Gonad maturation as an indication of seasonal cycles for several species of small copepods in the Barents Sea M. FREDRIKA NORRBIN Norrbin. M. F. 1991: Gonad maturation as an indimtion of seasonal cycles for several species of small copepods in the Barents Sea. Pp. 421-432 in Sakshaug, E., Hopkins, C. C. E. & Britsland, N. A . (cds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology. Trondhcim, 12-16 May 1990. Polar Reserch I O ( 2 ) . I n a remote oceanic area likc the Barents Sea. it is often difficult to follow the seasonal developmcnt of copcpod populations in detail. Information on the gonad maturation stage of older juveniles and adults of a species will reveal the immediate state of reproduction and the expected development of juveniles into reproductively active adults. Winter “resting stages” in juveniles can also be rccognised. Zooplankton were caught during Pro Mare cruises in early March. May. July/August, mid-Scptember and mid-October. Abundance and composition of developmental stages of small copepods were determined for several stations from each cruise. Samples consisting of Stages CIV to CVI of Pseudocalanus acupes (Giesbrecht 1881). P. minufus (Kroyer), Microcnlanus pusillus (Sars). and M . pygmaeus Sars were stained with carmine and analysed with respect to gonad maturation stage. length. width, and area of the prosome and the area of the gonad and the oil sac. Image analyses were performed from photographs or drawings of copepods using a digitising pad. With additional information on abundance and stage composition, and by comparing the present data set with information on Pseudocalanus spp. from Balsfjordcn, northern Norway, seasonal cycles for the species could be inferred. Fredriku Norrbin. Depurfment of Aquatic Biology. The Norwegiun College of Fishery Science. P . 0. Box 308.3 Guleng. N-9001 Tromsm, Norway (reuised February 1991). Introduction In Arctic and subarctic areas, where the yearly input from primary production to higher trophic levels is typically limited in time, the dominant species of large, herbivorous copepods are known to be active during only part of the year (Hopkins et al. 1984; Tande et al. 1985). They produce one or less than one generation per year and spend most of the time either building up energy stores or in a resting state, slowly consuming these reserves. The many smaller species of copepods have been described in several studies of high latitude zooplankton (e.g. Ussing 1938; Digby 1954; Fon- taine 1955; Grainger 1959; McLaren 1969). Although numerically dominating in nearshore Arctic areas (Horner & Murphy 1985), the small species have lately been found to be of little importance in offshore ecosystems (e.g. Cooney & Coyle 1982; Vidal & Smith 1986). However, many of the recent studies have been conducted primarily during the spring and summer seasons, while high abundances of small copepods have been recorded mainly during the second half of the year (Minoda 1967; Hassel 1986; the present study). In many cases the nets used have been too coarse to catch young stages of Microcalanus in particular. This suggests that smaller species may still play an important role in Arctic food webs during autumn and winter. Gonad maturation has been used to describe life cycles for large, herbivorous copepods and some smaller species in Balsfjorden, Northern Norway (Tande & Hopkins 1981; Tande & Gron- vik 1983; Norrbin 1987) and in the Barents Sea (Tande et al. 1985). Experimental work with a small copepod from temperate areas shows that the gonads of juveniles grow larger under good feeding conditions (Razouls et al. 1987). This is reflected in seasonal cycles in gonad size and maturation of Pseudocalanus in Balsfjorden (Norrbin 1987). In Balsfjorden, the drop in gonad 422 M . Fredrika Norrbin size of juvenile Pseudocalanus in late summer anticipates the beginning of the overwintering resting period long before signs of this are visible in other factors (Norrbin 1987; unpubl. data). During the latter part of the Pro Mare programme, some of the cruises took place during autumn and winter, enabling the study of biologial material from the “unproductive” season. The aim of this project within Pro Mare has been to evaluate the importance of the smaller species of calanoid copepods in the Barents Sea ecosystem, with an emphasis on their significance during the time when larger calanoids are inactive. In this paper information on stage composition and gonad maturation is used to infer seasonal life cycles for four copepod species; Pseudocalanus acuspes, P . minutus, Microcalanus pusillus, and M . pygrnaeus. Materials and