iv + 32 p., 22 figs., 24 tables. 24. Tfemperature conditions on the cold pool 1977-81: A comparison between Southern New England and New York transects. By Steven K. Cook. February 1985, iii + 22 p., 5 figs., 5 tables, 14 app. figs. 25. Parasitology and pathology of marine organisms of the world ocean. By William J. Hargis, Jr. (Editor). March 1985, iv + 135 p. [38 papers.] 26. Synopsis of biological data on the sand perch, Diplectrum formosum (Pisces: Serranidae). By George H. Darcy. March 1985, iv + 21 p., 20 figs., 7 tables. 27. Proceedings of the Eleventh U.S.-Japan Meeting on Aquaculture, Salmon Enhancement, Ibkyo, Japan, October 19-20, 1982. By Carl J. Sindermann (Editor). March 1985, iii -i- 102 p. [15 papers.] 28. Review of geographical stocks of tropical dolphins (Stenella spp. and Delphinus delphis) in the eastern Pacific By William F. Perrin, Michael D. Scott, G. Jay Walker, and Virginia L. Cass. March 1985, iv -i- 28 p., 26 figs., 4 tables. 29. Prevalence, intensity, longevity, and persistence ofAnisakis sp. larvae and Lacistorhynchus tenuis metacestodes in San Francisco striped bass. By Mike Moser, Judy A. Sakanari, Carol A. Reilly, and Jeannette Whipple April 1985, iii -i- 4 p., 6 figs. 30. Synopsis of biological data on the pink shrimp, Pandalus borealis Kr^yer, 1838. By Sandra E. Shumway, Herbert C. Perkins, Daniel F. Schick, and Alden R Stickney May 1985, iv -i- 57 p., 46 figs., 36 tables. Some NOAA publications are available by purchase from the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. ERRATA Fishery Bulletin, Vol. 82, No. 2 Epperly, Sheryan P., and Walter R. Nelson, "Arithmetic versus exponential calcula- tion of mean biomass," p. 446-448. — R + R Page 446, left column, equation should read: Bt = - 2 Bt {e^'-^' - 1) Gt -Z, Page 446, right column, equation should read: Bi = AAl, left column, line 1, correct to read: recruit. In one, Bf was computed arithmetically. Page 447, right column, line 3, correct to read: F-multiples and ages of entry, when B^ was cal- Page 447, Figure 1, second line, correct to read: when dt = 1.0 and Sj = 1.0, DELTA = J?,(5( ^^ - B( ^^^) = B^ * {0.0061 + 0.0037 (G( - Z^) Page 448, paragraph 1, line 8, correct to read: ing a need to minimize the G^ - Zi difference if B^ Page 448, paragraph 1, last line, correct to read: ommend that 5, be calculated exponentially. U S Po«»l Ssrvic* STATEMENT OF OWNERSHIP, MANAGEMENT AND CIRCULATION Required hy 39 U S C J6S5I 1 A. TITLE OF PUBLICATION Fishery Bui let in 3 FREQUENCY OF ISSUE Quarterly IB. PUBLICATION NO. 3A NO OF ISSUES PUBLISHED ANNUALLY 2. DATE OF FILING . su 38 ANNUAL SUBSCRIPTION PRICE S21,Q0 4 COMPLETE MAILING ADDRESS OF KNOWN OFFICE OF PUBLICATION fSlreet. Ctrv. Cnuntv. State and Z1P*4 Codei iSot pnnteni Scientific Publications Office, NMFS 7600 Sand Point Way N.E. BIN C15700 (bldg. 1), Seattle, WA 98II5 5 COMPLETE MAILING ADDRESS OF THE HEADQUARTERS OF GENERAL BUSINESS OFFICES OF THE PUBLISHER iSoi pnnieri National Marine Fisheries Service, NOAA, DOC 2001 Wisconsin Ave., NW , Washington, DC 20235 6. FULL NAMES AND COMPLETE MAILING ADDRESS OF PUBLISHER, EDITOR, AND MANAGING EDITOR (Tha iiem MUST yOT be blank/ PUBLISHER f.Wameand Complete Mailing Address) Jack McCormicK, Scientific Publications Office, NMFS 7600 Sand Point Way N.E. BIN C15700, Seattle, WA 98115 iOVrOH iNarie and Compleu yifiiliniAtiiiress) u^r-c Mary FuKuyama, Scientific Publications Office, NMFS 7600 Sand Point Way N.E. BIN C15700, Seattle, WA 98115 MANAGING EDITOR iName and Complete Mailing Address) Mary Fukuyama, Scientific Publications Office, NMFS 7600 Sand Point Way N.E. BIN C15700, Seattle, WA 98H5 7 OWNER (If owned by a corporation, its name and address must be stated and also immediately thereunder