Developmental anatomy of the Drosophila brain: neuroanatomy is gene expression.

On-line databases of anatomical information are being compiled for a number of genetically manipulable organisms, including the fruit fly, Drosophila melanogaster. Based on the success of the molecular databases that preceded them, they face formidable problems in data cataloguing, storage, and retrieval. The prospect for such databases, which is apparent already, is to alter permanently the approach to neuroanatomy in such species. Experience with Drosophila indicates the possibility to arbitrate controversies over and, in some cases, to redefine the borders drawn in the brain by conventional neuroanatomical methods. Two publications in this issue of Journal of Comparative Neurology by Nassif et al. and by Hartenstein et al. highlight one of the first demonstrations of a further opportunity in Drosophila. In some cases, it is already possible to suggest how individual cellular elements--neurons, tracts, and neuropil regions--might be traced from the time when they first express a precocious marker, such as the product of the fasciclin-II gene, through the metamorphic pupal stage, and into the adult. In this way, it becomes possible to identify the structures of the adult brain from the time of their first emergence in the embryo and to follow their transitional positions throughout the course of development. Critical in this process is the neuroanatomical organization of the larval brain, which contains not only the fully functional central nervous system of the larva but also the developing elements of the adult brain, because this holds the key to understanding both the cellular elements that are inherited from the embryo and the elements that are in the process of forming the adult nervous system.