[308] K. Brandt (“Über den Nitratgehalt des Ozeanwassers and seine biologische Bedeutung,” _Abh. d. kais. Leop. Carol. deutsch. Akad. d. Naturfoscher._, 1915) accounts for this fact by the assumption that through the greater activity of the denitrifying bacteria in the tropical waters the amount of available nitrates is here comparatively smaller than in the polar oceans. The writer fully appreciates the importance of this fact but nevertheless is inclined also to see a limiting factor in the enormously rapid decline of the duration of life at the upper temperature limits. P. and N. Rau state that in the cold certain butterflies live longer, and similar statements exist for the silkworm, but these statements are not based on exact experiments, which are difficult. Dr. Northrop and the writer have started experiments on the influence of temperature on the duration of life of the fly _Drosophila_. Newly hatched flies were kept first without food except water and air at 34°, 28°, 24°, 19°, 14°, and 10°, and second with cane sugar. The average duration of life was as follows: _With water_ days _With cane sugar_ days 34° 2.1 6.2 28° 2.4 7.2 24° 2.4 9.4 19° 4.1 12.3 14° 8.3 10° 11.9 These experiments show that there is a definite temperature coefficient for the duration of life and that this coefficient is of the order of magnitude of that of a chemical reaction. We are continuing these experiments with animals in the presence of food. It should, however, be remembered that the fly carries with it a good deal of reserve material from the larval period. We have carried on simultaneously determinations of the temperature coefficients of the duration of the larval and pupa stage of these organisms at the same temperatures and found ratios similar to those given above for the duration of life with water only.