Hurt et al, 2004 wrote:Two species of the endangered Sonoran topminnow, Poeciliopsis occidentalis and P. sonoriensis, occur in two isolated drainage systems in southeast Arizona, U.S.A., and are allopatric throughout their range. Although these two taxa are morphologically very similar, and have been previously described as the same species or subspecies, several molecular studies have since indicated that they differ more than their morphology suggests. To determine whether the behaviours of the two species function as premating barriers to reproduction, we investigated their mating preferences and behavioural patterns in a laboratory setting. Results from no-choice mating observations showed that the mating behaviours of the two species differ. Observations conducted during multiple-choice mating trials provided evidence of assortative mating, suggesting an early stage of premating reproductive isolation.

Lindholm & Braden, 2002 wrote:Abstract: We propose that the evolution of female preferences can be strongly influenced by linkage of attractive male traits to the Y chromosome and female preferences to the X chromosome in male heterogametic species. Such linkage patterns are predicted by models of the evolution of sexually antagonistic genes. Subsequent recombination of attractive male characters from the Y to the X would create physical linkage between attractive male trait and preference. A literature survey shows that Y linkage of potentially sexually antagonistic traits is common in poeciliid fishes and other species with sex chromosomes that are not well differentiated, but may also occur in taxa with degenerate Y chromosomes. In the guppy, attractive male traits are primarily Y and X linked; a literature review of the inheritance of sex-limited attractive male characters suggests that 16 are Y linked, 24 recombine between the X and Y, two are X linked, and two are autosomal. Crosses and backcrosses between high female preference (Endler’s live-bearers) and low female preference (Rio San Miguel) guppy populations show that this character has a strong additive genetic component and that it will be possible to investigate the physical linkage of male and female sexually selected characters in this species through mapping studies.

Magurran, 1998 wrote:

Population differentiation is often viewed as an important step towards speciation, and part of the rationale for conserving variation at the intraspecific level is that the potential to generate more biological diversity should be retained. Yet, speciation is not an inevitable consequence of population divergence. This paper reviews recent work on the Trinidadian guppy, Poecilia reticulata, a species that is renowned for its capacity for population differentiation. Guppy populations evolve rapidly, within 101 to 102 generations, as a response to changes in selection exerted by predators. The rates of evolution involved can be up to seven orders of magnitude greater than those seen in the fossil record. Sexual selection, particuarly female choice, appears to reinforce the divergence that natural selection has generated. Perplexingly, however, there is no reproductive isolation (either prezygotic or postzygotic) between populations, even those that have been separated for at least 106 generations. Sexual conflict may be the key to explaining this absence of speciation. Male reproductive behaviour, particularly the high incidence of sneaky mating, may be instrumental in producing suffcient gene flow to prevent reproductive isolation. Sneaky mating has the potential to undermine female choice, and is known to be an important means of sperm transfer in wild populations. Sexual dimorphism, also a result of sexual conflict in guppies, may inhibit speciation in another way. Morphological di¡erences between the sexes, that have arisen for reproductive reasons, mean that males and females are pre-adapted for di¡erent foraging niches. This, in turn, reduces the opportunity for the development of feeding polymorphisms, a mechanism that seems to have been important in the sympatric speciation of other fish species.

Calilasseia wrote:For example, Xiphophorus maculatus, the Platy, can hybridise with Xiphophorus hellerii, the Swordtail, even though assortative mating probably keeps the two separate in the wild.

Berthold wrote:

Calilasseia wrote:For example, Xiphophorus maculatus, the Platy, can hybridise with Xiphophorus hellerii, the Swordtail, even though assortative mating probably keeps the two separate in the wild.

This brought to my mind that I have read somewhere (sorry, can't give a source) that the females of non - sword bearing species of this genus are strongly attracted by sword bearing males. Considering that non - sword bearing species continue to exist: would you say that the researchers had an exceptional sample (or that their work was flawed in other respects)?