In discussing this subject I will raise only the question whether or not the specific character of the inaugural symptoms of some infectious diseases may be due to phylogenetic association. These inaugural symptoms are measurably a recapitulation of the leading phenomena of the disease in its completed clinical picture. Thus, the furious initiative symptoms of pneumonia, of peritonitis, or erysipelas, of the exanthemata, are exaggerations of phenomena which are analogous to the phenomena accompanying physical injury and fear of physical violence. Just as the acute phenomena of fear, or those which accompany the adequate stimulation of nociceptors, are recapitulations of phylogenetic struggles, so may the inaugural symptoms of an infection be a similar phylogenetic recapitulation of the course of the disease. A certain amount of negative evidence is supplied by a comparison of the response to a dose of toxins with the response to a dose of a standard drug. No drug in therapeutic dosage except the iodin compounds causes a febrile response; no drug causes a chill; on the other hand, all specific toxins cause febrile responses and many cause chills. If a species of animal had been poisoned by a drug during vast periods of time, and if natural selection had successfully established a self-defensive response, then the administration of that drug would cause a noci-association (chemical), and a specific reaction analogous to that following the administration of Coley's toxins might be expected. Bacterial noci-association probably operates through the same law as that through which physical noci-association operates. Natural selection is impartial, however. It must be supposed that it acts impartially upon the microscopic invader and upon the host. On this ground one must infer that, in accordance with the same law of natural selection, the bacteria of acute infections have met by natural selection each advance in the struggle of the host for immunity. Hence the fast and furious struggle between man and his microscopic enemies merely indicates to what extent natural selection has developed the ATTACK and the DEFENSE respectively. This struggle is analogous to the quick and decisive battles of the carnivora when fighting among themselves or when contending against their ancient enemies. But when phylogenetically strange animals meet each other, they do not understand how to conduct a fight: natural selection has not had the opportunity of teaching them. The acute infections have the characteristics of being ancient enemies. On this hypothesis one can understand the high mortality of measles when it is introduced into a new country. By natural selection, measles has become a powerful enemy of the human race, and a race to which this infection is newly introduced has not had the advantage of building up a defense against it by the law of natural selection. May not the phenomena of anaphylaxis be studied on associational lines?