Book - Contributions to Embryology Carnegie Institution No.56-6

From Embryology
Embryology - 21 Oct 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Mall FP. and Meyer AW. Studies on abortuses: a survey of pathologic ova in the Carnegie Embryological Collection. (1921) Contrib. Embryol., Carnegie Inst. Wash. Publ. 275, 12: 1-364.

In this historic 1921 pathology paper, figures and plates of abnormal embryos are not suitable for young students.

1921 Carnegie Collection - Abnormal: Preface | 1 Collection origin | 2 Care and utilization | 3 Classification | 4 Pathologic analysis | 5 Size | 6 Sex incidence | 7 Localized anomalies | 8 Hydatiform uterine | 9 Hydatiform tubal | Chapter 10 Alleged superfetation | 11 Ovarian Pregnancy | 12 Lysis and resorption | 13 Postmortem intrauterine | 14 Hofbauer cells | 15 Villi | 16 Villous nodules | 17 Syphilitic changes | 18 Aspects | Bibliography | Figures | Contribution No.56 | Contributions Series | Embryology History

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Chapter 6. Sex-Incidence in Abortions

By Adolph H. Schultz

The numerical proportion of the sexes has always been a subject of great interest. The question whether man produces males and females in equal or unequal numbers bears a close relation to the problem of sex determination and must be studied in connection with the death-rate in different periods of life and the comparative sex-mortality. While sex-ratio is conditioned originally upon sexdetermination, mortality may change it in the course of time. It will be shown herein that sex-ratio is not constant, but varies in the different periods of life and under manifold influences. Most of the information at hand concerning it is derived from statistics of the new-born and adult, but as regards intrauterine life, as far back as conception, our knowledge is limited to a few incomplete statistics and conflicting estimates. The main purpose of this study is to elucidate as far as possible this problem of the proportion of the sexes in utero, both living and aborted. The short review of the sex-ratio after birth, given herein, is intended to show the fluctuations in the numerical proportion of the sexes throughout life and may serve for a comparison of the relative mortality before and after birth. The last part of the study is a compilation from the literature of factors which may possibly have an influence on the original sex-ratio and of conditions which may change it during the course of prenatal life.


In view of the generally accepted supposition that sex is determined either before or at the time of fertilization, one may speak of a sex-ratio at conception or a sex-ratio of fertilized ova. This may also be called an original, physiological, or primary sex-ratio; the sex-ratio of the new-born may be termed secondary and, finally, that of adults is the tertiary sex-ratio. The latter term, in contrast to the preceding ones, does not apply to a definite time, such as conception or birth, but may be used collectively to designate the changing numerical proportion of the sexes after they reach maturity.

Tertiary Sex-Ratio

The consideration of sex-ratio in adults will be limited to generalities. According to BUcher (1892), for Europe the sex-ratio[1], including all ages, is 97.6; for Asia 104.5; for Africa 103.3; for America 102.8; for Australia 117.4. However imperfect these computations may be, they nevertheless show that Europe, with its excess of females, occupies a unique position. There are only a few European countries with a greater proportion of men than women. Among the American negroes there is an excess of females, and among the American Indians an excess of males (Cummings, 1918). According to Brooks (1887), the Australasian colonies had in 1881 a sex-ratio of 143.7 for the aborigines, and one of 118.6 for the population of foreign descent.


Table A. Number of females per 1,000 males. (From Rauber.)

Mall Meyer1921 tableA.jpg

Table B. Changes in sex-ratio with advancing age. (From Prinzing.)

Mall Meyer1921 tableB.jpg

Table A, taken from Rauber (1900), represents the number of females to every 1,000 males in some of the European countries. As will be noted, the excess of females increases with age. This is still more clearly shown in table B, from Prinzing (1905), taken from the census of 1900 in Germany. Inasmuch as in Europe sex-ratio at birth favors males, its subsequent decrease must be the result of a greater mortality among the latter, and also, in some countries, of a greater emigration of males. The unequal mortality is shown in table C, from Ploss (1887).

The sex-ratio of mortality may differ with age and under various conditions of life. For example, between the ages of 25 and 40 years (according to Knopfel, 1907) in rural districts it is below, while among urban population it is above 100. A difference in the comparative death-rate of the sexes in city and rural populations is shown by table D (Baker, 1910, compiled from the United States census of 1900) to exist throughout life, but to vary at different ages.


