For a long time it has been widely believed that the human chrom some number was 48 and that this number applied to both men and w the controversy of the twenties and thirties regarding the presen absence of a Y-chromosome being forgotten. However, earlier this ye Tjio and Levan [1] announced that they had regularly counted chromosomes only in dividing cells of tissue cultures established from human embryos. Although the possibility of a regular loss of 2 chrom somes during embryonic development or culture was almost too fanc to be considered, they recommended that new counts should be made preparation from testicular material before the obvious conclusion drawn. This we have done.

We examined testis tissue from three males, aged 47, 53, and 63 years, respectively. The material was obtained at the Churchill Hospital, Oxford, from fresh operative specimens at the moment of removal from the body. After pre-treatment in hypotonic fluid the specimens were fixed in acetic alcohol and stained by the Feulgen procedure. Squash preparations were then made. First spermatocytes at diakinesis or meta-phase were the most suitable stages for observation and accurate counts were made on 188 cells. Of these, 174 contained either 23 bivalents, or 22 bivalents plus univalent X and Y. The remaining 14 cells contained 22 bivalents or less, and presumably had been damaged during the making of the preparations with consequent loss of one or more bivalents. No cells contained more than 23 bivalents. There were relatively few spermato-gonia in mitosis and most of these were damaged, nevertheless a few clear counts of 46 chromosomes were obtained.