Short title PLANKTONIC COPEPODS OF THE CANADIAN BASIN, ARCTIC OCEAN. MSc the sis Elizabeth R. Bulleid Marine Sciences Centre March, 1972

Transcription

1 Short title PLANKTONIC COPEPODS OF THE CANADIAN BASIN, ARCTIC OCEAN MSc the sis Elizabeth R. Bulleid Marine Sciences Centre March, 1972

2 ABS TRAC T SOME ASPECTS OF THE BIOLOGY AND DISTRIBUTION OF THE PLANKTONIC COPEPODS OF THE CANADIAN BASIN IN THE ARCTIC OCEAN Elizabeth R. Bulleid March 1972 MSc Thesis Marine Sciences Centre Fifty plankton sampl~3 were taken in the upper 300 m of the Beaufort Sea in 1966, from which thirty-nine species of planktonic copepods are identified. Their depth distributions within the Arctic layer are analysed; one group is shown to be characteristic of the upper 200 m, replaced by another group below this depth. Twelve species have bipolar distributions; fifteen are circ~polar. It is suggested that sorne have failed to colonise the North Pacifie owing to strong competition from congeneric species south of Bering Strait. The reproductive cycles and depth distributions of the copepodite stages of four species are analysed; two Calanus species have multi-year cycles and may have developed a continuous low reproductive rate which compensates for mortality when intensive breeding is not possible for long periods. Metridia longa and Euchaeta glacialis have annual cycles and breed independently of the phytoplankton bloom.

3 '- SŒ'lE ASPECTS OF THE EIOLOGY AND DISTRIBUTION OF THE PLANKTONIC COPEPODS OF THE CANADIAN BASIN IN THE ARCTIC OCEAN. by Elizabeth R. Bulleid A thesis submitted ta the Faculty of Graduate Studies and Research in partial fulfillment of the re~uirements for the degree of Master of Science. March 1972 Marine Sciences Centre McGill University Montreal Elizabeth R. Bul1eid 1972

4 - i - t TABLE OF CONTENTS ACKNOWLEDGEMENTS ii LIST OF FIGURES iii LIST OF TABLES iv I. INTRODUCTION 1 Physical Oceanography 3 Sound Scattering Layers 8 II MATERIALS AND METRODS 12 III RESULTS AND DISCUSSION 15 Systematics 15 Reproduction Breeding Cycles Depth Distribution of Reproductive Stages 45 Depth Distributions 57 Zoogeography 68 IV SUMMARY 82 V BIBLIOGRAPRY 85 APPENDIX 91 J

5 1 - ii - ACKNOWLEDGEMENTS Thanks are due to Dr. M.J. Dunbar of the Marine Sciences Centre, McGill University for giving me the opportunity to work on this project, and for his support and helpful criticism. l am grateful to Dr. Carol Lalli for reading the final draft of this thesis and for her suggestions. Very special thanks must go to Mrs. D.C. Maclellan and W. Gordon Tidmarsh of the Marine Sciences Centre for their help in the identification of specimens, discussions of results and above ail, for their continued encouragement. This work was supported by research grants to Dr. Dunbar from the Office of Naval Research and the National Research Council of Canada.

6 '- - iii - LIST OF FIGURES 1 Curves of temperature, salinity and sigma-t Surface circulation of the Arctic Ocean Drift track of Fletcher's Ice Island 11 during 1966 sampling period. 4. Technique for multiple net horizontal tows Relative occurrence of the copepodite stages 33 of Calanus hyperboreus. 6. Relative occurrence of the copepodite stages 36 of Calanus glacialis. 7. Relative occurrence of the copepodite stages 40 of Euchaeta glacialis. 8. Relative occurrence of the copepodite stages 44 of Metridia longa. 9. Depth distribution of the copepodite stages 48 of Oalanus hyperboreus. 10. Depth distribution of the copepodi te stages 50 of Calanus glacialis. 11 Depth distribution of the copepodite stages 52 of Metridia longa. 12. Depth distribution of the copepodite stages 55 of Euchaeta glacialis.

7 .. i - iv - LIST OF TABLES 1. Relative occurrence of the copepodite stages 32 of Calanus hyperboreus. 2. Relative occurrence of the copepodite stages 35 of Calanus glacialis. 3. Relative occurrence of the copepodite stages 41 of Euchaeta glacialis. 4. Relative occurrence of the copepodite stages 43 of Metridia longa. 5. Depth distribution of the copepodite stages 47 of Calanus hyperboreus. 6. Depth distribution of the copepodite stages 49 of Calanus glacialis. 7. Depth distribution of the copepodite stages 51 of Metridia longa. 8. Depth distribution of the copepodite stages 54 of Euchaeta glacialis. 9. Number of copepods caught per hour of sampling 58 effort in the depth intervals sampled.' 10. Number of Microcalanus pygmaeus caught per hour 59 of sampling effort in the depth intervals sampled. 11. Number of copepods other than M. pygmaeus caught 59 per hour of sampling effort in the depth intervals sampled. 12. Geographie distribution of the species in this 81 collection.