methods Plankton was sampled during the Pro Mare cruises of 1987 and 1988; Cruise 11 in March 1987, Cruise 13 in October 1987, Cruise 14 in May 1988, and Cruise 16 in September 1988 (Fig. 1). Additional samples derive from Cruise 2 in July- August 1984 (Fig. 1). The region covered by sampling typically consisted of latitudinal tran- sects into the ice edge followed by a number of stations in ice-covered waters. Sampling equipment usually consisted of 90 pm WP-2-nets. On Cruise 12, a 200 pm WP-2 net was used and on Cruise 10. a 90 pm Juday net (for a description of the nets, see e.g. Unesco 1968). Zooplankton was concentrated and preserved in a solution of 4% formaldehyde and 5% propylene glycol immediately upon sampling. This was exchanged for a weaker preservative upon return to the laboratory. Stage and sex composition of small copepod species were estimated from subsamples. A mini- mum of 1000 individuals were counted from each station. Gonad maturation of males and females of copepodite Stages IV to V was analysed for the species Pseudocalanus a c w p e s (Giesbrecht, 1881; see Frost 1989), P . minutus (Kroyer), Micro- calanus pusillus (Sars), and M . pygrnaeus Sars (Fig. 2). (The genus Microcafanus is currently subject to some taxonomic confusion. The species in this study have been determined as far as possible l i Fig. 1 . The western Barents Sea. Stations from Pro Mare cruises. Starting and end stations are indicated. the lines connecting the stations signify different cruises; small dots = Cruise 10; larger dots = Cruise 13: unbroken line = Cruise 14; lines and dots = Cruise 16. In addition. stations from Cruise 2 relevant for this study are denoted by triangles. according to Sars (1903), Brodskii (1950) and Vidal (1971). Occasional specimens of Micro- calanus s p . b were found in the samples (Vidal 197 1)). There was a great variation in the abundance of different species and developmental stages in samples from different stations and seasons. The aim was to analyse at least 20 copepods of any given stage and sex from each station, but this was often not possible. Copepods were stained in ethanolic borax car- mine for several days, then rinsed in distilled water, dehydrated in ethanol and finally trans- ferred to cedar wood oil (Norrbin 1987). Groups of stained copepods were arranged in a lateral position in cedar wood oil on a glass slide (Figs. 2 and 3). They were then photographed through a Wild Heerbrugg dissecting microscope. The resulting colour slides were projected down on a CalcComp 2500 digitising pad. Several rneasure- ments were taken (see below) and fed directly into a computer file using a specially adapted Seasonal cycles of small copepods in the Barents Sea 423 taneously, notes were made of gonad maturation stage (after Corkett & McLaren 1978; Tande & Hopkins 1981; Norrbin 1987). Analysis of the data was performed using Mic- rosoft Excel spreadsheets and Statgraphics Stat- istical Graphics System, Version 3.0. A C D 0 1000 pm I -I Fig. 2. Species in the present study. appearance of stained copepods. The figure shows prosome and urosome in lateral view. A. Pseudocalanus acuspes CV female in advanced state of maturation. B . P . minufus CV female in gonad resting state. C . Microcalanus pusillus CV male with medium-sized gonad. D. M. pygmaeus CV male in gonad resting state. Basic programme which calculated distances and areas. A small number of the copepods were drawn on paper using a camera lucida instead of photography. These were digitised in the same way as above. The measurements taken were prosome length and greatest width, area of prosome, area of gonad, and area of oil sac (Fig. 3). Simul- PROSOME AREA GONAD AREA / GUT OIL SAC AREA LENGTH Fig. 3. Measurements taken on copepods using digitising table. Results Species composition All four species in this study co-occurred on sta- tions in the western Barents Sea (Table 1). There was a tendency for Pseudocalanus to be dominant in shallow areas and for Microcalanus to be more dominant in water of more than 200m depth (Table 1). Abundances were generally higher in the south- ern part of the investigated area and copepod numbers increased during the year, reaching a maximum in September and October (Table 1). Stage distribution Some features of the stage distributions of all copepod species in the present study are especially noteworthy; presence of adult females indicates ongoing or recently ceased reproduction, and presence of adult males indicates that mating and reproduction have recently begun. A large num- ber of early copepodites (CI) is a sign of recent reproductive activity