Maximum-likelihood identification of fossils: taxonomic identification of Quaternary marmots (Rodentia, Mammalia) and identification of vertebral position in the pipesnake Cylindrophis (Serpentes, Reptilia)

摘要: We applied a Maximum-Likelihood (ML) criterion to the problem of identifying unknown specimens using database whose identity was known. Our approach based on shape, quantified two-dimensional Cartesian landmarks. technique two specific problems: (1) Quaternary marmot skulls (Marmota, Sciuridae, Rodentia) species, and (2) position individual elements within vertebral column Red-tailed pipesnake, Cylindrophis ruffus (Serpentes, Alethinophidia). The ML finds best by choosing sample that fits unknown. Cross-validation tests indicated identifications marmots were correct about 80%–90% time. Fossil from sites (Meyer Cave, Illinois Little Box Elder Wyoming) could be assigned species (M. monax M. flaviventris respectively), but several other localities not species-level taxon. Snake vertebrae allocated their proper columnar interval more than 80% time, with incorrect assignments rarely being 10% out place. is widely applicable in palaeontology, where isolated morphological can acute often ignored. allows palaeontologists base securely data, recognize conditions under which confident identification or cannot made.