V HUMAN FECUNDITY

1. As a preliminary to the separate inquiries into the two aspects of the problem, it is necessary briefly to consider the question of human fecundity.

We may ask, in the first place, whether, taking a broad view of human history, there is any reason to believe that human fecundity has changed since the epoch of the pre-human ancestor, and, if so, whether it has increased or diminished. Secondly, there are a number of habits and customs which have been practised in different degrees at different times, such as prolonged lactation, polygamy, and so on, and we may also ask what bearing, if any, such habits and customs have upon fecundity.

There is much connected with the subject which is difficult and obscure. It is, in particular, often difficult to ascertain whether any differences in fecundity, as may be found to exist, are of the nature of racial or of environmental differences. Into such points, however, we need not go. We are concerned only with the question as to whether there are differences in fecundity, and we may confine our attention to a notice of any broad differences which do exist and to the question whether certain habits and customs widely practised at one time or another and supposed to bear upon fecundity do or do not have the influence attributed to them. We may begin with the first question set out above. In order to obtain an answer to this question we have to rely upon evidence derived from various sources, partly from primitive races, which we assume in general to exhibit conditions prevalent among prehistoric races, and partly from evidence derived from a study of the mammalian species which have been subject to changes comparable with those to which man has in the course of his history been subject.

2. From what has been said regarding the nature of reproduction in Chapter II, it follows that it is the female rather than the male that we have to consider. The male gametes vastly outnumber the female, and, generally speaking, the male is always ready for 89intercourse. When studying the possibility of increasing the fecundity of domestic species, attention can, as pointed out by Marshall in the case of sheep,^[104] be limited to the female. Taking a broad view of the matter, it is clear that there has been no change in fecundity due to any change in male reproductive power.

There are, nevertheless, two points concerning the male which we may notice before we pass to the consideration of the female. First, with regard to the problem as to changes in fecundity during the course of history, we shall bring forward evidence to show that good conditions favour an increase in the reproductive power of the mammalian female. There is reason to think that good conditions have a similar effect on males. Thus among some mammalian species in a state of nature there is a special ‘rutting season’ for the male, and it is not infrequently found that in captivity the males of such species will rut all the year round.^[105] We shall also bring forward evidence to show that with improvements in conditions, sexual excitement is more easily aroused and that there has been an increase in the development of the generative organs, and we may note here that these observations apply to the male as well as to the female. As, however, any increase in male reproductive capacity can have had no general influence upon fecundity, we may, so far as the question of any general increase or decrease in fecundity is concerned, neglect the male and confine our attention to the female.

Secondly, with respect to the influence of certain factors at particular times, it is in nearly all cases the female and not the male which is affected. But it may be observed that sterile marriages are in part due to male sterility. Mayer estimates that this is so in a third of such marriages.^[106] Kelly gives the result of an investigation into 110 cases, of which 59 per cent. are attributed to the male.^[107] Male sterility is due either to inborn or acquired conditions. So far as sterility due to the former is concerned, there do not appear to be any grounds for thinking that it has to any important extent varied from one epoch to another. So far, however, as sterility due to disease is concerned, there have been in all probability certain changes. Gonorrhoea is perhaps the most important of those diseases which bring about sterility. Of 90the 59 cases mentioned above 12 per cent. were due to gonorrhoea. Other diseases may cause sterility, such as tubercle and epididymitis. Relatively to the historical period as usually defined, gonorrhoea is an old disease; but we shall find reason for thinking that the great majority of diseases have evolved since the time of the pre-human ancestor, and to the extent to which this is true it must be allowed that male sterility due to disease is more or less recent. With these remarks we may leave the question of male sterility, as it is relatively of little importance.

3. It is therefore to factors which influence the female generative process that we must look for the causes of such increase or decrease of fecundity as are of importance. Certain factors may be put aside on the grounds that there is no reason to think that they have been of more importance at one time than at another. Such are malformations of any kind that prevent the meeting of the male and female gametes.^[108] With regard to the factors of importance it is by their influence upon different aspects of the female generative process that they are best classified. There are three main aspects of this process, variations in any of which will influence fecundity. These are the length of the mature period, the interval between births, and the number at a birth.