Table C. Sex-ratio of mortality in European countries. (From Ploss.)

Mall Meyer1921 tableC.jpg

Table D. Excess of male death-rate (per 1,000 population) over female death-rate. (From Baker.)

Mall Meyer1921 tableD.jpg

Bell (1918), in his studies on the Hyde genealogy, found that during the years covering the child-bearing period the death-rate was greater among females than among males. His conclusions were based, however, upon comparatively small statistics. In opposition to this we have the more reliable conclusion of Willcox (1904) that, according to the life tables of several entire countries, the death-rate of women between 20 and 30 years of age (a period during which probably fourfifths of the child-births occur) was less than that of males of the same age.

Secondary Sex-Ratio

The sex-ratio of the living-born for all European countries is, roughly averaged, 105 to 106, but this may vary markedly even in comprehensive statistics. Manifold explanations for these variations have been offered and these will be considered briefly farther on. In the individual countries small differences in the secondary sex-ratio may exist, as shown by table E, taken from a table by Ploss.


We have very little data concerning the sex-ratio of new-born among races other than white. Newcomb (1904) states that in Japan the excess of males in over 1,000,000 births was practically the same as in European countries. According to the same author, it seems probable that among the negroes of the United States there is a slight excess of female births. In a table by Nichols (1907) one finds the following sex-ratios among the living-born of primitive races: Japan (1877-1902), 104.7; India (1891-1900), 107.5; negroes of United States (1900), 99.8; and an average since 1880 of 100.9.

Table E. Secondary sex-ratio. (From Ploss.)

Mall Meyer1921 tableE.jpg

Table F. Sex ratio of mortality during first year of life. (Prinzing.)

Mall Meyer1921 tableF.jpg


As a result of the unequal mortality of the sexes, this secondary sex-ratio becomes reversed early in life. Kroon (1917) states that in Holland the sex-ratio of mortality during the first year of life is 119; that is, the mortality among male infants is greater by one-fifth than among females; also, that during the first two months of life this ratio reaches even 139. Analogous numbers for the first year of life are given in table F (Prinzing, 1906).


Kroon's statistics show that this high ratio of mortality of the first year of life soon undergoes a decrease, reaching its minimum (approximately 80) between the fourteenth and fifteenth year. According to Prinzing (1905), and also Willcox (1904), the mortality from the ages of 5 to 20 years is greater in females, and indeed in the latter part of this period it is the result chiefly of tuberculosis, for which disease the common occurrence of anaemic and chlorotic conditions at the time of puberty furnishes an excellent soil. After this age the sex-ratio of mortality increases rapidly and results in the reversion from an excess of males to an excess of females, and the higher the death-rate the earlier will this reversion take place, the death-rate being influenced by various factors. It can be said, for instance, that as a general rule the death-rate of infants is especially high in countries with a high birth-rate, such as Russia and China. This, in connection with the high sex-ratio of infantile mortality, will tend to decrease very rapidly the sex-ratio of the surviving infants.

Primary Sex-Ratio

The sex-ratio of fertilized ova can not be determined directly; however, there is an indirect method of solving the problem of the original sex-ratio by means of computing the mortality of cyemata.[2] In case the sex-ratio of fetuses dying in uie.ro is equal to the secondary sex-ratio, then only will the latter and the primary sex-ratio be equal. If the number of males and females that were aborted or still-born were absolutely the same, then the primary would be lower than the secondary sex-ratio. If, however, the sex-ratio of mortality during pregnancy exceeds the sex-ratio of the living-born, then the primary sex-ratio must of necessity be greater than the secondary, and indeed increasingly so as the total intrauterine mortality becomes relatively greater. It is necessary, therefore, to consider two factors in order to deduce the primary sex-ratio from the secondary: (1) The sex-incidence in abortions, and (2) the relative frequency of abortions. Both of these factors i. e., the sex-ratio and the relative rate of mortality differ in the various periods of pregnancy, thus complicating the calculation of the primary sex-ratio. I have endeavored to gather from the literature estimates and statistics concerning the intrauterine mortality, beginning with the relative frequency of still-births and abortions.


