Complete cytochrome b gene sequences from all but one species of delphinid plus four outgroups were analyzed using parsimony, maximum likelihood, and neighbor-joining methods. The results indicate the need for systematic revision of the family; a provisional classification is presented and compared to previous studies. Among the suggested revisions are removal of Orcinus from the Globicephalinae, placement of Grampus within the Globicephalinae, removal of all Lagenorbyncbus spp. from the Delphininae, and placement of Soma in the Delphininae. The genus Lagenorbyncbus is found to be polyphyletic. L. albirostris (type species for the genus) and L. acutus are not closely related to each other or to nominal congeners. L. acutus is therefore assigned to the genus Leucopleurus. The remaining four Ldgenmbyncbus species are closely related to Lissodelpbis and Cepbalorbyncbus and are placed in the genus Sagmatias. These three genera constitute the revised Lissodelphininae. Within the Delphininae, a well-supported clade includes the two species of Delphinus, Stenelld clymene, S. frontalis, S. cowuleoalba, and the aduncus form of Tursiops truncatus. Accepting the monophyly of this group renders the genera Stenella and Tursiops polyphyletic. Apart from this finding, phylogenetic resolution within the Delphininae was poor, so comprehensive taxonomic revision of this group awaits further study.

No abstract

The “stock” is the fundamental population unit of legally mandated conservation efforts, yet its formal definition in the scientific literature and in two U.S. conservation acts is varied and so general that attempts to apply it in practice are arbitrary. Because choice of stocks deserving management protection is sometimes politically contentious, improvement of the working definition is important. A key element should be the degree to which a population can be considered an evolutionarily significant unit. We propose that a hierarchial classification scheme be applied to stock designations. Category I populations, having the highest probability of being evolutionarily significant units, are characterized by a discontinuous genetic divergence pattern where locally adapted and closely related genome assemblages are separated from others geographically and by significant genetic distances. Category II populations are similarly characterized by significant genetic diversity, but with weak geographic partitioning. Category III populations are the converse of II, having little genetic differentiation between assemblages that are clearly separate and likely to be reproductively isolated. Category IV assemblages have the lowest probability of being evolutionarily significant units and are characterized by extensive gene flow and no subdivision by extrinsic barriers. In addition to phylogeographic designation, the following information is used in the classification, as indicated by single‐letter abbreviations: distribution (a), population response (b), phenotypic (c), and genotypic (d) information. Included are evidence both for and against designating population as a separate stock. In the designation “Type II a/bc,” for example, information to the right of the solidus would be evidence for “lumping,” to the left would be for “splitting.” Missing letter abbreviations would signify lack of reliable data. Note that phylogeographic designation depends on the results of selection operating to produce a locally adapted genome (indicated by differences in demographic, phenotypic, and genotypic measures) and on gene flow (indicated by differences in distribution or by movement data). Hierarchial stock categorization allows resource managers to direct limited resources to the populations most deserving of protection, that is, the populations that are most likely to be evolutionarily significant units. Using this comprehensive classification of stock allows preliminary, conservative splitting of assemblages where data are lacking without the danger that these divisions will become entrenched as biological dogma.

We quantified progesterone in 110 blubber samples from dolphins of known reproductive status in order to test the accuracy of a method to determine pregnancy status in wild cetaceans. The samples were collected from fishery‐bycaught delphinids of three species (Delphinus delphis, Lissodelphis borealis, and Lagenorhynchus obliquidens). We ascertained that blubber progesterone concentrations could clearly distinguish pregnant D. delphis (range 132–415 ng/g, mean 261 ng/g) from non‐pregnant mature and immature ones (range 0.92–48.2 ng/g, mean 15.2 ng/g). We found similar dramatic differences in L. borealis and L. obliquidens. These results were insensitive to various blubber sampling depths and anatomical sampling locations on the body, suggesting relative homogeneity of progesterone levels throughout the blubber. However, no trend was found in blubber progesterone concentration with fetal length, indicating that although blubber progesterone appears to distinguish pregnancy status, it is unlikely to differentiate pregnancy stage. Based on the findings presented here we suggest that this method, when coupled with projectile biopsy procedures, can be used to assess the pregnancy status of free‐ranging cetaceans and thus provide a new tool to determine pregnancy rates of wild populations.