The evidence suggests that hybrids in hybrid zones usually are disadvantageous.

In 1989, Barton and Hewitt reviewed 170 reported hybrid zones and concluded that most of them were 'tension zones'; a hybrid zone in which the hybrids are selected against, and in which the two races on either side of it are adapted to different environments. The hybrid zone exists because of dispersal of individuals out of the area to which they are best adapted.

This being so, natural selection should have favored mating preferences. Within the zone, the mating preference should be stronger (because of reinforcement) than outside it where only one form is found. However, there is little evidence of reinforcement. In a 1985 review of 32 hybrid zones in which there was known to be partial isolation, Barton and Hewitt found evidence of reinforcement in only three of them. It is not known why reinforcement of prezygotic isolation is not more powerful.

In summary, the theoretical argument for the two stages of parapatric speciation - divergence along an environmental gradient, and reinforcement - is persuasive; but its application to natural evidence is less complete. It can usually be argued that real hybrid zones are secondary, not primary, and further work is needed to sort out how reinforcement is (or is not) operating within them.