Some reference to the nature of the female sexual cycle in mammals has been made. The beginning of maturity is usually measured by observing the beginning of menstruation. It is, perhaps, worthy of mention that the beginning of menstruation does not always coincide with the beginning of ovulation. Thus the estimation of the duration of the mature period made by watching for the beginning and end of menstruation is not always exact; the difference is not of consequence here, but the fact that a difference is possible emphasizes that what is essential in the whole process is ovulation. The true mature period is the period during which ovulation occurs. The interval between births is also dependent upon ovulation and so is number at a birth.^[109] In fact what we are asking is in the main what factors influence ovulation.

4. The information regarding the duration of the mature period is not satisfactory. Though there is a large amount of information about the beginning of menstruation, there is little information 91about the age at which it ends. There are at least three factors which influence the age at which menstruation begins—climate, race, and general nature of the surrounding conditions.

Generally speaking, the hotter the climate the earlier menstruation begins. Englemann gives the following figures:^[110]

Zone. Average Age for Menstruation.

Tropics 12·9 years

Temperate 15·5 „

Cold 16·5 „

and Krieger gives the following figures:^[111]

Place. Average Age for Menstruation.

Christiania 16 years 9 months 25 days

Berlin 15 „ 7 „ 25 „

London 15 „ 1 „ 14 „

Lyons 14 „ 5 „ 29 „

Marseilles 13 „ 11 „ 11 „

Calcutta 12 „ 6 „ 0 „

Sierra Leone 10 „ 0 „ 0 „

It further appears that there is a tendency for menstruation to end earlier in hot climates; so much so that upon the whole the mature period is shorter in hot than in cold climates.

There is, however, no very close or definite connexion between climate and menstruation. In the United States it has not been found possible to detect any influence of climate upon menstruation, though this country stretches from 29° to 45° of latitude, and has a temperature the annual average of which varies from 40° F. to 70° F. But if the conditions concerning the different racial elements in the population of the country are examined, it is found that in whatever part of the country they may be the average age for each racial element remains constant, and usually varies slightly from the average age for the whole population.^[112] There are other indications that the age at which menstruation begins are connected with differences in race. Krieger quotes the following from Joachim for girls in Hungary:^[113]

Race. Average Age of Menstruation.

Slavonic 16 to 17 years

Magyar 15 „ 16 „

Jewish 14 „ 15 „

Styrian 13 „ 14 „

92Other figures show that Jewesses menstruate earlier than the average age for menstruation in the country in which they live.^[114]

What is of chief interest to us is that good conditions also influence the age at which menstruation begins. The better the conditions, the earlier does it begin. Further, it is also known that the mature period tends to be prolonged where conditions are good. It is known, for example, that the mature period comes to an end earlier among the labouring than among the richer classes.^[115] Therefore good conditions tend to be connected not only with an earlier beginning but also with a longer duration of the mature period. Thus Mayer found the average age at which menstruation began among 3,000 women of the upper classes to be 14·69 years, and among 3,000 poor women to be 16·0 years.^[116] Several observers have recorded the average age for the beginning of menstruation among the different classes in the population. The following table is given by Krieger:^[117]

Brienne de Boisment. Tilt. Krieger. Ravn.

Yrs. Mths. Yrs. Mths. Yrs. Mths. Yrs. Mths.

Upper Class 13 8 13 5½ 14 1⅙ 14 3

Middle Class 14 5 14 3½ 15 4–2/6 15 5½

Lower Class 14 1 16 8⅙ 16 5¼

5. With regard to the question as to the interval between births some interesting indirect evidence is obtained from what we can learn regarding the former condition of the sexual cycle in man. A sketch was given in the second chapter of the main features of the mammalian sexual cycle. These cycles may follow one another without any prolonged interval, or there may be a lengthy period of rest. In monkeys and also in man the interval is short; but the former cannot breed at every heat period; special sexual seasons are experienced at definite times of the year when breeding takes place. It is probable that ovulation is limited to this special season, and that, although copulation may take place at other seasons of the year, conception can only follow copulation at this special season.^[118] In civilized man, so far as is known, ovulation is 93never confined to any particular season of the year; conception can, therefore, follow copulation at any period. But we do find evidence of the former existence in man of a special sexual season, and this is of particular interest because it suggests that at one time there was in man, as there usually is in monkeys, one season only during which conception could follow copulation.

