Spectrophotometers are used for estimating sperm concentrations from raw ejaculates in semen processing laboratories. Unfortunately, these instruments have a limited detection spectrum and do not permit accurate quantification of sperm numbers in highly diluted or concentrated samples. The objectives of this study were to validate a DNA assay for quantification of sperm numbers in extended or undiluted semen samples and to determine precision of the assay. The principle of the assay is based upon a fluorescent dye that binds to adenine-thymine base pairs in double-stranded DNA. Semen samples and calf thymus DNA standards were sonicated in 2 M NaCl buffer with 1 mM EDTA. The DNA content of samples was compared to standards of calf thymus DNA using fluorometry. Sensitivity of the assay was determined to be 1.4 x 10(5) sperm cells. Concentrations of sperm estimated from DNA assay values did not differ from flow cytometric cell counts. Assays were performed in three different laboratories, using different equipment, to assess the assay's repeatability. Estimates of sperm concentrations determined by the DNA assay were similar, regardless of location and source of equipment used to perform the assays. This assay fulfills statistical criteria for being sensitive, accurate, and repeatable, and it can be employed in laboratories processing semen for artificial insemination as a tool for spectrophotometer calibration, a check for straw filling accuracy, or to quantify sperm numbers in extended, packaged semen.