The effect of testosterone (T) replacement on changes in mood was studied for 60 days in 51 hypogonadal men. All patients were withdrawn from their prior T replacement for at least 6 weeks before enrollment. Of these patients, 18 received T enanthate 200 mg im every 20 days, 16 received sublingual T cyclodextrin (SLT) at a dose of 2.5 mg three times daily, and 17 received SLT at a dose of 5.0 mg three times daily. The total treatment period was 60 days. The patients were asked to respond to a questionnaire on 7 consecutive days before the start of treatment and on 7 consecutive days before their visits to the clinic on days 21, 41, and 60 of treatment. The following mood parameters were assessed using a 7-point Likert rating scale: angry, alert, irritable, full of pep (energy), sad/blue, tired, friendly, nervous, and well/good. When compared with the baseline period, T replacement led to significant decreases in anger (P = 0.0045), irritability (P = 0.0009), sadness (P = 0.0033), tiredness (P = 0.0035), and nervousness (P = 0.0291), and significant improvement in energy level (P = 0.0020), friendliness (P = 0.0072), and sense of well-being (P = 0.024) in all subjects as a group. Analyses of the area under the curve (AUC) of baseline serum T levels before T replacement showed significant positive correlations between serum T (AUC) and friendliness (r = 0.29, P < 0.05) and sense of well-being (r = 0.27, P < 0.05), and significant negative correlations with nervousness (r = -0.27, P < 0.05), irritability (r = -0.29, P < 0.05) and tiredness (r = -0.28, P < 0.05). Similar correlations were found between serum dihydrotestosterone (DHT) and some of the mood parameters. After T replacement in the hypogonadal men, these correlations between AUC of serum T levels and the positive and negative mood scores disappeared. These results were corroborated in a subsequent study in which 30 hypogonadal men were supplemented with SLT 5 mg three times daily for 6 months. The patients were less nervous (P = 0.0025) and more alert (P = 0.0004), friendly (P = 0.042), and energetic (P = 0.0001) during the 6-month treatment period compared with baseline. We conclude that T replacement therapy in hypogonadal men improved their positive mood parameters, such as energy, well/good feelings, and friendliness and decreased negative mood parameters including anger, nervousness, and irritability, and direct correlations between serum T and DHT with mood scores were only observed in the baseline period when serum androgen levels were below the normal range. The latter observation suggests that once a minimally adequate serum T/DHT level was achieved by T replacement therapy, further increases in serum T/DHT levels did not further contribute to the improvement in mood variables.