8 ', I. INTRODUCTION Our knowledge of the oceanography and marine biology of the area of the high polar basin which is permanently icecovered was very meagre before Since then, submarines have been used for some research under the polar ice. It was and ice floe-s the introduction of the use of floating ice islandsaas platforms for oceanographic work that brought about the recent expansion of our understanding of the Canadian Basin. The first work on the plankton of the Arctic Ocean was the result of the voyage of the "Fram", under Nansen, in 1893 to 1896 (Sars, 1900). In 1931, the "Nautilus" became the first submarine to be used for research in the north; zooplankton collections were made north of Spitzbergen and were reported on by Farran (1936). The first major zooplankton collections in the Eurasian Basin were made by the Russian "Sedov" expedition in 1937 to 1939 (Bogorov, 1946). The U.S.S. "Burton Island" made surface plankton collections in the coastal and continental shelf areas of the Beaufort and Chukchi Seas in 1950 and 1951 (see Johnson, 1953, 1956; Rand and Kan, 1961), but it was not until the Russian ice station NP-2 was occupied during the same two years that the first year-iong study of the vertical distribution of zooplankton in the Arctic Ocean was made. This was also the first time that plankton had been collected from depths greater than 300m in the Arctic; until Brodskii and

9 , Nikitin (1955) reported on this work, only the plankton of the upper layer of the Arctic Ocean had been known. The submarines "Seadragon" and "Skate" made cruises in the Canadian Basin and employed automatic, multiple net samplers to collect plankton from the upper 200 m. The results of these cruises were reported by Grice (1962), Mohr and Geiger (1962), and Geiger (1966). In 1957 the American Drift Station Alpha repeated the type of survey made by the Russians on NP-2 (Johnson, 1963). Fletcher's Ice Island (T-3) was first occupied in 1952, and since then has provided a platform for oceanographic research. Marine biology in general (Mohr, 1959), the epipe;.agic amphipods (Barnard, 1959), and the pelagic polychaetes Knox (1959) have been studied on T-3. The results of comprehensive studies of the vertical distribution of plankton made from T-3 in the Canadian Basin have been published by Johnson (1963), Grainger (1965), and, most recently, by Harding (1966) and Dunbar and Harding(1968). In these latter two studies, the distribution of the plankton has, for the first time, been related to the hydrography of the area. Of the zooplankton collections made in the Arctic, only three have been used to study the vertical distribution of these organisms to depths of 2000 m or more in the Canadian Basin (Brodskii and Nikitin, 1955; Johnson, 1963; Harding, 1966)

10 - 3 - The collection dealt with here was made on T-3 during the summer of It is unusual in that it is one of the few collections made by horizontal towing below Arctic ice (Grice, 1962), and, for the first time in the Arctic, "multiple horizontal tows" which provide simultaneous samples of plankton from a variety of depths were used. Physical Oceanography A brief discussion of the hydrography and circulation of the water in the Canadian Basin follows. More detailed analyses of these aspects of Arctic oceanography are found in Coachman and Barnes (1961, 1962, 1963), but sorne basic description is necessary for interpretation of the biological results to follow. The three main water masses of the Arctic Ocean were described by Nansen (1902) at the beginning of this century as the Arctic Layer, the Atlantic Layer, and the Arctic Deep Water. A fourth water mass of Pacifie origin modifies this general three-layered system, especially in the Chukchi and Beaufort Seas (Coachman and Barnes, 1961). The Arctic Layer, with which this study is concerned, comprises the upper 200 to 300 m of the waters of the Canadian Basin. It can be divided into four distinct water masses. j

11 - 4 - These have been studied by Coachman and Barnes (1961, 1963) and their description of them was confirmed by the data taken on T-3 by Hansen (Hansen and Dunbar, 1970; Hansen, Bulleid and Dunbar, 1972). The Arctic Surface Layer occupies the upper 50 m, and is isothermal and isohaline. This is the coldest ( C) and the least saline ( %0) water in the depth range under consideration (0-300 m). At 50 m there is a sharp increase in both salinity and temperature which creates a strong density interface. The 50 to 200 m layer is known as the Arctic Intermediate Layer and, in the Beaufort Sea area, is modified by Pacifie water flowing through Bering Strait. A temperature maximum at 75 m identifies the upper half of the layer ( m) as Arctic water modified by water which entered the Arctic Ocean from the Bering Sea during the summer. Below this ( m), the temperature decreases to a minimum indicating that the denser winter water from the Pacifie is influencing the lower pa=t of the Arctic Intermediate Layer. The salinity increases throughout the whole Intermediate Layer to 33.2%0. From about 150 m, the temperature and salinity begin to increase more rapidly to maxima of OOC and 34.6%0 at 300 m (the deepest extent of this study). This is the area of mixing of the Arctic water above with the Atlantic water below. Dunbar and Harding (1968) have discussed the difficulty of

12 - 5 - defining an exact boundary between the Arctic and Atlantic Water Layers. It should be noted that the northern Chukchi and Beaufort Sea area is the furthest away in space and time from the point of entry of Atlantic water into the Arctic Basin between Spitzbergen and eastern Greenland. The water has lost many of its identifying characteristics of temperature and salinity (Coachman and Barnes, 1963), although a temperature maximum of C is observable at 500 m. The broad layer from 175 to 300 m, then, is an area of mixing of two water masses: the Arctic water mass above, and the weak Atlantic water mass which extends from about 300 to 900 m. Beneath the Atlantic Layer, from 900 m to the bottom, lies the Deep Water, with which this study is not concerned. Fig. 1 shows curves of temperature, salinity, and sigma-t for a typical station from T-3 in 1967, and indicates the stratification of the upper 300 m of the Canadian Basin. In general, the water of the Arctic Layer (0-300 m) moves around the Canadian Basin in a clockwise, slow moving gyre. Pacifie water entering the Arctic Intermediate Layer in the Chukchi Sea may flow west along the Siberian Continental Shelf with offshoots in the form of small counter-clockwise gyres. It may also flow from the Chukchi Sea across the central part of the Arctic Ocean via the North Pole. The se two flows join and leave the Arctic Ocean as part of the East Greenland Current. A small outflow from the Canadian Basin

13 Figure 1. Curves of temperature, sa1inityand sigma-t for a station in the Canadian Basin. T-3 Station 6-02; May 3, 1967 Cafter Dunbar and Hansen, 1970).

