Patterns in the evolution of gall traits and gallwasp lifecycles
Our evolutionary work is aimed at revealing how and why oak gall diversity has evolved, and much of it has been done in collaboration with Dr. James Cook, at Imperial College at Silwood Park and Dr. Antonis Rokas, now in the Carroll Lab at the university of Wisconsin-Madison.
We're interested primarily in why there is such an incredible diversity of oak gall forms (follow this link to our recent review in TREE). Our current favoured hypothesis is that the evolution of this diversity has been driven by selection to escape the natural enemies (particularly chalcid parasitoid wasps such as Megastigmus stigmatizans) that kill many gallwasp. Many gall traits can be interpreted as defences: examples include - including nectar secretion and the recruiting of ant guards, coatings of spines or sticky resin, and internal airspaces and dummy chambers. Evidence that these traits have evolved through natural selection comes from the fact that all of these traits have evolved repeatedly in oak gallwasps (follow this link to Stone and Cook 1998 on gall trait evolution).
Gallwasps also show diversity in other things -particularly the parts of host oaks that they develop on, and the group of oaks they attack. Current evidence shows that moves between different parts of an oak tree (for example, from catkins to buds) have happened relatively frequently in evolution, but that shifts between different oak groups are much rarer (Cook et al 2002). If you would like to know more about patterns of evolution in all gallwasps, the best place to start is Ronquist and Liljeblad (2001).
Finally, we are also using molecular genetic techniques to study the complex lifecycles of oak gallwasps (described for all gallwasps in Stone et al 2002). Most oak gallwasps are cyclically parthenogenetic, meaning that their lifecycles involve alternation between sexual and asexual generations. Among lifecycles of this type (also found in some aphids, waterfleas and rotifers) gallwasps are unusual in having strict alternation between one generation of each type. No-one knows how this alternation is controlled. We have used phylogenetic and population genetic techniques to either match known sexual and asexual generations, or infer the existence of unknown sexual generations (Atkinson et al 2002, 2003).
