Concern about the effect of environmental changes on male reproductive health has grown in recent years to become a major preoccupation in some developed countries. A possible decline in human sperm concentration was suggested in the early seventies following studies in the US. In 1992 a meta-analysis of 61 articles published by Carlsen et al. concluded that the mean sperm count of healthy men had declined by 1% per year over the previous 50 years. From 1995 and onwards, some retrospective, longitudinal analyses of the sperm count of fertile or infertile men contradicted this while others did not. The demonstration of a geographical variation in sperm concentration, between and within countries or regions, appears to be less controversial. The amplitude of the difference observed cannot only be explained by methodological or confounding factors, and must to some extent be attributed to ethnic, genetic or environmental factors. As many of the published studies suffer from imprecision regarding the description of population characteristics and confounding factors, and were not designed with controlled and standardised methodology, the debate remains open. Prospective studies in well-defined cohorts of men in various populations are required to evaluate the potential effect of external factors on male reproductive health. These studies should not be limited to the analysis of sperm concentration, as this may not be the best biomarker of testis function and human fertility.