Locusts exhibit an extreme form of phenotypic plasticity and can exist as two alternative phenotypes, known as solitarious and gregarious phases. These phases, which can transform from one to another depending on local population density, show distinctly different behavioural characteristics. The proximate mechanisms of behavioural phase polyphenism have been well studied in the desert locust Schistocerca gregaria and the migratory locust Locusta migratoria, and what is known in these species is often treated as a general feature of locusts. However, this approach might be flawed, given that there are approximately 20 locust species that have independently evolved phase polyphenism. Using the Central American locust Schistocerca piceifrons as a study system, we characterised the time course of behavioural phase change using standard locust behavioural assays, using both a logistic regression-based model and analyses of separate behavioural variables. We found that for nymphs of S. piceifrons, solitarisation was a relatively fast, two-step process, but that gregarisation was a much slower process. Additionally, the density of the gregarisation treatment seemed to have no effect on the rate of phase change. These data are at odds with what we know about the time course of behavioural phase change in S. gregaria, suggesting that the mechanisms of locust phase polyphenism in these two species are different and may not be phylogenetically constrained. Our study represents the most in-depth study of behavioural gregarisation and solitarisation in locusts to date.