and ongoing development of the population. McLaren (1974) and Corkett & McLaren (1978) give a development time from egg to CI of approximately 45 days at 0°C for Pseudocalanus). Absence of the youngest stages and a predominance of CIII to CV imply an accumulation of overwintering stages. Because stage distribution varied between regions, the figures representing this distribution are divided into five areas (Figs. 4 to 7): Atlantic Water north and south of approximately 76“N, Polar Water north and south of approximately 7YN, and shallow areas at 75 t o 77”N (on Spits- bergenbanken or Hopenbanken), usually with mixed water masses. Representative stations from each sampled region are shown for March, May, September, and October (Figs. 4 to 7; Table 1). Northerly Atlantic Water stations and southerly Polar Water stations overlap due to the fluc- tuation of the position of the Polar Front. Most stations were localised in the marginal ice zone, 424 M . Fredrika Norrbin Table 1. Zooplankton sampling data for the Pro Mare cruises and abundance of small copcpods. "Region" refers to regional divisions used in Figs. 4 to 7. P.a. = Pseudocalanus acuspes. P . m . = P. mbiurus. M.pu. = Microcalarius pusillus. M.py. = M . pygmaeus. Total = total small copepods including Cyclopoida and Harpacticoida. A plus sign signifies an abundance of less than one individual per cubic meter. Positron Depth Abundance of small Copepods (individuals m-:) Date Station Region Lat Long ( m i ~ . a . Cruise 2 , 1984 21 Jul 11 01 Aug 111 01 Aug 118 01 Aug 137 M A u g 160 Cruise 10. 1987 01 Mar 2 02 Mar 6 06 Mar 13 Cruise 13, I987 15 OCI 1 I6 Oct 5 16 Oct 7 20Oct 12 22 Oct 15 24 Oct 17 Cruise 14, 1988 21 May 3 21 May 5 21 May 6 22 May 7 23 May 9 24 May 10 27 May 13 3OMay 16 Cruise 16, 1988 G9Sep 1 10 Sep 3 10 Sep 4 1 1 Sep 6 1 1 Sep 8 I3 Sep 12 14 Sep 13 17Sep 16 - - - - - C D C E D B B C E E D D D B C C C D B A A - - - - 78"4X'N 80"12'N 80'52" RI"I8'N 79"40'N 75"16'N 75"49'N 76"l I'N 73"10'N 76'00" 77"W'N 78"Ol ' N 77"40'N 77"15'N 74"W'N 75"00'N 76"03'N 76"22'N 76"16'N 76%" 76"10'N 75"2Y'N 76"W'N 76"34'N 77"W N 78"00'N 79"M'N 80"33'N 80"12'N 7953" 28"IY' E 24"16'E 23"OO'E 22"20'E 26"26'E 23"3 1 ' E 30"M'E 22"5 1 ' E 27"58'E 32"IO'E 32"00'E 31"41'E 3020'E 29"OO'E 31"13'E 3173'E 3 1 W ' E 31"03'E 29Y7'E 27"lI'E 2355'E 23"31'E 28"Oo' E 2 8 2 8 ' E 29"26' E 3 1 W ' E 31"56'E 30"28'E 29"42'E 28"05 ' E 30 195 55 220 150 114 325 48 320 320 210 204 208 187 279 358 335 304 24 1 91 42 85 167 72 232 233 80 207 215 311 0 23.121 98 6.369 38.270 5.931 25.134 0 15.971 12.015 1.492 2.738 300 1.307 1 .OX5 505 557 1.258 4.832 133 10.369 IW,210 53,897 2.871 16.332 27.269 26.742 54.055 - 88. I 30 P . m . ~ 4go 1.861 11,757 18.693 2.246 4.397 982 - + 10.394 11.348 1,014 16.81 1 573 1.502 980 1.883 842 658 802 392 333 8.791 23.739 9.503 269.8 10 29,705 13.639 8.413 11.605 . . M.pu. M.py. Total 0 0 9.013 392 0 + 2,492 + - - 8.630 + 34.200 614 82 + 27.840 0 246.493 9.328 60,744 5.340 457 477 10.443 225 1.592 + 82.028 601 8.479 + 12,673 1.199 11.765 787 5 , 5 0 9 692 350 + 0 0 2,332 loo 7.664 676 + + 13.640 638 1,816 306 0 157 9.471 2.567 9.915 3.906 11,146 2.266 833 54.766 28.412 46.663 109,134 243.645 29,882 127.069 1,470,505 580.742 9.728 326.286 28. I46 232.864 213,115 214.836 172.019 125.871 27,003 8.687 9.260 664.957 456,158 2.000.267 1.631,141 177,806 148.026 182.986 193.279 - but the southernmost stations of each cruise were usually set up in open water: stations one to seven of Cruise 13. one t o six of Cruise 14. and stations one to six of Cruise 16. During Cruise 2 all stations except no. 111 were localised in open water. Pseudocalanus acuspes In March the population of P. acuspes was still dominated by overwintering copepodites (CIII to CV; Fig. 4). In May the population was in a developing phase at most stations, as evidenced by the presence of CI (20 to 40%) and older juvenile stages (Fig. 4 ) . The shallow water sta- tions (Fig. 4C) where adults dominated were an exception. In these areas there were high densities of Cirripedia nauplii which may have preyed on young copepodite stages. The southernmost, open water station (Fig. 4E) had the highest number of young copepodites. In July/August there seemed to be developing populations of P. acuspes at stations to the west of Nordaust- landet. North of Svalbard the populations were either i n a developing phase or dominated by older juveniles. In September, there was a clear predominance of early copepodites in most areas (Fig. 4A, C and D). Very few older copepodites and adults were present, except for a lingering population of adult females at the southerly polar water stations (north of 77'"; Fig. 4B). The Stage 111 copepodites may have represented the build- Seasonal cycles of small copepods in the Barents Sea 425 r P . acuspes f- 50 0 5 0 5 0 0 50 50 0 5 0 5 0 0 50% MARCH MAY SEPTEMBER OCTOBER F;g. 4. Seasonal stage distributions for Pseudocalanus acuspes. The presentation is divided into sevcral gcographical areas (from north to south): A . 