Lake Untersee is the largest freshwater lake in the interior of East Antarctica. It is a perennially ice-covered, max. 169 m deep, ultra-oligotrophic lake. In contrast to earlier studies, we found clear evidence for physical and chemical stratification in the summer of 1991–92. However, the stratification was restricted to a trough, c. 500 m wide and up to 105 m deep, in the south-western part of the lake. There, the water body was distinctly stratified as indicated by sharp vertical gradients of temperature, pH, dissolved oxygen, and electrical conductivity. The water column was anoxic below 80 m. The chemical stratification is also indicated by changes of ionic ratios. Moreover, there was some evidence for methanogenesis and bacterial sulphate reduction in Lake Untersee.

Patchiness in the horizontal distribution of sympagic organisms was studied at an Antarctic coastal site during autumn. A hierarchical sampling design (nested ANOVA) was used to assess variation in the biota on scales from metres to kilometres. Metazoan abundance, chlorophyll concentration and salinity were measured in 54 sea ice cores. The metazoan fauna was dominated by nauplii of the copepod Paralabidocera antarctica (6 × 104 to 4 × 105 m−2). Other copepods present included Stephos longipes, Oncaea curvata, Oithona similis, Ctenocalanus citer, and unidentified harpacticoid copepods. Chlorophyll a concentrations were generally much higher than values recorded at other sites at the same time of the year, reaching a maximum of 78 mg m−2. Metazoan abundances did not correlate strongly with chlorophyll or salinity. Significant variability in abundance of P. antarctica and O. similis, and chlorophyll concentration occurred at the scale of kilometres, whereas salinity and other metazoan abundances were not significantly variable at any of the scales examined. Considerable variation was evident at scales of less than one metre.

Human contamination of Antarctic environments is a sensitive issue and has been the focus of many research articles over the past 35 years. The majority of these studies have targeted waste materials and various hydrocarbons, with assessment of microbial contaminants being largely restricted to sewage outfalls. The present study aimed to detect bacteria of human origin in the area surrounding Halley research station. It was apparent from both molecular and culture methods that bacteria of human origin are extremely difficult to detect outside the immediate surrounding of the buildings, though recommendations are made for increasing the probability of determining the presence of organisms in the environment. The results also indicate that molecular methods are more sensitive than cultural techniques, in that the only evidence for organisms in the environment surrounding the buildings came from positive PCR reactions. PCR would appear to be a useful method for studying the microbial ecology of Antarctic environments.

The stomach contents of demersal fish in late January 1982 were analysed. Samples were taken at 100, 300 and 500 m depth south of Elephant Island, Bransfield Strait and north of Livingston Island, and at 800 m to the east of Smith Island. Fifty four taxa of fish belonging to 11 families were collected. The diets of 2101 fish representing 38 taxa were examined. These were classified into three categories, fish feeders, krill feeders and benthos feeders. Fish prey species fed on krill and/or benthos. Krill was a major dietary component for 32 (84.2%) out of 38 taxa. Gobionotothen gibberifrons was distributed at all 10 stations (100–800 m in depth) and its diet comprised krill and benthos. The present findings verify the importance of krill in the Antarctic marine ecosystem and indicate that krill is consumed by benthic fish at greater depths than previously reported.

Magnetic, gravity and bathymetric data from Marguerite Bay are used to study the relationships between oceanic and continental tectonic evolution in the arc and fore-arc of the Antarctic Peninsula. The data are used to redefine the crustal structure of the Marguerite Bay area, providing evidence for a northward continuation of George VI Sound and the Alexander Island Mesozoic accretionary prism almost to the continental shelf edge. A two-stage model of extension, associated with changes in spreading rates and approaching ridge segments, is proposed to explain the crustal structure and Cenozoic tectonic evolution of the area. The model involves the opening of George VI trough by Tertiary dextral transtension, followed by oblique extension in an area bounded by the Tula and Adelaide fracture zones. This interpretation confirms previous work linking oceanic tectonic processes with continental arc and fore-arc structural development.

Rates of tag-loss are determined for Dalton Jumbo Rototags applied to the hind flippers of 4343 (2208 males, 2135 females) southern elephant seal (Mirounga leonina) pups at Marion Island over an eight year period from 1983–1990 as part of a demographic study of the species. Loss rates were the lowest recorded to date for this species (range 0.0–9.1%). No significant relationship existed between age and rate of tag-loss, neither was there any sex or year related differences in age-specific tag-loss rates. The low rates of loss highlight the value of tagging as a marking technique, and allow for high levels of confidence in the reliability of the population parameters that are derived from the tagging data collected for the Marion Island population.

