I found this paper by E. B. Taylor and colleagues from a link on evolgen. The paper is titled "Speciation in reverse", and it is about the loss of distinction between two stickleback morphs in Enos Lake, British Columbia.
From the abstract:
Bayesian analyses of population structure in a sample collected in 1994 indicated two genetically distinct populations in Enos Lake, but only a single genetic population was evident in 1997, 2000, and 2002. In addition, genetic analyses of samples collected in 1997, 2000, and 2002 showed strong signals of 'hybrids'; they were genetically intermediate to parental genotypes. Our results support the idea that the Enos Lake species pair is collapsing into a hybrid swarm.
I love this paragraph from the introduction:
The persistence of sympatric species that still occasionally hybridize implies that a dynamic balance exists between the occasional production of hybrids and their removal by natural selection. Presumably, there have been cases where completion of speciation in sympatry fails because gene flow overwhelms factors promoting divergence. Perhaps more infrequently, speciation proceeds to completion only to be undone later when environmental conditions change. Such reversals can occur, for instance, if the fitness of hybrids is suddenly improved under the new environmental conditions (e.g. Grant and Grant 2002). Alternatively, environmental cues upon which premating isolation is based may suddenly be altered, leading to a burst of hybridization that selection can no longer overcome (discussed in Coyne and Orr 2004). In either instance, the frail integrity of species that lack complete postzygotic isolation demonstrates the contribution of the environment to their maintenance and, perhaps, provides insight to the identity of factors that initiate speciation.
"The frail integrity of species that lack complete postzygotic isolation."
There is probably a reason why these particular populations have merged. The authors suspect that the introduction of nonnative crayfish may be a factor -- the crayfish eat some of the deep-water sticklebacks and may compete with them for food.
These stickleback populations have postglacial origins -- the age of Enos lake. Elsewhere, the divergence of stickleback populations in postglacial lakes has been found to be independently repeated, in terms of the origin of distinct benthic and limnetic (deep and surface) forms.
The most interesting aspect of this for me is the initial morphological divergence of the populations. They clearly have been adapting to distinct roles in these lakes even with substantial possibility of ongoing gene flow -- even if they were partially isolated by their use of different breeding strategies and water depths.
Highly adaptable. Similar morphological forms repeatedly evolve in similar habitats. Rapid speciation with ongoing genetic exchanges. Rapid collapse of morphological diversity (and genetic distinctiveness) in response to environmental change.
Taylor EB, Boughman JW, Groenenboom M, Sniatynski M, Schluter D, Gow JL. 2006. Speciation in reverse: morphological and genetic evidence of the collapse of a three-spined stickleback (Gasterosteus aculeatus) species pair. Mol Ecol 15:343. DOI link