Satellite cells and myonuclei of neonatal rat muscles were differentially labeled with 3H-thymidine according to the procedure of Moss and Leblond (1971). Minced muscles fragments containing either labeled satellite cells or labeled myonuclei were cultured until multinucleated myotubes grew out from the explants. Reutilzation of isotope released from degenerating nuclei was competitively inhibited by using a culture medium containing excess (0.32-0.41 mM) cold thymidine. after an 8-10 day growth period, the explants were fixed and prepared for autoradiographic (ARG) examination to determine whether labeled satellite cells or myonuclei had contributed to the myonuclear population of the developing myotubes. Counts were made of the number of labeled myotubes in the explants and compared with the number of labeled satellite cells and myonuclei in samples of the original muscle tissues fixed at the time of explantation. The original muscles showed a mean satellite cell labeling index of 51.7% and gave rise to myotubes with a mean labeling incidence of 40%. In contrast, myonuclear labeling in the original muscle tissues showed no correlation with subsequent myotube labeling. Only 3.4% myotube labeling was found in explants in which over 30% of the original tissue myonuclei had been labeled. Under conditions controlled for isotope reutilization, these observations confirm results of in vivo ARG studies indicating that satellite cells are the only significant source of regenerating myoblasts in injured muscle tissue.