Errata Page 33: Footnote 16, drift should read drifts . Page 93: Line 11, p. 92 should read p. 96. Page 239: Under the heading "Halithalestris croni (Kroyer) , paragraph 1, line 4, tables, P« 207 should read t ables, p. 29 71 Page 267: Last line before table, p. 298 should read p. 297 . Page 330: Legend for Figure 92, contou r should read contours. " PLANKTON OF THE OFFSHORE WATERS OF THE GULF OF MAINE ::::::: By Henry B. Bigelow From BULLETIN OF THE BUREAU OF FISHERIES, Vol. XL, 1924, Part II Document No. 968 : : : : : : : : : : : : : : : : : : : : : : price 31.2s Sold only by the Superintendent ol Documents. Government Printing Office, Washington, D. C. WASHINGTON GOVERNMENT PRINTING OFFICE FROM THE ARCHIVES OF THE WOODS HOLE OCEANOGRAPHIC INSTITUTION WOODS HOLE. MASSACHUSETTS FROM THE ARCHIVES Of THE WOODS HOLE OCEANOGRAPHIC INST1TUTI6JJ WOODS HOLE. MASSACHUSETTS PLANKTON OF THE OFFSHORE WATERS OF THE GULF OF MAINE By HENRY B. BIGELOW Museum of Comparative Zoology, Harvard University With tables of copepods by C. B. Wilson, and tables of diatoms by Albert Mann CONTENTS Page Introduction 5 The plankton 15 Section 1. — General survey of the animal plankton (zooplankton) 16 Vertical distribution 24 Neritic and oceanic plankton 31 Seasonal fluctuations in the planktonic communities 37 Immigrant planktonic communities 51 Tropical visitors 53 Arctic visitors 59 Other immigrants 62 Migrations of pelagic fish eggs and larva? 69 Quantitative distribution 78 Density of association 90 Annual variations in abundance 96 Plankton as food for whales and fishes. 97 Food of the plankton 106 The more important groups of plank- tonic animals 112 MoUusks 112 Cephalopods 112 Pteropods 116 Limacina retroversa 116 Limacina helicina 125 Clione limacina 125 Other pelagic mollusks 129 Crustaceans 130 Adult decapods 130 Pasiphsa 131 Euphausiids 133 Thysanoessa inermis 135 Thysanoessa longicaudata 139 Thysanoessa gregaria 142 Thysanoessa raschii 143 Nematoscelis megalops 146 Page Section 1. — General survey of the animal plankton (zooplankton) — Continued. The more important groups of plank- tonic animals — Continued. Crustaceans — Continued. Euphausiids — Continued. Euphausia krohnii 146 Meganyctiphanes norvegica 147 Thysanopoda acutifrons 155 Other euphausiids 155 Hyperiid amphipods 156 Euthemisto 156 Other hyperiids 165 Hyperia 165 Hyperoche 165 Parathemisto oblivia 166 Oceanic hyperiids 166 Copepods 167 Acartia clausi 171 Acartia longiremis 177 Acartia tonsa 181 ^Etidius armatus 182 Anomalocera pattersoni 182 Asterocheres beecki 187 Calanus finmarchicus 188 Calanus gracilis 211 Calanus hyperboreus 212 Candacia armata 218 Centropages bradyi 219 Centropages hamatus 220 Centropages typicus 221 Dacty lopusia thisboides 226 Dwightia gracilis 226 Ectinosoma neglectum 227 Eucalanus attenuatus 228 Eucalanus elongatus 228 Euchceta media 230 1 BULLETIN OF THE BUREAU OP FISHERIES Section 1. — General survey of the animal Pa s e plankton (zobplankton) — Continued. The more important groups of plank- tonic animals — Continued. Crustaceans — Continued. Copepods — Continued . Euchceta norvegica 230 Euchseta spinosa 237 Eucheirella rostrata 237 Eurytemora herdmani 238 Gaidius tenuispinis 238 Halithalestris croni 239 Harpactieus littoralis 241 Harpacticus uniremis 242 Heterorhabdus spinif rons 242 Idya furcata 242 Labidocera restiva 243 Lucicutia grandis 244 Metis ignea 244 Mecynocera clausi 245 Metridia longa 245 Metridia lucens 253 Monstrilla serricornis 263 Oithona similis 264 Paracalanus parvus 264 Parathalestris jacksoui 271 Phyllopus bidentatus 27 1 Pleuromamma 272 Pseudocalanus elongatus 275 Rhincalanus cornutus 283 Rhincalanus nasutus 284 Scolecithricella minor 285 Temora longicornis 287 Temora turbinata 293 Tortanus discaudatus 294 Undeuchceta major 295 Undeuchoeta minor 295 Zaus abbreviatus 296 Zaus spinatus 296 Table of copepods in vertical hauls, Mav to October, 1915, by Dr. C. B. Wilson 297 Table of copepods in vertical hauls, February to May, 1920, by Dr. C. B. Wilson 299 Table of copepods in surface hauls, February to May, 1920, by Dr. C. B. Wilson 303 Table of copepods in horizontal hauls, December, 1920, and January and March, 1921, by Dr. C. B. Wilson 304 Supplementary note on the cope- pods, by Dr. C. B. Wilson ... 305 Daphnids (Cladocera) 307 Section 1. — General survey of the animal Page plankton (zooplankton) — Continued. The more important groups of plank- tonic animals — Continued. Worms 308 Glass worms (chretognaths) 308 Sagil ta elegans 308 Sagit ta serratodentata 320 Sagitta maxima 324 Sagit ta lyra 327 Sagit ta hexaptera 328 Eukrohnia hamata 328 Other chaetognaths 334 Tomopterids 334 Tomopteris catharina 334 Tomopteris septentrionalis 340 Pelagic coelenterates 340 Hydroid medusae 341 Melicertum campanula 34 1 Staurophora mertensii 342 Pty chogena lactea 348 Mitrocoma cruciata 348 Phialidium languidum 350 Trachomedusre 352 Aglantha digitale 352 Scyphomedusae 357 Cyanea capillata var. arctica — 357 Aurelia aurita 362 Other scyphomeduste 364 Ctenophores 365 Pleurobrachia pileus 365 Mertensia ovum 371 Bolinopsis inf undibulum 372 Beroe cucumis 372 Other ctenophores 376 Siphonophores 376 Stephanomia cara 377 Diphyes arctica 379 Other siphonophores 379 Pelagic hydroids 379 Section 2. — General survey of the plank- tonic plants (phytoplankton) and unicellular animals 381 Phy toplanktonic communities 383 Quantitative distribution of the phyto- plankton 397 Peridinians 40? Ceratium 407 Other peridinians 416 Diatoms * 17 List of diatoms at representative localities 423 PLANKTON OF THE GULF OF MAINE Section 2. — General survey of the plank- Page tonic plants, etc. — Continued. Diatoms — Continued. Notes on the dominant genera of diatoms 43 1 Asterionella 43 1 Biddulphia . 432 Chsetoceras 433 Coscinodiscus 436 Coscinosira 438 Ditylium 438 Eucampia 440 Guinardia 440 Lauderia 44 1 Nitschia 44 1 Rhizosolenia 442 Skeletonema 448 Thalassiosira 449 3 Section 2. — General survey of the plank- Page tonic plants, etc. — Continued. Diatoms — Continued. Notes on the dominant genera of diatoms — Continued. Thalassiothrix 454 Other diatoms 457 Notes on other unicellular plants and animals 458 Phaeocystis 458 H alosphsera 459 Acantharian radiolarians 460 Tintinnids.. 463 Other unicellular organisms 465 Notes on the biology of the phytoplank- ton 465 Bibliography 487 INTRODUCTION This memoir is the second part of the report on the oceanographic and biologic survey of the Gulf of Maine, the account of the fishes ' forming the first. The vessels of the bureau have carried out the following oceanographic and plankton cruises in the Gulf of Maine since 1912, when the systematic survey was begun : Schooner Gram-pus: July to August, 1912; July to August, 1913; July to August, 1914; May to October, 1915; and July, August, and October-November, 1916. Steamer Albatross: February to May, 1920. Steamer Halcyon: December-January, 1920-21 ; March, 1921 ; and August, 1922. In addition, tows were taken at intervals during the winter of 1912-13 off Gloucester and between Cape Ann and Cape Elizabeth in April and May, 1913. The Fish Hawk also carried out an extensive program of towing in Massachusetts Bay during the winter and spring of 1924-25, but only a few of the catches have been examined. The locations, hydrographic data, and types of nets employed, and the depths of the hauls have been published for all the stations up to May, 1920, in the follow- ing reports : July-August, 1912, stations 10001 to 10046, in Bigelow, 1914, p. 135. November, 1912-May, 1913, stations 10047 to 10056, in Bigelow, 1914a, p. 416. July- August, 1913, stations 10057 to 10061 and 10085 to 10112, in Bigelow, 1915, p. 342. July-August, 1914, stations 10213 to 10264, in Bigelow, 1917, p. 330. May-October, 1915, stations 10266 to 10339, in Bigelow, 1917, p. 331. July-November, 1916, stations 10340 to 10355, 10398, and 10399 to 10404, in Bigelow, 1922, p. 176. February-May, 1920, stations 20044 to 20129, in United States Bureau of Fisheries Document No. 897 (1921). For ready reference the locations of all the tow-net stations for these cruises are given on the accompanying charts (figs. 1 to 6) ; also on figures 7 and 8, the Halcyon tow-net stations of the winter and spring of 1920 and 1921, and of August, 1922, the data for which have not yet been published. As the value of any regional account of the plankton depends largely on the amount of data available, it may be of interest to add that more than 1,000 tows have been made in the Gulf of Maine region since 1912, at various depths from the surface down to the bottom, some with horizontal and others with vertical nets. In a few cases the tows were made with the horizontal closing net (Bigelow, 1913a). The area covered in this report is the same as that covered in the report on the fishes; that is, the oceanic bight from Nantucket on the west to Cape Sable (Nova 1 Fishes of the Gulf of Maine, by Henry B. Bigelow and William W. Welsh. Ft. I, Vol. XL, Bulletin, O. S. Bureau of Fisheries, 1924 (1925), 567 pp., 278 figs. Washington. Bureau of Fisheries Document No. 965. 5 O BULLETIN OF THE BUREAU OF FISHERIES Scotia) on the east. These natural boundaries are continued offshore by Nan- tucket Shoals on the one side and by Browns Bank on the other, which roughly demark the boreal waters of the gulf from the warmer coastal water off southern _y«~" f /sf n / bAtan J yrPy V^ SuJt ""'■ > srf J) *j>*^B®"* > *^ / NOVA /!fi*« 6 «89.' X * 97 /"'" / „ > V) /) ^P"* \SCOTIA M «■ *■ ''ft-/ tS •96-* (Yarmouth v^ /O «I00 \ I. t\^j^\ .-'»IOI Wj mA^ ; rorttandC f w saJ 5 VI CAPE ) P/Z™ / .102 1 V A * L IS \^s' .91 *8» .83 rM f / - u? 86 / / «104 r/-**. X 4T ♦ / < Jr I C «106 i •89 . + + "4- «r U . 80 / s^ ~?~^ .87 N . \ \^__ - .88 ^V j. P 106 X / x/ $ ^*\ i 86 s^Bftsaw- «r « 68 /'"' / B6-. /Sj ./ > - -J \ ♦ \ ~* =« f | <:, «69 i + ♦ *■ / * «r u-H ♦ 4 J •60 / „ ,» ,^x - ...->-'' / ^--- . " "■**. 40' ■*- -I- + 4- * *■ W 7t 70" i il i »9" «r or ««' Fig. 2.— Locations of Grampus stations 10057 to 10060 and 10087 to 10106, July 8 to August 22, 1913, and general location of stations 10047 to 10056 and 10053 to 10056, November 20, 1912, to April 14, 1913 (X) includes the whole of the offshore rim formed by Georges and Browns Banks and the two main deep channels — Eastern and Northern — that pierce it. Brief notes on the plankton collected on the several cruises have already been published (Bigelow, 1914, 1914a, 1915, 1917, and 1922). 8 BULLETIN OF THE BUREAU OF FISHERIES The present report gives a general account of the planktonic communities (animal and plant) of the open waters of the gulf outside the outer headlands (such as must precede the intensive survey of the plankton of any region), with such .269 •260.262 • 261 • 216 • 217 • 218 4- • 219 Fig. 3.— Locations of Grampus stations 10213 to 10263, July 19 to August 28, 1914. Stations where no tows were made are underlined notes on the occurrence of the more important groups and species as a preliminary examination of the large amount of material collected has afforded. The plankton of the many harbors and estuarine situations around the shore line of the gulf, and within 1 to 5 miles of the land generally, is barely touched on. almost all our towing PLANKTON OF THE GULF OF MAINE 9 having been done well out at sea; and when this is studied the communities will no doubt prove quite different from those of the open gulf, with neritic forms domi- nating instead of oceanic, and with larval forms of various parentage playing a far 7V 70" 69' 68' 67' 68* ^* 2 8 1 ) ^^ /<% And f /yf cv/ "A; ? A N / sY/ 301 / x?7 Of * 318 1 Jr r PEN)>B® \S 328. 285*212 ' 383 302.284 306 * 273 +. y «490 «c-~"^»604 .(489 r 7 »(488 IB 1 1 «i 1506 *o * H 43' M + -£* sr \V "491 ■* + ■' 4- \ t- (12- t ft -^K^-f •••■. -■•"'••..- ..-■ s y 41' 4 + + +■ + / + 41" '"'^ y x 40' + + + + J- 4-40' 71" 70" 69' 68* 67" 66' Fig. 7.— Locations of Halcyon stations 10488 to 10503, December 29,1920, to January 9, 1921; station 10504, February 9, 1921; and stations 10505 to 10511, March 4 and 5, 1921 unpublished lists of the plankton taken in towings at frequent intervals at St. Andrews, New Brunswick, from November, 1915, to October, 1916, data repeatedly referred to below. I also owe thanks to Dr. A. G. Huntsman, who has offered many PLANKTON OF THE GULF OF MAINE 13 unpublished notes and much information on conditions in the Bay of Fundy region; to Dr. C. J. Fish, who has contributed a preliminary note on the phytoplankton 71* 70* 42°- CAPE ANN 42' 71° 70* Fig. 8.— Location of Halcyon stations 10631 to 10645, August 22 to 24, 1922 collected by the Fish Hav)Tc in Massachusetts Bay during the winter and spring of 1924 and 1925; to Dr. A. H. Leim; and to Capt. John McFarland for towings taken from his schooner Victor. 14 BULLETIN OF THE BUREAU OF FISHERIES 45' 71* 15' -45' -30' -15' CAPE -42* -45 w Fig. 9.— Location of Fish Hawk stations, 1924-25 PLANKTON OF THE GULF OF MAINE 15 THE PLANKTON Although of rather recent birth as words go, 3 the term "plankton" filled so obvious a need that it is now in general use to cover a whole assemblage of organ- isms, plant and animal, related by their manner of life though they may be far apart in the systematic scale. By it we understand all such forms as float or swim freely in the water, but which, however active, are unable to carry out voluntary horizontal journeys of any extent, though certain of them perform considerable ver- tical migrations under the directive influence of sunlight or of some other physical stimulus. Among the three major faunistic groups into which the inhabitants of the sea may be divided — bottom dwellers, free swimmers, and plankton — the im- portance of the last in the economy of nature was slowest in gaining general appre- ciation. Within the last half century, however, biologists have come to realize both that the number of species of this category is past all counting and that the microscopic pelagic plants are the chief producers — that is, are capable of elaborating simple inorganic compounds into complex organic matter — in the sea. They serve as food supply for many larger marine animals at one stage or another, and thus play a most essential role in the general nutritive scheme of marine life. As it chances, the planktonic plants (producers) as a whole are unicellular and microscopic; the planktonic animals (consumers) are multicellular and comparatively large, so that the oft-employed terms " microplankton " and " macroplankton" are not em- piric, but do classify the plankton roughly as vegetable or animal, more technically as phytoplankton or zooplankton. In the following pages I have attempted to place before the reader a general survey of these two great planktonic divisions as they occur in the Gulf of Maine, followed by more particular accounts of the status of such groups of each as loom large in its pelagic communities at one time or another. Many other groups are also represented in the tow nettings, but time and the assistance available have so far allowed examination of those only that are dominant or numerically important in the Gulf at one time or place or another. Study of the occurrence of buoyant fish eggs is not sufficiently advanced to warrant more than a few preliminary notes here. The present knowledge of the breeding grounds and seasons and of the distribution of the eggs and larva? of Gulf of Maine fishes is summarized by species in the first part of this report (Bigelow and Welsh, 1925). > The term was coined in 1886 by Hensen. 75898—26 2 SECTION 1.— GENERAL SURVEY OF THE ANIMAL PLANKTON (ZOOPLANKTON) Few living zoologists have been as fortunately placed as were we on setting sail on the Grampus from Gloucester on our first ocean ographic cruise in the Gulf of Maine on July 9, 1912, for a veritable mare incognitum lay before us, so far as its floating life was concerned, though the bottom fauna can be described as compara- tively well known. Not but what an extensive list of pelagic crustaceans, ccelenter- ates, and other planktonic animals had been recorded thence, but everything was yet to be learned as to what groups or species would prove predominant in the pelagic fauna; their relative importance in the natural economy of the Gulf; their geographic and bathymetric variations; their seasonal successions, migrations, and annual fluctuations; their temperature affinities, whether arctic, boreal, or tropic; and whether they were oceanic or creatures of the coastal zone. We even had no idea (incredible though it may seem at this place and day) what we should prob- ably catch when we first lowered our tow nets into deeper strata of Massachusetts Bay, for, so far as we could learn, tows had never previously been tried more than a few fathoms below its surface. Nor did we at first realize, when the catch was examined in our floating laboratory, that the little reddish copepods (Calanus) darting to and fro in the glass dish, with a few large Sagittae (S. elegans) and young euphausiids among them, would prove the backbone of the local planktonic fauna. Such, however, has proved to be the case; for station after station, cruise after cruise, year after year, have yielded cumulative evidence that (taken by and large) the calanoid copepods are its predominant members at all seasons, except where deposed from the leading role by the local or temporary swarming of some other and usually larger animal. Our first summer's cruise was enough to show that Calanus finmarchicus (large among copepods but small if judged by more familiar standards) is the most important member of the plankton of the Gulf of Maine, if bulk and numbers both be taken into account, and that it plays much the same role there that it does in North European waters (Bigelow, 1914, p. 99). Calanus, as "red feed" or "cayenne," is well known to the local fishermen, who are quite aware of its importance as food for fishes. 4 Side by side with Calanus we have everywhere found its relative, Pseudocalanus elongatus (p. 275) ; but even where the latter outnumbers the former, as sometimes happens, it adds but little to the bulk of the catch, so tiny is it. We have so constantly found the copepod Metridia lucens (p. 253), the chsetognath, or "glassworm," Sagitta elegans (p. 308), the amphipod genus Euthemisto (p. 156), the euphausiid genera Thysanoessa (several species, p. 133) and Meganyctiphanes (p. 147), the pteropod Limacina retroversa (p. 116), the ctenophore Pleurobrachia pileus (p. 365), and (in deep water) the larger copepod Euchseta (p. 230), associated with Calanus, that all these together may be spoken of as the " Calanus community" (figs. 10 and 11), a community that domi- nates the animal plankton from the Grand Banks on the north to Cape Cod (in winter even to Chesapeake Bay) on the south, and from the coast line, on the one hand, out to the continental slope, on the other. ' See page 188 for a further account of this copepod. 16 Boll. V. S. B. F., 1924. i Doc. 968.1 Fig. 10.— Calanus community, chiefly Calanus finmarchicus, with C hyptrhoretts, Euchzta vorvegica, Sagitta elegant, Tomopteris, Tbysanoessa, and Aglantha, Western side of basin, March 24, 1920, haul from 20X) meters (station 200S7). X 1.5 Fig, m. Calanus community, chiefly Calanus tinman incus, with Sagitta elegant, larval euphausiids, and larval witch floundei (Glyptoeephalus cynoglossus), M issachusctts Bay, Juls 19, 1916, haul from 30 n meters (station 20340). X 3.5 PLANKTON OF THE GULF OF MAINE 17 Although copepods usually dominate, the other boreal animals just mentioned are so nearly universal in the Gulf in summer that the planktonic community is then surprisingly uniform qualitatively, with the list of prevalent species varying hardly at all from station to station over its inner parts, as is illustrated by the two fol- lowing tables of catches made north of the Cape Cod- Cape Sable line during the summers of 1913 and 1914, seasons that may serve as representative because the plankton of the upper water layers was of the same general type during the sum- mers of 1912, 1915, and 1916, as I have pointed out eleswhere (Bigelow, 1917 and 1922). Occurrence of representative species in the Gulf of Maine, August, 1913 Species Calanus fiDmarchicus Pseudocalanus elongatus Metridia lucens Anomalocera pattersoni Eucbjeta norvegica Meganyctiphanes norvegica.. Thysanoessa inerrais ' Euthemisto compressa Euthemisto bispinosa Hyperocne kroyeri Limacina retroversa Tomopteris catharina Sagitta elcgans _. Phialidium languidum Pleurobrachia pileus Stations X X X X Per cent lot* a stations for each species 100 80 80 80 70 40 90 50 40 80 60 100 80 50 ' Data [or Th. inermis are not available for 1913; it can, however, be assumed to occur in at least 80 per cent of the cases, since i t was taken at 14 of our 18 midsummer stations in 1914. Occurrence of representative species north of Georges and Browns Banks, July and August, 1914 July August Species M o © 8 o © $ o s © © o SO © X X X X X X X X X X X X X X o X X X X X "x" "x" X X X X X o en o o & o o © o <0 © Calanus finmarchicus X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X "x" X X X X X X X X Metridia lucens Anomalocera pattersoni X Euchfpta norvegica. X X X X X X X X X X X X X X X X X X X X X "x" x X X X X X X X X X X X X x" X X X X X V Meganyctiphanes norvegica.. "x" X X X X X X X "x" X X X X X X X X X X X x X X X X X X .... X x Thysanoessa longicaudata Euthemisto compressa.. X X X "x" X X X X X X X X X X X x Euthemisto bispinosa Limacina retroversa.. X X X Sagitta elegans.. X X X X X X X X X X X x Sagitta serratodentata __ Notwithstanding the quahtative uniformity of the animal plankton of the waters of the Gulf of Maine in summer, the actual aspect of the catches of the tow nets often differs markedly from station to station, according to the relative abundance of their several components and especially of the copepods. As a rule these (chiefly Calanus, Pseudocalanus, and Metridia, with Euchseta in the deepest layers of 18 BULLETIN OF THE BUREAU OF FISHERIES water) are the dominant factor, and it occasionally happens that they practically monopolize the water locally. Such, for instance, was the case in the Eastern Basin on August 13, 1914 (station 10249), when the net from 50 meters captured only 3 or 4 Sagittse, 2 pteropods (Limacina), 3 or 4 larval rosefish (Sebastes), a few small medusas (Phialidium) , 51 euphausiid shrimps, and an odd Euchseta, among millions of Calanus (3 to 4 liters, by measure; no other copepods were detected in sample examined by Doctor Esterly). Near Mount Desert Rock, too, on the same day (station 10248), a cursory examination of about 3 quarts of copepods, among which Calanus, Metridia, and Euchseta were represented in the proportion of about 30, 5, and 2, revealed only a few Pseudocalanus, 21 Thysanoessa longicaudata, odd amphipods (Euthemisto), 24 Meganyctiphanes, 7 Thysanoessa inermis, 6 or 8 ptero- pods (Limacina), 1 worm (Tomopteris), a few Sagittse, 1 Pleurobrachia, and frag- ments of the ctenophore Beroe. Similarly, the only other animals detected in a preliminary examination of the 2 to 3 quarts of copepods 5 captured in the 60-0 meter haul on the eastern part of Georges Bank, on July 23 of that same year (station 10224), were 89 euphausiid shrimps (Thysanoessa inermis), a few amphipods (Euthemisto), half a dozen young fish, and one caprellid, the latter being an accidental straggler from the bottom. The most notable shoal of Calanus we have encountered was off Cape Cod on July 22, 1916 (station 10344), where a 15-minute haul with a net 1 meter in diameter captured 6 quarts at 40-0 meters, together with many thousands of silver-hake larvae (Merluccius), but nothing else except a few small Sagitta elegans, an odd pteropod (Limacina), and an occasional larval crab and euphausiid, though the deeper waters, as exemplified by a haul at 90-0 meters, supported comparatively few copepods but many Sagittse. We have found Calanus (with its relatives, Pseudocalanus and Metridia) hardly less dominant at enough other localities 6 to prove that it is a common event for these copepods to monopolize the plankton of any part of the Gulf in summer. As a rule, however, the animal plankton is more diversified at all levels by the hyperiid amphipods, euphausiids of several species, pteropods (Limacina), Sagittse, etc., men- tioned above, even though copepods may dominate the planktonic community as a whole (figs. 10, 11, and 12). Some of these other groups may be a major element in the plankton locally. For instance, the chsetognaths (Sagitta elegans) often rival the copepods in bulk (if not in actual numbers) at the mouth of Massachusetts Bay and in the Isles of Shoals regions; indeed, our second towing station, 12 miles or so off Cape Ann (10002), yielded a swarm of these arrow worms on July 10, 1912 (Bigelow, 1914, p. 100), and we have encountered similar swarms of Sagittse at other localities since then (fig. 13). An abundance of the large pelagic shrimps Meganyctiphanes (fig. 14) and Thy- sanoessa is regularly characteristic of the deep northeastern corner of the Gulf throughout the year and of the Eastport-St. Andrews region in summer (p. 134), while various larval forms (crustaceans, especially) are extremely numerous locally near shore in their appropriate seasons, as noted elsewhere (p. 31). As other instances of the swarming of one characteristic boreal animal or another we may add that the • Sample examined by Doctor Esterly was nearly pure Calanus finmarchkus. • Notably off Gloucester on Aug. 9, 1913 (station 10087); in the Western Basin on July 15, 1912 (station 10007); near Platts Bank on Aug. 10, 1913 (station 10089); ofl the slope of German Bank on Aug, 12, 1913 (station 10095); northeast of Mount Desert Rock on Aug. 13, 1913 (station 10100); and ofl Cape Elizabeth on Aug. 15, 1913 (station 10104). Bull. CJ. S. B. !•'.. 1924. i Doc. 968.) Fig 1-' —A mnnntonous plankton of Calanus finmarchicut, Massachusetts Bay, July 19, 1916, haul from 30-0 meters (station 10341). In connection with Figure 13 it illustrates a striking example "f vertical stratification, with Calanus dominating the shoaler and Saaitta elegam the deeper levels. X 1.75 FIG 13 Plankton dominated by Sagitla degam, Massachusetts Hay. July 19, 1916, haul fr SO meti 1 1. >i i 10341) In connection with Figure 12 it illustrates local abundance "f tins chaetognatb at the deepei levels at a station where the plankton at shoaler levels was almost pure Calanus. X 1.75 Bull. U. S. B. F., 1924. (Doc. 968.) Fiu. 14. Plankton dominated by the pelagic shrimp Meganyclipltanes norvegica and by the glassworrus Eukrohnia hamata and Sagitta rfrgans, with Calanus and other eopepods. Northeast part of basin, March 23, 1920 (stalion'2(iosi), haul from 140-0 meters. X 1.25 Fig. 15.— Plankton dominated bv juvenile amphipods (Euthemisto). Southern slope of Georges Bank. July 21 ,1914 (station 10219), surface haul. X 9 Bull. U. S. B. I'.. L924. (Doc. 968.) 7 I r Vv <2S Fig Hi Plankton dominated by adull amphipods (Euthemisto) and by Calanwfinmarchicus. Southwestern edge of Georges Bank, July 24, 1916, haul from 160-0 meters (station 10351). X about 1 Fig 17. — Pteropods (iimocina retroversa) and Co/onus n»- morcMcus, northwest part of Georges Bank, July 20, 1914, haul from 50-0 meters (station 10215). X about 2.5 Bull. U. S. B. F.. 1924. (Doc. 968.) Fig. ik.— Plankton dominated by the ctenophore Phurobrachia pileus, with barnacle (Balanus) larva; in the "nauplius" stage. Browns Bank, April 16, i»20. haul from 40-0 meters (station 20106). X 1.5 Fig. 19.— An unusually rich catch of haddock eggs, with the glassworm Sngitta elegant, the pteropod Limacina relromsn, Calanus, and other copepods. Eastern pari of Georges Bank, April 17, 1920, haul at the surface (station 20111). X4 PLANKTON OF THE GULF OF MAINE 19 surface waters were alive "with young amphipods (Euthemisto) as well as with young stages of Calanus finmarchicus, in the proportion of about one of the former to four of the latter" (fig. 15), off Penobscot Bay and off Mount Desert Island on August 11, 1913 (Bigelow, 1915, p. 274, stations 10091 and 10092); that older Euthe- misto (fig. 16) were plentiful (though not rivaling the copepods) off Cape Ann and in the western basin on August 31, 1915 (stations 10306 and 10307), and at several sta- tions along the outer edge of the offshore banks (p. 156); that the pteropod Limacina retroversa (fig. 17), which, as a rule, is but sparsely represented in our tow nettings, swarmed off Penobscot Bay on August 11 and 14, 1913 (stations 10091 and 10101); that fragments of a siphonophore (Stephanomia) formed fully half the catch of the 40-meter haul off Cape Cod on July 8 of that same year (station 10058); and that the ctenophore Pleurobrachia pileus often fills the water to the exclusion of almost every- thing else in the neighborhood of German Bank (fig. 18). In summer and early autumn the large medusEe Cyanea, Aurelia, and Stauro- phora often gather in vast numbers in narrow lanes or windrows, though usually for brief periods (p. 362), and at this same season the hydroid medusa Phialidium lan- guidum is often so abundant on the surface that it fills the tow net to the brim (p. 350). Young fish, too, sometimes occur in numbers sufficient to loom large in the total catch, notable instances of which have been the swarming of young silver hake off Cape Cod, mentioned above (p. 18); likewise of young rosefish (Sebastes) near Cape Elizabeth on July 19, 1912 (station 10019), when several hundreds were taken (Bigelow, 1914, p. 101), off Massachusetts Bay on August, 9, 1913 (station 10087), and near Cashes Ledge, September 1, 1915 (station 10308). Occasionally we have encountered notable quantities of fish eggs, particularly of squirrel hake ( Urophycis chuss), in Ipswich Bay, July 16, 1912 (station 10008); of silver hake (Merluccius) near Monhegan Island and off Mount Desert, on August 4 and 18, 1915 (stations 10303 and 10305); of dinners (Tautogolabrus) at many localities along shore in sum- mer, especially in Massachusetts Bay 7 (station 10340-10343); and of haddock over their spawning grounds on Georges Bank during the early spring (fig. 19). In summer, generally speaking, copepods are relatively most abundant in the western side of the gulf, less so in the eastern, the result being that, in spite of the qualitative uniformity of the tow nettings from station to station, their general aspect is usually most monotonous off the coasts of Massachusetts and southern Maine and out thence to the western basin, and most diversified in the central parts of the gulf and in its deep eastern trough. The only notable exception to the mid- summer dominance of calanoids anywhere in the open gulf north of its offshore banks (local swarmings of other animals, such as those just mentioned, seldom rival the copepods in actual abundance, whether measured by bulk or by numbers) is the Pleurobrachia swarm of the German Bank region, which I have already described in the several preliminary reports on our cruises (Bigelow, 1914, 1915, and 1917). Since we have found this ctenophore in abundance at that same general locality dur- ing the successive Augusts of 1912, 1913, and 1914, and again on September 2, 1915, this is evidently a regular phenomenon of summer. Having occasion to recur to it in a later chapter (p. 365) , I need add here only that Pleurobrachia, large and small, 7 The ledges oft Cohasset are a very productive nursery for this fish, judging from the quantities of its eggs that are to be found there. 20 BULLETIN OF THE BUREAU OF FISHERIES were so abundant on these occasions that every haul yielded quarts of them, and that they fish through the water so thoroughly with their trailing tentacles that a great scarcity of all smaller pelagic animals regularly characterizes this part of the gulf in summer. In fact, a more striking contrast would be far to seek than between the masses of these glassy sea marbles, which have filled our nets there, and the abundant crustacean plankton of the deeper basin a few miles to the westward. Although spring, not midsummer, is the chief season of reproduction in the Gulf of Maine (p. 41), certain of the planktonic groups of animals breed in sufficient numbers there in July or August for their larvoe to loom large in the summer plankton. This is true of the euphausiids, for we have found their larval stages common in Provincetown Harbor on July 20, 1916 (station 10343); on the surface off northern Cape Cod, August 28, 1914, in company with large Calanus (station 10264; Bigelow, 1917, p. 2S3). Young euphausiids were also abundantly represented in the hori- zontal haul at 40-0 meters on August 31, 1915 (station 10306), but so closely re- stricted to the upper stratum that a haul from 110-0 meters brought back very few among a half liter or so of calanoid copepods. Euthemisto is likewise produced in great numbers well within the gulf in August — witness rich hauls of the newly- hatched larvce off Penobscot Bay on August 11, 1913 (station 10092), and in the western basin two summers later (p. 160). Copepods, too, breed throughout the sum- mer, as noted below (p. 46) , and in sufficient numbers for their young stages to char- acterize the plankton locally. Most of the medusas spawn during the late summer or early autumn (pp. 358, 364). We may also point out, what is discussed at some length below, that larva? of coastwise origin and of the most diverse natures are likewise produced during the warm season , though few of them color the aspect of the plankton more than a few miles out from the land (p. 32) . In a later section the seasonal plankton cycle is discussed in some detail (p. 37) ; however, it may clarify the account to note here that very little change takes place in the general composition of the Calanus community during the period (July to August) covered by our midsummer cruises, except for the disappearance of the earlier and the appearance of the later maturing species of medusa? (p. 46). For example, the only notable change during the interval between hauls made at the same location off Cape Cod on July 8 (station 10057) and again on August 5 (station 10086) in 1913 was that Staurophora, Stephanomia, and Beroe, which were prominent in the tow on the first occasion, were no longer to be found on the second, the lists be- ing practically identical otherwise. 8 Three years later we found Calanus and its companion copepods as overwhelmingly predominant in the upper 40 meters or so off Cape Cod on August 29 (station 10398), among such boreal animals as Pleuro- brachia, Aglantha, Sagitta elegans, Euthemisto compressa, and larval euphausiids, as we had five weeks previous (station 10344, July 22) in the corresponding stratum of water a few miles to the south. One very notable event does take place during the summer, however; that is, the entrance of Sagitta serratodentata into the gulf and its westward dispersal there, which are described in a later chapter (p. 322) . The foregoing remarks have reference chiefly to the inner waters of the gulf — ■ that is, north of the offshore banks that form its southern rim — but the same ele- ments unite to form the general planktonic assemblage over all but the outermost 8 A typical Calanus community with Sagitta elegans, Euthemisto, a few euphausiids, and Limacina. PLANKTON OF THE GULF OF MAINE 21 slope of the latter. Thus, a typical Oalanus community, with Clione, Limacina, and the other boreal forms characteristic of the inner parts of the Gulf, occupied the waters over Nantucket Shoals on July 14, 190S (Bigelow, 1909, p. 201), and at the same time of year in 1913, when we found no decided change in the boreal character of the plankton (Calanus predominating) until we had sailed westward nearly to New York (Bigelow, 1915, p. 269). During the summer of 1914 we again found Cal- anus, with its usual companions, predominant over the greater part of Georges Bank in July, and across the mid-zone of the continental shelf abreast of Marthas Vine- yard in August; also in August, 1915; and from Cape Cod out to the continental slope in July, 1916. But although Calanus is as universal over the offshore banks as within the gulf, it does not dominate the plankton so constantly there. Thus we found Sagitta elegans as important, faunally, as were the copepods over the central part of Georges Bank during our summer cruise of 1914, and swarming both over the northeast corner of the bank on July 23 (station 10224 9 ) and in the Northern Chan- nel on July 25 (station 10229), practically to the exclusion of everything else, except for an abundance of adult Euthemisto, which (we may suppose) are sufficiently large and active to protect themselves from the glassworms, voracious though the latter are (p. 107). Even when copepods, as a group, are the chief factor in the summer plankton over Georges Bank, it is sometimes the little brown Temora longicornis (fig. 20), not Calanus, that is the dominant species there. This was the case at a station on the northwestern part of the bank in July, 1913 (station 10059), while the frequency with which Kendall, in his field notes for August, 1896, describes "small brown copepods" (which could only be Temora) as abundant, side by side with "red feed" (Calanus) and "green copepods" (Anomalocera), or even as constituting the bulk of the surface tow, suggests that such dominance on its part is a common event on the northern part of the bank (lat. 41° 45' to 42°, long. 66° 30' to 68° 30'). His records suggest that Temora increases in number there with the advance of the summer, 10 which parallels its seasonal history in the Massachusetts Bay region (p. 289) . Hyperiid amphipods (two species of the genus Euthemisto, p. 156) have often been reported as plentiful over the outer part of the continental shelf off Marthas Vineyard. We found them in abundance over the corresponding zone off Nantucket Shoals and over the western end of Georges Bank, side by side with the copepods, in July of 1913 and 1916 and August of 1913 and 1914. They are equally charac- teristic of the outer parts of the banks eastward across the mouth of the Gulf of Maine and off the Nova Scotian coast, where they breed in abundance (p. 160) and grow larger than within the gulf to the north. The outer part of the continental shelf is the offshore limit to the occurrence of copepods in abundance abreast of the Gulf of Maine; but the pelagic amphipod genus just mentioned is perhaps most plentiful along the upper part of the continental slope, where it mingles with the oceanic planktonic community of the warmer waters of the Atlantic basin. It has likewise been our experience (though fresh observations may give cause to alter conclusions drawn from a single summer's cruise) that in mid- • The catch o( one-half hour's haul of the Helgoland net at 40-0 meters was about 5 liters of Sagitta cltgam, and very little else except a few Calanus, Temora, Pseudocalanus, 3 or 4 Euthemisto, 2 Limacina, young crabs and other decapods, and some floating hydroid fragments described below (p. 380). 10 Kendall's tows were taken during the last week in August. 22 BULLETIN OF THE BUKEAU OF FISHERIES summer Euthemisto is to be expected in abundance over Browns Bank, largely replacing the copepods there, for on July 24, 1914 (station 10228), the surface waters were alive with them, while on June 24, 1915 (station 10296), the tows on the bank yielded large numbers of these amphipods among the still more abundant Calanus (more abundant in bulk as well as in numbers). Euthemisto is also an important factor in the plankton close in to the land off Cape Sable, where they increased in relative abundance in 1914 from July 25 (station 10230), when they were overshad- owed by Calanus, until August 11 (station 10243), when they were dominant in the plankton. A seasonal change of the same sort took place in the shoal coastal waters off Shelburne, Nova Scotia, during the summer of 1915; for Euthemisto dominated a very scanty plankton there on September 6 (station 10313), where it had been out- bulked both by copepods and by Sagittse on June 23 (station 10291), though domi- nating the plankton farther out over the shelf on that day (10293). Although euphausiid shrimps of one species or another (p. 133) are practically universal within the gulf — may, indeed, be constantly plentiful locally, as off the Eastport-Grand Manan region, and temporarily so elsewhere (p. 133) — we have never found them dominating the water of the gulf at any level except over Browns Bank, where the tow net working at 60 meters depth yielded a quart or more of these pelagic shrimps 11 on July 24, 1914 (station 10228), diversified only by an occasional Sagitta, three Bero'e cucumis, a few copepods, and no amphipods at all, notwithstanding the abundance of the latter at the surface at this same station. Though not strictly within the limits of the gulf, I may add that four days later euphausiids occurred in great numbers over the slope abreast of Cape Sable 12 (station 10233), and in this same general region on March 19, 1920 (station 20076, fig. 21). It is not safe to assume, however, that these shrimps are constantly abundant over Browns Bank in summer, for we found none at all there on our only other visit during the warm half of the year (June 24, 1915, station 10296), but in their stead made a very rich haul of calanoids (3 to 4 liters bulk), with a few Euchasta, many large Euthemisto, small Sagittae, and occasional tropical organisms, such as Phronima and Salpa zonaria. To close this brief survey of the chief planktonic communities of midsummer, I must remark that a sprinkling of Gulf Stream animals — sometimes, indeed, a typi- cally tropical plankton — is to be expected all along the upper part of the continental slope at that season, corresponding to the high temperature of the Gulf Stream, the inner edge of which lies but a few miles farther offshore. This tropical plankton and such members of the general bathypelagic community of the Atlantic basin as approach the slope are the subject of a later section (p. 53) . The accompanying photographs (figs. 10 to 21), illustrate certain of the more characteristic communities as they occur in nature, and the distribution of the more characteristic communities, for July-August, 1914, is outlined on the chart (fig. 22). The great majority of the species of pelagic animals that unite to form the bulk of the zooplankton of the gulf are endemic in origin, breeding sufficiently regularly and abundantly within its limits to maintain the local stock by local pro- duction. This generalization, which the reader will find discussed in more detail under the accounts of several of the species concerned, applies to most of the com- 11 Chiefly Afeganyctiphanes norvegica, Thysanassa inermis, Th. longicaudata, with fewer Th. gregaria and Nematoscelis megalops. " Chiefly Euphausia and Nematoscelis and fewer Th. longicaudata at 100 meters; Nematoscelis at 400 meters. Bull. V. S. B. F., 1924. (Doe. 968.) Fig. -'» Plankton dominated by the small brown copepod Temora longicornis, with a few of the larger Calunvji Unmarehicus, juvenile Euthemisto. and glassuorms i Smiilln (Injuns). Western part, at Georges Hank. Julv '.I. I'.iKi, haul from 2.5-0 fathoms (station 1005a). X !t FlQ. 21. -Plankton dominated by .the pelagicTshrimp Thysanoesm'.ongkaudata, with Call marehiaa, glass worms (Sag'itta elegant), and the naked pteropod Cliom limacina. Outer pan of continental shelf oil shelhurne, Nova Scotia, March 19, 1920, haul from m letei ti a 20076). X1.75 PLANKTON OF THE GULF OF MAINE 23 mon copepods, notably to Calanus finmarchicus, Pseudocalanus elongatus, Metridia lucens, Euchfeta, and to sundry others (see the chapter on copepods, p. 167); like- wise to Sagitta elegans (p. 308), both the local species of Euthemisto (E. compressa Fig. 22. — Distribution of the more characteristic types of animal plankton in the odshore waters of the Gulf of Maine, July and August, 1914. O, calanoid copepods dominant; #, glass worms (Sagittfe) dominant; X, amphipods (Euthe- misto) dominant; A, euphausiid shrimps dominant; A, cteuophores (Pleurobrachia) dominant; Q, hydromedusse (Phialidium) dominant; P, swarm of pteropods {Limacina retrovasa) and E. bispinosa, p. 156), the euphausiid shrimps Meganyctiphanes and probably Thysancessa inermis (p. 139), and the pteropod Limacina retroversa (p. 124), to men- tion only a few. It also applies to a whole category of animals of coastwise nativity 24 BULLETIN OF THE BUREAU OF FISHEBIES It does not follow from this, however, that all parts of the gulf are equally favorable as marine nurseries. On the contrary, few if any animals breed indifferently or equally plentifully over its whole area, and different parts of the gulf may run the whole gamut from extreme productivity to almost complete sterility for one species or another. Our work has not progressed far enough to give more than a ghmpse of such local differences; enough, however, has been done to show that the south- western corner of the gulf generally, and the Massachusetts Bay region in particular, stand at one extreme, with innumerable copepods and a great abundance of pelagic fish eggs produced there (not to mention other planktonic animals), while certain small areas in the Bay of Fundy exemplify the other, where few if any animals with floating eggs breed successfully. Broadly speaking, our hauls have demonstrated that the coastal belt, out to the 100 or 150 meter contour, is more prolific than the deep trough in the production of planktonic animals. VERTICAL DISTRIBUTION OF THE ZOOPLANKTON In the foregoing lines the various planktonic communities are treated as though their several component groups or species were indifferently distributed from the surface downward, independent of depth; the various lists, that is, are such as would be yielded by vertical hauls from surface to bottom at the respective stations. Such is by no means a true picture, however, for it often happens that, although the species from any given locality occur side by side geographically, they may be far apart bathymetrically, and especially so in the deeper parts of the gulf. Nor is it astonishing, with a pelagic fauna as varied as that of the Gulf of Maine, and with its sundry members responding variously in their vertical occurrence to the physical conditions under which they live, that we have usually found the plankton of mid- summer more or less stratified even in the upper 100 meters or so, either by the concentration of one group of animals at one level, another group at another, or by a comparatively barren state of the immediate surface contrasted with great pro- ductivity in the underlying strata of water. The stratification between depths less than 100 meters, on the one hand, and the bottom waters of the guff, on the other, is still more significant, being one of kind as well as of degree, as I shall endeavor to make clear later (p. 26). Indeed, it would not be too much to say that the local zooplankton is never quite uniform from the surface downward to any considerable depth, unless it be in very shallow water or in localities where vertical circulation keeps the whole column effectively stirred from top to bottom. With so many subjects involved, stratification, whether quantitative or quali- tative, may occur in infinite variety, and many instances of the sort have forced themselves on our notice, though our hauls have not been particularly directed toward the detection of such. Perhaps the most interesting phase of the subject, as it is certainly the most widespread, is the scarcity of adult pelagic animals of the Calanus community, including most of the species which together make up the preceding plankton lists (p. 17), at the surface during the daylight hours of summer. No matter what nets we have used on the surface between sunrise and sunset in the offshore waters of the gulf at this season, they have usually yielded very little zooplankton of any kind, and often practically nothing except larval PLANKTON OF THE GULF OF MAINE 25 forms and the smallest Crustacea and phytoplankton. In fact, had we relied on surface hauls by daylight alone, we would hardly have suspected the existence of the abundant and varied planktonic fauna which peoples its deeper water layers. True, we have occasionally made rich catches of Calanus, with its companion animals, right on the surface in the middle of the day, as, for example, near Gloucester on July 22, 1912 (station 10012), near Lurcher Shoal on August 12, and off Penobscot Bay and Cape Elizabeth on August 14, 1914 (stations 10245, 10250, and 10251), and near Seguin Island on August 4, 1915 (station 10.303) 13 ; while the extraordinary abundance of Calanus that characterized the 40-1 00 meter stratum in the western side of the gulf during late July, 1916 (p. 18), was reflected in the presence of consid- erable numbers of these little crustaceans on the surface at the time, by day as well as by night. However, such occurrences have been exceptional. Huntsman, similarly, has characterized " the presence of Calanus en masse at the surface between 3 and 4 p. m., under a bright sun," in the Bay of Fundy in September as an unusual event (Willey, 1919, p. 181). On the other hand, surface tows made in the gulf during the hours of darkness, especially if near midnight, have usually yielded an abundance of the calanoid copepods (even including the deep-water genus Euchasta). And the geographic locations of the stations where we have made rich surface catches by night point to a general diurnal migration of the Calanus community — upward after dark, downward about daylight — in the inner parts of the Gulf of Maine in summer, such as Esterly (1911 and 1912) and Michael (1911) describe for the San Diego region, 14 and with all the major planktonic groups sharing in it more or less, though perhaps none so regularly as the copepods. The data bearing on this point are not extensive, no particular attention having been paid to it in arranging the stations. We have occasionally found the surface practically barren some hours after sunset and before the first sign of sunrise, even at localities where the deeper waters supported a rich and varied plankton, as was the case in the western basin on August 9, 1913 (station 10088), and again on the 22d of that month a year later (station 10254). Of course, there is nothing novel in a vertical migration of this kind, similar phenomena having long been known and widely heralded in other seas; nor is it necessary to seek far afield to find a parallel in New England waters, for Peck (1S96) long ago described the copepods as deserting the surface of Buzzards Bay almost completely during the daytime, to reappear there after dusk. It is unfortunate that our hauls have not been arranged to show at what precise time after sunset the copepods rise to the surface in largest number or how soon after midnight they sink again, a question of great interest from the physiological standpoint (p. 204). Few data have been gathered as to the actual vertical range through which this migration takes place in the Gulf of Maine; that is, how far up and down any individual animal may swim, or how universally or regularly the members of any group of animals indulge in it. It must be very widespread occasionally, at least among the copepods, for at times we have towed them in great numbers right 11 In an earlier report (Bigelow, 1914a) it was stated by error that a large haul of Calanus was obtained on the surface by day at station 10027; actually this station was occupied at about midnight. 14 Data on the euphausiids, amphipods, pteropods, etc., will be found summarized in the accounts of these several groups. 26 BULLETIN OF THE BUEEAU OF FISHERIES on top of the water after dark, notably near Mount Desert Rock on August 16, 1912 (station 10032), where the 4-foot net, towed for half an hour, yielded nearly 3 liters of plankton, chiefly copepods, with Calanus finmarchicus dominating, besides Euchseta, Centropages typicus, Metridia, Anomalocera, and Pseudocalanus ; also the shrimps Meganyctiphanes, Thysanassa inermis, Th. longicaudata, Th. gregaria, and Nematoscelis ; the pteropods Limacina and Clione; Euthemisto of both species; the two common chsetognaths Sagitta elegans and S. serratodentata; Tomopteris; Stephanomia; and larval redfish in lesser number; in short, a typical Calanus com- munity. A second instance of this sort came to our notice off southern Cape Cod on July 22, 1916 (station 10346), when the surface net yielded about as much Calanus (nearly a liter) , with a sprinkling of Pseudocalanus and Metridia, an odd Euthemisto, Sagitta elegans, and Clione, as did the 30-meter net, although the mouth area of the latter was four times the greater, and it was towed for an equal period. As a rule, however, this vertical migration does not bring nearly so large a proportion of the zooplankton to the top of the water at any time during the night, for our catches have almost always been far richer (more varied, as well) at some little depth than im- mediately on the surface. This is illustrated by a station off Cape Cod on August 23, 1914 (station 10256), where the catch of Calanus, Euchasta, Meganyctiphanes, Euthemisto, S. elegans, and Stephanomia was several times larger in the 130-0 meter haid than in the surface haul, even after allowing for the use of nets of different diameters. Whatever the precise physiological stimulus may be which causes so many of the copepods and other pelagic animals to rise at sunset and to sink again soon after midnight — and this is still an open question (p. 204) — its results are certainly confined to a far shoaler stratum in the Gulf of Maine, where it is never necessary to lower the net deeper than 40-100 meters to find the Calanus community at full strength at any time of day, than in the San Diego region off southern California, where Calanus in particular congregates as deep as 200 fathoms by day, to swim upward nearly or quite to the surface in the darkening hours (Esterly, 1911). Nor is it probable that the daily vertical migration in the Gulf of Maine often covers more than 100 fathoms even for Euchasta, which sinks considerably deeper in the daytime than does Calanus but less often reaches the surface at night. Until more extensive data are avadable it is idle to do more than touch on this interesting question. Apart from these vertical diurnal migrations our hauls have afforded glimpses of vertical stratifications of three other sorts (sometimes all three of them are exem- plified at a given station) : (1) As between young and adult communities as a whole: (2) between the adults of the several groups, genera, or species, even within the rather narrow depth limits (say, 10 to 100 meters) where the Calanus community as a whole attains its most abundant development; and (3) between the planktonic communities of the upper 100 meters or so, on the one hand, and of the deepest water of the gulf, on the other. Perhaps as illustrative a case as any that has come under our notice, and one typical of the western side of the gulf as a whole in early summer, is afforded by a station off Cape Cod on July 8, 1913 (station 10057), where it was the surface haids alone that yielded any considerable number of copepod nauplii and eggs; the haul at 15-0 fathoms (27-0 meters) caught swarms of Calanus and many PLANKTON OF THE GULF OF MAINE 27 euphausiids and hyperiids, but only a few Sagittae; the haul from 60-odd meters contained almost no euphausiids, hyperiids, or pteropods, but yielded large numbers of Sagittae, and Euchosta was taken in it alone. Thus, the Calanus, euphausiids, and pteropods were mostly above 30-50 meters, the Euchaeta and Sagittae below that depth, with Beroe, Pleurobrachia, and Stephanomia more evenly distributed (Bigelow, 1915, p. 267). A similar bathymetric segregation as between the copepods and the large adult Sagittae prevailed in Massachusetts Bay on July 19, 1916 (station 10341; figs. 12 and 13), when the haul at 30 meters yielded a practically pure Calanus plankton with many larval fishes and some young euphausiids but very few Sagittae, whereas the net working at SO meters captured a swarm of large S. elegans but not nearly so many Calanus as the shoaler haul. This condition must have been general over a considerable area at the time, for we had much the same experience two days later off Cape Cod (station 10344), where Calanus and young silver hake were extraordinarily abundant at 40 meters (the largest catch of young fishes we have ever made — Bigelow and Welsh, 1925, p. 394), but evidently concentrated in a narrow depth zone centering at about that level, for both were practically absent on the surface, on the one hand, and very much less numerous in the 90-0 meter catch, on the other, whereas Sagittas, equally absent from the surface, were scarce in the 40-meter hauls but abundant in the catch from 90 meters. A depth relationship of the same sort (between copepods and euphausiids) obtained on August 9, 1913, off Cape Ann (station 100S7), where the 30-0 meter haul brought back a rich gathering of the former (chiefly Calanus, with many Pseudocalanus) and manj- larval rosefish, but only an occasional euphausiid, whereas we captured a con- siderable number of the latter (small Thysancessa) at 80-0 meters, but only a fraction as many copepods as at 30 meters, and an occasional Sebastes. On the other hand, lest the reader conclude that the Sagittae and the euphausiids invariably congregate below the densest shoals of copepods when stratification occurs between these groups, I may point out that we found the 40-0 meter haul on the northwest slope of Georges Bank, July 20, 1914 (station 10215), practically monopolized by S. elegans and Limacina retroversa, with very few copepods, whereas a rather rich haul from 70-0 meters brought in about as great a bulk of copepods (about equal numbers of Calanus and Pseudocalanus) as Sagittae, but no Limacina at all. Similarly, there were about sLx times as many Calanus and Pseudocalanus at 110-0 meters as at 40-0 meters off Cape Ann on August 31, 1915 (station 10306), with just the reverse holding in these same hauls for Euthemisto and for young euphausiids. The latter, indeed, were almost wholly confined to the shoaler level, where they about equaled the copepods in bulk if not in numbers. The copepod plankton of the western basin must also have been much denser below than above 100 meters on May 5, 1915 (station 10267), when the vertical haul from 250-0 meters yielded great numbers, whereas the catch of the horizontal net working at 85 meters and up to the surface was scanty (total catch less than % liter). As still another instance of vertical stratification in summer, I may mention our station of August 12, 1914, on German Bank (10244), where the surface water con- tained an abundance of small Euthemisto but only a few Calanus (besides the Pleuro- 28 BULLETIN OF THE BUREAU OF FISHERIES brachia so common there, p. 19), whereas the haul from 40 meters yielded copepods chiefly, with only occasional Euthemisto. No doubt a more intensive examination of the zooplankton of the Gulf of Maine will multiply such instances indefinitely, but enough have been mentioned to show that a definite vertical segregation may occur at certain times and places between animals having the same faunal status. On other occasions the contents of hauls at different depth levels, between, say, 10 and 100 meters, are often almost precisely alike, as was the case near Lurcher Shoal on August 15, 1912 (station 10031), when copepods, euphausiids, Sagittse, Staurophora, Euthemisto, and even Salpse (p. 56) occurred in proportions so similar in hauls from 50-0 and from 100-0 meters that it would have been difficult to distinguish samples of the one catch from the other had it not been for the presence of the large copepod Euchseta in the deeper one. Many other instances of this same sort might be mentioned also. Our experience has been that young and larval forms of all sorts, from fish eggs to copepod nauplii, are usually most plentiful at or very near the surface. For example, in May, 1920, which is the season of their greatest abundance, nauplii were far more abundant in the surface catch and in closing-net hauls from 10-15 meters in Massa- chusetts Bay (stations 20120, 20121, and 20124) and off the Merrimac River (station 20122) than in the deeper catches. It is safe to say that the great majority of the copepods breeding in the Gulf of Maine pass through their early stages in the upper 40 meters of water. Similarly, the nauplius and cyprid larva? of the common barnacle, so prominent in the plankton for a brief period in spring (p. 43), are usually condensed at and near the surface, rarely at some lower level (station 20105, figs. 23 and 24). Larval and even half -grown euphausiids are also far more plentiful above than below 50 meters; and this is even more true of larval amphipods (Euthemisto), which live close to the surface at first (p. 163), to sink to deeper levels with advancing age; likewise of young S. elegans, as described elsewhere (p. 316). Since most of the fish produced in the gulf live in this same zone during their first weeks, it may, not inaptly, be named the nursery of the gulf. Certain conspicuous adult animals are also as typically characteristic of the sur- face of the gulf as are the innumerable, larval forms. Such, for instance, is the large blue copepod Anomalocera which may often be seen darting to and fro in the sun- light immediately in the surface film and which seldom sinks more than a few fathoms. The small brown copepod Temora longicornis likewise occurs in greatest numbers near the surface; for instance, a surface tow near Nantucket Lightship, on July 9, 1913 (station 10060), "yielded thousands, while the haul from 20 fathoms caught only 25 specimens, and it was not taken at all in hauls from depths greater than 23 fathoms" during that summer (Bigelow, 1915, p. 294). Much the same rule holds for the little copepod Centropages typicus, of which "the surface haul at station 10088 yielded ten times as many specimens as the haul from 80 fathoms, though made with a net of only one-sixth the mouth area" (Bigelow, 1915, p. 293), and which we twice found common at the surface during August, 1914, but not at all in the catches at 25 meters and deeper (Bigelow, 1917, p. 291). It is our surface hauls, too, that most often yield Evadne and appendicularians; indeed, we question whether the latter ever sinks to any great depth in the Gulf of Maine. One of the Bull. C. S. H. I'.. 