SHORT COMMUNICATION
A second examination of predation on pelagic copepods by the brittle star Astrotoma agassizii
Frank D.Ferrari and John H.Dearborn'
Smithsonian Oceanographic Sorting Center, Museum of Natural History, Smithsonian Institution, Washington, DC 20560 and' Department of Zoology and Center for Marine Studies, University of Maine, Orono, ME 04469, USA
I- Abstract. Euchaeta antarctica and Catanoides acuius, common pelagic copepods of the Southern
^ Ocean, dominate stomach contents of the brittle star Astrotoma agassizii from the South Georgia shelf in November-December. This finding does not support our earlier hypothesis of restricted H feedmg by Astrotoma in austral fall on these copepods during their seasonal migration.
Previously we reported on the feeding biology of Astrotoma agassizii, a large Antarctic brittle star whose prey in March-June (austral fall) is dominated by pelagic calanoid copepods (Dearborn et al., 1986). Astrotoma agassizii is found throughout the Southern Ocean in depths of -70-1000 m and occurs irregularly on the shelves of sub-Antarctic islands and the Antarctic continent. Its disc diameter can exceed 60 mm and long, unbranched arms can reach 700 mm in length. The arms are extremely flexibile and can be extended, coiled tightly or held in various sinuous positions. This brittle star frequently climbs off the substrate and clings to other organisms such as sponges, hydrocorals, gorgonians or bryozoans; one or two arms are used to hold its position while the other arms are extended for feeding.
Stomach contents of animals collected in March/April along the Antarctic Peninsula and May/June around South Georgia were dominated numerically by pelagic calanoid copepods, with a carnivore Euchaeta antarctica (56%) and a herbivore Calanoides acutus (23%) comprising almost 80% of copepod prey (Dearborn et al., 1986). Copepodids V and VI were the dominant stages. We suggested that this feeding resulted in carbon transfer between pelagic and I benthic ecosystems, and further, that the presence of this combination of calanoid copepods in brittle star stomachs resulted from near-bottom aggreg- i- ations of lipid-rich C.acutus and its predator, the midwater E.antarctica, during
, the former's seasonal, downward migration to deep waters.
Cruise 8601 in 1986 of the Polish research vessel Professor Siedlecki to survey nototheniid ground-fishes around South Georgia, and funded by the United States Antarctic Marine Living Resources Program, provided us with an opportunity to determine prey identities oi A.agassizii in early austral spring and to test the hypothesis of its episodic feeding on C. acutus and E.antarctica mediated by the former's austral fall migrations. Brittle stars were collected using a beam trawl with nominal mouth width of 17.5 m and estimated sweep area of 70 m^; the forward half of its 24-m-long cod end was fitted with 51.5-mm
Stretch mesh, the aft half with 42.9-mm mesh. SampUng data are shown in Table 1. Bottom temperatures varied from 6 to 13°C, with salinities from 34.16%c to 34.40{)%o. Aboard ship, benthic invertebrates were fixed and temporarily preserved in 10% formalin/90% sea water.
Brittle star disc diameters were measured to the nearest 0.1 mm prior to dissection, and gut contents were analyzed following the procedures of Fratt and Dearborn (1984) and Dearborn et al. (1986). Presence or absence of food provided a measure of the relative number of brittle stars feeding. Frequency of occurrence of individual food items was expressed as a percentage of feeding brittle stars. Each stomach was assigned an estimated overall fullness value ranging from 0 (empty) to 16 (maximum fullness), with intermediate values of 1, 2, 4, 8 and 12, following the method of Fratt and Dearborn (1984) and Dearborn et al. (1986). Food items in each gut were assigned a point value from the same scale according to their estimated volume contribution. Results were expressed as the proportion of total points awarded to a particular item compared to the points assigned to all food items (percentage composition by relative volume).
Finally we multiplied the point value of each food item by its fuUnes index to provide a fullness relative volume for each item, an index we consider more useful in describing contributions of various prey to brittle star diet.
Summary statistics from analyses of the stomach contents of 63 A.agassizii are presented in Table II. Mean disc diameter of the brittle stars was 46.2 mm;
71.4% of all animals contained at least one food item. The mean stomach fullness value for animals containing food was low (1.4). Table III lists prey items, occurrences and percentage composition by relative volume and fullness relative volume. For all samples copepods were the dominant food in Astrotoma stomachs, accounting for 90% relative volume and 86.5% fullness relative volume. Unidentified crustacean remains, chaetognaths and larvaceans each occurred with 4.4% frequency of occurrence, although their contribution to total food bolus by relative volume and fullness relative volume varied.
One hundred and thirty-four copepods were identified from stomachs of the brittle stars, and are listed by stage and sex in Table IV. All are calanoids and none have been reported associated with the benthic interface. Many were in good condition with unbroken exoskeletons and undigested musculature.
Calanoides acutus was most abundant; copepodid stage V (CV) dominated.
Euchaeta antarcdca was second in abundance, with CV dominant. These two species with Rhincalanus gigas comprised 90% of copepod prey.
