Several recent articles have reported the presence of a gammaretrovirus, termed "XMRV" (xenotropic murine leukemia virus-related virus) in prostate cancers (PCa). If confirmed, this could have enormous implications for the detection, prevention, and treatment of PCa. However, other articles report failure to detect XMRV in PCa. We tested nearly 800 PCa samples, using a combination of real-time PCR and immunohistochemistry (IHC). The PCR reactions were simultaneously monitored for amplification of a single-copy human gene, to confirm the quality of the sample DNA and its suitability for PCR. Controls showed that the PCR assay could detect the XMRV in a single infected cell, even in the presence of a 10,000-fold excess of uninfected human cells. The IHC used 2 rabbit polyclonal antisera, each prepared against a purified murine leukemia virus (MLV) protein. Both antisera always stained XMRV-infected or -transfected cells, but never stained control cells. No evidence for XMRV in PCa was obtained in these experiments. We discuss possible explanations for the discrepancies in the results from different laboratories. It is possible that XMRV is not actually circulating in the human population; even if it is, the data do not seem to support a causal role for this virus in PCa.