Abstract
Cladistic relationships were assessed for ten Middle American and eleven South American pit vipers of the genus Bothrops (sensu lato) based on biochemical and anatomical characteristics. Agkistrodon piscivorus, Crotalus durissus and Lachesis muta were included as reference taxa for outgroup rooting purposes. The Middle American species were analyzed separately from the South American species. In each analysis a combined data matrix of allozyme, isozyme, external morphology and cranical anatomy was subjected to Wagner parsimony procedures using the Phylogenetic Analysis Using Parsimony (PAUP) computer package.For the Middle American species, three alternative minimal length cladograms were constructed. Topological differences resulted from the ambiguous placement of Bothrops godmani and Lachesis muta. The remaining Middle American taxa conformed to stable groupings. Four lineages were identified: (1) B. asper and B. jararaca, (2) arboreal taxa, B. lateralis, B. nigroviridis, B. schlegelii and B. undulatus, (3) the hognose group including B. nasutus and B. ophryomegas and (4) the jumping vipers B. nummifera and B. picadoi. Based on these results generic rank names are proposed to accommodate each lineage. These are Bothrops, Bothriechis, Porthidium and Atropoides respectively.The cladistic relationships of the South American species indicate that at least two major lineages exist. Also, the South American forms are much less divergent in their characteristics than are the Middle American forms. The South American lineages include: (1) the B. atrox group (B. atrox, B. brazili, B. jararacussu, B. moojeni and B. pradoi) and (2) the B. neuwiedi group (B. alternatus, B. erythromelas, B. itapetiningae, and B. neuwiedi). The relationships of two species, B. castelnaudi and B. jararaca were not fully resolved.The cladistic relationships of the Middle American and South American taxa are compared to earlier proposals of pit viper phylogeny and found not to be concordant. In addition, these results are discussed in light of their implications to understanding vicariant aspects of Neotropical biogeography.