Astrotoma captured planktonic calanoids <3 mm (CVI Drepanopus forci- patus, cm E.antarctica); the largest measurable copepodids captured were
>7 mm (CV, CVI R.gigas and CV E.antarctica). Results of this and our previous study, in which CVI E.antarctica were common prey, suggest that these brittle stars can capture copepods up to 10 mm, and sUghtly larger prey as well, because some juvenile mysids and euphausiids were taken by Astrotoma in austral fall.
Percentages of feeding Astrotoma in our austral fall and spring studies were similar, 67.8 and 71.4% respectively. Although there was much variation between collections from individual stations, the mean percentages of feeding
M c u o .1
"o
6 a
2 D.
J=
c H.
i
•a^ jz
«! a.
S •o o
J=
T3 C D.
-T -^ r- ON a^ O CM rt (N
2 8-
c/5 -a
o
c
D. ID
•a _aj o
c/1 *a
Ov -^ 00 fN a\ ^
(^ •— fNJ
B
E cd lyj n a>
1
ea Q. e a 6H in c
2 ^ S5
O w-i O f*^ CN rn
"O O lA) Q w-1 m
^ r- ON
« O O
^O -o ^O 00 00 oc 0\ ON ON
3S
1317
Table 11. Summary statistics from stomach analyses of 63 Astrotoma agassizii, all stations Disc diameter range (mm)
Mean size (mm)
Percentage of animals feeding
Percentage of feeding animals containing more than one food type Mean number of food types per feeding animal
Mean stomach fullness
Mean stomach fullness (feeding animals only)
10.1-64.5 46.2 71.4 8.9 1.1 1.0 1.4
Table III. Summary of stomach contents of 45 Astrotoma agassizii containing food, all stations Food item
Copepoda 43 Unidentified crustaceans 2 Chaetognatha 2 Larvacea 2 Unidentified debris 1
No. of Percentage frequency Percentage composition by occurrences of occurrence Relative Relative
volume fullness volume 95.6
4.4 4.4 4.4 2.2
90.0 86.5
4.4 3.2
2.5 4.8
1.4 3.2
1.7 2.4
Table IV. Identified copepodids
Name Sexed (1 [emale/mals :) lotal
copepodid stage Unsexed copepodid stage
CVI CV CIV CV CIV
cm
Euchaeta antarctica 4/0 13/8 12/4 4 45
Calanoides acutus 4/0 47 4 55
Rhincalanus gigas 1/5 4/8 1/0 19
Drepanopus forcipatus 8/0 8
Pleuromamma robusta 7/1 8
Calanus propinquus 1 1
Calanus simillimus 0/1 1
Gaetanus tenuispinus 1/0 1
Total 138
animals containing more than one food type were quite different (55.5% in fall, 8.9% in spring). This suggests that although in both seasons copepods were dominant, brittle stars in fall were taking a greater variety of prey; stomachs of some fall Astrotoma contained ostracods, mysids, amphipods and euphausiids not represented in spring material. However, mean stomach fullness values of 1.8 and 1.4 for fall and spring are low, suggesting that Astrotoma probably feeds more or less continuously rather than taking a large volume of prey over a short period of time. No data are presently available on digestion rates for Astrotoma.
Data from May/June 1975 off South Georgia (Dearborn et al., 1986, Table 5) indicate that E.antarctica (56%), C.acutus (23%) and D.forcipatus (10%) comprised almost 90% of that prey group. Comparisons of percentages for the three most abundant prey species in spring with their fall percentages indicate decreased relative abundance of E.antarctica and D.forcipatus with increased
C.acutus and R.gigas. Changes in percentages of copepodid stages are distinctive only for E.antarctica. Percentage of CV is similar, but there is a relative increase in CIV (2 to 36%) and a decrease in CVI (57 to 8%) in spring.
Marin (1986) has provided the most complete description of the changing seasonal distribution of C.acutus in open ocean waters. He reports mating in deeper waters during late winter to early spring followed by upward migration of fertilized females to surface waters, where eggs are shed. Progressively deeper dispersion accompanies ontogenetic development and results in a protracted downward migration. However, dominance of CV animals in Astrotoma stomachs in November-December does not reflect Marin's stage distributions at the northern edge of the Weddell Sea Gyre where CIII and CIV were abundant (Mann, 1986, figure 30). Comparisons between summer and winter distributions of C.acutus over deep waters around South Georgia point to its seasonal population shift from above 250 m in summer to 250-500 m in winter (Atkinson and Ward, 1988). Over shelf waters of South Georgia in winter C.acutus was relatively more abundant (Atkinson and Peck, 1989) than in adjacent, epipelagic, offshore waters (Atkinson and Peck, 1988).
An increasing amount of information now is available about the biology of E.antarctica. Reports from oceanic stations show E.antarctica with a widespread distribution throughout the Atlantic and Pacific sectors of the Southern Ocean (Ferrari and Dojiri, 1987; Fontaine, 1988). Data from Hopkins (1985a) indicate open ocean vertical occurrences between 200 and 1000 m, with some upward dispersion at night. Marin and Antezana (1985) extended its distribution beyond the Southern Ocean to shallow waters of Chilean fjords in early austral spring.
