Theodosius Dobzhansky, in his essay, “On species and races of living and fossil man” (1944):
The opinion is often expressed that species and races are arbitrary categories. This opinion is false. If given the opportunity to secure the necessary data, a biologist is able in a majority of cases to decide beyond a reasonable doubt whether the forms under study are distinct species or only distinct races. Lion, tiger, leopard, and domestic cat are species ; Angora cat and alley cat are surely not species but races. However, “borderline cases”, in which it is impossible to decide whether one is dealing with species or with races, do exist. Indeed, their existence was used by Darwin to demonstrate organic evolution. If species are the primordial units of creation, or else if they arise by sudden leaps (as thought by G. St. Hilaire and recently by Goldschmidt), then we should be able to find methods to decide whether any two forms are still races or already species. If, on the other hand, species evolve gradually from races, then the decision will be possible only in some, perhaps in a majority, of cases, but at least some instances must be found in which forms are too distinct to be races but not distinct enough to be species. Evolutionists have concentrated their efforts on proving that such borderline cases do exist; by indirection they conveyed to biologists in general the impression that there are no other but borderline cases.