Imprint cytology of the gastric mucosa has been found to be very simple, inexpensive, and rapid for diagnosing Helicobacter pylori infection. However, there is a fear that preparing imprint smears may damage the biopsy specimen for subsequent histologic examination. This study was planned to investigate whether this damage happens. Four antral biopsy specimens were obtained from each of the 100 patients undergoing upper gastrointestinal endoscopy. Imprint smears were made from two biopsy specimens, which were then fixed in 10% formal saline and sent for histologic study. The third and fourth biopsy specimens were directed fixed in 10% formal saline for histologic examination. Two pathologists examined the imprint smears. Agreement between the two observers was observed in 97% of cases. Beyond-chance agreement was good with a kappa index of 0.90. H. pylori organisms were seen in 82% of biopsy specimens from which imprint smears were prepared and in the same percentage of biopsy specimens that were processed directly. The pathologists could not identify the histologic sections form which imprints were made. It is concluded that imprint cytology is an excellent method of diagnosing H. pylori infection and that preparing imprint smears does not alter the quality of the tissue. The same biopsy specimen can be used for histologic studies.