In waters around South Georgia its distribution is centered at 250-500 m. It is the most common euchaetid encountered above 250 m, and it is the most abundant euchaetid found in waters over the South Georgia shelf (Ward and Wood, 1988). Hopkins (1985b, 1987) reported copepodids of a calanoid, Metridia gerlachei, and poecilostomes, Oncaea spp., as well as unidentified copepods in guts of E.antarctica. Metridia gerlachei, ~3.5 mm (Vervoort, 1957), and the common Southern Ocean Oncaea spp., <2.0 mm (Heron, 1977), are both much smaller than CV C.acutus.
It is clear from these data, particularly the continued presence of E.antarctica and C.acutus in stomachs of A.agassizii during November-December, the protracted downward migration of C.acutus reported by Marin and the presence of copepodids smaller than CV C.acutus in guts of E.antarctica reported by Hopkins, that the dominance of E.antarctica and C.acutus as prey items of A.agassizii in austral spring did not result from the seasonally episodic downward migration of the pelagic herbivore and its predation by the pelagic carnivore as we previously suggested (Dearborn et al., 1986). Analysis of A.agassizii from winter months will determine if this pelagic-to-benthic trophic link is seasonally continuous.
Acknowledgements
We thank Dr Kenneth Sherman, National Marine Fisheries Service, Narragansett, RI for inviting participants from the Smithsonian Oceanographic
Sorting Center on the R/V Professor Siedlecki, and Mr Jerome Finan and Mr T.Chad Walter for collecting the brittle stars. Mr Peter Ward, British Antarctic Survey, kindly reviewed the manuscript.
References
Atkinson.A. and Peck.J. (1988) A summer-winter comparison of zooplankton in the oceanic area around Soutli Georgia. Po/arBio/., 8. 463-473. . , . ., c ,u n^^mi, Atkinson A and Peck.J. (1989) The distribution of zooplankton m relation to the South Georgia sheif in summer and winter. Proceedings Fifth SCAR Symposium On Antarcic Biology, in press_
Atkinson.A. and Ward,?. (1988) Summer-winter differences in copepod distribution around South Georgia. Hydrohiologia, 167/168, 325-334. . u- j i„
Dearborn,:. (1977) Foods and feeding characteristics of Antarctic asteroids and ophmro.ds. In Llano,G. (ed.). Adaptions Within Antarctic Ecosystems, Proceedings of the Third StAK Symposium On Antarctic Biology. Washington, DC, pp. 293-326.
Dearborn,J., Ferrari,F. and Edwards,K. (1986) Can pelagic aggregations cause benthic satiation.' Feeding biology of the Antarctic brittle star Aslrotoma agassizii (Echmodermata:Ophiuroidea;.
Antarct. Res. Ser.,M,\-2%. c, .u n ^„„
Ferrari F and Dojiri,M. (1987) The calanoid copepod Euchaeta antarctica from Southern Ocean Atlantic sector midwater trawls, with observations on spermatophore dimorphism. J. Crustac.
BIO/., 7, 458-480. , ,,
Fontaine,M. (1988) The taxonomy and distribution of the antarctica species group ot the genus
£MC/I(JW(I (Copepoda, Calanoida)./lma/-c(. /?«. Ser., 47, 27-57.
Fratt.D. and Dearborn,!. (1984) Feeding biology of the Antarctic brittle star Ophionotus victonae (Echinodcrmata: Ophiuroidea). Po/ar BIO/., 3, 127-139. , . u , Heron,G. (1977) Twenty-six species of Oncaeidae (CopepodaiCyclopoida) from the southwest Pacific Antarctic area. Antarct. Res. Ser., 26, 37-96. , n / Hopkins,T. (1985a) The zooplankton community of Croker Passage, Antarctic Peninsula. Polar Hopkins f (1985b) Food web of an Antarctic midwater ecosystem. Mar. Biol., 89, 197-212.
Hopkins.T. (1987) Midwater food web in McMurdo Sound, Ross Sea, Antarctica. Mar BioL, 96, 93-106. , .^ _ , Marin V (1986) Distribution and life cycle of three antarctic copepods {Calanoides acutus, Calanus
propinquus and Rhincalanus gigas). Ph.D. dissertation. University of California, San Diego, CA.
Marin,V. and Antezana,T. (1985) Species composition and relative abundance of copepods in Chilean fjords./. /'/aw/cion Rev., 7, 961-966.
Vervoort.W. (1957) Copepods from Antarctic and sub-Antarctic plankton samples. Reports of the British Au.iiralian, and New Zealand Antarctic Research Expedition. 1929-1931. (ser. B). Vol. 3^
Ward,P. and Wood,A. (1988) The distribution of the Euchaetidae (Copepoda:Calanoida) around South Georgia with notes on their reproduction. Polar BioL, 9, 45-52.
Received on January 11, 1989; accepted on July 4, 1989
1320