14 6 D E p T H M E T R E S '0 ' , ' UO HO \ \ -...,,,~ o c "" \., \ \ \ \ "\ "- ~ ----,- " \ \ \ S\ \ \ \, U.. f W 1 N 1 E } ARCTIC SURFACE WAHR PACIFIC MODIFIED INTERMfDIATE WAHR ATLANT IC WAHR 11ft , tri c.''''o:'''o:-~-...,.;t,.q;;----~'~'"''o;;-~--:.'':3."''0:-~_"""'''~1 -=,.:-- S %0

15 Figure 2. Diagrammatic map showing the general surface circulation of the Arctic Ocean Cafter Harding, 1966)..1

16 - 7 - «a «z «(.)

17 - 8 - gyre passes through the Canadian Archipelago. More detailed discussions of the circulation of the surface waters of the Arctic Ocean m~y be found in Gordienko (1961), Dunbar (1951), and Collin and Dunbar (1964). The simplified account given above is illustrated in Fig. 2. Sound Scattering Layers Two types of sound scattering layers have been found in the Canadian Basin as a result of work on T-3. One, the deep Scattering Layer (D.S.L.), was first detected by Hunkins (1965) using a 12-kHz recorder. It exhibited the same characteristics as D.S.L. 's in other parts of the world ocean, except that it. was shallower ( m) and seemed to migrate vertically in an annual, rather than a diurnal, pattern. This may be the result of the Arctic light regime since Hansen and Dunbar (1970) have shown that the organism responsible for the sound scattering in the D.S.L. is likely to be Arctogadus glacialis, the polar cod. The D.S.L. was first recorded during the summer months when the vertical migrations of the cod's zooplanktonic prey may be expected to be disrupted. In fact,hunkins (1965) has shown that the D.S.L. does undergo diurnal vertical migrations at the time of the autumnal equinox when the hours of light and dark are equally divided.

18 ' The second scattering layer was detected only with the more precise 100-kHz recorder (Hansen and Dunbar, 1970). As it coincides with the strong density interface at 50 m mentioned above, it is referred to as the Pycnocline Scattering Layer (P.S.L.). Hansen found large concentrations of shelled pteropods associated with this layer and has shown that Spiratella helicina is the most likely cause of the sound scattering (Hansen and Dunbar, 1970), but the reason for their accumulation at this intect'ace is not clear. The change in salinity is weil within the tolerance range of Spiratella (Harding, 1966). Harder (1968) postulates that a discontinuity may act as a trap for organic detritus and phytoplankton, and that zooplankton may aggregate at these interfaces for feeding. Hansen and Dunbar (1970) deny the validity of Harder's theory since Hunkins, Thorndike and Mathieu (1969) did not report any definite nepheloid (light scattering) layer at the same depth as the P.S.L. which would have indicated an accumulation of detritus. On reading the paper by Hunkins, Thorndike and Mathieu (1969), it becomes evident that there is a great deal of light scattering in the upper layer, although they do not define a distinct layer at 50 m. They point out that the nephelometer is very inaccurate and do not state whether a P.S.L. was in fact present when the measurements were made. It appears that this paper is not sufficient justification for criticism of Harder's ideas. They receive support from other

19 ,-.. i work done on T-3. Kinney, Loder and Groves (1971) studied the particulate and dissolved organic matter content of water samples taken from T-3 in 1968 and They found a distinct peak in the particulate matter concentration (55 pg/l) at the 50 m density interface. Particulate organic carbon and nitrogen concentrations were also shown to be highest at this depth. A much more detailed description of the sound scattering layers will be found in Hansen, Bulleid and Dunbar (1972). J

20 Figure 3. The drift track of Fletcher's Ice Island (T-3) during the 1966 sampling period (June 25 to August 28).

21 ,.. i ~ USSR ALASKA

22 ,-.i II. MATERIALS AN]) METHO])S Zooplankton collections were made from thirteen stations occupied by T-3 between June 25th and August 27th, ])uring that period, the ice island drifted along an erratic westerly track from a position at 'N, 'W to 'N, 'W, about 400 miles north of the Alaskan coast; the overall distance covered was very small (Fig. 3). Most of the zooplankton collections made at these thirteen stations were horizontal tows using the drift speed of the ice island attained during periods of high wind. At these times, the ice island drifted at speeds of up to 0.5 knots relative to the water beneath it. Since wind speeds and directions were never constant, and as no flow meter was used with the plankton nets, it was impossible to estimate the amount of water filtered by the nets. Simultaneous collections were made from various depths using a multiple horizontal tow method. Six nets were hau.led at the same time, the rings being suspended at the desired intervals from a heavily weighted cable (Fig. 4). Nets of mesh numbers 6 and 0 were used mounted on 0.5 or 1 m rings. (For detailed station list and sampling information, see Appendix 1.). The samples were preserved in formalin and were sorted at the Marine Sciences Centre of McGill University during 1970 and The larger samples were split using a modified Folsom J

23 ,- _i Figure 4. Techni~ue used for the multiple net horizontal tows. Nets may be removed from a moving cable by the svring clip (after Hansen and Dunbar, 1970).

