The feeding behavior of antarctic knll Euphausia superba on ice algae was observed in situ and in the laboratory. Field observations by divers confirm that knll uLhze natural sea ice microalgae for food. Laboratory investigations show that melting ice releases algae into the water column which induces area-intensive foraging behavior in M1. This behavior is characterized by high speed swimming and rapid turning, accompanied by rapid opening and closing of the thoracic appendages, also known as the feedmg basket. Presentation of increased concentrations of ice algae to laboratory populations of knll significantly increased euphausiid responsiveness which led to location of and direct grazing upon the undersurfaces of ice containing microalgae. Foraging behavior of krill on ice algae appears to be affected by the spatial patchiness of the algae withln the ice and on the rate of algal cell release from ice. We propose that sea ice algae is an abundant and predictable food resource for knll during austral winters, when phytoplankton food resources are depleted.

Juvenile Chironex fleckeri medusae were maintained in aquaria for several months. One individual was raised over a nine-month period to subadult condition for the first time. The medusae did not feed naturally in most aquaria but they accepted and digested prey items placed by hand onto the manubrium. Medusae maintained in planktonkreisels, however, extended tentacles and captured and ingested live Acetes, large prawns and fish by subumbrellar flexing of the pedalia. Digestion of prey was rapid and food particles were circulated directionally through functional canals and lacunae primarily by contractions of the bell but also by peristaltic contractions of interradial gastrovascular tissues. In the laboratory, medusae visually reacted to dark objects by swimming away from them. Swimming behaviour of medusae in the sea and natural predation by green turtles (Chelonia rnydas) on C. fleckeri are described.

Eight of the 10 photophores of the Antarctic krill, Euphausia superba, are located at the ends of muscular stalks and exhibit coordinated orientation responses to incident white light; light emitted from the photophores is directed away from the incident light. Moreover, eye rotation occurs synchronously with photophore movement. Immobilization of one or both eyes eliminated the photophore light-following response in 40% of the trials, but in the remaining 60%, photophores continued to exhibit oriented, but less stable responses. In the presence of a stationary light source the eyes could be passively rotated without affecting photophore position. Furthermore, eye removal or covering the head with an opaque hood eliminated coordinated photophore movement. We conclude that vision is necessary for light-following responses by the photophores. In addition, the control signal for that movement is CNS-derived, may occur spontaneously or may be lightinduced, and appears to be accompanied by a parallel signal governing eye rotation. Subtle differences in photophore response when krill were oriented other than horizontally imply that krill may have a gravity sense that could help them orient in darkness.