Eight sensory structures (campaniform sensilla), appearing identical in the light and scanning electron microscopes, are found in specific locations on the wings of Drosophiia. Their axons enter one of 2 central tracts, a medial one or a lateral one. The topographic arrangement of the sensilla on the wing is not reflected in this central projection pattern. There is, however, a strict correlation between the time when a sensillum develops and the path its axon follows: The 4 sensilla whose axons form the medial projection are born and differentiate early during the development of the wing, while the other 4 sensilla, all of which project laterally, arise during a second wave of differentiation.This time-related projection pattern remains stable in the face of a variety of genetically induced alterations in the precise number and location of sensilla.The sensory systems of insects, like those of vertebrates, are frequently organized in topographic fashion (e.g., Johnson and Murphey, 1985; Murphey, 198 1;Strausfeld, 1976). Because differentiation in a sensory field often proceeds in a topographically ordered way, there can be a simultaneous correlation between the position of a sensory element in the periphery, its age, and its central projection (Murphey et al., 1980;Walsh and Guillery, 1984).We report here an analysis of the central projections of a set of 8 identified sensory elements in the wing of the fiuitfly Drosophila melanogaster, whose cuticular components appear identical by light and scanning electron microscopy, and whose cell bodies normally differentiate in a sequence that does not correspond with their position. We find that in this system the region of the CNS to which a neuron projects is related to the time of its birth and differentiation, rather than to its peripheral position. This time-related axonal distribution pattern remains stable even when the numbers or locations of the neurons contributing to it are altered by genetic manipulation. A preliminary account of these results has been published (Wigston et al., 1984).Ongoing studies indicate that the physiological properties of these receptors are not identical, in spite of the structural similarity of the cuticular elements, so that a three-way correlation between time of differentiation, central projection, and physiological function is established (Dickinson and Palka, 1985, and unpublished observations).Received Oct. 2, 1985; revised Dec. 9, 1985; accepted Dec. 18, 1985. We thank W. A. Harris. C. E. Holt. E. R. Ma-o. R. K. Murnhev. and members of our laboratory for commenting bn the ma&s&ipt; Eric &e&r painstaking selection of useful genetic variants; Usha Rani for patient technical help; and especially Peter Kareiva for carrying out the statistical analysis.