Taxonomy is an imprecise science that delimits the evolutionary continuum into discrete categories. For marine mammals, this science is complicated by the relative lack of morphological data for taxa that inhabit remote and often vast ranges. We provide guidelines to promote consistency in studies relying primarily on molecular genetic data to delimit cetacean subspecies from both populations and species. These guidelines identify informational needs: basis for the taxonomic hypothesis being tested, description of current taxonomy, description of relevant life history, sample distribution, sample size, number and sequence length of genetic markers, description of measures taken to ensure data quality, summary statistics for the genetic markers, and analytical methods used to evaluate the genetic data. We propose an initial set of quantitative and qualitative standards based on the types of data and analytical methods most readily available at present. These standards are not expected to be rigidly applied. Rather, they are meant to encourage taxonomic arguments that are consistent and transparent. We hope professional societies, such as the Society for Marine Mammalogy, will adopt quantitative standards that evolve as new data types and analytical methods become widely available.

Uncertainty in marine mammal taxonomy is increasingly being addressed using molecular genetic data. We examined 32 peer‐reviewed articles published between 1994 and 2011to review methodological practices, consistency of markers and analytical methods, and overall quality of arguments used when genetic data have been employed to delimit new species and subspecies of marine mammals. The mitochondrial DNA (mtDNA) control region was the primary genetic marker used in these studies, but analytical methods varied greatly across studies. Diagnosability, a common metric for delimiting subspecies with morphological data, was only used through citing of fixed differences in mtDNA sequences. Assignment tests based on microsatellite data were less common but were applied at both taxonomic levels. Nuclear DNA sequence data were rarely used. Basic background material needed to evaluate the strength of arguments, such as distribution and sampling maps, were often missing. For most studies, sample sizes were good, but adequate geographic sampling for broadly distributed taxa was often lacking, diminishing the strength of evidence for taxonomic distinctness. Examining these empirical cases revealed a mixture of sound and inadequate practices for genetic studies of cetacean taxonomy and suggested that improvements could be made to the field by developing standard guidelines.

Mitochondrial DNA (mtDNA) control region sequences and microsatellite loci length polymorphisms were used to investigate genetic differentiation in spotted dolphins (Stenella attenuata) in the Eastern Tropical Pacific and to examine the intraspecific structure of the coastal subspecies (Stenella attenuata graffmani). One-hundred and thirty-five animals from several coastal areas and 90 offshore animals were sequenced for 455 bp of the mitochondrial control region, resulting in 112 mtDNA haplotypes. Phylogenetic analyses and the existence of shared haplotypes between the two subspecies suggest recent and/or current gene flow. Analyses using v 2 , F ST (based on haplotype frequencies) and F ST values (based on frequencies and genetic distances between haplotypes) yielded statistically significant separation (randomized permutation values P<0.05) among four different coastal populations and between all but one of these and the offshore subspecies (overall F ST =0.0691). Ninety-one coastal animals from these four geographic populations and 50 offshore animals were genotyped for seven nuclear microsatellite loci. Analysis using F ST values (based on allelic frequencies) yielded statistically significant separation between most coastal populations and offshore animals, although no coastal populations were distinguished. These results argue for the existence of some genetic isolation between offshore and inshore populations and among some inshore populations, suggesting that these should be treated as separate units for management purposes.