This issue is addressed in the article:

Whether the coywolf actually has evolved into a distinct species is debated. Jonathan Way, who works in Massachusetts for the National Park Service, claims in a forthcoming paper that it has. He thinks its morphological and genetic divergence from its ancestors is sufficient to qualify. But many disagree. One common definition of a species is a population that will not interbreed with outsiders. Since coywolves continue to mate with dogs and wolves, the argument goes, they are therefore not a species. But, given the way coywolves came into existence, that definition would mean wolves and coyotes should not be considered different species either—and that does not even begin to address whether domestic dogs are a species, or just an aberrant form of wolf.

In reality, “species” is a concept invented by human beings. And, as this argument shows, that concept is not clear-cut. What the example of the coywolf does demonstrate, though, is that evolution is not the simple process of one species branching into many that the textbooks might have you believe.

This puzzled me as well. Do we not have a word for this? Is _subspecies_ really the word? In linguistics mutually intelligible but distinct `languages' are called dialects. Similarly is there a term for mutually interbreedable but distinct `species'?

Technically until the "forthcoming" paper by Jonathan Way is published, and designated type specimens exist, it isn't clear what it is -- certainly not a new species or subspecies right now.