A simple and reproducible injection technique has been developed for inducing an in vivo experimental astrocytoma model in rats. Newborn rats were injected with a suspension of astrocytoma C-6 cells in doses ranging from 5 to 1 X 105 cells. The optimum dose of tumor cells was found to be between 104 and 105 cells, with a consistent 100% take rate seen above 1 X 104 injected cells. The injection of less than 104 cells resulted in a decreased ability to induce tumors, and a prolonged survival. The pathology was consistent with that of a glioblastoma multiforme. The tumor showed a diffuse infiltrating border, necrosis, and pseudopalisading. Light microscopy revealed undifferentiated tumor cells while electron microscopy demonstrated rough endoplasmic reticulum, Golgi complexes, mitochondria, and the occasional cilium and centriole. The nature of the astrocytoma C-6 cell line, and the advantages, disadvantages and possible uses of the model are discussed.