A study of dog-coyote hybrids and implications concerning hybridization in canis.

A female, mongrel, black and white, 25-pound terrier dog bred to a captive male Kansas coyote gave birth on 20 April 1954 to six F1 hybrid “coy-dogs” From 1956 through 1961 these produced four litters of F2 hybrids. Five of the F1 generation resembled melanistic, short-legged coyotes; the sixth was similarly colored but shaggy like the mother. The F2 generation was more varied, dog-like to somewhat coyote-like animals. Behavior varied, but all of the animals were intermediate, with some coyote-like traits, including howling. They were aggressive among themselves and had a dominance hierarchy in males and probably in females, with males dominating females. They displayed no submissive behavior adequate to inhibit a dominant individual from attacking an inferior. No trace of male parental care (which is strong in coyotes and absent in dogs) was noted in the coy-dogs. The generally small size of the F2 litters (mean 2.25, range 1 to 3) and their small number in relation to the possible number, suggests some decrease in fecundity but might also have resulted from crowding or other suboptimal conditions inducing prenatal mortality. There was a rather high incidence of dental anomaly among the hybrids. Both sexes of the hybrids, as in the comparatively few comparable studies, displayed a late autumnal, annual mating season (in this case in December), differing from that of coyotes, which breed in late winter (mainly February). The literature of dog-coyote hybridism is summarized and the implications of the peculiar reproductive timing of hybrids are discussed with respect to the possibility of hybridization in the wild leading (a) to the establishment of hybridswarms of coy-dogs and intermediate individuals; (b) to the introgression of dog genes into coyote populations; and (c) to the probable source of the extremely large size and remarkably great variation of a population of wild, coyote-like animals recently established in New England. It is concluded that the phase shift in the breeding season of F1 hybrids, requiring the young of animals presumably less fit than coyotes to be reared in midwinter and in the absence of male parental care should form an effective, if not absolute, block to the development of hybrid swarms and to the introgression of dog genes into the coyote gene pool. Application of discriminant functions analysis—here applied also to various hybrids—to a series of wild Kansas coyotes (which are as much exposed to dogs as any coyotes) shows that these specimens have none of the unusual variability characterizing the New England population. It is therefore concluded that the variability and large size of the New England canids result from the introgression of wolf genes, probably in Ontario, into coyote stock. Recent proof that these species can in fact hybridize is cited. Also, two captive-reared coyote-wolf hybrids, long ago reported but later widely over-looked, seem likely in fact to represent this cross if judged by discriminant functions analysis and by their small size.