clearly-definable behaviours that could be subject to a mutational screen had immense heuristic value and influenced generations of neurobiologists. The pioneer British behaviour geneticists whose Drosophila work immediately preceded that of Benzer, in particular Bastock and Manning, are sometimes forgotten, partly because the complex, naturalistic behaviours they studied, and the theoretical frameworks they employed, seem to have been less amenable to the kind of explanations sought by Benzer and his followers. It could even be argued that, although Bastock was the first to identify how a gene could alter a behaviour [17,20], their work led to a dead end. This latest paper [1] shows this was not the case. The beautifully detailed research by McKellar et al. [1] indicates that one of the most complex and integrated behaviours shown by Drosophila, courtship, is slowly beginning to reveal its cellular substrates. This paper, and the model it presents of how sequential actions may be encoded, suggests a general framework for investigating stereotypical behaviour in Drosophila species and indeed in other animals. As well as developing this model in males, the next challenge will be to employ it to investigate something that is far less well understood, and which Bastock and Manning repeatedly emphasised all those years agocourtship involves female behaviour, too.