79-81"N. Polar Water. B. 76-78"N. Polar Water. C. 75-77"N. shallow (bank) areas with polar or mixed water. D. 76-77"N. Atlantic Water. E. 7S75"N. Atlantic Water. Distributions arc givcn as pcrccnrages of all copepodites. Females are placed to the right of the vertical bar and males to the left (sex was determined for copcpoditc Stages IV to VI). Each distribution represents one to three similar stations. See Table 1, "Region" column, for information on stations used in the construction of this figure. Where a distribution is based on less than 100 copepodites, the number is given in the figure. P . minutus 11 II A I . I1 I t I i 7 L * n=39 n=32 P 1 7 n = 5 9 1 1 I A n=26 50 0 50 SEPTEMBER 50 0 50% OCTOBER 50 0 50 MAY 50 0 50 MARCH Fig. 5 . Seasonal stage distributions for Pseudocalanus minutus. See Fig. 4 for clarification 426 M . Fredrika Norrbin M . pusillus 50 0 MARCH L n=2 1 f n=70 I ?€.!.I n = 1 4 r n=34 50 0 50 MAY 50 0 50 SEPTEMBER Fig. 6. Seasonal stage distributions for Microcolonus purillus. See Fig. 4 for clarification up of the overwintering population. No adult males were present. I n October early copepodites were still present. but the adults had disappeared (Fig. 4 ) . Pseudocalanus minutiis I n March adult females of P. minutus were present at several stations (Fig. 5). Copepodite Stages I1 and 111 at these stations may have repre- sented reproduction earlier in the winter or part of the overwintering population. I n May there was a dominance of adults and older juveniles at all stations (Fig. 5). As in P . acuspes, the highest concentrations of younger copepodites were found at the most southerly station (Fig. 5E). In July/August the situation was similar to that of P. acuspes. In September the accumulation of overwintering stages had proceeded further than for P. acuspes. with 50 to 70% of the population consisting of copepodite Stages 111 to V . As in P. ncuspes, there were no adult males, but adult fema!zs were still present at northerly stations (Fig. SA). In October the dominance of over- wintering stages was almost total (Fig. 5 ) . There were, however. females present at the north- - f T n=25 I 50 1- 0 5 0 % OCTOBER ernmost stations (Fig. 5B). where grazing seemed to occur. MicrocaIanus pusillus I n March all stages of M . pusillus except adult males were present, suggesting that reproduction had taken place earlier in the winter (Fig. 6C and D). All developmental stages were found at most stations in May (Fig. 6), but at shallow water stations adults and older juveniles dominated (Fig. 6C). In July/August populations seemed to consist largely of adult females. but the younger copepodites are likely to have slipped through the 200pm net because CI to CIII of this species measure n o more than 200 to 340 pm in prosome length. In September the younger copepodite stages comprised 10 to 30% of the populations (Fig. 6). Adult females were abundant especially in areas influenced by Polar Water (Fig. 6A, B and C) and adult males were present on the northernmost stations (Fig. 6A). All devel- opmental stages were still present in October, but older juveniles and adults predominated (Fig. 6). Adult males were present at most stations. Seasonal cycles of small copepods in the Barents Sea 327 II I A in M . pygmaeus 50 0 5 0 5 0 0 5 0 5 0 0 50% MARCH MAY SEPTEMBER OCTOBER Fig. 7. Seasonal stage distributions for Microcalanus pygmaes. Scc Fig. 4 for clarification. 401 x o 10 20 30 40 m 80 I m m 4 0 60[ f # \ 2010 0 0 I 10 20 30 40 F " I I I I I 10 20 30 40 r D t A 0 A A t A J A O*& A A Oil sac index I I I I I 10 20 30 40 50 Fig. 8. Seasonal variations in gonad size and oil content. Station averages for CIV females. CIV males. CV females and CV males. Oil sac index = oil sac area as percentage of prosome area. G/G,., = Gonad area as percentage of maximum observed gonad area in the specific copepodite stage and sex. Open squares = March. solid squares = May, diamonds = July/August. solid triangles = September. and open triangles = October. Arrows show the seasonal cycle of resource allocation to either gonads or oil. A. Pseudocalanus acuspes. B. P. minutus. C . Microcalanus pusillus. D. M . pygrnaeur. 