1924. (Doc. 968.) Fia. 23. Plankton at the surface, northern channel, April 15, l«20 (station 20105), dominated bj the copepods Calanus.finmarchicus, Acartia, and Metridia. In connection with Figure 24 it illustrates vertical stratification of the plankton. X 10 w ^ Fio. 14 Plankton of thi deeper levels, :.i the same station as the surface haul m Figure 23 This deep haul (150 meters) was dominated bj the "naupliu " lar\ f barnacles (Balanus), with fewer Calanus ihi othei i opepods < lo Bull. U. S. B. F., 1924. (Doc. 968.) Fh,. 25. -Surface plankton dominated by Oikopleura at a st ition on the continsnl il slop' south wesl of Georges Hank, May 17, L920 (station 20129), where the deepsr water (fig. 26) was dominated by euphausiids and copepods. X 3.5 Fig. 26. -Plankton in the 100-0 meter haul, dominated by Calanus fimnaTchicus and other copepods, and by euphausiid shrimps (Thysanoessa) al a station ofl the southwest slope of Georges Bank, May 17, 1920 (station 20129), where the surface catch (fig. 25) was dominated by Oikopleura. X 4 PLANKTON OF THE GULF OF MAINE 29 most striking instances of vertically stratified plankton we have ever encountered resulted from a swarming of large appendicularians (fig. 25) on the surface and down perhaps to 40 or 50 meters over the southern edge of Georges Bank on May 17, 1920 (station 20129), overlying a moderately abundant Calanus and young euphau- siid community in the deeper strata down to about 100 meters (fig. 26). Various medusa?, among them the largest (Aurelia and Cyanea), likewise seek the surface even in bright sunlight, while smaller species, notably the common hydroid medusa Phialidium languidum., sometimes swarm there in such numbers as to fill our tow nets to the brim. In fact, the latter seldom, if ever, sinks more than a few meters deep. Ctenophores, too, of several species, come up to the top on smooth days, where they can be seen drifting along like crystal balls (p. 372), and on occasion even the large euphausiid shrimps may swarm on top of the water, day as well as night, probably to avail themselves of a particularly succulent food supply; in the Eastport region, for instance, in summer (p. 147), and in the Isles of Shoals- Boon Island region in spring (p. 145), though they are no more characteristic of the superficial layers elsewhere and at other seasons than are the adult Sagittse. Since most of the deep-water members of the plankton (e. g., Euchaeta, the largest of local copepods, and the chaBtognath Eukrohnia hamata) have occasionally been taken on the surface in the Gulf of Maine (pp. 235, 328), any number of this faunal group may be expected to appear at that level occasionally. It needed very few hauls from the deep trough of the gulf to show that there is a decided cleavage in composition between the zooplankton of the upper and of the lower water layers, with the 100 to 150 meter level roughly delimiting the two. No hard and fast line can be drawn between these communities, for the gap is bridged, on the one hand, by such occasional excursions of the deep-water dwellers upward even to the surface as have just been mentioned and, on the other, by the presence of Calanus, Metridia, Thysanoessa inermis, Tomopteris, Sagitta elegans, Euthemisto, Limacina, etc., in decreasing numbers right down to the bottom, even in the deepest parts of the gulf, a fact demonstrated by the closing-net hauls listed below (p. 50). Nevertheless, the two communities are so characteristic in general aspect that it is usually possible to tell at a glance whether any particular sample came from much above or far below 100 meters. The features making this possible are the abundance and regular occurrence of Euchseta norvegica in the deep basin of the gulf. This copepod is so much larger than any of its relatives and is made so conspicuous by the blue egg clusters of the female that it gives a distinctive appear- ance to the entire catch. It is regularly accompanied by the chaBtognath genus Eukrohnia (p. 328) ; more rarely by the larger glass worm S. lyra (p. 327) ; fre- quently by the large pelagic decapodous shrimp Pasiphasa; and locally by large numbers of the euphausiid shrimp MeganyctipTianes norvegica (the latter, however, occurring in shallow water also). On the other hand, this "Euchseta" community includes only a sparse representation of Euthemisto, Calanus, or Pseudocalanus, and practically no Pleurobrachia or pteropods. Unfortunately we have made only one successful closing-net haul deeper than 100 meters during all our summer cruises, for it was not until the spring of 1920 that our closing apparatus for horizontal hauls was developed to a dependable state; 30 BULLETIN OF THE BUREAU OF FISHERIES hence, except for that one instance, the catches in the deep summer hauls have all been contaminated by the Calanus community captured by the open nets on their journeys up and down. For this reason I can not claim that the Euchasta, Eukrok- nia, etc., taken at any given station necessarily came from the deepest levels. But the Euchasta community has been consistently represented in our midsummer hauls below 100 meters, no matter in what part of the basin of the gulf these have been made (see the following tables, pp. 40 and 50), and as we have never found it in any abundance in hauls shoaler than 100 meters it would be merely academic to dispute the general thesis that it is actually characteristic of the deepest stratum of the Gulf of Maine. Whether the occasional excursions of Eukrohnia and Euchseta to the surface, such as I have just mentioned (p. 29) and discuss at greater length elsewhere (pp. 235, 328), are sporadic events induced bysome temporarily or locally active vertical circulation, or whether they are mqre regular concomitants of regularly recurrent physical states than now appears probable, the fact remains that it is only below 100 meters — that is, in the saltest water of the trough of the gulf, which is never very cold — that the Euchseta community occurs regularly. 15 The Euchasta com- munity similarly characterizes the corresponding level along the continental slope abreast of the gulf. The use of the closing net is requisite to show in what relative amounts these deep- water animals are mingled with Calanus and its companions in the deeper strata of the inner parts of the gulf. In one such haul just mentioned (off Cape Cod, August 29, 1912, station 10043) at a station where Calanus outnumbered Euchseta at least 2,000 to 1 in the 20-0 meter haul (Bigelow, 1914, p. 116), these two copepods were about equally numerous at 125 to 120 meters, with Euchasta bulking the larger, thanks to its great size. The total volume of the catch was small, however (less than one-half liter), and we have never found the deep-water Euchseta community even approaching the swarms of Calanus of the upper 100 meters, or so, in volume of plankton present in the water. Unfortunately we lack precise data on this point. To recapitulate, three chief bathymetric pelagic communities of animals can be distinguished in the Gulf of Maine in summer, not, of course, sharply outlined, but still sufficiently so to be recognizable. First is that of the surface, with its juveniles, small copepods, etc., which receives accessions of large copepods, Sagittse, euphausiids, etc., by night and rarely by day; second, the general boreal community of the upper and mid depths, with Calanus, Metridia, and Pseudocalanus, Euthemisto, Thysa- noessa, and Sagitta elegans as its index species; third, the Euchseta community of the deepest waters of the gulf. The distinctions between these communities, and espe- cially between the last two, are greatest when and where the water is most stratified in density and temperature — that is, in the southwestern part of the gulf in mid- summer — least when and where the water is most uniform vertically. This is the case in all parts of the gulf during late winter and early spring; and throughout the year in regions of very active vertical circulation, such as the neighborhood of Eastport, the St. Andrews region at the mouth of the Bay of Fundy, and locally on the offshore banks. 11 See p. 236 (or precise temperatures and salinities. PLANKTON OF THE GULF OF MAINE 31 To answer a question that has often been asked me by zoologists as well as laymen, I may remark that there is no level in the Gulf of Maine but supports a varied pelagic fauna. NERITIC AND OCEANIC PLANKTON None of the criteria by which the plankton can be subdivided ecologically (e. g., relation to temperature, season of reproduction, depth of habitat, etc.) is more fundamental than whether its members do or do not depend on the coast line with its shallows and great supply of foodstuffs; that is, whether they are neritic or oceanic. This distinction is as interesting to the oceanographer as to the biologist, a know- ledge of the mutual distribution of the two groups on the high seas often going far to reveal the mutual relationships and fluctuations of waters of coastal and of offshore origin. The pelagic larva? of various familiar bottom-dwelling animals (a host in them- selves), including most of the worms, bivalve and gastropod mollusks, decapod crustaceans, barnacles, starfishes, and sea-urchins, so abundant in the bays and shallow waters along the coasts of the Gulf of Maine, belong to the neritic category. The adults of many medusas, including the largest and most conspicuous species as well as others minute, are equally neritic, for they pass through a fixed stage in shallow waters during early life. Here, also, fall certain small phyllopod crustaceans (e. g., Evadne), which, though pelagic for most of their lives, survive unfavorable seasons in the form of resting spores on the bottom, a life history analogous to that of many diatoms, which consequently fall in the neritic category also, as do various other pelagic plants less prominent in the plankton. There is also a whole series of planktonic animals, particularly among the copepods, bound to the neighborhood of the coast by some unknown bond (perhaps by dependence on a particular food supply), and hence to be classed as neritic, although they are pelagic throughout life both as larvas and as adults. Here, too, must be classed the pelagic eggs of all the species of fish that spawn in shallow water, such as cod, haddock, pollock, silver hake, cunners, and flounders of sundry species. Contrasted with this coastwise population of the open sea are all the oceanic animals and plants, which are not only free floating or swimming throughout life but show no apparent relation to the coast line in their distribution — to borrow a nautical term, they form its "blue water" population. It is, of course, impossible to draw a hard and fast distinction between the neritic and oceanic categories, the border line being bridged in too many instances by the many pelagic forms occurring indifferently both near shore and out at sea, and also by animals that are dependent on the bottom in deep water at some stage of existence but not in shallow water; for example, by the hydromedusan genus Calycopsis, wluch probably passes through a fixed stage but has never been found nearer shore than the continental slope. However, the division holds fairly well for the Gulf of Maine. In northern seas, generally, neritic elements form a large part, if not practically the whole, of the plankton of sheltered bays and estuaries and off river mouths — 75898—26 3 32 BULLETIN OF THE BUREAU OF FISHERIES indeed, in all locations where conditions may be described as estuarine — and dominate for a mile or two out from the coast line generally. No detailed study of the plank- ton of any such situation tributary to the Gulf of Maine has yet appeared, but Willey's (1913 and 1915) and McMurrich's (1917) observations at St. Andrews, with the lists contributed by Doctor McMurrich (p. 12) and the record that might be collected from many sources of the abundance of various medusae and of larval forms of many kinds inshore, show that the gulf is no exception to the general rule. The complexion of the plankton at Woods Hole recently described by Fish (1925) may serve as an indication of the preponderance of neritic forms that may be expected in the Gulf of Maine bays and harbors and close along its coast line generally. Thus, Fish classifies 42 of the characteristic diatoms as neritic and only 16 as oceanic, while at least 13 out of 15 hydromedusae described by him as " occurring commonly in surface towings" (Fish, 1925, fig. 26) are characteristic of the neritic group and only one oceanic. Two neritic scyphomedusae occur in abundance. Only two of the many annelids listed from his tows (Sagitta and Tomopteris) are truly pelagic when adult, for the others swim only during the breeding season or as larvae. Molluscan larvae are at times abundant in the Woods Hole plankton. The neritic phyllopods Evadne and Podon are characteristic of the local tows, as are the larvae and sometimes the adults of neritic mysids. Fish found barnacle larvae abundant in their season, bottom-dwelling amphipods were taken in large numbers in the tow during their breeding season, and the larvae of decapod Crustacea — shrimps, prawns, crabs, and hermit crabs — are dominant. On the other hand, no euphausiid is a permanent member of the local plankton, though several species have been recorded at Woods Hole. Thus, aside from the copepods, the oceanic element of the Woods Hole plankton is wholly overshadowed by the neritic. If one were to turn to the Gulf of Maine de novo, one might naturally expect the plankton of its central portion to be so largely recruited from the coastal zone that neritic elements would loom large there also, judging from the form, length, and complexity of the shore line with the abundant and varied bottom fauna which it supports; from the confinement of the gulf by the extensive and shallow offshore banks on the ocean side; from the great volume of river water that pours into it; and from the fact that the tides are strong enough in places to stir the water thoroughly. Our first summer's cruise (in 1912) was enough to show that this is not the case but that the pelagic communities of the gulf a few miles out to sea are predominantly oceanic, except over the offshore banks. Our subsequent cruises have corroborated this for summer, autumn, and winter for all the years of record, and for the whole offshore basin of the gulf, where we have never found neritic forms, plant or animal, playing a role of any importance in the plankton except for a brief period in spring, as pointed out below. The rarity of animals of coastwise origin or affinity in the open gulf in summer (except within a trivial distance of land and over the shallow banks) will appear from the following facts of distribution, already summarized in an earlier report (Bigelow, 1917, p. 251). PLANKTON OF THE GULF OF MAINE 33 The most conspicuous planktonic inhabitants of the gulf, of neritic nature, are the two large scyphomedusan genera Aurelia (p. 362) and Cyanea (p. 357). Their value as indices of coast water has long been appreciated in north European seas, and they are both so large that they are usually visible as they float on or near the surface, if present in any numbers; consequently, notes on their local presence or absence, as seen from the vessel, afford a closer record of their distribution than do the actual captures of specimens at the tow-net stations. Both of these medusae are abundant along the shores of the gulf in summer, but Aurelia is so closely con- fined to the immediate vicinity of the land that we have seldom seen it more than a mile or two outside the 100-meter contour (or more than 15 miles from land), while the zone within which it occurs regularly, if not abundantly, extends hardly 10 miles seaward beyond the outer headlands and islands (p. 363) ; nor have we found it on Georges Bank, though the shallowness of the water there suggests this as a possible breeding ground for it. Cyanea, the common "red jellyfish," which often grows to a breadth of 3 feet across the disk and sometimes to a tremendous size (A. Agassiz, 1865), is not so closely confined to the immediate vicinity of the land as is Aurelia, for it occurs regularly in the coastal zone, on Nantucket Shoals, and on Georges Bank, which must be important centers of production for it, judging from the abundance of the young medusae there in spring and summer (p. 359) . However, it is a rare occurrence to find a Cyanea outside the 100-meter contour in the Gulf of Maine (on July 15, 1912, we captured a very large Cyanea in a haul from 120-0 meters in the western basin). The hydromedusa Melicertum campanula, 10 so abun- dant all along the coasts of the Gulf of Maine (p. 341), is an even more precise neritic indicator than Aurelia, for it is still more closely confined to the coastal zone, not because the waters of the open sea are fatal to it (its abundance in Massachusetts Bay proves the contrary), but because it passes through its fixed stage only in sheltered localities, estuaries, etc., and because its free-floating (medusa) stage is of shorter duration. Although Melicertum often swarms in localities as open to the ocean as Massachusetts Bay and the outer parts of Penobscot Bay, as well as in more inclosed waters, a single example from the western basin (August, 1913, station 10088) is our only record of it more than 15 miles from land. The medusae of the genus Sarsia, which are plentiful in season (p. 43) in bays and estuarine situations all along the shallow coastal zone of the gulf, where they are detached from their hydroids in great numbers in spring, are similarly restricted to the coast line, for we have never taken them in the offshore parts of the gulf and rarely more than 4 or 5 miles from land. This is equally true of many other small hydroid medusae, most of which appear in the gulf for a brief period only, and then far more numerously close to shore than outside the outer islands. As I have pointed out elsewhere (Bigelow, 1917, p. 252), an interesting example of neritic occurrence among Coelenterates is afforded by the hydroid colonies we have found floating in considerable numbers over Nantucket Shoals and Georges Bank in July of 1913, 1914, and 1916, and in February, 1920, as well (p. 379). These are so closely confined to the immediate vicinity of the localities where they are torn from the bottom that we have never found them or their free medusae (which some- times swarm on the banks) anywhere in the deeps of the gulf to the north. >• Large catches of Melicertum 38 miles off Cape Co I and near Browns Bank on August 12 and 19, 19211, prove that it drift farther offshore. 34 BULLETIN OF THE BUREAU OF FISHERIES There are other species of hydroid medusas that are not so closely confined to shoal water, probably because they are able to pass through their fixed stage at greater depths and consequently at a greater distance from land. Staurophora and Phialidium, for example, bear much the same relationship to the 100-meter contour in their distribution (p. 345) as Aurelia, Melicertum, and other forms more dependent on shoal water bear to the immediate coast line. Other typical examples of the neritic habit are afforded by the larva? of various decapods among the pelagic Crustacea, young crabs, in particular, being instructive because so conspicuous and so easily recognized in the tow. These (provisionally identified as the common rock crab, Cancer a?n. dominating the plankton; X, occurred] Mar. 1, 1920, station 20050, 75-0 meters Apr. 9, 1920, station 20090, 60-0 meters May 4, 1920, station 20120, 40-0 meters July 9, 1916, station 10341, and 80-0 meters Oct. 31, 1916, station 10399, 60-0 meters Feb. 13, 1913, station 10053, 20-0 meters X X X X X X X X X X X X X X X X X X X X X X X D X D D X D D D X X X X X X X X X X D X X X X D D X X X D D X X X X X X X X X X X X X X X X PLANKTON OF THE GULF OF MAINE 39 The most striking event in the seasonal cycle of the zooplankton of the Gulf of Maine (if a negative one) is that a very decided decrease, amounting on occasion almost to complete disappearance of the pelagic fauna, takes place early in spring over the whole area of the gulf, coincident with the tremendous vernal flowering of diatoms (p. 3S5), an event the precise date of which varies locally and from year to year. The quantitative aspect of this change is discussed elsewhere (p. 82), but it also exerts an adventitious influence on the qualitative composition of the plankton, for with all its members sharing in the impoverishment, the rare as well as the com- mon, the less abundant forms practically disappear and the scanty catches become extremely monotonous. We first observed this impoverishment in Massachusetts Bay during the late winter and early spring of 1913, when the zooplankton fell to so low an ebb, quantitatively, as the water began to warm from its winter minimum, that the total volume of the catch of a net about 1.2 meters in diameter, towed for half an hour at 40-0 meters on March 4, was only about 15 cubic centimeters. In this catch an occasional Pseudo- calanus elongatus, 12 Sagitta elegans, 9 Tomopteris catharina, an odd Euthemisto, and some haddock eggs were the only variants detected among the Calanus finrnar- chicus, of which the general mass consisted. On April 3, following, the net yielded only a few dozen copepods,one Euthemisto, and two Clione, with a few unrecognizable siphonophore bells and Balanus nauplii; while the catch of planktonic animals made on April 14 was no more varied (a few Calanus, one Tomopteris, one 5. elegans, one Beroe, one young Staurophora, and a few Balanus nauplii), whereas the water was thick with diatoms on both these occasions. Subsequent experience during the spring of 1920 has shown that this vernal impoverishment of the zooplankton, which takes place to a greater or less degree in the upper strata of water over the entire area of the gulf, is especially characteris- tic of the coastal belt aDd of Georges Bank, where it culminates in March. It in- volves no qualitative alteration in the plankton, however, for the spring community, sparse though it be near land, is of essentially the same type as the more abundant pelagic population of midsummer, with the same groups and species (notably Calanus finrnarchicus) predominant. Practically all the common oceanic animals of mid- summer except Sagitta serratodentata, which is a seasonal immigrant (p. 320), may be found represented in late winter and spring, if a sufficient mass of plankton be ex- amined from any given locality in the gulf, though many are so rare then that the net is more apt to miss than to catch them. Winter adds few extralimital visitors to the local pelagic fauna, never (in our experience) enough to give a distinctive aspect to the plankton. The essential qualitative unity between the zooplankton of summer and that of spring may be illustrated by the horizontal hauls off Cape Elizabeth on March 4, 1920 (station 20059), which yielded Calanus finrnarchicus (dominant), Sagitta elegans, Thysanoessainermis, Th. raschii, haddock and plaice eggs, Pleurobracbia, and Tomop- teris catharina, although the water was then so barren that the vertical net caught nothing at all (p. 82). The typical boreal fauna was still more fully represented on the same day off Penobscot Baj^ (station 20057), although the plankton was hardly denser there numerically, viz, by C. finrnarchicus (dominant), Pseudocalanus, 40 BULLETIN OF THE BUREAU OF FISHERIES Euchreta, Sagitta elegans, Eukrohnia, Euthemisto of both species, Clione, Limacina retroversa, Tomopteris, Meganyctiphanes, Thysanoessa inermis, and Th. longicaudata. This is a list that might be expected in summer or autumn, and the same was true of the hauls made in Massachusetts Bay during the winter of 1912-1913, mentioned above (p. 39). The plankton is as uniform, qualitatively, from season to season in the deeper parts of the gulf as the following table shows for a location in the western basin about 30 miles off Cape Ann. Zooplankton in the western basin, various months [D, dominant; X, occurred] Febru- ary, station 20049 March April, station 20115 May. station 10267 June, St ; 1 1 ion 102U9 July, station 10007 August Decem- ber, station 10490 Station 20087 Station 10510 Station 10088 Station 10254 Station 10307 Calanus flnmarchicus.. D D D X D X D X X X D D D D D D Calanus hyperboreus X Pseudocalanus elongatus X X X X X D X X X X X (?) X X X X X Metridia lucens __ X Metridia longa Euchseta norvegica. X D X D X X X X D X X X X Anomalocera pattersoni Centropages typicus.. Pasiphtea X X X X X X X X X X X X X X X X X X Meganyctiphanes norvegica X X X X Thysanoessa inermis Thysanoessa longicaudata Thysanoessa gregaria. Euthemisto compressa X X X X X X X Euthemisto bispinosa Limacina retroversa X X X X X Clione limacina X X X X X X X D Sagitta elegans X X X X X X X X X X Sagitta serratodentata Sagitta lyra X X X X X X X X X X X X Eukrohnia bamata X Tomopteris catharina. X X X Aglantha digitale X X X X X X Beroe cucumis X Stephanomia X X X Pbialidium languidum Broadly speaking, our March hauls have paralleled those made in midsummer in the relative importance of the several groups of animals in different parts of the gulf, as well as in the qualitative composition of the catches. Thus, Pleurobrachia was dominant on German Bank both on March 23 and on April 16, 1920 (stations 20085 and 20103), just as it usually is in summer and autumn, and its area of abun- dance extended from abreast of Yarmouth, on the north, to the shoals off Cape Sable, to the south, on both these visits. On both these spring visits there was a second center of abundance for Pleurobrachia on Browns Bank, where our June and July tows have yielded only an occasional specimen; but although the area of abundance for Pleurobrachia in this general region was more extensive in March and April, 1920, than we have found it in summer, these ctenophores were less plentiful in actual number; nor had they so thoroughly exterminated the other smaller animals, for we found the German Bank-Cape Sable swarm accompanied by copepods in fair numbers on the April visit, besides barnacle (Balanus) nauplii (in abundance), Sagitta elegans, euphausiids, Euthemisto, and Tomopteris. Bull. U. S. B. !•'., 1924. (Doc. 968.) '£ %% Pin 27— Surface catch illustrating abundance of larval copepods in the "nauplius" stage, in Massachusetl ' ' Hay in May oration 20121, May 4, 19211). X9 Fra. 23. T'a ■ -.hi ■, more highly magnified. < about 100 Bull. U. S. B. F., 1924. (Doc. 968.) Fig. 29. — Plankton dominated by half-grown Calanus finmarchkus, Massachusetts Bay, -May 4, [915 (station 1026(1), vertical haul from 125-0 meters. X 9 Km. :!0.— Plankton dominated hy largo Culnniis finmarchicui oft' Cape Cod, July 22, 191fi, haul from 40-0 meters (station 103441. This sample is from the most productive catch of Calanus vet made in the Gulf of Maine. X 9 PLANKTON OF THE GULF OF MAINE 41 Similarly, the spring cruise of 1920 suggests that S. elegans may be expected to rival the copepods in abundance over a large part of Georges Bank in February, March, and April, just as it does in July; for it was a large element in the catch at a station on the southwest part of the bank on February 22 (station 20046), on the northeast part on April 17, and had been so plentiful at a third station on the eastern part of the bank on March 11 (station 20066) that the "glass worms," with a great abundance of haddock eggs, dominated the catch (fig. 19). In short, Georges Bank is apparently a center of abundance for S. elegans throughout the year (p. 310), and the presence of a shoal of large Limacina retroversa on the northern part of the bank on March 11, 1920 (station 20065), reproduced our experience of July 20, 1914, though the exact localities in question were about 80 miles apart. Late in the winter and early in the spring the scanty zooplankton of the gulf is chiefly composed of fully adult animals, a fact made evident by the predominantly large size of its calanoid copepods and Sagittae, giving the catches a distinctive aspect when compared with those of July or August. The recrudescence which charac- terizes the advance of spring results primarily from the local propagation of its several component groups, not of replenishment by immigrants from any extra- limital source. This has been proved by repeated observations. In Massachusetts Bay this vernal augmentation is earliest apparent at stations close in to the land, in the shape of a sudden appearance of hosts of copepod nauplii (figs. 27 and 28). This event commences some time late in March off the mouth of Boston Harbor, for we found few nauplii there on the 5th of that month in 1920 (station 20062), but an abundance of them on the 5th of April (station 20089), besides many copepods in the older larval stages. As the season advances this vernal wave of reproduction on the part of the copepods spreads seaward; and the nauplii appeared in multitudes at the mouth of the bay during the last half of April, 1920, where we had found only an occasional copepod — egg, nauplius, or juvenile — on March 1 or April 9. In 1920 the swarms of larval copepods, together with the various other larvas that appear about the same time, produced a decided increase in the volume of animal plankton present in the water of the Massachusetts Bay region by the first week in May. This was our experience in 1913, also, when W. W. Welsh found the water in Gloucester Harbor reddened for areas of about a square yard, several yards apart, with what proved to be swarms of copepod nauplii and young copepods on May 3. The peak of production of copepods, however, is so soon passed in Massachusetts Bay that our nets brought back proportionally more of the older juveniles and fewer nauplii off Gloucester on May 16, 1920, than 12 days earlier, while the hauls off Magnolia, Mass., on May 17, 1913, yielded only a few copepod nauplii but an abundance of the later stages (chiefly Calanus, with some Eurytemora), besides many crab larvce in the zcea stage. The vernal replenishment of the zooplankton follows much the same course in the coastal belt immediately north of Cape Ann as in Massachusetts Bay, with a few copepod nauplii among the swarming diatoms off the mouth of the Merrimac River as early as March 4 in 1920 (station 20060). The nauplii were again noted there on April 9, and on May 7 hauls made close by with the closing net yielded 42 BULLETIN OF THE BUREAU OF FISHERIES nauplii (besides copepod eggs), larval Anemones, and young Staurophora down to 30 meters, overlying a sparse adult Calanus-Sagitta-Pleurobrachia community in the deeper strata of water. There is some evidence that the wave of reproduction of copepods continues to spread offshore with the advance of the season until it covers the southwestern part of the gulf generally; and it certainly endures later into the spring in the open gulf than in Massachusetts Bay, for the presence of nauplii showed that in 1920 these little crustaceans were breeding actively from Cape Cod to Georges Bank as late as May 16 and 17. In the spring of 1915 nauplii were abundant on the surface off the Cape, with older stages deeper down, as late as the 26th of the month (station 10279), although they had been almost entirely replaced by the older larvse and by half-grown Calanus (fig. 29) as early as the 4th of that month off Gloucester (station 10266). Similarly, the presence of copepod nauplii in the sink off the Isles of Shoals on May 14, 1915 (station 10278), coupled with a decided increase in young copepods between April 26 and May 14 to 16, 1913 (Bigelow, 1914a, p. 407), though with diatoms still abundant there on both these occasions, 19 suggests that copepods do not begin to multiply this far offshore until well into May, although repro- duction is under way more than a month earlier than this inshore off the Merrimac River. We have no evidence that the coastal waters east of Penobscot Bay ever see a local reproduction of copepods comparable to the waves of production just described for Massachusetts Bay. As to local production of copepods along the eastern (Nova Scotian) side of the gulf, I can only say that our hauls near Lurcher Shoal on March 23 (station 20082) , and again off Yarmouth, on German Bank, and near Cape Sable on April 13 to 15, 1920 (stations 20102, 20103, and 20104), yielded nauplii and older larval copepods in some numbers, which probably marks the beginning of a period of active propaga- tion, for in 1915 we found both nauplii and the older juvenile stages of Calanus plentiful on the surface of the eastern basin near by on May 6. The vernal wave of production of these little crustaceans reaches its apex by the end of May or the first of June in the northern and eastern parts of the gulf, for we found a typical Calanus plankton reestablished off Boothbay (station 10280), in the Fundy Deep (station 10282), and off Mount Desert Island (station 10284) by May 31 to June 11 in 1915. An important problem in the natural economy of the gulf is how far the vernal augmentation of the zooplankton of the offshore parts of the gulf — say, outside the 100-meter contour — is due to local propagation there and how far to a migration of the copepods out from the coastal zone where they are produced in such enormous numbers. To answer this question definitely demands a more critical study of our towings than opportunity has yet allowed. One thing is clear, however. None of our offshore hauls at any season has ever yielded copepod nauplii or the later larval stages in numbers to compare with their abundance in Massachusetts Bay. It is equally suggestive that in May, when the coastwise copepod plankton is juvenile, large Calanus have invariably been an important element in the total copepod catches in the deep basin, just as is the case in summer, which points to the coastwise waters 10 In 1913 they were diminishing in numbers locally by that time. PLANKTON OF THE GULF OF MAINE 43 of the gulf, especially its southwestern part including the Massachusetts Bay region, as the chief source of the copepod plankton of its center. It is probable, also, that Georges Bank is an important nursery lor copepods, since nauplii occurred in some numbers among the adult calanoids off its northern slope on March 11, 1920 (station 20064). The vernal increase in the numbers of copepods present in the Massachusetts Bav region, and wherever else reproduction takes place actively, is many times greater than the bulks of the catches might suggest, the production of young coupled with the dying off of the parent stock giving the copepod plankton of the coastal waters a juvenile character in spring with relatively few large adults. Thus, there were only about 8,000 adult Calanus per square meter among some 500,000 copepods, mostly young Calanus, off Gloucester on May 4, 1915 (station 20066) — that is, a little less than 2 per cent. After the peak of production is past, however, and with the growth of its product toward maturity, the percentage of large Calanus and adults of other species once more increases, until they form about one-third of the copepod popula- tion at the mouth of Massachusetts Bay by the end of June or first week in July (Bigelow, 1922, p. 136). During the late summer, when the stock of copepods of all species and ages dwindles, adults may locally amount to as much as one-half or two- thirds of the total (fig. 30). Coincident with the vernal propagation of copepods various young medusae commence their period of pelagic existence, as, for example, Staurophora, which ap- pears in swarms in Massachusetts Bay in May. Although we have never found young medusas more than a minor factor in the zooplankton of the gulf outside the outer headlands in spring, they often dominate inclosed waters for a brief period in May. This, for instance, was the case in Gloucester outer harbor on May 3, 1913, when Sarsia tubulosa, Bougainvillea superciliaris , Rathkea blumenbachii, Tiaropsis dia- demata, Obelia, and Staurophora were all abundant, and .ZEquorea and Cyanea tolerably common — all of them, no doubt, liberated close at hand, and certainly very recently, for none was found there a month earlier. We also found young hydro- medusae swarming in the harbor of Yarmouth, Nova Scotia, in May, 1915, and this probably applies to similar situations all along the complex coast line of the gulf from Cape Cod to Cape Sable; also to the shallow waters of Georges Bank, where young Hybocodon and Staurophora are sometimes sufficiently plentiful to "color" the tow in April (Bigelow, 1914a, p. 414). The larvae of echinoderms, worms, and mollusks of many kinds likewise appear in the plankton along shore in spring. Most of these, in fact most of the pelagic animals of coastwise origin, are confined to estuarine situations- in the Gulf of Maine, to sounds and bays among the islands, or to a coastal belt only a few miles wide at most, as noted above (p. 32), and hence may be passed over without further comment here. The early stages of the common rock barnacle (genus Balanus), however, are so abundant and so conspicuous that they deserve a word of mention. In 1913, as I have elsewhere described (Bigelow, 1914a), barnacle nauplii 20 were taken in large numbers in the Isles of Shoals-Boon Island region 31 » Here let me correct an error in an earlier paper, namely, that "barnacle" eggs were taken in the tow in March and April of 1913 (Bigelow, 1914a, p. 108). Barnacle eggs are not set free to float, but are nursed by the mother until the nauplii hatch out. For accounts and figures of the early stages of Balanus see Hceck, 1909. » No doubt young barnacles are as common in Massachusetts Bay as in any part of the gulf, though somehow we have chanced to miss their season there. 44 BULLETIN OF THE BUREAU OF FISHEBIES on April 5; the cyprid stage in abundance on the 9th, with only a few nauplii; while by the 19th cyprids alone were taken. These dominated the surface plankton during the last week of April, after which their numbers diminished, though some persisted in that region until mid-May. The reproduction of barnacles is at its height at about the same season along the eastern shores of the gulf, for their nauplii occurred at all our stations over the shallows from Yarmouth to Browns Bank on April 13 to 15, 1920 — abundantly in the North Channel (station 20105; fig. 24). At St. Andrews, in the Bay of Fundy, where because of the violent tides the surface waters warm slowly in spring, barnacle larva? (either nauplii, cyprids, or both) are recorded by Doctor Mc.Murrich in his plankton lists as early as the last week of January, regularly after mid-February, reaching their maximum abundance during April, occurring in diminishing numbers until June 8, and occasionally still later in that month. In 1917, according to Willey (1921), barnacle nauplii dominated the plankton at St. Andrews on April 7; nauplii and cyprids in subequal numbers formed nearly the entire catch on May 1 ; and cyprids alone on the 17th. The season is about the same for them in the Irish Sea. The spring season, likewise, sees striking additions to the plankton of the coast- wise and shoaler waters of the gulf generally, in the shape of buoyant fish eggs. Haddock eggs in particular are produced in such numbers locally during March and April (which is the height of the breeding season) that they may be a considerable element on the more prolific spawning grounds, such as the eastern part of Georges Bank, the neighborhood of the Boon Island ground, and locally in Massachusetts Bay. The extremely characteristic eggs of the plaice (Hippoglossoides plate ssoides) appear early in March (that is, slightly later than those of the haddock) and are taken until mid-June, with the height of the spawning season during April and May. Rusty-flounder (Limanda) eggs are first seen in the tow toward the end of April, most numerously in June and July, and rarely as late as mid-September. The spawning season of the witch flounder (Glyptocephalus) likewise follows hard on that of the haddock. Spring is the season most prolific in fish eggs in the Gulf of Maine, but they are seldom numerous except in the immediate vicinity of the spawn- ing grounds, or anywhere over the central deeps of the gulf, outside the 100-meter contour. 22 The most obvious effect of the very active reproduction of copepods just described, coupled with the scarcity of most other planktonic animals in the offshore waters of the gulf at the time, is that soon after its inception the zooplankton in the more productive centers of propagation becomes almost pure copepod; and, whether by local breeding or by drifting out from the coastal belt, as seems more likely, their numbers so multiply offshore as the water warms with the advance of the season that they overwhelmingly dominate the pelagic community of the whole gulf north of a line from Cape Cod to Browns Bank in May and during the first half of June. Since, furthermore, the other planktonic groups of animals that assume faunal importance later on in the year (e. g., Sagitta?, amphipods, euphausiids) do not commence multiplying actively until later in the season, it is during late spring and the first weeks of summer that the zooplankton of the upper 100 meters (empha- " For the chief spawning grounds and breeding seasons of Gulf of Maine fishes see Bigelow and Welsh (1925). PLANKTON OF THE GULF OF MAINE 45 sizing this depth limit for reasons which will appear presently) of the offshore parts of the gulf is the most monotonous. Although our records for this season are not all that might be desired, it seems certain that copepods (Calanus in particular) reach their high-water mark early in June, the exact date varying locally and with the forwardness of the season. So completely did the calanoids (chiefly ft finmarchicus) monopolize the upper strata of water right across from Cape Cod to Cape Sable during May, 1915, that the only other animals to be found among a liter of copepods off Cape Ann on May 4 (station 10266) were a few Sagitta eltgans, one young fish, two tiny Euthemisto, a few euphausiid larvae, and a few fish eggs, with the zooplankton of the western basin (station 10267), where diatoms were still swarming, so monotonous that a haul from S5 meters yielded nothing but copepods and one Tomopteris. Nor was the catch more varied in the central deep (station 10269), only one euphausiid, one Euthemisto, six or seven large Clione, and an occasional Limacina being detected among the copepods in the 85-meter tow on May 6, while we found only a few Euthemisto, euphausiids, and Sagitta?, with an arctic planktonic element to be discussed else- where (p. 59), among swarms of copepods in the eastern basin on that same day (station 10270). In that year (which was apparently a typical one) the plankton of the upper 100 meters was as monotonously calanoid in June as it had been in May. In the Grand Manan Channel, for example, on the 4th (station 10281), the 50-meter catch consisted of copepods varied only by 1 Euthemisto, 2 Clione, 1 Aglantha, 1 young fish, 1 fish egg, 2 Sagitta elegans, and a single specimen of Tomopteris. Much the same condition prevailed in the Fundy Deep on the 10th (station 10282); likewise near Mount Desert Island on the 11th (station 10284), when a cursory examination of more than 2 liters of Calanus and other copepods in the 70-0 meter haul revealed only one Clione and a single Sagitta as the sole variants. On the 26th of June, too, the upper strata of the western basin were similarly occupied by a calanoid plankton in extraordinary abundance (about 40,000 large Calanus per square meter). In the western and northern parts of the gulf, where copepods monopolize the water more completely at their peak season than they do the deep basin offshore, it is an unusual event for Sagitta?, amphipods, euphausiids, or pteropods, etc., to be of any importance in the plankton in spring or early summer, with the notable exceptions of the swarms of the euphausiid shrimp Thysanoessa raschii near the Isles of Shoals in April and May, 1913, and (with its relative, Th. inermis) on April 9, 1920 (station20093) , described below (p. 145) ; with the exception, too, of Meganycti- phanes, which is so plentiful in the northeast corner of the trough off Grand Manan that we captured no less than 1J^ liters there on June 10, 1915 (station 10283), in half an hour's haul at 100-0 meters, and of Pleurobrachia, which swarms on German Bank in May and June just as it does in summer (p. 19). Even where copepods so dominate the contents of the net, however, that nothing else strikes the eye at the first glance, a more careful examination of the catch will reveal some few amphipods, euphausiids, Sagitta?, etc. June 19 is the earliest date on which we found large Euthemisto in any abundance in 1915 (eastern basin, haul from 85-0 meters, station 10288). The interesting 46 BULLETIN OF THE BUREAU OF FISHERIES hydroid medusa Mitrocoma cruciata reaches maturity during this same month, when it may appear near shore in numbers sufficient to give a distinctive aspect to the tow, as was the case at the mouth of Penobscot Bay on June 14, 1915 (station 10287 p. 348). For the sake of clarity I should point out, at the risk of repetition (p. 389), that diatoms still swarm along a narrow coastwise belt east of Penobscot Bay in June. The advance of summer (from June on) sees an actual decrease in the number of copepods, owing, no doubt, to the destruction wrought among them by fishes and other enemies (p. 97). In part this decrease is made good by constant reproduction, evidence of which was afforded by an abundance of copepod nauplii near Cape Cod on July 8, 1913 (station 10057, surface), on July 7, 1915 (station 10300), and on August 29, 1916 (station 10398) ; likewise by the presence of large numbers of juvenile Calanus 23 between Cape Ann and the Isles of Shoals in July, 1912. The offshore banks also serve as a copepod nursery in July — at least locally — for copepod eggs, nauplii, and juveniles abounded on the surface near Nantucket Lightship on the 25th of that month in 1916 (station 10355), while the presence of young Calanus at various stages in development in most of the summer towings proves that this copepod breeds more or less regularly throughout the summer. Our experience, however, does not suggest that sufficient reproduction takes place during the warm months to maintain the local stock of calanoid copepods against depletion by the many dangers to which it is subjected. As copepods dwindle in numbers the other groups of common boreal animals increase, lending an increasing diversity to the plankton of the offshore parts of the gulf during the summer, most noticeably in the western side, where the plankton is most monotonously calanoid in May and June, thus producing the midsummer state already described (p. 17). Events notable in this gradual alteration are a great production of Euthemisto, resulting from local centers of reproduction such as I have just mentioned (p. 20) ; the active propagation of euphausiids (p. 20) ; a general penetration toward the western and northwestern shores of the Gulf on the part of the pteropod Limacina retroversa (p. 119); the appearance of shoals of the white and red jellyfishes (Aurelia and Cyanea) in the coastal belt as they disperse and drift seaward from their estuarine nurseries (pp. 360, 362) ; the presence of large Stauro- phora, often in abundance (p. 342) ; and the offshore swarming of the hydroid medusa Phialidium languidum (p. 350). It is during the summer, too, that the large and conspicuous arrow-worm Sagitta serratodentata first appears in any number in the gulf as a visitor from warmer waters to the south and east outside the edge of the continent, and spreads its range northward and westward as described elsewhere (p. 322). The copepod population, also, becomes diversified as the summer advance by increasing numbers of Anomalocera and Centropages, not only within the gulf but also on Georges Bank, where the former (which we did not find in spring) is practically universal and comparatively abundant in August. 24 The ctenophore Pleurobrachia pileus reaches its maximum abundance on the German Bank ground " Identified by Dr. C. O. Esterly. '< The "green copepod" of Doctor Kendall's field notes. PLANKTON OF THE GULF OF MAINE 47 and may almost completely monopolize the water there during the summer. In June and July, too, the eggs or larvae, or both, of sundry summer-breeding fishes, such as silver hake, rosefish, cunner, and witch flounder, appear in the appropriate parts of the gulf to take the place of such spring spawners as the haddock and plaice. As summer passes into autumn Sagitta serratodentata continues to spread west- ward right into Massachusetts Bay (p. 322). The hyperiid-amphipod genus Euthemisto likewise works inshore in September and October, so that it is more numerous in the bay then than at any other time of year, and Pleurobrachia may swarm locally, notably off the coast of eastern Maine and at the mouth of the Bay of Fundy. It is during late summer or early autumn, too, that Phialidium is most plentiful and that Salpa? and other tropical forms (p. 53) are most often encountered in the gulf. Hand in hand with the autumnal cooling of the surface, the small Phialidium languidum disappears first and then the larger scyphomedusae, either dying at the close of their natural period of life or being destroyed by the fury of the autumn storms. The large, blue copepod Anomalocera likewise vanishes from the waters of the gulf (p. 1S4). On the other hand, ctenophores may be locally abundant until well into the autumn, witness the swarms of Pleurobrachia that appeared off Cape Cod during October, 1916 (p. 367) ; and the small brown copepod Temora longi- cornis becomes so plentiful locally near the land at this season that it dominated the surface catch off Cape Ann on October 31, 1916 (station 10399), when a sample of the copepods consisted of over 100 Temora with but 2 Centropages and 1 Calanus. Doctor McMurrich, likewise, found Temora most regularly and in greatest abun- dance in October, November, and the first half of December at St. Andrews (p. 289), but in the open Gulf no definite seasonal periodicity has been established for it (p. 2S9) . Centropages was the most numerous copepod on the surface off Cape Cod in November, 1916 (station 10404), but all our deeper hauls in autumn have been dominated by Calanus, Pseudocalanus, and Metridia, with Euthemisto of both species, Sagitta elegans, Meganyctiphanes, Thysanoessa, and Limacina. In fact, they have paralleled the community characteristic of summer. So few of the bot- tom dwellers of the Gulf breed in October or November that their larvae are practi- cally nonexistant in the plankton at that season; but the presence of juvenile Calanus in the western basin on November 1 (station 10400), of young Aglantha and young Sagitta elegans, of eggs probably referable to the latter, and of an abundance of small as well as large Limacina off Massachusetts Bay at that time (stations 10399 and 10403) proves that all these pelagic animals reproduce in the Gulf during October, though probably not in any great abundance. I have already pointed out that no general alteration takes place in the zoo- plankton of the Massachusetts Bay region during late autumn and early winter, for our tows gave us much the same yield off Cape Ann at the end of November and in December, 1912, and in January, 1913, 25 as is to be expected there in August, Sep- tember, or October — that is, Calanus dominant, with such other copepods as Pseudo- calanus, Metridia lucens, Centropages, and Euchaeta; the chsetognaths, Sagitta elegans and occasional S. serratodentata; Euthemisto compressa and E. bispinosa; the common " These hauls are described in an earlier report (Bigelow, 1914a, p. 404) 7589S— 20 4 48 BULLETIN OF THE BUREAU OF FISHERIES boreal pteropod Limacina retroversa; and the ctenophores Pleurobrachia and Beroe. This also applies to tow-net catches at 12 stations between Cape Cod and Yarmouth (Nova Scotia) for the midwinter of 1920 and 1921, listed below. These lists vary somewhat from station to station, as is always to be expected, but there is no charac- teristic qualitative difference between the western and the eastern stations, the Calanus community (and chiefly C. finrnarchicus) dominating the same general assemblage of boreal animals as occurs in summer at the localities in question. Location, date, and depth of hauls Species J oa Boston, Dec. 29, 1920, station 104S8, 15-0 meters Off Cape Ann, Dec. 29, 1920, station 10489, 75-0 meters Western Basin, Dec. 29, 1920, station 10490, 240-0 meters Off Cape Cod, Dec. 30, 1920, station 10491, 125-0 meters Off the Merrimac, Dec. 30, 1920 , station 10492, 20-0 meters Off Isles of Shoals, Dec. 30, 1920, station 10493, 75-0 meters X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 X X X X X 2 X X X X X X 3 X X 1 1 1 X 1 X X X X X X X Location, date, and depth of hauls Species ' Off Cape Elizabeth, Dec. 30, 1920, station 10494, 75-0 meters Off Seguin Island, Dec. 31, 1920, station 10495, 60-0 meters Off Matini- cus Island, Jan. 1, 1921, station 10496, 100-0 meters Off Mount Desert, Jan. 1, 1921, station 10497, 50-0 meters Fundy Deep, Jan. 4, 1921, station 10499, 150-0 meters Off Lurcher Shoal, Jan. 4, 1921, station 10500, 60-0 meters X X X X X X X X X X X 3 X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 X X 1 1 X X 1 X 4 1 X 1 X 7 X 1 X X 1 X 1 X X 1 X 1 1 X 12 4 X X X X 1 For complete lists of the copepods at these stations see p. 304. PLANKTON OF THE GULF OF MAINE 49 The winter plankton of 1920-1921 differed from that of 1912-1913 in the rarity of the amphipod genus Euthemisto, both species of which not only occurred regularly dining December, January, and February, 1912 and 1913, but usually in consider- able numbers. Sagitta elegans, though it occurred regularly, was also far less numerous in the midwinter of 1920-1921 than at that season in 1912-1913, when it was an important factor in the tows made in Massachusetts Bay from December untd February. Whether these differences were actually the result of annual fluctua- tion in the stock of these two animals present or whether both are normally more abundant in Massachusetts Bay and its vicinity than in other parts of the gulf in winter remains to be learned. Other features of the winter plankton of the gulf worth mention are that the buoyant eggs of the American pollock (PollacTiiusvirens) appear in great numbers from November until February over its restricted breeding grounds; that cod eggs are to be expected throughout the winter (Bigelow and Welsh, 1925, p. 424) if the nets be towed near where the fish are spawning — seldom otherwise or in large numbers; and that some few copepods (probably Calanus) continue to reproduce right through the cold season, for their nauplii were detected at most of our December- January stations of 1920 and 1921, most plentifully in Massachusetts Bay. Euthemisto, too, must breed then (though probably in small numbers) to account for very young specimens taken off Gloucester on December 29, 1920. In this connection I may also call attention to numbers of large Calanus hyperboreus (5 per cent of all the cope- pods) among a very rich catch of C. finmarchicus in the western basin on December 29, 1920 (station 10490, p. 304), and of Stephanomia bells in the eastern basin and in the shoal water off Yarmouth (Nova Scotia), which was nearly barren otherwise, on January 5. On the other hand, the arrow-worm Sagitta serratodentata vanishes from the gulf sometime during late winter, our latest seasonal record of it being for January 16, 1913 (off Gloucester). Judging from the tow-net hauls made during 1913, the zooplankton of the Massachusetts Bay region continues decidedly uniform in composition throughout January and February, when the successive hauls reproduced one another with monotonous regularity, until early in March, when the quantity of animal plankton present in the water decreased to its annual minimum (p. 39) coincident with the vernal augmentation of vegetable plankton described elsewhere (p. 385), a change soon followed by the wave of reproduction on the part of the copepods which I have just discussed. It may safely be assumed that this is equally true of the northeastern part of the gulf, for although, unfortunately, we have no plankton records from its outer waters during the period January 9 to February 22, Doctor McMurrich found Calanus jinmarchic us and Pseudocalanus, with Temora longicornis and the neritic copepod genus Acartia, the chief animal constituents of tow-net catches during this season of the year at St. Andrews. The seasonal planktonic cycle in the deep waters of the gulf below 100 meters calls for separate discussion, because the Euchseta community is largely below the reach of the wide fluctuations of temperature to which the inhabitants of the shoaler strata of the gulf are subject. Data on this for the early winter consist of two tow-net hauls, one from 240 meters in the western basin, December 29, 1920 50 BULLETIN OF THE BUREAU OF FISHEEIES (station 10490), and the other from 150 meters in the eastern basin on January 5, 1921 (station 10502). On the former occasion the only, members of the Euchreta community detected among a great abundance of large Calanus Jinmarchicus and Calanus liyperboreus (p. 304) were a few Eucheeta and Eukrohnia; on the latter date the whole catch was extremely scanty (not over one-tenth liter) , consisting chiefly of debris of the siphonophore genus Stephanomia, with Calanus and other copepods, among which there were a few Euchseta, Meganyctiphanes, Thysanoessa inermis, Th. longicaudata, Sagitta elegans, pteropods {Limacina retroversa), two Euthemisto com- pressa, but none of the deep-water chaetognaths. These hauls suggest that a decided impoverishment of the deep-water plankton takes place during the autumn, but this may have been accidental. The Euchseta community probably persists unal- tered in qualitative composition throughout the winter, as widespread over the deep trough then as it is in summer, judging from the following catches made with the closing net in the central and eastern parts of the basin on March 2 to 3, and in the Fundy Deep on March 22, 1920. [D, dominant; M, many; X, occurrence] Species Station 20052. central basin, 160 meters Station 20053, Station 20055, southeast part, 175 meters east basin, 180 to 140 meters Station 20079, Fundy Deep, 180 meters Calanus finmarchicus .. Metridia lucens Euchfeta norvegica Meganyctiphanes norvegica. Thysanoessa i nermis Pasiphsea Euthemisto compressa Euthemisto bispinosa Tomopteris catharina Sagitta elegans... Sagitta lyra Eukrohnia hamata Limacina retroversa.. Clione limacina Beroe' _ _ Aglantha D X M 11 X X X 1 1 1 X M D X X 2 1 1 1 M X "x" 12 X X X 1 X X X M 22 X 12 1 1 1 X ' 1 20+ > 1 l 1 1 In open-net haul from 200 meters. Occurrence of characteristic animals in the Eastern Basin, various localities and months [D, dominant; M, many; X, occurrence] Location, date, and depth of hauls Species Station 20081, 140-0 meters. Mar. 22, 1920 Station 2O086, 150-0 meters, Mar. 23, 1920 Station 20112, 200-0 meters, Apr. 17, 1920 Station 10270, 150-0 meters, May 6, 1920 Station 10288, 200-0 meters, June 19, 1915 Station 10246, 150-0 meters, Aug. 12, 1914 Station 10093, 170-0 meters, Aug. 12, 1913 Station 10310, 175-0 meters, Sept. 2, 1915 Stations 10500 and 10502, 150-0 meters, Jan. 4 and 5, 1921 D D D D X D M X D D X M M X D X X X X D X M D X D X D X X X X X X M X X X Megan vctiphanes norvegica X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 1 X x 1 For further lists of the copepods see p 297. PLANKTON OF THE GULF OF MAINE 51 A similar community (notably Euchseta and the deep-water choetognaths) also occupied the deeper water layers in the western basin in February and March, 1920 (p. 40), and deep hauls made there and in the southeastern part of the basin that April gave much the same yield. Judging from hauls made in 1915, however, the deep-water chaetognaths Eukrohnia Tiamata. and Sagitta maxima disappear altogether from both the western and the northeastern deep troughs in May, not to. reappear there until August, 20 a phenomenon interesting for its bearing on the lines of immigration of these two species, neither of which breeds in the gulf, and as evidence of the seasonal fluctuation of the bottom current. But it is possible that they persist in the southeastern deep and in the eastern channel. It is probable that the Euchseta community of the western basin is at its lowest ebb in May or June, for if the euphausiid shrimp Meganyctiphanes norvegica was not wholly wanting there during those months in 1915, it was at least so rare that the nets did not chance to pick up any specimens, although it was plentiful in the eastern trough at the time. Meganyctiphanes repopulates the deep waters of the western side of the gulf by midsummer, however, for we have found it there at all our stations for July and August (p. 151), and the mammoth copepod Euchseta norvegica is as constant, though not as abundant, an inhabitant of the deepest waters of the gulf, season in and season out, as Calanus is of the upper strata. IMMIGRANT PLANKTONIC COMMUNITIES Besides the endemic boreal animals so far discussed (chiefly the Calanus com- munity), which are the most important members of the animal plankton of the Gulf of Maine, various immigrants enter it from time to time, as might be expected in any maritime area where waters of diverse origin meet and mix, the details of such immigrations varying with the ocean currents that give them birth and in which their participants normally pass their existence. According to their adaptability to the temperatures and salinities which they meet in the gulf, these involuntary visitors exhibit every degree of success as col- onists, from inability even to survive for more than a few days or weeks to perfect success in existing, growing, and breeding. The majority, however, occupy a middle ground — able to live and grow to large size in the gulf but not to reproduce them- selves there because of unfavorable temperatures or salinities, or at most breeding so seldom that their continued presence in the gulf depends absolutely upon successive waves of immigration from outside. Associated with their essentially exotic origin, most of these immigrants are decidedly seasonal in their appearance within our limits. To place clearly before the reader the faunal status of such wanderers, I must emphasize here (what is perhaps the most essential factor in the biology of all pelagic animals below the rank of fishes, and a truism to the oceanographer) their utter inability to cany out voluntary migrations of more than a few miles at most from place to place by swimming, for want of a continuous directive stimulus, though they often perform extensive vertical movements. The horizontal migrations of :« Possibly in July, a month for which we have but one deep station. 