24 WIND MJD.OMUT SEA C U R R E N T net net net ne. ring DETAIL OF NET ATTACHMENT MULTIPLE NET HORIZONAL TOW. J

25 ' Plankton Splitter, and one-half or one-~uarter of each sample was counted while the remainder was searched for the rarer species only. Fifty plankton samples, representing a total of 870 hours of sampling effort, were collected and examined for this study. The plankton collections were made in 1966 in conjunction with a study of sound scattering layers in the Arctic Ocean in cooperation with the Lamont-Doherty Geological Observatory. They were taken by Mr. William Hansen of the Marine Sciences Centre, McGill University. A sound scattering layer had been reported from T-3 by Hunkins (1965) and plankton collections were to be made in 1966 to enable comparative studies of plankton abundance and distribution before and after the seasonal appearance of the D.S.L. However, in 1966 no D.S.L. was detected with the 12-kHz recorder. Therefore, the collection investigated here represents a variety of experiments in collecting techni~ues: horizontal boat tows and multiple net horizontal tows. The duration of the tows and the towing speed varied greatly; in addition, two ring sizes and two mesh sizes were used.

26 III. RESULTS AND DISCUSSION While the sampling methods used for this collection were designed for other purposes and were so varied that very few quantitative analyses can be made, it has been possible to relate the hydrography as discussed above with the vertical distributions of the copepods. Some conclusions may also be drawn about the abundances and distributions of the reproductive stages of sorne of the most common species. Systematics: Thirty-eight species of copepods have been identified during the course of thisstudy. Of these, five need further clarification. In addition, there are nine unidentified specimens; of these, eight are sub-adult copepodites and therefore difficult to identify, and one is a mature female, referred to below as "Unidentified Species A." Copepodite stages were identified for ail species except the smallest (Microcalanus pygmaeus, Oithona similis, Oncaea borealis, and Spinocalanus ~.). Nauplii, when caught, were not identified to species as the nets used in the sampling were of too large mesh sizes to collect nauplii adequately. Whenever possible, the copepodites as weil as the adults were sexed. Sources used as aides to identification included Sars (1900, 1903, 1918, 1925), Brodskii (1950), Park (1970), Johnson (1963), and Grainger (1963).

27 The following is a systematic list of the species found: Order Ca1anoida Family Calanidae Genus Calanus 1. C. hyperboreus Kr~yer 2. C. glacialis Jaschnov 3. C. cristatus Kr~yer Family Eucalanidae Genus Eucalanus 4. E. bungii bungii Johnson Family Pseudocalanidae Genus Pseudocalanus 5. P. minutus (Boeck) Genus Microcalanus 6. M. pygmaeus (G.O. Sars) Genus Spinocalanus 7. S. abyssalis var. pygmaeus Farran 8. Q. magnus Wolfenden 9. Q.~. Family Aetideidae Genus Aetideopsis 10. A. multiserrata (Wolfenden) 11. A. rostrata G.O. Sars Genus Chiridius 12. C. obtusifrons G.O. Sars. Genus Gaidius 13. fr. brevispinus 14. fr. tenuispinus Genus Pseudochirella (G.O. Sars) (G.O. Sars) 15. P. spectabilis (G.O. Sars)

28 ,i Genus Chiridiella 16. Q. abyssalis Brodskii Family Euchaetidae Genus Euchaeta 17. ~. glacialis (H. J,. Hansen) 18. E. polaris Brodskii Family Scolecithricidae Genus Scolecithricella 19. ~. minor (Brady) Genus Scaphocalanus 20. ~. magnus (Th. Scott.) '21. ~. brevicornis G.O. Sars Family Tharybdidae Genus Undinella 22. Q. oblonga G.O. Sars Family Temoridae Genus Temora 23. T. longicornis (Muller) " Genus Eurytemora 24. E. ~. Family Metridtidae Genus Metridia 25. M. longa (Lubbock) 26. M. lucens Boeck (pacifica?) Family Centropagidae Genus Centropages 27. C. hamatus (Lilljeborg) Family Heterorhabdidae Genus Heterorhabdus 28. H. norvegicus (Boeck) 29. g. compaetus (G.O. Sars)

29 .. i Family Augaptilidae Genus Haloptilus 30. g. acutifrons (Giesbrecht) Genus Augaptilus 31.!. glacialis G.O. Sars Genus Pseudaugaptilus 32. R. polaris Brodskii Genus Pachyptilus 33. P. eurygnathus G.O. Sars Family Bathypontiidae Genus Temorites brevis G.O. Sars Order Cyclopoida Family Oithonidae Genus Oithona 35. Q. similis Claus Family Oncaeidae Genus Oncaea 36. o. borealis G.O. Sars 37. o. ~. Genus Lubbockia 38. ~. glacialis (G.O. Sars) 39. Unidentified Species A This appears to be an unusually lengthy spec~s list for the upper waters of the Arctic Basin. Brodskii and Nikitin (1955) reported finding forty-eight species of copepods from the Russian drifting station NP-2 which was in much the same area in 1950 as T-3 was for this study. Harding (1966) has

30 , reported forty-five species from T-3. These two studies, however, included samples from 4000 and 3000 m respectively, while the present study is concerned with the upper 300 m only. Comparable studies have reported eighteen species from the upper 200 m of the Arctic Ocean (Grice, 1962), and twelve from T-3 ta 300 m (Grainger, 1965). Johnson (1963) found thirty-six species down ta 2000 m at Drift Station Alpha, and Zenkevitch (1963) cited thirty-three species of planktonic copepods as being common in the Chukchi Sea; this number is now known ta be low. Grainger (1965) cited evidence from the literature ta show that about thirty copepod species are known in the upper 300 m, fifty have been taken between 300 and 1000 m, and twenty from below 1000m. Brodskii (1950) listed thirty-nine species of calanoids from all depths of the central part of the Arctic Ocean, twelve of which are endemic ta the Polar Basin, eight also occur in the Norwegian and Greenland Seas, and nineteen are also found in the Atlantic Ocean. Ta these should be added the five species which Johnson (1956) listed as being expatriates from the Bering Sea of the Pacifie Ocean. An explanation for the high number of species taken in the upper 300 m in this study may be that since the winds (and therefore the drift speeds of the ice island) tended ta be fasteet at night, many of the plankton tows were made during the night. During the middle of summer the reduction of light at night is minimal, and vertical migration may be reduced or absent in sorne species (Bogorov, 1946); but for at least part

