Secular trend in reported sperm counts.

Claims that there has been a recent secular decline in splerm counts have been disputed. Protaglonists on both sides of the discussion have based their arguments on data from a few large samples at different times and (mostly) in different places. It is not disputed that the differences between the mean counts in the various samples are statistically significant: what is a tissue is whether a secular decline is responsible. The present not adops a different approach to the problem: an attempt has been made to locate representative data on mean sperm counts of unselected men over the last 45 years. Twenty-nine such means are cited here. There can be no reasonable doubt that these reported mean sperm counts show a decline with time of publication, at least since 1960. It is not easy to see what artifact - other than a real secular trend - might responsible. Accordingly it is concluded that - at least in some places - it seems likely that a secular decline has occurred.