The relative number of still-births differs to some extent according to the various authors. This is in part explained by the fact that the statistics are taken from different countries and at various times. Von Winckel (1903) reported that in Berlin, during a period of 70 to 80 years, the number of still-born males varied from 3.2 to 7 per cent of all births, that of females from 2.5 to 5.5 per cent. Rauber states that in Germany the still-births amount on an average to 4 per cent of 1,800,000 annual births. According to Carlberg (1886), the proportion of stillbirths to the total number of births in Livland lies between 2.58 and 2.90 per cent, while the percentage for western Europe lies between 4 and 4.5. According to Prinzing (1907), from 1891 to 1900 the following proportions of still-births occurred in every 100 births: Austria 2.9, Switzerland 3.6, Italy 3.9, Holland 4.3, Belgium 4.5, France 4.6. These figures are somewhat increased when expressed in percentage of the living-born. Computing from Auerbach's (1912) statistics of over 100,000 births in Budapest, the still-births amounted to 3.3 per cent of the living-born. Bucura (1905) found that among 40,169 births in the Clinic Chrobak in Vienna, 5.8 per cent were still-born; Le Maire (1906) found 5.7 per cent among 40,339 births in Copenhagen. Both of these figures are too high, inasmuch as these authors did not use the term still-born in the usual sense, a small number of abortions being included. Nichols (1907) quotes the following figures from registries of vital statistics: 3.6 per cent of total births of whites in the United States are still-births, and among the negro births of the District of Columbia 13.8 per cent are still-born. The percentage of still-births in Japan is 8, according to the same source. These figures from Nichols also include abortions, but inasmuch as these are not nearly so regularly reported as miscarriages in the last few months of pregnancy, they do not considerably increase the percentages.


The relative number of abortions is extremely difficult to determine, inasmuch as everywhere large numbers of them, especially of the earlier months, remain unknown, and even the most careful statistics will always be far from complete. The most reliable source of information is probably the estimate of the experienced obstetrician. Williams (1917) expresses himself on the question of frequency of abortions as follows:

"A conservative estimate would indicate that about every fifth or sixth pregnancy in private practice ends in abortion, and the percentage would be increased considerably were the very early cases taken into account, in which there is profuse loss of blood following the retardation of the menstrual period for a few weeks."

Other authors give different estimates. Franz (1898) found 15.4 per cent of pregnancies ending in abortion; Malins (1903) found 19.23 per cent; Taussig (1910) estimates that one abortion occurs to every 2.3 labors; Pearson (1897) one to every 2.5 labors; Mall (1910) calculates that there is one abortion to every 4 births at term; Ahlfeld (1898) estimates the same proportion; Whitehead (1848) assumes that every seventh pregnancy is interrupted by abortion, and states later on that only 13 per cent of married women who reach the end of the childbearing period escape having an unsuccessful pregnancy. Auerbach (1912) reports that, according to estimates for Berlin, abortions amount to one-sixth to one-tenth of the number of living-born. According to the same author, in Budapest, from 1901 to 1905, there were 111,139 living born, and from 1903 to 1905, 7,702 abortions. Assuming an approximately equal number of living-born for each year, they would amount to 66,678 for the years 1903 to 1905, the proportion of abortions to this number of living-births being 11.55 per cent. This percentage is doubtless too small. Auerbach himself assumes that many abortions of the earlier months are concealed. Under certain circumstances the proportion of miscarriages and abortions may increase tremendously. Bluhm (1918) reports that among working women in Berlin in 1915 and 1916 there were 190 abortions to every 100 full-term births, and attributes this partly to the harmful occupations in factories.


From these quotations it would seem most probable, on a rough average, that out of every 100 fertilized ova only 78, or even less, develop to term, the remainder being aborted. This intrauterine mortality may appear to many to be very high, but if we consider that during infancy this waste of life continues unabated the above estimate will not seem so improbable. Infant mortality during the first year of life varies in different countries; according to Phelps (1910), from slightly below 10 to 27 (in Russia) per 100 living-born; i. e., the number of infants dying within the first year of life may be more than one-fourth the number of living-born.[3]


It appears that the frequency of abortion is greatest in the first three months of pregnancy. Auerbach, in considerable material, found that half of the abortions fell within that period, yet it is for this period that our statistics are most incomplete. In reality, therefore, more than half the number of abortions must occur in the first three months. According to Franz, 42.6 per cent, and according to Diihrssen, even 59 per cent of abortions occur in the third month alone. Nichols tabulated the relative frequency of abortions and still-births at different periods of pregnancy from extensive statistics of Paris, Brussels, and the District of Columbia. Table G is based on this table by Nichols, the figures representing the percentages of the total number of abortions and still-births. These statistics are quoted for the sake of completeness and not because the writer believes that they represent the actual conditions; all that they show is that the farther back we go into fetal development the less regularly are abortions reported. There can be no question, for instance, that the abortions occurring in the third to the fifth month inclusive amount to very much more than 11.6 per cent of all the intrauterine deaths, and yet this percentage is given for the whites of the District of Columbia.