52 BULLETIN OP THE BUREAU OF FISHERIES planktonic animals, so often recorded and occasionally so extensive, are invariably the result of actual and corresponding movements of the water masses in which they live. Utterly at the mercy of tide and current, they drift as helplessly as buoys with the latter, able to escape from an unfavorable environment only by swimming up or down in response to light or to gravity. For them there is no such thing as the geographic migration in the true sense, with which we are familiar among birds and fishes. It follows from this that to state the currents or the more diffuse movements of water that enter the Gulf of Maine is to list the sources from which occasional visitors can reach it. These are, first, but least important, the surface stratum of tropical water, popularly known as the Gulf Stream, lying close outside the continental edge, proverbial both for high temperature and salinity and for the tropical pelagic fauna it carries with it, and which enters the gulf regularly, though in small amounts, as a component of the general surface indraught into its eastern side, besides flowing directly across Georges Bank on rare occasions. Second, and equally characteristic both hydrographically and biologically, is the ice-cold water of the Cabot or Nova Scotian current that flows past Cape Sable in considerable volume in spring, carry- ing arctic inhabitants. Greater in amount than either of these, though not always so clearly characterized by its plankton, is the complex mixture between coastal, northern, and tropical oceanic waters, which is constantly being manufactured along the outer edge of the continental shelf and over the upper part of the continental slope, and which composes the major part of the influx into the eastern side of the gulf. To this the name "cold wall" has often been applied. Finally, the mid-depths of the Atlantic basin contribute an occasional straggler, which must enter via the deepest trough of the Eastern Channel. None of these sources, except the third, adds appreciably to the gulf plankton, in which, as I have pointed out, endemic animals are overwhelmingly preponderant; but so important are the exotic forms as indicators of the respective waters that give them birth that they deserve more attention than their numerical strength of itself would warrant. Several of the commonest and most characteristic inhabitants of the different ocean currents are among the largest and most easily recognized. For example, the presence of a Salpa or of a bit of gulf weed (Sargassum) anywhere in the Gulf of Maine is as sure evidence of an actual influx of Gulf Stream water as if the latter could actually be seen, and the same is true of the Arctic pteropod Limacina helicina for northern waters. Note, also, that whatever the origin of an exotic immigrant, whether Tropic or Arctic — or any driftage, for that matter — it travels the same route, once it is caught up in the inflow into the eastern side of the gulf, a fact well illus- trated by the striking resemblance between the distribution (within our limits) of the cold-water Aglantha, on the one hand (p. 353), and the whole category of tropical organisms, on the other (fig. 31). So close, in fact, is the parallel, that the one chart mifht almost be substituted for the other, so far as the inner parts of the gulf are concerned, were the seasonal element ignored. Immigrants in the upper strata, whatever their source, rarely reach the central part of the gulf unless their numbers be fortified and their period of existence within our limits lengthened by local repro- duction; but those entering in the deeper strata of water do follow the troughs (p. 64). PLANKTON OF THE GULF OF MAINE 53 TROPICAL VISITORS The term "tropical visitors" is used here for such animals as are native to the Gulf Stream and are able to survive only in its warm surface waters outside the edge of Portland Fig. 31.— Locality records for certain of the more typical planktonic animals of tropical or warm-Atlantic origin. A, Salpae, • Thysanoessa gregaria; X, tropical copepods; O. Portuguese man-o-war (Physalia); A Physophora hydrostatica; 0, gulf weed (Sargassum); — , many tropical species the continent. Others equally of tropical origin, but which find conditions more favorable for growth (though not for reproduction) in the mixed water, are discussed as belonging to the latter, for it is by that route that they enter the Gulf. 54 BULLETIN OF THE BUREAU OF FISHERIES Ever since the early eighties it has been known (from many collecting trips carried on by the vessels of the United States Bureau of Fisheries from the laboratory at Woods Hole) that the inner edge of the tropical water, carrying with it an extra- ordinarily rich and diversified tropical plankton, lies only a few miles south of the 100-fathom contour off Marthas Vineyard in summer, just as is the case farther west and south. Hence, although actual records of the pelagic fauna and flora at this same relative position farther east have been very scanty up to within the last few years, there was no reason to doubt that a tropical community occupied the same relative position along the slope off Georges Bank; while the deep-sea explorations of the National and Michael Sars, of the Canadian fisheries expedition of 1915, and of the international ice patrol (Fries, 1922), have shown that the same assemblage of warm-water planktonic animals and plants characterizes the inner (northern) edge of the Gulf Stream to and beyond the southern corner of the Grand Banks of New- foundland. It was therefore to be expected that any lines we might run seaward as far, say, as the 1,000-meter contour, would bring us into warm water, where our tow nets would yield a tropical plankton instead of the boreal community charac- teristic of the Gulf of Maine to the north. And so it has proved, as the follow- ing brief notes on our offshore hauls will illustrate. On July 10, 1913, for instance, we saw fragments of gulfweed on the surface near Nantucket Lightship, and the neighborhood of the stream was made evident over the 150-meter contour to the south (station 10061) by "the presence of Salpae, Phronima, and the amphipod genus Vibilia, though the bulk of the plankton still con- sisted of Calanus Jinmarchicus, with such other boreal forms as Euchseta norvegica, Euthemisto, and Sagitta elegans " (Bigelow, 1915, p. 268) . We had a similar experience over the 1,000-meter contour, some 70 miles farther east, about a week later in the season the following year (station 10218), when we found the water of the high tem- perature 27 characteristic of the inner edge of the Gulf Stream, more properly the tropical water (p. 52), with a typically tropical plankton including Salpa fusiformis and its relative genus, Doliolum; the tropical amphipod genera, Phronima, Vibilia, and Oxycephalus; the copepods Rhincalanus and Sapphirina; the chastognaths Sagitta enflata, S. kexaptera, and Pterosagitta draco; with the 11 species of tropical pteropods and 19 species of tropical medusas and siphonophores listed below, and gulfweed (Sargassum) floating on the surface, as I have elsewhere noted (Bigelow, 1917, p. 245). Tropical pteropods and ccelenlerates taken over the continental slope off Georges Bank, July 21, 1914, station 10218 Species Mollusks: Limacina rangii, d'Orb Creseis conica, Eschscholtz Creseis acicula, Rang Hyalocylis striata, Rang Cuvierina columnella. Rang — Diacria trispinosa, Lesueur Cavolina longirostris, Lesueur.. Cavolina uncinata, Rang Peraclc reticulata, d'Orb Corolla calceola, Verrill. Firoloida desmarestia, Lesueur. Pleurobranchea tarda, Verrill. . Medusae: Stomotoca ptcrophylla Toxorchis kellneri.. Laodicea cruciata 60-0 meters X X X 300-0 meters 1 2 X X X 400-0 meters 1 1 1 1 2 1 1 Species Medusa; — Continued. Rhopalouema lunemrium. Rhopalonema velatum Liriope scutigera Liriope tetraphylla ... Aglaura hemistoma. Nausithbe punctata - Siphonophores: Hippopodius hippopus Diphyes spiralis Diphyes appendiculata Diphyes bojani Diphyopsis dispar.. Diphyopsis mitra Agalma elegans.. Anthophysa formosa Physalia physalis 60-0 meters X X X X X X X X X 300-0 meters X X x X x X x x 400-0 meters " Temperature 17.7° and salinity 36.04 per mille at 40 meters; 20.48° at the surface. PLANKTON OF THE GULF OF MAINE 55 Rather scanty catches at the same relative position on the slope 100 miles far- ther east on July 22, 1914 (station 10220), likewise included tropical animals (Rhin- calanus, a phyllosome crustacean larva, Phronima, Doliolum, and four specimens of the warm-water pteropod Limacina rangii) as well as boreal, while the tropical ele- ment was similarly represented by Phronima and Sagitta enjlata in the plankton over the slope off Marthas Vineyard a month later (August 26, stations 10260 and 10261), although the catch was chiefly boreal (Bigelow, 1917, p. 245). In the cold summer of 1916 the tropical water lay farther out from the edge of Georges Bank in July, with the 50-meter temperature ranging from 4.85° to about 8° over the slope between the 175 and 1,000-meter contours on the 23d (stations 10349-10351, and 10352). Corresponding to this, the plankton along this zone was typically boreal (much the same as in on the bank and in the gulf), Calanus finmarchicm dominating, with Pseu- docalanus, Metridia lucens, Euchxta norvegica, large Euthemisto compressa and E. bispinosa abundant (as is usually the case along the slope), Limacina retroversa, Thysanoessa inermis, Th.raschii, and Sagitta elegans. Indicative of the zone of mix- ture between coastal and ocean water was the fact that Sagitta serratodentata was about as numerous as S. elegans over the 200-meter contour (station 10349) and Nematoscelis megalops at the outer station; but the only planktonic animals or plants to which a tropical origin could safely be credited were a few Salpa Jusiformis at station 10349, many at station 10352, a single Physophora hydrostatica (station 10353), a large Pyrosoma (station 10352), and a few fragments of gulfweed (Sargassum, station 10352). This poverty of warm-water forms contrasted strongly with what we had found there in July, 1914, listed above (p. 54). None of our three lines off Cape Sable (where high temperatures are separated from the slope by a still broader wedge of cold mixed water) has run out far enough to reach Gulf Stream water. Nevertheless we have taken Rhincalanus and Sagitta enjlata over the 500 to 1,000 fathom contours in summer even there (station 10233), and have seen Physalia (June 24, 1915). No doubt the boreal forms would be left behind altogether a few miles farther out to sea along this line in summer also, to give place to tropical forms on the surface and to typically oceanic plankton in the shadow zone of the mid-depths. In winter and early spring it is necessary to go considerably beyond the 1,000- meter contour to find surface water as warm even as 10° or tropical pelagic animals in any numbers abreast of the Gulf of Maine. For example, on February 22, 1920, the only representatives of this community in hauls made off the western end of Georges Bank (station 10244) were an occasional copepod (Rhincalanus) and amphi- pod (Phronima), with Phronima and the medusan genus Rhopalonema at the corresponding location off Cape Sable on March 19 (station 10277). The tow off the southeast face of Georges Bank on March 12 (station 10269) produced no dis- tinctively tropical forms, but by May 17 of that year the Gulf Stream community had again approached so close to the western end of the bank that our nets yielded several Salpse, subtropical copepods (Eucheirella), amphipods, and medusas (Rhopalonema) among the boreal organisms of which the bulk of the plankton con- sisted at the outermost station (20129). 56 BULLETIN OF THE BUREAU OF FISHERIES Tropical pelagic, animals as conspicuous as Salpa and the Portuguese man-of- war (Physalia), together with others less noticeable, are often carried close in to the coasts of southern New England during the summer, west and south of longitude 70°, by sporadic movements of Gulf Stream water, with the topographic bight west of Nantucket Shoals serving in particular as a trap for them, as the common occur- rence of Physalia at Woods Hole and the considerable list of tropical pelagic fishes that have been taken there (H. M. Smith, 1898; Kendall, 1908; Sumner, Osburn, and Cole, 1913) bear witness. Occurrences of this sort are far less frequent east of Cape Cod, however, and when invasions of the inner part of the Gulf of Maine by tropical planktonic animals do take place it is usually in the persons of but few individuals and fewer species. How slightly this tropical pelagic community encroaches on Georges Bank even in midsummer, when abundantly represented only 15 to 20 miles seaward from its 200-meter (100-fathom) contour, was brought forcibly to our attention in July, 1914, when only occasional warm-water animals or plants (e. g., Pterotrachea Icerau- denii, Doliolum, Phronima, a phyllosome larva, and the tropical pteropod Cavolina tridentata) occurred over the southern edge of the bank (station 10219) where the plankton was otherwise boreal, in spite of the rich and varied tropical plankton we have just mentioned (p. 54) as occupying the warmer water over the continental slope only a few miles farther out. Tropical pelagic animals have been found even more rarely in the inner parts of the Gulf of Maine than along the offshore banks, as might be expected. In fact, the euphausiid shrimp Thysanoessa gregaria (p. 142) is the only member of this com- munity occurring regularly there (but see, also, Sagitta serratodentata, discussed on p. 320). Except for these, the complete list of tropical planktonic animals so far detected in our catches in the gulf proper is brief. Among copepods the genera Eucalanus, Dwightia, Eucheirella, Pleuromamma, and Rhincalanus may be so classed, because all of them undoubtedly enter the gulf from the inner edge of the Gulf Stream, and, judging from their rarity, are unable to establish themselves in its cool waters, though properly speaking they are oceanic-Atlantic rather than typically tropical. The status of each in the gulf is given in detail in the chapter on copepods. The euphausiid shrimp Nematoscelis megalops, often plentiful along the continental slope, appears only as a stray in the interior parts of the gulf (p. 146). Salpa? (perhaps the best tropical indicators of all) have been taken at a number of stations, usually represented, however, by few examples. This was the case with Salpa fusiformis near German Bank and off Lurcher Shoal, August 14, 1912 (stations 10030 and 10031), though other scattered speci- mens were seen floating on the run from one station to the other. A few Salpa tilesii were also taken in the tow near Lurcher Shoal, August 12, 1913 (station 10096). Huntsman (1921) records five S. fusiformis found on the beach at Campobello Island (New Brunswick) in the autumn of 1913, and two S. zonaria taken in that general region (probably near Grand Manan) in 1910. On September 30, 1912, Capt. John McFarland, of the fishing schooner Victor, to whom the Bureau of Fisheries is indebted for other interesting tow-net hauls, made a large catch of S. mucronata 25 miles off Chatham, Cape Cod; and fishermen reported great PLANKTON OF THE GULF OF MAINE 57 numbers of large Salpse (probably S. tilesii) in Massachusetts Bay in November and December, 1913, which, so far as I can learn, are the only occasions when Salpsa have been found in such numbers within the gulf, though they are often reported in abundance south and west of Cape Cod. Local swarms, such as this, probably result from their very rapid asexual multiplication (there is no evidence that they can reproduce sexually in cool waters) in summer and early autumn (A. Agassiz, 1866). The Portuguese man-of-war (Physalia) , with its translucent float, is even more apt to attract attention than Salpa, as it drifts on the surface, and it is equally a tropical visitor, though at the mercy of wind as much as of current. We have only one record of Physalia within the gulf, viz, in the eastern basin, June 19, 1915 (Bigelow, 1917, p. 246; a single specimen seen but not captured). In the summer of 18S9, however, a year when Physalia was unusually plentiful off the coast of southern New England, many were seen in the Bay of Fundy and several were taken near Grand Manan and submitted to Doctor Fewkes for identification (Fewkes, 1889 and 1890). The only other tropical coelenterates so far recorded within the gulf are two examples of the siphonophore Physophora hydrostatica on German Bank (station 10030) in August, 1912 (Bigelow, 1914, p. 103), 28 while the "Venus girdle" (Cestum), a warm-water ctenophore, is known from off the southeast slope of Georges Bank (Smith and Harger, 1874; Bigelow, 1914b, p. 31). We have one record for a tropical pteropod (Limacina inflata) off Cape Cod on July 19, 1914 (station 10213), while two living specimens of the pteropods Diacria trispinosa and Atlanta, genera that are of warm Atlantic if not strictly tropical origin (Meisenheimer, 1905), were taken in a haul near Gloucester on July 8, 1913. The warm-water hyperiid amphipod Phronima sedentaria was taken on Browns Bank on June 24, 1915 (station 10296), which, with a fragment of gulfweed near German Bank (September 2 of that year), completes the list. The geographical locations of these records, the most characteristic of which are shown on the accompanying chart (fig. 31), and their dates prove that occasional planktonic immigrants from the inner edge of the Gulf Stream may be expected anywhere in the Gulf of Maine at any season. Aside from Thysanoessa gregaria, however, which may, perhaps, be endemic in small numbers in our waters, or which at least is able to survive there for a long time if it does not reproduce (p. 143), and omitting Sagitta seiratodentata, which falls in a different category (p. 58), there is a decided preponderance of tropical records in the eastern part of the gulf, though fewer hauls have been made there than in the western, a concentration, that is to say, where the salinity curves locate the chief influx of offshore water. The great majority of the records lie in the peripheral zone corresponding to the anticlockwise oceanic eddy that dominates the circulation of the gulf. In spite of the considerable tropical list, we have never made anything that could be called a tropical haul in the gulf or encountered a community of animals of warm- water origin there. In fact, most of the records are for single specimens; seldom has the tow net yielded as many as half a dozen at any one station, and, except for certain » Also taken off the southern face of Georges Bank on July 24, 1916, station 10352. 58 BULLETIN OF THE BUREAU OF FISHERIES copepods (p. 56), never more than two tropical animal species among the hosts of boreal animals. This scarcity of planktonic visitors of the tropical category within the Gulf of Maine and even over its shallow southern rim, when so rich a tropical surface fauna inhabits the inner edge of the Gulf Stream along the outer edge of the continental slope only a few miles without the 100-fathom contour, is fundamentally due to their inability to survive or to reproduce in the low temperatures of the coast water. Their sporadic and solitary occurrence there, contrasted with the considerable numbers and even communities of tropical planktonic animals that often drift close inshore west of Cape Cod, is explicable only on the assumption that the surface waters of the Gulf Stream very seldom overflow the barrier formed by Georges Bank, an assumption corroborated by the physical character of the water. Nevertheless, the Gulf of Maine does owe to the tropical water indirectly, if not directly, one common and very characteristic summer visitor, the large chretognath Sagitta serrato dentata. This species, which is the dominant member of its systematic group in the coastal waters south of New York, occupies a rather peculiar faunal niche in the Gulf of Maine, for while it breeds only in the high temperatures of the Gulf Stream (so far as the area under discussion is concerned), great numbers drift into the cooler mixture zone along the edge of the continental shelf, where they thrive and grow to a much larger size than they do in the warmer waters farther offshore, either because lower salinities and temperatures especially favor their growth (though not their reproduction), or perhaps because of a richer food supply (p. 323, and Hunts- man, 1919). As a denizen of this mixed water, S. serratodentata is swept in abundance into the Gulf of Maine, where, because of its size and abundance, it is the most prominent of all the exotic immigrants, though it never attains a more permanent status there. Owing to its peculiar relationship to oceanic temperatures, all the Gulf of Maine records so far obtained for S. serratodentata have been for large specimens, the locali- ties of capture indicating considerable longevity for it within the gulf. It is strictly seasonal in its presence there, however, being so rare in winter and early spring that we have taken it only twice between December 1 and May 1, viz, in Massachusetts Bay on December 4, 1912 (station 10048), and again on January 16, 1913 (station 10050). It appears in the eastern side of the gulf as early as the first week in May (p. 320, and Bigelow, 1917, p. 296), and by June it has spread generally over the eastern basin and into the Bay of Fundy as well as over the outer edge of the shelf off Cape Sable, and probably also all along the southern and eastern parts of Georges Bank, where we found it in July, 1914. This species penetrates the inner parts of the gulf so slowly during the early summer that in five years we have found it only once in the western and southwestern parts prior to August 1. Thereafter, however, it spreads so rapidly westward and southward along the coast of Maine that our August and September records for it cover the whole northern half of the gulf from Cape Ann right across to Cape Sable, including Massachusetts Bay, where it occurs regularly in late summer and autumn. The locations of the stations of capture and the fact that S. serratodentata is usually more numerous in the eastern than in the western side of the gulf (p. 322) are PLANKTON OF THE GULF OF MAINE 59 sufficient evidence that its invasion takes place chiefly into the eastern side and from the southwest and south; that is, across the eastern end of Georges Bank and via the Eastern Channel. It is probable (as suggested by Doctor Huntsman in a recent letter) that S. serratodentata also comes to the gulf from the east, drifting with re- current movements of mixed water along the outer edge of the continental shelf oft* Nova Scotia and entering across Browns Bank or through the Eastern Channel, but there is no reason to suppose that any come by way of the Northern Channel or around Cape Sable across the coastal shallows; in fact, it would be very surprising to find any warm-water species journeying along that route. Our failure to find S. serratodentata off Cape Cod in autumn, although Septem- ber, October, and November are the months when it is widest spread in the northern parts of the gulf, suggests that the individuals of the species taking part in the successive waves of immigration inward past Nova Scotia seldom survive long enough in the eddy-like circulation of the gulf to journey much beyond Massachusetts Bay in their circuit. The fact that specimens from the outer edge of the continental shelf have been much larger than is usually the case in the Gulf Stream, or in tropical seas generally, corroborates this view, for it indicates a considerable sojourn in the cool band of banks water on the part of S. serratodentata before it enters the Gulf of Maine. ARCTIC VISITORS In the Gulf of Maine the Arctic, like the Tropic, immigrants fall in two categories, depending on whether they are able to survive for a considerable period and even to reproduce to some extent there, or whether they find the high temperature of the water so fatal that they soon perish. The latter group — most typically Arctic — has not been represented within the gidf in our midsummer, autumn, winter, or early springhaulsexceptforanoddMertensia 29 off Penobscot Bay on June 14, 1915 (p. 371), though this ctenophore and the Arctic medusa Ptychogena lactea have previously been recorded in Massachusetts Bay and at Grand Manan in September (A. Agassiz, 1865; Fewkes, 1888) ; but in early May of 1915 both of these cold-water ccelenterates, with the large shelled pteropod Limacina helicina and the appendicularian OiJcopleura vanhoffeni, which are equally characteristic of a northern origin, were taken in the eastern side of the gulf at localities where temperature and salinity gave clearest evidence of an influx of the cold Nova Scotian water past Cape Sable into the gulf at the time (fig. 32). Since each of these species was represented by several specimens, their capture just then and there can hardly be looked upon as accidental. As I have pointed out elsewhere (Bigelow, 1917, p. 248), "the appearance of the Arctic Oikopleura in the gulf is especially noteworthy, since it has not been recorded previously on this side of the Atlantic south of Baffins Bay, though known in European waters as far south as the Shetland Islands (Lohmann, 1896 and 1901). Thanks to Lohmann's excellent descriptions and figures (1896, p. 72, Taf. 14, figs. 6, 7, and 10; 1901, p. 15, figs. 16 and 17), it is easily recognized, its chief difference from the closely allied 0. labradorien.sis being the presence of many small dendritic chordal cells. Its very large size (rump length upward of 4 millimeters) is likewise diagnostic, while the red margin of the tail makes it a conspicuous object in the water." " Mertensia occurred over the outer half of the continental shelf oft" Shelburne, Nova Scotia, on Mar. 19, 1320 (p. 371). 60 BULLETIN OF THE BUREAU OF FISHERIES It was for only a brief period, however, that these Arctic animals persisted in the plankton of the gulf during the spring in question, for none of them were captured there during our later cruises (June to October) that year, except for the single Mer- tensia just mentioned; and although Mertensia, Limacina, and OiTcopleura van- hoffmi were all present over or outside the continental shelf abreast of Cape Sable as late as June 24, available data suggest that the planktonic species of this category disappear, from west to east, successively, from the coast water between Cape Sable and Halifax with the advance of the summer, as I have noted elsewhere (Bigelow, 1917, p. 249). Whether the Gulf of Maine is annually invaded by these species is yet to be deter- mined, but what little is known of the seasonal expansion and contraction of the 7V 70" 69* 68" 67* 66* ^ ■• y f Gifkvto f Syf t~U/ MAJ^AN /' y+sj PENStmScpir /bayB^ ^' ,./s,,.,...- $ j/y (r J NOVA * .-^ ( \ SCOTIA 44 ■1- + / '- + +■ 1 + w 9 h / ^y MP Yarmouth t\ -/^tl -N °^l X$\ J| Portland C " w\. 1(1 CAPE ) "(SABLE/ i /' *-» P s f H 43 + *G + \ w 42' + \ nN + + ..•■•• *- *■ UT 7V 70" 69" 68" 87" 66' Fig. 32. — Localities at which certain planktonic animals of Arctic origin were taken in May and June, 1915. H, Limacina helicim, M, MeTtensia ovum; 0> Oikopleura vanhojteni; P, Ptychogeila lactea Nova Scotian current makes this seem probable. Nor does the fact that the more delicate of the Arctic planktonic animals are scarce, if not absent, from the gulf in any given summer mean that no such invasion occurred during the year in question, for Mertensia (A. Agassiz, 1865) is extremely sensitive to water that is too warm. And since, judging from my own experience, this applies equally to Limacina Jielicina. and to the Arctic Oikopleura, it is only while a direct and considerable influx of northern water is taking place around Cape Sable into the gulf (distinguished from the increment it contributes to the general inflowing drift) that they are likely to appear in the catches of the tow nets. Consequently, failure to find them in mid- summer has no bearing on their presence or absence a month or two earlier in the season. PLANKTON OF THE GULF OF MAINE 61 Judging from our cruise during the spring of 1915, they reach their greatest abundance and their widest dispersal in the gulf some time in May. The localities of capture, with what data are available on the currents at that season, suggest that after they have once passed Cape Sable their general line of drift is westward toward the center of the gulf, not northward along the west coast of Nova Scotia, which is the route followed by most visitors from the south (e. g. by Sagitta serratodentata), and that they keep near the surface. Alexander Agassiz's (1S65) discovery of Mertensia and of Ptychogena in Massa- chusetts Bay in early autumn, of Mertensia in abundance at Eastport, Me., in the early sixties of the past century, and Fewkes's (1888) record of the latter as plentiful there in the summer of 1885 and at Grand Manan in July and August, 1886, are contrary to our experience during the period 1912 to 1915; nor does Doctor McMur- rich mention Mertensia at all in his plankton lists for St. Andrews. It is probable that such an abundance of Mertensia and its presence in the inner part of the gulf so late in the season were the visible evidence of a greater influx of northern water past Cape Sable than has taken place at any time during the past decade, and that this inflow turned more northward toward the Bay of Fundy. Unfortunately, however, no record was taken of the temperatures of the gulf during the years in question, and, conversely, no collections were made of the plankton during the abnormally cold summer of 1884. The group of northern animals that better resist high temperature is repre- sented in our catches with some frequency by the two calanoid copepods Calamus hyperboreus and Metridia Tonga, occasionally by a third large copepod, Gaidius tenuispinis, and regularly by the naked pteropod Clione limacina (p. 125). The status of each of these in the gulf is discussed below. I need only add here, of Metridia longa, that while it reaches the gulf chiefly as an immigrant with the Nova Scotian water, it is able to survive there for a considerable period and to thrive "amazingly in their wanderings," says Willey (1921, p. 194), speaking of the species at St. Andrews, in the Bay of Fundy, " if we may judge from their store of oil." Prob- ably, as he suggests, most of them perish eventually in the gulf without leaving de- scendants, and thus, though the animals concerned are diametrically opposite in faunal origin, the distributional status of this copepod within the gulf is analogous to that of Sagitta serratodentata , the specimens that penetrate the gulf as driftage from the north, surviving there long enough to scatter far and wide and to be picked up in the tow net, still flourishing though far from Cape Sable and long after they have passed by it. Metridia longa can not be looked upon as a regular annual visitor to the gulf, for while it has been taken at many stations in some years, in others it has been sought in vain (p. 247). There is some evidence that in the years when it passes west of Cape Sable in greatest number it succeeds in breeding to some extent in the gulf, and the result of its longevity there, coupled with this local reproduc- tion, is that in its years of plenty it becomes so widely distributed that the locality records do not mirror its lines of immigration and of dispersal. For further dis- cussion of this point see page 249. 62 BULLETIN OF THE BUREAU OF FISHERIES The copepod Oalanus liyperboreus affords a second example of an Arctic immi- grant that finds an environment in the gulf favorable for the growth of the indi- vidual and to some extent for reproduction. Its recorded occurrence in the Gulf of Maine illustrates the care with which such data must be analyzed before general conclusions can be drawn from them, for if its Arctic nature were not well estab- lished, the fact that there is a center of abundance for it in the western side of the gulf and a second in the eastern might easily lead one to assume a totally erroneous faunal status for it. In reality it is probable that its comparative abundance off Massachusetts Bay is the result of a certain amount of local reproduction, though replenishment of the stock depends directly on immigration via the Nova Scotian current, as emphasized hereafter (p. 215). The routes by which C. liyperboreus enters the gulf are discussed in the general account of the species. Once past Cape Sable they spread so generally over the gulf that it is impossible to trace their further drift from the actual locality records, probably because the large ody adults, on which most of our records have been based, live long enough to become dispersed far and wide, as well as because of the local production just mentioned. OTHER IMMIGRANTS The indraft of water through the eastern channel and over the neighboring parts of the banks is not only fairly constant in its physical characters but carries with it various planktonic animals as characteristic of this source as those previously discussed are of an Arctic or Tropic origin. They include in their ranks, however, perfectly successful colonists, which, consequently, are also regularly endemic in the gulf (for example, the mammoth copepod Euchasta and the amphipod genus Euthemisto), as well as species that evidently find the gulf a less favorable environ- ment than the Salter and heavier mixed water, as evidenced by their comparative scarcity near shore and the smaller size attained there at sexual maturity. Others, too, are included, which are unable to breed at all in the gulf, though they may live there for some time, in which respect they correspond to S. serratodentata, of the Tropic group, and to L. helicina, of the Arctic category. The influx of this mixed water into the gulf being more or less continuous through- out the year, either via the two channels, Northern and Eastern, or across Georges Bank, the mechanical agency for replenishing the stock of visitors from this source is always available, their life histories and chiefly their seasons of reproduction determining whether they are in evidence in the gulf at any given season of the year. As I have pointed out, Tropic and Arctic visitors are brought into the gulf chiefly in the superficial water stratum, but the whole column of water down to the bottom of the deepest trough of the eastern channel serves as a medium for the dis- persal of the immigrants entering with the mixed water, the precise "sailing routes" (to borrow a nautical term) followed by its inhabitants depending upon the courses of the inflowing water at the different levels at which they live. For the most in- structive animal index to the movements of the surface layers of the mixed water, because the most abundant and conspicuous, we need only refer back to Sagitta serratodentata (p. 58) ; for, although this cha^tognath primarily originates in the Gulf PLANKTON OF THE GULF OF MAINE 63 Stream, it is not direct overflows or influxes of the latter across the offshore banks that maintain the large stock within the gulf during its season of abundance, but the general indraft of mixed water. The euphausiid shrimp Nematoscelis megalops (p. 146), which is less common than S. serratodentata in the inner parts of the gulf but is equally characteristic of the upper strata of water along the continental slope, occupies the same faunal status. The large and easily recognized clurtognath Eulrohnia Jiamata (p. 328) is a characteristic inhabitant of a lower level in the mixed water (say, below 50 meters), though not of the deepest. Its faunal relationship is diametrically opposite to that of its relative, S. serratodentata, for while it is widely dispersed over the ocean basins in the mid-depths, it is only in the Arctic or at least in cold seas that it comes to the surface regularly (Apstein, 1911). It enters the Gulf of Maine by the same route followed by S. serratodentata, but below it, and is equally unable to breed within the gulf, 30 though in its case this failure is because the temperatures it experiences there are too high instead of too low. The eastern channel entrance to the gulf is deep enough to include a part of the vertical zone in which this species is most plentiful in the mixed water over the slope, where it appears in considerable numbers between 100 and 300 meters as well as deeper (p. 329, and Huntsman, 1919) ; hence it is not surprising that it should occur commonly in our deeper hauls in the gulf though seldom on the surface. The vary- ing sizes of the individuals taken there suggest that it is able to "carry on" through- out its natural span of life anywhere in the gulf below, say, 100 meters, though unable to reproduce. Our records do not show the migration routes for Eukrohnia as clearly as they do for Sagitta serratodentata, because the former is a year-round member of the plankton of the gulf. For this reason (coupled, as I believe, with longevity within the gulf), it is to be expected anywhere within our limits below 100 or 150 meters and at any season, though the extreme southwest corner of the deep basin off Cape Cod and also certain isolated sinks to which its access is more or less obstructed, may prove exceptions to this rule. If all our records of Eukrohnia for all seasons are united, however, there is a decided preponderance in the eastern, and particularly the ex- treme northeastern, parts of the gulf contrasted with its western side, not only in the number of stations at which it has been taken but also in its local abundance, which agrees with the general anticlockwise direction of the inflowing eddy. The distribu- tion of Eukrohnia (p. 328) illustrates how closely its inward route follows the Eastern Channel and the slope of Browns Bank. Although Eukrohnia is a constant con- stituent of the plankton all along the seaward slope of Georges Bank, the latter must by its shoalness, oppose an absolute barrier to its dispersal, for we have not found a single specimen at any of our stations on the bank at any season. Consequently, none of the Eukrohnia that have passed the mouth of the Eastern Channel as they drift westward can enter the gulf on their farther journey. Finally, I may point out that the regularity with which Eukrohnia appears in the gulf is as good evidence 30 Although Gulf of Maine specimens are oiten large, we have found none there with sexual organs developed. 75898—26 5 64 BULLETIN OF THE BUREAU OF FISHERIES as the salinity and temperatures that its native water is a large if not the major constituent of the inflowing current, for it is not abundant even along the continental slope (p. 333, and Huntsman, 1919). The cold-water siphonophore Diphyes arctica, which occasionally penetrates the Gulf of Maine (p. 379), does so at about the same level as Eukrohnia (about 50 to 150 meters), and it is probable that, like the latter, it journeys with the mixed water, in which we have found it over the slope off Shelburne both in March and in June and off the slope of Georges Bank in July, but not along the Nova Scotian coast. The Eastern Channel is, no doubt, the route by which it enters the gulf, judging from the concentration of the localities of capture along the eastern slope of the gulf basin in March and April, 1920. The ultimate origin of D. arctica is not clear as concerns the Gulf of Maine, for while it was formerly supposed to have been one of the most charactersitic of Arctic indicators, captures of it by the Gauss in deep hauls off Cape Verde (Moser, 1915) suggest that it may also range widely in the cold mid-layers of more southern seas, just as Eukrohnia does, and thus reach the gulf from the inter- mediate depths abreast its mouth. Sagitta maxima, the largest of local chastognaths, is perhaps the most useful animal indicator of the deepest stratum of the water entering the gulf via the Eastern Channel, both because its habitat is well known offshore, and because it neither breeds in the gulf nor can long survive there, being unfitted for life in water of low salinity no matter what the temperature (Huntsman, 1919, p. 433). S. maxima is so closely con- fined to depths of 150 meters or deeper, both in the Gulf of Maine and in neighboring parts of the Atlantic Ocean, that its presence anywhere in the inner parts of the gulf is unmistakable evidence of the existence of an inflowing current then, or shortly previous, and close to the bottom of the trough. The locality records for S. maxima are concentrated correspondingly in the Eastern Channel, in its immediate debouehe- ment into the general basin of the gulf, and thence northward along its eastern trough as far as the Grand Manan deep, on the one hand, and in the deepest part of the western basin, on the other. As might be expected from its faunistic status, S. maxima is no more periodic (seasonally) than Eukrohnia in its occurrence in the gulf; but although specimens drift in more or less constantly throughout the year, it has invariably been so sparsely represented in hauls made within the gulf, contrasted with considerable abundance at 200 to 300 meters along the continental slope to the east and north, that the indraft can tap only the uppermost levels of its natural habitat offshore at any season. The lines of dispersal followed, respectively, by Sagitta serratodentata, Eukrohnia, and S. maxima within the gulf correspond closely with the dominant drift of water at as many levels — that is, surface, mid, and deepest — as made evident by the physical data afforded by temperature and salinity and by drift bottles. Thus, while S. serratodentata not only spreads widely over the offshore parts of the gulf in its season, it also sweeps right around the coast to Massachusetts Bay (which apparently serves more or less as a cul-de-sac for it, as it has for certain drift bottles released in the Bay of Fundy), and Eukrohnia has much the same distribution except that it lives PLANKTON OF THE GULF OF MAINE 65 so much deeper that it is prevented from entering Massachusetts Bay by the contour of the bottom, and, in fact, hardly encroaches at all on the shallow coastal belt within the 100-meter contour. Furthermore, the two agree in their scarcity in the Fig. 33.— Chief routes followed by planktonic immigrants entering the Gulf of Maine at different levels. \\\, immigrants at the surface; 111, immigrants at intermediate levels; =, immigrants at the deepest level southwestern part of the basin of the gulf— that is, just where the physical data, to be discussed elsewhere, locate the "dead water" in the anticlockwise eddy that occupies the gulf. However, S. maxima, living in the deepest waters of the basin, must follow 66 BULLETIN OF THE BUBEAU OF FISHEKIES its two diverging troughs, in both of which there is a dominant though perhaps not a constant indraft along the bottom, the result being that while its route parallels those of the two preceding species in the eastern part of the gulf, it crosses below them at a lower level in the western, an interesting phenomenon illustrated in the accompanying chart (fig. 33) . No doubt this applies in general to the three bathy- metric groups which these three chretognaths typify. The possibility that visitors may occasionally penetrate the gulf from the mid- depths of the Atlantic basin below, say, 300 meters, deserves a word. The successive deep-sea expeditions, from the Challenger in 1872 to 1876 down to the Michael Sars in' 1910, have found an abundant and varied pelagic fauna in the Atlantic below the level to which strong sunlight penetrates. Generally speak- ing, the adults of this community live well below 200 meters (many of them chiefly below 400 to 500 meters) and many of them are characterized by a peculiar coloration. Thus, those dwelling so deep that red light reaches them feebly, if at all, often exhibit a very dense pigmentation (Hjort, 1911 and 1912; Bigelow, 1911a), many fishes of this category being black with phosphorescent organs, decapods dark red, and medusas either of a beautiful, translucent, deep claret color or opaque chocolate, tints quite unknown among jellyfishes in shallow water. This extreme development of pigment is so characteristic of this whole faunal group that the latter is often referred to as the "black fish-red prawn" community. At a higher level (that is, in the zone between 150 and 500 meters, but neverthe- less below the reach of the wide diurnal fluctuations in illumination to which the surface waters are subject) there exists an entirely distinct series of fishes of quite different aspect, which as a rule are "laterally compressed, with a mirrorlike silvery skin; when colored, the back is generally blackish brown, and the resplendent mirror- like sides of the body blue or violet. The eyes are large, very often telescopic, and the body i s provided with a number of light organs" (Hjort, 1912, p. 628). They are accompanied by sundry medusas, which parallel them in their pale pigmen- tation but brilliant iridescence, as I have pointed out elsewhere (Bigelow, 1911a, p. 6). It is a fortunate chance for the oceanographer that many of the bathypelagic animals are so distinctively colored, because their presence in any numbers any- where in shoal water over the continental shelf would be the best of evidence of an upwelling of Atlantic water from the mid-depths or deeper, a type of oceanic circulation that has evoked considerable discussion as a possible factor in maintain- ing the low temperature of the coastal waters off the eastern United States. Conse- quently, the presence or absence of the black fish-red prawn community within the Gulf of Maine is a question of some moment, and it is in the hope of encouraging others to keep a sharp lookout for it there that I have devoted the preceding lines to the general appearance of its members. No doubt this planktonic community is represented at the appropriate level all along the continental slope off the United States, for it occurs generally over the whole Atlantic basin from high latitudes to low. We encountered it over the 1,500-meter contour off Cape Sable on March 19, 1920 (station 20077), the following being a partial list of its more noticeable repre- sentatives in hauls from 500 and 800 meters: Several black lantern-fishes (genus Myctophum) ; a specimen of the curious deep-sea snipe eel (Serrivomer beanii) , 45 PLANKTON OF THE GULF OF MAINE 67 centimeters long; 32 the wine-red medusa Periphylla hyacinthina; 13 specimens of its chocolate-colored relative JEginura grimaldii; the iridescent medusae Halicreas papillosum and Rhopalonema funerarium; and many red prawns; side by side with the ehsetognaths Eukrohnia and Sagitta maxima, the large copepod Euchseta norvegica, and the euphausiids Nematoscelis and Thysanoessa, besides boreal animals such as S. elegans, Tomopteris, Limacina balea, and Calanus. Scanty though the catch just listed is, compared with the abundant pelagic fauna that has been encountered by the National, the Valdivia, and the Michael Sars at many stations in the North Atlantic, and by the Albatross on many occasions and in localities in widely separated parts of the Pacific, it is the only one in which the black fish-red prawn community has been represented by more than an occasional example even at our outermost stations, though we have towed down to 400 meters or deeper at several other localities off the slope abreast of the Gulf of Maine in February, May, June, July, and August. In fact, to complete our list of captures of this category I have only to add two genera of fishes (Cyclothone and Myctophum) and one red medusa (Atolla) from 750 meters off the southwest face of Georges Bank, February 22, 1920 (station 20044); a few black fish and bathypelagic medusae (JEginura) from 1,000-0 meters southeast of the bank three weeks later (March 12, 1920, station 20069); a scattering of bathypelagic fish (mostly juvenile wSternop- tychids and Myctophids) at our summer stations along the same zone off the bank in June and July, and off Cape Sable. With bathypelagic animals so scarce in the cool water that washes the continental slope abreast of the Gulf of Maine, and with both the Eastern Channel (the bottle- neck through which, alone, the deeper strata of oceanic water flow into the gulf) and the basin into which it debouches considerably shoaler than the levels at which they attain their maximum development offshore, it would be surprising to find any of them in the inner parts of the gulf except as the rarest of stragglers. As a matter of fact, our cruises have yielded only two such records — viz, one Cyclothone signata 23 millimeters long on Browns Bank, station 10296, June, 1915, and a muti- lated specimen, probably of this same species, taken in an open-net haul from 180 meters in the Fundy Deep on March 22, 1920. Nor have other students been more successful in this respect so far as I can learn. Thus it is evident that members of this community occur only accidentally within the limits of the gulf, for did they enter the latter as often even as the tropical animals discussed above, they would have been sure to attract attention in the tow net by their striking appearance. In short, the plankton of the gulf receives practically nothing from the deeper layers of the Atlantic at any season. Even the most temporary invasion on their part would be so important an event, both faunistically and hydrographically, that sharper and more constant watch should be kept for them in the gulf than their rarity there would warrant otherwise. The several Tropic and Arctic visitors and immigrants from the continental slope touched on above illustrate the less successful degrees of colonization, ranging from utter failure in the cases of sporadic visits of exotic tropical animals and the equally » For a description of this eel see Goode and Bean, 1896, p. 155, fig. 168. It is not included in the report on the fishes o( the Gulf of Maine (Bigelow and Welsh, 1925), because the localities of record lie outside the limits covered therein. 68 BULLETIN OF THE BUREAU OF FISHERIES short-lived incursions by the more delicate Arctic forms, to the more successful though equally temporary immigrations by animals that are able to survive under the physical conditions which they encounter in the gulf and even to grow there, but not to breed; such, for example, as Sagitta serratodentata and Eukrohnia. The next step toward successful colonization would be the ability to breed in the gulf in small numbers or during especially favorable years, which would still leave the species concerned dependent on immigration from prolific centers elsewhere for the main- tenance of the local stock. In the nature of the case instances of this sort are difficult to demonstrate without intensive and long-continued studies of the plankton, but it is evident that the copepods Calanus hyperboreus and Metridia longa both fall in this class (p. 61) ; also the curious pelagic worm Tomopteris catharina, the continuous and rather common occurrence of which in the gidf and its wide dispersal there depend chiefly on immigrants of northern origin (it is a north-boreal form), for while it breeds in the gulf in some summers it fails to do so in others (p. 338). It is probable, also, that the large naked pteropod Clione limacina has this same faunal status, breeding in sufficient numbers for the local production, coupled with individual longevity, to give it a uniform distribution over the gulf and so to obscure the routes followed by the immigrants from colder waters east and north of Cape Sable, on whose visits its continuous presence in the gulf equally depends (p. 127). The amphipod genus Euthemisto stands a rung higher on the ladder of pro- gressive colonization, for it neither breeds so abundantly (though it does so regularly) in the gulf nor grows to so large a size there as it does over the outer edge of the offshore banks — Georges and Browns (p. 158). Local fluctuations in the abundance of animals of this status throw no direct light on their waves of immigration, being due, as often as not, to local centers of reproduction within the gulf itself and even close up to the land, such as we have occasionally encountered for Euthemisto (p. 160) ; but greater abundance in the eastern part of the gulf than in the western, especially if coupled with prolific centers of reproduction in the zone of mixed water over the outer part of the continental shelf abreast of it (and this is true of Euthe- misto) , shows that the stock produced within the gulf receives frequent accessions to its numbers from outside. No doubt one or other member of the plankton might be found to represent every conceivable intergradation from utter failure to perfect success in colonizing the waters of the Gulf of Maine (for all members of the plankton are colonists in the last analysis) were the known record sufficiently complete. The copepod genus Euchseta, for example, may be taken as representative of animals that breed indifferently and grow equally large along the continental slope, in the Eastern Channel, and in the gulf wherever the depth is sufficient, as proven by the occurrence of sexually adult males, of females with large egg clusters, and of juveniles. For this copepod the gulf basin is simply a diverticulum from its general geographic range. Most successful of all are those that find a more favorable environment in the inner parts of the gulf than in the waters immediately tributary to it, and it is to this group that such members of the local zooplankton as the copepods Calanus f.n- marchicus and Pseudocalanus elongatus and the chastognath Sagitta elegans belong. It is true that most, if not all, the animals of this category have equally prolific centers of PLANKTON OF THE GULF OF MAINE 69 abundance elsewhere (chiefly to the eastward and northward), connected with the gulf by a continuous zone of occurrence, but all of them are regularly more abundant in the particular temperatures, salinities, densities, etc., that characterize the Gulf of Maine than immediately outside it, whether to the east or the west or offshore. Indeed, such multitudes of several of these species (Calanus, especially) arc pro- duced there that the small accessions which the gulf may receive from the north must be far outnumbered by the emigrants that emerge from it to journey either northward along the inner edge of the continental slope, on the one hand, or around Cape Cod to the westward and southward over the outer part of the continental shelf, on the other. It is probable that the boreal winter plankton of the coast water south of New York draws more from this source than from local production. MIGRATIONS OF PELAGIC FISH EGGS AND LARWE One of the most interesting and economically important fields of study to which our Gulf of Maine explorations are introductory is the involuntary migrations of the early stages of fishes, with the effects of such journeyings on the fish population of different parts of the gulf. Any information obtainable on this subject is instructive from the point of view of the migration of the plankton within the gulf, because every buoyant fish egg floats from spawning until hatching, wherever the current may carry it, rising or falling vertically according to specific gravity of the water only, with the young larvae equally at the mercy of tide and current until after the yolk sac is absorbed. Even the older pelagic fry of most fishes are hardly less helpless, so far as voluntary horizontal migration is concerned, until they attain considerable size (some species become contranatant — that is, turn to swim against the current — at an early stage), even though they are able and do swim up and down and thus exercise a choice of level at which they live. Now the water of the open sea never being at rest (no area as large as the gulf lacks some dominant movement, if not a definite current, in one direction or another), it follows that only in the rarest instances does a fish hatched from a buoyant egg ever grow large enough to descend to the bottom in the precise locality where the egg that gave it birth was spawned. The drift during its pelagic life may be only a few miles if spawning occurs in some bay or sound sheltered from the free circulation of the sea by off-lying islands; it may, indeed, be almost nil in this case, should the tidal currents in the two directions be of equal strength. Outside the outer head- lands, however, the journeyings of floating fish eggs are, generally speaking, so considerable that they are often measured better by degrees of latitude and longitude than by miles. Such, to quote only a couple of the more striking and better known examples, is the case with the cod eggs spawned south and west of Iceland, for most of the fry resulting therefrom drift right around to the north and east coasts of the island before they seek the bottom (Schmidt, 1909). Off Norway, too, cod eggs and fry have long been known to carry out long journeys with the current (Damas 1909a; Hjort, 1914). Indeed, events of this sort are inevitable, given the indicated factors of animals able to swim but weakly, caught up in the set of any current. 