31 '-.. \ of the sampling period, the reduction of light was probably sufficient to induce vertical migration in most of the copepod species (Digby, 1961). Previous studies on vertical migration in polar seas have been in ice-free areas. However, Hansen (pers. comm.) reports that in 1966, T-3 was surrounded by open water. In this case, some of the species found in the present collection may be expected to be weil above their normal daytime vertical ranges. For instance, Lubbockia glacialis, found by Harding (1966) only below 900 m, was taken in six of the fourteen samples taken below 70 m with the number 6 nets. The method of sampling used in this study may also be of importance in this regard. A net towed for several hours at the same depth is more likely to catch the rarer species at that depth than a net hauled vertically through a large dep ch interval. This may be the explanation for the fact that many species (Aetideopsis multiserrata, Temorites brevis, Scaphocalanus magnus, Chiridius obtusifrons, etc.) were found 50 or 100 m above the depths at which they were found by Harding (1966) or Johnson (1963). The five species mentioned above as being in need of taxonomie clarification are Metridia lucens (pacifica?), Spinocalanus magnus (males), Spinocalanus ~., Oncaea ~., and Eurytemora~. In addition, several comments are in order with regard to the taxonomy of certain other species. Metridia lucens (Boeck, 1965) is a common Atlantic species and has been reported from East Greenland (Jespersen, 1939), )

32 .i north west of Spitzbergen (Farran, 1936), Oslo Fjord (Wiborg, 1940), and Davis Strait (Jespersen, 1934). It was discovered in the Pacifie, and Giesbrecht (1895) and others noted differences in the morphology. Brodskii (1950) placed the Pacifie form in a separate species as M. pacifica. Damkaer (1964~ cited in Park, 1968) studied both Atlantic and Pacifie specimens and could find no distinct morphological differences which would warrant the formation of a new species. It is important that this controversyin the literature be resolved as Johnson (1956) has listed N. pacifica as one of the "Pacifie expatriates" (see below) found in the Chukchi and Beaufort Seas. If in fact M. pacifica and N. lucens are one species, it becomes impossible to determine whether any one specimen has entered the Polar Basin from the Pacifie or the Atlantic Ocean, and the whole question of expatriatism for this species becomes more complex. It may be that as more biological work is done in the Arctic Ocean, the mode of entry of this species will be clarified. As only one individual was collected, and it was not possible to examine the taxonomy in great detail, l have placed the specimen in the species N. lucens, especially as the validity of the Pacifie species is doubtful. Spinocalanus magnus (males): Johnson (1963) reported finding two calanoid males which he recorded only as "Spinocalanus? males", since parts of the swimming feet were missing, and because the structure of the fifth legs was unusual in being uniramous which is not the case for other males

33 of the genus. Harding (1966) found complete specimens and was able to assign the males to the species Spinocalanus magnus. In this study, many of these males were caught - mostly submature individuals (stages IV and V) and a few mature specimens. Morphological examination appears to confirm Harding's findings, although, in most cases, the swimming legs are not all present. Perhaps of greater significance is the observation that these males only occurred in samples which also contained large numbers of female S. magnus and that although the mature males were considerably smaller than the stage VI females, the stage V individuals were less so. The stage IV males and females were the same size and were indistinguishable except for the presence of the partially formed fifth legs in the young males. Spinocalanus ~.: One species of Spinocalanus present in the collection could not be identified. It occurred in one sample from 250 m with large numbers of Microcalanus with which it was easily confused. The sample as a whole was not well preserved and in most specimens the swimming legs were damaged. The overall size of the mature females ( mm) is smaller than the size ranges given by Brodskii for any Spinocalanus species. The smallest of them is S. longicornis (1.1 mm) and it may be that the present specimens belong to that species. For the time being they are listed as Spinocalanus ~. Oncaea ~.: Two mature female specimens of the genus ~1

34 ' Oncaea were found in one haul from 300 m. That they are not o. borealis is immediately obvious due to the lack of the conspicuous dorsal hump found in that species. In addition, they are somewhat larger than Q. borealis which Sars (1918~ reported as measuring 0.70 mm at the most. These specimens both measure 0.91 mm (total length). Other Oncaea species known from the Arctic are Q. minuta (which is much too small to be confused with these specimens) and O. notopus (Bogorov, 1946; Jespersen, 1939). o. notopus is cited by Sars (1900) as having a maximum size of 0.70 mm in Oslo Fjord. These two specimens may belong to a new species, or the size range given by Sars for Q. notopus may be too small. It is not an uncommon phenomenon to find that Arctic representatives of species also found in subarctic of temperate regions are larger than their southern counterparts. For the reasons given above, these specimens have been listed as Oncaea~. until their position can be further clarified. Eurytemora ~.: One mature female, found at a depth of 15 m, agrees with ail the characteristics of the genus Eurytemora, but with none of the species descriptions in the current literature. Its overall length is 1.76 mm and the structure of the fifth thoracic leg is unusual in that the distal segment bears three, rather than two, spines; two apical (as is usual) and one lateral. The pterygoid processes of the terminal thoracic segment are obtuse compared with those of other species reported from the Arctic such as E. herdmani (Grainger, 1965;