This investigation reports the first simultaneous measurement of methanesulphonate (MSA) and non-sea salt sulphate (NSSS) in aerosols, surface snow, and ice core samples for a continental site in Antarctica (78°S, 139°E, elevation 2849 m). Aerosol MSA concentrations ranged from 0.09–0.43 nmol m−3 STP (median 0.14 nmol m−3) and were generally lower than those observed at coastal Antarctic sites. NSSS concentrations varied from 0.66–1.32 nmol m−3 stp (median 0.88 nmol m−3), comparable to those reported for other continental Antarctic locations. Whereas the MSA:NSSS molar ratio in aerosol samples was in the range 12.7–32.5% (median 17.0%), the ratio down a snow pit and ice profile varied from 1.14–55.6% (median 3.50%), reflecting the variability to be expected over a period of a decade. The chemical composition and low MSA content suggests an origin of aerosols consistent with long range transport from mid-latitudes.

The Permian–Triassic boundary within the Amery Group of the Lambert Graben is placed at the contact between the Bainmedart Coal Measures and overlying Flagstone Bench Formation, based on the first regular occurrence of Lunatisporites pellucidus and the first appearance of Aratrisporites and Lepidopteris species. The Permian-Triassic boundary is marked by the extinction of glossopterid and cordaitalean gymnosperms, and by the disappearance or extreme decline of a range of gymnospermous and pteridophytic palynomorph groups. Earliest Triassic macrofloras and palynofloras of the Flagstone Bench Formation are dominated by peltasperms and lycophytes; corystosperms, conifers, and ferns become increasingly common elements of assemblages through the Lower Triassic part of the formation and dominate floras of the Upper Triassic strata. The sedimentary transition across this boundary is conformable but marked by a termination of coal deposits; overlying lowermost Triassic sediments contain only carbonaceous siltstones. Typical red-bed facies are not developed until at least 100 m above the base of the Flagstone Bench Formation, in strata containing ?Middle Triassic palynofloras. Across Gondwana the diachronous disappearance of coal deposits and appearance of red-beds is suggestive of a response to shifting climatic belts, resulting in progressively drier seasonal conditions at successively higher palaeolatitudes during the Late Permian to Middle Triassic. The abrupt and approximately synchronous replacement of plant groups at the Permian–Triassic boundary suggests that factors independent of, or additional to, climate change were responsible for the turnover in terresrtial floras.

A single specimen of palinurid lobster, possibly Linuparus, from Pliocene rocks on Marine Plain, Princess Elizabeth Land, documents the only post-Miocene occurrence of decapod crustaceans in the Antarctic. The badly crushed, partial remains, probably an exuvia, are encrusted with the serpulid worm, Spirorbis Daudin.

Octachlorostyrene (OCS) and trans-nonachlor (TNC) were detected in cape petrels (Daption capense) of King George Island, which tallies with their presence in samples of gentoo penguins (Pygoscelis papua) of the Falkland Islands. The detection of TNC in a sample of the Antarctic southern fulmar (Fulmarus glacialoides) implies that the Antarctic region has been contaminated by this compound. Mirex and photomirex were also detected in samples of the cape petrels and southern fulmar, as well as in Adélie penguins (Pygoscelis adeliae) from Hop Island. The ratios of the mirex and photomirex concentrations in the truly Antarctic species from different locations are similar, which suggests that these compounds are diffusely distributed over the continent. The detection of organochlorine pollutants in Antarctic seabirds is an indication that these compounds have a global distribution. Screening of subcutaneous fat of Antarctic seabirds for organochlorines provides an excellent indication of the occurrence of organochlorine pollutants in Antarctica, and as such an ‘early warning’ for the global dispersion of these compounds.

For the first time complete swath bathymetric maps of central and eastern Bransfield basins are presented at a medium scale and detailed morphological descriptions are provided. Bathymetry reveals morphological structures which provide important information about the structure, volcanism, and kinematics of these basins. The central basin is dominated by two roughly orthogonal sets of extensional faults (N065° the main set and N145° the secondary set), displays several discontinuous along-axis large volcanic cones and ridges trending about N059°, and undergoes extension roughly normal to the basin. The structure of the eastern basin has a roughly rhomboidal pattern (fault sets along N053° and N103°), displays four deep troughs and a much smaller amount of volcanism than the central basin, and is affected by extension with a significant sinistral strike-slip component.

A total of 69 diatom taxa, belonging to 24 genera were recorded from thirteen freshwater and moss samples collected in the Antarctic Peninsula region. Marine taxa comprised 26%. Cluster analysis revealed three assemblages. The Nitzschia acidoclinata assemblage occurs on wet mosses; the Luticola muticopsis assemblage is found near penguin rookeries whereas the Pinnularia krookii–Pinnularia microstauron var. elongata assemblage occurs on drier places such as dry mosses and drying mud.