This evidence has been reviewed by Westermarck, by Ploss and Bartels, and by Havelock Ellis.^[119] Some examples of this evidence may be cited in order to show upon what facts this suggestion is founded. According to Mr. Johnston, the Indians of California “have their rutting season as regularly as have the deer, the elk, the antelope, or any other animals”. And Mr. Powers confirms the correctness of this statement, at least with regard to some of these Indians, saying that spring ‘is a literal St. Valentine’s day with them, as with the natural birds and beasts of the forest’. As regards the Goddanes in Luzon, Mr. Foreman tells us that ‘it is the custom of young men about to marry to vie with each other in presenting to the sires of their future brides all the scalps that they are able to take from their enemies, as proof of their manliness and courage. This practice prevails at a season of the year when the tree—popularly called by the Spaniards “the fire tree”—is in bloom.’ Speaking of the Watch-an-dies in the western part of Australia, Mr. Oldfield remarks, ‘Like the beasts of the field, the savage has but one time for copulation in the year. About the middle of the spring the Watch-an-dies begin to think of holding their grand semi-religious festival of Caa-ro, preparatory to the performance of the important duty of procreation. A similar feast was, according to Mr. Bonwick, celebrated by the Tasmanians at the same time of the year.’^[120] Similar evidence is forthcoming as regards many other primitive races. It is interesting to note that among civilized races there are traces of customs which point to the former existence of a season of sexual licence. Such were the mediaeval Feast of Fools, and the classical festivals of Brumalia and Rosalia.^[121] ‘Feasts, similar to the erotic feasts which were indulged in by the ancients ... were still practised to some extent in Russia and in some parts of India at a much later date, while such customs as “gwneyd Bragod” and possibly our own “bean 94feasts” may not improbably be the modern representatives of these ancient customs in our own country.’^[122]

This evidence all points to a former condition when conceptions could only take place at one season of the year. We must suppose that it has gradually become possible for copulation to be followed by conception at any period, and that therefore fecundity has increased.^[123] Some light is thrown upon the causes that underlie these changes by a reference to certain facts regarding other species. Before we turn to this question reference may be made to the phenomenon of ‘mittelschmerz’—or intermenstrual pain. It is not uncommon and is sometimes accompanied by a discharge. The interpretation of this phenomenon is doubtful. One explanation, however, has some interest for us. It is suggested that it may represent the first steps in the establishment of a different variety of sexual cycle in which the dioestrous periods occur fortnightly instead of monthly. If this interpretation is correct, it would mean that the sexual cycle is evolving in the direction of still greater fecundity.^[124]

6. It has frequently been observed that when members of a wild species are brought into captivity there is a change in the sexual season. This change may be in the direction of an increase or of a decrease in the generative power. If the bringing into captivity involves such an alteration of conditions as to disorganize the generative process, then a decrease in fecundity may result. It frequently happens, for instance, that carnivores are infertile in captivity, and this may be due among other things to the absence of fresh meat.^[125] Generally speaking, however, if the alterations do not produce disorganization, an increase in fecundity results, and as there is no reason to suppose that the change in human circumstances was ever of such a nature as to produce disorganization, we must suppose, on the analogy of what happens among other mammalian species, that, if the betterment of the conditions has had any effect, it has been in the direction of an increase in fecundity.

A few examples of what is meant may be given. The wild dog of South America, the wolf, and the fox breed only once a year under 95natural conditions; in captivity they experience two annual heats like the dog.^[126] The otter in the wild state breeds but once a year; in captivity oestrous may occur at regular monthly intervals all the year round.^[127] Bears are monoestrous in the wild state; in the Zoological Gardens oestrous may be experienced for two or three months.^[128] With regard to the Ungulata we have numerous facts of the same nature. The red deer offers the best example; under natural conditions there are two dioestrous cycles lasting three weeks; in captivity the sexual season extends over the whole year.^[129] The facts are somewhat similar in respect to cattle and horses. Examples might also be taken from other mammalian classes. The wild rabbit breeds from February to May; in captivity the sexual season lasts nearly the whole year. Further it may be noticed that among lower groups good conditions are, as is well known, markedly favourable to increased fecundity.

