Study design
We used data originally collected in Dennis et al. 2011. This included photographs of second and third instar D. pulex exposed to six different concentrations of C. flavicans kairomone, including a control (0, 0.1, 0.25, 0.5, 0.75 and 1µl ml-1). Animals were photographed at the second and third instar because this is when the defence is usually expressed the most . For each treatment, third-generation mothers of at least their third brood had been exposed to the relevant cue concentration and offspring of the subsequent brood were transferred to glass jars which contained 50ml of hard artificial pond water, food (2 x 105 cells ml-1 of algae) and the appropriate concentration of predator cue. Animals were transferred to a new jar containing fresh media and cue daily, until they reached maturity. Photographs of the animals were taken at the second and third instars.
Three genotypes were analysed for shape plasticity from this dataset. Two (clones Chardonnay and Cletus) showed a characteristically low level of induced morphology and one (clone Carlos) a characteristically high level. Animals from second and third instars were pooled so the data formed a factorial design of six predation risk levels x three genotypes. There were 5-10 replicate photos (i.e. replicate individuals) at each level of predation risk for each clone (totalling 518 photos).