In the registration area of the United States in 1900 there were the following death-rates for infanta under 5 years of age (Billings, 1904): White: males 5.33, females 4.43; negroes: males 11.85, females 10.58 per 100. In the negro race infant mortality is more than twice as high as in whites.



Table G. Percentage distribution of abortions in different months of pregnancy. (Adapted from Nichols.)

Table H. Sex-ratio of still-born. (From Morgan.)

Scattered through the literature are reports on the sex-ratio of abortions. Those concerning the numerical proportion of still-births are more numerous and more reliable than those relating to younger fetuses and embryos that have been aborted. The sex-ratio of the still-born is much higher than that of the living-born. Table H, taken from Morgan (1913), gives the sex-ratios of still-born for some of the European countries. According to Nichols, the sex-ratio in over 11,000,000 still-births in Europe was 134.2, being highest in Spain (152.3). In Japan ample statistics of still-births yielded the unusually low sex-ratio of 107, a proportion which is only a little higher than that for living-born. It should be mentioned that in the statistics used by Nichols still-births comprise "in the main fetuses of more than six months' gestation." Such high sex-ratios of still-born as that of Walter (260) or that of Tschuprow (400) are probably based upon relatively limited material and do not represent true ratios.


Auerbach gives detailed information concerning the sex-ratio of abortions. His material is distributed among the different months of pregnancy as shown in table I. The sex-ratio of abortions during the first three months he assumes to be at least equal to that of abortions in the fourth month, i. e., 229. It seemed to him more probable, however, that it increases in constant proportion, and he therefore estimates 322 for the third month and 452 for the second month. Nichols tabulated, by sex and period of gestation, large statistics of "still-births" comprising almost 60,000 cases from Paris (1891-1902), and 4,400 whites and 7,500 negroes from the District of Columbia (1874-1902). In these statistics neither sex nor age determination is any too reliable, and the ratios in the different periods have therefore to be taken for what they are worth. Nichols's figures, in abstracted form, are given in table J. His conclusion is that the "ratio of male fetuses born dead is much the highest from the third to the fifth month; much lower from the sixth to the eighth month; and at term the ratio again rises." According to Carvallo (1912), the sex-ratio of abortions up to the fourth month is 250; this figure he calculated from the statistics of Paris in 1908. Korosy (1898)


Table I. Sex-ratio of abortions in different months. (From Auerbach.)

found the sex-ratio in 3,781 abortions to be 152.4 Pinard and Magnan (1913) report on 1,229 abortions, the ages of which are not stated. This material showed a sex-ratio of only 101.1. Rust (1902) also found the sex-ratio in 454 abortions from the first six months to be very low, i. e., 101.8. It is apparent how greatly these figures vary. A new contribution towards the knowledge of the sex-incidence in abortions is therefore not valueless, especially as great care has been taken to determine age and sex.


Table J. Sex-ratio of abortions and still-births in different months. (Adapted from Nichols.)

Table K. Sex-ratio of 1,410 fetuses from different months.

The last paper of the writer on this subject (1918) was based upon a relatively small amount of material; since then it has been more than doubled, increasing correspondingly the dependability of the conclusions drawn from it. The material in all consists of 1,410 fetuses, 1,249 of which are from the embryological collection of the Carnegie Institution of Washington. For data on 32 fetuses, I am indebted to Dr. Ingalls, and in addition I have made use of 57 fetuses published by Rauber and of 72 Filipino fetuses tabulated by Ruth (1918). Of this material only a small percentage is derived from induced abortions; the great majority were spontaneous ones and therefore represent the inevitable mortality. Age classification was based upon the sitting height (Keibel and Mall, 1910). Both normal and pathological fetuses were used in this study. These were for the most part white. In a limited number of cases no parental history was available; however, it is certain that a great majority of these also were white. Among the specimens other than white there is a preponderance of negroes, with a total of 201. The sex-ratios of these fetuses in the various months of pregnancy are given in table K. The pathological fetuses alone show a sex-ratio of 103.7, a proportion which indicates that the two sexes become pathological with about the same relative frequency. The sex-ratio of all the negro fetuses is 105.1, and of all the Filipinos 182.1. The material on which these latter ratios are based is, however, too small to draw any safe conclusion from them.


