In the protozoa and some other lowly forms of animal life multiplication by binary fission occurs indefinitely, so that one organism is succeeded by two, and as there is no vestige of a corpse, as Weissmann expressed it, the organism is immortal. Maupas’s9 investigations indicated that without occasional conjugation of two individual protozoa and the resulting rejuvenation the organisms undergo senile degeneration and die; but Woodruff10 has shown by observations extending over 13½ years that Paramoecium, and so presumably infusoria in general, can multiply indefinitely in favourable conditions without conjugation, though periodically an internal reorganization (endomixis11) of the individual takes place. It seems not improbable that endomixis provides the rejuvenation otherwise resulting from conjugation. C.M. Child12 brings forward reasons for modification of the explanation that death is escaped simply by division; he shows that in certain circumstances senescence occurs in the protozoa as it does in the higher animals, but that death is avoided by the rejuvenescence in each process of reproduction, reconstitution occurring and new organs being formed in the place of the old. In other words, all organisms from the protozoa to man undergo senescence and die unless rejuvenescence intervenes; this compensation holds good in the lower forms and death may never occur; but in the higher forms rejuvenescence is so much restricted by the evolutionary increase in the physiological stability of the protoplasmic substratum and the resulting higher degree of individuation that senescence is much more continuous and death is inevitable. It is the penalty for high individuation, and is the result of conditions and processes that have produced the complicated mechanism of the higher animals and man. While holding out the faint hope that the advance of knowledge and of experimental technique may make it possible at some future time to bring about a greater rejuvenescence and retardation of senescence in man and the higher animals, Child points out that the present condition of their protoplasmic substratum is the result of millions of years of evolutionary equilibration, and that this task must therefore be one of extreme difficulty.

Modern biologists such as Driesch, E. Schultz, Child, Steinach, and Julian S. Huxley, to the last of whom I am personally much in debt, have indeed shown that experimentally the life cycle and growth, which are so closely correlated, may be modified dramatically by various methods. By starvation planarian flat worms become smaller and their life cycle is reversed by the process variously called dedifferentiation, reduction, or involution, whereby their structure becomes simpler; similar retrogressive changes have been produced in the social ascidians, Clavellina and Perophora (J.S. Huxley13), this process of dedifferentiation or reversible differentiation being the primitive reaction of organisms to unfavourable circumstances. By alternate feeding and starvation Child has kept planarian flat worms at the same size while controls passed through nineteen generations, thus showing that the life of cells is not a matter of time but of metabolism. In the higher animals the conditions are of course more complex on account of the self-regulating mechanisms, especially the ductless glands. But by disturbing the endocrine balance experimental biology has produced some remarkable results: by feeding white mice with tethelin (a phospholipin obtained from the anterior lobe of the pituitary) Robertson and Ray14 found that epithelial proliferation was accelerated, though increase in weight of the animals was retarded, and that the duration of life was prolonged. Drummond and Cannan,15 however, failed to confirm these results, and feeding with the anterior lobe of the pituitary has given rise to very contradictory reports; acceleration (Clark, Robertson, Goetsch), retardation (Pearl, Wulzen), and no effect (Gudernatsch, Lewis and Miller, Hoskins, Sisson and Broyles16) have been described. Steinach’s rejuvenation of senile rats by ligature of the vas deferens, which is followed by increase in the interstitial cells of the testis and definite prolongation of life, will be referred to later (vide p. 76). The retardation of growth due to a diet deficient in vitamins, first proved by Gowland Hopkins17 in 1912, may be maintained for long periods, but when the diet is appropriately altered growth is resumed and proceeds to the full standard (Osborne and Mendel18); it remains to be proved that the total duration of life can thus be prolonged. Though every living species of animal may in normal circumstances have its allotted span of life, this is so subject to accidents and modifications by external influences that the average duration of existence, especially in the long-lived species, does not represent the natural term of years. Thus the expectation of life at birth in England and Wales is 51·5 years for males and 55·35 years for females, as contrasted with the ideal physiological duration of life for 100 years. Violent death in one form or another—traumatic or infective—is so much the rule that the occurrence of natural death in the animal kingdom has been questioned, just as de Candolle denied the natural death of trees and ascribed it to injury or disease. Metchnikoff,19 who paid special attention to this question, described natural death in the Ephemerids, day-flies, the imagines of which are incapable of feeding and after an hour or two of “aerial life devoted to love” die, their ova falling into the water; Sir Ray Lankester accepts the view that they are “wound up” for a few hours only of life, but according to Child death is probably due to starvation. In man natural death from old age—a physiological ending—is very rare, much less frequent indeed than death certificates and ordinary parlance would suggest; for in such cases there is very commonly some latent disease, such as pneumonia, which just turns the scale in the trembling balance. Montaigne warns us “what an idle conceit it is to expect to die of a decay of strength, which is the last of the effects of the extremest age, and to propose to ourselves no shorter lease of life than that, considering that it is a kind of death of all others the most rare and hardly ever seen.” While it must be admitted that this still holds good, it is a stimulus to our efforts to render it no longer true.