70 BULLETIN OF THE BUREAU OF FISHERIES Extensive migrations of fish eggs and of young fishes, in fact of all the plankton, are therefore to be expected as characteristic events in the Gulf of Maine with the dominant anticlockwise eddy that governs its circulation — not their occurrence, but their absence would cry for explanation. And so interesting is this question, and so directly does it bear on the practical problems of the fisheries, that it deserves passing notice, even granted that we can not yet outline the travels of so much as a single species of fish in the gulf. No matter how little related the various species are, it is justifiable to consider as a unit all fishes that are subject to similar influences during their pelagic lives, the precise routes they follow at this early age depending not on themselves but on the locations and times of year where and when their eggs are spawned, in relation to the circulation of water in the gulf, and on the duration of the pelagic stage as govern- ing the length of time during which they drift before they abandon this nomadic life for a more stationary habitat on or near bottom. Several of our gadoid and flat fish are particularly suitable for such a combined survey, because while they do not spawn on precisely the same grounds or at just the same seasons, cod, haddock, silver hake, and such common flounders as plaice, dab, and witch, agree in breeding only in the peripheral belt of the gulf and on the offshore banks, seldom, perhaps never, in its central deeps outside the 200-meter contour. As the composite chart (fig. 34) shows, buoyant gadoid and flatfish eggs of one kind or another have been found all around the coastwise belt of the gulf, likewise widespread on Georges and Browns Bank, the richer clusterings of egg records mirroring the greater number of hauls made at particular localities rather than any demonstrable preponderance of eggs as compared with the intervening stretches. If there were no dominant drift of current in one direction or the other, but only the tide to disperse the eggs in these shoaler parts of the gulf, the distribution of the. larvae would simply parallel that of their parent eggs ; but year after year and voyage after voyage we have come to see more and more clearly that such is not the case, but that the young pelagic stages of the cod and flounder families are much less plentiful in the northeastern corner of the gulf than in its southwestern waters in general or in the Massachusetts Bay region (fig. 35) in particular. The considerable number of towings carried out along the coast of Maine from spring until autumn, in 1915, fairly rule out the possibility that the discrepancy in distribution between eggs and fry is only apparent and results from an imperfect record. To suppose that the same nets would catch young fish in Massachusetts Bay and as consistently miss them off Mount Desert and to the eastward is absurd ; nor can the depths of the hauls be made responsible, seeing that we have towed at various levels, surface to bottom, as well as vertically, at many stations along the coast. A difference of this sort between the locations where the eggs are spawned and where the resulting larvse are to be found is not a novelty, for Petersen (1892) long ago reported a precisely similar phenomenon for Danish waters. In short, I am convinced that the scarcity of larval and post-larval fishes in the one corner of the gulf as contrasted with their abundance in the other is real. It is, of course, possible that the northeast part of the gulf is so ill fitted for a fish nursery that only a small proportion of the pelagic eggs spawned there ever PLANKTON OF THE GULF OF MAINE 71 hatch or the resultant larvae survive. The researches carried on during the past few years at the Canadian Biological Laboratory at St. Andrews point unmistakably to the conclusion that few if any floating eggs of any groups of animals hatch success- 71" 70" 69" 68' 67' 66' i t 1 1 *\v ) / .A CjfAN'D f /sf ruJ fAyAN / A^y V-C*« ^75h**J// • •■''■ ^ A .[ £ 1 NOVA SCOTIA 44' *■ + 44" ^W . v" ! Yarmouth Portland C .•*? * V - A "' |fl CAPE y ~J .* ! • U /SARLE< y'" . V* • r -f • .'i • i m H *■ ""V • J>'* * (y-^... y, •• ! / 4t ♦ / • ••• * + r , + * '"+. ta • V* v. N • /?• \ ~ "-v» • v /■■ «i •*' » \ / jf\^ '• '■ O ^"^""V •• •• - • * ^ams"" "">**« •■ \ -'.;' + <* *^rf + + .-■"* *- \ 4- « ^*\ u\ • • •'■•. r v^ ^s^i ..,- ~?y\ -n - ; « • \< y • «' * + + + + 4- / . Of • ^ • ..' ..-■ .•-x ,■•-" •..-.... -"" ""--■' */~~ . 40' 4 4- + 4- 4- 4- '.'j- 71* 70* 69" 68' 67" 66" Fig. 34. — Locality records for buoyant flounder (pleuronectid) and gadoid eggs combined (a dot for each record of each species), 1912 to 1922 fully in certain parts of the Bay of Fundy, this being particularly true for chretognaths and fishes (Huntsman, 1922; Huntsman and Reid, 1921). As evidence of the un- suitability of the bay as a breeding ground for fishes with buoyant eggs, Huntsman 72 BULLETIN OF THE BUREAU OF FISHEKIES (1918, p. 65; 1922) offers the extraordinary rarity of the larvae, for example, of the plaice (Hippoglossoides), witch (Glyptocephalus), cod, haddock, hake (Urophycis) , or pollock (Pollachius virens), although the adults of all of these are plentiful there; 71* 70* 6 9' 68' 67* 66 - Wfe 1 ) y « ->^gtQnd :/ Jf 44 PEN^BS-COlr /V) " « J? 10 J] •»! ,-' 4- V^y NOVA ( SCOTIA j + 44' (Yarmouth Portland C .; ^ : • In CAPE \ "(SABLE/ J * ? **/7 "■■-"'\ • ! 43 US / c^i .. Q 3^ * ) + ^' •*%)> . + + .. 7 /~? CAPERS-?' X ■■'''"' gW*£*~ + lOOMcters _ • 4- r x • \ : \ 4- i 4r «■ ^° v^ 1 ( ' + + + + + / + 4V • • • • • • • .'"""■*■. • ..■*-"" .,■-'" — \ 40' + -i- + +. 4- 4- V 71' 70' 69" 69* 67' 66' Pig. 35. — Locality records for flounder (plcuronectid) and gadoid larvae (a dot for each record of each species) to illustrate the probable drift of buoyant fish eggs and larval fishes all, in fact, spawn in the bay, for cod and plaice eggs have been recognized there in the plankton (Huntsman, 1922), and floating fish eggs of some species were noted by Doctor McMurrich as occurring occasionally during January, February, April, and early May, and regularly thereafter until the end of August at St. Andrews. PLANKTON OF THE GULF OF MAINE 73 Taken by itself, the absence of larvae, contrasted with the presence of eggs, could as well result from a drift of the latter out of the bay before hatching — such, indeed, as the circulation of water would call for — as from their failure to hatch locally or of the larva? to survive. But there are two objections to this view, to my mind unan- swerable; first, that larva? and young fry of these several species are fully as rare along the eastern shores of Maine — that is, in just the waters into which the outflow from the bay debouches — as within the latter; second, that the drift into the southern entrance of the bay would naturally bring with it gadoid and flatfish eggs from the shallows off western Nova Scotia. Some of the cunner (Tautogolabrus) larva? produced in St. Marys Bay, which Huntsman (1922) has found to be an important site of repro- duction for this fish, must likewise find their way into the Bay of Fundy either around Brier Island or through the passages; but so few of them survive the conditions they encounter in the Bay of Fundy, that none have been recorded from all the winter and summer towing which has been done from the St. Andrews station. Most of the common fishes that do succeed in breeding in large numbers in the bay lay demersal eggs; for instance, the several sculpins (Cottida?), the lumpfish (Cyclopterus), the rock eel (Pholis gunnellus), the winter flounder {Pseudopleu- ronectes americanus), and the herring. The rosefish (Sebastes) and the eelpout (Zoarces), which are viviparous, produce young far advanced in development. The evidence just summarized justifies the hypothesis that while young fish hatched in the bay from demersal eggs, or such as are far developed as to size and fins at hatching, thrive there, most of the very small and helpless larva? produced in the bay from pelagic eggs, or which enter it as immigrants from the south, perish. Hence we may speak of the Bay of Fundy as a deathtrap to buoyant eggs and larva? drifting northward along the eastern shores of the gulf, and it contributes none of these to the coastal waters to the westward. Even the very abundant stock of young herring produced about the mouth of the bay (notably at Grand Manan) do not spread far to the westward, Huntsman having found that they soon become contra- natant and begin to work back against the current, which takes them out of the planktonic category. An understanding of the causes that prevent successful development in the bay would make it possible to estimate the probable suitability, from east to west, of the waters along the eastern coast of Maine, where eggs are certainly produced in some abundance but where few larva? have been taken. Huntsman (1918) suggests the violent tidal stirring in the bay as responsible, by preventing vertical strati- fication of the water. The low surface temperature may also be an effective check to species such as the cunner, which spawn in high temperatures. Neither of these factors, however, would seem likely to interfere with the successful breeding of late autumn, winter, or spring spawners — the American pollock and the haddock, for instance. Further light on this interesting question, to which our own work has contributed nothing, is to be expected from the investigations now being carried out at St. Andrews by the Biological Board of Canada. From Mount Desert eastward the coastal belt of the gulf more and more closely approximates the Bay of Fundy hydrographically, owing to the increasing strength of the tides and the consequent activity of tidal mixing. Correspondingly, 74 BULLETIN OF THE BUREAU OP FISHEBIES the general neighborhood of Mount Desert Island is the most easterly location along the northern shores of the gulf where we have found gadoid or flatfish eggs in any numbers. The rather uniform transition in the state of tidal mixing, with its consequent effect on salinity and temperature, which characterizes the coastal belt from the Bay of Fundy to Casco Bay, indicates an improvement from east to west in condi- tions for buoyant fish eggs and larvae; but outside the outer islands 33 salinities and temperatures vary so little from Penobscot Bay westward and southward to Massa- chusetts Bay, especially during winter and spring when most of the more important gadoid and flatfish species spawn, that there is nothing in the physical state of the water to suggest one part of this zone as notably more suitable for their successful reproduction than another. With the dominant set of the water tending to drift all fish eggs and larvae produced along the northern shores of the gulf toward the west and south, and with few or no accessions coming from the east to the coastal zone between Mount Desert and Cape Elizabeth because of the sterility of the Bay of Fundy in this respect, tows there might be expected to take eggs and very young larvae, but seldom older ones or the post-larval stages. Actually, most of our tow nettings there have yielded eggs alone (fig. 34) ; but the larvae hatched from buoyant fish eggs are so small and soft until two weeks or so old that they are apt to be mashed past recog- nition amongst the mass of other plankton, hence may very well have been over- looked, and by the time they are large and resistant enough to be noticed among the hard-shelled copepods, etc., they may have drifted for a considerable distance. Mavor's (1920 and 1922) recent experiments with drift bottles give some idea of the actual speed with which the surface water, and consequently the fish eggs and larvae floating with it, may travel westward and southward around the gulf, indicating that a drift of about 4 nautical miles per day is not unusual in summer and autumn, although more or less intermittent. The rate is probably higher than this during the spring. On this basis, buoyant eggs spawned off Mount Desert Island and far enough out from the land to be caught up in the general peripheral eddy of the gulf (how far this means is not yet known) might drift well beyond Cape Elizabeth during the two weeks interval that may be set as a fair average incubation period for gadoids and flatfishes in general in Gulf of Maine temperatures. Whether the eggs actually equal the drift bottles in the speed of their journey depends on whether they float at the same level — that is, in the upper two meters or so. Many of them, and perhaps most, taking the year as a whole, do so; but locally, and especially when the surface is at its lightest after the river freshets, many eggs float deeper down where the dominant drift probably is slower, notably those of the haddock, which is spawning actively at that season (Bigelow and Welsh, 1925). During the interval after hatching, when the larvae are so small that they are seldom recognized in ordinary tow nets, the small proportion of them that survives the vicissitudes of pelagic life very likely drifts another 50 miles or so, so that Mount » Low surface temperature close in along the land between Penobscot Bay and Casco Bay in summer may be a bar to the local breeding of the cunner, though this would not apply up the many estuaries that indent this section of the coast. PLANKTON OF THE GULF OF MAINE 75 Desert fish may well reach Massachusetts Bay in their journey by the time they are 10 to 15 millimeters long, if they remain in the superficial water layers. If they sink to lower levels, as it is practically certain that many of them do, their involuntary migration during this stage probably is not so extensive, there being reason to believe that the general set is more rapid above than below 40 to 50 meters ; but whatever depth they seek within the 100-meter contour (which in general limits the offshore dispersal of both eggs and larvre in this side of the gulf), the majority of them will tend in the same general direction. Similarly, the larva? hatched from buoyant fish eggs spawned off Machias, where considerable numbers are produced, might well travel as far as Cape Elizabeth before attaining the sizes we have recog- nized in the tow nettings. The distribution of the buoyant eggs of the cod and flatfish families in the gulf bears precisely the relationship to that of the older larval stages (fig. 35) which involuntary migration of this sort would produce. In fact, something of the kind might safely have been prophesied from what is known of the circulation of the gulf; and I believe it safe to assert that the great majority of the larval fishes hatched from buoyant eggs spawned in the zone from 10 miles or so outside the outer islands out to the 100 or 150 meter contour, between Cape Elizabeth and the Bay of Fundy, drift a greater or lesser distance around the periphery of the gulf toward the west and southwest (if they survive as long as three weeks or a month), though this drift may be interrupted or even reversed on any given day or over a period of several days. They may tend to hug the coast, as it seems Mavor's (1920) first series of drift bottles did in 1919 (this probably is the usual event in spring), or swing more offshore, and so, if they live pelagic long enough, come around to the northeastern corner of the gulf as other drift bottles released in the summers of 1922 and 1923 have done. The variations in the dominant set are not well understood, but in any case they will tend to follow an anticlockwise and eddying course. Thus, fish eggs and larva?, and for that matter every member of the plankton, animal or vegetable, tend to follow the same peripherical migration zone as do the immigrants that enter the eastern side of the gulf in the upper 50*meters (p. 64). Only such buoyant eggs as are spawned among the islands, in bays, or close in along shore (as most of the dinners are) are likely to escape this dominant set. At the times when the dominant drift of the surface water follows the coast line closest, south toward Cape Ann, Massachusetts Bay probably acts to some extent as a catch basin for all sorts of flotsam from the north, living, of course, as well as dead, as it did for certain of Mavor's drift bottles. The chart (fig. 35) sug- gests that larva? that pass Cape Ann tend to be caught up in the back water of the bay, to remain there until they abandon the pelagic life for the bottom. Thus, it is probable that the rich fish fauna of the bay and its adjacent waters is regularly recruited from the north and east. Similarly, the abundant occurrence of young pollock at Woods Hole in late spring (fry so small that they are evidently the product of the previous winter's spawning) is clear evidence of a migration southward along and around Cape Cod from the very productive spawning grounds at the mouth of Massachusetts Bay, 76 BULLETIN OF THE BUREAU OF FISHERIES because no important spawning is known for this fish south of the Massachusetts Bay region (Bigelow and Welsh, 1925). There is no evidence that the larval stages of the cod or flatfish families acquire a contranatant (that is, up-current swimming) habit, as the herring does. Conse- quently the extent of their involuntary journeyings depends on the duration of the pelagic stage as much as on the velocity of the drift with which they travel. Very little information has been gathered on this in the Gulf of Maine, but in north European seas both the American pollock (PoUachius virens) and the haddock are pelagic for about three months; most of the cod hatched in the Gulf of Maine prob- ably are so for at least two months, if not longer, before they take to the bottom. So far as the elapsed time goes, experience with drift bottles suggests that this may be long enough for some of them to make the entire round of the gulf — that is, from off Mount Desert or Penobscot Bay around to the Bay of Fundy — but whether any of them actually do so is not known. The extent of the actual drifts of different species would be governed largely by the levels in the water at which the larvas live. Schmidt's (1909) classic and oft-quoted study of the distribution of cod and American pollock (Pollachius virens) eggs and fry around Iceland illustrates how far apart the fry of different species, hatched from eggs spawned in the same general regions, may travel before abandoning their pelagic life, if living at different levels and pelagic for different lengths of time. The two fishes in question spawn at the same season (maximum egg production about April), and both of them mainly, if not exclusively, off the southwest and south coasts of the island, while the fry of both show a tendency to drift thence westward and northward. But while the American pollock mostly descend to the bottom in practically the same waters where spawned, either because their span of pelagic life is short or because living at such a level that they drift slowly, the young cod generally travel right around the island (a trip of something like 500 miles for many of them) , and the result is a scarcity of the youngest bottom stages on the south and west but a great predominance of them over those of the pollock off the northeast and east coasts. The Icelandic haddock likewise perform; a similar involuntary migration, enduring from May until July. The great abundance of young pollock only a few inches long along the littoral zone in the Gulf of Maine suggests that the involuntary drift of the pollock is also shorter with us than is that of cod or haddock. Here, again, definite evidence, one way or the other, is lacking for want of systematic towing during January and February. Very few definite observations have been made on the depths at which the various young fish live while pelagic in the Gulf of Maine, and it is not safe to assume that these will be the same as in the northeastern Atlantic, the vertical distribution of temperature and of salinity being different. It is probable that the young pollock frequent the surface layers more than either cod or haddock (except for such of the latter as live commensal with medusae), this being the case in European waters; but the involuntary migrations of the Gulf of Maine pollock take place in winter when the circulation of the gulf is believed to be at its minimum. Drift bottles released during the period from January to March would be extremely instructive in this connection. On the whole, the drifts of young cod may be expected to follow PLANKTON OF THE GULF OF MAINE 77 deeper, and of young haddock still deeper currents, but to what extent this differen- tiates the dispersal of their fry in the gulf from those of the pollock can not be stated until a sounder knowledge of the circulation of the waters of the gulf has been gained. It has long been known that the larval and post-larval stages of the hakes (genus Urophycis) are apt to be right at the surface in the Gulf of Maine in summer. They might therefore be expected to follow very closely the tracks of the drift bottles released at that season. Silver-hake (Merluccius) larvas, on the contrary, which are among the most abundant of young fishes in the southwestern part of the gulf in July and August, usually have been taken in hauls from 40 meters or deeper (seldom at the surface), and it would seem that they must -therefore travel with the under- current. In the case of silver hake it is not improbable that some of the larvae that journey down past Cape Cod drift on past Nantucket Shoals toward the south- west. Consequently, eggs spawned in the Gulf of Maine may contribute to the fry found west of Nantucket in summer, though most of these are the result of local propagation (Bigelow and Welsh, 1925, p. 395). It is equally possible that part of the young silver hake circle eastward over the northern part of Georges Bank, and so northward into the gulf again, for drift bottles released on a line running southwest from Cape Cod have shown a division in this respect, many of the outer ones having gone westward and some of the inner ones eastward, but we have found no Merluccius larvae in any of our July towings over the banks, although they are abundant off Cape Cod during that month. I have previously (Bigelow, 1917, p. 279) suggested the possibility of a passive migration of cod and haddock from the western part of the gulf out onto Nantucket Shoals and to the western parts of Georges Bank, where we have since found young haddock in some abundance floating commensal with medusae in July (Bigelow and Welsh, 1925). The drift of the haddock eggs that are spawned in enormous numbers on the eastern part of Georges Bank in spring (p. 37; and Bigelow and Welsh, 1925, p. 439), and of the resultant larvae, is a question of great interest. A considerable propor- tion of these may take to the bottom on more westerly parts of the bank, because the northern part of this spawning ground seems to be affected directly by a set from the northeast during the critical season; but at the time of our Marcli and April visits thither in 1920 the presence of newly spawned eggs in abundance right out to the 1,000-meter contour proved that a drift out to sea was then taking place from the southern point of the bank. Eggs subject to this drift must suffer one of two fates. Probably they would be caught up in the band of cool mixed water along the continental slope, in which case the eggs and larvae might again be swept in on the shelf somewhere to the westward by some incurving swirl in the complex interaction of warm and cold waters, or, circling to and fro, come in again on Georges Bank. If they drifted farther offshore, but still not far enough out to reach water of fatally high temperature, they would probably tend to travel to the northeast. Therefore, as Doctor Huntsman suggests in a recent letter, it is possible that the Georges Bank spawning ground, which is AM ■"•§ '^&W j £*- ™™rv S HQ.LE QCEANOGRAPHIC I N S~ 1TUT ION WOOD* H«V« £"!' M.^e. CHUBS'- 3 ■ 78 BULLETIN OF THE BUREAU OF FISHERIES certainly one of the most important off the American coast, may even contribute to the fish stock of the Grand Banks. Haddock or any other bouyant eggs spawned on Browns Bank, or German Bank to the north of it, would probably tend either northward into the gulf or west- ward toward Georges Bank, depending upon the precise state of the Nova Scotian current at the time; and it is probable that this was the source of the cod-haddock eggs towed over the eastern side of the basin on May 6, 1915 (station 10270), and on April 17, 1920 (station 20112). Larva? hatched on Browns and German Banks might be expected to follow the same route during the spring, if living at about 40 to 50 meters, which it is probable that most of them do. Eggs spawned on Browns and German Banks after the rush of water past Cape Sable has slackened, would be more apt to be drifted northward toward the Bay of Fundy, but this would apply mostly after the spawning season of the haddock had passed. It is obvious that if practically no production of the species of gadoids and flatfishes that lay buoyant eggs takes place in the Bay of Fundy, and if most of those produced along the northern side of the gulf drift away to the southwestward, as the evidence marshalled above seems to prove, there must be as regular an immigra- tion of the older fry back again to maintain the stocks of adult fish. However, this subject does not immediately concern the plankton. It is interesting to compare the chart of gadoid and flatfish fry (fig. 35) with the corresponding chart for the rosefish (Sebastes), a viviparous species (Bigelow and Welsh, 1925, fig. 120), as an illustration of the degree to which the dispersal of larval fishes depends on the precise locality where they are produced. In the case of the former this happens chiefly inside the 100-meter contour, with the result just described. No doubt, when young rosefish are born in that belt and chance to rise near the surface they follow the same route, journeying with the dominant set. But rosefish also produce their young generally over at least the northern half of the deep basin of the gulf, where the dominant anticlockwise eddy is felt less. It is also probable that in most cases the young Sebastes, like their parents, live rather below the level of the most active currents, hence are less apt to be caught up by them. Further (though less important in its effect than is the location of the breeding grounds in relation to the circulation of the gulf) , Sebastes is so compara- tively large and strong at birth that its involuntary migrations cover a shorter period than those of most of the fishes that lay floating eggs, and consequently its larvae are to be found widespread, except close to land, and not concentrated in any one part of the gulf. QUANTITATIVE DISTRIBUTION OF THE ZOOPLANKTON To give an adequate quantitative picture of the plankton would require a far greater number of vertical hauls than have yet been made in the Gulf of Maine. Not only are the seasonal gaps in the series serious, but hauls should be located closer together than has been feasible for us, even in July and August, unless the plankton is more uniform than our work suggests. However, even a cursory examination of the zooplankton, if extended over a considerable area or through a considerable period of time, is certain to reveal wide fluctuations in abundance as well as in its qualitative PLANKTON OF THE GULF OF MAINE 79 composition, both from season to season and from place to place; and inasmuch as an understanding of the causes of the fluctuations in the numerical strength of any group of marine animals would clarify the interaction of the many physical factors that govern pelagic life in the sea, information along this line is never amiss. Quantitative data regarding the plankton run the whole gamut from the most casual to the most accurate and precise, depending on the method of collection and enumeration employed, which in turn depends on whether it is the absolute numbers of individuals of any group that is sought or merely their abundance relatively and in a rough way. Perhaps I shall not be taken to task when I add that no wholly satisfactory method has yet been devised for estimating the abundance of the larger and more active members of the zooplankton. With immobile objects such as fish eggs, or weak swimmers such as ctenophores and copepods, vertical nets of the more modern patterns yield counts of reasonable accuracy; but when we attempt to deal with animals whose powers of directive swimming are as well developed as those of Sagittse, euphausiids, young fish, etc., the certainty that some of them — it may be many or it may be few — escape the net introduces an unavoidable source of error and one that is far more serious than the clogging of the meshes, resulting in only partial filtration of the column of water through which the nets fish, and one that must always be reckoned within quantitative work. For this same reason enumerations of the plankton contained in samples of sea water of known volume, collected by water bottle or by pump, a method that has proved fertile for the study of the phytoplankton (p. 398), are of no value whatever for any animals except the smallest. In short, any absolute census of the total plankton in the open sea will, we think, long remain something of a will-o'-the-wisp. If the goal be no more than a comparative (not an absolute) estimation of the amount of zooplankton present in the water, these difficulties fade. If the same type of net is employed for all the hauls and of a mesh calculated for the general size of the plankton elements for which it is intended, and if the length of the column of water fished through is either known accurately or is the same on all occasions, the catches will be fairly comparable one with another, and the net error (that is, failure to filter perfectly) becomes secondary. If the nets are large enough in diameter 31 (say half a meter or more), with filtering surfaces sufficiently extensive in proportion to the mouth area, and of a shape proper for the rapid passage of water, they will certainly capture a majority of the animals in their path up to the size of amphipods, Sagittae, and euphausiids. In the case of the copepods, which, after all, are the backbone of the zooplankton of the Gulf of Maine, the catch will be suffici- ently representative of the actual population for comparative purposes, 35 even if the few individuals that chance to lie near the outer rim of the mouth of the net dodge it and escape. With this end in view we have, since 1914, abandoned vertical nets of the Hensen pattern, with their small mouths, for a vertical net half a meter in diameter, of the Michael Sars pattern; 36 and I may add that in making vertical hauls the net has 3 * The larger the better. « A whole literature, from the hands of its sponsors or critics, has arisen about the reliability or the reverse of the vertical net, which has been the classic engine for quantitative plankton studies ever since Hensen (1887) first sponsored it. » For specifications of this pattern see Murray and Ujort , 1912. 7.J80S— 26 6 80 BULLETIN OF THE BUREAU OF FISHERIES invariably been lowered as near to bottom as feasible, so as to sample the whole column of water. As yet we have not attempted a quantitative survey of any particular stratum, though, from the nature of the case, the hauls in the shallow coastal zone have been confined to a thin layer of water. The results of the vertical hauls are supplemented by the much more numerous horizontal hauls, made with various nets and covering the gulf generally at most seasons of the year. Inasmuch as the quantitative value of horizontal hauls has often been disputed, I must admit at once that they seldom fulfill the basic requirement of fishing through a column of water of known length. Furthermore, while the level at which an ordinary open net works for the major part of the haul can be determined within reasonable limits if it is used at moderate depths, its yield can not be depended upon as an index of the richness of the plankton at that particular depth unless cor- roborated by other evidence, because it may have passed through a swarm of copepods or what not on its way up or down. Horizontal hauls made in deep water, say of 500 meters or more, have little quantitative value if of short duration, because the horizontal journey made by the net may then be little if any longer than the vertical, which, of course, may be equally true of individual hauls in shallow water under exceptional circumstances. In general, however, it is safe to assume that when the horizontal distance through which the net works exceeds the vertical manyfold, as is the case for shallow hauls of considerable duration (for example, our standard of half an hour at 100 meters or shallower), considerable weight may be given to the average quantitative results of several hauls, the more so the greater the discrepancy between their horizontal and vertical portions, hauls at the surface being entirely satisfactory in this respect. In short, while everyone agrees that it is idle and misleading to expect precise quantitative data from ordinary tow nets used hori- zontally from a moving vessel, there is no need of going to the other extreme, as some students have done, and discarding a method that is not only so convenient but so often available when rough weather prohibits vertical hauls. 37 As a matter of fact, if they are interpreted with common sense and made at appropriate levels in the water, the catches of the horizontal tow nets often throw much light on the quan- titative distribution of the animal plankton, especially in preliminary surveys. At the worst they can be trusted to reveal the existence of areas of markedly rich or of very scanty plankton, for no one can deny that the plankton must be more abundant where tows are uniformly productive than where the same nets as regularly yield little or nothing, especially at times and places when and where the larger animals occur in local shoals, which the vertical net may miss altogether but which a long horizontal tow is almost certain to encounter. Thus, to quote only one example, Jespersen (1924) was able to demonstrate very wide differences in the abundance of zooplankton in different parts of the Atlantic, from horizontal hauls of long duration with large nets, especially the general poverty of the so-called "Sargasso Sea." •' An excellent example of the light which horizontal hauls may throw on the fluctuating abundance of the plankton is afforded by the long-continued series of tow nettings carried out by the Marine Biological Laboratory at Port Erin, on the Isle of Man, under Professor Herdman's direction. PLANKTON OF THE GULF OF MAINE 81 The choice of a unit and of a method of measurement by which to express the quantitative abundance of the zooplanktonic community as a whole, as distinguished from its several component groups, is a matter of real difficulty. The easiest thing to do is simply to let the .whole catch settle in suitable jars or graduates until visible shrinkage ceases and to record the volume of the resulting mass. Unfortunately, however, this does not give a true measure of the actual content of the net, much less (owing to the sources of error just mentioned) of the total column of water fished through, because it likewise includes the gaps between the individual animals composing it, together with any detritus that may have been in suspension in the water. This introduces a serious error, for plankton settles more or less closely according to the shapes of the individual animals composing it, smooth, round, fish eggs, for example, packing far more closely and regularly than do copepods with their long appendages. Nevertheless, even such simple measure- ments as this yield rough pictures of the abundance of the animal plankton, hence they have been made for all our vertical tows and for many of the horizontal ones. Jespersen (1924) measured the volume of the catch after draining the water from it. The process may be rendered more accurate if after draining a known amount of water is added, when the resultant increase in the volume will correspond to that of the catch plus the small amount of liquid which still adhered to the plankton after the draining. I have employed this method in a few cases where it seemed likely that the direct measurement of volume would be seriously misleading because of the character of the organisms concerned. The use of the centrifuge would be stdl better, but this has not been attempted for the Gulf of Maine hauls. 38 Counting is the most instructive method of estimating the catch from most points of view, though it entails much labor and time, and this is the only method by which the actual numerical strength of the several groups of animals composing the zooplankton can be learned. Various types of apparatus have been devised for this purpose, most of them by the Kiel School of Biologists, the process followed for the Gulf of Maine hauls being as follows: The catch of the vertical net (its volume having been measured as above) is first diluted to a volume of 150 cubic centimeters, well mixed, and then, while the plankton is still in suspension, 3 cubic centimeters are taken with a suitable pipette and the copepods, fish eggs, etc., counted. The ordinary pipette, familiar to every biologist, will seldom serve for taking this sample; but it is not necessary to em- ploy the complicated "Stempel" pipette, for one of the shape shown in the accompanying sketch (fig. 36), with large rubber bulb, tube opening about 3 millimeters in diameter, and total volume of Fig. 3fi.— Volumet- ric pipette used for sampling cope- pods for counting aB For an excellent account of these and of other methods of plankton estimation see Johnstone, 1908, p. 129. 82 BULLETIN OF THE BUREAU OF FISHEBIES about 25 cubic centimeters, graduated as required, serves well for copepods and all smaller animals. The chief difficulty is that it is not always easy to make sure that the diluted plankton is evenly distributed in the fluid while the sample is being taken, because the various animals settle at different rates. Therefore, it is usually advisable to take two or sometimes three samples from each haul and average the results. Animals as large as amphipods, Sagittae, and euphausiids are seldom so numer- ous but that it is easy to count the entire number caught in a vertical haul, and as a rule it is necessary to remove them before taking the sample of copepods, etc., lest they clog the mouth of the pipette. Fish eggs, also, can usually be counted directly from the entire catch, though they sometimes occur in such numbers that it is neces- sary to take a sample for this purpose. The copepods have been counted for most of the vertical hauls, the results being discussed in the chapter on that group (p. 167). Notes on -numerical strength of other animals will be found under the particular species. The unit of measurement best available for the volume depends upon whether horizontal or vertical nets are used. If the former, calculation of the amount per hour's hauling, as employed by Jespersen (1924), can hardly be bettered; but vertical hauls lend themselves to a somewhat more exact measure, namely, the amount present under some chosen area of the surface of the sea, which is usually expressed in cubic centimeters of plankton per square meter. This would be a sufficient index to the total productivity of any locality at any given time, and hence is often extremely instructive from the biologic viewpoint; but, as I shall have occasion to emphasize later (p. 90), it does not necessarily throw any light on the density with which the plankton is aggregated, since it neglects the possible stratification of the latter at different levels. On this basis the animal plankton of the gulf as a whole, like the phytoplankton (p. 399), is apparently at its lowest annual ebb late in February and during the first half of March, when it was only in the western basin and over a tongue extending from the Eastern Channel and eastern edge of Georges Bank northward along the axis of the eastern basin to the 100-meter contour off Grand Manan (fig. 37) that we found as much as 75 cubic centimeters per square meter in 1920. Nor did we make any rich hauls then even in these comparatively productive zones, judged by mid- summer standards (p. 83). In all other parts of the gulf at the time, both inshore and over the basin, except as just qualified, and on Georges Bank as a whole, the water supported less than 25 cubic centimeters of plankton per square meter of sea surface, with several of the catches too small to measure, while on one occasion (off Cape Elizabeth, March 4, station 20059) the vertical net yielded nothing whatever. If the minimal catches of February and March, 1920 (less than 25 cubic centi- meters), be credited with 15 cubic centimeters of zooplankton per square meter (probably an excessive estimate), the average for the whole gulf at this season was only about 40 cubic centimeters, contrasted with about 100 cubic centimeters in midsummer, and the distinction between rich and barren was decidedly more sharply marked than we have found it during the more productive seasons of the year. PLANKTON OF THE GULF OF MAINE 83 The few data available suggest that April sees a general augmentation in the amount of animal plankton across the southern half of the gulf from the mouth of Massachusetts Bay to the coastal bank off Cape Sable, including the eastern part of Georges Bank. Over this zone the plankton volumes per square meter averaged about 100 cubic centimeters during the second and third weeks of that month in 1920; but north of a line from Cape Cod to Cape Sable, where diatoms were flowering freely (p. 385), our hauls, horizontal as well as vertical, certainly yielded no larger amounts of animal plankton in April than in March and an unmistakable decrease in the amount of zooplankton took place from March to April in the northeastern part of the basin coincident with the local flowering of diatoms. However, the swarms of microscopic plants which are then present make quantitative, measure- ments of the larger forms diflieult or even impossible, both by clogging the meshes and by overshadowing the copepods, etc., in the catches of the tow nets. Unfortunately we have not been able to follow the planktonic cycle through the whole of any one spring. But if the May state of 1915 represents the normal sequence to the April state of 1920 (a reasonable working hypothesis unless shown to be false), the zooplankton increases to volumes of 200 to 235 cubic centimeters off Massachusetts Bay and northward toward Cape Elizabeth, on the one hand, and in the eastern basin off German Bank, on the other, during the last half of April and first half of May, as tabulated elsewhere (Bigelow, 1917, p. 312), an increase caused by the tremendous production of copepods which succeeds the vernal flowering of diatoms (p. 41). In fact, it will probably be no exaggeration to set the average volume of zooplankton per square meter by the last of May at 100 or more cubic centimeters for the whole gulf outside the 50-meter contour and north of the Cape Cod-Cape Sable line, 39 with the exception of the coastal zone from Penobscot Bay eastward, where the water still remained extremely barren on May 11 and 12 (volumes of 10 to 20 cubic centi- meters at stations 10275 and 10276). Except for this barren zone, where the catches have been so small as hardly to be measurable, the gulf as a whole probably supports a greater mass of animal plank- ton during the last week of May and the first part of June than at any other season, though we have few quantitative records for the latter month. The considerable Dumber of vertical hauls made in July and August during the summers of 1912 to 1916 (listed in table on p. 84) make it possible to outline with some confidence the major geographic variations in the amount of zooplankton present in the guff in midsummer. During the summer of 1914, which may serve as representative, the animal plankton was most plentiful (volumes of 100 cubic centimeters or more per square meter) in three distinct and separate regions, which I have described elsewhere (Bigelow, 1917, p. 308, fig. 91) — first, over a belt running diagonally across the gulf from the Massachusetts Bay-Cape Cod region to the northeast corner of the basin off the mouth of the Bay of Fundy, as outlined on the accompanying chart (fig. 38) ; second, over the northeast corner of Georges Bank; and, third, from Cape Sable out across the northern channel to Browns Bank, which, on the evidence of the hori- zontal hauls, should include German Bank, because of the Pleurobrachia which wo '» We have no quantitative data (or May and June from Georges Bank. 84 BULLETIN OP THE BUREAU OF FISHERIES found swarming there in 1912, 1913, and 1914 (p. 19) . 40 While 1914 is the only summer for which we have quantitative data from the offshore banks, all the most productive (100 + cubic centimeters) of the summer hauls of 1913, 1914, 1915, and 1916 41 were hkewise similarly concentrated in the Cape Cod-Bay of Fundy belt just outlined (fig. 38). So uniformly productive has this "rich zone" proved in summer that only 3 of the 25 vertical hauls, which we have made there in June, July, and August, have failed to yield upwards of 100 cubic centimeters of animal plankton per square meter, although the waters both immediately to the north and to the south of it have often proved decidedly barren, as the chart illustrates. The average volume of plankton for aU the vertical summer hauls in this rich zone has been nearly 170 cubic centimeters per square meter including those for 1916 (an exceptionally rich year), and more than 150 cubic centimeters if the 1916 hauls are omitted. Approximate volume of plankton per square meter of sea surface. July and August hauls, 1912 to 1916 Volume Volume Year Station in cubic centi- meters Depth Year Station in cubic centi- meters Depth Meters Meters 1912 10002 10004 250 50 119 55 1914 10213 10214 210 120 110 175 10007 65 265 10215 60 70 10008 50 41 10216 30 70 10011 20 110 10218 50 500 10015 10 37 10223 170 75 10021 10 110 10224 240 55 10022 30 82 10225 30 260 10025 80 91 10226 200 85 10027 30 165 10227 50 220 10031 30 128 10229 170 100 10035 Trace. 73 10230 140 50 10036 30 165 10243 100 55 10038 20 73 10244 15 50 10043 15 165 10245 10246 60 200 110 190 Fathoms 10247 10 30 1913 10087 10089 180 80 128 183 10248 10249 100 105 190 220 10090 120 164 10250 350 145 10092 160 219 10253 60 140 10095 60 37 10254 200 260 10096 120 91 10255 70 175 10098 10099 70 30 55 37 1915 > 10304 10306 275 no 200 140 10100 220 165 10307 165 235 10101 10102 100 90 73 128 1916 10340 10341 125 250 45 80 10103 70 73 10342 250 55 10104 90 146 10344 225 80 10105 55 110 10345 200 150 10346 200 62 1 For a list of the hauls for other months of this year see Bigelow, 1917, p. 314. Contrasting with the rich belt, the entire coastal zone of the gulf, from Cape Ann on the south and west to Grand Manan Island at the mouth of the Bay of Fundy on the east and north, has invariably proved far less productive of zooplankton in midsummer — never with more than 90 cubic centimeters per square meter, usually *> These ctenophores had shrunk in the preservative to only a fraction of their natural bulk before the vertical hauls were measured. " In 1916 the zooplankton was unusually abundant in the waters off Cape Cod and in the southwest corner of the gulf in July, a fact discussed on p. 97. PLANKTON OF THE GULF OF MAINE 85 with less than 70 cubic centimeters, and ranging from this down to traces too small to measure. North of Cape Ann the general rule has been the closer to land in summer the scantier the catch (fig. 38), while the coastal belt as a whole then sup- 7(C Fig. 37.— Volumes of plankton, in cubic centimeters, below each square meter of the surface of the sea in February and March, 1920, as calculated from the catches made in the vertical hauls. In the shaded area the volumes were uniformly greater than 75 cubic centimeters. ports less zooplankton to the north and east of Cape Elizabeth than to the south and west, with the Grand Manan Channel the most barren part of the open gulf. We have no quantitative data from the immediate vicinity of the western coast of Nova 86 BULLETIN OF THE BUREAU OF FISHERIES 9 O 7T Fig. 38.— Volumes (cubic centimeters) of animal plankton below each square meter of the surface of the Gulf of Maine in summer, as calculated from the vertical hauls made in 1912-1916. 0, 100 cubic centimeters or more per square meter; ©, 50 to 100 cubic centimeters or more per square meter; O. 50 cubic centimeters or less per square meter; A, stations where horizontal hauls showed an abundant plankton, but where no vertical hauls were made. The hatched curve includes areas where we have usually found more than 100 cubic centimeters per square meter; the stippled curve where the catches have usually been less than 50 cubic centimeters per square meter. PLANKTON OF THE GULF OF MAINE 87 Scotia, but in 1914 the neighborhood of Lurcher Shoal proved far less productive than the deeper basin near by. Were all parts of the gulf equally favorable for the existence and multiplication of animal plankton, the catches of the vertical hauls might be expected to vary in direct ratio to the depth — that is, to the amount of water filtered by the net — and, speaking broadly, there usually is more plankton below any given unit of the sea's surface in moderately deep water (say 50 meters or more) than in very shoal water. Notwithstanding the comparative barrenness of the greater part of the coastal zone, however, the regional differences in the abundance of plankton in the Gulf of Maine do not correspond closely to the depth; nor can they be correlated with the distance from the coast, per se, because we have repeatedly found the plankton very plentiful in moderate depths both near land, as in Massachusetts Bay, and close in to Cape Sable, and as far offshore as Georges and Browns Banks, while, on the other hand, some of our deep hauls have proved unproductive in spite of the considerable length of the column of water fished through. Such, for example, was the case in the Eastern Channel and the neighboring part of the basin in July, 1914. In fact, the vertical hauls made in the southeastern deep of the gulf in summer (July 23, 1914, station 10225, and June 25, 1915, station 10298), have both proved extremely barren, with only 30 to 70 cubic centimeters per square meter in spite of the considerable depths of the hauls (175 to 260 meters), showing that both in June of 1915 and July of 1914 the rich zone was bounded on the east by much less prolific waters. It is on the strength of these hauls that I have laid down the demarcation between the two zones on the accompanying chart (fig. 38), but the volume of plankton present in the water varies so widely from season to season and from year to year that the lines must not be drawn too finely in plotting its regional variations, and the future alone can show whether it is regularly characteristic of the summer season for such a barren wedge to separate the rich waters to the north from the equally prolific shallows of Georges and Browns Banks. The presence of more than 200 times as much animal plankton beneath each square meter of the surface of the sea at the mouth of Massachusetts Bay on July 20, 1916, as in water nearly twice as deep in the Grand Manan Channel on August 19, 1912 (only a trace), and the fact that there were 200 cubic centimeters per square meter in 85 meters of water on the northeastern edge of Georges Bank on July 24, 1914, but only 50 cubic centimeters per square meter that same day in the Eastern Channel, 15 miles distant, where the depth was 220 meters, illustrate the contrast between productive and barren waters. Vertical hauls in the Massachusetts Bay region, the only part of the gulf where our data warrant even a tentative account of the quantitative, fluctuations that take place during late summer and autumn, suggest a diminution in the volume of zoo- plankton during the late summer followed by an autumnal increase, which was so considerable in 1915 that there was over twice as much plankton per square meter in water only 80 meters deep by the end of October as we had found at a neighboring station in 140 meters depth two months previous. 88 BULLETIN OF THE BUREAU OF FISHERIES Zooplankton volumes, mouth of Massachusetts Bay Date Station Depth of haul in meters Approxi- mate volume, cubic centi- meters per square meter Date Station Depth of haul in meters Approxi- mate volume, cubic centi- meters per square meter July 10, 1912 10002 10340 10341 10342 10087 10253 119-0 45-0 80-0 55-0 128-0 140-0 250 125 250 250 180 60 Aug. 31, 1915 10306 10324 10338 20050 20090 10266 140-0 140-0 80-0 150-0 120-0 125-0 110 Oct. 1, 1915. 150 Do Oct. 27, 1915 250 Do Mar. 1, 1920 ±25 Aug. 9, 1913 Apr. 9, 1920 10 Aug. 22, 1914 May 4, 1915 270 Evidence that a similar augmentation spread generally throughout the coastal waters west of Penobscot Bay in 1915 is afforded by volumes as great as 100 to 150 cubic centimeters per square meter off Penobscot Bay, off Cape Elizabeth, and near the Isles of Shoals during that October. However, we have yet to learn whether this increase is an annual event, nor does our experience suggest that it extends east of Penobscot Bay, because vertical hauls yielded only 30 cubic centimeters per square meter off Mount Desert Island and 20 cubic centimeters off Machias on October 9 (stations 10328 and 10327). We have made no quantitative hauls in the gulf during the period between Octo- ber and late February, but the comparative scantiness of the yields of the horizontal nets in Massachusetts Bay during the cold months of 1913 (Bigelow, 1914a) and at all our inshore stations from Cape Cod to Yarmouth, Nova Scotia, in December, 1920, and January, 1921, points to an ebbing zooplankton as characteristic of the coastal belt in late autumn and early winter, leading progressively to the extremely barren state of the water typical of the first weeks of spring (p. 82). Hauls made near Mount Desert Island and in the northeast corner of the gulf from January 1 to 5, 1921 (stations 10497, 10500, and 10502) were equaUy unproductive, 42 but I hesitate to conclude from this that the water was actually so barren there, because horizontal hauls were hardly more productive in that general region in March, 1920, although the vertical nets yielded large catches, a fact suggesting that the former missed the level at which the plankton was most concentrated. However this may be, it seems that in winter and early spring the zooplankton is far more plentiful in the western side of the basin than near shore, because we made a rich horizontal catch there on December 29, 1920 (station 10490), a rich vertical haul (though a rather scanty horizontal) on February 23, 1920 (station 20049), and a rich horizontal and a comparatively rich vertical on March 24 of that year (station 20087). The results of both vertical and horizontal hauls point to the Massachusetts Bay region and the neighboring part of the basin, on the one hand, and to the deeps off Lurcher Shoal and the eastern part of Georges Bank, on the other, as the parts of the gulf uniformly most productive of zooplankton; wbile the deep water in the » Yield of half an hour's haul with a M-meter net was only about 100 to 150 cubic centimeters in each case at 50-0, 75-0, and 150-0 meters. PLANKTON OF THE GULF OF MAINE 89 southeastern corner of the gulf, where vertical hauls have yielded only 25 to 65 cubic centimeters per square meter on four visits (March 11, 1920, station 20064; April 17, 1920, station 20112; June 25, 1915, station 10298; and July 23, 1914, station 10225), although made in depths of from 200 to 340 meters, and the coastal zone east of Penobscot Bay would seem to be the least productive. Recapitulating for the Massachusetts Bay region, the zooplankton is at its scantiest some time in March, earlier or later according to the forwardness of the season; it increases very rapidly in amount during May, reaches its annual maximum of abundance late in May or early in June, when there may be from 10 to 20 times as much animal life in the water (200 to 300 cubic centimeters per square meter) as in March, and wanes in August. A second well-marked pulse is noticeable in Sep- tember, culminating in October, after which the plankton diminishes once more. Our experience during the cold months of 1912 and 1913 (Bigelow, 1914a) was that a moderate amount of zooplankton is to be found in the bay throughout the winter, but that it suddenly declines almost to the vanishing point late in February or early in March. The plankton passes through a corresponding quantitative cycle throughout the entire coastal zone from Massachusetts Bay to the mouth of the Bay of Fundy; but although the waters east of Cape Elizabeth are as barren as the region from the Isles of Shoals to Cape Cod in early spring, they are never as productive of zooplankton as is the latter in late spring and early summer, and, consequently, the difference between the seasons of maximum and of minimum abundance of plankton is not as great. The fact that the northern corner of the eastern basin proved extremely barren on April 20, 1920 (station 20100), whereas we have found an abundant animal plankton there in summer, suggests that this region, like Massachusetts Bay, is the site of a wide seasonal fluctuation, with a brief period of barrenness in spring coin- cident with the vernal flowerings of diatoms. This applies likewise to the shallows off Cape Sable and over the eastern part of Georges Bank, where the zooplankton is extremely plentiful in midsummer but sparse in March. So far as our experience goes, the seasonal fluctuation in the amount of plank- ton present is widest in the neighborhood of the Isles of Shoals, with a range of from practically nil to upwards of 300 cubic centimeters per square meter. The coastal belt along the outer islands east of Penobscot Bay illustrates the opposite extreme. Here the catches of the vertical nets may be but little larger (25 to 30 cubic centimeters per square meter) in summer (the richest season) than in spring, and we have only once made a reasonably productive vertical haul in this zone (70 cubic centimeters per square meter at station 10098). The quantitative fluctuations are also comparatively narrow from season to season, or at least no pronounced impoverishment takes place in spring, in the deep waters of the western basin, so that the plankton of that part of the gulf is classed as "rich," not "scanty," the year around, as shown by the following table. 