The genus Crocodylus comprises an evolutionarily recent radiation (12 extant species) of large, aquatic Archosaurian predators. The group is likely to have an African saltwater-tolerant common ancestor and trans-oceanic migration is required to account for the current circumtropical distribution of the group. However, the phylogeny and historical biogeography of the genus remains poorly resolved. Many species have suffered recent severe declines due to a combination of habitat destruction and overharvesting associated with the crocodilian skin-trade. All species are now protected under CITES, but several remain critically endangered in the wild, including in the Indo-Pacific, the Siamese crocodile, C. siamensis, and the Philippine crocodile, C. mindorensis. Other Indo-Pacific species are intensively managed for skins, including the New Guinea crocodile, C. novaeguineae, and the estuarine crocodile, C. porosus.
A complication for conservation and management of crocodiles in the Indo-Pacific region is that the distribution of genetic variation within species is unclear, and species boundaries are uncertain on the basis of morphology. There are three areas of particular concern: (i) the southern New Guinean population of C. novaeguineae may comprise an undescribed taxon on the basis of divergence in morphology and life history traits from the allopatric northern population; (ii) recent anecdotal evidence exists for a biogeographically distinct population of C. siamensis in Borneo; and (iii) a putative new species of palustrine crocodile from Borneo, C. raninus, has recently been resurrected. Clarification of the taxonomic status of freshwater Crocodylus species in the Indo- Pacific has been identified as a conservation priority by the IUCN Crocodile Specialist Group. However, no phylogeographic data has yet been collected for any Indo-Pacific crocodilian.
In this thesis I combined Bayesian and maximum likelihood phylogenetic analyses of relationships within Crocodylus (Chapter 2), with population genetic and phylogeographic analyses of several Indo-Pacific species, the estuarine crocodile, C. porosus (Chapter 3), the New Guinea crocodile, C. novaeguineae (Chapter 4), and the Siamese crocodile, C. siamensis (Chapter 5). The aims of the study were to: (i) resolve the phylogeny and biogeographic history of the genus; (ii) clarify the taxonomy of phenotypically (southern-form C. novaeguineae; C. raninus) or biogeographically (Bomean C. siamensis) distinct populations of uncertain status; and (iii) discriminate historical and contemporary determinants of population structure within species in order to delimit conservation units for management.
Bayesian and maximum likelihood phylogenetic analyses of mtDNA sequences (Chapter 2) revealed a monophyletic Central American clade (C acutus, C. intermedius, C. moreletii, C rhombifer), and supported trans-Atlantic colonisation of this region by an African founder closely related to the Nile crocodile, C. niloticus. Monophyly of the six Indo-Pacific species was weakly inferred, but there was strong evidence for a monophyletic lineage in which the Australian freshwater crocodile, C. johnstoni, was basal to the sibling species C. novaeguineae and C. mindorensis. A second Indo-Pacific lineage comprising C. palustris, C. porosus and C siamensis received moderate support, but relationships within this clade were poorly resolved. The putative Bomean species, C. raninus, appears to be conspecific with northern-form C. novaeguineae, although this result requires confirmation with genetic analysis of museum type specimens.
The phylogenetic analysis provided a fi-amework for phylogeographic and population genetic studies within species. Intraspecific genetic diversity was minimal relative to mean interspecific divergence, particularly for C. novaeguineae, and there was no evidence for departures from existing species boundaries. Different historical and contemporary processes were found to underlie contrasting phylogeographic patterns in the three species, with less structure evident for the estuarine species, C. porosus, than for co-distributed populations of the two primarily freshwater species C. novaeguineae and C. siamensis. A strong phylogeographic break was revealed between C. siamensis on the SE Asian mainland and a sample of conspecific captive individuals purportedly sourced from the wild in eastern Borneo. This result is consistent with the existence of a remnant population in this region and with a history of past fragmentation between Borneo and the SE Asian mainland. For C. novaeguineae, mtDNA and microsatellite analyses revealed strong, but shallow, population structure that was congruent with the north-south partitioning of morphological and life history variation in this species. This suggested that the Central Highlands have not acted as a long-term biogeographic barrier to genetic exchange, but that contemporary gene flow is very limited. This recent isolation may have facilitated the development of phenotypic differentiation in the species. Both southern-form C. novaeguineae and the putative C. siamensis population in Borneo should be considered as distinct evolutionarily significant units (ESUs) for future management.
In comparison, no phylogeographic structure was identified for the estuarine crocodile, C porosus, either among populations that were co-distributed with C novaeguineae and C. siamensis, or at a broader geographic scale encompassing the Indo-Malay Archipelago and Western Pacific Ocean. This suggests that the sampled C porosus populations comprise a single ESU. Nonetheless, populations were strongly structured on the basis of allele frequency differences and at least seven regional populations qualify as management units. A highly asymmetric pattern of historical gene flow orientated from western to eastern populations, and involving several episodes of range expansion coupled with recurrent gene flow, was found to underlie the distribution of mtDNA variation in C. porosus. A pattern of isolation by distance was identified for microsatellite markers, although there was no evidence for male biased gene flow. Differences in phylogeographic pattern between species may reflect a greater capacity for long-distance marine migration by C. porosus, relative to the two primarily freshwater species.
The integration of phylogenetic and phylogeographic analyses in this study has addressed one of the IUCN Crocodile Specialist Group's conservation priorities by clarifying species boundaries in the Indo-Pacific freshwater species complex. Existing species boundaries were supported, but distinct ESU's were identified in the New Guinea crocodile, C. novaeguineae, and the Siamese crocodile, C. siamensis. The purported Bomean endemic species, C. raninus, does not appear to be a valid taxon. These findings will have important implications for conservation and management of crocodiles in the Indo-Pacific region.