Geomorphological observations show no detectable uplift (i.e. falling relative sea level) of Amery Oasis since the establishment of relatively stable sea level during the mid-Holocene. The observations around the basin of Beaver Lake include an absence of raised shoreline features, the presence down to the present tidal limit of in situ ventifacts and residual landforms, the cliffed southern shoreline and adjacent shallow subhorizontal floor of Beaver Lake, and the composition of recent moraines on the basin's north eastern edge. This lack of Holocene uplift is consistent with low uplift rates observed from coastal oases of East Antarctica and suggests minor, rather than major, changes to the Antarctic ice sheet during the most recent Quaternary glacial cycle. The formation of Beaver basin is attributed to late Cenozoic glacial excavation by south flowing ice of the palaeo-Nemesis Glacier, initially eroding when relative sea level was higher than it is today. The basin containing Radok Lake was excavated by the palaeo-Battye Glacier probably when most effective during the numerous long cold periods of the late Cenozoic. The field evidence from landforms and the presence of marine fossil deposits suggests Amery Oasis was not overrun by erosive ice since at least the Pliocene, major ice streams such as Lambert Glacier flowing then, as now, around the oasis.

A series of benches on nunataks of Alexander Island (Antarctica) are described. An increase in bench size with distance away from the retreating glacier suggests an age spectrum. The benches have thermal contraction cracks (in bedrock) on shallower, upper sections of the risers as well as salt encrusted runnels on the steeper lower section of the tread. The benches also show a distinct orientational preference (orientated to the north through to west) and, from first principles, these seem to be the aspects with optimal freeze-thaw cycles and temperatures conducive to thermal stress fatigue. The extensive dilatation associated with the retreating glaciers is thought to play a significant role in the origin and development of the benches as the combination of extensive jointing and optimal process conditions are thought to constrain where benches begin. The jointing, aided by the thermal contraction cracking, then facilitates extension and continued weathering of the treads. It would appear that these benches are examples of so called “cryoplanation terraces” that have been reported as fossil forms in Europe and North America. The study of such active forms in the Antarctic may provide good analogues for fossil features found in the Northern Hemisphere.

The coast of Victoria Land extends from Williamson Head (69°11'S, 158°E) to McMurdo Sound (77°S, 163°E). A comparison of various documents and images spanning several decades has allowed the ice front fluctuation and the iceberg calving flux during this century to be estimated. During the periods from 1956–65 to 1972–73 the floating glaciers underwent a reduction of 978 km2 with an iceberg calving flux of about 134 km2 yr−1. After this, during the periods from 1972–73 to 1989–91, the floating glaciers underwent an advance of 272 km2 with an iceberg calving flux of about 53 km2 yr−1. Glacier tongues with bottom accretion calve less often than those with bottom melting. Most floating glaciers have shown cyclic behaviour without a strong trend. Exceptions to this general style are Hells Gate ice shelf, McMurdo Ice Shelf and floating glaciers of Cape Adare which have undergone a significant retreat since the beginning of the 20th century. The different behaviour of these floating glaciers has been hypothesized as being due to: increased energy available for meltwater production of marine ice that progressively warmed these thin ice shelves and then increased iceberg calving (Hells Gate and McMurdo), or to increased melting at the ice-ocean interface related to a major intrusion of Circumpolar Deep Water from the nearby continental slope (Cape Adare). An estimate of the mass balance of East Antarctica from which these glaciers are fed shows a positive value, that is significant despite all the uncertainties of balance measurements.

We present two time series of bottom current and temperature collected 8 m above the seabed on either side of a large sediment drift located on the continental rise of the Pacific margin of the Antarctic Peninsula. The mean current speed is comparable (6.2 cm s−1 and 6.1 cm s−1 respectively), but the mean direction differs by about 121°. The direction of mean flow follows the bathymetric contour, and the maximum speed never exceeds 20 cm s−1 (below the typical benthic storm threshold). The potential temperature is remarkably stable (0.11 ± 0.01°C and 0.13 ± 0.02°C at the two sites). The cross-covariance indicates a significant peak at 20.2 days lag, slightly longer than the travel time of 18.7 days calculated between the two stations following the isobaths (98.4 km) and thus providing evidence for the topographic control on bottom water flow. The observed bottom water flow is consistent with deposition of Holocene hemipelagic sediments of the ‘drift maintenance’ stage. Indicators for palaeoceanographic conditions during glacial periods of the ‘drift maintenance’ stage and the older ‘drift growth’ stage are at present too scarce to understand fully how the past oceanographic conditions influenced the evolution of the drifts.

