In December/February 1990/91 a survey was conducted by RV Kaiyo Maru in the vicinity of the South Shetland Islands and Elephant Island. The main objective of the survey was to investigate the mechanisms responsible for the concentration of krill and to estimate krill biomass in the area surveyed. Krill biomass estimation was done by means of hydroacoustics. Phytoplankton distribution was assessed by measuring surface concentrations of chlorophyll a. Water circulation was studied with satellite-tracked buoys equipped with curtain drogues deployed at a level of 30 m from the surface because at this level krill concentrations are most frequently observed. Main krill concentrations were observed in shelf waters north of the islands as usual. Tracking of buoys demonstrated the existence of convergent complex eddies in shelf waters of these islands. A high concentration of chlorophyll a was also recorded in shelf waters. A spatial correlation between distributions of krill and chlorophyll a was observed. It is therefore considered that both hydrodynamic and food availability factors may be responsible for krill concentration. On a traditional fishing ground north of Livingston Island, krill density had increased 3.4 times over a period of 40 days from late December 1990 to early February 1991. This increased krill density in early February 1991 (157 g/m²) was 54% lower than that observed in late January 1988. Information from studies of krill predators by US scientists also indicated that krill abundance during the first half of the 1990/91 summer was lower than that in previous seasons. Total biomass of krill in shelf waters of the area surveyed was estimated at 1.78 million tonnes (56 g/m²) in late January 1991. The obtained estimate of krill biomass should be regarded as the lower level for this time of the season. From mid-February and onwards it was reported that krill abundance had increased to the normal level.