Trachytherus, MENDOCENSIS, Ameghino, 1889

Billet, Guillaume, Muizon, Christian De & Quispe, Bernardino Mamani, 2008, Late Oligocene mesotheriids (Mammalia, Notoungulata) from Salla and Lacayani (Bolivia): implications for basal mesotheriid phylogeny and distribution, Zoological Journal of the Linnean Society 152 (1), pp. 153-200 : 179-180

TRACHYTHERUS ? MENDOCENSIS

The poorly known species T? mendocensis comes from the Divisaderan of eastern Argentina ( Fig. 20 ). The hypodigm includes two specimens: MHNM 2494 PV (the holotype, right P2–P3, M1–M2) and MLP 45-VII- 10-2 (an isolated M3). It was considered to be pre- Deseadan or even pre-Tinguirirican ( Simpson et al., 1962; Flynn & Swisher, 1995; Flynn et al., 2003). However, recent investigations on the Mariño Formation in Divisadero Largo have yielded new remains of undoubted Mesotheriinae whose M3 is very similar to that of T? mendocensis ( Cerdeño, López & Reguero, 2005; Cerdeño, González Riga & Bordonaro, 2006). On this basis, it has been suggested that at least the M3 (MLP 45-VII-10-2) from Divisadero Largo previously attributed to Trachytherus ? mendocensis might not be referable to the genus Trachytherus but to a Mesotheriinae and came from Miocene beds of the Mariño Formation ( Cerdeño, López & Reguero, 2005; Cerdeño, González Riga & Bordonaro, 2006). Remains of a hegetotheriid known from Divisadero Largo might also belong to Miocene deposits ( López & Manassero, 2006). Before more evidence is provided, we consider Trachytherus ? mendocensis as valid among Trachytheriinae and have coded it in our cladistic analysis from the two specimens known for it [MHNM 2494 PV holotype probably lost ( Cerdeño et al., 2006) and MLP 45-VII-10-2]. In this case, the analysis is not consistent with a plesiomorphic status for this taxon compared with other basal mesotheriids. The holotype of T? mendocensis, as figured in Simpson & Minoprio (1949), appears to belong to Trachytheriinae especially because the M1 almost encloses a central fossette. If the M3 (MLP 45-VII-10-2) has to be removed from this taxon, then Trachytherus ? mendocensis would be coded as unknown concerning character 12 on the shape of M3 whereas it was coded (1) (derived state) in the present analysis. It would result in six trees of 20 steps (CI = 0.95 RI = 0.96), the strict consensus of which presents T? mendocensis in a basal clade of mesotheriids with T. subandinus , T. alloxus as an isolated basal mesotheriid, and T. spegazzinianus as the sister taxon of mesotheriines.

TEMPORAL AND PALAEOBIOGEOGRAPHICAL

IMPLICATIONS

The pattern of repartition of basal mesotheriid species raises interesting temporal and/or geographical implications. Three different species from three different sites are known from the Deseadan of Bolivia, emphasizing the diversity of basal mesotheriids in this period and region. Little can be said concerning the age of the locality of Rio Pluma, which has yielded the holotype and single specimen of T. subandinus ( Villarroel et al., 1994) .

The distinction of their mesotheriid remains highlights an important difference between the Salla and Lacayani faunas. Hoffstetter et al. (1971) mentioned a possible slight time discrepancy between Lacayani and Salla. Vucetich (1989, 1991) argued that the Lacayani rodents were more similar to their Patagonian counterparts than to Salla species, which is the same for basal mesotheriids.

Salla in Bolivia and Patagonian localities (e.g. Cabeza Blanca, La Flecha, Gran Barranca, Pico Truncado, Laguna del Mate) in Argentina have yielded by far the best known Deseadan faunas. Even considering only higher taxa, these faunas share many forms. However, they always differ enough to be considerated as different species (see Shockey & Anaya, 2004, for pyrotheres). The question is which factor (ecology, geography, age) explains the observation that the Lacayani fauna resembles the Patagonian fauna more than the Salla fauna.