The greatest deviations in the ratios obtained by Auerbach and by Carvallo, on the one hand, and by the author on the other, occur in the third and .fourth months, in which Auerbach found the ratios to be 322 and 229, and Carvallo 250. The author's corresponding figures are very much lower, namely, 121 and 117, respectively. The great excess of male abortions in the early months of pregnancy, as found by Auerbach and Carvallo, may find its explanation in the fact that in the statistics used by them the sex was determined by different individuals who had not the specialized knowledge necessary for such determination on young fetuses. The same source of error exists in case of statistics of abortions throughout whole countries as used, for instance, by Nichols.[4] Early in the differentiation of the external genitalia only the expert can state the sex with certainty. At this time, and even later, the inexperienced, misled by the size of the clitoris as well as by other factors, may erroneously determine the female fetus as male. Fewer errors would be made on specimens from the latter part of the third and the fourth months if only those definitely male were reported as such, and all the doubtful ones were designated as female. Even granted that larger statistics might raise the sex-ratio of abortions, the latter would never reach the high figures stated by Auerbach, by Carvallo, and assumed by others. Just as the sex-ratio of mortality following birth varies according to age, so it is found to be true during the prenatal period. The author's material showed a high sex-ratio in the third and fourth months. In the fifth to the seventh month it became very much lower, to rise again in the eighth to the tenth month, attaining a higher figure than in any previous period. An analogous changing of the sex-ratio of intrauterine mortality was found by Auerbach and by Nichols with only slight differences in the duration of the periods of high and low ratios, but with considerable differences in the ratios themselves. Nothing is known in regard to the sex-ratio of abortions during the first two months of pregnancy; however, that of the third month might be used hypothetically for this period. The author's material from the fifth to the seventh month shows an average of 101.9, but owing to the variability in the individual months it is quite probable that the number of male and that of female fetuses perishing during the period from the fifth to the seventh month is relatively, not absolutely, equal. At any rate, it is apparent that during the middle third of intrauterine development the excess of male abortions is much lower than at the beginning and at the end of pregnancy.


In order to make use of the above citations and figures in computing the primary sex-ratio, rough and approximately average values must first be established. The following appear to be most probable: For each 100 living-born, with a sex-ratio of 105.5, there occur in the eighth to the tenth month 4 still-births with a sex ratio of 130; in the fifth to the seventh month 7 abortions with a sex-ratio of 106; in the fourth month 2 abortions with a sex-ratio of 120; and from conception to the end of the third month 15 abortions with a sex-ratio of 125. This makes altogether 28 abortions and still-births to every 100 living-born; i.e., 100 living-born to every 128 fertilized ova.[5] The primary sex-ratio found from these averages by simple mathematical operations is 108.74. The writer's last calculation of the primary sex-ratio (1918) resulted in 108.47. The very small increase in the corresponding value of the present study serves as a confirmation of the previous finding. The sex-ratio at conception was estimated by Bernoulli as 108.2. Slightly higher (108.7) is the ratio computed by Jendrassik (1911) from statistics collected by Bodio. Both of these figures are strikingly similar to that of the author. Lenhossek (1903) estimates the primary sex-ratio as 111, Auerbach as 116.4, but the latter believes that it would reach at least 125 if corrections were made.


Even if these approximate averages, which will become more exact only when based upon more extensive, careful statistics, must be accepted cum grano salis, it may nevertheless be stated with certainty that more males (not exceeding 10 per cent) are conceived; that at certain periods of pregnancy the relative mortality of males exceeds that of females by as much as one fourth or more; and that this, in connection with the very high intrauterine mortality, especially at the beginning of pregnancy, serves to lower the primary sex-ratio considerably throughout prenatal life.

