Cite?
As noted earlier, ability to interbreed is not the only criterion in classifying animals together as the same species. Not only are they unlikely to breed in the wild, they reach sexual maturity at different ages and have different amounts of estrus cycles per year.
Chronos has summed up the situation relative to pygmies and other human populations rather well. As long as two populations are linked by a continuous chain of interbreeding populations, they can be considered members of the same species. Pygmies can and do interbreed with other African populations (female pygmies often marry taller Bantu men in the areas in which they live; the reverse rarely happesn), which in turn interbreed with their neighbors, and so on. It is completely irrelevant if a pygmy has ever interbred with a white person, an Australian aborigine, or an Eskimo, since all these groups are linked by continuous series of interbreeding populations.
Regarding dogs and wolves, domestic forms are problematic as to classification, since they can’t really be said to exist “in the wild,” and their “natural” interactions with ancestral populations are uncertain. Most domestic animals have been give separate species names in the past, but the current tendency is to consider them the same species as the ancestral forms. Domestic dogs (known as Canis familiaris) that have gone feral can interbreed with both wolves (Canis lupus) and coyotes (Canis latrans), and the offspring are fertile. However, as they are known to be descended from wolves (through genetic analysis) in the fairly recent past, domestic dogs should probably just be considered a form of Canis lupus.
Incidently, I have spent some time in the company of pygmies. When I was in the Ituri forest in Zaire in 1993, I spent about a week in field camp with some, and went out net hunting with a group of 20 when they caught a duiker (forest antelope). Unfortunately, I couldn’t converse much with them, since I didn’t speak the local language and none of them spoke French.