MacFadden (1990) argued that the environment in Salla was semi-arid but Vucetich (1991) stated that evidence from rodents did not indicate an arid environment. Cenograms even indicate a closed, moist habitat in Salla ( Croft, 2001), but the strong particularities of Cenozoic South American faunas (e.g. small number of carnivorous mammals) might affect this result. Vucetich (1991) also underlined that Salla contains less hypsodont rodent species than Lacayani and Patagonia. Therefore, the Patagonian localities were a more open landscape than Salla ( Vucetich, 1991). However, low-crowned ungulates are much less abundant in Salla than in Patagonia ( Shockey, 2005). Nevertheless, no clear pattern of difference can be extracted from general hypsodonty levels between notoungulates from Salla and Patagonia (the small number of notoungulates in Lacayani precludes such comparisons). Thus, more palaeontological data are needed to establish the potential ecological differences between these sites. A recent interpretation of the Salla palaeoenvironment indicates an ‘open woodland experiencing seasonal rainfall’ on the basis of the low species richness of primates, the common occurrence of high-crowned species and the presence of pedogenic carbonate nodules ( Kay, Williams & Anaya, 2002), very different from a semi-arid environment.

A geographical barrier may also have existed between Salla and Patagonia (see Croft, 2007, for an example of provinciality in the Miocene of South America). In this case, the Lacayani deposits, close to Salla geographically, would then reflect a period during which faunal communication between Bolivia and Patagonia was more important. To solve this problem, more precise ages for these localities are needed.

On the basis of faunal evidence, MacFadden et al. (1985) argued that the Salla fauna is older than the Patagonian ones. Among others, their argument was based on the presence in Salla of presumed ‘archaic’ groups, absent in the Patagonian locality of Scaritt Pocket. However, the opposite hypothesis (Salla younger than Patagonian localities) has recently been proposed, also from faunal evidence ( Reguero & Cerdeño, 2005), highlighting the difficulty in evaluating the relative ages of these localities based on this type of argument. The only obvious conclusion that can be made is the diachronism between Salla and Lacayani. However, the relative ages cannot be determined until more detailed studies of its fauna are made.

Radiometric dates for Salla were provided by Kay et al. (1998). The richest fossiliferous levels were given ages of between 27.02 and 25.82 Myr, whereas the best sampled interval was dated between 26 and 25.8 Myr. Concerning Patagonia, Marshall et al. (1986) provided an age (40 K/ 40 Ar methods) between 23.4 and 21 Myr for Scaritt Pocket, considered to be the youngest of the Deseadan beds. For the oldest beds they studied (Pico Truncado), Marshall et al. (1986) provided an age of approximately 33.6 Myr. As far as is known, Trachytherus spegazzinianus is present at Pico Truncado whereas no remains of mesotheriids are known at Scaritt Pocket.

More recently, Flynn & Swisher (1995) revised these ages using 40 Ar/ 39 Ar methods. They estimated an age between 29 and 27 Myr for these localities, thus constraining the Deseadan SALMA sequence to this time span. The Salla beds would then be younger than the Patagonian sites, which is not in agreement with faunal evidence ( MacFadden et al., 1985). Unfortunately, radiometric dates are scarce in Patagonia. For example, nothing is known on characteristic localities such as La Flecha or Cabeza Blanca ( Marshall et al., 1986), and care should be taken with ages already estimated ( Kay et al., 1998). Better knowledge of absolute and relative ages of Patagonian sites is needed before it is possible to attribute faunal differences with Salla to temporal, geographical or ecological causes (or a combination of these). Also, additional studies on the Lacayani fauna would help to shed light on the particular faunistic distribution pattern observed in the Deseadan.

Figures

Figure 20. Locality map of some Oligocene mesotheriid-bearing localities. Lacayani (Trachytherus spegazzinianus); Salla (Trachytherus alloxus sp. nov.); Rio Pluma (Trachytherus subandinus); Curuzú Cuatiá (Trachytherus curuzucuatiense = Trachytherus spegazzinianus); Divisadero Largo (Trachytherus? mendocensis); Cabeza Blanca and La Flecha (Trachytherus spegazzinianus).