90 BULLETIN OF THE BUREAU OF FISHERIES Volumes of plankton per square meter, western basin Feb. 23,1920. Mar. 24, 1920. Apr. 18,1920. May 5, 1915 Cubic centi' meters of zooplank- ton per square meter 175 95 150± 250 Date Cubic centi- meters of zooplank- ton per square meter June 26, 1915. July 15, 1912. Aug. 22,1914 Aug. 31,1915 250 65 200 165 There is, likewise, less fluctuation with the seasons on the western part of Georges Bank than on the eastern. The largest volume of plankton per square meter yet recorded for the Gulf of Maine was 425 cubic centimeters in the eastern side of the basin on September 1, 1915 (station 10309), while the smallest was a bare trace. In fact, the animal population may be so sparse locally that a vertical haul may catch nothing at all, as has been our experience at several stations along the coast of Maine and in the Grand Manan Channel (p. 84) ; but even then, a half hour's tow with the horizontal net has invariably yielded a few copepods or other animals, proving that although the planktonic community may fall to a very low ebb, indeed, at its season of scarcity, it never vanishes wholly from any part of the gulf at any time of the year. DENSITY OF ASSOCIATION OF THE ZOOPLANKTON A statement of the volume of zooplankton existing in the total column of water below any chosen unit of sea area — e. g., each square meter — serves to illustrate the total regional and seasonal production of the gulf; but unless the water in question be very shallow, it throws little light on the density in which the animals concerned are congregated, because the catch of the vertical haul may be distributed generally over a column so long that even a considerable volume of plankton might mean only a sparse population. To meet this need, another unit of measurement is required, the one usually employed in other seas, and of which I have made use in previous re- ports (Bigelow, 1915 and 1917), being the volume of plankton present in each cubic meter of water. This , of course, is simply the product of the volume per square meter of sea surface divided by the depth (in meters) covered by the haul in question. Were the zooplankton of the gulf uniformly distributed from the surface down to bottom, this simple calculation would not only "establish the relative richness of different regions in plankton, and hence in food for the pelagic fishes " (Bigelow, 1915, p. 327), a question naturally of much importance in the economy of the gulf, but go far to explain many biologic problems even more far reaching. Unfortunately for the statistician, however, such is not the case, all our experience tending to show that the zooplankton is often more or less stratified and that the degree of stratification varies widely from place to place with the time of day and with the change of the seasons. Consequently, the results always require analysis in the light of any information bearing on the vertical distribution of the planktonic communities represented in the catches in question. Otherwise one is apt to be led to conclusions so widely astray as to be worse than none. PLANKTON OF THE GULF OF MAINE 91 On the whole, it is in late winter and early spring, when the physical characters of the sea water are most uniform vertically and when its vertical stability is least, that the zooplankton of the Gulf of Maine and of other boreal seas most nearly approaches vertical uniformity of distribution. At this season, as illustrated by the March cruise of 1920, the volumes of zooplankton present in the water are so small in all parts of the gulf, and the depth of water through which it was distributed at the more productive localities is so considerable, that the volume per cubic meter (by direct calculation) was only 0.7 to 1 cubic centimeter even where the plankton was densest — for instance, in the eastern and northeastern troughs of the basin, in the Eastern Channel, and over the northeastern and southeastern parts of Georges Bank. It ranged down from this to a minimum of practically nothing in the deep water in the southeastern corner of the gulf, the average for all stations being about 0.4 cubic centimeters, which is something less than half the summer average by the lowest possible estimate. Nor is it likely that this calculation seriously understates the density of aggregation of the zooplankton for any large portion of the gulf in March, because there was little evidence of vertical stratification during that month. Zooplankton volumes per cubic meter, March, 1920 Locality Date Station Cubic centi- meters per cubic meter Locality Date Station Cubic centi- meters per cubic meters Feb. 23 Mar. 1 Mar. 2 Mar. 3 ...do 20049 20050 20052 20053 20054 20055 20056 20057 20058 20060 20061 20062 20063 20064 0.6 .1 . 1 .3 .4 .5 t 2 .2 .5 .2 .1 .5 .1 .0 Georges Bank: Mar. 11 .do. . 20065 20066 20067 2IIDI ;s 20070 20071 20079 21111-11 20081 20083 20086 21 II 1ST 20089 0.3 .3 Mar. 12 do .5 .7 03 Mount Desert Rock ...do ...do Mar. 13 do. . 1.0 OH Mount Desert Island .7 Mar. 4 ...do Mar. 22 .do. .. .1 OB Machias (Me.)... .4 ...do do .7 Mar. 5 ...do. .. Off Yarmouth, Nova Scotia Mar. 23 do .4 OB Boston... .5 Mar. 11 ...do Mar. 24 Apr. 6 .4 Off Boston .4 With the advance of the spring the concentration of the plankton is augmented both by the increase in the total amount present in the gulf, just remarked, and by its stratification at one level or another. Not only does the first of these factors raise the volume per cubic meter to 2 to 4 cubic centimeters at the very least by midsummer in such prolific though rather shallow regions as the waters off Cape Cod, the neighborhood of Cape Sable, and the eastern part of Georges Bank, 43 but stratification may result in a far denser concentration of the plankton at some particular level while rendering other strata of water far more barren than the ostensible volumes per cubic meter (as derived from the usual calculation) would call for. We have encountered this phenomenon in its most extreme form in the deeper parts of the gulf, but experience has shown that a greater or less tendency on the part of the zooplankton, as a whole, to congregate at some particular level is to be expected anywhere in the gulf in summer, leaving the shoaler as well as the deeper * s Plankton volumes per cubic meter, calculated from our summer and autumn hauls, have been published already; those for the year 1913 in Bigelow, 1915, p. 326; for 1914 and 1915 in Bigelow, 1917, pp. 310 and 314; and for 1916 in Bigelow, 1922, p. 136. 92 BULLETIN OF THE BUREAU OF FISHERIES layers of water practically deserted except in regions where active vertical currents keep the water thoroughly mixed. Therefore, it is usually safe to assume that the plankton is far more densely aggregated at some level, though perhaps only through a very narrow vertical zone, than the calculation of volume per cubic meter would indicate; but since we have occasionally found it rather uniformly distributed from the surface downward, even in the more stagnant parts of the gulf, no hard and fast rule can be laid down in this respect. Vertical stratification may result from a definite vertical migration of various animals toward the surface during the hours of darkness and downward again at sunrise, but quite apart from this phototropic phenomenon, which has often been described in other seas and which I have touched on above (p. 24), the tendency frequently shown by animals of different systematic groups (one of which may be and often is far more plentiful than the others) to segregate at different levels during the warm half of the year — copepods, for instance, at one depth and Sagittse at another — often causes a very uneven quantitative distribution of the plankton vertically in summer and early autumn. In Julj T and August, 1913, for instance, it was invariably the shoaler subsurface haul that yielded the largest catch at stations where two such were made with the horizontal nets at different levels, even after making allowance for the use of nets of different types, although the reverse might have been expected because of the greater volume of water strained by the deeper hauls. " Evidently, then, the zooplankton was usually densest in the upper strata of water during that particular summer, say from 20 meters down to 50 at the localities of record, which were generally distributed over the offshore parts of the northern half of the gulf, and it was decidedly less abundant below 75 meters on the one hand or in the surface stratum on the other. This rule did not hold during the summer of 1914, however, when it was sometimes the deeper haul (stations 10215, 10246, 10248, and 10254), sometimes the shallower (stations 10214 and 10249), that yielded the largest catches, but usually one was much more productive than the other, as illustrated by the following table: Comparative catches of horizontal hauls of half an hour's duration (reduced to a column 1 square meter in cross section) during July and August, 191 4- |The depth is the level at which the major part of the haul was made "J Locality Southwest Basin Georges Bank, northwest part Southeast Deep Eastern Basin Northeast Deep Off Mount Desert Rock Western Basin _ Station 10214 10215 10225 10249 10246 10248 10254 Date July 19. July 20. July 23. Aug. 13. Aug. 12. Aug. 13. : Aug. 22. { Depth in meters 30 150 30 60 60 240 50 175 50 150 50 150 75 225 Volume in cubic centi- meters 3,550 250 150 375 160 125 2,180 500 150 1,000 150 1,250 150 625 a Assumed to have fished through three quarters of a mile. ** For discussion of these hauls, with necessary corrections, and for the tabulated results, see Bigelow, 1915, p. 327. PLANKTON OF THE GULF OF MAINE 93 Although it was often the deeper haul that yielded the larger amount of plankton, all the very rich tow-net catches (2,000 cubic centimeters or more) made in the ^ulf during that summer (six in number; see Bigelow, 1917, p. 312) were from depths of 100 meters or less, with the average volume (about 900 cubic centimeters) of all the subsurface catches made shoaler than 100 meters, almost three times that of the deeper hauls (about 350 cubic centimeters), although the latter fished through a longer column of water on their journey down and up. Thus, it seems that the gulf is usually richer in zooplankton above than below 100 meters depth during the summer season, and very rich catches were made in vertical hauls shoaler than that at the few stations which the Grampus occupied in the gulf during July, 1916 (p. 92; Bigelow, 1922, p. 136). With the plankton often concentrated at some one level, it becomes more or less a matter of chance whether a net fishing horizontally hits or misses the richest zone. Consequently, the yields of the two sorts of hauls, horizontal and vertical, are often far from parallel. When there is a wide discrepancy between the two it has usually been in favor of the horizontal net (especially in deep water), for we have usually made at least one horizontal tow in the productive stratum between 40 and 100 meters at each station, whereas the vertical catch mirrors the plankton content of the barren strata combined with that of the rich. Occasionally, however, the tables are turned, as was the case on July 23, 1914, on the eastern part of Georges Bank (station 10223), where the volume per cubic meter taken by the vertical haul was more than seven times as great (2.2 cubic centimeters) as that taken by the horizontal haul (about 0.3 cubic centimeter) although the depth of water — that is, the length of the column fished through — in the case of the former was only 82 meters, whereas the latter worked for about three-quarters of a rrule. Thus, the vertical net must have passed through water much more productive than the level at which the horizontal net was fishing. In 1913 and 1914, too, the richest catches with horizontal nets were not at the stations where the volumes per square meter or per cubic meter were largest, as calculated from the vertical hauls. It follows from these facts that while the ostensible volumes per cubic meter may be a satisfactory index to the density of the planktonic population of the Gulf of Maine in winter or early spring, and in summer at stations where no stratification is apparent from the yields of the horizontal hauls, and while this calculation may approximate the truth in very shallow waters generally at most times of year, as a rule it greatly understates the actual maximum density of aggregation of the plankton in deep water, making such regions appear much less prolific as feeding groimds for pelagic fishes than their richer layers actually are, while crediting far too high a plankton content to their more barren strata, as I have pointed out else- where (Bigelow, 1917). Owing to the tendency of the zooplanktonic community as a whole to con- gregate in the upper 100 meters of water during the warm months, but at the same time to keep some few meters down (p. 24), the seasonal difference between the volumes of plankton per cubic meter present in March, on the one hand, and in July and August, on the other, is actually much greater than the ratio arrived at by any calculation which fails to take account of its vertical stratification. A more 94 BULLETIN OF THE BUREAU OF FISHERIES nearly correct picture of the summer state results from the assumption that the entire catch of zooplankton in the vertical net at that season was taken below 10 meters at each station, but that it was only one-third as dense as the ostensible volume per cubic meter below 100 meters, and correspondingly concentrated above that level. The results of such a calculation for 1914 are given in the following table: Volumes of plankton per cubic meter (in cubic centimeters) between the depths of 10 and 100 meters, July to August, 191 4 1 Locality Date Station Total depth in meters Volume per cubic meter if calculated as above, in cubic centi- meters Volume per cubic meter if uniformly distrib- uted, in cubic centi- meters Off Cape Cod July 19 ...do.. . 10213 10214 10215 10216 10223 10224 10226 10225 10227 10229 10230 10243 10244 10246 10247 10248 10249 10250 10253 10254 10255 110 175 70 70 75 55 85 260 220 100 50 55 50 190 30 190 220 145 140 260 175 2.2 1 1 .5 2.6 5.3 2.6 .2 .4 1.9 3.5 2.2 .4 1.7 .5 !s 3.3 .6 1.4 .6 1 90 Southwest Basin 68 Georges Bank: July 20 ...do . 85 Southwestern part .43 Eastern part... July 23 .do 2.40 4.30 July 24 July 23 July 24 July 25 do Southeast Deep __. _ .12 Eastern Channel _ .23 North Channel... ._ 1.70 2 80 Do Aug. 11 Aug. 12 .do 30 Northeast trough _ 1. Off Machias, Me... _: .do .33 Off Mount Desert Rock Aug. 13 do .52 .48 Aug. 14 Aug. 22 . .do 2.40 Off Cape Ann .42 Western Basin .77 Aug. 23 .40 i For tables of the volume per cubic meter for July and August, 1913, and for May to October, 1915, see Bigelow, 1915, p. 328, and 1917, p. 314. The most instructive feature of this table is its demonstration that, although the total amount of plankton present below any given unit of the sea's surface rules larger in the deeper parts of the gulf than in the shallower water, as a rule it is most densely aggregated in the coastal belt within the 150-meter contour and in the shallows of Georges Bank, no. matter which calculation be employed. This was true, also, in the summer of 1913. In fact, the northeastern part of the deep basin, where the water has proved very productive on several occasions in summer and early autumn, as well as in late spring, has been the only exception to this rule for any time of year. Enough hauls have now been made to show that the zooplankton (especially the Crustacea) is usually most densely congregated, summer after summer, in four rather definite areas — (1) over the eastern end of Georges Bank, (2) in the shoal water south of Cape Sable, (3) in the deep northeastern basin, and (4) off Massachu- setts Bay out to the 100-meter contour (fig. 39). At the other extreme the western and southern parts of the deep basin and the coastal belt inside the 100-meter contour east of Penobscot Bay have never yielded as much as 2 cubic centimeters of plankton to the cubic meter of water at any season by either mode of calculation, nor has the water over the coast bank west of Nova Scotia proved productive except for the Pleurobrachia swarms so characteristic of that locality (p. 19). PLANKTON OF THE GULF OF MAINE 95 The most abundant concentrations of plankton which we have yet encountered in the Gulf of Maine have been off Cape Cod on May 26, 1915 (station 10279, nearly 4 cubic centimeters per cubic meter) ; on the eastern part of Georges Bank on July 7V 70- 69' 68' 67* 66* ' $Jst/) ' '"' ^ ' — / \ /v^ ) V^ .'.'. +- + PENbB&COff /Vj ^ , /"' ■■•■■'''$( /bay>5 vK ' < ' / NOVA / <£^ o ^f .' " / n y S3 /) ,-i ^ '•■ \ SCOTIA /L^f/ + ./ " + *■ 1 I + • ''4' Portland C wffv f-i * 01 .? (Yarmouth 1/1 CAPE \ ,-j' • ; (, 5ABL Ri -V, J% X* C ; x X ° ?.■■-. V X ; \ /" u ■ : + <2 yv u \ i ^ ,. 3 * ( ffl" + + + 4- +- / + or '-■,.....--,. ■ ' 40 + 4. + 4- 4- 4- w 71" 70- •»" 60' 67 - 66" Fig. 39. — Locations where vertical hauls have taken more than 2 cubic centimeters of animal plankton per cubic meter at different seasons, calculated by the method described on page 94. X, September to November; O. May; July to August 15 23, 1914 (station 10224, about 5 cubic centimeters per cubic meter); in the eastern basin on September 1, 1915 (station 10309, approximately 3.5 cubic centimeters per cubic meter, assuming some stratification) ; and at the mouth of Massachusetts Bay 7589S— 26 7 96 BULLETIN OF THE BUREAU OP FISHERIES in July, 1916 (station 10342, at least 4.5 cubic centimeters per cubic meter); but occasionally it is much more dense than this at one level or other, the volumes just listed being the minima possible. For example, a horizontal haul of 15 minutes' duration at 40 meters depth, with a net 1 meter in diameter, off Cape Cod on July 22, 1916 (station 10344), yielded over 6 liters, mostly copepods, which is equivalent to about 12 cubic centimeters per cubic meter for the water fished through (the tow covered about one-third of a mile). In fact, it was the richest tow-net catch we have ever made in the gulf, although the vertical haul indicated only about 2.8 cubic centimeters of plankton per cubic meter. ANNUAL VARIATIONS IN ABUNDANCE Annual variations in the amount of zooplankton living in the waters of the gulf will mirror the long-time fluctuations in its physical state — may, indeed, be the best clue to such — and exert an important influence on the growth, local repro- duction, and distribution of the adults of such important plankton-feeding fishes as herring, mackerel, and pollock. It is certain that considerable fluctuations of this sort in the plankton do take place from year to year, as illustrated by the following table of the volumes per square meter of sea surface for corresponding localities in the summers of 1913-14 and the first week of September, 1915. 45 Locality Stations Plankton, in cubic centimeters per square meter 1912 1913 1914 1915 1912 1913 1914 1915 i 10002 i 10007 10087 10089 10090 10092 10095 10096 10098 10100 10101 10253 10254 10255 10249 10244 10245 10246 10247 10248 10250 10306 10307 250 65 180 80 120 160 60 120 70 220 100 60 200 70 105 15 60 200 10 100 350 110 165 10028 10309 10311 10315 10316 30 30 30 '25 425 45 OH Lurcher Shoal 10031 10036 10033 50 12.5 10038 ' 10318 20 25 74 123 117 117 ' July hauls. ' A few miles west of the corresponding stations, 1912 to 1914. * From horizontal hauls. According to these measurements the volume of the plankton was greater in 1913 than in 1914 at all but two stations. As between 1913 and 1915, however, one year was the richer at some, the other at other localities. However, since the average is practically the same (or at least did not differ as widely as the probable error) for the three years, there was apparently no important general change in the amount of plankton existent in the gulf from 1913 to 1915, though both these years were apparently decidedly more productive, on the whole, than was 1912 during the corresponding months (Bigelow, 1915, p. 337). During the summer of 1916 (a year of low temperatures) the waters off Massachusetts Bay proved more produc- « Although different types of nets were used during these years, the results, reduced to the common standard, will allow a rough and ready comparison. PLANKTON OF THE GULF OF MAINE 97 tive than we have previously found them at that season, thanks to the abundance of large Calanus, with volumes of plankton per square meter for six stations along the shore from Cape Ann to southern Cape Cod (July 19, 1922) ranging from 135 to 250 cubic centimeters (average 208 cubic centimeters) , and it was then that we made the exceptionally rich horizontal net haul already mentioned (p. 96). Notes on the yearly numerical fluctuations in the local stock of the commoner copepods will be found under the discussions of the several species. PLANKTON AS FOOD FOR WHALES AND FISHES We might, figuratively, conceive of the swimming and floating life of the sea as a pyramid, with the microscopic plants as its base and the large sharks and whales as its apex, the latter few in numbers but each enormously destructive of the smaller organisms on which it preys. The general thesis that the smaller plankton, animal and vegetable, is practically the sole food supply for young marine fishes no longer requires further proof or argument. It likewise so serves for many species of fish when adult, especially for the schooling fishes, such as herrings, menhaden, mackerel, shad, and the like. The large adult gadoids, too, feed on plankton to a greater extent than is generally appreciated. The great basking shark (Cetorhinus maximus), which is still an occasional visitor to the gulf, is exclusively a plankton feeder throughout its life, and most of the northern whalebone whales have long been known to subsist largely on the smaller pelagic animals — several of them exclusively so — a fact widely heralded in zoological textbooks. The literature dealing with the dependence of the larger marine animals on the plankton has grown to formidable dimensions in the last half century, but very few first-hand observations have yet been made on the relationships between fish and plankton in the Gulf of Maine. So far as these go, however, they show that what is true of north European seas in this respect applies equally to American waters, as, indeed, might have been prophesied, allowing for the differences between the composition of the planktonic communities of the two sides of the north Atlantic Ocean. In the Gulf of Maine the groups of Crustacea that are of chief importance in the diets of adult fishes and whales are the copepods and the euphausiids. Exami- nation of stomach contents at European whaling stations has proved that instead of subsisting indiscriminately on all sorts of plankton, large and small (as has some- times been taken for granted) , or on pteropods (as the Arctic right whale often does) , the planktonic part of the diet of the other species of whalebone whales common in boreal seas consists almost exclusively of these two groups of Crustacea. While there is ample ground for the choice of a crustacean rather than a molluscan diet in the greater abundance of the former than of the latter on both sides of the north Atlantic, it is possible that the whales in question may voluntarily prefer the harder and more oily shrimps and copepods. The finback (Balxnoptera physalus Linne), commonest whale in the Gulf of Maine to-day, eats a mixed diet of plankton and fish, devouring the latter, particu- larly the herring, in great numbers, but probably depending more on the smaller pelagic animals in the long run. A considerable number of finback stomachs have 98 BULLETIN OF THE BUREAU OF FISHERIES now been examined by various observers, and in every case (apart from fish) they have been packed with euphausiids and with euphausiids alone. Thus G. M. Allen (1916, p. 200) writes that "on the Newfoundland coast stomachs of several finbacks which I examined contained enormous quantities of the small shrimplike schizopod Thysanoessa inermis." Lillie (1910), too, found the stomach contents of several finbacks taken off Ireland in July and August to consist altogether of euphausiids (in this case Meganyctiphanes) and of fish; and in more than 150 finbacks killed at the Belmullet whaling station on the west coast of Ireland, Burfield (1913) and Hamilton (1915 and 1916) found nothing but immense numbers of these same pelagic sin-imps (Meganyctiphanes), with occasional fragments of fish. Nor have I been able to find any definite evidence that this whale ever succeeds in capturing copepods, or any of the smaller plankton for that matter, though, according to Murie (1865), the stomach of one captured near Gravesend, England, contained fragments of medusae as well as of Crustacea. In short, euphausiids, and these alone, are its support, apart from fish. The Atlantic humpback (Megaptera nodosa), which is not uncommon off the New England coast, though never so plentiful there as the Atlantic right whale once was or as the finback now is, subsists on much the same diet as the latter — viz, fish and pelagic shrimps (euphausiids) — while Andrews (1909) found its close ally, the Pacific humpback, feeding on the latter alone; smaller planktonic animals have never been found in humpback stomachs so far as I am aware. The blue whale, or sulphur bottom (Balsenoptera rrmsculus), which is not un- common along the coasts of the Gulf of Maine and is numerous in Newfoundland waters, is even more dependent on euphausiids than are the two whales previously mentioned, for it is not known to eat fish at all, on the one hand, or copepods, on the other. All the sulphur-bottom stomachs recently examined (a considerable number in the total) have been packed with euphausiids alone — Thysanoessa in whales from Newfoundland (G. M. Allen, 1916), Meganyctiphanes in others taken off the west of Ireland (Lillie, 1910; Burfield, 1913; Hamilton, 1915 and 1916), and Euphausia in the Antarctic (Liouville, 1913). The destructiveness of these huge mammals is illustrated by Collett's (1877, p. 161) statements that sulphur-bottom stomachs frequently contain 300 to 400 liters of shrimps, and that occasionally one is taken crammed with up to 1,200 liters of Thysanoessa. Andrews (1916), too, writes that this whale feeds exclusively on euphausiids; Millais (1906), however, credits it with a copepod diet. The North Atlantic right whale (Eubalasna glacialis), once common in New England waters though now unhappily nearly extinct there (and with it the glories of the New England coastwise whale fishery), subsists largely on euphausiids, notably on Thysanoessa (Kukenthal, 1900). Collett (1909), indeed, found nothing else in right whales taken off the Hebrides and off Iceland. The only eyewitness's account of its feeding habits in New England waters, for which we must turn back nearly 200 years (Dudley, 1734, quoted by G. M. Allen, 1916) tells of "this whale, in still weather, skimming on the surface of the water to take in a sort of reddish spawn or brett, as some call it, that at some times will lie on the top of the water for a mile together." From its geographic situation and mode of occurrence this Bull. U. S. H. F.. 1924. i Doc. 968.) FlG. 40. — Marginal fringe on one of the whalebone plates of a finback whale (BaUcnnptrra physalus) from the Gulf of St. Lawrence. Natural size Fig. ii. -Marginal fringe ob one ol the whalebone plates of a pollock whale ( Bala noptera boretUis) from the Gulf of Maine. Natural size PLANKTON OF THE GULF OF MAINE 99 was probably Calanus or otber copcpods. Unfortunately, little is known of tbe babits of the Atlantic right whale, but it is well established that the pollock whale (Balsenoptera borealis) feeds chiefly on copepods at certain times and places, for Collett (1886, p. 26) found the stomachs of several, killed off East Finmark in July, "filled with a fine gritty mass, which consisted entirely of Calanus finmarchicus," with the Calanus occurring "in great numbers and in a tolerable state of preserva- tion" among the hairs of the baleen plates; and since he gives excellent figures of these copepods, their specific identification is assured. In West Finmark, however, this same whale has been reported as subsisting chiefly on euphausiids (Collett, 1886). Kukenthal (1900) likewise states that it feeds on these shrimps, and Andrews (1916) writes that most of the specimens which he opened in Japanese waters contained euphausiids only, while a few had eaten fish. G. M. Allen (1916) and Millais (1906) are therefore fully justified in crediting it with a mixed copepod (Calanus and Temora) and euphausiid diet. The fact that only two of the species of whalebone whales known to occur in the Gulf of Maine eat copepods, while all feed on euphausiids, seems not to have been appreciated, though established past cavil by the analyses of stomach contents just mentioned. It is, I think, impossible to explain this preference for shrimps on the ground of voluntary selection, for while it is not unreasonable to suppose that whales follow the schools of Crustacea rather than the soft-bodied Sagitte, ccelenterates, or mollusks, copepods (and particularly Calanus) usually abound in northern seas wherever euphausiids are plentiful, and finback, pollock whale, and right whale must gather them all, the large with the small, into their open and expectant mouths as they swim. With whales, however, just as with tow nets of different mesh, the fineness of the straining apparatus determines what part of the total planktonic population is retained to serve as food. If the whalebone be coarse or comblike, as it is in the finback whale (fig. 40), the blue whale, and the humpback, objects as small as copepods are driven out through the sieve with the outrush of water when the mouth is closed, while the much larger euphausiids are retained. The pollock whale, however, possesses, in the "unusually fine and curly, almost wooly bristles" on the inner side of the baleen plates (fig. 41), so well described by Collett (1886, p. 263), a straining apparatus so much more efficient as to sift out the copepods as well as the larger crustaceans. This is true also of the right whale, with its silky-fine baleen (Collett, 1909, p. 95) and ability to strain large volumes of water with little effort. 46 However, the finer the strainer and the better adapted for the capture of the smaller animals, the less effective it is for capturing fish, as witness the depend- ence of the pollock whale on plankton contrasted with the piscivorous habit of the finback. The fertility of the gulf as a feeding ground for whales depends, then, not only on the total amount and local concentration of the plankton or on its nature — whether or not crustacean — but equally on the size of the units of which it is composed. Thus, the abundance of Calanus in Massachusetts Bay and off northern Cape Cod " For a general account of its feeding habits see Beddard, 1900. 100 BULLETIN OF THE BUREAU OF FISHERIES provided an ideal pasture for the Atlantic right whale, of which it once fully availed itself, as early records show, but not for the finback, for which the bay is a desert except when herring or other fish are schooling there or during the brief local swarm- ings of euphausiids. It is common knowledge among fishermen that finbacks seldom appear in any numbers anywhere in the gulf except when in pursuit of fish. It is also probable, that the volumetric preponderance of copepods over euphausiids in most parts of the gulf explains the comparative rarity there of the shrimp-eating blue whale with its very coarse whalebone. Before leaving this subject I should emphasize that the large, easily recog- nized, pelagic amphipod Euthemisto, locally and temporarily so abundant, has never been recognized in the stomachs of any of the whalebone whales. Is it not eaten? And if not, why not? It is probable that copepods are the main dependence of the basking shark (Cetorliinus maximus), whose gillrakers perform the same service in filtering its crustacean food from the water taken into the mouth as do the baleen plates of the whalebone whales. I need merely point out that the alimentary canal of a speci- men taken at West Hampton Beach, Long Island, on June 29, 1915, contained a large quantity of minute Crustacea, "whose reddish bodies lent color to the entire mass" (Hussakof, 1915, p. 26). When we turn to the dependence of the smaller fishes on crustacean plankton, we are confronted by a published record so embarrassing for its wealth (mostly, however, based on experiences in European seas) that I shall lay only a few of the more typical examples before the reader, and those most applicable to the Gulf of Maine. The unicellular plants have been described repeatedly in zoological literature as the chief food supply of the youngest larval fishes, and a long list of diatom and peri- dinean species has, at one time or another, been recorded as having been eaten by them; but recent studies of the stomach contents of large series of various common fishes in the English Channel (Lebour, 1919, 1920, 1924) have proved that although many fish do take more or less diatoms, peridinians, etc., few depend on these uni- cellular forms to the extent that has been generally supposed, even during their earliest larval stage (cf. also Hjort, 1914, p. 205), but begin to take larval copepods and other microscopic animals by the time the yolk sac is absorbed, if not sooner. However, Lebour found the young European flounder (Pleuronectes flesus) subsisting chiefly on the green flagellate genus Phseocystis up to the time of its metamorphosis, with other flatfish taking a considerable proportion of peridinians and diatoms, and this proved true of young herring less than 10 millimeters long, which also take Halo- sphsera. Outside of the littoral zone, where the mummichogs (Fundulus Tieteroclitus) consume diatoms as well as other small organisms indiscriminately, the menhaden is the oidy important Gulf of Maine fish that continues throughout life to subsist chiefly on diatoms and peridinians, with the most minute of Crustacea and other animals. These it is enabled to sift out of the water by its fine branchial sieve, as Peck (1894) long ago described. 47 " On the feeding habits of the menhaden see also Bigelow and Welsh, 1925, p. 123. Bull. U. S. B. F., 1924. i Doc. 968.) 'i 1 '- 12 Segments of the branchial iieves of tin oinmon Ashe; thai feed "ii plankton, a. Menhaden, I ecoo tyrannus; b, herring, Clupea harengus; r, mackerel, Scombei scombrur; d, mackerel, side view ol Bill raker, with gill spines, x 25 PLANKTON OF THE GULF OF MAINE 101 The menhaden has no rival among the fishes of the gulf in its utilization of this pelagic vegetable pasture (indeed, Peck (1894) so noted) ; nor is any other local species possessed of a filtering apparatus comparable to that of the menhaden (fig. 42a) for fineness and efficiency, though in European waters its relative, the sardine (Clupea pilcTiardus) , feeds equally on microscopic plankton as well as on copepods". The Pacific anchovy also feeds on diatoms and peridinians as well as on zooplankton (W. E. Allen, 1921, p. 54). 48 Among clupeoids, as among whalebone whales, a direct relationship obtains between the fineness of the sieve through which the water taken in through the mouth is strained — in this case the gillrakers — and the minimum size of the organisms that can be retained and utilized; everything smaller passes through. Even the menhaden (though most of its food is microscopic) is unable to capture the very smallest organisms, such as coccolithophorids and infusoria; and the herring and alewife, with coarser sieves (fig. 42b), subsist chiefly on organisms with a longest dimension of at least 0.5 millimeter (copepods or larger animals), which they select individually and not by swimming open-mouthed, as the menhaden does 49 (Bigelow and Welsh, 1925, p. 103). Experience with the tow net shows that if diatoms are plentiful enough they will be picked up by a coarse mesh, and the mackerel, which carries broadly spaced spines on the long rakers on the foremost gill arch (figs. 42c and 42d) consumes more or less pelagic plants, and especially the diatom genera Lauderia and Chsetoceros, in British waters in winter when the fish are in deep water (Bullen, 1908 and 1912). I know of no direct evidence, however, that mackerel ever feed on diatoms or peri- dinians in the Gulf of Maine unless taken accidentally along with other plankton. Pelagic Crustacea of one kind or another form the major part of the diet of the adults of all plankton-feeding fishes other than the menhaden in the Gulf of Maine and in northern seas generally, and of the fry of all Gulf of Maine fishes, the sundry crustacean members of the plankton appearing in the lists of stomach contents with monotonous regularity. For most species of fish, indeed, this is true from the earlier larval stages onward, as just noted. In fact, Lebour (1920 and 1924) found that herring, and others as well, devour larval mollusks, small Crustacea, etc., even before the yolk sac is absorbed. Thereafter the diet of all the species of fish which she studied consisted chiefly of the latter, most frequently of copepods, adult and larval, and of Cladocera, with decapod and other larvae playing a secondary role and microscopic plants taken only vicariously, except that some larval herring had fed to some extent on unicellular organisms. Perhaps the most interesting result of Lebour's work, apart from her general conclusion (1920, p. 262) that copepods, other Entomostraca, and molluscan larvae are the chief food of nearly all young sea fish, is that "usually each species of fish selects its own favorite food, to which it keeps, indiscriminate feeding seldom or never taking place." It would not be safe to postulate the precise larval food of any of the Gulf of Maine flounders from that of their European congeners, so widely do the latter '» Mullets also subsist largely on unicellular plants, but they are only accidental visitors to the cool waters ef the Gulf of Maine. " It is easy to watch them doing so in the aquarium. 102 BULLETIN OF THE BUREAU OF FISHEBIES differ among themselves in their choice of diet, 50 nor were any of the gadoids common to American and North European waters studied by Lebour. However, several North Sea members of the family were feeding on small copepods — mainly Pseudo- calanus — and Calanus was taken freely as the larval fishes grew in size. Dannevig, too, wfites that numbers of newly-hatched cod placed under observation at the hatchery at Flodevigen, Norway, took no food until the yolk sac had been absorbed, and thereafter fed from the first on such animals as mollusk larva?, nauplii, etc., "seeming to despise the innumerable diatom forms which are likewise present in the water" (Dannevig, 1919, p. 48). Evidently this applies to the American cod as well, because young fish 12 to 20 millimeters long have been observed to feed exclusively on copepods at Woods Hole (Bumpus, 1898), and according to Mead (1898) copepods are likewise the favorite diet there for young sculpins and sand launce (Ammodytes). Judging from the general similarity between the planktonic communities of the two sides of the North Atlantic, there is every reason to assume that the dietary lists which Lebour gives for very young herring and mackerel would apply as well (in a general way) to the Gulf of Maine as to the North Sea. For the former species this diet consisted chiefly of larval gastropods, with copepods, particularly Pseudo- calanus, next in importance, barnacle (Balanus) and bivalve larvae in smaller amounts, and with unicellular forms, as just noted (curiously enough, out of about 1,000 specimens 8 to 15 millimeters in length over 700 contained no food); while the young mackerel had eaten copepod nauplii (chiefly Calanus and Temora) and crustacean (probably copepod) eggs, with a few ostracods, euphausiid larvae, and even young fish. In Norwegian waters, according to Nordgaard (1907), the older herring feed chiefly on euphausiids and copepods, especially the genera Calanus and Temora, with ostracods, tintinnids, larval barnacles, Halospha?ra, and other small members of the plankton consumed in smaller amounts. Copepods and -euphausiids together constitute almost the entire diet of the herring in the Gulf of Maine, with fish smaller than about 4 inches long taking chiefly the former and larger ones taking both at localities where they are available (Moore, 1898; Bigelow and Welsh, 1925, p. 103). Young herring, taken while feeding on the surface at Woods Hole, have been found full of copepods of several species. What is known of the feeding habits of the alewife {Pomolobus pseudoharengus) , and blueback (Pomolobus sestivalis), is to the effect that they also subsist chiefly on these two groups of Crustacea during the part of the year when they are in salt water, and that shad (Alosa sapidissima) subsist on copepods and mysid shrimps. Mackerel, in the Gulf of Maine, have also long been known to feed greedily on calanoid copepods (the "red feed" or "cayenne" of which fishermen often describe the fish as crammed full) . I have found fish, taken off Cape Elizabeth, August 12, 1912, packed with Calanus finmarchicus and Pseudoca- lanm elongatus; Goode (1884a) found the stomachs of mackerel, taken off Portland in 1874, full of large, copepods and euphausiids. The schools of mackerel frequenting the Bay of Fundy have also been reported as following and preying upon the shoals of " So far as I can learn there is no record of the stomach contents of the larval witch (Glyptocephalus) or American plaice (Hippoglossoides). PLANKTON OF THE GULF OF MAINE 103 shrimp (Meganyctiphanes and Thysanoessa) , which so often appear on the surface there (S. I. Smith, 1879). Richard Rathbun (1889) reports some of the mackerel that he examined from the southern fishery (off the coasts of Virginia and Maryland in latitudes 37° 4S' N. and 38° 01' N.; longitudes 74° 13' and 74° 21' W.) in 1887, as full of copepods and others of euphausiids. Dr. W. C. Kendall found the mackerel on the northern part of Georges Bank feeding on Calanus (probably also Pseudoca- lanus) and on small brown copepods (probably Temora), as well as on other plank- tonic animals (Bigelow and Welsh, 1925, p. 201) ; and many more instances might be mentioned where copepods, euphausiids, or both, have been reported as mackerel food in American waters as well as in European. The larger copepods also enter to some extent into the dietary of the American pollock (Pollachius virens) in the Gulf of Maine — witness Willey's (1921) record of a fish taken near Campobello Island with many Euchxta norvegica in its stomach and some Calanus JinmarcMcus and C. hyperboreus. Euphausiid shrimps offer as important a food supply for this large and active gadoid as do small fish. Thus, Moore (1898) describes pollock at Eastport as feed- ing chiefly on them and following them in their appearances and disappearances. Wflley (1921) also found pollock feeding on euphausiids at Campobello. Welsh saw great numbers of pollock schooling in pursuit of shrimps and greedily feeding on them in the neighborhood of the Isles of Shoals in spring, as I have described elsewhere (Bigelow and Welsh, 1925, p. 401). In the North Sea region medium-sized specimens of this gadoid (there called the "coalfish" or "green cod") eat considerable amounts of small pelagic Crustacea, such as Calanus, Temora, Centropages, Pseudocalanus, cirriped larva?, ostracods (Evadne), as well as euphausiids, in addition to the small fish and to the bottom- dwelling worms and Crustacea that form their staple food. It is probable that when euphausiids descend toward the bottom in the Gulf of Maine they become food for the hakes (genus Urophycis), which, in the main, are shrimp eaters (Bigelow and Welsh, 1925, p. 450), and which are known to gorge on euphausiids along the outer part of the continental shelf (Hansen, 1915, p. 94). So, too, the deep-water fish Macrourus (Bigelow and Welsh, 1925, p. 470) ; and even as typical a bottom and fish feeder as the cod is known to adopt a pelagic life and to feed on euphausiids off the north and east coasts of Iceland (Paulsen, 1909, p. 39; Schmidt, 1904). The common skate {Raja erinacea) also feeds on copepods on occasion (Linton, 1901, p. 279), though this is quite exceptional for it. In North European waters the hyperiid amphipods are a major food for herring (Brook and Calderwood, 1886), but although the genus Euthemisto is widespread and at times locally abundant in the Gulf of Maine, I have found no record of herring feeding on it there, and have recognized none in the stomachs of the Gulf of Maine herring I have opened. Probably this is due to the mutual geographic distri- bution of the two animals, Euthemisto being most plentiful offshore and herring along the coast. These amphipods may be expected to form an important item in the diet of herring on Georges Bank. This is certainly true of the mackerel there, for Dr. W. C. Kendall found the latter feeding on Euthemisto on the northern part of the Bank in August, 1896 (Bigelow and Welsh, 1925, p. 201) . Mackerel taken 104 BULLETIN OF THE BUREAU OF FISHERIES near Woods Hole in summer have also contained Euthemisto (Rathbun, 1896), and Rathbun (1889) found mackerel feeding largely on amphipods off Virginia and Maryland in the spring. European mackerel also feed on Euthemisto, and, generally speaking, the latter are no doubt more important as a source of fish food over the outer part of the shelf and along the continental edge (where they are constantly abundant) than in the inner part of the Gulf of Maine; but no evidence is at hand that any Gulf of Maine fishes depend on them to the extent to which the long-finned albacore (Germo alalunga) does off the French coast (Le Danois, 1921). Whenever and wherever the larvae of decapods are plentiful, all plankton- eating fishes feed on them greedily. In the Gulf of Maine the "megalops" stages of crabs are of considerable economic importance in this respect. Linton (1901 and 1901a), for example, found many young herring at Woods Hole full of them, and Doctor Kendall in his field notes records some of the fish in certain schools of Georges Bank mackerel as packed with them, almost to the exclusion of other plankton. Larval shrimps, prawns, and lobsters also enter regularly into the dietary of many fishes in European seas, notably the various clupeoids. In Swedish waters the young stages of bottom-dwelling shrimps are regularly consumed by mackerel (Nilsson, 1914); no doubt also in the Gulf of Maine, though definite information so far available on this point is scanty. Adult decapods hardly enter into the plankton of the Gulf of Maine, except for the large deep-water prawn Pasiphsea, which may be expected to prove a staple food for hake (genus Urophycis) . Sagittae are eaten in considerable quantity by mackerel. Rathbun (1889), for example, found them in fish taken in the southern fishery off the Middle Atlantic States, and Doctor Kendall, in his notes, records some of the mackerel taken on the northern part of Georges Bank during the last week of August, 1896, as full of them. Sagittse probably will be found to enter largely into the dietary of the mackerel in Massachusetts Bay in early summer; in fact, whenever they are plentiful (p. 18). They are also eaten by herring in Scottish waters (Brook and Calderwood, 1886), and probably this will also prove to be the case to greater or less extent in the Gulf of Maine. In the Adriatic Sagittse are also the chief dependence of the young goosefish (Lophius piscatorius) while it lives pelagic (Stiasny, 1911), which probably applies equally to the Gulf of Maine goosefish (Bigelow and Welsh, 1925, p. 526). The American pollock also consumes Sagittae in the Gulf of Maine (Willey, 1921). The shell-bearing pteropods, represented locally by Limacina retroversa, are seldom plentiful enough in the Gulf of Maine to be of much importance as a possible food supply for the schooling fishes there, but when these mollusks do swarm mackerel would no doubt feast on them, for they are an important food for this fish off the west coast of Ireland (Massy, 1909). According to Rathbun (1889), mackerel eat L. retroversa off the Middle Atlantic States, and mackerel taken off No Mans Land (an islet near Marthas Vineyard) have been recorded as full of them. In Norwegian waters, according to Nordgaard (1907), this pteropod also enters into the dietary of the herring, but as Limacina seems not to have been recorded as herring food else- where in north European seas it probably does not so serve to any great extent in the Gulf of Maine. Lebour's (1920) observation that young fish of various species not only had not eaten Limacina, although the latter were plentiful in the tow, but PLANKTON OF THE GULF OF MAINE 105 refused thorn when offered in the aquarium is interesting as suggesting that the mack- erel is rather an exception in feeding on this pteropod. Naked pteropods are never plentiful enough in the Gulf of Maine to be of any importance as food for larger animals. Probably all the fishes that eat plankton consume buoyant fish eggs to some extent, the amount taken depending chiefly on the local supply conveniently available. Thus Brook and Calderwood (1SS6) found fish ova more or less prominent in the diet of Scottish herring, according to the varying abundance of the eggs in the plankton, and although fish eggs have not actually been recorded from the stomachs of Gulf of Maine herring there is no reason to doubt that the latter consume them whenever they offer, as is also the case in the English Channel, according to Lebour's (1924a) recent studies. Mackerel also are known to take eggs of their own as well as of other species. Fish eggs have been found in small mackerel from the Woods Hole region, to quote a local instance, and in European seas medium-sized specimens of the American pollock {PollacTiius virens) eat considerable amounts of fish eggs among other plankton. The only groups of planktonic animals sufficiently plentiful in the Gulf of Maine to be of any importance in its natural economy, but which are not regularly con- sumed by its fishes in as large quantities as the supply allows, are the medusae, siphonophoras, and ctenophores. E. J. Allen (1908) and Goode (1884 and 1884a) record medusas and siphonophores from mackerel stomachs; but this is exceptional, and although they may bite out pieces of large medusas this is probably for the sake of the amphipods (Hyperia) living within the cavities of the latter (Nilsson, 1914). It would not be surprising to find mackerel gorging on Pleurobrachia in the Gulf of Maine at the places and times when this ctenophore swarms, for Andrew Scott (1924) reports mackerel in the Irish Sea full of them during one of their incursions. The spiny dogfish {Squalus acanihias) feeds to some extent on ctenophores (Pleurobrachia) in spring, the fish often containing them when they first appear at Woods Hole in May; and in north European waters this troublesome little shark sometimes devours ctenophores in such quantity that their stomachs are full of them (Mortensen, 1912, p. 72, fide Dr. C. G. J. Petersen). The lumpfish likewise feeds regularly on medusas and ctenophores in European waters, hence probably in the Gulf of Maine, and the sunfish (Mola mola), which is only an accidental visitor to the gulf, subsists chiefly on these watery organisms (Bigelow and Welsh, 1925, p. 303) ; but so far as is known neither the herring tribe nor any of the gadoids ever eat them — in fact, no Gulf of Maine fishes other than those just mentioned. With the young fry of the whole fish population of northern seas dependent for their existence on the supply of plankton, it is but natural that many attempts should have been made to correlate the movements and migrations of the more important food fishes with local and temporal Quotations in the supply, either of the plankton as a whole or of such members of it as serve as the chief diet of the particular species in question, as well as with the far-reaching physical phenomena that may be looked on as the ultimate causes of such fluctuations. Thus, to mention only a couple of examples, Bullen (1908) has established at least a plausible causal 106 BULLETIN OF THE BUREAU OF FISHERIES relationship between the fluctuations in the amount of zooplankton present in the sea and in the seasonal and yearly catch of mackerel, corroborated by experience for herring, also, in the Irish Sea (A. Scott, 1924) ; and E. J. Allen (1908) aroused an interesting discussion by his tentative hypothesis that the abundance of mackerel at any given locality depends on the amount of sunshine during the previous months, sunny weather favoring the multiplication of diatoms and thus affording a rich pasture for copepods, an abundant stock of which attracts mackerel. Dr. C. B. Wilson, in a letter, suggests that the diurnal migrations of copepods upward toward the surface at night and downward by day may be the reason why mackerel and herring most often school at the surface at night, following the daily migrations of their prey. To attempt to connect the fluctuations in the stock or the movements of the fish population of the gulf, even of such typical plankton feeders as the herring, with variations in the supply of plankton is as yet out of the question, neither digested statistics of the catch of the former nor sufficiently definite information as to the latter having been gathered. However, it is evident that a correlation between the two must exist, and, as Dr. C. B. Wilson writes, "anything that contributes to a detailed knowledge of the presence and movements of the copepods throughout the year will give us information as to the movements and distribution of the fish," and is therefore of as direct interest to the fisherman as to the scientist. FOOD OF THE PLANKTON The study of the stomach contents of the smaller pelagic animals, which to- gether make up the zooplankton, is, as Steuer (1910, p. 622) points out, beset by many obstacles, principal among which is the rapidity with which the various organic substances are digested after being eaten, leaving as recognizable in the masticated or half-digested state only such objects as are provided with spines, bristles, etc., or with calcareous or silicious shells of characteristic outline. Then, too, it is a common experience to find whole series of animals, even of the larger species, perfectly empty. In spite of these difficulties, however, so considerable a body of observations has been accumulated that the general diet of most of the important planktonic groups can now be stated with some confidence, and although little attention has yet been paid to the diets of the plankton of the Gulf of Maine, there is no reason to suppose that the feeding habits of its various members differ essentially from those of their north European representatives. Among the zooplankton, as among the pelagic fishes, some species or groups are carnivorous while others depend for subsistence on the unicellular vegetable life of the high seas, but within the various groups the smaller planktonic animals are decidedly uniform in their feeding habits. Perhaps as striking an illustration of the carnivorous habit as any is afforded by naked pteropods such as Clione limacina, which, so far as known, live exclusively on other pelagic animals and most often on their own shell- bearing relatives (for instance, on Limacina), which they devour by thrusting the protrusible proboscis into the shell and tearing the inmate to pieces in spite of its futile efforts to escape by contracting into the smallest possible compass, as Schie- menz (1906, p. 29) has so graphically described. PLANKTON OF THE GULF OF MAINE 107 Equally voracious, and far more destructive to smaller animals in the Gulf of Maine because of its greater abundance there, is the pelagic amphipod Euthemisto. The few Euthemisto stomachs which I have examined all contained copepods, often so nearly intact as to show that they had been swallowed whole and were not torn to pieces by their captor's mandibles. In seven Euthemisto upwards of 20 millimeters long, from several localities (stations 10294, 10296, and 10307), the stomachs were packed with copepods (mostly Calanus, but occasionally Temora) , with more or less other crustacean debris, parts of legs, antennae, etc., and in one instance a fish egg. The presence of an entire young Euthemisto in the stomach of one adult shows that this amphipod, like so many other marine animals, is cannibalistic when opportunity offers. Euthemisto is so large and so active that wherever it is abundant it must wreak havoc among the Calanus hordes among which it swims. Probably it materially decimates the stock of copepods existing all along the outer edge of the continental shelf (p. 165), and it may also be a serious enemy to them locally and temporarily within the gulf. Small individuals of Euthemisto feed on unicellular organisms as well as on Crustacea, specimens about 10 millimeters long 51 from the western basin, August 31, 1915 (station 10307), containing more radiolarians (Acan- thometron) than copepods. Decapod larvae, so abundant at times in shallows and in coastwise waters, are also, as a rule, carnivorous in their later stages (vide Steuer's (1910, p. 631) account of zoeas devouring young fish, smaller Crustacea, etc.). Lobster larvae also feed greedily on other young decapods of smaller size (Weldon and Fowler 1890), their cannibalistic habit being the bane of the fish-culturist. Lebour (1922), however, describes crab zoeas as also eating green plant cells, Phaeocystis, and diatoms, most often Coscinodiscus among the latter. The young lobster also consumes diatoms in large amount, likewise fragments of algae during its pelagic life (Herrick, 1S96), and this is probably true of most other decapods, if not of all Crustacean larvae at least when they are newly hatched and until they are large enough to capture and subdue more active organisms. Sagittae are strictly carnivorous and so active, fierce, and well-armed that it is no wonder they are recorded as feeding on things as far apart as tintinnids, crustaceans, other Sagittae, and young fish. Among the Gulf of Maine species, 5. maxima is notable in this respect, for while the commoner S. elegans and EulcroTinia hamata are usually empty or contain, at most, oil globules or unrecognizable debris, I have on several occasions found S. maxima that had perished in the preservative while in the act of devouring animals as large as Euchaeta and Tomopteris, as well as their own kind, or containing in their guts newly-swallowed copepods or smaller Sagittae of other species. Lebour (1922 and 1923) speaks of the larval herring as frequently falling victim to Sagittae, which may be serious enemies when as plentiful as they often are in the Gulf of Maine. It is probable that the comparative scarcity of copepods, often