Before we go on to look into the causes of this increase in fecundity, it may be noticed that the number of young produced at a birth among domestic species is on an average greater than that produced by the wild ancestor. This evidence, it will be observed, refers to the third factor which bears upon the increase in fecundity. The wild rabbit is said to produce at the most six young at a time; the tame rabbit has four to eleven, and a case is known in which eighteen were born, all of which survived.^[130] The wild sow bears from four to eight, and sometimes twelve, young. The domestic sow is considered to be of no value unless she produces at least eight at a birth. Equally remarkable examples are to be found outside the Mammals. ‘The hen of Gallus bankiva lays from six to ten eggs; the tame one in the course of a year from eighty to one hundred.’

7. It is clear that one common feature in the conditions surrounding domesticated species and civilized man, as compared with species in a state of nature, is the increase in the richness and regularity of the food-supply, and the general betterment of the environment. In fact, just as this increase in the richness of the surroundings may increase the length of the mature period, so it may increase the number at a birth and also decrease the 96interval between births. That this is the result of an improved environment in the case of the sheep has been shown by Marshall. He found, for instance, that in ‘Scotch Blackfoot, Cheviot, and other Scottish sheep the normal percentage of ova discharged at any single oestrous is not appreciably in excess of the usual percentage of births at the lambing season’.^[131] This, it may be noted, is an example of what was said in the second chapter regarding, in the first place, the measurement of fecundity by the number of ripe ova produced, and, in the second place, regarding the fertilization on the average of all, or nearly all, ripe ova. Marshall also found that ‘there was every reason for supposing that the processes of growth and maturation can be very largely influenced both by insufficiency of food-supply on the one hand and by artificial stimulation on the other’.^[132] What apparently happens is that insufficiency of food retards the development of the ova and may cause the degeneration of some of them. Therefore if the food-supply is good, there will be more ripe ova at the sexual season, fewer sterile females, and a greater number of twin births. ‘There is overwhelming evidence,’ says Heape, ‘that flocks in good condition at tupping time have a higher subsequent development of fertility than flocks in poor condition at tupping time.’^[133] By ‘good’ condition is meant not a ‘fat’ but a strong, healthy, and vigorous animal.

It may be asked whether there is any evidence as to the increase in numbers at a birth in the human female. Among civilized races about one birth in eighty to ninety is on the average a twin birth. Our knowledge of primitive races is not sufficiently precise to enable any estimate to be made regarding the frequency of twin births among them. There are, however, very numerous references in the accounts of these races to the superstitions attaching to twin births. The nature of these superstitions, and in general the mystery felt to surround twin births, very strongly suggests that the phenomenon is rare. It can only be said that it is probable that twin births are rarer among primitive races than among civilized races.

So far, therefore, as we have gone, the evidence points to the conclusion that two of the three factors which determine fecundity have varied in the direction of increasing fecundity, and that 97possibly the third factor may have varied in the same direction. It has also been noted that similar changes have been observed to take place in animals which have been subject to better conditions. The fact that human conditions have certainly been bettered suggests that the same cause may have been at work in the case of man. There is, however, no doubt that changes in the case of man are not wholly explicable in this way. In part the differences are racial differences, and to the degree in which the differences are racial they cannot be attributed to the direct effect of the surroundings.

8. Additional support to the view that fecundity has increased with civilization is given by certain other types of evidence. In the first place it appears to be a fact that the reproductive organs of the more primitive races of mankind are smaller and in all respects less well developed than those of civilized races.^[134] It is not meant that the organs differ qualitatively in any way; they are merely smaller relatively to the other organs than among civilized races. It is doubtful how we should interpret these facts; but it does not seem unreasonable to assume a connexion between a lesser development of the reproductive organs and a lower degree of fecundity.

Secondly, we have a considerable body of evidence with respect to the strength of sexual desire among these races.^[135] Many observers have recorded their opinion that the members of these more primitive races do not experience sexual excitement to the same degree as do the members of the more civilized races. There are other observations which record the difficulty experienced in obtaining sexual erethism. It seems reasonable to associate these observations with a lower degree of fecundity than that found among civilized races.

In this connexion the fact that lack of sexual excitement renders fertilization less likely is relevant. As fertilization has been achieved with the female in a condition of narcosis,^[136] it is clear that sexual excitement is not necessary. Nevertheless, if the 98question of sexual excitement has any bearing at all upon the degree of fecundity as between higher and lower races, it must be in the direction of rendering the lower races less fertile.