The length of days ascribed to Methuselah (969), Adam (930 or, if the “conceit urged by learned men,” as Sir Thomas Browne20 says, that he was 50 or 60 years old when called into being be accepted, 980 years), Seth (912), has aroused attempts at explanation on the basis of a difference in chronology. The suggestion that the “years” were lunar and not solar, in other words containing 30 instead of 365 days, was controverted at length by Eugenius Philalethes21 (not Thomas, twin of Henry Vaughan “the Silurist”) as too radical, for on this interpretation some of the patriarchs would have become fathers of children before they were ten years old. Long ago, Hensler, following Justin, put forward the view that the year consisted of but three months up to the time of Abraham, when it was extended to eight months until the era of Joseph, after which it contained our full complement of twelve months. This explanation was supported by Hufeland22 on the ground that some eastern nations still reckoned three months to the year. Adopting this explanation Methuselah’s age would be cut down to 243 years.

The Psalmist’s threescore years and ten and its sorrowful extension to fourscore are so often exceeded by human beings who appear normal that the possibility of a far longer existence has often been raised. John Sterne or Stearne (1624–69) who, though founder of the Irish College of Physicians, was, according to Mahaffy,23 more of a theologian than a physician, believed that there was no reason why men should not live as long as the patriarchs before the Flood; but he also argued that these patriarchal ages must have been exceptional, for had all the human beings between the Creation and the Flood, a period of more than 1400 years, lived to an average breeding age of 400 years and begun, as did the patriarchs, to beget children at the rate of one male every three years, the population of the earth could not have found standing room on its surface; his colleague in Trinity College, Dublin, Miles Symner, Professor of Mathematics, provided him with a calculation showing that there would have been less than one cubic foot available for each individual. This seems to be echoed in a slightly different connexion by Weissmann’s conception of natural death as due to natural selection and as an adaptation for the good of the species, the degenerated organisms being no longer fertile or useful, an hypothesis hardly necessary, as Metchnikoff points out, when degeneration which is the prelude to dissolution has appeared. Roger Bacon24 stating that man, originally immortal and after the Fall able to live for a thousand years, had his life gradually abbreviated by corruption of his own making, believed that it should be possible by taking care as to regimen to prolong life for a hundred or more years than the common duration. Since the time of Aristotle the vital cycle has been thought to be a multiple of the period of growth. Francis Lord Bacon considered that animals in general should live eight times as long as they take to come to maturity, and Hufeland,25 adopting this principle, and regarding 25 years as the termination of adolescence, concluded that man’s natural span should be 200 years. Buffon took 14 years as the age of puberty, and multiplying it by 7 decided that human life should naturally be 100 years. Assuming that union of the epiphyses of the long bones marked the time when growth was finally completed, Flourens26 calculated that throughout the animal kingdom life was five times as long, and that as in man the epiphyses united at 20 or 21 years of age the span of human life should be 100 years.

More modern estimates of the duration of human life under ideal conditions of protection against infection have been to the same effect, thus conforming to the estimate of Cornaro,27 who died at that age, that a hundred years was “the time allowed to man by God and nature.” Metchnikoff,28 while admitting that there must be variations, considered that man should live more than 100 years, and Luciani fixed the physiological duration of life at 100 years. Ebstein’s (1891) estimate of 70 years as the usual duration of life, which is the age at which the maximum number of deaths occur when the first few years of life are excluded, is different from the ideal duration under the best physiological conditions and corresponds more or less with B.W. Richardson’s rough rule that the individual’s probable age at death can be arrived at by taking the average of the ages of his parents and grandparents; in this procedure the important factor of heredity is taken into account, and the disturbing part played by accidents and infections in modifying the ideal expectation of life is to some extent neutralized by taking the average of the six lives.

Although 100 years may be regarded as the physiological life of man it is rarely reached, and one of several reasons is that the physiological requirements for this apparently exaggerated term of years are very seldom provided. Sir Ray Lankester29 suggested that just as giants in stature are exceptional variations so centenarians should be regarded in the same light, as giants in years instead of in inches. He meets the obvious objection that whereas the parents of giants are of normal height, longevity and centenarianism appear to run in families, by doubting if the actual quality of potential long life is transmitted and by ascribing family longevity to the inheritance of traditions and habits favourable to long life. In this connexion it should be remembered that many giants are pathological and due to acromegaly, whereas unusual height, over six feet, certainly runs in families; analogy therefore makes it reasonable to believe that length of days, like length in inches, may run in families, although these two approaches to giantism very seldom occur in the same individual.

