It is no mere accident that the most striking examples of natural death occur among insects, for these creatures display an unusual stability in their cellular structure with a corresponding lack of the power of regeneration, in these particulars resembling man and the higher animals. The cells of the nervous system are very complex, and are well adapted for the highest function, that is to say, the psychical function. These highly endowed cells, however, are devoid of the power of reproduction. Many experiments have been made in relation to this, and it has been proved clearly that in cold-blooded vertebrates the brain and spinal cord with the nerve cells contained in them are capable of regeneration, whilst among mammals only extremely rare cases are known in which there has been any regeneration of the nervous elements. It is to be expected, then, that cases of natural death occur in the higher animals and especially in man. However, no case is known so plain as that presented by the ephemerids. I have already stated that of deaths apparently due to senile debility in man, a large proportion are certainly due to various infectious diseases that affect the old, such as pneumonia and nephritis. Close examination of the tissues confirms this conclusion, for the destruction of the higher elements by phagocytes produces what is really violent death and not a natural death like that of the ephemerids. Natural death in man is probably a possibility rather than an actual occurrence. Old age is not a true physiological process but exhibits many morbid characters. That being the case, it is not surprising that it seldom ends in natural death. It is probable, however, that natural death occasionally occurs in very old men. Attempts have been made to estimate the natural limits of human life. Flourens[327] based a calculation on the duration of the period of growth. If the latter be taken as one fifth the natural life, then human life ought to last a century. As centenarians are rare, the vast majority of deaths, which happen before that age has been reached, must be regarded as violent or accidental. The rule of Flourens, however, is arbitrary, and there is no evidence to show that it is exact. Probably in the human race, as in the case of ephemerids, the natural duration of life varies and cannot be expressed by a definite figure. In most cases it ought to be more than a hundred years, and only in rare cases ought it to be much less than that term. Probably there is a variation in the duration of life just as there is a variation of the date of sexual maturity for which rules may be laid down but not without anticipating numerous exceptions.