Over 50 000 individual dive records collected by time-depth recorders were analysed with respect to sex of the seal, time of year and the approximate geographic location of the dive. Six distinct dive types were described on the basis of parameters such as the amount of time spent at the maximum depth of the dive, the rate of ascent and descent, and the general form of the dive profile. These dive types were 'rest' dives, 'travel' dives, 'surface' dives, 'general non-foraging' dives, 'pelagic foraging' dives and 'benthic foraging' dives. The seals spent 90% of their time at sea submerged. Less than 2% of the time was spent on the surface in intervals of more than 10 min. A further 20-30% of the time was spent on the various non-foraging types of dives. Most females performed only 'pelagic foraging' dives, while males performed both 'pelagic' and 'benthic foraging' dives. All the 'benthic foraging' dives occurred in Area 3 (defined by water-temperature data as lying over the Antarctic Continental Shelf) and were 400-500 m deep. 'Pelagic foraging' dives occurred in all three foraging areas and ranged in depth from 200 to 1100 m. These types of dives also exhibited marked diurnal variations in depth, unlike 'benthic foraging' dives. The seals spent 10-20 min at the bottom of each 'foraging' dive, where they generally displayed a series of small changes in depth (wiggles). The size of these 'wiggles' tended to be larger in 'pelagic foraging' dives than in 'benthic foraging' dives. The diving behaviour of southern elephant seals is related to the possible prey they exploit in the Southern Ocean.

Southern elephant seals (Miroungo leonina) were immobilised with a mixture of tiletamine and zolazepam administered intravenously at a mean (sd) dose rate of 0.46 (0.08) mg/kg. This dose provided a satisfactory degree of anaesthesia with no side effects, and the induction, duration and recovery times were short. The mean (sd) induction time was 26 (9) seconds and the mean level of anaesthesia was 4.4 units on an eight-point scale. Male seals were given less drug than female seals, remained immobilised for shorter periods and recovered sooner. The mean (sd) dose of drug administered to males was 0.44 (0.06) mg/kg and to females 0.48 (0.08) mg/kg, and the mean (sd) duration times were 14.9 (4.5) minutes and 16.1 (5.3) minutes. The mean (sd) time taken to recover from immobilisation was 14.5 (4.6) minutes for males and 15.7 (5.3) minutes for females. Physiological condition and size significantly affected the duration of anaesthesia. Thin seals remained immobilised for 18 (7) minutes whereas fatter seals remained immobilised for 15 (4) minutes (P<0.0001).

The diving behaviour of 14 adult southern elephant seals was investigated using time depth recorders. Each of the seals performed some dives that were longer than its theoretical aerobic dive limit. Forty-four percent of all dives made by post-moult females exceeded the calculated limit compared with 7% of those made by postbreeding females and less than 1% of those made by adult males. The extended dives displayed characteristics that suggested that they were predominantly foraging dives, although some were apparently rest dives. Dives longer than the calculated aerobic limits often occurred in bouts; the longest consisted of 63 consecutive dives and lasted 2 days. Postmoult females performed longer bouts of extended dives than postbreeding females. Extended surface periods (longer than 30 min) were not related to the occurrence of extended dives or bouts of extended dives. The possible physiological mechanisms that permit such prolonged continuous dives are discussed. Southern elephant seals may increase the aerobic capacity of dives by lowering their metabolism to approximately 40% of the resting metabolic rate on long dives. There is substantial interseal variability in the methods used to cope with long dives. Some animals appear to use physiological strategies that allow them to prolong the time available to them at the bottom of a dive, while others use alternative strategies that may limit the time available at the bottom of their dives.

Stomach contents were lavaged from 76 southern elephant seals (Mirounga leonina) at Heard Island between July 1992 and March 1993. Eighty-six percent of stomachs contained cephalopods of 17 species. Numerically the most important was Psychroteuthis glacialis (21.1%), and from estimated biomass the most important was Kondakovia longimana (40.4%). Three other species were also common prey: Moroteuthis knipovitchi (19.4% by estimated biomass), Moroteuthis ingens (13.0%), and Alluroteuthis antarcticus (10.2%). Sixty-six percent of stomachs contained fish remains, and four species, Dissostichus eleginoides, Electrona carlsbergi, E. antarctica, and Gymnoscopelus nicholsi, were identified from otoliths. The diet of adults differed from that of juveniles, particularly pups in their first year. Martialia hyadesi was the most important prey of juveniles and represented 57.1% of estimated biomass consumed. Furthermore, smaller seals ate smaller squid. The species and size of cephalopods eaten by southern elephant seals are similar to those of other Southern Ocean predators, particularly some beaked whales.

Determining where, when and how much animals eat is fundamental to understanding their ecology. We developed a technique to identify a prey capture signature for little penguins from accelerometry, in order to quantify food intake remotely. We categorised behaviour of captive penguins from HD video and matched this to time-series data from back-mounted accelerometers. We then trained a support vector machine (SVM) to classify the penguins' behaviour at 0.3 s intervals as either 'prey handling' or 'swimming'. We applied this model to accelerometer data collected from foraging wild penguins to identify prey capture events. We compared prey capture and non-prey capture dives to test the model predictions against foraging theory. The SVM had an accuracy of 84.95±0.26% (mean ± s.e.) and a false positive rate of 9.82±0.24% when tested on unseen captive data. For wild data, we defined three independent, consecutive prey handling observations as representing true prey capture, with a false positive rate of 0.09%. Dives with prey captures had longer duration and bottom times, were deeper, had faster ascent rates, and had more 'wiggles' and 'dashes' (proxies for prey encounter used in other studies). The mean (±s.e.) number of prey captures per foraging trip was 446.6±66.28. By recording the behaviour of captive animals on HD video and using a supervised machine learning approach, we show that accelerometry signatures can classify the behaviour of wild animals at unprecedentedly fine scales.