It was observed in the second chapter that the willingness of the female to receive the male at any time in the sexual cycle is a peculiarity of man. It may be observed in passing that evidence has lately been brought forward suggesting that the indifference of the female to the point in the sexual cycle at which intercourse takes place has been exaggerated, or rather that the existence of periods of desire has been somewhat overlooked.^[137] However this may be, what concerns us here is the peculiar fact that the male is received by the female at any time in the sexual cycle. It is possible that copulation at one time in the sexual cycle is more likely to result in conception than at other times, though the facts are obscure. But whether this is so or not is scarcely relevant to our purpose, because there is no reason to think that there have been any variations in the practice of copulation between one time and another or between one race and another, as could have sensibly affected fecundity even if the limitation of copulation to one period of the sexual cycle is of marked importance.

9. Lastly, we have to refer to what we may call the statistical evidence. The interpretation of the evidence as to number of children among primitive races is so difficult a matter that it cannot in itself be held to throw much light upon the question of fecundity. The trend of the evidence, however, certainly supports the conclusion that fecundity is less among primitive races than among civilized races. There are a vast number of observations recorded by travellers regarding the number of children among these races. For the most part these observations have reference only to the number of children seen alive, and therefore are not even a measure of the fertility—far less a measure of the fecundity.

Practically all such observations emphasize the small size of the families. When dealing with hunting and agricultural races in another connexion we shall have reason to refer in some detail to this evidence. Here we may give a few typical examples of this evidence, remembering that it is evidence for the most part of the number of children seen and not of the number of children born, though, in selecting the following examples for quotation, an effort has been made to pick out those in which the authors have 99attempted to discount the effects of infanticide and other factors of elimination.

Of the Australians Curr remarks: ‘I am of opinion that the Australian females bear on an average six children, or did before the advent of the whites and whilst living in their natural state.’^[138] Spencer and Gillen state that sterility is common among the Australians^[139] and that the number of children rarely exceeds four or five and is generally two or three.^[140] Of the Eskimos we read: ‘the women are not prolific. Although all the adults are or have been married, many of them are childless, and few have more than two children. One woman was known to have had at least four. Dr. Simpson heard of a “rare case” where one woman had borne seven children.’^[141] ‘On the average the pure breed Greenlanders are not prolific. Two, three, or four children to each marriage is the general rule, though there are instances of families of six or eight or even more.’^[142] Of the American Indians there is a large amount of evidence. Dr. Holder, who combined medical knowledge with exceptional opportunities for observation, says: ‘With Indians large families are the exception. The Crow tribe, of less than 2,500 people, is divided into 630 families, which gives less than four to each family, and this includes parents and often grandparents and relatives by affinity or adoption, leaving the average offspring, to each child-bearing woman decidedly lower than in white communities.’^[143] Speaking of the Indians of Vancouver Island, Sproat says: ‘As a rule they have few children.’^[144] Bancroft reports of the Nootka tribe: ‘Women rarely have more than two or three children.’^[145] Of the Chinooks: ‘Barrenness is common, the birth of twins rare, and families do not usually exceed two children.’^[146] Catlin says: ‘It is a very rare occurrence for an Indian woman to be “blessed” with more than four or five children during her life; generally they are contented with two or three.’^[147]

It is merely suggested here that evidence of this kind, which will be much amplified in the following chapters, may have to be interpreted as pointing to a lower degree of fecundity among these 100lower races than among the civilized races. The trend of the evidence is at least suggestive when it is remembered that according to the estimate made by Matthews Duncan a normal woman among civilized races living in wedlock throughout the mature period under favourable circumstances should bear from ten to twelve children.^[148]

If the view suggested is correct, we should expect to find in such countries as India and China a higher fecundity than among primitive races, but a lower fecundity than among European races. There is some indication that this is so. At first sight it might seem that the well-known fertility of these races indicated a higher fecundity than in Europe. When, however, in the case of India, not the crude fertility, but the fertility corrected for the number of married women of reproductive age is calculated, it is found to be lower than in Europe. The figures per 1,000 are 160 in India and 196 in England.^[149] It has, of course, to be remembered that there are in both countries certain factors bearing upon fertility—early marriage in India and restraint from intercourse and contraceptive methods in England. But it can hardly be supposed that the former is more effective than the latter in decreasing fecundity, and it is probable that we have here an indication of lower fecundity in India.

