Batches of Greenfinches were held for variable periods on daily light schedules of 7L (seven hours light) - 17D (17 hours dark) (6L+ 1L - 17D) the time when the lights initially switchedon (dawn) being kept constant. They were subsequently given new schedules whereby they received the 6rst 6 hours photostirnulation as before, but were given the seventh hour at variable times throughout the scotophase according to group. In this way cycles which ranged from 6L - fD - 1L - 16+D to 6L - 11D - 1L - 6D were given. It has already been established that a cycle of the former kind stimulates a peak in LH secretion, without spermatogenesis occurring. It is now shown that spermatogenesis could be stimulated with cycles in which the one-hour light phase occurred 91/2 to 18 hours from dawn, peak spermatogenesis being found with a cycle of 6L - 7D - 1L - 10D. Such a cycle does not correspond with much LH activity.

The testicular histology of all specimens is described and discussed and the conclusion made that LH and FSH are produced at different phases of a circadian type rhythm(s). In consequence, short constant photoperiods (winter days) may only engage the LH sensitive phase leading to interstitial cell rehabilitation, whereas long days engage both LH and FSH phases leading to spermatogenesis. With the special conditions of the experiment it was possible to engage one secretory phase to the partial exclusion of the other and this led to some unusual histological manifestations. Thus in some groups a Sertoli cell steatogenesis occurred while spermatogenesis was in progress, and in others there was an accumulation of primary-spermatocytes in very large tubules. This last condition was considered to result from gonadotropin secretion in the absence of an adequately prepared interstitium so that insufficient androgen was available to facilitate the maturation of the later germ cell stages. This aspect is discussed in detail, as is also the significance for birds of circadian based photoresponses of the Kind demonstrated.