Unilateral testicular torsion leads to bilateral testicular injury: most patients have sperm abnormalities after testicular torsion. Histological changes in the contralateral testis have been interpreted either as a consequence of a predisposing testicular pathology, or to a noxious effect from the twisted testis. 17 patients between the ages of 14–34 years (20·7 [SD 6·1]) were operated on to alleviate unilateral testicular torsion. Bilateral testicular biopsy specimens were taken, fixed in 1% glutaraldehyde, embedded in Epon, and semithin histological sections were stained with toluidine-blue. The duration of the time of torsion before surgery ranged from 0·5–11 hours (4·2 [3·0]). Testicular torsion injuries are ranked into three pathological grades, independent of the degree to which the spermatic cord was twisted, or the duration of the torsion. The biopsies of the patients with twisted testes were compared with specimens from testes from three patients who had undergone a surgical procedure for suspected infertility and whose testicular histology was judged to be normal. Extensive degradation of DNA is characteristic in the early stages of apoptosis and is preferentially labelled by TUNEL (TdT-mediated dUTP Nick End Labelling). All biopsy specimens were labelled with TUNEL. All the contralateral testicular biopsy samples of twisted testes that had torsion-injuries of grade 2 or 3 revealed atrophic Leydig cells, malformation of the late spermatids, often binuclear spermatids, reduction in the number of late spermatids, and pathological changes in Sertoli cells. Apoptosis was enhanced in all the contralateral testes that were examined particularly in spermatocytes, early and late spermatids, and to a lesser extent in Sertoli cells (figure). In contrast, spermatogonia, peritubular connective tissue (fibroblasts and myofibroblasts), and endothelial cells seldom underwent apoptosis. Leydig cells were affected less often than spermatocytes. Apoptosis seldom occurred in Sertoli cells of the testes of those in the control group although some spermatids and spermatocytes were occasionally apoptotic. Both the extent of apoptosis in the contralateral testis and degree of the necrotic change within the twisted testicle correlated with duration of the torsion. The notion that the twisted testis has the ability to shutdown a contralateral testis has interesting evolutionary implications. Therefore, apoptosis and the cytokines which activate it have evolved to eliminate the cells which are ostensibly at risk. We proffer the hypothesis that the increase in apoptosis in the contralateral testis in patients with testicular torsion is a consequence of a breakdown in the blood-testis barrier of the twisted testis. The structural changes in spermatids and Sertoli cells may be the result of this phenomenon; but it can not be the cause of Sertoli cell only syndrome in the contralateral testes observed in 7 of 50 patients with testicular torsion. Thus pre-existing changes, either congenital or as a cause of subtle chronic injuries (possibly subtorsions), also may cause impaired fertility. In conclusion, extensive apoptosis is a phenomenon that was observed regularly in the germinal epithelium of the contralateral testis of patients with testicular torsion. Other cells also may be affected; these include Sertoli cells, and, to a lesser degree, the Leydig cells. However, capillary endothelium, connective tissue, and peritubular fibroblasts rarely were affected.