Sub-Antarctic Marion Island's glacial history has acted as a control on abiotic terrestrial processes and the colonization and distribution of biotic species found on the island today. Recent chronological studies have shown an early deglaciation of the island and identified new geomorphological features associated with past ice dynamics. These permit a reassessment of ice extent during and after the island's last local glacial maximum. In this paper, we provide a revised reconstruction of the island's palaeo-ice extent by using a geomorphology-based approach to delineate palaeo-ice margins and demarcate possible glacial basins. The model presented here provides the needed spatial context for future studies on the variations in the distribution of species (e.g. microorganisms and plant species) and abiotic processes and forms (e.g. soil development and periglacial landforms). In addition, it highlights areas that require improved geophysical assessment in order to produce a more complete island-scale reconstruction of former ice extents (e.g. the west coast).

A vagrant black-browed albatross Thalassarche melanophris bred with a grey-headed albatross T. chrysostoma on Marion Island at least four times between 2000 and 2009 (and continued to return to the colony until at least 2019). The eggs failed to hatch in three breeding attempts, but the pair fledged a chick in the 2006/07 breeding season. Genetic sexing identified the black-browed albatross as female and she shared all eight sampled microsatellite alleles with the chick, whereas the grey-headed albatross social parent did not match the chick. The fledgling was banded and re-sighted in its natal breeding colony in 2016 and 2018, when it displayed an intermediate black-browed x grey-headed albatross phenotype, similar to a putative hybrid photographed at sea off Australia. These results suggest that the black-browed albatross cuckolded its social mate with another grey-headed albatross in 2006/07. The failures of the other three breeding attempts at the egg stage possibly indicate genetic incompatibility with the social partner.

To evaluate how avian influenza virus (AIV) circulates among the avifauna of the Antarctic and sub-Antarctic islands, we surveyed 14 species of birds from Marion, Livingston and Gough islands. A competitive enzyme-linked immunosorbent assay was carried out on the sera of 147 birds. Quantitative reverse transcription polymerase chain reaction was used to detect the AIV genome from 113 oropharyngeal and 122 cloacal swabs from these birds. The overall seroprevalence to AIV infection was 4.8%, with the only positive results coming from brown skuas (Catharacta antarctica) (4 out of 18, 22%) and southern giant petrels (Macronectes giganteus) (3 out of 24, 13%). Avian influenza virus antibodies were detected in birds sampled from Marion and Gough islands, with a higher seroprevalence on Marion Island (P = 0.014) and a risk ratio of 11.29 (95% confidence interval: 1.40–91.28) compared to Gough Island. The AIV genome was not detected in any of the birds sampled. These results confirm that AIV strains are uncommon among Antarctic and sub-Antarctic predatory seabirds, but they may suggest that scavenging seabirds are the main avian reservoirs and spreaders of this virus in the Southern Ocean. Further studies are necessary to determine the precise role of these species in the epidemiology of AIV.

Although spiders are a diverse and ecologically important group of predators across the sub-Antarctic islands, relatively little is known about their biology. Here we provide data on the abundance, body size variation and sex ratio of an indigenous spider, Myro kerguelenensis, across an altitudinal gradient on Marion Island. In so doing we test explicitly the hypotheses that density will decline with declining resource availability at higher elevations, and that a converse Bergmann body size cline will be found in this species. Density of M. kerguelenensis decreased with altitude and ranged from a mean density of 5.3 (SD 3.42) individuals per m2 at 50 m a.s.l. to a mean density of 0.83 (SD 1.15) individuals per m2 at 600 m a.s.l. Mean female sternum width was 1.39 mm (SD 0.44) and mean male sternum width was 1.40 mm (SD 0.22). No evidence for Bergmann or converse Bergmann clines was found. At increasing altitudes, sex ratios became increasingly female-biased with populations at 600 m a.s.l. comprising 0.87 (SD 0.28) females, on a proportional basis, possibly as a result of resource limitation and an increase in the prevalence of sexual cannibalism. The food web implications of this study are highlighted.

