1. Between January 24th and April 22nd, 32 male starlings, with females, were subjected to (a) no added light (controls), (b) 1.7 foot candles of filtered red light together with the total light from a General Electric Company's "sunlamp" at about 3 ft. distance, approximating to midsummer mid-day sunlight, and (c) 1.7 foot candles of filtered red light, after dark for 6 hours, in addition to a 9½ hours constant daily daylight period (about 18 min. reduction from the daylight period in nature at the beginning of the experiment).

2. Samples from (a), (b) and (c) groups were killed for study after 18, 25, 32 and 60 days, and from (c) group also at 88 days.

3. Testes of controls (a) underwent a slight regression, remained smaller than at the first killing, and did not increase in germ-cell or other gonadal activity.

4. Testes of birds in (c) group were slower to reach their maximum size and spermatogenesis (60 days) than were those from (b) group (32 days), with their more intense light and approximately double the amount of red rays. Both sets of birds failed to maintain their maximum size and spermatogenesis, though the more slowly reacting (c) group maintained it longer than did the (b) group.

5. The reaction differences are such as to indicate that differences in light intensity and amount of red rays caused the observed differences in relative length of the phases of the sexual cycle and, particularly, in the length of periods of maximum testis size and activity rather than any specific action of ultra-violet rays or excess of green light.

6. Even on the highly stimulating exposures to light used, the duration of maximum spermatogenic activity was limited and varied with different birds, with different exposures to light, and with different types of light, and the consequent rates of germ-cell activity. The duration of any phase of the sexual cycle bore an inverse relation to the effectiveness of the stimulating exposure to light, which, in turn, depended on the daily period, intensity, wave-length and rate of change of illumination.

7. Alternative modes of causation of these phenomena and the probable mediation of the anterior lobe of the hypophysis in the reactions are discussed.

"Studies on the Sexual Cycle in Birds; No. VIII." Aided by grants from the Committee for Research in Problems of Sex and the Committee for Grants-in-Aid of the National Research Council of the U.S.A. for 1930-2, administered by T. H. Bissonnette. Facilities for study were also received from the School of Agriculture and the Laboratory of Experimental Zoology of the University of Cambridge. For all of these aids the authors are very grateful.