the names and addresses of stockholders owning or holding J percent or more of total amount of stock If not owned by a corporation, the names and addresses of the individual owners must be given If owned by a partnership or other unincorporated firm, its name and address, as well as that of each individual must be given. If the publica- tion IS published by a nonprofit organization, its name and address must be stated.) {Item must be completed.) FULL NAME COMPLETE MAILING ADDRESS department of Commerce ]kth St., NW. Washington. DC 20230 8 KNOWN BONDHOLDERS. MORTGAGEES, AND OTHER SECURITY HOLDERS OWNING OR HOLDING 1 PERCENT OH MORE OF TOTAL AMOUNT OF BONDS. MORTGAGES OR OTHER SSCUR\r\£S (If therr are none, so state) FULL NAME COMPLETE MAILING ADDRESS None 9. FOR COMPLETION BY NONPROFIT ORGANIZATIONS AUTHORIZED TO MAIL AT SPECIAL R/^TSS (Section 42J 12 DMMonly) Th* purpoM, function, and nonprofit status of this organization and the exempt status for Federal income tax purposaa (Check one) E (1) HAS NOT CHANGED DURING PRECEDING 12 MONTHS D (21 HAS CHANGED DURING PRECEDING 12 MONTHS (If changed, publisher must submit explanation of change with this statement i 10. EXTENT AND NATURE OF CIRCULATION (See instructions on reverse side) AVERAGE NO. COPIES EACH ISSUE DURING PRECEDING 12 MONTHS ACTUAL NO. COPIES OF SINGLE ISSUE PUBLISHED NEAREST TO FILING DATE A, TOTAL NO. COPIES (Net Press Run) (handled by U.S. B. PAID AND/OR REQUESTED CIRCULATION Uash DC 1 Sales through dealers and carriers, street vendors anucountet saTeJ 2.2'«I 2,2i4l «f»e^ ZOijOZ) 2 Mail Subscription (Paid and/or requested) C. TOTAL PAID AND/OR REQUESTED CIRCULATION (Sum of I OBI and I0B2) printed for sqles 800 800 0. FREE DISTRIBUTION BY MAIL. CARRIER OR OTHER MEANS SAMPLES, COMPLIMENTARY, AND OTHER FREE COPIES ],kU\ \.hk\ E. TOTAL DISTRIBUTION (Sum of C artd D) Z.Z'tl 2.2'«1 F. COPIES NOT DISTRIBUTED 1. Office use. left over, unaccounted, spoiled after printing 2. Return from News Agents G. TOTAL (Sum of E, FI and 2-should equal net press run shown in A) Z.Z'tl 2.2l4l 11. I certify that the statements made by me above are correct and complete SIGNATURE AND TITLE OF EDITOR, PUBLISHER. BUSINESS MANAGER, OR OWNER PS Form 3526, Julv 1984 (See instruction on revene) ^ 77 INFORMATION FOR CONTRIBUTORS TO THE FISHERY BULLETIN Manuscripts submitted to the Fishery Bulletin will reach print faster if they conform to the following instructions. These are not absolute requirements, of course, but desiderata. CONTENT OF MANUSCRIPT The title page should give only the title of the paper, the author's name, his affiliation, and mailing address, in- cluding ZIP code The abstract should not exceed one double-spaced page In the text, Fishery Bulletin style, for the most part, follows that of the U.S. Government Printing Office Style Manual. Fish names follow the style of the American Fisheries Society Special Publication No. 12, A List of Com- mon and Scientific Names of Fishes from the United States and Canada, Fourth Edition, 1980. Text footnotes should be typed separately from the text. Figures and tables, with their legends and headings, should be self-explanatory, not requiring reference to the text. Their placement should be indicated in the right-hand margin of the manuscript. Preferably figures should be reduced by photography to 5% inches (for single-column figures, allowing for 50% reduction in printing), or to 12 inches (for double-column figures). The maximum height, for either width, is 14 inches. Photographs should be printed on high quality or glossy paper. Do not send original drawings to the Scientific Editor; if they, rather than the photographic reductions, are needed by the printer, the Scientific Publications Office