Did stick insects really regain their wings?

In a recent issue of Nature, Whiting et al. (2003) report a surprising and novel result: the re-evolution of insect wings from wingless ancestors in Phasmatodea (stick insects). If true, this represents an important advance in our understanding of insect and gene evolution. It overturns a long-held notion that wings evolved just once in insects and have been lost many times. The allegedly re-evolved wings have exactly the same detailed structure as other insect wings (Whiting et al., 2003, their Fig. 1b) and if this feature truly has been regained in stick insects it would demonstrate a remarkable maintenance of ‘wing genes’ despite inactivation of the developmental pathway over long periods of evolutionary time. Whiting et al.’s conclusions have already been promulgated in the popular literature (New Scientist, Scientific American, The New York Times, Anon, 2003a, b, c) but before this extraordinary evolutionary scenario reaches the entomology textbooks a re-examination is in order. Two lines of evidence are essential to Whiting et al.’s wing re-evolution hypothesis: (1) the topology in their stick insect phylogenetic tree, and (2) the distribution of wing/wingless transformations on that tree. The latter issue, character optimization, is addressed here. Our reappraisal of the evidence convinces us that Whiting et al. have overstated significantly the probability of wing re-evolution in stick insects. When this is taken into account we see no grounds for overturning the traditional view of stick insect evolution.