428 M. Fredrika Norrbin Microcalanus pygmaeus In May the stage distribution of M . pygniapirs was similar to that of M . piisillus (Fig. 7). In September, stage distribution at most stations resembled that of P. minutus. with a dominance of older copepodites and adult females (Fig. 7A and D). Some adult males were present, indi- cating that the populations were not completely "at rest". In October, older juveniles made up from 60 to 100% of the populations, but adult males were still present at Polar Water stations (Fig. 7B). Very few specimens were found in the March and July/August samples. Seasonal oariation in gonad size and oil content Seasonal cycles of energy allocation were evident i n stages CIV to CV of all four species (Fig. 8). I n May, when gonad sizes were at their maximum. the copepods contained little oil. Oil content increased and gonad size decreased as the season progressed. Maximum oil sac sizes were attained in September for all species. Minimum sizes in gonads were found from September to October for Pseudocalanus acuspes and Micro- calanus pygmaeus, from August to September for P. minutus, and in October or March for M . pusillus. In P . acuspes CIV and CV gonad sizes reached a maximum in May (Fig. 8A). At the same time, oil content reached a minimum i n this species. In September, copepodites from the northernmost stations tended to have smaller gonads than those from stations further south. Gonads of P. minutus copepodites reached their greatest size in May (Fig. 8B) but were large in some localities both i n early August and in mid-September (late summer situations). In early March, gonads increased in size while oil content decreased in all species except M . pusillus (Fig. 8 A and B ) . In P. acuspes CIV, but not in CV, gonads were larger in March than in the period September/October. indicating the start of spring maturation in the overwintering generation. In P. minutus Stage V copepodites also displayed this sign of early maturation i n March. Oil stores were nearly depleted i n this species in March (Fig. 8B). The seasonal cycles in both gonad size and oil content were much less-pronounced in the Microcalanus species (Fig 8C and D). Cope- podites from May, September and October exhibited some large gonads but also smaller and Fig. 9. Life cycle and gonad development of a small calanoid c o p e p d . Inner pathway represents direct development during productive season. Resting stages usually consist of CIV-CV (Pseudoco~anuc acuspes also CIII) and are formed before winter. In winter-early spring these stages enter the cycle of normally moulting and maturing copepodites. less mature ones. consistently small. Only in March were gonads Discussion Seasonal cycles in maturation The results presented here indicate trends in gonad maturation and in the population dev- elopment of small copepods. More frequent sam- pling is needed i n order to describe seasonal cycles in detail. It is obvious from earlier studies that small, undeveloped gonads can be used to identify resting copepodites, and that gonads in cope- podites during the productive season always appear in a wide size range with a predominance of larger size groups (Fig. 9; Norrbin 1987; unpubl. data). The size increase in gonads of female copepodites is due to the prolongation of the gonad into forward diverticulae and to maturation and migration of oocytes. Razouls et al. (1987) showed that gonads of CV females of Temora stylifera grew more i n size under good food conditions. This should lead to efficient util- isation of abundant food regimes, allowing for early maturation and reproduction in the adult stage. In May and August several specimens of Pseudocalanus CV females had ova in their ovi- ducts. In male CV t h e reproductive system could also be well differentiated, with a visible sperm sac and spermatheca. I t is likely that the seasonal cycles of the Pseu- docalanus species in the souhern parts of the Seasonal cycles of small copepods in the Barents Sea 429 O N D J F M A M J J A S o L ' " ' " ' ' l I I Fig. f0. Seasonal cycle in gonad size for CV females from Balsfjorden, Northern Norway 1985-86. Gonad size was measured as gonad length. A . Pseudocalanus acuspes. B . P . minutus. Data modified from Norrbin (unpubl. data). Barents Sea (72 to 76'N) are similar to those in Balsfjorden (69'30"). In the northern areas, which are ice-covered most of the year, there is likely to be a shorter productive period, which results in fewer generations per year. The seasonal variation in gonad length for Pseudo- calanus acuspes and P . minutus CV females in Balsfjorden (Fig. lo), together with monthly analyses of stage composition, indicates three