remarked at the precise levels, localities, or times when Sagittae abound, is direct evidence of the extent to which the latter may reduce the stock of their prey. But of all the members of the plankton, the most destructive to smaller or weaker animals are the " Euthemisto as small as this can contain but one or two large copepods at the most. 108 BULLETIN OP THE BUREAU OF FISHERIES several coelenterates, and especially the ctenophore genus Pleurobrachia, a pirate to which no living creature small enough for it to capture and swallow comes amiss. Small Crustacea of all kinds, other coelenterates, Sagittre, fish eggs, and even fish of considerable size all are devoured, and so clean does it sweep the water with its trailing tentacles that wherever these ctenophores abound practically all of the smaller animals are soon exterminated. The larger ctenophore Beroe is even more voracious, though, fortunately for the productivity of our seas, it is less numerous than Pleurobrachia. As Chun (1S80) long ago observed and graphically described, Beroe feeds on its own relatives, even on other ctenophores many times as large as itself, as well as on whatever else it can capture. Lebour (1922 and 1923) found it dieting chiefly on Pleurobrachia, also to some extent on other ctenophores and diatoms, while we ourselves have often found Calanus and other copepods in its gastric cavity. Mertensia is no less voracious, for I have seen one individual of this genus which "had entirely engulfed a young sculpin (Acanthocottus grcenlandicus Fabricius) no less than 21 millimeters long, the victim being doubled up so as to fit into the digestive cavity of its captor" (Bigelow, 1909a, p. 317). The various species of medusaj, large and small, all belong to the piratical category, and the total destruc- tion they wreak on euphausiids, copepods, appendicularians, the various larval forms, etc., is beyond any estimation. Even animals as active and themselves as voracious as Sagittae may fall victims to medusae (Obelia) far smaller, as Steuer (1910, p. 631) describes. The siphonophores, too, of which our waters support one species in abundance (p. 377), destroy countless copepods, etc. The common boreal euphausiids, important in the faunal community of the Gulf of Maine, may typify the planktonic animals that feed chiefly on pelagic vegetables, but which also consume animal food in less amount. Thus Lebour (1922) found bits of green weed, diatoms, and fragments of mollusks in Nyctiphanes couchii. Paulsen (1909, p. 48) records Thysanoessa inermis from Icelandic waters stuffed with the diatoms Asterionella, Chaetoceras, and Coscinodiscus, and describes Megany- ctiphanes as full of these same diatoms, with tintinnids (Cyttarocylis), peridinians (Dinophysis, Ceratium, and Peridinium), and Globigerina in addition; but his dis- covery of crustacean debris (Calanus antennae recognizable among it) in the stomachs of both these species of pelagic shrimps proved that they had also eaten smaller Crustacea — some of the specimens examined had, indeed, partaken of a purely animal diet. Holt and Tattersall (1905, p. 103) likewise found some examples of Meganyctiphanes with the leg basket more or less stuffed with prey, including copepods, schizopods, and decapod larvae, Limacina and other animal debris, and one with the tail of a young fish actually in its mouth. Lebour (1924a) reports Meganyctiphanes feeding on Sagittae, Crustacea, and dead specimens of its own kind in the aquarium. We can substantiate these observations in part, having recognized algal filaments and diatom debris among the mass of finely comminuted particles (themselves, to judge from their brownish green color, probably vegetable in nature) with which the alimentary tracts of numerous specimens of Meganycti- phanes from various parts of the gulf are packed, and we have often found specimens of this shrimp carrying loads of small crustaceans. For example, one taken off Cape PLANKTON OP THE GULF OP MAINE 109 Cod on December 29, 1920 (station 10491), had a dozen or more Metridia and as many Pseudocalanus. live or six large Calanus, the siphon and part of the stem of a Ste- phanomia, besides a considerable mass of diatoms (Rhizosolenia) and some unrecog- nizable animal debris clasped between its thoracic legs. Several others taken at random from a large catch of these shrimps, made in the northeastern corner of the gulf on June 10, 1915 (station 10283), carried packs consisting chiefly of Calanus, occasionally a Eucha?ta, and Pseudocalanus, matted together with unrecognizable vegetable debris. One had a starfish larva and two eggs, probably of its own species, with the young nauplius almost ready to hatch out. Lest the reader think this omnivorous diet is at all seasonal, I may add that most of the Meganyctiphanes taken in the eastern basin on August 7 of that year carried loads of Calanus, Metridia, and Temora, with the cladoceran genus Evadne in great numbers, besides algal filaments and debris, the origin of which I could not determine. At Eastport, too, I have seen Meganyctiphanes clasping bits of herring refuse from the sardine factories. Up to very recently the method by which euphausiids gather their food had not been actually observed in life, but since the preceding lines were written, Lebour (1924a, p. 405) has described the food as "brought to the thoracic limbs by a current from behind, set up by the movement of the abdominal limbs, the thoracic limbs forming a sort of basket-hke receptacle for the accumulated food." Thus with the bristly armature of their legs they sweep the water for their prey just as barnacles do, gathering whatever copepods, Cladocera, diatoms, peridinians, or indeed small animals or plants of any sort, come within their reach as they dart to and fro in the water. The nourishment of the marine copepods remained a riddle until Dakin (1908) found that the alimentary canals of hundreds of Calanus, Pseudocalanus, Centro- pages, and other genera of copepods from the North Sea contained chiefly diatoms. He counted up to 200 diatom shells in the stomach of a single copepod, with peridin- ians and a green substance (previously noted by other students), apparently the remains of shell-less unicellular plants. Esterly (1916) has similarly described the contents of the guts of several hundred copepods (mostly Calanus) from San Diego, Calif., as consisting chiefly of Coscinodiscus and other diatoms, silicoflagellates, Dinophysis, Peridinium and other peridinians, and of coccolithophorids. Lebour (1922) also found diatoms of various species, Phasocystis, coccoliths, and peridinians in Calanus; diatoms and green remains in Pseudocalanus; diatoms and flagellates in Temora; and Pha?ocystis in Anomalocera. Murphy (1923, p. 450) writes that the copepod Oiihona nana ate kelp and diatoms in the aquarium, and we have recognized remnants of Thalassiosira in sundry specimens of Calanus, and Thalassiosira, Chsetoceros, and Biddulphia in Metridia from Massachusetts Bay at the time of the vernal diatom flowering. Diatom frag- ments have also been detected repeatedly in the excreta of copepods, which are familiar objects in the catches of tow nets, but Esterly's (1916) discovery of an oc- casional nauplius and copepod fragment in copepod stomachs proved that they are not exclusively vegetarian. Lebour (1922) has more recently found that the large blue copepod Anomalocera may feed largely on micro- Crustacea, while 110 BULLETIN OF THE BUREAU OF FISHERIES smaller copepods form a considerable item in the diet of Temora. Calanus, however, she found chiefly vegetarian, and Pseudocalanus perhaps exclusively so. Marshall's (1924) more recent study of the gut contents of large numbers of Calanus taken throughout the year in the English Channel corroborates this, diatoms proving the chief article of diet in spring and autumn with peridinians (curiously enough, however, no Ceratium) in summer. Silicoflagellates were also eaten in small quantities, while a few of the Calanus had eaten other copepods, molluscan larvse, and tintinnids. All the Tomopteris I have examined have been empty, which has been the experience of most students, but it is probable that they are vegetable feeders chiefly, Lebour (1922 and 1923) having found diatoms their principal diet, with some green flagellates. Tomopteris, however, sometimes turns carnivorous, for she watched one swallow a Sagitta whole and saw another that contained a larval herring. All the shell-bearing pteropods (Limacina retroversa, for example) are also vegetarian, dieting chiefly on diatoms. The Salpse likewise feed on diatoms, peridinians, and other small organisms, animal as well as plant, their gut contents and foecal masses having long been a treasure house to the student of the microscopic plankton. For example, the "guts" of large S. tilesii collected south of Nantucket Lightship in July, 1913 (station 10061), contained a varied assortment of diatoms, Peridinium, and Ceratium, besides an occasional newly-hatched Euthemisto; but the most successful captors of the unicellular pelagic plants are the appendicularians, which, thanks to their very fine-meshed straining apparatus, are able to utilize gymnodinids, rhizopods, naked flagellates, coccolithophids, 52 etc., forms so tiny that for the most part they pass through the finest tow nets. Appendicularians likewise devour the larger protozoans and unicellular plants. For example, a large Oikopleura vanhqff'eni from the neighborhood of Lurcher Shoal (May 10, 1915, station 10272) was packed with the horns and other fragments of Ceratium, besides small Peridinium of several species, tintinnids, and silicoflagellates (Distephanus). None of the pelagic tunicates are plentiful enough in the Gulf of Maine to make serious inroads on the phytoplankton. In the Gulf Stream to the south Salpte sometimes occur in hordes, and on such occasions strain the water bare (Bigelow, 1909). Among the unicellular planktonic animals the infusorians are proverbially rapa- cious. The tintinnid genus Cyttarocylis has been found to contain a great variety of microsocopic organisms — e. g., Peridinium, Dinophysis, Goniaulax, and diatoms (Lebour, 1922) — and even the Infusoria, which are provided with chromatophores, are known to take solid food (Steuer, 1910, p. 627). Radiolarians engidf diatoms, tintinnids, and other Infusoria; hence, when Acanthometron swarms in the gulf (p. 460) it must locally take heavy toll of other microscopic animals and of planktonic plants. Foraminifera are also rapacious animals, but have never been found plentiful enough in the plankton of the Gulf of Maine to be of any great importance in the economy of its planktonic communities. On the border line between plant and animal, so far as their mode of nourishment is concerned, stand the peridinians, for while the shelled forms are typical producers "For an account of the food of appendicularians see Lohmann (1903, p. 23, pi. 4) and Johnstone (1908, p. 139). PLANKTON OF THE GULF OF MAINE 111 the naked peridinians have repeatedly been found to contain other peridinians, Phssocystis, and occasionally a diatom. 53 It is a question of moment in the economy of the sea, and of practical bearing on the fisheries problems of the gulf, to what extent the sundry carnivorous mem- bers of its plankton menace the survival of the stocks of larval fishes that are produced there. The preceding pages contain sundry instances of planktonic animals eating young fish, which could be multiplied many-fold from published reports, were this worth while. In the Gidf of Maine it is probable that the most deadly enemies of newly-hatched fishes are the medusas, ctenophores, and Sagittas. The rapacity of Mertensia and Pleurobrachia in this respect has been mentioned; when and where the latter are abundant (as is so often the case on German Bank) it is hard to see how any larval fishes can escape their constant fishing. Pleurobrachia is also known to devour buoyant fish eggs of various species. In view of its local abundance, this ctenophore must be a serious enemy to the propagation of cod and haddock over the banks to the south and west of Cape Sable. Lebour (1925) has also reported Bolin- opsis, another ctenophore plentiful in the gulf (p. 372), as devouring larval goosefish (Lophius) in the aquarium; no doubt it accepts a fish diet equally in nature. The two medusas which are most abundant in the open waters of the gulf — ■ Aurelia and Phialidum — are also proven fish eaters, as are others plentiful in the coastal zone, 51 and the swarms of both of these which we have frequently encountered (pp. 350, 362) must take heavy toll of the little fishes that cross their paths. With Sagitta elegans so plentiful and so widespread in the gulf, it, too, must de- stroy great numbers of young fish; must, then, be as serious a menace to the stock of herring, etc., in the Gulf of Maine as Lebour (1923) has found it in the English Channel. It may, perhaps, be named the most effective check among all the plank- tonic category to the local propagation of such fishes as pass through a prolonged planktonic stage, and this incudes most of the important food-species of the gulf. I have found no published record and have seen no actual instance of the amphipod genus Euthemisto eating fish; but in view of its known rapacity it is likely to do so when occasion offers. Decapod larva? certainly do (p. 107), and these are abundant locally near shore at certain seasons. Euphausiids also eat fish to some extent, though probably it is a minor article in their dietary (p. 108). It is fortunate, indeed, that the copepod species which so usually dominates the plankton of the gulf (Calanus finmarcliicus) is not a fish eater (at least, it is not known to eat fish). Were the blue copepod Auomalocera as plentiful as Calanus, hardly a young fish could survive. As it is, few can "run the gauntlet" of the medusas, ctenophores, Sagittas, and crustaceans that prey upon them; and so many species (and these plentiful in the gulf) of these groups are now known to prey on fish larvae that they are almost certainly the most effective check on the survival of the countless myriads of young fish that are yearly produced in the gulf. There is good reason, then, to believe that the fluctuations known to occur from year to year <* Lebour (1922) has recently given a considerable diet list for Amphidinium and Gymnodiniun). « Lebour (1923, 1924) found Aurelia, Phialidium, Aequorea, Obelia, Laodicea, Rathkea, and Bougainvillea feeding on young fish; likewise several other medusa; and Pleurobrachia. 75898—26 S 112 BULLETIN OF THE BUREAU OF FISHERIES in the stocks of herring, mackerel, haddock, etc., which are reared in the gulf, depend more on the abundance of the rapacious members of the planktonic com- munity (and especially on the abundance of Sagittse, medusae, Pleurobrachia, and Euthemisto) than on any other one factor. If plankton studies need any defense from the standpoint of the fisheries we need look no further. THE MORE IMPORTANT GROUPS OF PLANKTONIC ANIMALS MOLLUSKS In coastal and estuarine waters generally the larval stages of mollusks are abundant in the plankton, but in the open gulf they hardly figure in the catches, leaving the pteropods as the only molluscan group that is a regular factor in the planktonic community. The cephalopods are also considered briefly because of their importance in the natural economy of the sea, although so large and such active swimmers that they are not properly "plankton." Cephalopods Only two of the considerable list of cephalopods recorded at one time or another from the coasts of New England (for a complete list see Johnson, 1915) play a role of any importance in the pelagic life of the Gulf of Maine, but these two — Loligo pealii Lesueur and TUex iUecebrosa (Lesueur) — are extremely abundant locally in their proper season, when they form one of the principal sources of bait for fisher- men. While, on the one hand, their young provide an important element in the diet of various larger fishes, the adult squids devour innumerable fish fry. So active are these cephalopods and so easily do they avoid small or slow- moving gear that we have never taken a single specimen in our tow nets. Indeed, I can, from my own experience, verify Verrill's (1882, p. 306) statement that it is hard to capture them with a dip net, even when confined in a fish pond or weir. Hence I can offer the reader only a brief summary of accounts published pre- viously, with such notes as have been gleaned from personal observation on the beaches, and from accounts given me by fishermen and other observers. Loligo is the common squid south of Cape Cod, Illex north of Cape Ann, with the ranges of the two overlapping in Massachusetts Bay. Illex also occurs, if less commonly, as far south and west as the Woods Hole region (Sumner, Osburn, and Cole, 1913a). Loligo, on the other hand, has long been known occasionally as far north as Penobscot Bay, and Dr. A. G. Huntsman and Dr. A. H. Leim write me that it has recently been found to be quite common in summer in various estuaries of the Bay of Fundy; for instance, Passamaquoddy Bay, Scotsman Bay, and Cobe- quid Bay. Since more is known of the life history of Loligo than of Illex, it may be con- sidered first. Loligo is common in the Woods Hole region from April or May until November but disappears during the winter. During the 10-year period, 1900 to 1909, the earliest captures ranged from April 16 to May 7 (Sumner, Osburn, and Cole, 1913a), which probably applies to Massachusetts Bay, though, taking one year with another, this squid appears there later in spring and disappears earlier in autumn than it does along the southern coast of New England. During the late PLANKTON OF THE GULF OF MAINE 113 spring, summer, and early autumn Loligo is extremely common both south and north of Cape Cod, passing part of the time on or near the bottom, but often seen swimming in shoals near the surface, and it is taken in great numbers in fish traps and weirs and even in eelpots. Many specimens have likewise been dredged. Along the shores of southern New England it breeds from May until September, or later. I am informed by W. F. Clapp that he has frequently found its eggs in Duxbury and Plymouth Bays from June until October, and in the Bay of Fundy its eggs and larva? are reported by Doctor Leim in August and September. Since Verrill (1882) notes the capture of considerable numbers in breeding condition near Cape Ann as early as May in 1878, it is safe to credit it with a breeding season enduring throughout the warmer half of the year over the major part of its range. The eggs, which adhere together in bunches of hundreds of gelatinous capsules, attached to some fixed object, are laid chiefly (perhaps not exclusively) in depths varying from just below tide mark down to 50 meters or so and have been trawled in large numbers on every sort of bottom south of Cape Cod (Verrill, 1882; Sumner, Osburn, and Cole, 1913a). It has been estimated that individuals of the European representa- tives of this genus may lay as many as 40,000 eggs. According to Verrill, hatching takes place from June until October south of Cape Cod; probably during these same months along the shores of Massachusetts Bay, according to Mr. Clapp 's observations. We owe to Verrill (1882) an extensive series of measurements of the young squids at various seasons, and though he found it difficult to follow their rate of growth, owing to the protracted period over which spawning endures, his general conclusion was that June-hatched squids attain a mantle length of 60 to 85 millimeters by November; that the smallest have grown to about 150 to 180 millimeters when they reappear the next May; that the later- hatched summer broods are about 60 to SO millimeters long in the following spring; and that the largest adult breeding squids are probably from 2 to 4 years old. The young squids, from less than 6 up to 25 or more millimeters in length, often swim near the surface, where they have been taken in immense quantities with the tow net. Mr. Leim informs me that he towed young Loligo 2 to 4 millimeters long in Cobe- quid Bay, Bay of Fund} T , in September, 1921. Nevertheless, although young Loligo must be produced in myriads on their main breeding grounds, the larval stages are so closely confined to the coastal or inclosed waters of their nativity during their first summer that we have never taken them even in Massachusetts Bay (though they spawn abundantly in its tributaries) or anywhere in the open Gulf. It is not known whether this squid moves offshore as the water chills in autumn or whether it passes the cold season inshore on the bottom. There is, however, some slight presumption in favor of the latter alternative, for it seems to be strictly a coastal form, which, so far as I can learn, has never been reported from the offshore banks in summer or from deep water. North of Cape Ann Loligo is always far outnumbered, and, except for the small Bay of Fundy colony, is practically replaced east of Penobscot Bay by Illex illece- brosa, 55 a squid much resembling it in appearance but easily distinguished (indeed it » This squid has often been referred to the genus Ommastrcphes. Recent students of the cephalopods, however, unite in referring it to Illex, a genus founded by Steenstrup for the reception of its European relative, /. coindeti. For a recent discussion of Illex see Pfefler (1908 and 1912). 114 BULLETIN OF THE BUREAU OF FISHERIES belongs to a different family) by its perforated eyelid as well as by its shorter fins. It has long been known that this beautiful animal is very abundant from Massa- chusetts Bay northward to the shores of Newfoundland and Labrador, and my own observations lead me to believe that its numbers increase from southwest to north- east around the coasts of the Gulf of Maine. However, though its economic value has been fully appreciated by fishermen for over a century, and while it has often been referred to in scientific literature, practically nothing is known of its life history. Illex appears along the shores of the gulf in late spring or early summer (I have been unable to find any record of the exact date of its vernal arrival) , is found very plentifully there throughout the summer and early autumn, and vanishes from the coast some time in October or November. According to reports by fishermen it is present offshore in winter, though not to be found in the coastal zone at that season, a phenomenon to which I shall have occasion to recur. During its season Illex occurs even more abundantly than does Loligo farther south, the vast schools in which it visits the coast having been described long ago by Verrill. Owing to a habit of stranding, the presence of this squid is very evident, as it oftens comes ashore in large numbers on the beaches from Cape Cod to the Bay of Fundy. On the islands near the mouth of the latter, in particular, I have found them, as did Verrill, in windrows on the flats in August and September, stranded squids being a familiar sight there to everyone. At low tide shoals of squid may often be seen darting to and fro over the sand or struggling in the shallows. For some inscrutable reason the squid, once aground, seems forced by instinct to drive farther and farther ashore — throw it out ever so often into deeper water, and it shoots, arrowlike, back on the beach, to perish there as the tide ebbs. This fatal habit causes the destruction of multitudes of squid, as long ago recounted by Verrill and by Smith and Harger (in Verrill, 1882, p. 307), who tell us that when in pursuit of young mackerel many of the "squids became stranded and perished by hundreds, for when they once touch the shore they begin to pump water from their siphons with great energy, and this usually forces them farther and farther up the beach." " It is probable, from various observations," says Verrill (1882, p. 307), "that this and other species of squids are mainly nocturnal in their habits, or at least are much more active in the night than in the day." Certainly it is at night that they most often enter the weirs and pounds. During the dark hours in summer and autumn the presence of shoals of squid is often disclosed by their phosphorescent wakes, Hjort (1912, p. 649) describing the common Norwegian squid, of the genus Ommastrephes, as "moving in the surface waters like luminous bubbles, resembling large milky white electric lamps being constantly lit and extinguished." The Gulf of Maine Illex, however, is often seen swimming near the surface during the daytime as well. Whenever and wherever found, these squids are extremely voracious, and the schools that run ashore often do so in pursuit of fish fry. At the mouth of the Bay of Fundy, both in summer and in early autumn, I have seen them eagerly following the schools of young herring, which in their turn are feeding upon shrimps (euphausiids) , often so common in the surface waters there (p. 135). I can corroborate Verrill's observation that squid stomachs are then often distended, both with shrimp and PLANKTON OF THE GULF OF MAINE 115 with fragments of herring, having found this to be the case in dozens of specimens. Young mackerel, too, suffer from their attacks, and we owe to Smith and Harger (quoted by Yerrill, 1S82, p. 306) a graphic account of their pursuit of the latter among the wharves of Provincetown Harbor during the month of July. Particularly inter- esting is their activity at such times, the ferocity of the attack, and the deadly nature of the single bite. The cannibalistic habits of Illex have likewise been commented upon, its own young being a common article of diet. This squid, like so many of the pelagic fishes, is very erratic in its appearance, being here to-day in hordes and gone to-morrow, perhaps to reappear in a few days. Illex provides a valuable source of bait for the offshore fishermen. It has been estimated that at one time squid formed fully half the bait supply of the vessels resorting to the Grand Banks (Goode, 1884), and we have record of 30,000 to 40,000 taken in one Newfoundland harbor in a single day. Probably Illex never occurs in the Gulf of Maine (which is the southern outpost of its regular range) in such abundance as this, but as long ago as 1897 the squid fishery of Massachusetts Bay alone (no doubt this and the preceding species combined) yielded over a thousand barrels of bait, and in 1902 the catch of squid in Massachusetts was upward of 5,000,000 pounds. At one time or another large numbers are taken by various methods all along the coasts of the Gulf as well as on the offshore banks. So voraci- ous and active an animal, and one at the same time so numerous, must take a heavy toll of the young fish, not to mention the various planktonic animals. Illex is probably to be classed as an oceanic animal, for it occurs commonly on the Grand Banks far from land and is often plentiful on Georges Bank as well. Probably its vernal appearance and continued presence off the coasts of the gulf of Maine throughout the summer are to be explained as a feeding migration (certainly this has nothing to do with its spawning), while its disappearance from the coast in autumn is part of a general offshore movement. Mr. Clapp's capture of several large specimens on Georges Bank (taken in otter trawl) during the last week of Novem- ber in 1911 harmonizes with this suggestion. The fact that a whale (species unknown) that stranded on the south shore of Cape Cod on January 29, 1869, contained in its stomach thousands of Illex beaks 56 belonging to squids of about 12 to 15 inches body length throws no light on this point, for it may have eaten them many miles away from where it came ashore. We have no other winter records for Illex from the Gulf of Maine. Nothing is known of the breeding habits of this squid; its eggs have never been found, nor have its newly hatched young been recorded." However, it is safe to say that it does not spawn along the coast of the Gulf of Maine at any season, for all the adult squids examined by Verrill and all that I have seen have been sexually inactive. Neither did McMurrich find its young at any season in his tows at St. Andrews. Indeed, the smallest Gulf of Maine specimens of which we can learn are one of about 10 centimeters, reported by Capt. H. E. Calder near Campobello, at « Some hundreds of these are preserved in the collection of the Museum of Comparative Zoology. Their identity has been established by Mr. Clapp by comparison with the beak dissected from an Illex from Oeorges Bank, which measured about 14 inches in length from the edge of the mantle to tip of tail. •' One with a mantle measuring only 33 millimeters in length is recorded by PfefFer (1912). 116 BULLETIN OF THE BUREAU OF FISHERIES the mouth of the Bay of Fundy (date unknown), and others of 16 to 19 centimeters, taken off Shelburne, Nova Scotia, in July, 1921. 58 Very likely its eggs are pelagic, as are those of some of its relatives, but it is certain that they do not occur regularly among the plankton of the Gulf of Maine, pelagic squid eggs (at least such as I have seen in the West Indies) being very easily recognized at all but the very earliest stages by the characteristic embryo. In European waters Illex illecebrosa is replaced by the form /. coindeti, so closely allied that Pfeffer (1912) regards the difference between them as no more than subspecific. /. coindeti ranges from Scottish waters to the Mediterranean. No squids other than Loligo and Illex have ever been found in any numbers in the Gulf of Maine, nor is it likely that any other species are ever numerically important in its pelagic fauna, with the possible exception of the boreal-arctic Gonatus fabricii. There is only one actual record of this species from the Gulf, a single specimen taken from the stomach of a cod near Seal Island, off Cape Sable (Johnson, 1915) ; but since its larvae have been taken at several localities between Newfoundland and Ireland, once, even, close to the southern edge of the Grand Banks (Hjort, 1912), the adult (which resembles Illex so closely that it might well be overlooked among the shoals of the latter) may be more common along the coasts of Nova Scotia and even in the Gulf of Maine than the paucity of actual records suggests. Finally, we may note that no "giant squids" seem ever to have been found in the Gulf of Maine. Pteropods Limacina retroversa Fleming 69 This shelled pteropod, a boreal form known from latitude about 50° to northern Norway, off the European coast, and from latitude about 34° to the southern part of Davis Strait, in the western Atlantic, is one of the most characteristic of the permanent pelagic inhabitants of the Gulf of Maine, where its numbers depend on local reproduction and not on immigration from elsewhere. It is the only pteropod of which this can confidently be asserted. Although it has now been taken in all parts of the gulf at one season or another, it is, as I have previously pointed out (p. 45 ; Bigelow, 1917, p. 299), far less regular in its occurrence in the gulf than certain of the calanoid copepods, the amphipod genus Euthemisto, or Sagitta elegans. It has commonly been our experience to find it comparatively plentiful at one station but rare or absent at another hard by. Similarly, waters where the nets yield an abundance of Limacina on one visit may prove quite barren of it a few weeks later, as was the case in the spring of 1920 on the eastern part of Georges Bank, where large Limacina were plentiful on March 11 (station 20065), but were sought in vain on April 17 (station 20111). Limacina was present on one cruise and absent on the next, or vice versa, at several localities during the season of 1915, notably off Mon- hegan and Matinicus Islands and in the northeast corner of the basin of the gulf. «• Information supplied by Doctor Huntsman. »' I follow Meisenheimer (1905) in uniting under this name the L. retTOVtrsa and L. balea of the early malacologists. Bonnevie (1912), it is true, has separated the two once more, basing the distinction partly on the shape of the shell (in which character, however, her specimens intcrgraded) and partly on the structure of the radula; but W. F. Clapp writes that "a careful exami- nation of the quantities of Limacina from fche Gulf of Maine has shown that it is impossible to consider the material as belonging to more than one species." PLAKKTON OF THE GULF OF MAINE 117 As appears from the accompanying charts (figs. 43 and 44), this pteropod has been taken over all the offshore waters of the gulf, on Georges Bank, and over the continental shelf off Nantucket. During our summer cruises (the season for which 71* 70' 6 9* 68' «7* 66* 1 r 1 1 «W) ■• J ' ) S y j/\ / •***- S Gi AND / /S-*r r-1/ HAt^AN / A^y ■X"- 1 o Q f^V • -' \ .. '(/ A peAb^oVAJ ^ (':v ° °0 V .' / •<> •i V + / k ° /i + + 1° 8° i + 4- 43 O ... ...^ "j *. \ '"'■- ...*- Vi C - o» --„ / -V.' 4 4 4 * ' . ' ' "' • • • • -0' • • 40 * 4 a + 4. • 4 4 40 71* 70" 63- 68* 67* 66* Fio. 43.— Occurrence of the pteropod Limadna retroversa from July to September, 1912 to 1922. • , occurred; 0, swarmed; Oi not taken our records are most extensive) it has appeared at rather more than half of all the stations, but the regularity of its distribution differs from summer to summer. For example, it was practically universal over the deeper parts of the gulf in August, 118 BULLETIN OF THE BUREAU OF FISHERIES 1913 (Bigelow, 1915, p. 302), whereas in July and August, 1912, we found it only in the northwest part of the gulf, on the one hand, and over German Bank, on the other (Bigelow, 1914, p. 120). At the same season in 1914 we found no Limacina ;,^l>" ° x O ff j. (Yarmouth S Fig. 44.— Occurrence of the shelled pteropod Limacina retroversa in winter and spring. X, locality records for December, 1920-January, 1921; • February to April, 1920; O. May of 1915 and 1920. The hatched curve incloses its area of occurrence in early spring up to May off Penobscot Bay, where it had been plentiful during the two summers preceding, but towed numbers of them in the northeastern corner of the gulf (stations 10246 and 10247) not far distant, and likewise in the Eastern Channel, over the northwest PLANKTON OF THE GULF OF MAINE 119 part of Georges Bank, and off Cape Cod (Bigelow, 1917, pp. 298 and 299). We have not taken Limaeina on Browns Bank either in spring or in summer, but since it has appeared at several of our stations over the shelf farther east, as well as on German Bank, in June, July, and August, and in the eastern basin of the gulf in March and April, it is more likely that our failure to find it on Browns Bank was accidental than that this pteropod does not occur there. Our most productive summer catches of Limaeina retroversa have been as follows: On July 29, 1912, we encountered a swarm of juveniles off Casco Bay (station 10019); in 1913 great numbers were taken off Nantucket on June 21 (by Capt. John McFar- land, lat. 40° 45' N., long. 70° W.); off Penobscot Bay, August 11 (station 10091); and near Cape Elizabeth, August 15 (station 10104); while the largest haul of all, yielding about 125 cubic centimeters of Limaeina (besides other plankton), was made over the northeast edge of Georges Bank on July 20, 1914 (station 10215). Thus, the few rich stations just mentioned (fig. 43) show no definite grouping in any one part of the gidf, but are spread far and wide. We did not find Limaeina in numbers at any time during the spring, summer, or autumn of 1915, though it was taken at about 50 per cent of our stations for that year; nor was it more plentiful in the gulf at our few stations for July and August of 1916, though odd specimens were detected at about half of them. In spite of the erratic way in which Limaeina appears and disappears (or at least vanishes from observation) in the Gulf of Maine, the records for the five years 1912 to 1916 show that in summer this pteropod is much less plentiful in the coastal zone and out to the 100-meter contour, from Massachusetts Bay northward and eastward as far as Mount Desert Island, than it is farther offshore. Limaeina has appeared in less than 10 per cent of the June-August stations in this inshore zone, to which w r e have paid particular attention, but seldom in any of the hauls at that season in the inner part of Massachusetts Bay or in any of the other indentations of the coast west of Mount Desert. Close proximity to the coast and shoalness of the water do not necessarily imply a scarcity of Limaeina in summer, however, for this, it seems, is its period of maximum abundance at St. Andrews, where Doctor McMurrich found it at almost every station from mid-June until September in 1916. Limaeina is likewise a regular summer inhabitant of the. coastal waters along the outer shores of Cape Cod and of the shallows over German and Georges Banks, and south of Nantucket. Furthermore, it may occasionally appear in great numbers in Massachusetts Bay in summer, when it is usually rare or absent there, for Alexander Agassiz (1866) found it swarming at Nahant (some 12 miles from Boston) during the summer of 1863. A considerable number of records of Limaeina for September, October, and November show that this pteropod, like Euthemisto, tends to work inshore in the western side of the gulf in autumn. Thus, in 1915 00 it occurred at four out of six late October and early November stations in Massachusetts Bay, whereas we have only once found it inside a line from Cape Cod to Cape Ann in July or August of recent years (station 10342, July 19, 1916). Similarly, no Limaeina were taken in the hauls along the Maine coast inside the 100-meter contour in 1915 until Sep- *> See Bigelow, 1917, p. 299, for records of Limaeina in 1914 and 1915. 120 BULLETIN OF THE BUREAU OP FISHERIES tember, though in other years it has appeared in numbers off Casco Bay in summer, as just noted (p. 119). Apparently it partially withdraws from the Bay of Fundy in autumn, for McMurrich found only occasional examples at St. Andrews from the first week of October until the new year. It is not yet possible to plot the distribution of Limacina over the gulf as a whole for winter, our December- January cruise having been confined to the northern and western parts; but there, at least, Limacina is as widespread during early winter as it is in summer; and if the season of 1920-1921 be representative, it is even more regularly distributed, for it occurred at 10 out of 14 tow-net stations, both in Massachusetts and Ipswich Bays near land, and from Cape Cod to Nova Scotia offshore (stations 104S8 to 10491, 10493, 10495, 10496, 10497, and 10500 to 10502). Similarly, Stimpson (1854) described it as present in Massachusetts Bay from February until April, more than half a century ago, though the fact that it appeared in the tow near Gloucester late in November, 1912, and again in Feb- ruary, 1913, but neither in December nor in January of that winter, shows that it is as subject to sporadic fluctuations in abundance there during the cold season as dur- ing the warm. Failure to find Limacina in the Fundy Deep on January 4, 1921, with McMur- rich's record of it as only occasional at St. Andrews during the half-year from Decem- ber to May, 61 suggests that it occurs less regularly and is much less plentiful in the Bay of Fundy in winter than in summer, which is just the reverse of its seasonal history in Massachusetts Bay. If the season of 1920 can be taken as representative, Limacina withdraws from the whole northern and eastern part of the gulf and likewise from the immediate coastal zone in the western side during the last few weeks of winter or first days of spring, for we did not take a single specimen anywhere in the guff during that March or April north or west of the undulating curve laid down on the accompany- ing chart (fig. 44) ; although Limacina in various stages in growth then occurred irregularly along Cape Cod, in the western, southern, and southeastern parts of the basin, and over and off the slope of Georges Bank. Our records point to the months of March and April as the season when the geographical range of Limacina in the Gulf of Maine is least extensive, and to the area just outlined as the only part of the gulf where this pteropod is regularly present the year round. With the advance of spring it once more spreads over the northern corner of the gulf, occurring at four stations in the eastern side of the basin in May, 1915; but while a considerable augmentation in its numbers takes place in the St. Andrews region (which probably mirrors conditions in the Bay of Fundy generally) by late June, as reflected by the frequency of captures listed by Doctor McMurrich, this does not happen in the coastal zone of the gulf west and south of Mount Desert until three months later, as just noted. In this connection it is interesting that Limacina is present all the year round off the west and south coasts of Ireland, just as it is in the offshore waters of the Gulf of Maine, but is seasonal along the Irish shores, with its maximum in spring •> From his plankton lists for 1915 and 1916. PLANKTON OF THE GULF OF MAINE 121 and summer (Massy, 1909), and that it is as erratic in its occurrence in the North Sea as it is in the Gulf of Maine. Limacina has been taken at about 50 per cent of our stations over the conti- nental slope between the longitudes of New York and Cape Sable in late winter, spring, summer, and early autumn, though never in great numbers. Only one specimen was taken at our most oceanic station (1021S, July, 1914), where the plank- ton as a whole was tropical, nor did we find it associated with the warm-water pteropods at our outermost stations south of New York in 1913. Being typically boreal in its affinity to temperature, it is not to be expected in the warm waters of the so-called Gulf Stream off the American littoral except as an accidental and probably short-lived straggler from the cooler coastal zone, but in more northern seas Limacina occurs chiefly in what is generally known to European oceanographers as the "Atlantic" water. This, for example, is the case south of Iceland, where it appears in great shoals, and it is with the general drift of this water (which is warm in contrast to the polar currents) that Limacina penetrates the Norwegian sea (Paulsen, 1910), for it is not at home in the icy cold Arctic water of comparatively low salinity. Most of the records of Limacina in the gulf have been from subsurface hauls, for which the precise depths can not be stated because made with open nets; but most of them have apparently come from comparatively shoal levels, for when two hauls have been made at different depths below the surface the shallower has usually taken the most Limacina. On the whole, the most prolific depth zone may be stated as from 20 to 25 meters down to about 80, which corroborates Paulsen's (1910) generalization that Limacina lives chiefly shoaler than 50 meters in north European seas, though it has occasionally been taken much deeper. In summer we have never detected Limacina on the surface during the hours of bright sunlight. In August, 1913, for example, "it was only once taken on the surface (station 10103), although a surface haul was made at every station, usually with a net of the same mesh as the one in which Limacina was taken in the depths" (Bigelow, 1915, p. 303), that one occasion being at 7 p. m. On several occasions during August, 1914, however, and the summer and autumn of 1915 (stations 10247, 10264, 10294, 10295, 10308, 10329, and 10333), surface tows between sunset and sunrise have yielded it in some numbers. This suggests that Limacina, like many other planktonic animals, performs a more or less regular diurnal migration in summer, rising toward the surface during the dark hours, to sink again at sunrise. The fact that the surface captures of Limacina (10 stations) 62 on our March and April cruises of 1920 were made invariably either in the dark or during the twilight hours between sunset and sunrise shows that this also takes place in spring, but perhaps not in autumn and early winter, when the sun is at its lowest. 63 This habit certainly is not so characteristic of Limacina in the more northern seas, where the sunlight is « Limacina retroversa was taken at the following stations during the spring of 1920: 20044, 20045, 20046, 20O4S, 20063, 20057, 20060, 20001, 20064, 20065, 20067, 20068, 20070, 20071, 20088, 20091, 20094, 20105, 20107, 20110, 20114, 20116, 20119, 20120, 20126, 20129; and at the following in the winter and early spring of 1920-21: 10488, 10490, 10491, 10493, 10495, 10496, 10497, 10501, 10502, 10505, 10509, 10510, 10511. For earlier Gulf of Maine records of this pteropod see Bigelow, 1914, 1915, 1917, and 1922. •i We lack direct information on this point, our surface hauls for that season having been made with small, flne-meshed nets, through which so little water filters that the apparent absence of Limacina may not be significant. 122 BULLETIN OF THE BUREAU OF FISHERIES weaker. In fact, it may not be followed at all there, for this pteropod is occasionally met with in great shoals on the surface off Iceland in daytime, though usually not when the sun is high. The presence of Limacina retroversa in the Gulf of Maine throughout the year, together with its very general distribution there, proves that its local presence or absence is not governed by small variations in temperature or salinity. On the contrary, Limacina (both large and small) has been taken at one season or another in water varying in temperature from 2° to about 16.6° — that is, over practically the entire range proper to the gulf except for the very coldest and the very warmest. Probably its habit of coming up to the surface at night brings it into the latter also, on occasion. But the great majority of the Gulf of Maine records for this pteropod have certainly been from temperatures lower than 15° at all sea- sons, and since it has never been found regularly or abundantly in water warmer than this in any part of the ocean, 15° may be set arbitrarily as the upper tem- perature limit to its continued presence and prosperous existence. Thus, in our latitudes it is probably the high temperature of the oceanic water that is the offshore barrier to it, confining it to the continental edge and shelf off the coast of the United States. On the other hand, although Limacina occurs in temperatures as low as 2 to 3° in the gulf in winter, it does not tend to congregate in the very coldest water at that season, but rather the reverse, for it was either absent altogether or at least very rare during the spring of 1920 (one or two only at stations 20055 to 20061) wherever the major part of the column of water was colder than 2°, although it was present in the neighboring parts of the gulf at the time. We have found it equally lacking or very rare in early spring in the icy cold water over the whole breadth of the shelf abreast of southern Nova Scotia, and certainly it is very scarce, if it occurs at all, in the coldest water along that coast in summer. Furthermore, Doctor McMurrich's notes show that there is a very close agreement between winter chilling and scarcity, vernal warming and regular presence of Limacina at St. Andrews, where it practically disappears when the temperature falls below about 3°, not to reappear regularly in the tows until the water warms to 8 or 9° the following spring. Although the evidence is not so clear, it seems that the presence or absence of Limacina may be correlated similarly with temperature in Massachusetts Bay, whence it appears to vanish when the water chills below, say, 2 to 3°, as happened in February and March of 1920; whereas in warmer winters, as that of 1912-1913, when the temperature of the water did not fall much below 3°, Limacina may occur sporadically and in small numbers right through from autumn until February (p. 120). These facts obviously suggest that it is the local cooling of the water that drives this pteropod from the coastal waters of the gulf, and from its northeastern corner generally, in late winter and early spring. Temperature may also determine the bathymetric occurrence of Limacina. For example, we found it comparatively abundant on the surface over the outer part of the shelf abreast of Cape Sable early in the summer of 1915 (station 10294, June 23), when the superficial water had warmed to 9° to 10°, but with temperatures as low as 2° to 3° only 40 meters down it was certainly scarce at deeper levels. In PLANKTON OF THE GULF OF MAINE 123 fact, it may not have occurred at all, for the few specimens brought in by the deep hauls may have been picked up by the nets close to the surface on their journey down or up; and the scarcity, if not absence, of this species in the coldest water along Nova Scotia is sufficient evidence that it is not an immigrant to the Gulf of Maine by that route. The general thesis that it is not at home in water of Arctic temperatures is further corroborated by Doctor Huntsman, who informs me that Limacina retro- versa is scarce, if not wanting, in the Gulf of St. Lawrence, where, by contrast, its larger Arctic relative (L. helicina) is very plentiful. I have pointed out elsewhere (Bigelow, 1917, p. 299) that L. retroversa occurs in numbers in waters of widely varying salinity in the Gulf of Maine, which agrees with experience in European seas; but in spite of its tolerance for variations in salinity it is clearly characteristic of the Salter rather than of the fresher waters of the gulf. Thus, it has been detected at only five stations out of 55, where the upper 10 meters or so have been fresher than 31.5 per mille; never in any numbers except where the underlying layers were much Salter (e. g., station 10294, surface 31.06, 80 meters, 32.79 per mille). While such evidence is perhaps not conclusive for an organism so sporadic in its local appearances and disappearances, at least it justifies the working hypothesis that L-. retroversa is seldom to be expected in water fresher than, say, 31.5 per mille, and not likely to persist in much lower salinities. About 31.06 per mille is the lowest salinity in which it has certainly been taken within the limits of the gulf, and Paulsen (1910) has already suggested the probability that when this pteropod chances to stray into water much fresher than 30 to 31 per mille it perishes. The dependence of L. retroversa on comparatively high salinity may have as much to do with making Massachusetts Bay and the coastal belt of the gulf generally unfavorable for it in spring as has its avoidance of very low temperatures. Until the seasonal cycle of these two sets of phenomena — biologic and hydro- graphic — has been followed more closely, the dependence of the former on the latter can only be stated in the most general terms. However, it is important for an understanding of the biology of this pteropod to emphasize the probability that there is a causal relationship between the seasonal expansions and contractions in its geographic range in the Gulf of Maine, on the one hand, and local and seasonal differences in the salinity of the water, on the other. We find in this a resasonable explanation for the fact that while winter chilling to 2° to 3° probably is the cause which banishes L. retroversa from the coldest parts of the gulf in winter, 04 it does not reappear near the coast in regions where the effect of the spring freshets in lowering the salinity persists longest into spring and summer (Massachusetts Bay, for example) until several months after the water has warmed to a point favorable for its existence, and until a considerable increase has taken place in the salinity of the upper 40 meters or so. In such locations, therefore, low salinity is probably responsible for its protracted absence, which continues until the water is once more salt enough for its liking. Kepopulation of the coastal zone by Limacina after its annual period of scarcity might take place in one of two ways — either by local survival or by immigration. •< From parts of the Bay of Fundy and from the inner parts of Massachutests Bay and probably from all along the shore in cold winters. 124 BULLETIN OF THE BUEEAU OF FISHEBIES Alexander Agassiz's (1866) observation that Limacina often sinks to the bottom suggested to him, and to other students subsequently, that this habit may explain its sudden appearances and disappearances — that is, that it may endure unfavorable periods on the bottom, where salinity would always be sufficiently high for its existence in all parts of the gulf except in very shallow water. However, since this habit has not been observed in European waters, where L. retroversa is often far more abundant than we have ever found it in the Gulf of Maine, probably its disappearance from the coast water reflects either the death of the local stock or a migration out to sea, its reappearance there reflecting an actual immigration from offshore in toward land, which follows more or less closely on the reestablishment of a favorable environment in the coast water and depends on the precise distribution of Limacina at the time relative to the circulation in the central parts of the gulf. The upper limit of salinity for Limacina is certainly as high as 36 per mille (35.9 per mille is the most saline water in which I find it actually recorded), and inasmuch as it thrives in water of 34 to 35 per mille in the North Sea region no part of the Gulf of Maine could ever be too salty to afford it a favorable environment. Nothing is known of the reproduction of L. retroversa in the Guff of Maine except that young as well as old individuals have been taken repeatedly in spring, summer, autumn, and winter, proving it endemic. Very little information is as yet available as to the actual numbers in which L. retroversa occurs in the gulf, and comparison of the catches of the horizontal nets with those of the verticals shows that whether it be scarce or plentiful, it is so prone to congregate in shoals (which one net may hit but the other miss) that it would take a great number of vertical hauls to yield even an approximation of its actual numerical strength over any considerable area of the sea. For example, the vertical haul from 70 meters yielded none at all at the station where we made our largest catch in the horizontal net (station 10215, northwest part of Georges Bank, 125 cubic centimeters of Limacina in a 50-meter haul of one-half hour's duration). An instance of the opposite sort is afforded by a station in the center of the gulf (March 2, 1920, station 20052), where the quantitative haul yielded enough (58 specimens) to indicate comparative abundance (theoretically 240 Limacina under each square meter of the sea's surface), whereas the surface haul jdelded only a few dozen individuals, the horizontal net, working at 100 meters, none at all, and the closing net only a few at 160 meters. Instances of this sort, which might be multi- plied, make any attempt to plot its actual numbers from the data yet in hand not only idle but apt to prove misleading. However, it can be stated as a general propo- sition that only on the rarest occasions does L. retroversa form any considerable pro- portion of the plankton in any part of the gulf, judged either by numbers of individ- uals or by built. 65 Nor have we ever found it in abundance to compare with the shoals recorded by Paulsen (1910) from the waters south and west of Iceland. There- fore, it is not likely that this pteropod is ever of as much importance as pasturage for the pelagic fishes in the Gulf of Maine as it is in Irish waters, for instance, where, says Massy (1909), it regularly serves as an important item in the diet of both mack- erel and herring. w The richest catches of Limacina are noted above (p. H9). PLANKTON OF THE GULF OF MAINE 125 Limacina lielicina Pliipps The Arctic pteropod L. lielicina, a close relative of the boreal L. retroversa, though characteristic of a different zoogeographic province, appears but rarely in the gulf, and then only as an immigrant from the colder waters to the east and north. Its status as such and its importance as an indicator of cold currents being discussed elsewhere (p. 59), this mention may be confined to a list of its recorded occurrence in the Gulf of Maine. 68 May 6, 1915 — off Cape Sable, station 10270, 150-0 meters and 50 meters. May 10, 1915 — near Lurcher Shoal, station 10272, 60-0 meters, occasional specimens on each occasion. Clione limacina (Pliipps) The large shell-less pteropod Clione, beautiful in the water and easily recog- nized, may be expected anywhere in the northern half of the Gulf of Maine in winter, spring, or summer (fig. 45). During the cold half of the year — December to May — it has appeared at nearly 50 per cent of our stations, both over the gulf as a whole and on the individual cruises. Not only are the records for these months very generally distributed over the deeper basins and along the coastal belt, but Clione may be more universal than the actual records suggest, for we have usually taken it in numbers so small that its failure to appear in the tow nettings at other stations may have been purely accidental. In summer, too, we have found Clione repeatedly in the northern parts of the gulf, but during the period from June to August it has appeared at only about 20 per cent of our stations — that is, distinctly less regularly than in winter or spring. We have not found it at all in September, October, or November, though the few stations for those months have been occupied at localities where it has been taken at other times of year. From this it appears that Clione is distinctly seasonal in its occurrence in the gulf, reaching its maximum from February until May and its minimum in autumn. Although Chone is oceanic in its general biologic status as opposed to neritic or coastwise, it shows no apparent predilection for the deeper rather than the shoaler parts of the Gulf of Maine; and while we have not found it in inclosed waters, and Doctor McMurrich detected it only once at St. Andrews (on February 16, 1916), it has been known to appear in swarms in Portland Harbor, an event referred to below (p. 127). Neither do our records suggest any seasonal onshore or offshore migrations on its part, such as appear to be executed by its relative, Limacina retroversa. I should point out that Clione is no more regular in its occurrence and shows no more concentration in the eastern than in the western side of the gulf, such as might be expected of an organism the maintenance of whose numbers within our limits depends partly on immigrations around Cape Sable, and such as actually ob- tains for various Arctic animals (p. 59). On the contrary, no general portion of the open gulf north of a line from Cape Cod to Cape Sable appears more favored by it than another at its season of maximum abundance, but our few traverses of Georges " Also off Halifax, Aug. 2, 1914; near Shelburne, Nova Scotia, and over the continental slope off that port, June 23 and 24, 1915 (Bigclow, 1917, p. 300). 126 BULLETIN OF THE BUREAU OF FISHERIES Bank suggest that Clione is less common there than within the gulf proper to the north. Thus, in March, 1920, it was not detected at all at the three stations (20065 to 20067) on the eastern end of Georges Bank, though on the slope to the south 71' 70" 69" 60* 67* 66* 14 4- 4- Portland C V /BAY c^ Vr •-'•- O A r-V* / N ° V A ( SCOTIA *■ ! I + A X $ (Yarmouth In CAPE V WSABLEJ 44' r * J P f! X • • XX + x x X x • ■* +,•■■■•■'""' jx ^* X >>*. 4* iZ 4 r / xo/i 4 X.. -j / • rw x X I \ \ o x x o o 4- 4- X " -x ' X Vj *\'x V 43* ? T] c7pe3°3 X o /" x- ' ■-^^ • \ j X*5 ■J x ',1' + + + 4- t / X + «r * "' -■.. ^..— ■■ *■' w + 4- + 4- 4- 4- 40* 71" 70* 69' 68' 67* 66* Fig. 45.