Whether interspecific hybrids are fertile or sterile depends to some extent on the definition of a species. If one defines a species as organisms that look very different, then things someone has decided to call different species may actually not be, and may interbreed perfectly well - it depends on what one means by "look different" which is basically impossible to quantify. If one defines a species as a group of organisms that can interbreed productively, then by definition interspecific hybrids have to be sterile. The latter definition, while attractive (as per Ernst Mayr) does have problems in practice. What about populations of organisms that are very similar physically and perhaps genetically, but live in different regions so they never have a chance to meet. If one forces mating and the offspring are viable (and fertile), then by the latter definition these would not be different species. But for various other reasons it may make more sense to think of them as different species. Also, this species definition is by its nature inapplicable for extinct organisms, fossils and such. It seems pretty extreme to say that Tyrannosaurus and Triceratops can't be defined as different species because we have no way to test whether their hybrids could be made.
Even in living organisms, there are some intriguing situations, for example certain well-studied fish populations in certain lakes, that prefer to stay near the shore. Any fish is able to mate with fish living nearby, but fish on opposite sides of the lake are not fertile together. It's impossible to define where two (or more) different "species" boundaries are, it's more a quantitative effect of increasing genetic incompatibility in taking pairs of fish at increasing distance apart (up to half-way around). I'm not an evolutionary expert but a number of such situations have been described.