35 Johnson, 1956) and~. transversalis (Johnson, 1956). This specimen is therefore assumed to belong to a species new to science and will be fully described in a more appropriate publication in the near future. The occurrence of a member of this genus in this collection is in itself unusual (see below). Calanus hyperboreus: Grainger (1963) published length frequency histograms for the cephalothoracic length of Calanus hyperboreus in the eastern Canadian Arctic. The size range for mature females was given as 5.67 to 7.43 m. Jespersen (1937) published a slightly larger range of 5.54 to 7.70 mm for the same species in Baffin Bay and Davis Strait. He was able to relate this variation to water temperature, the smaller animais being found inoouth-eastern Davis Strait, and the larger ones in the colder waters of the Labrador Current and Baffin Bay. His conclusion was that the size of Q. hyperboreus was inversely related to water temperature. In this study, a number of small specimens were found which showed no morphological differences from the larger specimens. The size range for C. hyperboreus in the central Arctic is 5.35 to 7.50 mm. Tidmarsh (pers. comm.) has found an even larger size range for this species (5.12 to 7.68 mm) Kane Basin. in northern Baffin Bay and In these latter three areas the water temperature is generally colder than in the area of Davis Strait, and yet, contrary to what would have been expected from Jespersen's

36 conclusions, the size range of Calanus hyperboreus has been extended only toward the smaller end of the range. Combining the information given above, a new size range for the cephalothorax length of mature femajes of Calanus hyperboreus from all areas of the Canadian Arctic is proposed: 5 12 to 7. 7 mm. Genus Euchaeta: The genus Euchaeta was split by Scott in 1909 into two genera: Euchaeta and Pareuchaeta. The distinguishing characteristic was the shape of the inner bristle of the caudal rami. Vervoort (1957) examined representatives of many species of both genera over a wide geographical area and could find no constant morphological differences between them. He replaced Pareuchaeta in Euchaeta. For this reason the species in this collection are listed as Euchaeta glacialis and E. polaris, and are, in fact, identical with the species Pareuchaeta glacialis and ~. polaris reported from the Polar Basin by Harding (1966), Grainger (1965), Brodskii (1950), Johnson (1963), etc. Pseudochirella spectabilis: One mature male and one mature female of this species were identified from this collection, as well as stage IV and V individuals. The mature specimens fitted Brodskii's (1950) description in every detail except one. This one characteristic - the presence of five "massive spines" on the inner margin of the first basipodite of the fourth pair of swimming legs in the female - is important as it is a quick and easy means of making the identification. In this female speci-

37 , men, however, there were nine of these spines on the fourth leg. Jespersen (1934) has noted variations in the number of these spines in other members of the genus. The conclusion drawn here is that many morphological and size characteristics now used to define species (or even genera, as in the case of Euchaeta-Pareuchaeta) are too restricting. They may be subject to variations due to geographic isolation, temperature effects, reproductive state, feeding habits, or other ecological factors. Sorne characteristics may vary randomly in the same area as Jespersen (1934) noted for Pseudochirella. Variation in size and morphological traits has resulted in taxonomie confusion in many copepod genera such as Microcalanus, Pseudocalanus, Spinocalanus, and Metridia (Vidal, 1971~ Pseudaugaptilus polaris and Chiridiella abyssalis: Brodskii (1950) noted that the males of both of these species are unknown. A single mature male of each species occurred in this collection. A preliminary search of the literature has not revealed descriptions of these males. If none should be found, the males of Pseudaugaptilus polaris and Chiridiella abyssalis will be described in a future pu"blication.

38 ,,i Reproduction:' The coexistence of several of the copepodite stages and the mature individuals of any species indicates that it is breeding successfully in the area of collection. From the list below, it appears that ail the species in the collection, except six, are breeding successfully in the Arctic Basin. Three of these exceptions (Calanus cristatus, Metridia lucens and Eucalanus bungii bungii) are expatriates from the Bering Sea; that is, they are strays from their usual range of distribu~ion and area of successful reproduction (see below). Another, Temora longicornis, might be considered an expatriate from the Atlantic as it is known only from that ocean, although more samples in the eastern Canadian Arctic might link its previous known range with this finding from T-3. Two more, Eurytemora~. and Centropages hamatus,are members of brackish water, neritic genera. Although both are known from coastal areas of the Arctic Ocean (Johnson, 1956), these animais might be considered "expatriates" in that they are weil outside their normal salinity ranges. While the adults may survive in areas of higher salinity, successful reproduction would be doubtful. The following is the list of stages and sex,rècorded for each specie s: Aetideopsis multiserrata M F VI V V IV IV III 1

39 ' Aetideopsis rostrata M V IV F VI V IV Auga12tilus glacialis M VI V F VI V Calanus cristatus F V Calanus glacialis M VI F VI V IV Calanus h;y12erboreus M VI F VI V IV Centropages hamatus M VI Chiridiella ab;yssalis M VI Chiridius obtusifrons M VI V IV F VI V IV Eucalanus bungii bungii M VI F Euchaeta glacialis M VI V IV F VI V IV Euchaeta 1201aris F VI Eur;ytemora ~. F VI Gaidius brevis12inus M VI V IV F V IV Gaidius tenuis12inus M VI V IV F VI V IV Haloptilus acutifrons F VI V IV V III III III II III II III II III II l III II III Heterorhabdus com12actus F VI V IV III Heterorhabdus norvegicus M VI F VI V IV Lubbockia glacialis F VI Metridia longa M VI V IV F VI V IV Metridia lucens F VI Microcalanus ~aeus M VI juveniles F VI II.1 III l

