An examination of species limits in the Aulacorhynchus "prasinus" toucanet complex (Aves: Ramphastidae).

The number of species recognized in Aulacorhynchus toucanets has varied tremendously over the past century. Revisors seem to disagree on whether head and bill coloration are useful indicators of species limits, especially in the A. "prasinus" complex. Using morphometrics, I tested the hypothesis that the major color-based subspecific groups of A. "prasinus" sensu lato are simply "cookie-cutter" (i.e., morphologically nearly identical) toucanets with different head and bill colorations. Univariate and multivariate analyses show that they are not simply morphological replicates of different colors: a complex array of morphometric similarities and dissimilarities occur between the major subspecific groups, and these variations differ between the sexes. Latitude and longitude had a small but significant association with female (but not male) PC1 and PC2. Hybridization and intergradation were also considered using plumage and bill characters as a surrogate to infer gene flow. Hybridization as indicated by phenotype appears to be substantial between A. "p." cyanolaemus and A. "p." atrogularis and nonexistent between other major groups, although from genetic evidence it is likely rare between A. "p." albivitta and A. "p." cyanolaemus. The congruence and complexities of the morphological and color changes occurring among these groups suggest that ecological adaptation (through natural selection) and social selection have co-occurred among these groups and that species limits are involved. Further, hybridization is not evident at key places, despite in many cases (hypothetical) opportunity for gene flow. Consequently, I recommend that this complex be recognized as comprising five biological species: A. wagleri, prasinus, caeruleogularis, albivitta, and atrogularis. Four of these also have valid subspecies within them, and additional work may eventually support elevation of some of these subspecies to full species. Species limits in South America especially need more study.