The authenticity of persons alleged to be of great age rightly excites critical examination, and some of the most famous examples, such as Henry Jenkins (169), Thomas Parr (152¾), Katherine, Countess of Desmond (145), as Sir George Cornewall Lewis and Mr. W. Thom showed, as well as many of the 1712 centenarians in James Easton’s30 list covering the years A.D. 66 to 1799, cannot be regarded as established.31 In the same category we must place Petrasch Zorten (1537–1724), whose portrait at the reputed age of 185 was reproduced by Sir John Sinclair.32 Old people take a natural pride in their age and tend to exaggerate it; according to Mr. G. King33 the excess in the census returns of persons over 91 and the deficit in those between 85 and 90 years of age could only be explained by the conclusion that the temptation to overstate the age so as to appear among the nonagenarians had not been resisted. The vast majority of centenarians are naturally among the bulk of the population, namely, the poor, among whom natural selection is provided with a correspondingly greater opportunity of finding those “vigorous frames which promise a long life,” and in these circumstances the means of checking the exact age may be less easy than in the case of the well-to-do.

Sex.—Females are more long-lived than males; the 1911 census for England and Wales shows that after 10 years of age there are more females than males living in all the quinquennial periods, the proportion of females progressively increasing until at the age of 85 and upwards there are 645 females to 355 males in 1000 living at that age group. Among centenarians also the ratio of the sexes is much in favour of that popularly said to be the weaker. Out of the 691 reputed centenarian deaths registered in England and Wales during the 10 years 1910–19 inclusive, 504, or 73 per cent, were females, and 187, or 27 per cent, males. But during the same period in Ireland the disproportion was much less: among the 945 reputed centenarian deaths 545, or 58 per cent, were females, and 400, or 42 per cent, males. It has been said, though without sufficient statistical evidence, that although the female sex is favourable to longevity extreme length of days is attained only by males. The superior longevity of the female sex, in spite of the risks attaching to child-birth, depends on several factors, such as an existence less exposed to accident and infection, a more temperate life, and in addition, so Sir George Humphry argued, on a stronger inherent vitality, for during the first year of life when there is no difference as regards the first two circumstances more boys than girls succumb. The higher male death rate in the first year of life, however, may be partly due to a mechanical cause, namely, the large size of the male infant’s head, as a result of which the numerical superiority of male infants born is reduced by the effects of trauma. Another factor is, perhaps, that old women do not feel so old as old men do, and that therefore auto-suggestion plays a less powerful part with them.

The question must arise as to what is the object of the prolongation of life after the reproductive function has waned as it does in women—the more long-lived of the sexes—about half-way through the physiological term of years. In some lower animals life terminates very shortly after propagation of the species has been effected; but this is not the rule, and the biological meaning of the continuation of the life of individuals that but cumber the ground and may lead a parasitic existence requires explanation, though educational advantages and moral lessons may undeniably be provided by healthy old age with its store of what a Cambridge man, now justly famous, described in his rather rebellious youth as “that greatly over-rated property experience.” It has been suggested by Sir Ray Lankester that the inherent property resulting in longevity may be bound up as a “correlated variation” with some other characteristic useful in the struggle for existence and the perpetuation of the species. An obvious possibility is that the long-lived have a higher fertility rate than average mortals. Hufeland34 argued that as very old people are nearly always married more than once and generally at a very late period of life “a certain abundance in the power of generation is favourable to longevity.” There are few satisfactory data to determine the converse proposition with which we are now concerned, namely, is longevity accompanied by unusual fertility? As already mentioned, the great majority of centenarians are found among the bulk of the population, namely the poor, who are far more prolific than the well-to-do; but this does not prove much, for the inverse correlation of fertility and good social status is largely artificial and due to voluntary birth control. Among the 824 persons between 80 and 100 years of age analysed by Sir George Humphry35 in 1889 there were 335 married men and 292 married women, and in each group the average number of children was six, which is somewhat in excess of what is popularly supposed to be the average size of a family; but the question of average fertility is one on which Dr. Major Greenwood warns me that it is difficult to give an unambiguous answer, for it all depends on whether completed or incompleted fertilities are meant; thus in the 1911 census the completed fertilities average 5·8 and the uncompleted 4·2. No value could, he considers, be attached to the general figure obtained by dividing the total number of children by the total number of couples enumerated. Hence Sir George Humphry’s and other statistics are not really comparable. While finding that longevity was equally shared by the single and married women and by those who had children and by those who were barren, Sir George Humphry stated that among those who had children fertility was associated with longevity.