Determination and Changes of the Primary Sex-Ratio

The question now arises as to what determines the sexes and their unequal distribution at conception. Its discussion dates back into antiquity; since Aristotle, philosophers and physiologists searched in vain for the key to this problem. The most fantastic theories were advanced, one of the oldest being that sex is correlated respectively with the right or left ovary or testicle (Hippocrates, Galen). In recent times much work has been done in trying to solve this problem. Among the most interesting theories stands the idea of the possibility of two distinct varieties of spermatozoa. Wilson (1905) distinguishes between male-producing and female-producing spermatozoa. This might lead to an unequal distribution of sexes at conception. Morgan (1913) suggested that it may be due to a difference in the rate of travel of the two types of sperm, or that a disease process, or a factor such as alcoholism, might affect one type to a greater degree than the other.


Hertwig (1912) attributes sex determination to the ovum or the degree of its maturation, an advanced stage of maturation producing males. In this way he attempts to explain the difference in sex-ratio according to social class. Thury (1863) proposed the idea that ova which are fertilized late may produce more males. Thus he explained the high sex-ratio among Jews who, on religious grounds, refrain from intercourse for 7 days following menstruation. Here may also be mentioned the recent investigations by Siegel on the relation of menstruation to sex, according to which coitus from the first to the eighth day after menstruation yields 86 per cent males; from the ninth to the fifteenth day it results in 65 per cent females; from the sixteenth to the twenty-second day, 85 per cent of conceptions are females, and in the remaining premenstrual period woman is practically sterile. Pryl and Jaeger, working independently, have confirmed Siegel's observations.[6] These findings contradict, in a way, those of Hertwig and Thury.


Lorenz (1898), Lenhosse'k (1903), and Orschansky (1903) are of the opinion that sex is sujbect to hereditary influences, inasmuch as they found families in which males predominated and those in which females appeared in excessive numbers. Newcomb (1904), Woods (1906), and Heron (1906) deny this and show that inheritance plays no part in the sex-ratio. Numerous authors attribute its variation to the absolute and relative ages of the parents. According to Rosenfeld (1900), there is a decided preponderance of male children born to young and old fathers, as compared with those of middle age. Francke, from the statistics of Norway, found this to be true in respect to young fathers, but reached an opposite conclusion as regards old ones. Dumont (1894) found for Paris a sex-ratio of 101.9 when the fathers were from 18 to 25 years, 104.2 when the fathers were between 25 and 50 years, and 97.5 when they were over 51 years. According to E. Bidder (1878), the sex-ratio of births by mothers under 18 and over 40 years is especially high, and Specht (1916) reports that the large majority of births by mothers under 16 years are male. Sadler (1830) stated that the relative ages of the parents determine the sex-ratio. The latter is 86.5 when the father is younger than the mother, 94.8 when both are of equal age, and reaches 163.2 when the father's age exceeds that of the mother by 16 or more years. Kollman (1890) obtained an opposite result, drawing the conclusion, which was based on extensive material, that the sex-ratio is high when the father is younger and low when he is older than the mother. At the same time he opposes the view that the absolute age of the mother has any influence whatever upon sex-ratio. Stieda, on the basis of his investigations, came to the conclusion that any influence on the part of the absolute ages of the parents is out of the question, as he noted the highest sex-ratio when the parents were of equal age. Numerous other authors have occupied themselves with the question of parental age as an influence upon sex-ratio, but only two additional ones will be mentioned, Boudin (1862) and Stadler (1878). The conflicting views which have been presented suffice to show that nothing definite is known concerning a correlation between the age of the parents and sex determination; in fact, such a correlation is hardly to be expected.


Pearl (1908), in a very careful study, demonstrated that there are more males produced when the parents are of different racial stock. The same effect of hybridization on the sex-ratio has been found by King (1911) for rats and by Guyer (1903) for birds. The well-known assertion that the sex-ratio rises after wars has evoked various attempts at explanation. The following few examples show best how little the different authors agree on this improbable relation. Ploss (1858, 1861) ascribes it to malnutrition of the mothers. Berner (1883) believes it to be due to the decreased competition which follows wars and which brings about an increased prosperity. Busing sees the cause in the increased sexual demands upon the male, which is also said to increase the sex-ratio in polygamy. According to Newcomb, following the American Civil War of 1861-65 no increase in sex-ratio was observed. Nichols found no effect of war upon the sex-ratio in France from 1806 to 1872. On the other hand, Henneberg (1897) claims to have found such an influence; he reported that in Holstein, between 1835 and 1845, the sex-ratio was 105.76, and after the period of the war, from 1846 to 1853, it was 106.67. In the few cases where such a difference was confirmed, it was so slight as not to exceed the normal variations of the secondary sex-ratio (it is in the latter that any difference would be found) and therefore is to be considered as such. These natural variations were shown by Lehr (1889), Carlberg, Nichols, and others; they may be very considerable, as reported by von Winckel (1903), according to whom the sex-ratio of new-born in Berlin showed extremes in a period of 100 years of 104.79 and 100.64. Variations of sex-ratio have been determined not only for individual years and groups of years, but also for the seasons. According to Goehlert (1889), in autumn and winter relatively few conceptions take place, but of these a higher percentage is male. The same conclusion was reached by Sormani (1883). Inasmuch as the studies of these authors were based upon statistics of births, without consideration of the relative frequency of abortions, their conclusions in regard to the sex-ratio at conception must be treated with caution.

