Euphausia superba Dana was collected at stations encompassing 3" of latitude in icecovered waters west of the Antarctic Peninsula in winter of 1992 (July-August), and in the same region the previous summer (December-January). There was no significant change in any biometric index (dry weight vs length, carbon vs dry weight, nitrogen vs dry weight, or body C:N). Rates of ammonium excretion in winter were not significantly different from those previously reported for E. superba in summer. It was clear that E. superba fed as a carnivore in winter. Copepod body parts consistently occurred in M 1 guts. During winter shipboard experiments, freshly caught krill ingested copepods and produced faeces at rates comparable to those measured in summer. The low ratio of N:P excretion (2.02) also suggests carnivorous feeding. A physiological model of carbon and nitrogen metabolism indicates that E. superba must feed during winter to maintain its observed C : N ratio and rate of ammonium excretion. On a diet of less than 1 Metridia gerlachei female copepod per day, a 20 mm knll can grow with a growth efficiency comparable to that achieved in summer. We suggest that starvation, body shrinkage and reduced metabohsm are unusual behaviors for knll in winter, and that carnivory sustains growth until ice algae become available as food in late winter and early spring.

Calanoldes acutus (Giesbrecht) in 0 to 290 m stratified samples collected repetitively at 23 stations in Gerlache Strait, Antarctica, in November 1989 consisted of stage 111 to V1 copepodites (C111 to CVI) All stages except CV decreased in abundance through the 3.5 wk sampling period. Vertical distnbutions of all copepodite stages remained consistent, with no evidence of die1 vertical migrahon. Centers of mass were as follows: female CVl, 44 m; CV, 66 m; CIV, 127 m; male CVI, 189 m. Progressive ovarian maturation in adult females was not correlated with ambient chlorophyll a concentrations. Mean daily egg production at 16 stations over a 3 d period in late November was a hyperbolic function of 0 to 150 n~ integrated chlorophyll a concentrations.