10. All the evidence, therefore, points to the same conclusion. If there has been any general change in the strength of human fecundity in the course of human history, using that phrase in the widest sense, it has been in the direction of an increase. It is not necessary for our present purpose to attempt to be more precise. This view is, as we shall show below, that of the best authorities. It may be noticed that different opinions are frequently expressed by authors who are not professional biologists.^[150] Such opinions are in most cases so vaguely worded that it is seldom clear what 101precisely is meant. In many cases it is fairly evident that it is intended to imply that there is some connexion between growth of civilization and decrease in fecundity. So far as this is what is meant, there is no evidence for this view. (There is no indication whatever that increasing intellectual activity is accompanied by decreasing fecundity. On the other hand, inasmuch as intellectual activity is connected with an amelioration in the conditions, to that degree there is a connexion between it and an increase in fecundity.) Such views clearly owe their origin to the attention paid to the decline in the birth-rate. It is not always realized that a declining birth-rate may be due to a decline in fertility alone, wholly unconnected with a decline in fecundity.

That fecundity has increased was the opinion of Darwin. ‘There is reason to suspect’, he says, ‘that the reproductive power is actually less in barbarous than in civilized races.... It is highly probable that savages, who often suffer much hardship and do not obtain so much nutritious food as civilized men, would be actually less prolific.’^[151] According to Heape, ‘it would seem highly probable that the reproductive power of man has increased with civilization, precisely as it may be increased in the lower animals by domestication; that the effect of a regular supply of good food, together with all the other stimulating factors available and exercised in modern civilized communities, has resulted in such great activity of the generative organs, and so great an increase in the supply of the reproductive elements, that conception in the healthy human female may be said to be possible almost at any time during the reproductive period.’^[152]

It is interesting to note that no differences in fecundity can be observed as between modern civilized races.^[153] The conclusion we have reached is merely that there has been, broadly speaking, an increase in fecundity, and that in large measure this is explicable as a result of the betterment of conditions, and there is nothing to lead us to expect that there would on this account be any difference between modern European races.

11. We may now pass to a brief consideration of certain habits, customs, and other factors which, operative from time to time, have, or are widely supposed to have, an influence upon fecundity, 102and in some cases upon fertility; for though fertility is not, properly speaking, considered in this chapter, it is convenient to deal with certain matters here.

To polygamy has often been attributed a decrease in the number of children born to a woman. Whether it is supposed that the practice of polygamy in some way decreases fecundity or has an influence upon fertility is not clear. In any case there is no evidence that it has any influence either upon fecundity or upon fertility. There is no reason to think that polygamy is attended by any structural or physiological changes such as would influence fecundity, and the statistical evidence does not show any difference in the average number of children born, whether monogamy or polygamy is practised. Theal investigated this question as far as marriages among Bantu women are concerned. He collected a number of returns and sums up the returns as follows. ‘Altogether these returns embraced 393, the wives or widows of monogamists, mostly professing Christians, and 591 women, the wives or widows of polygamists. In a few instances it was noted that the women might not have passed the age of child-bearing. The 393 women, wives of monogamists, had borne 2,223 children, that is on an average 5·65 children to a woman. The 591 women, wives of polygamists, had borne 3,298 children, that is on an average 5·58 children to each woman. Thus monogamy made hardly any appreciable difference in the birth-rate.’^[154]

The question of the influence of lactation on fecundity is of considerable importance, but unfortunately no very definite conclusions can be drawn from the facts known. There is a considerable amount of evidence to the effect that the continuance of lactation to some extent inhibits heat in animals and menstruation in women. Though the effect of continued lactation is doubtless in this direction, it cannot be said to be of any very definite strength. It is stated that mares giving suck are liable to miss a season.^[155] ‘There can be no doubt that in the case of sows early weaning is conducive to a more frequent recurrence of oestrous and an increased number of litters.’^[156] ‘The return of menstruation during lactation in women has been dealt with recently by Heil and Dingwall Fordyce. Heil, who has studied the conditions of 200 nursing mothers, expresses the belief that 103the recurrence of menstruation and not the condition of amenorrhoea is the normal state during lactation, but that menstruation is not so frequent in the later lactations as in the earlier ones. Fordyce has reached similar conclusions, finding that menstruation occurred during lactation in 40 per cent. of the cases in which suckling was performed, while in 92 per cent. of the cases its return was within nine months of parturition, and that menstruation during lactation was commoner with the earlier than with the later lactations, showing that age is an important factor.’^[157]