Changes in the Secondary Sex-Ratio

The primary sex-ratio, as shown above, becomes transformed by an unequal intrauterine mortality of the two sexes into a different secondary ratio, and it is obvious, from a mathematical consideration, that the greater the proportion of abortions and still-births the lower will be the sex-ratio of living-born. Attempts have been made by a number of authors to explain the great mortality of males during certain periods of prenatal life. Carvallo simply says "les garcons sont plus fragiles"; Auerbach, also, considers the male fetus less resistant. Grassl (1912) gives as an explanation the supposition of a difference in the viability of the germ plasma. Jendrassik speaks of hereditary reduction of vitality among the excess of males. Ewart (1918) suggested that "it is possible here, of course, we have no data that the female conception may graft itself on the lining membrane of the uterus more easily than the male." Rauber explains the greater mortality of males by the greater demands of the larger fetuses upon the mother, the latter not always being able to meet them; the production of a female does not require as much from the mother. Lillie (1917) offers the suggestion that the greater mortality among male fetuses is a result of disturbance of the equilibrium that protects the male from the sex hormones of the mother. These are all more or less plausible hypotheses lacking in proofs. As to any real understanding of the unquestionably higher mortality of male fetuses we are still at a loss; attention may be called, however, to the fact that this sexual difference in vitality and power of resistance against disease is not restricted to the period of intrauterine life, but is found also in the first few years of postnatal life, at which time occupation, child-bearing, and other factors can not be held responsible for the difference between the male and female death-rate.


The excess of male still-births is ascribed by most authors to the more difficult labor attendant upon the greater size of the male,[7] especially the circumference of the head. Button (1910) is of the opinion that at the time of birth the bones of the male skull are, as a rule, more firmly ossified than those of the female. He states also that with the advance of civilization the pelvic development in women is not proportionate to the cephalic development in infants. This perhaps explains the conclusion reached by Bluhm (1912), that the relative number of therapeutically induced premature births is increasing. That labor in case of male children more often demands artificial aid from the obstetrician than in case of females is shown by Prinzing, according to whom 6.18 per cent of male births in Wiirttemberg called for operative measures, as compared with 4.67 per cent of the female births. This, however, is not due alone to the greater size of the male infant. Von Winckel found that in 566 new-born infants of over 4,000 grams weight operative aid was necessary in only 3 per cent more cases than in births of lighter babies. Furthermore, the more difficult labor of the larger male child can not in itself be held responsible for the high sex-ratio of still-born infants, inasmuch as, according to Treichler (1895), 29.6 per cent, and, according to Prinzing (1907), 32.6 per cent, of all still-births are premature, and in the sex ratio of these size plays but little part. According to Ladame (1904), in Switzerland the number dying during labor constituted only 36.4 per cent of the total still-births. Finally, Von Winckel found that the death-rate among 1,000 new-born of over 4,000 grams weight was only 4.17 per cent.

