OK, it took me a while to notice this paper:
Evolutionary Paths Underlying Flower Color Variation in Antirrhinum
Annabel C. Whibley,1* Nicolas B. Langlade,1* Christophe Andalo,2 Andrew I. Hanna,3 Andrew Bangham,3 Christophe Thébaud,2 Enrico Coen
Science 18 August 2006:
Vol. 313. no. 5789, pp. 963 – 966
To understand evolutionary paths connecting diverse biological forms, we defined a three-dimensional genotypic space separating two flower color morphs of Antirrhinum. A hybrid zone between morphs showed a steep cline specifically at genes controlling flower color differences, indicating that these loci are under selection. Antirrhinum species with diverse floral phenotypes formed a U-shaped cloud within the genotypic space. We propose that this cloud defines an evolutionary path that allows flower color to evolve while circumventing less-adaptive regions. Hybridization between morphs located in different arms of the U-shaped path yields low-fitness genotypes, accounting for the observed steep clines at hybrid zones.
Antirrhinum is the genus of snapdragons. And the paper is all about evolvability at basis. What the research shows is that there exist high-fitness paths between magenta and yellow flower morphs in snapdragons, even though direct hybridization produces a low-fitness orange flower. What would appear to be two isolated fitness peaks in one view of a fitness landscape are actually connected in a high-fitness “cloud” when visualized in a higher dimensionality. While this may seem pretty obvious from a theoretical perspective, what this paper provides is an empirical demonstration of an instance where both the low-dimensional view of the situation argues for isolated fitness peaks and the high-dimensional view shows that there is no actual isolation and that there does exist a continuous high-fitness path linking the disparate genotypes in snapdragons in Southern Europe. So when an antievolutionist argues for “isolated peaks”, one can usefully refer them to this work and inquire as to whether the situation being discussed has been studied in sufficient detail to exclude the higher-dimensioned connectedness of high-fitness genotypes demonstrated here.