Cryptic species of nomadic birds in the red crossbill (Loxia curvirostra) complex of North America

Studies of vocalizations, morphology, and allozymes in North American red crossbills (Loxia curvirostra) suggest that the complex is not a single species and actually contains at least eight cryptic species. Vocalizations of crossbills recorded from widely separated regions of North America fell into eight distinctive types, with each individual giving calls of a single type. Geographic distributions of call types overlapped broadly, showing that call types do not represent dialect populations or geographic races. External and skeletal measurements showed that each form occupied a narrow range of the total morphological variation in the complex, and although most forms were morphologically separable, some were nearly identical. All forms showed low levels of within-group morphological variation and similar levels of sexual size dimorphism. Character wear, especially in bill length, was an important contributor to within-form variation, but size differences due to age were not. Measurements on 2979 museum specimens of crossbills showed that the set of individuals of known vocalizations was a good representation of the total morphological variation in the North American complex. North-south clinal variation in Loxia curvirostra, which is apparently anti-Bergmann's rule, is explained as variation in frequencies of cryptic species across latitude. The cryptic species were found to be aligned on allometric evolutionary trajectories with bill size changing more rapidly than body size. Bill and body size differences were correlated to ecological differences in the complex. Little variation in plumage coloration was found among the different forms. Electrophoretic analysis of 561 L. curvirostra and 16 white-winged crossbills (L. leucoptera) at 35 genetic loci showed only slight differentiation within the genus, and forms of red crossbill were nearly identical in allozyme frequencies. Cryptic species of red crossbills are probably less than 100,000 years old. Speciation in the group appears to involve adaptive diversification in morphology and cultural evolution in vocal behavior and ecology. Nomenclature in the complex will require extensive revision.