will request them. Each table should start on a separate page Consistency in headings and format is desirable Vertical rules should be avoided, as they make the tables more expensive to print. Footnotes in tables should be numbered sequentially in arable numerals. Tb avoid confusion with powers, they should be placed to the left of numerals. Acknowledgements, if included, are placed at the end of the text. Literature is cited in the text as: Lynn and Reid (1968) or (Lynn and Reid 1968). All papers referred to in the text should be listed alphabetically by the senior author's sur- name under the heading "Literature Cited." Only the author's surname and initials are required in the literature cited. The accuracy of the literature cited is the respon- sibility of the author. Abbreviations of names of periodicals and serials should conform to Serial Sources for the BIOSIS Data Base. {Chemical Abstracts also uses this system, which was developed by the American Standards Association.) Common abbreviations and s3mibols, such as mm, m, g, ml, mg, °C (for Celsius), %, °Iqq, and so forth, should be used. Abbreviate units of measure only when used with numerals. Periods are only rarely used with abbreviations. We prefer that measurements be given in metric units; other equivalent units may be given in parentheses. FORM OF THE MANUSCRIPT The original of the manuscript should be typed, double- spaced, on white bond paper. Please triple space above headings. We would rather receive good duplicated copies of manuscripts than carbon copies. The sequence of the material should be: TITLE PAGE ABSTRACT TEXT LITERATURE CITED TEXT FOOTNOTES APPENDIX TABLES (Each table should be numbered with an arable numeral and heading provided). LIST OF FIGURES (Entire figure legends) FIGURES (Each figure should be numbered with an arable numeral; legends are desired) ADDITIONAL INFORMATION Send the ribbon copy and two duplicated or carbon copies of the manuscript to: Dr. William J. Richards, Scientific Editor Fishery Bulletin Southeast Fisheries Center Miami Laboratory National Marine Fisheries Service, NOAA 75 Virginia Beach Drive Miami, FL 33149-1099 Fifty separates will be supplied to an author free of charge and 50 supplied to his organization. No covers will be supplied. Contents— Continued Notes PETERSON, CHARLES H, R BRUCE DUNCAN, HENRY C. SUMMERSON, and BRIAN F. BE AL. Annual band deposition within siiells of the hard clam, Mercenaria mercenaria: consistency across habitat near Cape Lookout, North Carolina 671 SULLIVAN, LORETTA F, DENNIS A. EMILIANI, and K. NEAL BAXTER. Stand- ing stock of juvenile brown shrimp, Penaeits aztecus, in Tfexas coastal ponds 677 THOMAS, DAVID H. A possible link between coho (silver) salmon enhancement and a decline in central California Dungeness crab abundance 682 FELDKAMP, STEVEN D. The effects of net entanglement on the drag and power out- put of a California sea lion, Zalophus califomianns 692 PARSONS, GLENN R. Notes on the life history of the catshark, Scyliorhinus meadi 695 LIBBY, DAVID A. A comparison of scale and otolith aging methods for the alewife, Alosa pseudoharengus 696 MAULE, ALEC G., and HOWARD F HORTON. Probable causes of the rapid growth and high fecundity of walleye, Stizostedion vitreum vitreum, in the mid-Columbia River 701 TAYLOR, D. M., R. G. HOOPER, and G. R ENNIS. Biological aspects of the spring breeding migration of snow crabs, Chionoecetes opilio, in Bonne Bay, Newfoundland (Canada) 707 CREED, ROBERT P., JR. Feeding, diet, and repeat spawning of blueblack herring, Alosa aestivalis, from the Chowan River, North Carolina 711 Index 717 Notices NOAA Ibchnical Reports NMFS published during first 6 months of 1985. • GPO 593-096 11 MBL WHOI LIBRARY UH n UA F