generations of P . acuspes and two generations of P . minutus per year in this area (Norrbin 1987). Overwintering CV females of P. acuspes mature between March and April; the first new gen- eration of the year matures in May and there is a second generation which matures in July before the build-up of the new overwintering population begins in August (Fig. 10A; Norrbin 1987). The overwintering generation of P. minutus matures in the February-March period and CV cope- podites disappear until May (Norrbin unpubl. ; Fig. 10B). The first generation matures in May and from June onwards the overwintering gen- eration is formed, with gonads slowly developing during autumn and winter (Fig. 10B). This pattern is closer t o that of the larger copepod Calanus finmarchicus in this area (Tande & Hopkins 1981). Similar presentations of gonad size are given for CV females of the four species in the present study (Fig. 11). Since gonad length was not measured here, and area is proportional to length squared, the square root of the gonad area measurement has been used in Fig. 11. A constant relating area and length is lacking, giving lower values for the Barents Sea copepods (Fig. 11). The Barents Sea data for the Pseudocalanus species imply similar cycles. For P . acuspes the stage composition and state of maturation of over- wintering juveniles were almost identical in the Barents Sea and Balsfjorden in early March (Figs. 4, 10A and 11A; Norrbin 1987). In May there is an obvious maturation of the first generation (Fig. 11A). It is likely that two to three generations of 0 ' 4 [ A - 0.2 0'41 -€- E - L (II m t 0 0, o'21 0.1 \ 't 0.2 0.1 0 0 till,, O N D J F M A M J J A S Fig. If. Seasonal cycle i n gonad size for CV females from the Barents Sea (present study). Gonad sizes were measured as gonad areas. The square root of this measurement was taken to approximate a length measurement; due to lack of a scaling constant these values are smaller than in Fig. IOA. A. Pseu- docalanw acuspes. B . P . minutus. C. Microcalanus pwillus. D. M. pygmaeus. 430 M . Fredrika Norrhiri this species occur per year also in the southern parts of the Barents Sea. Further north there a r e probably fewer generations. Maturation of P. r n i n i r f u . ~ may occur somewhat later in winter in the Barents Sea. in the northern parts of the Barents Sea. the whole cycle is likely to be delayed. leaving room for only o n e generation per year. I n other Arctic localities. P . rnitzitiirs has been found to be mainly annual (Ussing 1938: Digby 1954: Fontaine 1955) or with a life cycle taking from o n e and a half t o two years t o com- plete (Digby 1954; Grainger 19%). However. since a revision of the genus P s r ~ k m l a r i i r s has taken place only recently (Frost 1989). several species may have been involved in these studies. Ps~irdoc~mlcitiiis females may be reproductively active for up to two months (Corkett & McLaren 1978). This may account for the presence of adult P. m m t m . s females in September and October. Late breeders were also found by Fontaine (1955) i n U n g a v a Bay. Offspring produced this late may have difficulty in surviving under the poor food condition\ during winter. Population densities in March art. only 5 to I O V of those i n September and October (Table I ) . I t is clear that Mic.roculurius pir.sil/ir.s cope- podites h a w a higher degree of maturity later i n the season than Pseirrlocczlrrnlr.( (Figs. 10 and 11). The extent of nlaturity of ,v, pFgr,luell,y cope- pndjres seen,s to var! l i t t l e c j r l r i n g t h e y e a r ( F , ~ , a higher degree in this species. as it does in Accirfio ~ o r ~ g I r u r ~ r i s ( N o r r b i n 1987) and in the larger cope- pods (Tande et al. 1085). PAOOUCTIVE SEASON PERIOD M . pusillus M . pygmaeus P. acuspes t P. minutus INCREASING SIZE: ~ i g . 12. P O ~ S ~ ~ I C beasonat cycich i n the productitin perl,,d f o r four~peciesofcopepods i n the western Barents Sea. The shaded area represents the period of reproduction and developing juveniles in the foulhern and central parts of the investigated area. The spccics arc arranged i n circles related to prosome the lnnermc>st clrcles. The diameterofeach circle is proportional the prosornc length of the specie\ i t reprcwnts. The length of the productive season is roughly inversely proportional t o ~ I L C ot the copepod specie\. D ) . Perhaps matur'ltion occurs in the to length. w i t h the smalleq specie\ ;ind longrst activity pcrlods in Oil siic s i z e I t is bigniticant that the hmallest species in the present study are those which vary the least in fat reserves (Fig. 8). Small copepods such as Micro- culuri~t.7 ( C V and adults of the smaller iw. yiczi/liis reach a prosome length of about 0 . 