The history of discovery of Australian mammals is reviewed, and it is concluded that although we may be approaching a reasonable knowledge of the diversity of Pleistocene mammals, the Tertiary record is only beginning to unfold. The Tertiary record and possible origins for monotreme, marsupial and placental groups are reviewed. Two families are so far known only from the Tertiary. All others survived into the Quaternary. The diprotodontids, macropodids, and thylacoleonids are proving useful as a basis for correlation. The oldest known Australian mammals are late Oligocene in age and reveal little about the actual origins of the groups they represent. Four Australian marsupial groups (the dasyuroids, perameloids, thylacinids and phalangeroids) have been considered to be more closely related to American than other Australian groups. Monotremes may be derivatives of multituberculates. Too little research into the affinities of Australia's placental groups has precluded attempts to determine their precise non-Australian origins: some may be Asian, such as the rodents, but some bat groups may have had a Gondwanaland origin. From extrapolations based on the much better-known history of the North American mammals, it is possible that we know of only a third of the total Australian mammal families. There may remain about 45 additional families to be discovered in the as yet unknown earlier Tertiary Australian fossil record.