Southern elephant seals Mirounga leonina within the Kerguelen province disperse between Marion Island (MI), Iles Crozet (IC) and Iles Kerguelen (IK) despite the high site fidelity characteristic of this species. From 1987–2002, 199 IC individuals, or 11.63% of the tagged population there, were resighted at MI. Resights of IC seals at MI peaked during the juvenile moult and seals aged 0–2 represent 66% of all resights made. Equal numbers of male and female seals (all age classes) were dispersing. Dispersing breeding females (n = 22, on 33 occasions) outnumbered dispersing breeding males (n = 6, on 16 occasions), after initially emigrating from IC to MI as juvenile seals (natal dispersal). Greater male-mediated gene flow was ultimately attained due to the polygynous mating system and some extremely successful males. Of 132 seals fitted with satellite-relay data loggers at MI, six hauled out at IC and three at IK. Two of the seals which migrated to IC were born at MI; all others were unmarked males likely to be native to IC or IK. Dispersal should ideally be considered when assessing vital rates for southern Indian Ocean elephant seal populations.

Management of non-native species introductions is a conservation priority in the Antarctic region. However, despite the recognised importance of intra-regional propagule transfer, the majority of studies have focused on inter-regional pathways (i.e. from outside of the Antarctic region). Here we quantify the number of seeds carried by expeditioners who have visited sub-Antarctic Marion Island. We recorded 420 seeds from 225 items of clothing, with seeds found on 52% of the items and soil on 45% of them. The median number of seeds for field-based and station-based personnel was 20.5 and 3 per person, respectively. Waterproof trousers and socks, particularly those of field workers, carry the greatest number of propagules (for field workers, medians of 5 and 6.5, respectively) and therefore should be the focus of intra-regional management interventions. Amongst the seeds found entrained within clothing several were from species which are widespread aliens in the Antarctic region including Agrostis stolonifera, Poa annua and Sagina procumbens, and indigenous zoochorous species (Acaena magellanica, Uncinia compacta) were also well represented. The present data provide quantitative evidence in support of previous, largely hypothetical concerns about the risks of intra-regional propagule transfer in the Antarctic.

The ecosystem of the isolated Prince Edward Islands, south of the African continent, is strongly impacted by ocean eddies that are associated with the eastward flowing Antarctic Circumpolar Current. Satellite altimetry has revealed that the archipelago lies in a region of enhanced eddy kinetic energy. In the late 1990s it became apparent that in order to understand the influence of these eddies on the islands’ ecosystem, the source, trajectory and nature of these eddies needed to be studied and understood. To this end a special research project with a strong ocean-going component was designed, the DEIMEC (Dynamics of Eddy Impact on Marion’s ECosystem) programme. In this review we focus on the physical oceanography and summarize the aims, the results and the successes of this South African research initiative. In the vicinity of the Prince Edward Islands, an average of three intense well-defined eddies is observed per year. Their advection speeds are of the order of a few kilometres per day and longevities of 7–11 months. These features, of c. 100 km in diameter and reaching depths of at least 1000 m, transport anomalous water masses across the Polar Frontal Zone.

The presence and distribution of alien plant species on subantarctic Marion and Prince Edward Islands is reviewed and updated. This study documents the finding of a well established population of Cerastium fontanum on Prince Edward Island. An additional taxa, Senecio sp., is recorded in association with a recurrence of Plantago lanceolata on Marion Island. Propagules of both these taxa have recently been dispersed to the island in building material imported from South Africa. The distribution of Agrostis stolonifera and Sagina apetala on Marion Island has increased markedly between 1981 and 1989.

The vascular vegetation of a mire-grassland community on Marion Island (47°S, 38°E) takes up c. 158 mg N m−2 d−1 in summer. Bryophytes take up c. 36 mg N m−2 d−1 during their peak growth period. Since inputs of N through precipitation and biological fixation are negligible, mineralization of organic N must have supplied the bulk of this N. From changes in peat inorganic N levels and rates of uptake by the vegetation we estimate mean mineralization rates of 178 mg N m−2 d−1 in summer and 55 mg N m−2d−1 in winter. In situ incubationof peat give a maximum mineralization rate of 48 mg N m−2 d−1. At this rate the small (700 mg m−2) pool of available N in the upper 25 cm of peat would be depleted by the vascular vegetation in about seven days and bryophytes would deplete the available N pool in the top 25 mm in two days. Hence the rate of N mineralization measured by incubation is much too low to account for the fluctuations in concentrations of inorganic N in the peat and the amounts taken up by the vegetation. This may be due to losses through denitrification or to the fact that soil macroinvertebrates were excluded from the incubation.

