The familiar increase of peripheral blood volume as a result of alcohol leads one to expect that very intoxicated persons cool more rapidly. Figure 14[28] shows an experiment from which we may conclude that _actually acceleration of the cooling_ does set in after partaking _liberally of alcohol before the experiment_. It is very remarkable that in such an experiment, _the only exception among all cooling experiments_, irregularity was absent in a cooling to 28.1° C. [82.6° F.]. Even if it was not possible to reproduce this apparent protection against irregularity caused by partaking of alcohol in control experiments on other subjects, there remains the possibility that the distending of the peripheral vessels delays the overloading of the heart, just as on the other hand it increases the speed of cooling. Our observations contradict the old seafaring custom of pouring _alcohol_ into a person _already cooled_, since, according to these observations the temperature tends, even in slight degrees of cooling, to sink further for a long time after rescue. As long as there is no active supply of heat from outside, the disadvantage of an increased heat loss will reduce the utility of stopping the peripheral vessel blockage. Also in _later stages_ of recovery one must obviously be very careful in giving alcohol; above all, this warning is emphasized by the possibility that one must reckon with a total irregularity after more than an hour, which may go unnoticed by the inexperienced experimenter. * * * * * _VIII. Preventive Measures_ * * * * * _IX. Concerning Life Jackets_ [_Schwimmwesten_] * * * * * _X. Summary_ 1. The curve of rectal temperature of human beings chilled in water of 2° C. [35.6° F.] to 12° C. [53.6° F.] shows a gradual drop to about 35° C. [95° F.], after which the drop becomes rapid. Death may occur at rectal temperatures below 30° C. [86° F.]. 2. Death results from heart failure. The direct damage to the heart becomes evident from the total irregularity observed in all cases, setting in at approximately 30° C. [86° F.]. This cardiac damage is due to overloading of the heart, caused by the marked and regular increase in the viscosity of the blood, as well as by the marked throttling of large peripheral vascular areas; besides, a direct injury to the heart by the cold is also probable. 3. If the neck is also chilled, the lowering of the temperature is more rapid. This is due to interference with the temperature-regulating and vascular centers; cerebral oedema also makes its appearance. 4. The blood sugar rises as the temperature falls, and the blood sugar does not drop again as long as the body temperature continues to fall. This fact suggests an intermediary disturbance of metabolism.