— Occurrence of the naked pteropod Clione limacina. ©, locality records for June, July, and August; O. the winter months; X* March, April, and May (20068) a haul from 150-0 meters yielded four; and while it appeared again there (station 20109) and on the bank to the north (station 20110) on April 16, only one specimen was noted at each station. Apparently Clione vanishes from all parts of PLANKTON OF THE GULF OF MAINE 127 Georges Bank as the season progresses, for we did not find it at any station there or along the continental slope abreast the gulf in July of 1913, 1914, or 1916. We have never found Clione assuming any faunal prominence in the open waters of the Gulf of Maine, where it is usually represented by occasional specimens only among the mass of other plankton brought in by the nets. For example, in Febru- ary, March, and April, 1920, all our hauls combined yielded not over 175 specimens of Clione, although it occurred at some 30 stations, whereas various other animals were captured in thousands — even millions in the case of the commoner copepods. Wood (1S69, p. 185), it is true, found Clione so abundant in Portland harbor in May, 1S68, that "the water appeared to be alive with them," but our experience ever since 1912 has been so consistent in this respect that I can only look on such local swarms of Clione as altogether exceptional for the Gulf of Maine, although this pteropod regularly appears in vast shoals in more northern seas. It is still uncertain to what extent Clione is endemic in the Gulf of Maine. There is every reason to suppose that it immigrates more or less regularly into the gulf around Cape Sable via the Nova Scotian current, as do the various Arctic organisms, because it is far more numerous off the east coasts of Newfoundland and Labrador — where I found it swarming among the floe ice in the summer of 1900 — about the Grand Banks of Newfoundland, and in the Arctic seas as a whole, than we have ever found it of late years in the Gulf of Maine or farther south. However, as I have elsewhere emphasized, in reality the local presence of Clione is not the sure index to Arctic currents many have supposed (Bigelow, 1917, p. 301, and 1922, p. 174), for it is as abundant in Atlantic as in Arctic waters around Iceland (Damas and Koefoed, 1907; Paulsen, 1910); and while Clione grows to a larger size in the latter than in the former, there is no reason to doubt, from their evidence, that it breeds successfully in both. Many authors have quoted its abundance south of Ireland, to which Massy (1909) called attention, and where there is no reason to credit it with an Arctic origin. According to Dr. A. G. Huntsman (in Bigelow, 1922, p. 135), its larvae are found over the whole region from the Gulf of Maine to the Gulf of St. Lawrence and the Newfoundland Banks, at sea but not in estuaries. Like many other animals, Clione decreases in numbers toward the boundary (in this case the southern) of its range, but it is probably impossible to draw any sharp line beyond which it can not maintain itself. No doubt as we pass from north to south it becomes more and more dependent on accessions of fresh blood from the north for the maintenance of the local stock, but in favorable seasons it may be expected to reproduce itself in unwonted numbers far beyond its normal zone of abundance. Probably the Portland swarm just mentioned resulted from an unusu- ally successful wave of local reproduction; and the generality of its distribution over the gulf suggests that more or less Clione are produced there yearly, though probably immigration via the Nova Scotian current is the more important source of supply. On the whole, I see no reason to alter the view, earlier stated, that it probably rarely succeeds in breeding south of Cape Cod. Even in the Gulf of Maine Clione can reproduce itself in abundance only on the occasions when hydro- graphic conditions conspire in its favor, conditions occurring so rarely that only the one instance of this is known. I must caution the reader that very few 7589S— 26 9 128 BULLETIN OF THE BUREAU OF FISHERIES observations have been made on the occurrence of larval Clione that might or might not survive to maturity. Even in European seas, where the plankton has been much more intensively studied, little is known of the conditions of temperature and salinity under which its reproduction normally takes place (Paulsen, 1910). Granting that Clione does reproduce itself to some extent in the Gulf of Maine, it follows that its presence at any particular time and place is not necessarily to be taken as evidence of a northern current; but in the last analysis Clione is essentially of northern origin in the gulf, and it is probable that a considerable proportion of the stock existing there at any given time are actual immigrants via the Nova Scotian current, some indirect evidence of which is yielded by the details of the records of its occurrence in the gulf. Thus, although the data yet at hand do not indicate any connection between the winter increase in the numbers of Clione and the fluctuations of the cold current (the latter is then at a low ebb), and although Clione shows no definite tendency toward concentration in the side of the gulf where this water is most in evidence, the spring maximum for Clione corresponds to the maximum annual intrusion of the latter into the gulf. West and south of Cape Cod Clione may safely be classed as primarily an immi- grant. As such it was long ago recorded as far south as the coast of Virginia (Rath- bun, 1889), and probably it is a more or less regular if usually uncommon visitor along this part of the continental shelf in winter and spring, for the Albatross towed it off Delaware Bay on February 20, 1920 (station 20042), and Rathbun (1889) recorded it from localities on the outer part of the shelf between the latitudes of New York and Chesapeake Bay in April and May of 1887. Occasionally large numbers of them may drift south, De Kay (1843, p. 66) describing them as very abundant in the bays near New York in April, 1823, but only for a few days, after which they vanished. In warm summers, such as that of 1913, it vanishes beyond Cape Cod by July, but in the cool summer of 1916 its presence off Chesapeake Bay, off Delaware Bay, and off New York in August suggested temporary breeding activ- ity under rarely favorable local conditions, a view supported by the fact that at one of these stations (10386) Clione larvae were taken with the adults (Bigelow, 1922, pp. 156, 174). Evidently, however, Clione did not succeed in maintaining itself there much later into the season, because it was not taken in these southern waters at any of the November stations for that year. The high temperatures of the tropical "Gulf Stream" water are a fatal barrier to the offshore dispersal of Clione a few miles outside the continental edge, from abreast of southern Nova Scotia southward. Probably Clione is never numerous enough, or locally numerous, in the Gulf of Maine for a long enough period to be of any importance in its natural economy. In more northern seas its great swarms afford a bounteous food supply for whales, and it is an important article of diet for both mackerel and herring in Irish waters, according to Paulsen (1910). 67 •' Station records of Clione in the Gulf of Maine have been published as follows: For July and August, 1912, in Bigelow, 1914, p. 118; for the winter of 1912-1913 and the spring of 1913, in Bigelow, 1914, pp. 403, 406, and 407; for the summer of 1913, in Bigelow, 1915, p. 302. In July and August, 1914, it was detected at stations 10213, 10243, 10249, and 10255; in the season of 1915 at stations 10276, 10277, 10278, 10280, 10281, 10282, 10286, 10287, and 10306; in July, 1916, station 10346; in October and November, 1916, not at all; in the spring of 1920, stations 20046, 20048, 20049, 20053, 20055, 20056, 20057, 20058, 20068, 2O074, 20079, 20081, 20086, 20087, 20091, 20094, 20095, 20097, 20100, 20101, 20103, 20105, 20106, 20109, 20110, 20112, 20113, 20114, 20115, 20119, 20122, 20124, and 20126; in December, 1920, and January, 1921, stations 10489, 10491, 10493, 10495, 10496, and 10497. plankton of the gulf of maine 129 Other pelagic mollusks Apart from the cephalopoda and the three pteropods (Limacina retroversa, L. Jtelicina, and Clione limacina) just discussed, very few adult pelagic Mollusca have ever been found within the southern rim of the Gulf of Maine. 08 The Grampus cruises have yielded an Atlanta and two specimens of the pteropod Diacria trispinosa from 10 miles north-northwest of Gloucester on July 8, 1913, and two of Limacina injiata taken off Cape Cod July 19, 1914 (station 10213). All these species are char- acteristic of the warmer parts of the North Atlantic, not of boreal waters, and hence reached the gulf as stragglers from the warm waters of the Atlantic to the south; but it is hard to account for their presence at the particidar times and places of cap- ture, because "they were taken with an otherwise typical boreal assemblage of plankton organisms" (Bigelow, 1915, p. 306). A Pneumoderma, or some closely allied pteropod too young for identification, was taken near Lurcher Shoal on August 12, 1914 (station 10245); and, under the name Pseudoclione, Danforth (1907) has described a pteropod of doubtful relationship from Casco Bay, which showed sexual maturity combined with various larval charac- ters (taken August 29 and again September 5 to 8, 1902). A Cavolina tridentata and two Ptero trachea from the southern edge of Georges Bank, respectively on July 21 (station 10219) and July 20 (station 10216) in 1914, complete the brief list. In contrast to the Gulf of Maine, the waters along the continental slope from the longitude of New York eastward have proved extremely rich in warm-water pteropods and heteropods carried thither in the sweep of the Gulf Stream, whence considerable lists of them were obtained by the early expeditions of the Bureau of Fisheries (Smith and Hargar, 1874; Verrill, 1885; Johnson, 1915), as well as on our more recent Grampus cruises (Bigelow, 1917, p. 302). However, since it is only in the rarest instances that any of these find their way into the inner parts of the Gulf of Maine, little space need be devoted to them here. The captures of this category made by the Grampus in July, 1913, and July, 1914, are noted elsewhere (p. 54; Bigelow, 1915, p. 301; Bigelow, 1917, p. 302). These two lists together comprise some 14 species, while Johnson (1915), in his more complete summary of previous records, mentions 25, representing the genera Firoloida, Carinaria, Atlanta, Clio, Cuvierina, Peracle, Corolla, and Glaucus. Others (e. g., Janthina) have also been recorded, but only from examples washed up on the beaches of southern New England or the outlying islands. To illustrate how seldom any of these oceanic Mollusca stray within the 500-meter contour and how sharply their range contrasts with that of their boreal relative L. retroversa, the accompanying chart (fig. 46), showing all records listed by Johnson (1915), is offered. All these are from summer and autumn. In winter and spring warm water, with its character- istic tropical-oceanic inhabitants, lies farther out from the continental edge. 88 Leaving out of account the various pelagic bivalve and gastropod larvae. 130 BULLETIN OF THE BUREAU OF FISHERIES CRUSTACEANS Adult decapods The Gulf of Maine supports a host of decapods — that is, crabs, shrimps, and lobsters — the larval stages of which often swarm in the plankton, most often along 71" 70* 69* 68' 67* 68* <% • ' ' ) y\ >^g^And /sf ] rO/ HAINAN / y^y /BAyB-* XT /Ljs + / *■ + \ ' r j NOVA \ SCOTIA 44 *■ + +- 1 + 44' / «/" ; ..■;• (Yarmouth " l"*} \-v-*-\ "'"' \, y\n . Portland C j In CAPE y P o w Wsable/ ,-•' ^ ;.-\ J / y /'"%... ,.■•■" 41 ( & J *0 + „_, orlOI>»*«-f 4- 43' v v ' , \ o \ \ 42 + \_ IT \ + +,-- \ 4 42- \ J *^° ^ \ ; i t \ 4 4V 4 + + + • • 4- ,'" •*• +- 4V • '.o + + • • • • • • 4- •1- 40' 71' 70' 69' 68' 67' 66' Fig. 46.— Locality records for oceanic pteropods and heteropods. • , one species; ®, 10 or more species shore, as noted elsewhere (p. 34). The adults of nearly all of them live on the bottom, except when some of the shrimps make brief swimming excursions upward when disturbed, as, for instance, by the passage of the bottom net or trawl, or when PLANKTON OF THE GULF OF MAINE 131 they are lifted by active vertical currents. The glass shrimps (genus Pasiphrea) are the only decapods regularly planktonic in the Gulf of Maine when adult. Pasipheea These shrimps are so much larger (80 to 90 millimeters long when adult) than any other crustaceans pelagic in the gulf that even a single specimen is sure to be detected in the tow. It is therefore safe to assume that the list presented herewith comprises our whole catch, which is not true of smaller organisms easily overlooked in the mass of other plankton unless abundantly represented in the catch. We towed our first glass shrimps (three in number) in the western basin in a haul from 150 meters on August 9, 1913 (station 10088). Since then they have been taken there on August 22, 1914; August 31, 1915; March 5, 1920; and April 18, 1920 (stations 10254, 10307, 20087, and 20115), and likewise at two stations in the deep water in the northeastern part of the gulf (March 3, 1920, station 20055, and March 22, 1920, station 20081) ; once in the southeast corner (April 17, 1920, station 20112), and once at the outer edge of the shelf off Cape Sable (March 19, 1920, station 20076). So far as I can learn, the only previous records of this genus for the Gulf of Maine are as follows: Western Basin, approximate latitude 42° 38', longitude 69° 38', two specimens dredged in 203 meters in August, 1877; two more near the same locality, 256 and 311 meters (dredge), on August 27, 1878 (Smith, 1879); others from Cape Cod Bay and from off Cape Cod, 25 meters and 212 to 223 meters, respectively (Rathbun, 1905). These early captures were recorded as Pasiphxa tarda, which has long been spoken of as the characteristic northern representative of the genus (Wollebagk, 1908). Sund (1913), however, has more recently shown that at least three perfectly distinct and easily recognizable species have been confounded under this name, Smith's own illustration (S. I. Smith, 1879, pi. 10, fig. 1) showing that in reality the early American records were not based on tarda but on the P. multidentata of Esmark, which has also proved to be the commonest glass shrimp in Norwegian waters. 69 All the recent specimens from within the Gulf of Maine likewise are multidentata, a perfectly transparent species, whereas P. tarda is commonly blood red. Our records of P. multidentata have been from comparatively deep hauls, though not invariably from the deepest stratum in the Gulf (fig. 47) as follows: Station Depth of haul in meters Depth of water in meters Station Depth of haul in meters Depth of water in meters 10088 148-0 / 75-0 \ 225-0 230-0 180-140 274 } 286 245 230 20076 200-0 140-0 2O0-0 200-0 200-0 250 20081 206 2O087 10307 20112._ 290 20055 20115 290 So far as I can learn, Pasiphrea has never been taken on the surface or in plankton hauls shoaler than 75 meters in the Gulf of Maine, though it has been dredged in as shallow water as 25 meters; hence, it is clearly bathypelagic in the • 9 The several species are easily separable by the form of the rostrum, which is high and coniform in multidentata. I refer the reader to Sund (1913). For details 132 BULLETIN OF THE BUREAU OF FISHERIES gulf, just as in the Norwegian fjords (Wollebaek, 1908), and very probably it lives on the bottom part of the time. The material at hand is not sufficient to throw any light on the breeding habits of Pasiphaea in the Gulf, except that females carrying the very large eggs were taken TV 70' 69" 68' 67' 66' 1 1 1 1 ■ m ,, ■ r-i / bAWaN r l/rfV 44 Ajf^ a /v r \ 7 - I e u J NOVA ( SCOTIA +• 4 V Portland C j -~~' ! * i ( Yarmouth 1/1 CAPE ) W SABLE { 43 + / ( 4n + p ,-^7^/^-^ ■' ..... v — . ..•■"' «' 42 { cs"7 ^ (j f rwi c + C i> \ S \ + +/■-" »• .■■• N. v\ \ \ I 4- r ^7 ^s^-f n ■"" ' .•■■■"" / ^.:£ x i , :; 1 / \s f 41' + + + 4- 4- ,'' 4- 41' ,y' • ,s • •'." + 4. + 4- 4- 4- 40' 71* 70" 69" 68' 67* 66' Fio. 47.— Locality records tor the decapodous shrimp Pasiphaea. X. P- multidentata: O P- tarda: A. S. I. Smith's record. (See p. 131) in August (station 10254) but not in March or April. The locations of capture suggest the western basin (where we have usually, though not invariably, found it in our deepest hauls) as the chief local center of abundance for Pasiphaea, but it is PLANKTON OF THE GULF OF MAINE 133 to be expected anywhere in the gulf below 200 meters — witness the records from the eastern basin and from the southeast deep. We have only two records for P. tarda, both over the continental slope off Georges Bank in hauls from 750 to 100 meters, February 22 and March 12, 1920 (stations 20044 and 20069), which agrees with Sund's (1913) experience that this species usually hves at a rather deeper level than P. multidentata, from which it is separable by the low rostrum, hardly rising above the general dorsal outline, and by its red color. We have not taken P. principis, but this species is recorded from south of Marthas Vineyard by Sund (1913). Euphausiids We are indebted to Dr. H. J. Hansen, who identified the collections made dur- ing the summer of 1912 and winter of 1912 and 1913, and to Dr. W. M. Tattersall, who undertook the same task for the gatherings of 1914, 70 for abihty to include a chapter on this economically important and faunistically instructive group of pelagic crus- taceans. I have attempted the identifications of the euphausiids contained in the tow nettings of our subsequent cruises by comparison with specimens named by these two eminent speciahsts and by the aid of Zimmer's (1909) very clear keys and descriptions; but while it is easy to name the adults of all the species occurring regularly in the Gulf of Maine, by easily recognizable anatomical features, the larval stages, occasionally abundant (p. 134), still await reference to their proper parentage. Knowledge of the occurrence of this group in the deep water outside the conti- nental shelf abreast of the gulf, between the longitudes of 71 and 65°, is chiefly based on the collections made by the Bureau of Fisheries' vessels in past years, recently reported upon by Doctor Hansen (1915). Only a few species of euphausiids are yet known to occur within the gulf, nor is it likely that the various oceanic members of the group will ever be found in its inner parts except as stragglers; but these few (to be treated in detail below) are among the most characteristic if not the most numerous members of its endemic plankton. True, they seldom dominate the catch, or even form any considerable part of it, except locally in the northeast corner of the gulf and near the mouth of the Bay of Fundy, and when they swarm in other parts of the gulf it is only for brief periods. But our tow nets have seldom failed to yield them in greater or less number, except at times and localities when the catch as a whole has been of the scantiest. Euphausiid shrimps are so important in the dietary of whales and of many fishes that pursue them eagerly (and indeed one can well believe them dainty morsels) that they are much more important economically than their small numbers, contrasted with the hosts of copepods, might suggest. This subject is discussed in another chapter (p. 97). The occasions on which we have made notably rich hauls of euphausiids within the limits of the Gulf of Maine have been as follows: On Browns Bank, July 24, 1914 (station 10228), the haul at 60-0 meters yielded about 500 cubic centimeters of small Thysanoessa, representing three species ( Thysancessa gregaria, Th. longicaudata, 70 For tables of occurrence ot the several species in these years see Bigclow, 1914a, p. 411, and 1917, p. 282. 134 BULLETIN OF THE BUREAU OF FISHERIES and Th. inermis), many large Meganyctiphanes, and a few Nematoscelis. Four days later we again encountered a euphausiid plankton over the continental slope off Shelburne, Nova Scotia (station 10233), where half-hour hauls on the surface, at 100-0 meters and at 200-0 meters, yielded, respectively, 125, 500, and 250 cubic centimeters, chiefly euphausiids. On this occasion the surface catch consisted mainly of Euphausia, but Nematoscehs dominated at 400 meters, with the two species mingled at the 100-meter haul. An abundance of these two genera is perhaps characteristic of this general location in summer, for we again found them in large numbers over the continental slope nearby on June 24, 1915 (station 10295). This does not apply to Browns Bank, however, which was barren of euphausiids on June 24, 1915 (station 10296), though productive of them the previous July; nor did we find more than an odd specimen there in March or April, 1920 (stations 20072 and 20106). Small Th. longicaudata were numerous over the northeast part of Georges Bank on March 13 of that year (station 20070). By April 16 (station 20108) they had vanished thence, but the fact that we once more found small Th. longicaudata very plentiful off the southwest face of the bank on May 17 (station 20129) sug- gests that the swarm had drifted westward from one end of the bank to the other during the interval from March to May. Turning now to the inner parts of the gulf, we have twice found the waters off northern Cape Cod supporting larval and very young Thysanoessa in abundance (July 8, 1913, station 10057, and August 28, 1914, station 10264). Medium-sized and adult specimens of this genus (particularly Th. inernis, p. 135) were also taken in large numbers in the eastern side of the basin in May (station 10270) and off Cape Ann in August, 1915 (station 10306). On August 22, 1914 (station 10254), we found Meganyctiphanes abundant in the deeper water layers of the western basin, but the most interesting swarming of shrimps of this group in the western part of the gulf was the sudden appearance of shoals of Thysanoessa raschii off the Isles of Shoals late in April, 1913, as described below (p. 145). Provincetown Bay was similarly invaded by "shrimps," very likely of this same species, in March, 1880, as described by A. H. Clark (1887), and in August, 1923, euphausiids of some sort were so plentiful at the surface off Penobscot Bay that Dr. George C. Shattuck wrote me of seeing "a good many shrimp in the water" while sailing from Isle au Haut to Matinicus Island during the last week of the month. All the congregations of pelagic shrimps mentioned so far have been sporadic, or at least of brief duration; but euphausiids are often enough plentiful in the ex- treme northeast corner of the deep basin, some 50 miles southwest of Grand Manan, at various seasons, for this local abundance to be regarded as characteristic. Our first visit to this locality (in August, 1912) did not suggest this (indeed, not a single euphausiid was noted in the tow on that occasion), but many large specimens of Meganyctiphanes norvegica were taken at this general location on August 13, 1913 (station 10097), in a haul from about 160-0 meters; again on August 13, 1914 (sta- tion 10246, 150-0 meters); on May 10, 1915 (station 10273, 125-0 meters); on June 10, 1915 (station 10283, 100-0 meters) ; and in the basin, a few miles to the south- ward, on August 7, 1915 (station 10304). If the year 1920 can be taken as typical, this local abundance of Meganyctiphanes is as characteristic of spring as of midsum- PLANKTON OF THE GULF OF MAINE 135 mer, for this shrimp was plentifully represented in that region on March 22 (station 20081) in hauls from 40 and from 200 meters, while the haul from 100 meters yielded about 50 on April 12 (station 20100), although the zooplankton as a whole was decidedly scanty on that occasion. I hesitate to extend this generalization to the winter, however, because only a few euphausiids were taken there on January 5, 1921 (station 10502). Euphausiids n are often extremely plentiful near the surface in the Eastport-St. Andrews region at the mouth of the Bay of Fundy, where the smaller-sized herring can be seen chasing them to and fro right up to the docks (p. 102), and they are so conspicuous when schooling that they must have been seen and commented upon by local fishermen from the first settlement of that coast. The earliest published reference to their local abundance there, or in any part of the gulf, for that matter, seems to have been in 1S79, when S.I. Smith (1879, p. 90) described Meganyctiphanes norvegica as occurring at the surface in the Eastport region in "swarms, filling the water for miles," and as "usually accompanied by schools of mackerel, young pollock, and other fish, and in autumn by immense flocks of gulls, the fish and smaller gulls appear- ing to feed almost exclusively on Thysanopoda at such times." Such occasions he recorded for April, August, September, and October, adding that Verrill found these shrimp swarming in myriads in the ripplings in the center of the Bay of Fundy in 1869, and that they are often so abundant among the wharves at Eastport that they may be caught there by the quart. Moore also wrote (1898, p. 401) that "during the summer and fall dense bodies of Thysanopoda are seen swimming about the wharves at Eastport and at other places in the vicinity, and they are also extremely abundant on the ripplings at Grand Manan, which has long been famous as a herring fishery. Excepting the eyes and the phosphorescent spots beneath, which are bright red, the bodies of these shrimps are almost transparent, yet such is the density of the schools in which they congregate that a distinct reddish tinge is often imparted to the water. In the summer and early fall of 1895 they were especially abundant about the wharves at Eastport, and on one occasion, at least, they were left at low water several inches deep over a considerable area of one of the docks." Moore believed that Thysanoessa inermis was the species chiefly concerned, but in the light of subsequent observations it is probable that then, as now, it was outnumbered there by Meganyctiphanes. Our own observations, with information communicated by Doctor Huntsman, show that the passage of time has seen no diminution in the abundance of the latter in the Eastport-St. Andrews region in summer and early autumn. It is only in the extreme northeast corner of the gulf, perhaps east of Machias, that euphausiids appear regularly in estuarine situations; farther west and south the group, as a whole, are creatures of the open sea. Thysanoessa inermis (Kr0yer) n Thysanoessa inermis, as I have stated elsewhere (Bigelow, 1917, p. 283), occurs more regularly over the gulf as a whole than any other euphausiid, though it is not the most abundant locally. In July and August, as exemplified by the summers of " Chiefly Meganyctiphanes, but Thysanoessa as well, according to Smith (1879), Moore (1898), and our own observations. '' I follow Hansen (1911) in including under this name both Th. neglecta and Rhoda inermis, which, as he has shown, are merely varieties of the one species. 136 BULLETIN OF THE BUREAU OF FISHERIES 1912, 1914, and 1915, it occurred at about 50 per cent of our stations (fig. 48), with the records for those months distributed generally throughout the offshore parts of the gulf as well as over Georges and Brown's Banks and over the shelf off Marthas Vineyard and Nantucket. 71" Fig. 48.— Occurrence of the euphausiid shrimp, Thysanoasa inermis, for June, July, and August. #, occurred; O. n ot taken; X, records by Hansen (1915) . The hatched curve incloses the area where it has occurred at 50 per cent of the stations This species (figs. 48 and 49) has occasionally been recorded close to land in Massachusetts Bay and may be abundant temporarily in Eastport Harbor, as just noted, but its presence in these estuarine waters is only sporadic in summer. Nor PLANKTON OF THE GULF OF MAINE 137 did Doctor McMurrich detect it at all at St. Andrews at that season, though it occurred there in November, December, and January, and occasionally in February and March. In fact, we have usually found it wanting in summer throughout the Fig. 49.— Occurrence of the euphausiid shrimp, Thysanoeesa inermis, February to Aprii, 1920. •, occurred; O. nemo taken. The hatched curve incloses the area where it occurred in about 50 per cent of the stations for March and April coastal zone from Cape Cod to Grand Manan, with the 100-meter contour roughly marking its shoreward limit from Cape Ann to the mouth of the Grand Manan Channel at that season. But its regular presence over the shallow southern rim of 138 BULLETIN OF THE BUREAU OF FISHERIES the gulf, as well as close up to the land off Cape Sable and in Eastport harbor during the warm months, shows that it is not the shoalness of the water which holds it offshore, but either some influence of the coast line itself or the physical state of the water. Thus it is rather more oceanic in the gulf than its omnipresent and much more plentiful companion, the copepod Calanus finmarchicus, for the latter thrives right up to the outer islands and headlands, though its adults are seldom abundant in inclosed waters. The term "oceanic," however, as applied to Thysanoessa inermis, does not imply that it reaches the Gulf of Maine from the warm water of the Atlantic Basin to the east and south. On the contrary, we have never found it in our hauls outside the continental edge, either east or west of Cape Cod, except at one station (10349, July 24, 1916), where low temperature proved that the inner edge of the "Gulf Stream" lay some distance farther offshore. Nor did Hansen (1915) find it in gatherings taken over the slope abreast of the gulf, where other euphausiids — e. g., Nemato- scelis — occurred in abundance, though he records it from various localities over the outer part of the continental shelf within the limits of the gulf — e. g., off Marthas Vineyard, near Browns Bank, and south of Nova Scotia. It is evident from this that the warm and highly saline tropical water, which is never far out beyond the edge of the continent in these latitudes, is an effective barrier to the offshore dispersal of Th. inermis off the eastern United States, although it ranges southward regularly to southern New England every summer, and even accompanies the Calanus com- munity as far south as the latitude of Chesapeake Bay in cool summers (e. g., 1916) and probably every winter. In all this its occurrence in American waters parallels its distribution on the other side of the Atlantic, where it is distinctively arctic-boreal, as Kramp (1913, p. 544) points out, occurring chiefly in the northern Atlantic and in the adjacent parts of the Arctic Ocean from Franz Josef Land to West Greenland, and southward as far as the North Sea and the waters around Ireland. Thysanoessa inermis is present in the Gulf of Maine throughout the year, as proven by the fact that we have taken it there throughout the spring and summer, at several stations in September and October of 1915, twice (out of five stations) in November in 1916, and at about half the stations occupied during our midwinter cruise of 1920 and 1921. As I have just pointed out, winter is its season of greatest abundance at St. Andrews, but it shows no apparent tendency to work inshore off the coasts of Massachusetts at that season, for we did not detect it at all in tows taken near Gloucester every two weeks throughout the winter of 1912 and 1913. 73 The most notable seasonal fluctuation in the distribution of Th. inermis within the gulf (supposing its status in 1920 to be representative) is that it almost totally disappears from the southern deeps, from the eastern channel, and from Georges Bank in March and April, although it occurred at about 50 per cent of our stations around the coastal belt at that season (fig. 49). Our failure to find it over the eastern « For its occurrence from 1912 to 1916 see Bigelow, 1914a, p. 411; Bigelow, 1917, pp. 282 and 283; and Bigelow, 1922, pp. 133, 136, and 150. In the spring of 1920 it was detected at Stations 20046, 20049, 20054, 20057, 20059, 20060, 20070, 20073, 20075, 20079, 20080, 20085, 20086, 20088, 20092, 20093, 20094, 20097, 20099, 20100, 20101, 20102, 20105, 2O100, 20116, 20119, 20122, 20125, and 20126; as well as at the following stations from December, 1920, to January, 1921: 10490, 10494, 10497, 10499, 10500, 10502, and at stations 10507, 10508, 10509, and 10510 in March, 1921. PLANKTON OF THE GULF OF MAINE 139 end of Georges Bank during these months certainly was not accidental, for we made two traverses of the bank four weeks apart, and it was equally wanting at our several stations on the western end of the bank on May 17, a month when we have previously found it widespread in the inner parts of the gulf. It will require more than the one year's data to prove whether this vernal con- traction of the range of Th. inermis on the offshore side, which must be followed by a corresponding expansion in June to repopulate these waters to the extent that obtains in midsummer, is an annual occurrence. We have yet to learn how far the maintenance of the local stock of Th. inermis in the Gulf of Maine depends on the reproduction which takes place there and how far on immigration around Cape Sable from the colder waters of the Nova Scotian current, no attempt having yet been made to trace the life history of this shrimp in the gulf. It is probable that Th. inermis breeds successfully at least as far west as Cape Cod, and that it is represented among the considerable numbers of larval euphausiids which we have taken there side by side with medium-sized specimens and large adults of this species. Thysanoessa inermis has never been found in abundance at the surface in any part of the gulf except at Eastport, though it has often occurred in small numbers in the catches of the surface nets. On the other hand, our deepest hauls in the gulf have never yielded many, and the largest catches have all been in nets working at 40 to 80 meters depth. Thus it tends to congregate at about the same level as Calanus and is not associated with the Euchseta community of the deep basins, as its relative Aleganyctiphanes norvegica so often is. I can offer no data bearing on the actual numerical strength of Th. inermis in the gulf, nor could much dependence be placed on the results of vertical hauls in the case of so active an animal unless with larger nets than we have used. Our largest catches of it have been made near Cape Ann (August 22, 1914, station 10253), on the eastern end of Georges Bank (July 23, 1914, station 10223), near Cape Sable (August 11, 1914, station 10243), and off Marthas Vineyard (August 25, 1914, station 10259). Tliysanoessa longleaudata (Kr0yer) 74 This species, as Kramp (1913) and Holt and Tattersall (1905) have pointed out, is generally distributed in Arctic Seas and in the northern part of the Atlantic, ranging south to the west coast of Ireland and northern North Sea in European waters. On the whole, it is more northern and more oceanic in its affinities than Th. inermis, but, like the latter, the records for it in the Gulf of Maine are so widely distributed that it is to be expected anywhere in the offshore parts of the latter in summer (fig. 50), late winter, and early spring. Only three times in all our experience, however, have we detected it in the coastal zone inside the 100-meter contour at any season, and never in inclosed bays or estuaries. Thysanoessa longicaudata is far less numerous in the gulf than its relative Th. inermis, and occurs there far less regularly, having been detected at fewer than 25 per cent of our summer stations (fig. 50), and then usually in small numbers; nor " For the occurrence of this species in 1912 to 1916 see Bigelow, 1914a, 1917, and 1922. In the spring of 1920 it was taken at Stations 20045, 20046, 20054, 20057, 20060, 20064, 20065, 20066, 20069, 20070, 20073, 20075, 20076, 20077, 20079, 20080, 20086, 20087, 20100, 20101, 20107, 20112, 20116, and 20129. It was also taken in December, 1920, and January, 1921, at stations 10490, 10494, and 10502. 140 BULLETIN OF THE BUREAU OF FISHERIES does there appear to be much change in its status from season to season, for it was found at about 20 per cent of the stations occupied by the Halcyon during December, 1920, and January, 1921, and at about 25 per cent of the Albatross stations of Feb- © •• • • * x ••• • •••• 71" Fig. 50.— Occurrence of the euphausiid shrimp Thysanoessa longicaudala. X. locality records, February to May, 1920; # July to September, including Hansen's (1915) records ruary to May, 1920 (fig. 50). Although the locations where Th. longicaudala has actually been taken are not concentrated in the one side of the gulf or in the other, we have usually made our largest catches of it in the eastern part, both in spring PLANKTON OF THE GULF OF MAINE 141 and in summer. For instance, it was abundant on the edge of Georges Bank on March 13, 1920 (station 20071), and on Browns Bank on July 24, 1914 (station 1022S) . This phenomenon and the fact that we have found it at most of our stations along the continental slope abreast of Georges Bank and south of Nova Scotia, where inermis has usually proved wanting, is no doubt correlated w T ith its oceanic nature, and Hansen (1915) records Th. longicaudata from many localities over the slope south of Marthas Vineyard, often in great abundance. Evidently this shrimp is a characteristic inhabitant of the cool band of water of mixed origin which separates the tropical Atlantic (so-called "Gulf Stream") water from the continental shelf. Probably it comes as a wanderer from the east and north, and it may follow the outer part of the shelf at least as far south as the latitude of Chesapeake Bay in cool summers, as in 1916 (Bigelow, 1922, p. 151) ; but we have never found it at any station where the presence of a tropical planktonic community has betrayed a large admixture of " Gulf Stream" water. Judging from the boreal- Artie affinities of Th. longicaudata, it is probable that high temperatures and salinities form an impenetrable offshore barrier to its dispersal off the coasts of Nova Scotia and the United States. Bathymetric range. — We have yet to find Th. longicaudata on the surface in the Gulf of Maine in summer, most of the records of it for the three months, July to September, being in hauls from 80 meters or deeper, the shoalest from 50-0 meters (two hauls). An interesting example of its preference for deep water is afforded by its vertical distribution in the western basin on August 22, 1914 (station 10254), when there were none on the surface, and, allowing for the use of different-sized nets, many more at 235-0 meters depth than at 75-0 meters (Bigelow, 1917, p. 282). Although it is not so closely confined to the deeper strata of water during the early spring (for we found many on the surface over the eastern end of Georges Bank on March 13, 1920 (station 20070), and a few on the surface in the western side of the basin 10 days later (station 20087)) most of the spring records of the species in the gulf have likewise been from depths greater than 75 meters. Thus, it finds its most favorable habitat at a deper level than that of Th. inermis. Judging from the rather conflicting statements of European students (Holt and Tattersall, 1905; Hansen, 1908; Tattersall, 1911; Kramp, 1913), Th. longi- caudata is equally a deep-water form on the other side of the Atlantic, though it comes right up to the surface of the water about Iceland (Paulsen, 1909). Probably the warm layer that forms over the surface of most boreal seas in late spring and summer acts as a barrier to its upward dispersal during the w r arm half of the year, just as high temperature confines it offshore, abreast of the Gulf of Maine. At any rate, its avoidance of the surface in summer and of the coastal zone at all seasons makes it an inhabitant of low temperatures and comparatively high salinities in the Gulf of Maine, where the water in which most of the stock lives ranges from about 2° to about 10° in temperature and upward of 32.5 per mille in salinity. Whether Th. longicaudata breeds in the Gulf of Maine or appears there only as an immigrant from the north is yet to be learned. Probably it is endemic there in small numbers, like other planktonic animals with a similar affinity for low temperature, but depends as much on more or less constant immigration from 142 BULLETIN OF THE BUEEAU OF FISHERIES northern sources, either around Cape Sable or from the mixed water along the outer part of the continental shelf, for the maintenance of its numbers within the gulf. 71' 70* 69' 68" 67* 66' 1 1 1 Lnd ' ' '■ >> ^T PAN /<-/XS \d f" • ^\ S» -^•■■" )/? PEN))B® ^ ,,,./" . X'V "/sable/ { /•'-•■•' X V^ f / ... x • X l --'\ J / (1 .-""--, 431 S^», c ... x ^~£ X '" ^ 5» / 4 4 " ; x f • V* A • x V./ fw • \ X 1 \ '••■ x 4Z r ^1 ^J&^§ -. <£» ^ \ ( ; \ r ■ O + ,.•-"" 4 /' \ 4 42- , ' 4V 4 + + X 4 4- 4 41' •x ..* '■•-,. J,-"* ■ ---.,. '"• v « • f* X 4 * + • A 40 4 4 4 40' • • % • •••• v • • 71* 70' 69" 68" 87" 66" Fig. 51. — Occurrence of the euphausiid shrimps Nematoscelis megalops and Thysanoessa gregaria. #, locality records for Nematoscelis, July to September, including Hansen's (1915); A, locality records for Nematoscelis, February to May, 1920; X, locality records for Thysanoessa gregaria Thysanoessa gregaria, G. O. Sars The fact that Thysanoessa gregaria occurs side by side with its boreal-Arctic relatives Th. inermis, Th. longicaudata, and Th. raschii in the Gulf of Maine is, as Doctor Tattersall writes me, an interesting phenomenon; for, unlike them, it is a PLANKTON OF THE GULF OF MAINE 143 tropical and warm-temperate form which undoubtedly reaches the gulf from the warmer waters offshore and not from the cooler seas to the east and north. Its local presence is sure evidence of an influx of such water into the gulf. As I have noted elsewhere (Bigelow, 1917, p. 284), Th. gregaria is much less common in the gulf than Th. inermis, or, I may add, than Th. longicaudata; but the records for 1912 (Bigelow, 1914a, p. 412), 1914, and 1915 (Bigelow, 1917, p. 285), show that in summer it is to be expected anywhere on Browns and Georges Banks, along the continental slope south of Nova Scotia, in the Eastern Channel, and in the inner parts of the gulf as well (fig. 51). We have never found Th. gregaria in any abundance anywhere in the gulf north of the offshore banks, but we took it in numbers on the western part of Georges Bank on July 20, 1914 (station 10216), and Hansen (1915) detected it in the gatherings from two deep stations south of Marthas Vineyard. Curiously enough, however, in spite of its well-established warm-water origin, we did not find it at our saltest and warmest station east of Cape Cod, where the plankton was distinctly tropical in aspect (station 10218, July 21, 1914), nor did it appear in the tow nettings along the slope from Georges Bank to the latitude of Chesapeake Bay during July, 1916. Our records for this species 75 prove that it is more seasonal in its occurrence in the Gulf of Maine than are its northern relatives, nearly all being for August; and its history in 1915 in particular, when it was not detected until August, although we made frequent tows in various parts of the gulf during the spring and early summer, shows that it increases in numbers and penetrates farther and farther into the gulf with the advance of summer. Its presence there seems short lived, however, for we did not find it at all during October, 1915, or November, 1916; and although the tow yielded an odd specimen off Glouces- ter on December 23, 1912, we sought it in vain in December, 1920, and January, 1921, and during the late winter and spring of 1920. Probably the correct explana- tion for its absence from the Gulf of Maine during the cold half of the year is that the species vanishes thence when the stock that has entered the gidf during the summer perishes at the onset of autumnal cooling. It does not reappear until the surface waters are once more sufficiently warm for its existence, which means mid- summer. Thus it closely parallels Sagitta serratodentata (p. 58) in its status in the gulf, and there is no reason to suppose that Th. gregaria ever breeds successfully there. Thysanoessa rascliii, M. Sars This species (fig. 52) resembles Th. longicaudata in its Arctic-boreal nature (Kramp, 1913; Zimmer, 1909), and ranges southward along the European coast to the northern part of the North Sea, to the longitude of Nantucket and probably still farther, off North America; but, as I have noted in an earlier report (Bigelow, 1917, p. 284), it is much less common in the Gulf of Maine in summer than is either Th. inermis or Th. longicaudata. It was not detected there at all in the hauls of July and August, 1912, and appeared at only three stations within the limits of the gulf during the summer of 1914 — two of them in its northeastern part and the third off Marthas Vineyard (Bigelow, 1917, p. 282). It was not detected at all during the " For lists of the Gulf of Maine records of Th. gregaria, 1912 to 1915, see Bigelow, 1914a, p. 411, and Bigelow, 1917, p. 282. 75898—26 10 144 BULLETIN OF THE BUREAU OF FISHERIES summer of 1915, was represented by occasional specimens only in Massachusetts Bay and over the continental slope south of Nantucket in July, 1916 (Bigelow, 1922, pp. 133 and 138) , 76 and Hansen (1915) adds only one station on Browns Bank 71" 70" 69' 68" 67' 66' u$ ' ' y/\ /b $ And j>AN /& 44' f +- PEi>j)>B»cpTr /j3 " X +- e" } NOVA ( SCOTIA 1 + U - ■'/ (Yarmouth Portland C ; • * 1(1 CAPE \ m *l SABLE/ / * N^s/r / ' <) • X ( ""\ w + / » * 4 .-■ i- • '"■s. , \ ;' 43' <5 \ <7*s • -^ 42' • ■» +,"■'-' 4- • •\ t" 4Z 41' + + X + + +- + ;.r 40' •4- 4- X + 4- 4- 4- 40' 71" 70' 63' 68' 67' 66' Fig. 52. — Occurrence of the euphausiid shrimp Thysanoessa rasdtii. 0, locality records, February to May, 1915, 1920, and 1921; X, August, 1912 to 1916; A- August records by Hansen (1915) (August, 1877) and a second off the northern end of Cape Cod (for the same month in 1881) to this brief list. 77 Even during the cold July of 1916 we found no Th. raschii west of Nantucket, either near shore or over the slope, though the range of 76 Doctor McMurrich did not detect it at St. Andrews. '" He lists many localities for it in the Gulf of St. Lawrence, where it is evidently a common species. PLANKTON OF THE GULF OF MAINE 145 Th. lorujicaudata, a species equally northern in its faunal status, then extended south- ward beyond the latitude of Delaware Bay. In short, the Gulf of Maine and the continental shelf abreast of Marthas Vineyard and Nantucket together form the southern outpost of Th. raschii in summer. Thysanoessa raschii is apparently no more plentiful in the gulf in autumn, for we have not noted it either in October or November and only twice during our December-January cruise of 1920-1921 (occasional specimens off Cape Elizabeth on December 30, station 10494, and off Lurcher Shoal on January 4, station 10500). Neither did we detect Th. raschii in any of the tows made off Gloucester from November, 1912, until March, 1913, but it swarmed a few miles north of Cape Ann during that April. The first specimens were noted on the 22d in the neighborhood of the Isles of Shoals; on the 23d (when, as it chanced, none were taken) Mr. Welsh wrote in his field notes of " the pollock schools feeding on shrimps, which were also in dense schools" (Bigelow, 1914a, p. 408); and a large catch of them made off Boon Island on the 25th, when Welsh saw "the feed (shrimps) breaking water trying to get away from the pollock, which are after them," estab- lished their identity as this species. At that time the shrimp, as he noted, were concentrated "in dense swarms apparently 6 inches to a foot below the surface," and although these schools had dispersed by the first week in May, so that they were no longer in evidence from the vessel, he still found them near the Isles of Shoals in abundance on the 12th and 13th of the month. There is no knowing how much longer they persisted there, for we did not revisit that region until the following August, when they had disappeared. We have never found this species so plentiful in the gulf since then, but in 1920 it appeared at about 25 per cent of the stations occupied by the Albatross in March and April, 78 twice in considerable numbers — that is, off Cape Elizabeth on March 4 (station 10059), and a few miles north of Cape Ann on May 8 (station 20122). It again appeared in abundance in this same general region in the spring of 1925, when tows from the Fish Hawk at two stations 5 to 7 miles southwest from the Isles of Shoals yielded large catches of Th. raschii on April 7, with a few Th. inermis. The facts just outlined are enough to show that the spring is the period of maximum abundance, the summer and autumn of minimum abundance, for Th. raschii in the Gulf of Maine, and the coastal zone between Cape Ann and Cape Elizabeth a center of abundance for it. Most of our records for it have been located either around the periphery of the gulf within or close to the 100-meter contour or in the shoal waters over Georges Bank (fig. 52) , but more data are needed to show whether this apparent concentration in the coastal zone is significant. Most of the specimens of Th. raschii that Welsh took during its period of abund- ance in April and May, 1913, were large, and we again found large adults in Ipswich Bay — that is, in the same general region — on May 8, 1920 (station 20122); but with this species so rare in the gulf in summer, few, if any, of the larvae resulting from such local centers of reproduction can survive there. Thus it is chiefly as "Stations 20044, 20059, 20060, 20070, 2O073, 20075, 20080, 20085, 20092, 20093, 20096, 20097, 20099, 20102, 20105, 20116, 20122, and 20125. 146 BULLETIN OF THE BUREAU OF FISHERIES an immigrant, not as a regular inhabitant, that Th. raschii occurs within the Gulf of Maine, where it occupies much the same faunal niche as the northern copepods, Calamus Jiyperboreus and Metridia longa (pp. 212 and 245). Nematoscelis megalops, G. O. Sars The presence of this euphausiid at our outermost stations has been mentioned in an earlier chapter (p. 56), and we have also found it occasionally within the Gulf— that is, off Mount Desert Rock on August 16, 1912 (station 10032), and at eight stations during July and August, 1914 (Bigelow, 1917, p. 282), as illus- trated on the accompanying chart (fig. 51). Most of these scattering records are from the eastern and southeastern parts of the gulf, as might be expected of a visitor from offshore, and it is probable that the few Nematoscelis that were present over Browns Bank and in the Eastern Channel in July, 1914, represented the innermost fringe of a swarm of this species that populated the waters over the continental slope southeast of Cape Sable at the time. Our summer records for Nematoscelis within the gulf are based on very few specimens in each case; nevertheless, this is the season at which it most often occurs, for we have never detected it there or even on Georges Bank during autumn, winter, or spring; but the fact that the Albatross towed it in fair numbers off the western end of Georges Bank on February 22 (station 20044) and southeast from Cape Sable on March 19, 1920 (station 20077), is sufficient evidence that it is to be expected along the continental slope abreast of the gulf during the cold half of the year as well as the warm. It not only occurs more constantly along this belt than within the gulf, but is much more abundant there in actual numbers— witness the large catches made at our outermost stations off Cape Sable by the Grampus on July 28, 1914, and June 24, 1915, and off the southern slope of Georges Bank on July 24, 1916 (Bigelow, 1922, p. 13S). Hansen (1915) likewise records it from many localities over the contmental slope off Marthas Vineyard, but not from the Gulf of Maine, from Georges Bank, or from anywhere on the continental shelf east of Cape Cod. This evidence supports the general thesis (Hansen, 1915; Zimmer, 1909; Kramp, 1913) that Nematoscelis megalops is typically an oceanic form of warm-temperate affinity, at home in the open Atlantic Basin; and since it is known to range as far north as Iceland and to the waters east of Newfoundland during the warm season, it is not surprising that it should occasionally enter the Gulf of Maine with the general indraught into the eastern side of the latter. We have no evidence that Nematoscelis ever breeds there successfully, however, nor is this at all likely, the probable fate of these rare im- migrants being either to withdraw once more to warmer regions as the water cools in autumn (if they have been able to survive the vicissitudes of life in a foreign environment so long), or to perish like other visitors from offshore, such as Thy- sanoessa gregaria and Sagitta serratodentata (pp. 142 and 320). Euphausia krohnii, Brandt Euphausia Irohnii (the only species representative of this large genus so far detected in the Gulf) has not been taken in the inner parts of the Gulf of Maine but was sparsely represented off the southern slope of Georges Bank (station 10220) PLANKTON OF THE GULF OF MAINE 147 and in the Eastern Channel (station 10227) in July, 1914. As has been noted above (p. 134), it occurred in abundance over the continental slope southeast of Cape Sable (station 10233) a few days later. We also found it at this general locality on June 24, 1915, which, with one record at the same relative position off Marthas Vineyard on August 26, 1914 (station 10261), completes the list for the Gulf of Maine cruises. All the records given by Hansen (1915) are from well outside the continental edge, though he lists so many captures of E. Jcrohnii that the species is evidently one of the commonest of euphausiids off the slope abreast of Cape Cod and at least as far east as off La Have Bank, and perhaps still farther. Thus, on the basis of actual record, Euphausia is hardly to be expected inside the outer rim of the Gulf of Maine except as a straggler from the warmer Atlantic. Meganyctiphanes norvegica (M. Sars) 79 While this brilliantly phosphorescent shrimp, the largest and most familiar of all euphausiids in the Gulf of Maine, has not appeared as regularly in our tow nets in most parts of the Gulf as has Thysanoessa inermis, it occurs locally in such abundance that it is far more important economically than the latter. The locality records for Meganyctiphanes are distributed generally enough to show that it may be ex- pected anywhere within the gulf north of the Cape Cod-Cape Sable line during the summer and early autumn, both in the deep basin and along shore. Nor does the chart (fig. 53) show any apparent concentration in distribution in one or the other side of the gulf at that season, if the considerable number of stations which the Grampus has occupied in the Massachusetts Bay region be allowed for. I have just mentioned (p. 135) the swarms of Meganyctiphanes that regularly appear during the warm months about St. Andrews and in Eastport Harbor, where numbers of these shrimps can usually be seen darting to and fro at the surface on almost any calm day in August. It seems that this region of violent tidal currents is the only part of the Gulf of Maine where Meganyctiphanes regularly enters the estuaries, but it appeared in the shallows at the head of Frenchmans Bay for a brief period in June, 1923, when a number were collected by Dr. Ulric Dahlgren. Me- ganyctiphanes appeared there again in abundance in the summer of 1924 (Dahlgren, 1925, has already reported these incursions). We have never taken it in our tow nettings inside the off-lying islands west or south of this at any season, and although neither comparatively shoal water, per se, nor the general neighborhood of the coast is any bar to its presence — witness its occurrence in Massachusetts Bay and in the Eastport-St. Andrews region — most of the Grampus, Albatross, and Halcyon records for it have been from the basin of the gulf outside the 100-meter contour. We have found it only once on German Bank (August 14, 1912, station 10029), once on Browns Bank (July 24, 1914, station 10228) and twice on Georges Bank (station 10223, July 23, 1914, and station 20124, May, 17, 1920), although it has been taken in the Woods Hole region and in shoal water south of Long Island (Hansen, 1915). '• For station records for this species from 1912 to 1916, see Bigelow, 1914, p. 118; 1914a, p. 411; 1915, p. 273; 1917, p. 282; and 1922, p. 133. During the spring of 1920 it was taken at stations 20049, 20052, 20053, 20054, 20055, 20050, 20057, 20076, 20079, 20081, 20087, 20088, 20093, 20097, 20098, 20100, 20102, 20113, 20114, 20115, 20122, 20126, and 20127. In December-March, 1920-1921, it was taken at Stations 10490, 10491, 10494, 10497, 10499, 10500, 10502, 10507. 10509, and 10510. 148 BULLETIN OF THE BUREAU OF FISHERIES The Gulf of Maine is the most southerly important center of abundance for this shrimp, and although it ranges much farther southward along the continental slope, most of Hansen's (1915) locality records of it from abreast of Cape Cod to the latitude Fig. 53. — Occurrence of the euphausiid shrimp Meganydiphanes norvegica, July to September 15. #, locality records. The hatched curve incloses the area of regular occurrence in summer and early autumn of Delaware Bay (37° 25' N. lat.) were based on odd specimens only, and we did not detect it west of Cape Cod in the summers of 1913 or 1916. The frequency with which it has been recorded in deep water off Cape Cod and off southern New England PLANKTON OF THE GULF OF MAINE 149 reflects the number of tow nettings that have been carried out along that part of the slope rather than any general abundance of Meganyctiphanes there, corresponding to which we have found it at only one of our stations off the slope of Georges Bank. The scarcity of Meganyctiphanes over Georges Bank and in the southeastern deeps of the gulf generally, in spring as well as in summer, suggests that the few specimens that drift westward beyond Nantucket Shoals along the continental slope are migrants, either from along the Nova Scotian coast to the eastward (and possibly even from as far away as the Gulf of St. Lawrence) or from the western side of I ho Gulf of Maine, not from the eastern or central parts of the latter. The alternation of the seasons sees a corresponding expansion and contraction in the area of distribution of Meganyctiphanes in the inner part of the Gulf of Maine. Probably this is at its narrowest late in the winter and early in the spring, for from February to April, 1920, we had only two records of it anywhere inside the 100-meter contour in the whole coastal zone on both sides of the gulf — one for half a dozen specimens near Mount Desert Island on March 3 (station 20056), and the other for a single specimen off Yarmouth, Nova Scotia, on April 9 (station 20102) — although we took it at many stations marked on the chart (fig. 54) in the central and northeast deeps of the gulf during that period. Nor did we find it anywhere on Georges or Browns Banks during these months. In fact, it is seldom that the local presence or absence of any one of the larger members of the zooplankton can be defined so sharply as in this instance. Thus it is evident that Meganyctiphanes withdraws altogether from the shallows of the gulf within the 100-meter contour during the coldest season, unless, perhaps, it persists locally around the shores of the Bay of Fundy; and our failure to find it at any of our February-May stations over the continental slope abreast of the gulf suggests that it vanishes similarly from this portion of its range in late winter and spring. Thus its area of distribution in the Gulf of Maine is then cut off from its more northerly centers of occurrence by an extensive zone off southern Nova Scotia and extending around Cape Sable, where there are no Meganyctiphanes at that season, which is not the case for Thysanoessa inermis (p. 135) or for Th. longicaudata (p. 139). During the later spring and early summer Meganyctiphanes disperses in all directions in the Gulf of Maine, to occupy the much more extensive range over which we have found it occurring in midsummer, and reappears over the slope off Marthas Vineyard. The contraction of the range of Meganyctiphanes, from its maximum in summer and early autumn to the spring state just outlined, may commence as early as October in the western side of the gulf, for we have not taken it anywhere in the Massachusetts Bay region in October, November, December, or during the winter of 1912-1913. It persists until later in the coastal belt north of Cape Ann, where we towed it near the Isles of Shoals and off Monhegan Island on November 1 and 2, 1916 (stations 10400 and 10402) ; off Cape Elizabeth, near Mount Desert Island, in the northeastern part of the basin, in the Fundy Deep, and off Lurcher Shoal during the last days of December and first week of January of the winter of 1920-1921 (stations 10494, 10497, 10499, 10500, and 10502). 