Gondwanatheria
Scientific classification
Kingdom: Animalia
Therapsida
Class: Mammalia
Suborder: Gondwanatheria
Genera
  Gondwanatherium
  Ferugliotherium
  Lavanify
  Sudamerica

Ref.

Gondwanatheria is an extinct taxon of mammals from the Upper Cretaceous to the Eocene of the southern hemisphere. The affinities of the group are not clear.
Gondwanatherians were first interpreted as early edentates; 'toothless' mammals such as anteaters. This sounds somewhat ironic, given that they were originally known only from teeth. ('Toothless' shouldn't be taken overly literally!) They were also treated as members of Multituberculata. Latest thinking is apparently returning towards the edentate affinities, or something else entirely.
Though generally no longer seen as multituberculates, "a few specimens described as ?Ferugliotherium," are from multis, (Kielan-Jaworowska & Hurum, 2001, p.411). "These poorly known specimens (not discussed herein) demonstrate that a branch of multituberculates apparently lived during the Late Cretaceous in South America."

There are two known families within Gondwanatheria. Sudamericidae Scillato-Yané & Pascual, 1984 (Sudamerica, Gondwanatherium and Lavanify) and Ferugliotheriidae Bonaparte, 1986 (Ferugliotherium). Further fossils have come from India and Antarctica, where gondwanatherids once lived in the lush forests of the Eocene.

The Gondwanatherians of Seymour Island, Antarctica

Antarctica has more to offer than just marsupials. "The occurrence on Seymour Island of sudamericids, that had become extinct in South America in the Paleocene, also indicates that isolation may have allowed extended survival of this Gondwanan group in the Eocene of Antarctica and the factors that caused their extinction did not affect this continent," (Reguero et al 2002, p.189).
From the same paper, (p.203): "For the sudamericids, Koenigswald et al. (1999) inferred a semi- aquatic and perhaps a burrowing way of life, similar to that of living beavers. Regarding this, the presence of two Antarctic taxa at Seymour Island (Goin, personal communication, 2000) suggests an important paleoecological constraint related to dietary preference of this group.”
The Antarctic Peninsula of the "late Early to latest Eocene", (La Meseta Formation -about 40 million years ago), seems to have been a lively place. The fossils and the geological conditions of the site suggest a near-by forest populated by a diverse fauna, which had many similarities with the slightly earlier residents of Patagonia; small, arboreal, fruit and insect eating possums, 10kg sloths, middle to largish grazers (sparnotheriodontids and Trigonostylops), falcons, ratites (rheass and the like, aka big, flightless birds) and penguins. At the end of the Eocene the climate seems to have taken a turn for the worse...
With thanks for Martin Jehle for the notification and interesting abstract and David Marjanovic for digging out the paper.

Link
[1]Marsupial teeth from Antarctica

Page references: Kielan-Jaworowska Z & Hurum JH (2001), Phylogeny and Systematics of multituberculate mammals. Paleontology 44, p.389-429.
Reguero MA, Sergio AM & Santillana SN (2002), Antarctic Peninsula and South America (Patagonia) Paleogene terrestrial faunas and environments: Biogeographic relationships. Palaeogeography- Palaeoclimatology-Palaeoecology, 179, p.189-210.

(This information has been derived from [1] MESOZOIC MAMMALS; Gondwanatheria, an internet directory. As that's my webpage, there are no issues of copyright. Trevor Dykes)