DISCUSSION

Prior to 1994, docodont mammals were known only from the Upper Triassic, Middle and Upper Jurassic, and Lower Cretaceous deposits of North America and Europe. This points to a typical Euramerican distribution for this group (Kron 1979). Tegotherium gubini Tatarinov 1994 was the first docodont mammal described from Asia (Late Jurassic of Mongolia). More recently, a number of docodont taxa have been reported from the Jurassic and Cretaceous rocks of Asia. These include Castorocauda lutrasimilis Ji et al. 2006 from the Middle Jurassic of Inner Mongolia, Tashkumyrodon desideratus Martin and Averianov 2004 from the Middle Jurassic of Kyrgyzstan, Itadodon tatarinovi Lopatin and Averianov, 2005 from the Middle Jurassic of Western Siberia, Dsungarodon zuoi Pfretzschner et al. 2005 from the Late Jurassic of Junggar Basin of Northwest China, and Sibirotherium rossicus Maschenko et al. 2003 from the Early Cretaceous of Western Siberia. These discoveries demonstrated that docodont mammals had wide geographic distribution and formed an important component of Mesozoic vertebrate fauna of Asia.

Based on phylogenetic analysis of docodont lower molar characters, Averianov and Lopatin (2006) suggested that Borealestes, Haldanodon, and Docodon are plesiomorphic taxa restricted to Europe and North America, whereas Tashkumyrodon, Itatodon, Tegotherium, and Sibirotherium are more derived taxa confined to Asia. According to them, certain European taxa, such as Krusatodon and Cyrtlatherium, were Middle Jurassic immigrants from Asia. Reigitherium bunodontum Bonaparte 1990 from the Upper Cretaceous (Campanian – Maastrichtian) La Colonia Formation, Argentina, originally described as a dryolestoid, was allocated to a highly derived docodont family Reigitheriidae, a possible sister group of Docodontidae and was considered as the first record of docodonts from the Gondwanan continents (Pascual et al. 2000). More recently, based on new material of Reigitheriidae from the La Colonia Formation, Rougier et al. (2003) transferred Reigitherium back to dryolestoids. In light of this, the new specimen from the Kota Formation is considered as the first record of docodont mammals from the southern continents indicating a wide geographic distribution for this group. Phylogenetic relationship of Gondtherium to the Asian taxa is difficult to assess as it is represented by an upper premolar, whereas a majority of the Asian taxa are known by lower molars and in a few cases by upper molars, but not by premolars. Within the clade consisting of Borealestes, Haldanodon, and Docodon (Pfretzschner et al. 2005, Averianov and Lopatin 2006), the holotype of Gondtherium compares well with Haldanodon pattern of premolar morphology.

Because paleobiogeographic interpretation using a single premolar with poorly established taxonomic relationship would rest on a tenuous ground, the biogeographic significance of Gondtherium needs to be approached cautiously. However, the associated mammalian taxa help us in reconstructing a generalized paleobiogeographic scenario. The more recent discovery of Dyskritodon from the Kota Formation (Prasad and Manhas 2002), first recorded from the Early Cretaceous of Morocco, represents an example of faunal continuity across India and Africa. Furthermore, Indotherium pranhitai, another mammalian taxon documented from the Kota Formation (Yadagiri 1984, Prasad and Manhas 2002) exhibits some similarities to morganucodontid upper molars from the Middle Jurassic Kirtlington Quarry, England. Finally, Kotatherium yadagirii, originally described as a "symmetrodont," is now referred to a new combination Paikasigudodon yadagirii (Prasad and Manhas 2002) within the "triconodont" family Morganucodontidae with close morphological similarities to Late Triassic Megazostrodon of South Africa (Crompton and Jenkins 1968). The occurrence of closely related mammals in the Jurassic of India and Late Triassic and Early Cretaceous of Africa, as well as Middle and Late Jurassic of Europe points to biogeographic connections between these regions. This is not surprising because paleogeographic maps show Europe in close proximity of NW Africa and India adjacent to Africa in the Early/Middle Jurassic (Scotese 1997, Chatterjee and Scotese 1999). The cosmopolitan distribution of the Kota fauna has also been corroborated by the non-mammalian vertebrate groups, ostracods, and charophytes. Barapasaurus and Kotasaurus, sauropod dinosaurs from the Kota Formation of India, and the Early Jurassic sauropod Vulcanodon of Zimbabwe appear to be closely related to the Late Triassic sauropod Isanosaurus of Thailand (Gillette 2003). The pterosaur Campylognathoides indicus is supposed to be closely related to Dimorphodon of Upper Liassic of Holzmaden, Germany (Jain 1974b). The ostracod Darwinula cf. D. sarytirmenensis has been described from the Lower Jurassic Kayenta Formation of North America (Kietzke and Lucas 1994), Middle Jurassic of Russia, and upper Middle to Upper Jurassic rocks of China (Govindan 1975). The charophyte taxon Aclistochara cf. A. jonesi is also known from the Upper Jurassic and Lower Cretaceous strata of Colorado and Wyoming (USA) and China (Feist et al. 1991). Presence of large sauropod dinosaurs similar to those of Africa, Europe, and Asia in the Jurassic of India has been cited in support of land connections that facilitated faunal interchanges between these landmasses (Chatterjee and Hotton 1986).

In a comprehensive account on the biogeography of Indian plate during the Mesozoic, Chatterjee and Scotese (1999) argued that, from Late Triassic to Late Jurassic, the Indian plate was tectonically a part of Gondwanaland and maintained two biogeographic corridors with central Africa (via Madagascar) and southern Africa (via Antarctica), and one with western South America (through Antarctica), resulting in the influx of several Laurasian as well as Gondwanan taxa. They predicted Morocco as a possible biogeographic link between the Laurasian and Gondwanan continents. The latest discovery of Dyskritodon from the Kota Formation (Prasad and Manhas 2002) is consistent with their interpretation. A Late Jurassic trans-Tethyan dispersal for primitive mammals such as "triconodonts," dryolestoids, and peramurids of Laurasian origin into Africa was suggested recently (Rauhut et al. 2002). A somewhat similar dispersal during the Late Jurassic–Earliest Cretaceous was suggested for albanerpetontid and discoglossid amphibians and early mammals (Thereudon, Tribotherium, Hypomylos, Gobiconodon, Hahnodon and Denisodon) based on Early Cretaceous (?Berriasian) terrestrial vertebrate fossil record of Morocco (Kielan-Jaworowska et al. 2004). In view of this continuity of mammalian as well as non-mammalian taxa during the Jurassic and Early Cretaceous across Gondwanan continents, it is predicted that early docodonts might have existed on other southern continents as well. The possible reasons for not finding them on the southern continents until now are 1) restricted occurrence of Jurassic continental sequences in this part of the globe, 2) low intensity sampling of the known deposits, and 3) taphonomic factors. Meanwhile if the Early Jurassic age is accepted for the Kota Formation, the tooth described earlier in this report would represent the oldest report of docodonts. However, there are conflicting views on the age of the Kota Formation. It has been dated as Early Jurassic (Upper Liassic) based on the fish fauna (Jain 1973), whereas the ostracod fauna favoured a Middle Jurassic age (Govindan 1975). Palynoflora, on the other hand, indicate a late Middle Jurassic to Early Cretaceous age (Vijaya and Prasad 2001). Therefore, a concerted effort by vertebrate paleontologists, micropaleontologists, and palynologists is necessary to resolve this issue. Notwithstanding the disparities in these datings, the new find from India is of paramount importance as it testifies to the presence of typical docodont mammals in Gondwanan continents.