Hartman 9523567 The goal of the current proposal is to substantially increase knowledge of the Late Cretaceous vertebrate fauna of Madagascar, which is currently very poorly known. With the discovery and analysis of additional and more complete vertebrate fossils from the Mahajanga Basin of northwestern Madagascar, it will be possible to address three critical (but not mutually exclusive) hypotheses relating to the physical and biotic connections of the component parts of eastern Gondwana during the Late Cretaceous and the biogeographic origins of the extant vertebrate fauna of Madagascar: HYPOTHESIS I: Madagascar, Africa, and the Indian subcontinent, owing to their physical proximity, shared vertebrate faunas in the Late Cretaceous that were very similar, in some cases at the species level. Many of the vertebrates from the Late Cretaceous of Madagascar are thought to be closely related to or, in some cases, conspecific with those of Africa and India. Correspondingly, there is general agreement among geophysicists that the Indian subcontinent was close or adjacent to the eastern margin of Madagascar at this time. Despite apparent similarities in the Late Cretaceous vertebrate faunas of Madagascar, Africa, and India, detailed comparisons have been difficult due to inadequate material. More complete material of vertebrate fossils from Madagascar (as well as India and Africa) are needed for such comparisons to be made. HYPOTHESIS II: The Indian subcontinent was faunally isolated from Madagascar (and Africa) in the later stages of the Late Cretaceous and, in the latest Cretaceous, contained elements from Laurasis (as a result of collision with Eurasia) that were not shared with Madagascar (or Africa). Some of the evidence used in support of the recent and controversial hypothesis that the Indian subcontinent had already collided with Eurasia by the latest Cretaceous is paleontological. The presence of purportedly Laurasian groups in the Maastrichtian of India indicated that a te rrestrial route between the Indian subcontinent and Eurasia may have existed by this time. However, the material used to support this conclusion consists of isolated, fragmentary, and rare specimens with questionable identifications. Preliminary identifications of some specimens form the Mahajanga Basin indicate that at least some of the "Laurasian" taxa found in the Late Cretaceous of India also occurred in Madagascar. If these identifications are confirmed, it is more likely that these "Laurasian" elements entered the Indian subcontinent by way of Africa and Madagascar, rather than from Eurasia. HYPOTHESIS III: Some vertebrate higher taxa that comprise the extant fauna of Madagascar (e.g., Lemuriformes, Tenrecidae, Cordylidae) colonized the island prior to the beginning of the Cenozoic. The biogeographic origins of the extant, highly endemic and imbalanced vertebrate fauna of Madagascar are unknown and have been the subject of speculative debate for many decades. This is largely the result of the virtual absence of nonmarine, pre-Holocene Cenozoic strata on the island. As a result, the best opportunity to shed light on this major problem is to discover vertebrate fossils in the richly fossiliferous Upper Cretaceous nonmarine strata of the Mahajanga Basin. Much of the significance of this project lies in discovering and understanding more about the vertebrates that lived in Madagascar during the Late Cretaceous. Fossil collections obtained through the proposed two field seasons will far exceed all previous efforts (combined) and will provide a reasonable estimate of the diversity of vertebrates in the "Maevarano" and "Marovoay" formations. In the process of accumulating and analyzing data to test the hypotheses above, we expect to: 1) contribute substantially to knowledge of the diversity, anatomy, and phylogenetic relationships of the Late Cretaceous vertebrates of Madagascar; and 2) better constrain the lithostratigraphy and biostratigraphy of the fossiliferous beds (and to d efine formally the "Maravoay", "Maaevarano", and "Berivotra" formations).