40 Oithona similis F VI juveniles Oncaea borealis M VI F VI Oncaea ~. F VI Pachyptilus eurygnathus F VI V copulae,pseudaugaptilus polaris M VI F VI V IV Pseudocalanus minutus M V F VI V IV Pseudochirella spectabilis M VI V IV F VI IV Sc aphoc al anus brevicornis M VI V IV F VI V IV Scaphocalanus magnus M VI V IV F VI V IV Scolecithricella minor M VI F VI V IV III juveniles III II III II Spinocalanus abyssalis M VI V ( var. pygmaeus) II F VI V IV! Spinocalanus magnus M VI V IV F VI V IV Spinocalanus ~. M VI F VI Temora longicornis M VI III juveniles Temorites brevis M VI V IV F VI V IV Undinella oblonga M VI V F VI V Unidentified sp. A. F VI III While other species besides the six mentioned above were also caught only as mature individuals or in one or two young

41 stages (Lubbockia,glacialis, Chiridiella abyssalis, Euchaeta polaris, Oncaea ~., and Unidentified Species A), the limitations of the sampling methods must be borne in mind. Of these five species, only Lubbockia glacialis was taken in more than one sample. The others were all extremely rare, and it was not unusual to find only one or two individuals. The younger stages of ~. glacialis are small enough to be missed by a number 0 to 6 net; even the mature individuals of this species were only captured with the number 6 nets. In fact, when interpreting the above results, the efficiency of large mesh plankton nets in sampling small size organisms or the smaller juvenile stages of some species should always be considered. In spite of this, Harding (1966) has previously shown that ~. glacialis, Chiridiella abyssalis and Euchaeta polaris do in fact breed successfully in the Arctic Ocean. Of the thirty-nine species in the collection, four were caught throughout the depth range under consideration, and in sufficient numbers that some quantitative analyses might be made of their reproductive cycles and of the depth distribution of their developmental stages. These species are Calanus hyperboreus, C. glacialis, Metridia longa and Euchaet~ glacialis.

42 ' Breeding Cycles: There were five more or less distinct time periods during which most of the samples \'J"ere taken in the summer of These were late June and early July, late July, early August, mid August and late August. Enough plankton samples were taken in each of these time intervals to make possible an analysis of the reproductive cycles of each of the four above species. In presenting the relative numbers of growth stages, percentages rather than absolute numbers have been used in all cases in order to avoid the bias introduced by differences in the numbers of samples or in overall towing times in each interval. Calanus hyperboreus: Harding (1966) found that Calanus hyperboreus does not complete its development in one year in the Arctic, but may take two years or even longer to reach maturity. This phenomenon has been noted for other copepod species in Tanquary Fjord and the Arctic Ocean (Cairns, 1967, 1969). A slower development rate to maturity may be a method of conserving energy and diverting it to be used in the strenuous process of adaptation to life in the rigorous Arctic environment (Dunbar, 1968). Cairns (1969) noted the difficulty that he and other workers have encountered in following the developmental patterns of Calanus species and contended that in Tanquary Fjord and the central Arctic Ocean, Calanus hyperboreus and Calanus glacialis are opportunistic breeders with multi-year cycles,

43 ,-,.i reproducing whenever environmental conditions will allow. Table 1 Relative occurrence of the copepodite stages of Calanus hyperboreus in each of five sampling periods in 1966, expressed as percentages of the total copepodite population. Copepodite June 25- July 23; August August August Stage July l II III IV V VI F VI M N 2, ,276 2,136 Calanus hyperboreus is known to overwinter as stage III or IV (Grainger, 1959); in Arctic areas where this species has a two- or multi-year breeding cycle, subsequent winters must be spent as stage V or VI. Breeding occurs whenever environmental conditions are favourable and there are fecund females present (Cairns, 1969). The increase in numbers of stage IV and V Calanus hyperboreus during the first part of the sampling J

44 Figure 5. Relative occurrence of the copepodite stages of Calanus hyperboreus in each of five sampling periods in Shaded areas represent males. i

45 , Calanus hyperboreus 80 (%) r JUNE 25 - JULY 10 N= JULY N:: AUGUST N= AUGUST N= AUG UST N= 2136 l n m IV V VI

46 '-,.i period, shown in Table 1 and Fig. 5, reflects the maturation of the first year overwintered individuals. Johnson's (1963) data indicated that spawning took place in late December off Drift Station Alpha; Cairns (1969) found that, off T-3, Q. hyperboreus spawned in early September in 1964; in the eastern Arctic, Grainger (-1959) found that it occurred in May. Brodskii and Nikitin (1955) found the largest numbers of stage l present in June, indicating that spawning took place in the Arctic Basin in May in 1950 because Grainger's (1959) data indicate a three or four week development time from the first appearance of the nauplii to that of copepodite!. Thls variation in spawning time substantiates Heinrich's (1961) observation that Calanus hyperboreus belon 3to a group of copepods which are capable of breeding in the absence of an abundance of phytoplankton. The presence of higher numbers of mature females and males in June and July indicates that, in 1966, spawning of Calanus hyperboreus occurred before sampling began, perhaps in late Mayas observed by Brodskii and Nikitin (1955). Although no stage l copepodites were caught, stage II was present throughout the period of sampling except in late August, by which time moulting to stage III appears to have occurred. Bogorov (1938) found that "biological spring" (the phytoplankton b1oom) occurs in August in the Arctic Ocean. The development of Q. hyperboreus to stage III and feeding in preparation for over-

47 , wintering coincide with the bloom. Fat stored at this time has an energy content of over 10,000 cal/gm (Conover, 1964). Increases in the numbers of stage V and VI in August reflect the two-year cycle of this species as the one-year-old individuals mature in preparation for a second winter and spawning in the following May. Calanus glacialis: Many of the comments made above regarding two- or multi-year cycles also apply to Calanus glacialis. Table 2. Relative occurrence of the copepodite stages of Calanus glacialis in each of five sampling periods in 1966, expressed as percentages of the total copepodite population. Copepodite June 25- July 23- August August August Stage July J l II III IV V VI F VI M N ,