Brown trout were introduced to the Van den Boogaard River on subantarctic Marion Island in 1964, and a small population became established. The last individual was seen in 1984, and the species is now considered to be extinct on the island. Their diet was exclusively allochthonous, with snails and spiders predominating. Ages estimated at six to eleven years showed that spawning must have occurred since the original introduction. Since the Van den Boogaard River enters the sea via a waterfall, it is postulated that trout were not able to practice an anadromous life-style, and that this, as well as other factors connected with the impoverished nature of the stream, led to dwarfing of the resident population. No further introductions of alien fish to Marion Island should be contemplated.

Seasonal movements and foraging areas of postbreeding (n = 9) and postmoulting (n = 3) adult southern elephant seal females from Marion Island were studied using Geolocation Time-depth Recorders. Movements were classified into three phases – an outbound transit phase, distant foraging phase, and an inbound transit phase. The longest residence time of postbreeding females during their foraging migrations was in areas at the outer edge of their feeding range (± 1460 km) both to the north and south of the island, largely within inter-frontal zones south of the Antarctic Polar Front (APF) and between the Sub-Tropical Convergence (STC) and the Sub-Antarctic Front (SAF). Postmoulting females travelled further afield (2122–3133 km distant) to the APF, to inter-frontal zones south of the APF (within the pack ice outer edge), as well as to the Antarctic Continental Shelf. This study provides additional information on the putative function of dive types in relation to the movement phases of elephant seal females from Marion Island. The relative frequency of assumed ‘foraging’, ‘exploratory’ and ‘transit’ dive types, as well as the duration and location of the different phases of movement suggest two seasonal foraging strategies. Sea floor topography could possibly cue the transit phases of both postbreeding and postmoulting females from Marion Island.

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.

Estimates of births and standing age-distributions were combined to estimate the size of the southern elephant seal population at Marion Island at various times during its decline. To estimate births each year from 1986 through 1997 we used the number of adult females hauled out on 15 October, which is the peak haulout date for breeding elephant seal females at all breeding sites in the Indian Ocean. A conversion factor (3.15) was derived from the standing age-distributions, and applied to estimates of annual births to yield total population size. The population at Marion Island declined 37.5% overall from 1986–97 at an annual rate of 4.3% from 1986–91 and c. 2.5% yr−1 afterwards.

Southern elephant seals haulout on land to moult, breed and for a third, unknown reason, which we refer to as the winter haulout. We used long-term mark-recapture data to estimate participation levels in the winter haulout by southern elephant seals at Marion Island. There was no evidence that participation levels varied between cohorts or between years. Participation differed between sexes, with males being more likely to haulout in winter, except in the first year of life where participation was equal. Within each sex, both age and reproductive status influenced participation, but age seemed to be the most influential determinant. Generally, immature male individuals hauled out year after year in winter. The results did not allow speculation as to the purpose(s) of the winter haulout.

Five populations of habitat-specific, morphologically distinct Algophagus mites (Astigmata: Algophagidae) were found to occur on the sub-Antarctic Prince Edward islands. Although these mites did not differ among themselves with respect to idiosomal setal characteristics, which were also similar to those of known sub-Antarctic Algophagus taxa, morphometric analyses revealed leg and idiosomal size differences. In addition to morphometric differences, leg setal positions and porose axillary organ structure, indicated that the morphotypes were taxonomically discrete. The taxa were found to include the known A. antarcticus Hughes, 1955, A. laticollaris Fain, 1974 (new status), a sub-species of A. semicollaris Fain, 1974, and two new species, A. brachytarsus sp. nov. and A. macrolithus sp. nov. With the exception of A. semicollaris, the taxa are described or redescribed in this paper. Morphological features of the females in particular, and notably leg proportions, claw dimensions, and apodeme positions, corroborate discrete terrestrial (A. brachytarsus and A. macrolithus) and aquatic forms (A. antarcticus, A. laticollaris, and A. semicollaris), and vary in a manner suggesting a functional relationship to habitat occupation by the various taxa.