5 m m ) are likelk t o be vulnerahle to starvation due to rela- tively large metabolic needs. Smaller organisms have higher weight-specific basal metabolic rates (Kleiher'\ Law). increasing their needs for a con- stantl! renewed food supply. This may be the prime reason why small copepods have a longer a c t i v i t y period during the year. o r . to be precise. why some copepods have undergone a n adap- tation to a prolonged activity season consisting of a decrease i n size. T h e Microcalati~is species are therefore unlikely to abstain from feeding at any part of the year: their likely food source is detritus a n d small organisms found in deeper waters. T h e main distribution of Microccrlunus lies in sub- surface waters (50 to 1200m. Buchanan & Sek- erak 1982: 200 to 500 m. S a m r o t o et al. 1986: below 1 0 0 m . Norrbin unpubl. d a t a ) . Fat stores may be an adaptation to short term food varia- bility (Hikansson 1985) rather than a means t o early maturation i n these species. The largest seasonal variation in oil sac size i s found in Pwrdocnlnrws rninutirs. the small copc- pod species which most resembles the large, her- bivorous copepods (Fig. 8B). T h e fat stores seem to be deleted during the winter, probably con- tributing both to survival and maturation (Figs. Seasonal cycles of small copepods in the Barents Sea 431 8B and 10B). During the productive season, in May, energy appears to be allocated both to matu- ration and fat stores (Fig. 8B). In P . acuspes, on the other hand, the smallest oil sacs are found in May, not in winter. At this time of abundant food, energy seems to be directed into rapid growth and maturation instead of reserves (Fig. 10A). This is consistent with information on the lipid class and fatty acid composition found for this species in Balsfjorden (Norrbin et al. 1990). In conclusion, the seasonal allocation patterns of these four species reflect other aspects of their ecology. P . acuspes is essentially a coastal species, growing quickly when food is abundant, but accumulating energy for winter. Females carry egg-sacs until hatching, implying elements of “K- strategy” (sensu MacArthur & Wilson 1967). P. minutus is even more of a “K-strategist”, an oce- anic species resembling the large herbivores, with a shorter developing season and a long resting period. Microcalanus spp. are deeper-living species relying on fat storage and detritivory. Eggs of Microcalanus are released directly into the water column (Norrbin unpubl. data). The relationship between copepod size and productive periods has been constructed from the details on stage distribution and gonad maturation in this study and is represented in Fig. 12. Further investigations into feeding habits, dev- elopment times and other aspects of the biology of Microcalanus and other little studied species are needed. The tendency to lump species together, which is often done with the smaller species, brings confusion into the interpretation of their life histories and general ecology. The fact that closely related species frequently co- occur, for example, is an observation of great ecological importance, touching upon themes such as the niche theory and hypotheses for sym- patric speciation. Acknowledgements. - I wish to thank S. Diel-Christiansen for inspiration and advice during the early part of the preparation of this manuscript and J . Marks for linguistic corrections. Thanks are also due to the crews of the Norwegian Coastguard vessels and R/V LANCE, to K.S. Tandc for access to his 1984 samples, to L. Dalsboe for assisting with the dehydration of many stained samples. to U. Norrmann for adapting the Basic digitising program to my needs, and to L. Olsen for drawing part of the figures. I am also grateful to my fellow participants in Pro Mare for cooperation and good company on the cruiscs. This study is part of the Pro Mare programme and was funded by the Norwegian Research Council for Scicnce and the Humanities (NAVF) and the Norwegian Ministry of the Environment. References Brodskii, K. A. 1950: Calanoida of the fareastern seas andpolar basin of the USSR. Keys to the fauna of the USSR. No 3 5 . Israel Program for Scientific Translations (Translated from Russian t o English, 1967). 440 pp. Buchanan, R. A . & Sekerak, A . D. 1982: Vertical distribution ofzooplankton in eastern Lancaster Sound and western Baffin Bay, July-October 1978. Arcric 35, 41-55. Cooney, R. T . & Coyle, K. 0. 1982: Trophic implications of cross-shelf copepod distributions in the southeastern Bering Sea. Mar. Biol. 70, 187-196. Corkett, C. J. & McLaren. 