48 Figure 6 Relative oc'currence of copepodi te stages of Calanus glacialis in each of five sampling periods in Shaded areas represent males.,-

49 Calanus glacialis (%) JUNE 25 - JULY 10 N=-738 JULY N= AUGUST 8-11 N a L- ~==~======~_ 10 AUGUST N o L-~---r--~~---.~.- Il III IV V VI AUGUST Nr:174

50 The data presented in Table 2 and Fig. 6 show that stages V and VI of C. glacialis were dominant throughout the 1966 season, as was also found by Harding (1966). This species is known to overwinter as stage III (Grainger, 1965) during its first year, and this is reflected in the presence of stage III in the first half of the sampling period only. The second winter is spent as stage V or VI as shown by the decrease in numbers of stage IV and simultaneous increase in stage V throughout most of the summer. Heinrich (1961) has shown that Calanus glacialis breeds only during periods of high concentrations of phytoplankton. The appearance of increased numbers of males and the larger numbers of females in August indicate that spawning occurred during the "biological spring." The stage l copepodites had not yet appeared before the sampling period ended. One question arises with respect to these Arctic species which do not have annual reproductive cycles. What happens to the population during long periods when the environmental conditions are not favourable for reproduction? The constant losses due to mortality as a result of predation, aging, etc., woùld be expected to exert a considerable toll on the population over a period of two, three, or more years. While Brodskii and Nikitin (1955) found the largest numbers of stage l copepodites of Calanus hyperboreus in June, indicating that spawning took place in May, their data also showed j

51 that stage l was present in small numbers in every month of Johnson (1963) found young copepodites of c. hyperboreus and C. glacialis over a "wide season" - longer than would be expected for these two species. He also mentions the "seemingly precarious existence" of the Calanus species. Since breeding in Calanus glacialis has been shown not to occur before the phytoplankton bloom, the presence of small numbers of stage II in late July and early August is puzzling. This, and the evidence from the literature cited above, suggests that a constant, very low rate of reproduction takes place in these populations as weil as the short period of intensive breeding which is environmentally controlled. Such a mechanism would allow a slow, but constant recruitment into the older stages to compensate for the losses mentioned above. low level of reproduction (as distinct from the periods of A constant intensive reproduction analysed above for the two Calanus species) would be an obvious advantage as a means of preventing the total extinction of a population of opportunistic treeders during long periods of conditions unfavourable for intensive reproduction. Careful collections of Calanus over a pe~iod of at least a year, using fine mesh nets (to ensure efficient catching of the small stages), is needed to settle this matter. Euchaeta glacialis: Unlike the three other species discussed here, Euchaeta glacialis is predatory, and is therefore

52 not directly dependent on an abundance of phytoplankton for successful breeding. Johnson (1963) observed egg sacs and spermatophores associated with mature individuals in both winter and summer samples. This indicates that breeding may occur throughout the year and substantiates Brodskii and Nikitin's (1955) finding of stage l copepodites in neally every mon th. Both of these reports note that the greatest number of stage l occurred in early June, leading to the conclusion that, although breeding probably occurs ail year round, there is a peak season shortly before June. Heinrich (1961) points out that the asynchronous breeding cycles of the various copepod species - of short or long duration, and dependant or independent of phytop~ankton blooms - are useful to the communit y as a whole as it reduces interspecific competition for food. Table 3 and Fig. 7 present the 1966 T-3 data for this species. While the number of stage l copepodites found in late June and early July was small, the high percentages of stage II at this time indicate that a successful spawning had occurred and that the majority of the offspring had already passed through stage l before sampling was begun.

53 Figure 7. Relative occurrence of the copepodite stages of Euchaeta glacialis in each of five sampling periods in Shaded areas represent males. '- 1

54 , E_uchaeta glacialis 80 (0/0) JUNE 25 - JUL y 10 N= ao , r-- JULY N= AUGUST 8-11 N= AUGUST N= JJ CL AUGUST 0 N=549 Il III IV V VI 27-28

55 '-, i Table 3. Relative occurrence of the copepodite stages of Euchaeta glacialis in each of five sampling periods in 1966, expressed as percentages of the total copepodite population. Copepodite June 25- July 23- August August August Stage July l II III IV F ' IV M V F V M VI F VI M N 1, Since E. glacialis is a carnivore, it is presumably able to continue its development during the winter months rather than overwintering at a single stage as do the other species discussed. Accordingly, the development to maturity seems to take only one year in this species, and early copepodites caught in 1966 would have been responsible for the generation spawned in The data for the stages IV, V and VI are more difficult to interpret. The relatively small numbers of these stages

56 ' may indicate that the June spawn of the previous year was not particularly successful. In this case, the populacion may have been augmented by the year-round reproduction noted by Johnson (1963) and by Brodskii and Nikitin (1955). Some of the mature animais represented here are probably those which spawned just before sampling began in late June of 1966, and these may be expected to die out during the remainder of the season. Mileikovsky (1970) has found exactly the same type of year-round breeding cycle with a period of intense spawning in the spring for the predaceous pteropod Clione limacina. Metridia longa: Since the size of the stage l copepodites of Metridia longa approaches the sampling limits of number 0 mesh nets, only the data from the number 6 nets have been used here. The data from NP-2 (Brodskii and Nikitin, 1955) seemed to indicate that N. longa may breed during a large part of the year, even during the winter. This species, like Q. hyperboreus, was found by Heinrich (1961) to breed independently of the phytoplankton bloom. Grainger (1959) found that nauplii of ~. longa appeared from March to July at Igloolik in the eastern C anadian Arc tic.