The long and complex Archaean evolution of the Napier Complex of Enderby Land, characterized by high-grade metamorphism and several strong deformations, is reassessed in the light of new SHRIMPU–Pb zircon dating results bearing on the ages of protoliths and possible regional extents of distinct Archaean tectonothermal events. Initial felsic igneous activity occurred over a significant time interval c. 3800 Ma ago. An age of 2980±9 Ma for the emplacement of charnockite at Proclamation Island might date the oldest tectonothermal event to be recognized in the Napier Complex. An ensuing, very-high grade, previously imprecisely dated tectonothermal event occurred at 2837±15 Ma. U–Pb zircon ages ranging from 2456+8/−5 Ma to 2481±4 Ma date a subsequent, protracted high-grade tectonothermal event. Whereas the ~2840 Ma event is of regional importance in the Amundsen Bay-Casey Bay area, it is possible that the ~2980 Ma event was of only moderate grade, minor importance, or even absent, in that part of the Complex. If so, the apparent trend to very-high temperature metamorphism in the Tula and Scott mountains compared with the Napier Mountains may reflect two distinct metamorphic events rather than a simple baric and thermal gradient. The oldest crustal component in the Napier Complex appears to have been of igneous derivation. Zircon populations in paragneisses at Mount Sones are similar to those in the nearby orthogneisses, which therefore may have been basement. Another paragneiss, in the Casey Bay area, yields no zircons older than 2840 Ma, probably indicating that pre-3000 Ma crust, which is now located nearby, was not exposed at the time of sedimentation there. The isotopic data are quite complex, particularly in rocks that experienced postcrystallization metamorphic temperatures of 1000°C or more. It is postulated that this complexity, which was largely the product of migration of radiogenic Pb within the zircon grains in ancient times, and produced local excesses of this element with respect to its parent U, was caused by volume diffusion at these abnormally high regional crustal temperatures.

Alpine-type valley and cirque glaciers occur in many massifs in the northern Prince Charles Mountains. A total of forty-seven glaciers have been investigated using maps and aerial photographs, and in the summer of 1991–92 seventeen of these were examined in the field. The distribution of these glaciers and their present-day snowline line altitudes appear to be influenced by their location with respect to snow-bearing winds, particularly the summer winds that bring moisture from the open waters of Prydz Bay. Moraine morphologies indicate that these glaciers advance and retreat out-of-phase with the larger ice sheet outlet glaciers. During the last glacial maximum the alpine-type glaciers retreated while the ice sheet outlet glaciers showed a minor expansion. This is believed to be due to the alpine-type glaciers being starved of snowfall as the expanded last glacial maximum sea-ice cover around the continent would have removed their maritime moisture sources. Recent contrasts in the behaviour of the alpine glaciers may reflect changes in summer sea ice extent in Prydz Bay.

A study of early winter first-year sea ice conditions and development in the western Ross Sea in May and June 1995 included measurements of snow and ice thickness, freeboard, ice core structure and stable isotopic composition. These variables showed strong spatial variability between the Ross Ice Shelf and the ice edge 1400 km to the north, and indicate that the development of the Ross Sea pack ice is quite different from that observed in other Antarctic sea ice zones. The thinnest snow and ice occurred in a 200 km wide coastal zone. The thickest snow and ice were observed in a continental shelf zone 200–600 km from the coast where the average ice thickness (0.8 m) determined by drilling is as thick as first-year sea ice later in winter elsewhere in Antarctica. A zone of moderate snow and ice thickness occurred on the deep ocean from 600 km to the ice edge at 1400 km. Thermodynamic thickening of the ice in the inner pack ice, <800 km from the coast, was dominated by congelation ice growth, which occurred in a greater amount (65%) and in thicker layers (mean: 20 cm) than was observed in the outer pack ice >800 km from the coast (amount: 22%; mean layer thickness: 12 cm) and elsewhere in the Antarctic pack ice. The preponderance of congelation ice in the inner pack ice might be due to a low oceanic heat flux on the Ross Sea continental shelf, and a colder, less stormy environment which favours the more frequent and prolonged calm conditions necessary for significant congelation ice growth. In the outer pack ice, thermodynamic thickening occurred mainly by snow ice formation (mean layer thickness: 20 cm) while dynamic processes, i.e., rafting and ridging, caused the thickening of frazil ice and columnar ice (mean layer thickness: 14 cm and 12 cm respectively). A greater amount of snow ice (37%) occurred in the outer pack ice than in the inner pack ice (15%), and both values indicate that in the Ross Sea, unlike other Antarctic sea ice zones, there can be significant seawater flooding of the snow/ice interface and snow ice formation before midwinter.