OBJECTIVE Evaluation of the testis biopsy specimen in male infertility has been hampered by the qualitative rather than quantitative nature of routine histological techniques and interpretation. Although useful, flow cytometry has demonstrated significant limitations due to its inability to differentiate among haploid cell types, spermatozoa and spermatids. Thus, deoxyribonucleic acid image analysis was performed on testis biopsy specimens to combine the advantages of quantitative ploidy and morphological characteristics to develop a more effective quantitative system. METHODS Image analysis using a specifically designed filter was performed on Feulgen stained 5 microns. sections of paraffin embedded testicular tissue. The archival testicular tissue had been obtained using standard biopsy techniques from patients with azoospermia and normal spermatogenesis on routine hematoxylin and eosin processed tissue. RESULTS The overall findings (mean plus or minus standard deviation) in 20 biopsies from 18 men are haploid 54.2 +/- 5.6, diploid 31.2 +/- 5.2 and tetraploid 15.0 +/- 4.6%. Spermatozoa and spermatids contributed 23.2 +/- 5.7 and 30.9 +/- 7.6% to the total cell content, respectively. A mean of 2,462 +/- 254 cells were counted per case. CONCLUSIONS We used image analysis to evaluate testicular biopsies. This technique can quantitatively determine the percentages of various cell types within the seminiferous tubules. Furthermore, it can differentiate between spermatozoa and spermatids based on morphological characteristics. This technique, which combines ploidy and cell morphology characteristics, represents a significant advancement in the interpretation of the testicular biopsy, and has implications for redefining the presently used system of qualitative interpretative assessment.