The modified Arneth Count of ninety healthy students was examined and the average weighted mean was found to be 2•628, and the standard deviation 0–19. When this was compared with the results previously obtained by Cooke and Ponder on a population selected by a rigid health standard, a significant but small difference was found which was not considered to upset the validity of the earlier work. The reasons for this are discussed.This investigation was carried out during my tenure of a Beit Memorial Research Fellowship. The cost of the experiments was defrayed by a grant from the Ella Sachs Plotz Foundation.