Denervation stimulates quiescent satellite cells in skeletal muscle to reenter the cell cycle. In the androgen-sensitive rat levator ani muscle (LA), this mitotic response to loss of neural input fails to occur in castrated animals. To elucidate the role of androgens in denervation-induced satellite cell proliferation, the denervated LA of castrated rats (Group A) was compared with that of animals infixed with testosterone implants after castration (Group B). Mean myofiber cross-sectional areas (Group A: 362.95 m 2 Ϯ 27.74; Group B: 403.13 m 2 Ϯ 53.87) and linear nuclear densities (Group A: 74.07 mm Ϫ1 Ϯ 17.58; Group B: 104.13 mm Ϫ1 Ϯ 4.06) were similar (P Ͼ 0.05) in both groups. The androgen-deprived myofibers of Group A, however, had a significantly lower nuclear content (271.0 Ϯ 74.91 vs. 1,285.80 Ϯ 81.74 in Group B; P Ͻ 0.05) on account of their considerably shorter mean length (3.44 mm Ϯ 0.29 vs. 12.31 mm Ϯ 0.92 in Group B; P Ͻ 0.05). The proportional representation of satellite cells in hormone-replaced, denervated muscle was more than twice that in the untreated group (Group B: 5.15 Ϯ 0.83% vs. Group A: 2.28 Ϯ 0.23%; P Ͻ 0.05). In absolute terms, the satellite cell number in Group B was approximately an order of magnitude greater than in Group A (408.4 ϫ 10 3 vs. 38.08 ϫ 10 3 ). The results confirm the absence of testosterone as the factor responsible for the inability of satellite cells in the LA of castrated rats to respond mitotically to the withdrawal of neural input after denervation.