Duclaux has called attention to a fact which is of importance for the investigation of the upper temperature limit for the life of organisms. According to this author it is erroneous to speak of a definite temperature as a fatal one; instead we must speak of a deadly temperature zone. This is due to the fact that the length of time which an organism is exposed to a higher temperature is of importance. Duclaux quotes as an example a series of experiments by Christen on the spores of soil and hay bacilli. The spores were exposed to a stream of steam and the time determined which was required at the various temperatures to kill the spores. It took at 100° over sixteen hours " " " 105-110° two to four hours " " " 115° thirty to sixty minutes " " " 125-130° five minutes or more " " " 135° one to five minutes " " " 140° one minute In warm-blooded animals 45° is generally considered a temperature at which death occurs in a few minutes; but a temperature of 44°, 43°, or 42° is also to be considered fatal with this difference only, that it takes a longer time to bring about death. This fact is to be considered in the treatment of fever. It is generally held that death in these cases is due to an irreversible heat coagulation of proteins. According to Duclaux, it can be directly observed in micro-organisms that in the fatal temperature zone the normally homogeneous, or finely granulated, protoplasm is filled with thick, irregularly arranged bodies, and this is the optical expression of coagulation. The fact that the upper temperature limit differs so widely in different forms is explained by Duclaux through differences in the coagulation temperature of the various proteins. It is, _e. g._ known that the coagulation temperature varies with the amount of water of the colloid. According to Cramer, the mycelium of _Penicillium_ contains 87.6 water to 12.4 dry matter, while the spores have 38.9 water and 61.1 dry substance. This may explain why the mycelium is killed at a lower temperature than the spores. According to Chevreul, with an increase in the amount of water, the coagulation temperature of albuminoids decreases. The reaction of the protoplasm influences the temperature of coagulation, inasmuch as it is lower when the reaction is acid, higher when the reaction is alkaline. The experiments of Pauli show also a marked influence of salts upon the temperature of coagulation of colloids. The process of heat coagulation of colloids is also a function of time. If the exposure to high temperature is not sufficiently long, only part of the colloid coagulates; in this case an organism may again recover.