Featured White Papers
- Oct. 14th: Simplified IT with Software-as-a-Service (SaaS) (ZDNet)
- PCI DSS therapy for the smaller retailer (McAfee)
- The rise of Web commuting (Citrix Online)
Early mammalian radiations
Journal of Paleontology, Nov 2001 by Cifelli, Richard L
How ancient is the marsupial-placental dichotomy (Lillegraven, 1969, 1974, 1975; Lillegraven et al., 1987)? Using stem-based definitions (and the respective terms Metatheria and Eutheria), generally-accepted eutherians are now known from the late Early Cretaceous of both Asia (Kielan-Jaworowska and Dashzeveg, 1989; Sigogneau-Russell et al., 1992; Averianov and Skutschas, 2000) and North America (Cifelli, 1999). The earliest uncontested metatherian is from the Albian-Cenomanian of North America (Cifelli, 1993b; Cifelli and Muizon, 1997). Recent studies suggest that Deltatheroida, a predominantly Asiatic group, may be related to marsupials (Marshall and Kielan-Jaworowska, 1992; Rougier et al., 1998). Yet the teeth of deltatheroidans are of remarkably primitive construction (Cifelli, 1993a); if they are related to marsupials, an extremely early divergence of Metatheria and Eutheria is implied (Kielan-Jaworowska, 1982, 1992). Molecular studies (Kumar and Hedges, 1998; Penny et al., 1999) place the metatherian-eutherian dichotomy as far back as the Middle Jurassic, long before the first appearance of either group in the fossil record.
Metatheria.-Only a handful of metatherian fossils was known from the Mesozoic when the Journal of Paleontology was founded (Simpson, 1929a). That record underwent explosive expansion in the second half of the twentieth century, thanks in large part to widespread use of underwater screen-washing and associated techniques (e.g., Clemens, 1966; Lillegraven, 1969; Fox, 1971, 1979a, 1979b, 1989; Archibald, 1982; Storer, 1991; Eaton, 1993; Cifelli, 1990a, 1990b). In North America, a modest diversity of taxa (represented mainly by jaws and teeth) is now known for much of the Late Cretaceous. Metatheria did not fare well on this continent after the close of the Mesozoic, however (Archibald, 1996a), and their subsequent radiations were mainly a Gondwanan phenomenon. Metatherians have also recently been described from the Late Cretaceous of Asia (Szalay and Trofimov, 1996; Averianov and Kielan-Jaworowska, 1999), suggesting a much broader early distribution pattern than previously envisaged (Lillegraven, 1974).
The record of Mesozoic metatherians remains negligible or non-existent for southern landmasses, though the Tertiary record has greatly improved. Particularly noteworthy are discoveries of Paleocene fossils from Brazil (Marshall, 1987; Szalay, 1994) and Bolivia, including remarkably complete skulls and skeletons (Muizon, 1995, 1998; Muizon et al., 1997). The Tertiary marsupial record from Australia has expanded dramatically, though problematic gaps persist (Archer, 1984; Archer et al., 1999), and fossil marsupials are now also known from Antarctica (Woodburne and Case, 1996; Goin et al., 1999). This vast increase in knowledge and appreciation of marsupial diversity, coupled with increased focus on relationships among major groups, has resulted in profound changes in systematics and taxonomy. Simpson's (1945) synthetic classification included only six superfamilies, all grouped within a single order. All workers since Ride (1964) have recognized the enormity of marsupial diversity by placing them in multiple orders, though relationships remain debated in many cases. An influential contribution is that of Szalay (1982; see also Szalay, 1994), who linked Australian taxa to one group (Microbiotheriidae) among the South American radiation, based on structure of the tarsus. Most later phylogenies and resultant classifications follow Szalay in recognizing a fundamental split of marsupials into australodelphian (including the microbiotheres) and ameridelphian Glades. However, the results of these studies (e.g., Archer, 1984; Aplin and Archer, 1987; Marshall, 1987; Reig et al., 1987; Marshall et al., 1990) are at variance on many points, and the interordinal relationships among marsupials remain unstable, particularly when molecular data are considered (e.g., Burk et al., 1999).