Injections of 5-7 ,ug (6-9 nmol) of colchicine into the dentate gyrus of the hippocampus of mature rats result in widespread destruction of dentate granule cells with little, if any, damage to other cell populations, including hippocampal pyramidal cells. Selective destruction of dentate granule cells is also observed after intraventricular injections. The destructive effects of colchicine appear as soon as 12 hr after the injection and lead to the disappearance of the granule cells over a period of days. Whereas the effects on nongranule cell populations in the hippocampus appear to be reversed by approximately 11 days after injection, the granule cells are almost completely absent at long intervals after injection. At the long postinjection survival intervals the disappearance of the granule cells is accompanied by elimination of their terminal projections, the mossy fibers, as revealed by Timm staining for heavy metals. Because the preferential neurotoxic effects of colchicine do not result in morbidity or obvious behavioral debilitation, the toxicity may prove useful for studying the functional consequences of removing specific cell populations in the central nervous system.In recent years, the ablation technique for studying brain function has received much criticism due to the lack of selectivity of the electrolytic or aspiration techniques traditionally used. This lack of selectivity is a particular problem when one attempts to dissociate the effects of the removal of specific neurons from the interruption of fibers having distant origins and terminations (fibers of passage), or when one attempts to ascribe the effects of lesions to the removal of one specific population of cells contained within a structure composed of heterogeneous cell types. Because of these problems, considerable recent interest has been generated by drugs or other treatments that are selective in their toxic effects on given neuronal populations or that can destroy neuronal cell bodies without interrupting fibers of passage. Several means of inducing more or less selective lesions have been described, including x-irradiation during development (1-3) and injections of toxic analogs of presumed neurotransmitters, including norepinephrine and dopamine (4-8), serotonin (9), and glutamic acid (10-12). These selective methods have provided new insights into the behavioral, biochemical, and physiological changes associated with the removal of specific cell populations in situations in which more traditional methods for producing lesions have largely reached the limits of their usefulness (13)(14)(15)(16)(17)(18)(19). Because of the tremendous potential of methods with a selective neurotoxic effect, particularly when these effects can be wrought in mature animals, we were intrigued by an apparently selective toxicity of colchicine for granule cells of the dentate gyrus of the hippocampal formation. The present study describes this effect.The publication costs of this article were defrayed in part by page charge payment. This article...