1. A. 1978: The biology of Pseu- docalanus. Adu. Mar. Biol. 15. 1-231. Digby, P. S . 1954: The biology of the marine planktonic cope- pods of Scoresby Sound. East Greenland. J. Anim. Ecol. 23, Fontaine. M. 1955: The planktonic copepods (Calanoida. Cyclopoida. Monstrilloida) of Ungava Bay, with spccial ref- erence to the biology of Pseudocalanus minutus and Calanus finmarchicus. J. Fish. Res. Ed. Can. 12. 858-898. Frost, B. 1989: A taxonomy of the marine calanoid copepod genus Pseudocalanus. Can. J. Zool. 67, 525-551. Grainger. E. H . 1959: The annual oceanographic cycle at Igloolik in the Canadian Arctic. I . The zooplankton and physical and chemical observations. J. Fish. Res. Ed. Can. 16. 453-501. Hassel. A. 1986: Seasonal changes in zooplankton composition in the Barents Sea. with special reference to Calanus spp. (Copepoda). J. Plankton Res. 8. 32S339. Hopkins, C. C. E., Tande, K. S.. Gronvik, S. & Sargent. J. R. 1984: Ecological investigations of the zooplankton com- munity of Balsfjorden, northern Norway: An analysis of growth and overwintering tactics in relation to niche and environment in Metridia longa (Lubbock), Calanus fin- marchicus (Gunnerus). Thyssanoessa inermis (Kroyer) and 7. ruschi (M. Sars). 1. Exp. Mar. Biol. Ecol. 82, 77-99. Horner, R. & Murphy. D. 1985: Species composition and abundance of zooplankton in the near-shore Beaufort sea in winter-spring. Arctir 38. 201-209. Hikansson, J . L. 1985. Thelongandshortterm feeding condition infield-caught Calanus pacificus. as determined from the lipid content: relationship with phytoplankton pigments in the Cali- fornia current. Dissertation, Univ. of Gothenburg. 92 pp. MacArthur. R. H . & Wilson, E. 0. 1967. The theory of island biogeography. Princeton University Press, Princeton. 203 pp. McLaren, I . A. 1969: Population and production ecology of zooplankton in Ogac Lake, a landlocked fiord on Baffin Island. J. Fish. Res. Ed. Can. 26. 1485-1559. McLaren, I. A . 1974: Demographic strategy of vertical migration by a marine copepod. Am. Nut. 108. 91-1U2. Minoda. T . 1967: Seasonal distribution of Copepoda i n the Arctic ocean from June to December. 1964. Records of Oceanographic Works in Japan 9 ( 1 ) . 161-168. Norrbin, M. F. 1987: Overwintering and energetic strategies of marine pelagic copepods in high latitudes. M S c . thesis. University of Gothenburg, Gothenburg. Sweden. 101 pp. Norrbin, M. F., Olsen. R.-E. & Tande. K . S. 1990: Seasonal variation in lipid class and fatty acid composition of two small copepods in Balsfjorden. northern Norway. Mar. Biol. 105. 205-211. Razouls, S . . Nival, S. & Nival, P. 1987: Development of the genital system in the copepodid stages of the calanoid cope- pod Temora srylifera Dana. J . Mar. B i d . Ass. U . K . 67.653- 661. 296338. 432 M. Fredrika Norrbin Sameoto. D.. Hermann. A . & Longhurst. A . 19x6: Relations between the thermoclinc meso and microzooplankton. chlorophyll a and primary production distributions in Lancas- ter Sound. Polar. Biol. 6. S%61 Sars. G . 0. IW3: An account of the Crustacea of Norway. Vol. I V . Copepoda. Calanoidu. Bergen Museum. Bergen. Tande. K. S. & G r ~ n v i k . S. 1983: Ecological investigations on the zooplankton community of Balsfjorden. northern Norway: sex ratio and gonad maturation cycle in the copepod Merridia longa (Lubbock) 1. Exp. Mar B i d €col. 71. 4% 54. Tande. K . S . . Hassel. A . & Slagstad. D. 1985: Gonad matu- ration and possible life cycle strategies in Calanus fin- marrhirus (Gunnerus) and Calanusglacialrs (Jaschnov) in the northwestern part of the Barents Sea. Pp. 141-155 in Gray. J. S. & Christensen. M . E (eds.): Marine biology of polar regions and the effect of stress oil marine organisms. John Wiley & Sons. Chichcstcr. Tande. K. S . & Hopkins. C . C . E . 1981: Ecological invcs- tigations of the zooplankton community of Balsfjorden, Northern Norway: the genital system in Calanmfinmarchicus and thc role of gonad dcvclopment in thc overwintering strategy. Mar. B i d . 63. 15F164. Unesco 1%8: Zooplankton sampling. lmprimeries Populaircs. Genkve. 174 pp. Ussing. H. H. 1938: The biology of some important plankton animals in the fjords of East Greenland. Medd. Cr@nland loo. 1-108. Vidal. J . 1971: Taxonomic guides to arctic zooplankton ( I V ) : K e v to the calanoid copepods of the central Arctic Ocean. University of Southern California Tech. Rep. No. 5 . 128 pp. Vidal. J. & Smith. S . L. 1986: Biomass. growth. and dev- elopment of populations of herbivorous zooplankton in the southeastern Bering Sea during spring. Deep-Sen Res. 33. 523-556.