I disagree with some of the conclusions of a letter by Dowling et al. (1992) regarding the use of characters to delineate species. The authors of the letter maintain that: (1) data cannot be used to disprove the species status of closely related taxa especially if data provides evidence of distinction; and (2) interspecific hybridization must not be regarded necessarily as an undesirable event between endangered species. Hybrids should not be uniformly condemned. The authors use our mitochondrial DNA results on the red wolf (Wayne & Jenks 1991) to provide illustrations of their points. Regarding the use of genetic data to determine species status, Dowling et al. make no mention of the intense debate surrounding species concepts (Baum 1992). The species concept that is adopted largely determines the appropriate use of genetic and characters to define species. Current philosophical extremes are represented by the phylogenetic species concept which defines species as clusters of organisms possessing uniquely shared characters (McKitrick & Zink 1988; Baum 1992) as opposed to reproductively isolated groups of interbreeding organisms, often termed the biological species concept (BSC) (O'Brien & Mayr 1991). Although other canid species we have examined can be defined under the phylogenetic species concept by mtDNA analysis, the red wolf as a separate species cannot because we find no unique mtDNA restriction site or nucleotide substitutions that unite all red wolves in a single clade (Wayne & Jenks 1991). Similarly, conclusions based on past analyses are invalid under the phylogenetic species concept because they utilize phenetic similarity rather than character state data (Nowak 1979). It remains for the morphologists to use modern phylogenetic techniques (Wiley 1981; Mayr & Ashlock 1991) to demonstrate that the red wolf can be defined by discrete character states uniquely shared by all red wolves. Under the biological species concept, evidence for reproductive isolation is often inferential and thus reproductive isolation is difficult to test. Indeed this is a common criticism of the BSC. Dowling et al. are correct in stating that only a few genetic changes may lead to speciation under this definition of species and therefore measures of divergence are not valid measures of specific status. However, in this respect, measures of divergence are not sufficient either as they may not correspond with reproductive isolation. A critical point is that both morphological and molecular techniques intend to uncover heritable differences among organisms. Dowling et al. make an unnecessary distinction in this respect between techniques which I assume they mean to involve direct assessment of DNA sequence divergence and morphologic techniques which more indirectly assess variation at the DNA level. In fact, if we accept the BSC, species status cannot be assessed directly because there are no general methods for directly assessing genetic variation in genes that govern interfertility. However, in the absence of gene flow, populations might be expected to diverge through drift in both morphology and DNA sequence. Thus, the presence of significant or molecular divergence between closely related sympatric forms can be used to infer that a long history of reproductive isolation has existed between them. Selection complicates this conclusion because it may affect the extent of and molecular differentiation.
Read full abstract
Aug 1, 2000
J.T Cook + 4
