We studied the efficacy of hysterosalpingography in the detection of peritubal adhesions without tubal occlusion by using laparoscopy as the gold standard for the diagnosis. Peritubal adhesions were diagnosed on hysterosalpingography by using the following radiographic criteria alone or in combination: (1) convoluted fallopian tube, (2) loculation of spillage of contrast medium into the peritoneal cavity, (3) ampullary dilatation, (4) peritubal halo effect (double-contour appearance of the tubal wall), and (5) vertical fallopian tube. The first 100 patients with the diagnosis of adhesions on hysterosalpingography who also had laparoscopy were included in the study. Seventy-five patients had the diagnosis confirmed on laparoscopy, resulting in a 25% false-positive diagnosis of adhesions with hysterosalpingography. Thirty-nine (52%) of these 75 patients had one radiographic finding alone, and 20 of these patients had loculation of spillage of contrast medium. Although the patients with adhesions tended to have two or more of the radiographic findings, the difference was not statistically significant. The most frequent radiographic findings found in combination were convoluted fallopian tube and loculation of spillage of contrast medium. These findings suggest that hysterosalpingography can be the diagnostic procedure of choice in the initial investigation of infertility due to peritubal adhesions.