150 BULLETIN OF THE BUREAU OF FISHERIES I have already mentioned the fact that the deepest water in the northeast corner of the basin, off Grand Manan, has yielded an abundance of Meganyctiphanes in March, April, May, and June, as well as during the later summer (p. 134). Consider- y <*•■" Fig. 54.— Occurrence of the euphausiid shrimp Meganyctiphanes norvegka, February to April, 1920. •, locality records; O. not taken. The hatched curve incloses the area where it occurs regularly in early spring able numbers were also taken by the Halcyon in the deepest haul (150-0 meters) near-by on January 5, 1921 (station 10502), proving that this serves as a reservoir for Meganyctiphanes throughout the year. This shrimp has also been taken at most PLANKTON OF THE GULF OF MAINE 151 of our stations in the western side of the basin of the gulf, except on May 5 and June 26, 1915 (stations 10267 and 10299). The triangular extremity of the deep trough north of latitude 44° is the only offshore locality in the gulf where we have found it constantly abundant. Moderate catches of Meganyctiphanes were also made on Browns Bank on July 24, 1914 (though our hauls at about this same location just one month earlier in 1915 yielded none), in the Fundy Deep on March 22, 1920 (station 20079), in the center of the gulf on April 17 of that year (station 20113), and it has been found swarming in Massachusetts Bay at least once in the past (Hansen, 1915). However, we have never taken more than a few specimens at any station there in all our cruising; and the fact that, with the exceptions just recorded, our hauls in other localities have usually yielded only from one or two to a couple of dozens of these shrimps is evidence that Meganyctiphanes seldom swarms anywhere in the gulf except in the northeastern part. It is not possible to estimate the actual numerical strength of Meganyctiphanes at any of our stations, because the small nets that have been used for the vertical tows in the Gulf of Maine do not yield reliable data for so active an animal and one which so commonly occurs in shoals. Two stations occupied by the Albatross in the center of abundance for this shrimp off Grand Manan during the spring of 1920 illustrate this imperfection of the record, for the vertical haul of April 12 (station 20100) did not yield a single specimen — that is, missed the school of shrimps alto- gether — although the catch of the horizontal haul — about 50 specimens — was about the same as on March 23 (station 20081), when the vertical haul indicated a Meganyctiphanes population of about 275 below each square meter of sea surface. Although Meganyctiphanes is not neritic (for it is not dependent on the bottom at any stage in development or associated with the coast line in its distribution), it is a creature of the banks water on both sides of the Atlantic and is not oceanic in the typical sense, finding the high temperatures and sahnities outside the edge of the continent an absolute barrier to its offshore dispersal along the American littoral. At one place and season or another Meganyctiphanes occurs over a very wide range of temperature in the Gulf of Maine, certainly from upward of 15° to as low as 2 to 3°, and possibly even colder; but it was rare at the coldest stations (0.5 to 2.5°) during March and April, 1920, with only three records from water as cold as 2°, s0 the temperature being higher than 3° and in most cases as warm as 4° to 5° at the five localities and at the deeper levels where it was most abundant during those months, although the surface strata might be colder. 81 It follows that almost the entire local stock of the species was then living in tempeartures of 3.5 to 5°. Therefore 3 to 4° may be set tentatively as the coldest favorable for the existence of Meganyctiphanes in the Gulf of Maine, a thesis corroborated by its absence from Ipswich Bay on April 9, 1920 (station 20092), when the temperature at 20 to 30 meters was still only 2.5°, coupled with its presence there on May 8 (station 20122), by which date the temperature had risen to 3 to 4° at that level. » One specimen at station 20054, 100-0 meters, temperature 1.7 to 2.5°; occasional examples at station 20056, whole column of water, 0.5 to 1.9°; 3 specimens at station 20057, whole column of water, 1.9 to 2.2°. " Station 20079, 180 meters, about 4°; station 20031, 140 meters, 4.5°; station 20100, 100-0 meters, about 4.5°; station 20113, surface, 3.3, and 4.5° at about 130 meters; station 20114, 110 meters, about 4°. 152 BULLETIN OF THE BUREAU OF FISHERIES These observations make it probable that Megancytiphanes deserts the shallow coastal zone as winter draws to its close, in order to avoid the extreme chilling to which this part of the gulf is subject; but data for a single year, and especially for one as cold as 1920, are not enough to settle this point definitely. On the other hand, the great majority of our captures of Meganyctiphanes have been from water colder than 12°, both in the offshore parts of the gulf and on the surface about Eastport and St. Andrews. But off Cape Cod, on August 23, 1914 (station 10256), we found it indifferently on the surface at a temperature as high as 19.5° and in the much cooler (5 to 6°) layers deeper down, and probably the Massachusetts Bay swarm mentioned below (p. 153) was likewise living in water at least as warm as 16°. Evidently the highest temperatures that ever obtain in the open waters of the Gulf of Maine are not immediately fatal to Meganyctiphanes, though it is doubtful whether it could long survive water so warm; nor does it always avoid it, although it may cease its upward swimming to do so or sink a few fathoms to escape it once it has come up to the surface. Nevertheless, judging from the distribution of Mega- nyctiphanes in other seas, it is probable that a constant high temperature is not favorable for it, and I think it safe to set 12 to 15° as the upper limit for its per- manent existence, and especially for its reproduction. Within the limits of 3 to 15° it is practically eurythermal in the Gulf of Maine, both horizontally and vertically, and its distribution there is equally independent of local and vertical differences in salinity, for it occurs indifferently over the whole range — that is, from 31 per mille or less to 34 per mille — except perhaps in the very freshest water at the time of the spring freshets. This parallels its distribution in European seas, where it is common in the Skager-Rak in salinities ranging from as low as 28 to 30 per mille to as high as 34 to 35 per mille at different seasons (Kramp, 1913). Apparently there is nothing in the physical state of the water over Georges Bank to account for the scarcity or absence of this euphausiid there, nor can a cause be assigned for this apparent anomaly in its distribution until its life history has been traced in more detail. The bathymetric distribution of Meganyctiphanes in the Gulf of Maine remains puzzling. Most of our summer records for it in the offshore parts of the gulf have been from deeper than 40 meters or so, and when this shrimp has occurred on the surface at that season it has usually been represented more numerously at some deeper level, a rule Illustrated by two stations in the western basin (August 22 and 23, 1914), when the number of Meganyctiphanes taken in the several hauls was as follows : Station Depth in meters Number of speci- mens Station Depth in meters Number of speci- mens 45-0 225-0 13 38 50 10256 45-0 8 35 Not only have we taken it right down to the bottom of the deepest trough of the guff, but it is only in the lowest strata of the latter that it occurs regularly and in numbers throughout the year, except in the Eastport region. To balance against PLANKTON OF THE GULF OF MAINE 153 this apparent preference for considerable depths is the fact that the small surface net captured no fewer than 111 large specimens in the center of the gulf on April 17, 1920, at 2 p. m. (station 20113), while the haul from 120 meters took only three, though there were many of these shrimps at 1 10 meters, but none on the surface only 35 miles distant to the westward (station 20114), that same day. S. I. Smith (1879 p. 89) likewise found it in shoals on the surface "on the mackerel ground" off Casco Bay, both day and evening during the warm months 40 years ago. It swarms on the surface in the Eastport-St. Andrews region in midsummer and earl}- autumn, as just remarked (p. 147), and although recent records for it in Massachusetts Bay have all been from depths of 40 meters or deeper, quantities of Meganyctiphanes were taken at the surface at the mouth of the bay on July 7, 1894, in dip nets from the rail of the Grampus; and they were so abundant there at a depth of less than 2 fathoms two days later that a large number found their way into the fish well of the vessel (Hansen, 1915). Thus, while the normal habitat of Meganyctiphanes is in the low temperatures and darkness of the deeper strata in the trough of the gulf, it may rise to the surface anywhere at any time. In the Eastport region it may be brought up involuntarily by the active stirring of the water which takes place there, and the constancy of this type of vertical circulation may account for the regularity of its presence at the top of the water there, expecially in view of the low surface tem- perature that characterizes that locality (10 to 12° in summer and early autumn). The Massachusetts Bay region, with surface readings of 16 to 18°, is nearly the warmest part of the gulf in midsummer, so Meganyctiphanes is not prevented from making occasional excursions upward to the top of the water even by temperatures so high that a prolonged stay would probably prove fatal. Furthermore, such excur- sions in this part of the gulf during the warm months involve voluntary upward swimming, the vertical currents being weak and the water highly stable, with its density much the lowest at the surface. Neither do they correspond to the diurnal vertical migrations shared in by many copepods (p. 25), because the appearances of Meganyctiphanes at the surface appear to be independent of the time of day. There- fore, the actual captures so far recorded do not indicate any definite phototropism on its part, positive or negative, although it is doubtful whether it could long survive the full illumination of bright sunlight. Experience in most parts of the Gulf of Maine is therefore in line with Paulsen's (1909) conclusion that when Meganyctiphanes visits the surface in Icelandic waters it is not as a direct response to temperature (to which I may add salinity) or to the degree of illumination, but in pursuit of food. It is also brought up by vertical currents, where these are active. The depth at which Meganyctiphanes is most plentiful is more definitely limited, and the relationship between its vertical occurrence and temperature is closer in North European waters than in the Gulf of Maine. Off Ireland, for instance, and in such parts of the North Sea as it visits, this euphausiid lives chiefly in the deeper layers of water, reaching its maximum, according to Tattersall (1911), at about 200 meters. In the Skager-Rak (Kramp, 1913, p. 542) it carries out a more or less definite vertical seasonal migration, always seeking the coldest level, which leads it to the surface in winter and down to lower levels in summer. 154 BULLETIN OP THE BUREAU OF FISHERIES Breeding habits. — The spawning of Meganyctiphanes has not actually been observed either in American or European waters, but it seems certain that this genus either does not carry its eggs with it at all after they are extruded, as some other euphausiids do, or that it nurses them only for a brief period at most, both because ovigerous females have never been seen, so far as I can learn 82 (Holt and Tattersall, 1905), and because eggs probably ascribable to this species have been found free floating in the one-celled stage by Sars (1898) and by Lebour (1924a). It is true that the eggs of Meganyctiphanes have not been identified with absolute certainty from among the plankton. Sars (1898), however, thought it probable that at least some of the euphausiid eggs 83 about 0.7 to 0.8 millimeter in diameter, which he found in Christiania Fjord where Meganyctiphanes is plentiful, had that parentage. Similar eggs had already been recorded from the Clyde area, a center of abundance for Meganyctiphanes, by Brook and Hoyle (1888). Holt and Tattersall (1905, p. 103), too, have assigned to this genus certain loose ova found side by side with Meganyctiphanes and occasionally even clasped between its thoracic legs, among various articles of prey, though without describing the dimensions or appearance of the eggs in question. Lebour (1924) has recently ascribed to this same parentage certain euphausiid eggs from the English Channel, because of the characters of the larvas hatching therefrom. Brook and Hoyle, Sars, and Lebour all agree in describing these eggs (the correct identification of which is made practically certain by cumulative evidence) as inclosed by a perfectly transparent capsule 0.7 to 0.8 millimeter in diameter, the ovum proper having a diameter of approximately 0.3 to 0.4 millimeter. Thus, when first set free in the water they much resemble buoyant fish eggs with wide perivitelline membrane; but cleavage being holoblastic and the development of the nauplius plainly visible within the egg, thanks to its transparency, their crustacean nature is apparent almost from the beginning. Euphausiid eggs are so characteristic in appearance, also, that there is no danger of confusing them with any other buoyant eggs. Our own hauls in the Gulf of Maine have yielded considerable numbers of eggs of this same type and size in various stages of development. We first detected them in a surface tow in the Grand Manan Channel, off Campobello Island, August 19, 1912 (in the report for that year (Bigelow, 1914, p. 104) they were referred to through error as "balanus" eggs). These were for the most part in early cleavage stages, a few in various stages up to the fully formed nauplius ready to hatch. Eggs of this same type, as well as the recently hatched nauplii, were again taken on the 22d of the month off Penobscot Bay (station 10039). Since that time we have detected similar eggs in the Fundy Deep and off Mount Desert Island in June (stations 10282, 10284, and 10286, June 10 to 14, 1915) and off the mouth of the Grand Manan Channel on July 15, 1915 (station 10301). It is not safe to say that all these eggs are Meganyctiphanes, for Lebour (1924) found eggs of Thysanoessa inermis indis- tinguishable from them; but the strong probability that at least part of them belong » The considerable series of large adults which I have examined contained none. « Metschinkoff (1871, pi. 34, fig. 1) first described the peculiar and very characteristic buoyant eggs of this group of pelagic Crustacea. PLANKTON OF THE GULF OF MAINE 155 to the former suggests that Meganyctiphanes spawns in summer, which fits in with the season of abundance of euphausiid larva? (p. 134) and points to the noitheaatern part of the gulf, where this shrimp is so abundant, as its chief spawning ground. Nothing is yet known of the seasonal occurrence or distribution of the larvse of Meganyctiphanes in the Gulf of Maine except that juveniles of the species were taken in some numbers off Cape Cod on July 19, 1914, in a haul from 70 meters (Bigelow, 1917, p. 282, station 10213). Very likely this genus was represented among the larval euphausiids taken on the surface off Cape Elizabeth on August 14, 1913 (station 10103); in Massachusetts Bay and off Cape Cod in July, 1916 (Bige- low, 1922, p. 133, and station 10343); and off the cape in August, 1914 (Bigelow, 1917, p. 2S3). These, however, have not been studied. 84 McMurrich, too, found young (unnamed) euphausiids common at St. Andrews from April until August, probably the offspring of the two pelagic shrimps Meganyctiphanes and Th. inermis, which are so plentiful in that region. However, larval euphausiids of any sort have always been very rare in our offshore catches in the northeastern part of the gulf, notwithstanding the constant presence of the adults there. Hansen (1915, p. 6S), I may add, records "immense numbers of older larva?" of Meganyctiphanes taken on May 25, 1891, over the 50-meter contour south of Shinnecock Light, Long Island, which is more than 2° of longitude farther west than the adults of this euphausiid have ever been found in any number. The possibility that adult Meganyctiphanes, in company with the general Calanus com- munity, may spread farther west and south over the shelf during the cold season than it does in summer makes it unsafe to assume that the larva? in question had drifted to the locality of capture from a more easterly birthplace. (Compare, in this connection, the status of Thysanoessa inermis west of Cape Cod, p. 138.) Although the evidence that the Gulf of Maine is a successful breeding ground for Meganyctiphanes still lacks something of proof positive, it is probable that this shrimp is not only regularly endemic there but that the northeastern part of the gulf is one of the most important centers of production for it off the American coast, and one, too, which receives few accessions from the north but forms a distinct and practically isolated colony. The relative distribution of euphausiid eggs and larvae, like that of pelagic fish eggs and larvse, is consonant with a general drift around the shore of the gulf with the dominant anticlockwise eddy, from the Bay of Fimdy to- ward Cape Cod, on the part of the developmental stages. Thysanopoda acutifrons, Holt and Tattersall The claim of this species to mention here rests on a single record — five specimens from the southeast corner of the gulf, July 23, 1914 (station 10225), identified by Dr. W. M. Tattersall (Bigelow, 1917, p. 282). Other euphausiids The species discussed above are the only euphausiids actually identified from within the Gulf of Maine or from the shoal waters over its southern rim up to the present time. Sundry other members of this group have been taken at one time or 81 According to Lebour (1924a) the larval stages or Meganyctiphanes and Thysanoessa are easily recognized. 156 BULLETIN OF THE BUREAU OF FISHERIES another at the outermost stations, between longitudes 71 and 65° and north of latitude 39°, both in the earlier collections of the Bureau of Fisheries, reported on by Hansen (1915), and during the more recent Gulf of Maine explorations, the latter identified by Doctor Tattersall. 85 The combined list is as follows: Beniheuphausia ambylops, Thysanopoda orientalis, Euphausia americana, E. mutica, E. brevis, E. tenera, E. hemigibba, Stylocheiron carinatum, S. abbreviatum, Thysanoessa parva, Nematoscelis atlantica, N. microps, and N. tenella. These are all oceanic species, any of which may be expected to occur occasionally in the southeastern corner of the gulf; hence a lookout should be kept for them in future collections from that region. HYPERnD AMPHIPODS Euthemisto The genus Euthemisto is one of the most characteristic, if not abundant, mem- bers of the plankton of the offshore waters of the Gulf of Maine. How regularly it is distributed there in summer (fig. 55) and over the shore banks as well appears from the fact that it has been taken at at least 90 per cent of our stations outside the immediate coastal zone, as bounded by the 100-meter contour on our July and August cruises of 1912, 1913, 1914, 1915, and 1916. Inside this zone, on the con- trary, it fails almost as regularly at this season, with only four or five summer records for it from water shallower than 100 meters along the western side of the gulf. Simi- larly, it is so rare at St. Andrews that it finds no place in Doctor McMurrich's local plankton lists, and this is true, to a less extent, off western Nova Scotia as well, judging from its irregular occurrence on German Bank. Euthemisto is usually only a minor factor in the plankton of the inner parts of the gulf. This rule has its exceptions, however, for we encountered swarms of its larvae off Penobscot Bay on August 11, 1913 (station 10090), and of adults as well as young in the deep basin farther east (station 10092), while it was so plentiful in the western basin on August 31, 1915 (station 10307), that the haul from 40 meters yielded about 200 cubic centimeters of adults and multitudes of newly-hatched larvae. We have usually found Euthemisto an important element in the tow nettings at the mouth of the gulf and over the outer part of the continental shelf generally from off Halifax to abreast of New York. For example, E. compressa abounded on the south side of Nantucket Shoals on July 9, 1913 (station 10060), while young bispinosa swarmed in the water southwest of Nantucket on August 22 of that same year (station 10112). We took about 1,000 cubic centimeters of medium-sized Euthemisto in a half hour's tow at 40 meters near Cape Sable on August 11, 1914 (station 10243), an equal volume of large specimens in a surface haul of the same duration with a net 1 meter in diameter on Browns Bank, July 24. 1914 (station 10228), and 750 cubic centimeters on the surface off Shelburne, Nova Scotia, three days later (station 10231). Euthemisto "again formed a considerable part of our catches on the shelf south of Nova Scotia (stations 10291 to 10294), on Browns Bank (station 10296), and off Marthas Vineyard (stations 10332 and 10333) in *> For the actual details of capture I refer the reader to Hansen (1915) and Bigelow (1917). PLANKTON OF THE GULF OF MAINE 157 the summer of 1915" (Bigelow, 1917, p. 286), as well as over the southwest part of Georges Bank in July, 1916 (stations 10351 and 10353), which substantiates the tow nettings made by vessels of the Bureau of Fisheries in past years. >r Git And V /v. ® tr 1 NOVA •-' + • 4- © j ( SCOTIA 1 + (Yarmouth N® 9 / J i .......y • o %■■■ ® > ®" ,.,-®' © © + © Flo. 55. — Occurrence of the ampnipod genus Euthemisto, July, August, and the first week of September. •, locality records for E. compressa; O, locality records for E. bispinosa; Q, locality records for both species together. The large symbols are for the more notable swarms This zone of abundance can hardly extend out beyond the continental edge, for, generally speaking, we have found Euthemisto decidedly less common over the continental slope and rare at the deep stations where the plankton is characterized 158 BULLETIN OF THE BUREAU OF FISHEBIES by a large tropical element (e. g., station 10218, July 21, 1914). Thus its abundance along the outer edge of the shelf does not imply an oceanic origin, but, hke Calanus, it is typical of the water of the coastal banks off the Gulf of Maine and along the American litoral as a whole, finding the inner edge of the so-called Gulf Stream a fluctuating barrier to its seaward dispersal, which is in line with its boreal nature. Euthemisto is not only more numerous over the outer part of the shelf than within the Gulf of Maine, but it grows larger there, although very large specimens occasionally occur even close to land. When adult females with eggs are taken in our coastwise hauls they are seldom over 10 millimeters long, with the general run of the catch still smaller, whereas the numerous adults taken over the offshore banks are often as long as 20 millimeters. Although we know little of the status of Euthemisto in the offshore parts of the gulf in autumn, there can be little doubt that an inshore movement of greater or less extent takes place at that time, for in 1915 this genus occurred in some numbers in October in Massachusetts Bay, where it is usually scarce or absent in summer (p. 156). Apparently it reaches its maximum abundance in the coastal zone of the gulf in October and November, and during the third week of November in 1912 it was comparatively common near Gloucester (Bigelow, 1914a, p. 403). To judge from the season of 1920 and 1921, however, this autumnal increase is followed by shrinkage in its numbers with the onset of winter, for in late December and early January we took Euthemisto at only 5 out of 14 stations in the northern and western parts of the gulf — never more than a few specimens in any haul — nor did it appear in any abundance later than November during the winter of 1912-1913, though a few were noted at all our stations until February. In February and March, 1920 (fig. 56), Euthemisto was as generally distributed over the gidf and over Georges and Browns Banks, as it is in summer (fig. 55) ; but it was far less numerous, for it appeared at only about half the February and March stations (occasional examples only) , the only exception to this rule being the waters off south- ern Nova Scotia (not strictly within our limits) , where it was taken in some numbers on two occasions (stations 20074 and 20075). Its numbers in the gulf fell to an even lower ebb in April, when we detected it (in very small numbers) at only 6 out of 30 stations, a shrinkage due to an actual decrease in the stock and not to an emmigra- tion out of the gulf, for, as it happens, these few records were near Cape Elizabeth, on the one hand, and off the western shores of Nova Scotia, on the other, with no Euthemisto whatever taken at our stations farther out at sea during the month. In 1920 none were detected in the western side of the gulf in May (stations 20120 to 20126), though a few (both bispinosa and compressa) were taken off the seaward slope of Georges Bank on the 17th (station 20129), in a haul from 100-0 meters; but in 1915 (which was also an earlier season in other respects) a scattering of Euthemisto was noted at most of the May and June stations at the mouth of Mas- sachusetts Bay, in the gulf generally outside the 100-meter contour, off Lurcher Shoal, on German and Browns Banks, and over the outer part of the continental shelf outside the continental edge off Shelburne, Nova Scotia. 86 During these months "« Recorded in my field notes from stations 10269, 10270, 10272, 10273, 10278, 10279, 10281, 10282, 10284, 10288, 10290, 10291, 10293, 10294, 10295, and 10296. PLANKTON OF THE GULF OF MAINE 159 it was noted at only one of the stations (10287) inside the 100-meter contour along the eastern coast of Maine. ../ © Fig. 56.— Occurrence of the amphipod genus Euthemisto from February to April, 1920. © locality records for Euthemisto comprasa: Q, locality records for E. compressa and E. bispinosa; O, stations where neither occurred; X, locality records for larva? too young for identification as the one species or the other. Euthemisto thus exhibits a more or less definite summer and early autumn maximum contrasted with an early spring minimum in the Gulf of Maine, disappear- ing from the coastal zone, as its numbers dwindle in late winter or early spring, to 75S9S— 26 11 160 BULLETIN OF THE BUREAU OF FISHERIES reappear there in October and later. This seasonal cycle is just the reverse of what obtains in the North Sea region, where Euthemisto compressa occurs commonly in winter with the indraught of Atlantic water (Tesch. 1911), but only in small numbers at other seasons. The presence of adults with eggs, of larvae, and of immature specimens at various stages in development shows that Euthemisto 87 breeds successfully over the entire area of the Gulf of Maine outside the outer islands and headlands — perhaps even in Massachusetts Bay. Large numbers of young are sometimes produced in the inner parts of the gulf — for instance, the swarms of young off Penobscot Bay in August, 1913, mentioned above (p. 20) — as well as in the surface waters of the western basin, where newly hatched as well as medium-sized Euthemisto were plentiful on August 31, 1915 (station 10307). The chief breeding areas, as indicated by relative abun- dance, lie over the outer edge of the continental shelf, extending as far west at least as longitude 71°, where we found shoals of young specimens as well as of adults late in August in 1913 (Bigelow, 1915, p. 281); likewise on the central, northwestern, and southwestern parts of Georges Bank, on Browns Bank, and in the coastal waters off Cape Sable. In this general zone we have not only found breeding adults as well as young on many occasions, but more than once have taken young in abundance on the surface and adults with eggs in the deeper hauls (p. 163). The breeding season of Euthemisto certainly extends over a large part of the year, for we have found its larvae in every month from February until October. Probably it also breeds during the late autumn, when we have not visited its chief offshore areas of reproduction, for occasional young specimens appeared in our tows near the Isles of Shoals and off Cape Cod in the first week in November, 1916 (stations 10400 and 10403), and in the deep near Cape Ann late in December, 1920 (station 103S9) ; but young are produced in greatest number in June, July, and August. No attempt has yet been made to estimate the actual numerical strength of Euthemisto in the Gulf of Maine, but at times the local population must be con- siderable to yield the abundant tow-net catches mentioned above (p. 156). In the preceding lines the genus has been treated as a unit. The relative fluctuations of its two local representatives, the species compressa and bispinosa,** are next to be considered. Although these two species of Euthemisto are often taken side by side, they occupy somewhat different faunal niches, with bispinosa the more oceanic of the two and showing a more definite seasonal movement toward and away from the coast than compressa does. 89 During the period February to May, when the genus as a whole is at a low ebb in the Gulf, compressa is decidedly the commoner member of the pair in its inner waters, while on Georges Bank and south of Nova Scotia the two occur in roughly equal numbers at that season (at least such was the case in 1920). In June, when the numbers of the genus as a whole increase, compressa still predominates within the gulf, but we found bispinosa 87 Both E. compressa and E. bispinosa. t» For descriptions and the distinguishing features of these two see Sars, 1895. I have elsewhere given tables of the relative abundance of the two for several of our cruises (Bigelow, 1914a, p. 4; 1915, p. 279; 1917, p. 287; 1922, pp. 133 and 148). •• For tables of the relative abundance of the two species of Euthemisto from 1913 to 1915 see Bigelow, 1915, p. 282, and Bigelow 1917, pp. 287 and 288. PLANKTON OF THE GULF OF MAINE 161 outnumbering it off Shelburne (station 10294) and on Browns Bank (station 10296) during that month in 1915. Station Species present Station Species present 20044 Compressa. Do. Compressa and bispinosa. Juveniles. Compressa and bispinosa. Compressa. Compressa and bispinosa. Do. Juveniles. Compressa and bispinosa. Juveniles. Compressa. 20074 Compressa and bispinosa. 20045 20075 Do. 20046 20077 20050 20079 Do. 20052 20087 .. Do. 20055 20095... Do. 20057 20102 20065 20104 Do. 20067... 20112 20113 Compressa. 20068 20071 20114.. 20072 20129 With the advance of summer the ratio of bispinosa to compressa increases. Thus, in July, 1914, bispinosa outnumbered the latter on the southern part of Georges Bank (stations 10216 and 10223) and on Browns Bank (station 10228) and about equalled it on the northwest part of Georges Bank (station 10215) and in the eastern channel (station 10227) ; but compressa was still the dominant member of the pair off Massachusetts Bay (station 10213), in the southeastern part of the basin of the gulf (station 10225), over the northeastern edge of Georges Bank (station 10226), along the continental edge off the southeast and southwest slopes of Georges Bank (stations 10220 and 10218), and abreast of Shelburne, Nova Scotia (station 10233). In August of that year bispinosa was the dominant member of the pair near Cape Sable (station 10243) and in the eastern side of the basin (stations 10245 and 10249). The two species were about equal off Mount Desert and Penobscot Bay (stations 10248 and 10250). In the deep water off Cape Ann (station 10254) com- pressa was the more numerous at the surface, but bispmosa predominated in the haul from 225-0 meters. Compressa still dominated at the mouth of Massachusetts Bay and in the south central parts of the basin (stations 10253, 10255, and 10256), but bispinosa was much the more numerous of the two at two stations on the conti- nental shelf off Marthas Vineyard at this time (stations 10258 and 10259), and while it dominated at one station at the continental edge (station 10260), compressa out- numbered it at another station a few miles farther out (station 10261). Bispinosa is not so important, relatively, in the inner parts of the gulf every summer, for in 1913 compressa outnumbered it at all the August stations east of Cape Cod and north of Georges Bank, though bispinosa was more plentiful then than it had been a month previous (we have no autumn records for that year in the gulf), and with the same center of abundance as in 1914 — that is, the central and eastern parts of the deep basin. Bispinosa outnumbered compressa in Massa- chusetts Bay, off Cape Cod, and locally south of Marthas Vineyard in October, 1915 (stations 10258 to 10267); and in the first week of November, 1916, it again predominated off Cape Cod (station 10404) but was detected at only two of five stations farther north in the guff at this time, whereas compressa was at all of them. Compressa was also the only Euthemisto noted close to land near Marthas Vineyard 162 BULLETIN OF THE BUREAU OF FISHERIES on November 10 (station 10405), but farther out on the continental shelf on this line bispinosa predominated in the rich catches of these amphipods (stations 10406 and 10407). In Massachusetts Bay, which may be taken as fairly representative of the western coastal waters of the gulf, E. bispinosa attains its greatest numerical strength, com- pared to E. compressa, during late autumn or early winter, dwindling rapidly there- after, as appears from the following table of the relative abundance of the two species in samples of the catches made off Gloucester during the winter of 1912-1913. Station Date Com- pressa Bispi- nosa Station Date Com- pressa Bispi- nosa 10047 _ Nov. 20, 1912 Dec. 4, 1912 Dec. 23, 1912 Jaa. 16,1913 20 15 15 30 12 25 12 2 10051 Jan. 30,1913 do. 4 25 30 20 10048 10052 3 10049 . 10053 Feb. 13, 1913 Mar. 4.1913 5 10050 10054 Although it is not yet possible to outline the relationship of the two species more in detail, it is safe to say that E. compressa is a permanent and characteristic inhabitant of all parts of the Gulf of Maine except the immediate coastal zone, occurring there wherever the genus is known at all, and at all seasons. E. bispinosa is to be found over the outer parts of the continental shelf throughout the year, but it is only a seasonal visitor to the inner parts of the gulf, spreading first into its eastern half in summer. By autumn and early winter it may rival compressa locally right up to the western and northern shores of the gulf, but in the western coastal zone it is usually outnumbered by the latter even at that season, and either perishes or withdraws seaward once more with the advance of winter. Thus, E. bispinosa is decidedly more oceanic than E. compressa, as it occurs in the inner parts of the gulf, which corresponds to the fact that it usually equals or predominates over the latter in the coast waters south of Nova Scotia, over the whole southern part of Georges Bank, and in the shallow waters south of Marthas Vineyard and Nantucket. It is also more oceanic than compressa on the European side of the Atlantic, seldom appearing within the North Sea, but regularly present off the west coast of Ireland (Tesch, 1911; Tattersall, 1911), well out from the west coast of France, at least in autumn (Le Danois, 1921), and in the colder waters of the Norwegian and Arctic Seas. But with the two species in roughly equal numbers in the rather scant catches outside the continental edge, or with compressa and not bispinosa predominating there (sometimes, in fact, the only member of the pair represented, as at station 20064 on March 11, 1920), the relative status of the two species off the North American littoral can not be established without further study. As a general rule, when bispinosa outnumbers compressa its preponderance is greatest in the deep hauls, whether in the gulf, over the banks, or west and south of Cape Cod. The adult Euthemisto are not characteristic of any precise depth level in the water, as is the large copepod Euchseta norvegica, for example (p. 29), but occur at all depths from the surface down to the deepest strata of the Gulf of Maine. PLANKTON OF THE GULF OF MAINE 163 Large ones, however, especially the females with eggs, have rarely been taken in our surface nets; and even medium-sized individuals have usually been but sparsely represented in the surface hauls, although we have occasionally met exceptions to this rule, notably in the northeastern part of the gulf during August in 1912 and 1913 (stations 10032 and 10096) and off Marthas Vineyard on July 10, 1913 (station 10002). On the other hand, E. compressa, like Calanus, has usually proved more abundant above than below 100 meters depth whenever two or more subsurface hauls have been made at different levels. The bathymetric distribution of the larvae of Euthemisto differs from that of the adults, for they are usually most numerous at or close to the surface. The fact that we have taken them in swarms in the surface nets at several stations where their parents (or at least females with eggs) were plentiful at deeper levels is evidence that they rise through the water immediately after they are hatched — one of the innumerable provisions of nature for the perpetuation of the species, for otherwise they would inevitably be devoured by their own voracious progenitors (p. 107). Examples of a bathymetric stratification of this sort as between adults and larvae were noted in the eastern part of the gulf (stations 10092 and 10093) and off Marthas Vineyard (station 10112) in August, 1913; over Georges Bank in July, 1914 (sta- tions 10215 and 10219); off Shelburne in June; in the western basin in August, 1915 (stations 10293 and 10307); and off Marthas Vineyard in July, 1916 (station 10353). Both species of Euthemisto — compressa and bispinosa — like Calanus finmarchicus and Sagitta elegans, tolerate very wide fluctuations of temperature and salinity, as, indeed, they do in European waters as well (Tesch, 1911). So far as actual occur- rence goes, we have taken them over the whole range of temperature prevailing within the limits of the gulf, from the icy waters of winter and of the Nova Scotian current, on the one hand, to the summer-heated surface of the western basin and the warm waters along the outer edge of the offshore banks, on the other; likewise over the entire range of salinity proper to the open waters of the gulf, except for the very lowest. It is not possible to draw any close parallel between the abund- ance (or reverse) of Euthemisto and the temperature from the data so far obtained, but we have never found it abundant in the coldest season, and most of the rich catches have been made in temperatures warmer than 5°, as appears from the follow- ing list of the readings at and above the levels at which the horizontal parts of the hauls were made, at several stations productive in large Euthemisto. General locality Station Date Depth in meters Temper- ature in degrees 10092 10307 10229 10243 10296 10228 10228 10216 10219 10258 10351 10231 Aug. 11,1913 Aug. 31, 1915 July 25, 1914 Aug. 11, 1914 June 24, 1915 July 24,1914 do July 20, 1914 July 21,1914 Aug. 25, 1914 July 24, 1916 July 27,1914 170 40 80 40 50 (■) 60 50 40 25 160 0) 5+ 7-8+ Off Cape Sable 5-6+ Do.... 7.5+ 3+ Do. 14.72 Do 8.3+ 12+ Do 13+ 12+ Do _ 4.8+ 6.62 ' Surface. 164 BULLETIN OF THE BUREAU OF FISHERIES The last of these records is especially instructive, because there were very few, if any, Euthemisto in the icy water below the surface at that station. The autumnal augmentation of the stock of Euthemisto in the coastal belt of the gulf likewise takes place in comparatively high temperatures (e. g., 7 to 11° on October 26 and 27, 1915, in Massachusetts Bay, stations 10337 to 10339), and our largest November catch was on the surface in water of about 10.3° (station 10404). Thus, whether or not the relation be a causal one (and this is not safe to postulate, in view of the wide distribution of Euthemisto in northern seas), the maximum abundance of Euthemisto in the Gulf of Maine coincides with rather high temperature, both in season and in the depth at which it congregates, corroborating Le Danois's (1921) observation that off the French coast E. bispinosa is common only in water as warm as 14°. The adults, however, whether of compressa or of bispinosa, certainly show no tendency to accumulate in the warmest waters of the gulf, which they could easily reach by swimming upward for a few meters. On the contrary, when they have been found in any number on the surface it has been at times and places where the water was at least no warmer than 15°. Only once have we found large Euthe- misto in any number at a temperature higher than 14°. For the adult, then, the optimum range of temperature in the Gulf of Maine is from 4° to about 12°. We have no evidence that any considerable reproduction of Euthemisto takes place in the gulf in temperatures lower than 5° or higher than 12 to 14°, but the fact that we towed occasional very small specimens in February, March, and April, 1920, both off Massachusetts Bay, in the western basin, near Cape Sable, on Browns Bank, and on the southwest part of Georges Bank (stations 20045, 20048, 20050, 20072, and 20104), proves that a certain amount of breeding takes place in water as cold as 2 to 3°. The larvse, however, are most often abun- dant in considerably warmer water, thanks to the fact that summer is the chief breeding season, and to their habit of rising to the surface. Here, again, we hesitate to assume any causal connection between temperature and the depth which they seek, it being as likely that their tendency to congregate at the warmest level is due to some quite different cause; such, for example, as the available supply of food, the density of the water, or the influence of sunlight. Within the Gulf of Maine Euthemisto is usually most numerous in compara- tively high salinities, say, upwards of 32.5, per mile, and while we have made very rich catches in water as little saline as 31.6 per mille along the Nova Scotia coast, this is the lowest salinity in which we have found it in any numbers. Hence, 31.5 per mile may be set arbitrarily as the lower limit to its common occurrence in the Gulf of Maine. When the superficial layers of the coastal zone of the gulf are fresher than this — that is, throughout the period of spring freshets and in early summer — Euthe- misto is usually rare there, if not absent ; but it would be no surprise to meet excep- tions to this rule, for Euthemisto has been found swarming off the English coast in water of only 30.26 per mille (Tesch, 1911). It is questionable whether high salinities ever act as a barrier to the migrations of Euthemisto in the one direction as low salinities do in the other. It certainly occurs regularly in water as saline as 35 per mille in the eastern North Atlantic, and while it is not a characteristic inhabitant of Salter seas (the highest salinity we have actually found it in was about 35.2 per mille (Bigelow, 1915, p. 283) ) it is PLANKTON OF THE GULF OF MAINE 165 more likely that constantly high temperature, not high salinity, is its outer barrier off eastern North America, and bars it from the warmer parts of the Atlantic in general. Within these wide limits, however, Euthemisto is very tolerant of varying salinity, both in the western Atlantic and in the eastern. At times and places where Euthemisto is abundant it probably serves as a valu- able food for pelagic fishes in the Gulf of Maine, though little information is avail- able. In Irish seas Tattersall (1906) found it forming a very large part of the food of two of the principal food fishes — herring and mackerel — as well as of the sea trout, while at times it forms the chief sustenance of the long-finned tuna {Germo alalunga) off the French coast (Le Danois, 1921). Euthemisto, in its own turn, is extremely destructive to copepods and to other small planktonic animals (p. 107). Before closing the brief account of this genus, I must emphasize our failure to find even a single specimen of the arctic Euthemisto (E. libeUula) witliin the limits of the Gulf of Maine. Certainly it does not reach it unless as the rarest of stragglers. »&* Other hyperhds The two species of Euthemisto are the only hyperiids that are of any numerical importance in the plankton of the Gulf of Maine. Their relatives, Hyperoche and Hyperia (similarly boreal in faunistic status), have been taken at several stations but always in small numbers. Hyperia Hyperia is represented locally by two species — galba and medusarum — both of which usually live commensal with the large medusae Aurelia or Cyanea. This is not invariably the case, however, for Hyperia has repeatedly appeared in the catches of the tow nets at stations where no medusae were taken or seen — for example, on German Bank, August 14, 1912 (Bigelow, 1914, p. 103). Associated with their occasional independence of the medusae we have found one or other species of the genus widely distributed in the northern half of the gulf, over deep water as well as shallow, but our nets have never yielded more than four or five specimens of Hyperia at any one station. Hyperia medusarum has been taken both in summer and in winter, but H. galba has so far been taken only in July and August. In the case of animals as comparatively scarce as Hyperia is in the Gulf of Maine, captures in tow nets are so largely a matter of accident that they do not give a reliable picture of the numerical strength of the species in question from season to season and from place to place. It seems, however, that Hyperia was decidedly more numerous in 1913, when we found it at some half dozen stations in the gulf (Bigelow, 1915, p. 279), than in the summer of 1914, when it was not found at all at the same localities and season (Bigelow, 1917, p. 289), or in 1915, when only odd individuals were taken during the summer. Hyperoche Hyperoche tauriformis 90 has appeared rather more commonly in our tow net- tings than has either species of Hyperia, having been taken at 10 stations in the "In an earlier report (Bigelow, 1915) this amphipod appears as " H. kroyeri Bovallius," but recent students of the group— e. g. Tesch, (1911) and Tattersall (1906)— agree that while it has passed most often as "kroyeri" or as "abyssorum" Boeck, its cor- rect designation is "H. tauriformis" Bate and Wcstwood. This name is accepted here for the sake of uniformity, the question not being of specific identity but simply of the distribution of the only species of Hyperoche known to exist in northern seas. 166 BULLETIN OF THE BUREAU OF FISHERIES gulf during August, 1913 (Bigelow, 1915, p. 279). Like Hyperia, it was far less com- mon in 1914, when we took it only once within the gulf limits and occasionally off the Nova Scotian coast east of Shelburne (Bigelow, 1917, p. 289); in 1915 it was taken at several stations, but never more than one or two specimens at any. Judg- ing from the regularity with which it appeared in Massachusetts Bay during the winter of 1912-1913 (.Bigelow, 1914a, p. 410; six out of nine stations, but only one or two examples on each occasion), Hyperoche is at least as common during the period from November to February as during the warm months; but it has not been detected at all at any of the stations occupied in late February, March, April, or May, suggesting that it becomes very rare in the gulf, if it does not entirely vanish thence, when the water is at its coldest for the year. Our captures of Hyperoche in the Gulf have all been near shore, for the most part within the 100-meter contour (Bigelow, 1915, p. 284), but the numbers of specimens concerned are too small to throw any light on its bathymetric distribu- tion or on the relationship which its occurrence bears to the physical state of the waters of the gulf. Parathemisto oblivia Parathemisto oblivia has been detected twice in our hauls in the open gulf (sta- tions 10032 and 10036, August 16 and 20, 1912) and at three stations off the outer coast of Nova Scotia (Bigelow, 1917, p. 289), all in late summer. Doctor Huntsman informs me that it breeds locally under estuarine conditions in the Bay of Fundy also. This amphipod is far more abundant in North European waters, where it plays much the same role as does Euthemisto in our gulf and sometimes occurs in shoals right up to the land (Edward, 1868; Tattersall, 1906; Tesch, 1911). Oceanic hyperiids Our stations along the continental slope have occasionally yielded oceanic and warm-water hyperiids in some numbers, but it is only on the rarest occasions that any of them encroach more than a few miles on to the shelf within the limits of the gulf, nor are any of them known from within Georges and Browns Banks (p. 56). For the sake of completeness, such records as have been obtained within the geo- graphic limits of the present study since 1912 are listed below 91 (for earlier records for New England waters, see Holmes, 1905). Date and stations Species July, 1913. • 10061 July and August, 1914 ' June to Au- gust, 1915 February to May, 1920 10218 10219 10220 10260 10261 10296 10333 20044 20045 20076 20129 3 4 X "~x" 3 2 1 Phronimasp Phrosina semilunata 1 1 X X X X X X 1 • For records between the latitudes of New York and Chesapeake Bay during that summer see Bigelow, 1915, p. 279. t Previously listed in Bigelow, 1917, p. 289. » For descriptions and an account oi the general distribution of these hyperiids on the high ;eas see Bovalhus, 188, to 1899. PLANKTON OP THE GULF OF MAINE 167 The distribution of these and of other warm-water planktonic animals is dis- cussed in a preceding chapter (p. 53). Copepods Except in certain restricted localities, or for brief periods when some other animal swarms, the animal plankton of the Gulf of Maine consists chiefly of copepods at all seasons. The seasonal fluctuations of the group as a whole are touched on above. The following chapter gives brief discussions of most of the species so far detected in the plankton of the open gulf or at St. Andrews (Doctor McMurrich's lists, p. 12). The great majority are forms that are not only typically pelagic but widespread in northern seas; but at St. Andrews, where strong tides stir the water from bottom to top, sundry dwellers in the littoral zone are brought up to or near the surface, and probably this takes place more or less in estuarine situations all around the shore line of the gulf. Samples of the copepods collected in 1912, 1913, and 1914 were identified by Dr. C. O. Esterly, and lists for those years have been published elsewhere (Bigelow, 1914, p. 115; 1914a, p. 409; 1915, p. 287; 1917, p. 290). It is not necessary to repeat them here. Only a preliminary survey has been made of the copepods towed by the Grampus in 1916 (Bigelow, 1922), but Dr. C. B. Wilson has supplied lists for the vertical hauls made in 1915 and the spring of 1920 and for the horizontals for the winter of 1920-21, which are tabulated below (p. 297). Doctor McMurrich's manuscript lists of plankton for St. Andrews, New Brunswick, have been especially instructive for the seasonal periodicity of the copepods. Previous to the inception of the Grampus cruises in 1912, almost no attention had been paid to the copepods of the Gulf of Maine, the only published data for that precise region being a few notes on species from Plymouth Harbor, Mass. (Wheeler, 1901). Subsequently Willey (1919, 1920, and 1921) has given some notes on the copepods of the St. Andrews region in the Bay of Fundy. The Copepoda of southern New England have been studied by Wheeler (1901), Williams (1906 and 1907), Sharpe (1911), and Fish (1925) ; those of the outer coasts of Nova Scotia and of the Gulf of St. Lawrence by Herdman, Thompson, and Scott (1S98), by T. Scott (1905), and by Willey (1919), whose lists of the species collected by the Cana- dian fisheries expedition of 1915 are referred to repeatedly in the following accounts of the several species. All living copepods are small — the largest up to 10 to 11 millimeters, the smallest less than 1 millimeter in length. The commonest Gulf of Maine species (Calanus finmarchicus) is about 2 to 5 millimeters long when adult. They are present in such immense numbers in the plankton, and they reproduce so rapidly, that they are the most important of all pelagic invertebrates from the economic viewpoint, furnishing the primary food for the young of most marine fishes until these attain considerable size, as well as for many of the larger planktonic animals of various groups. Copepods are the major article in the diet of the adults of such plankton-feeding species as the mackerel and all the herring tribe. This aspect of copepod economy is touched on in another chapter (p. 97). I need only emphasize here that evidence is con- stantly accumulating to prove that the fertility of any part of the northern seas in 168 BULLETIN OF THE BUREAU OF FISHERIES commercial fishes depends very largely on the stock of copepods. As Dr. C. B. Wilson writes, it is not too much to say that "their presence and abundance count as much for the higher animal life in the ocean as does that of nitrates in the soil or carbon dioxide in the air for plant life upon the land," for they are the chief intermediary through which the elemental foodstuffs elaborated by the marine plants on which the copepods feed are made available for the support of the larger marine animals that feed on them. Copepods are the only animal group that has been systematically counted in the catches of the vertical nets in the Gulf of Maine; and while the numerical calculations include so many indeterminate sources of error that they can be taken only in a general way, they have proved undeniably instructive in tracing the seasonal perio- dicity and relative regional abundance of several of the more common species. I must emphasize, however, that the counts given are only a rough indication of the relative abundance or scarcity of the several species, and that the "probable error" (unknown) may amount to as much as 80 to 100 per cent in extreme cases. (For a discussion of the allowance that must be made on this account see Johnstone, Scott, and Chadwick, 1924, p. 180.) For the group as a whole the numbers present per square meter have varied from next to none at occasional stations in the coastwise zone during the early spring, when diatoms are flowering and copepods are scarcest (p. 39), to upwards of 500,000 in May, when Calanus finmarchicus is swarming (e. g., station 10266, May 4, 1915). Copepods are at their lowest ebb in the gulf in February and March, when the maxi- mum per square meter at any station within the edge of the continent in 1920 was 37,500 (station 20049, in the western basin), the minimum 55, in the inner part of Massachusetts Bay, and the average about 6,600. Generally speaking, at this season there are more copepods under any given area of the sea surface in the deeper parts of the gulf than in the shoal, the numbers caught being roughly proportional to the amount of water strained by the net in its journey from the bottom up to the surface. Thanks to a swarm of Calanus (p. 189), there were more copepods outside the south eastern edge of Georges Bank than anywhere within the gulf. In April, 1920, the average within the continental waters of the gulf was about twice as large (13,300) as it had been in March, the maximum more than three times (130,000 in the northern channel), and the minimum had risen from 55 to 900. In another chapter (p. 41) I have commented on the tremendous augmentation of copepods which takes place in May and for which the vernal wave of reproduction of Calanus finmarchicus is chiefly responsible. In 1920 this was hardly under way by the middle of the month, but in 1915 it had raised the average number of copepods over the inner parts of the gulf to upwards of 140,000 by the 4th to the 14th (stations 10266 to 10278), with maxima of 511,000 off Cape Ann on the 4th and 411,500 in the eastern side of the basin on the 6 th. Fewer copepods were taken in June, the average being only about 23,000 per square meter. The fact that the vernal reproductive activity commences later in the northeastern and eastern shallows of the gulf, where most of the June stations were located, than in its western side is chiefly responsible for this apparent shrinkage; but with only about one-seventh as many copepods in the eastern basin on June 19, PLANKTON OF THE GULF OF MAINE 169 1915 (station 10288) as at a near-by location (station 10270) on May 6, it seems that the swarm resulting from this local center of active reproduction had dispersed in the interim. Unfortunately no vertical hauls were made later than June in the summer of 1915, but in Jul} 7 and August, 1914, the average number of copepods per square meter for the gulf, as a whole, inside the continental edge but including the offshore banks, was between 72,000 and 73,000 (see Bigelow, 1917, p. 315, for table of counts) — i. e., something less than half the May average for 1915, with a maximum of 227,000 in the northern channel and a minimum of 6,000 on the northern edge of Georges Bank at this time. Copepods were then most numerous per square meter (70,000 + ) in four distinct regions as follows: (1) Over a V-shaped area, with one arm extending from Cape Cod Fig. 57. — Number of copepods per square meter of sea area, July and August, 1914, as calculated from the catches of the vertical hauls. 1, scanty (less than 20,000); 2, intermediate (20,000 to 70,000); 4, rich (70,000 to 150,000); 6, very rich (150,000 or more). Reproduced from Bigelow, 1917, fig. 94. toward Penobscot Bay, the other to the eastern part of Georges Bank; (2) off Cape Sable; (3) in the extreme northeast corner of the basin of the gulf; and (4) south of Marthas Vineyard (fig. 57). The maxima were off Cape Cod, off Cape Sable, and in the northern channel (stations 10213, 10243, and 10229; Bigelow, 1917, p. 316). On the other hand, we have found very few copepods in the coastal zone in the ex- treme northeast corner of the gulf, in the southeastern part of the basin, in the eastern channel, or in the oceanic water outside the edge of the continent during the summer. The distribution of copepods on the basis of numbers per cubic meter has paralleled this, except that the region northeast of Cape Cod was shown to be relatively less productive by this than by the other calculation in July, 1914. The numbers per 170 BULLETIN OF THE BUREAU OF FISHERIES square and cubic meter for that summer and for the season of 1915 are tabulated in an earlier report (Bigelow, 1917, pp. 315 and 319). September stations for 1915 yielded an average of about 65,000 copepods per square meter in the northern half of the gulf — no noticeable change, that is, from the midsummer state — but the fact that the maximum (173,000) was considerably less and the minimum (14,700) con- siderably greater is interesting as evidence that copepods tend to become progres- sively more and more nearly equalized in number over the gulf as the season advances. In the earlier chapter I have pointed out that we have observed an autumnal increase in the amount of plank