1. For the heart rate in Pterotrachea coronata, intermediate temperatures disclose a thermal increment of 11,200 ±. This value is identical with the one reported by Crozier and Stier for the lamelli-branch, Anodonta. In the pteropod, Tiedemannia neapolitana the same temperatures typically reveal in the heart rate a µ value of 16,200 ± This agrees quantitatively with 16,300 found by Crozier and Stier for the heart of the slug, Limax maximus.
2. At high temperatures the average value of µ for Pterotrachea is 7,300: for Tiedemannia, 7,400. The corresponding averages at the lower limits are 22,000 and 23,000.
3. The great variability found near the edges of the temperature field are explicable in two ways. During intermissions characteristic of high temperatures and occurring also at low, we can assume a restorative process; while at both the upper and lower limits we may, in addition, find that reactions assume control which under ordinary circumstances never do so. Special evidence indicates that the highest temperatures employed, 27°C., and the lowest, 4°C., caused no irreversible changes in mechanism.
4. The theoretical analysis of the experimental facts makes use of Meyerhof's conception of carbohydrate metabolism and projects the cyclical nature of rhythm into the substrate of control. Assuming as a source of energy an original supply of material O, the value of 22,000 ± is assigned provisionally to a mobilization hydrolysis while 11,200 ± and 16,000 ± are attached to oxidative reactions influenced respectively by OH' and possibly Fe, or some other catalyst. The lowest value, 7,300 ± is assumed to indicate a synthetic process (lactic acid → glycogen?), possibly limited by CO2 excretion. In the present state of our knowledge, this distribution and interpretation seems to account reasonably for the experimental facts, but until we know more about the neurogenic controls, is entitled to rank only as an hypothesis.