There is another question of importance. It may be asked what influence age at marriage has upon fecundity. As far as the age of the husband is concerned, there is scarcely any influence at all. For any given age of wife the fecundity remains nearly the same, whatever the age of the husband. The age of the wife, however, is of importance, apart from the fact that delay in marriage reduces the use made of the limited period of maturity. For a woman is not equally fecund throughout the mature period. The earlier years are the most fecund years, and therefore postponement of marriage reduces the fecundity, other things being equal, by more than the fraction of the mature period which is passed in celibacy. According to Dunlop’s observations on marriage in Scotland ‘the effect of one year’s delay of marriage is to reduce the average family by fully one-third of a child, or that three years’ delay may be expected to result in the family being one child less. This result may be fairly correct in general, but it cannot be strictly applied, for the crude observations show that the effect of one year’s delay is not constant through the fertile period of the woman’s life, but is greater for the younger and less for the later years. Thus a year’s delay when the woman is aged from 20 to 25 averages 0·45 of a child, 0·37 when she is aged from 25 to 30, 0·32 when she is aged from 30 to 35, 0·29 when she is aged from 35 to 40, and 0·19 when she is aged from 40 to 45.’^[158] It requires, it may be noticed, a delay of about forty years on the part of the husband to decrease the number of children by one child.^[159]

A further problem is the influence of sexual intercourse before puberty upon fecundity. Exactly what influence early intercourse has upon the generative organs and their functions is not clear. 104It is known, however, that early intercourse is injurious to the general health, and it is not difficult to understand in a general way how, if this is so, the reproductive functions would be adversely affected. In the Punjab Census Report ‘it has been shown that the states which practise early marriage on an extensive scale have generally a smaller proportion of females at the age period 12 to 15. Inquiries into a large number of cases show that, when the marriage of young people is consummated at an early age, a fairly large number of wives dies of phthisis or some other disease of the respiratory organs or from some ovarian complication within ten years of the consummation of marriage.’^[160] It is also known that when of two races both living a similar kind of life under similar conditions, one practises early marriage and the other does not, as for example the Hindus and Mohammedans in India, fertility is higher among the latter than among the former.^[161]

Finally, it may be noted that the development of fat may lead to sterility. It is a fact well known to breeders that excessive fatness is accompanied by sterility. Animals that have been fattened for agricultural shows are often barren. There is no doubt that the development of fat may have the same effect in women. It is not quite clear why the development of fat should be inimical to fecundity; formerly it was supposed that the sperm was prevented from reaching the ovum owing to the presence of a mechanical obstruction. Although this may at times be the cause, it seems that the presence of an excessive amount of fat has a deleterious influence upon the metabolism of the organism and that the maturation of the ova must be in some way affected. Marshall found signs of abnormal ovarian metabolism in interstitial tissue of the ovaries of fat cows and heifers.^[162] Sterility, it may be remarked, is not to be regarded as an ultimate consequence of ‘good’ conditions which have been made too ‘good’. ‘Good’ conditions do not merely consist in abundant food, but also in sufficient exercise and so on. The development of fat in such a degree as to cause sterility is due to an excessive amount of food—an abundant food-supply being one only of the factors going to make up ‘good’ conditions—to the exclusion of other factors.

10512. In conclusion, some calculations may be conveniently added to illustrate the strength of human fecundity, which, as we shall see, is constantly under-estimated. Let us consider a population of a million born in the same year, half of whom are males and half females. Let us suppose that they all marry, each couple before the age of twenty producing two children, half of whom are girls and half boys. For the sake of simplicity we may imagine that at the end of each twenty-year period the parents die simultaneously with the birth of their offspring. Then, if the children marry and produce offspring as did their parents, we shall have a standard population of 1,000,000, which will neither increase nor decrease so long as these conditions are fulfilled. If, however, the average number of children is 2½ per couple, then in 100 years the population will be 3,050,000; if three, 7,954,000; if four, 32,000,000; if five, 97,650,000.

106