Sex-ratio has frequently been studied in relation to the pelvic diameters of the mother. The results are somewhat conflicting. Hoffmann (1887), Dohrn (1888), and Orschansky (1894) may be mentioned, according to whom the sexratio in children of mothers with narrow pelves is low. In contrast to this, Linden (1884) states it to be 133 in 360 births in which the mothers had narrow pelves. In case the size of the pelvis really has an influence, this can be exerted only upon the secondary sex-ratio in the way of elimination. In the same manner it seems evident that many of the factors which apparently affect the sex-ratio do not influence it at the time of conception; that is to say, they do not have any sexdetermining effect, but by their influence upon intrauterine mortality they change only the sex-ratio of the living-born. The well-known fact that the secondary sex-ratio among Jews is relatively high is explained by Diising on the ground of incest, blood marriages being of frequent occurrence in that race. Schultze (1903), on the other hand, concluded that inbreeding has no effect upon sex determination and the same conclusion was reached by King (1918). Busing, in his conclusions, failed to make a distinction between the sex-ratio at birth and that at conception; the latter is probably not different in Jews from what it is in other white races, but changes less by reason of the relatively fewer abortions and still-births among Jews, resulting in a higher secondary sex-ratio. The relative infrequency of abortions among Jews has been shown, for instance, by Auerbach. One finds frequently the assumption that the negro produces fewer sons than other races another conclusion drawn from statistics of the new-born alone. Nichols pointed out that in the Bistrict of Columbia still-births and abortions among the colored population amount to 13.8 per cent of the living-born, whereas in the white it is only 6.5 per cent. This difference is responsible for the different secondary sex-ratios of the two races (103.1 in negroes and 106.2 in whites). Punnett (1903) and others have shown that the births among classes of lower social status present a lower sex-ratio than those of the rich. The explanation lies again in the fact that the greater frequency of abortions among women of the working classes, who can spare themselves less during pregnancy and in whom pregnancies occur in more rapid succession,[8] results in a corresponding reduction in the sex-ratio, which probably was originally equal in the two classes. In addition, this greater reduction of the primary sex-ratio in the poorer classes is due to their higher percentage of stillbirths. According to Conrad in Halle, among laborers it was 5 per cent, while among the upper classes it was only 2.1 per cent; and according to Verrijn Stuart (1901), in Holland, among the poor, it was 3.16 per cent and among the rich 2.5 per cent of all births.

A further example illustrating how the primary sex-ratio was erroneously thought to be influenced is shown in its difference in legitimate and illegitimate children. Heape (1909) states that the sex-ratio of legitimate births among the white population of Cuba is 109.0, still-births included; that of the illegitimate only 105.95. There is even a greater difference among negroes, the sex-ratio being 97.91 for illegitimate children and 107.73 for legitimate ones. Heape immediately draws the conclusion that illegitimate unions result more often in the conception of females than do legitimate ones. According to Busing, the sex-ratio of legitimate births in Prussia, between the years 1875 and 1887, was 106.37; that of illegitimate only 105.54. The still-births in legitimate unions amounted to 3.91 per cent, in illegitimate ones to 5.32 per cent. A corresponding difference was demonstrated by Bertillon (1896) in the frequency of legitimate and illegitimate abortions. The greater mortality of illegitimate fetuses reduced the sex-ratio to a greater degree. The rule that the sex-ratio is greater in legitimate than in illegitimate births is not, however, without exceptions. Srdinko (1907) found that the sex-ratio of legitimate births in Austria was lower than that of the illegitimate, and explains this by the fact that the illegitimate are for the most part Jewish, in which race abortions are less frequent. Further exceptions are reported by Nichols in the case of England and Scotland. According to the last-mentioned author, there is an especially high sex-ratio in legitimate births as compared with illegitimate ones in Rhode Island (104.7 to 98.8), Portugal (107.1 to 100.5), and Greece (114.0 to 96.9). He, too, found a greater frequency of still-births and abortions in illegitimate pregnancies. Obviously, in such cases there is more concealment and consequently still less complete statistics are available than in the case of legitimate pregnancies ending in abortion.


According to a number of authors, the sex-ratio of first-born is greater than that of subsequent births, as demonstrated, for instance, in a table by Newcomb (1904). Lewis and Lewis (1906) report that in Scotland the sex-ratio of the .firstborn was 105.4, and that of subsequent births 104.8. The secondary sex-ratio is especially high as regards older primiparae, as shown by Ahlfeld (1872, 1876), Janke (1888), and Bidder (1893). That this is also due, in part at least, to different intrauterine mortality may be supported by the citation from Franz that abortions are more than twice as frequent in multipart as in primipara?. Moreover, the first-born children are appreciably smaller than subsequent ones, as demonstrated by Schaetzel (1893)[9] and others, a condition which might suggest a lower rate of mortality during labor. According to Duncan and Duke (1917), however, stillbirths are more frequent among first-born than among second and third-born, in spite of their smaller size; only in the case of children from the sixth pregnancy does the percentage of still-births exceed the one of the first-born. However, inasmuch as abortions are much more frequent than still-births, comparatively little importance can be attached to this. The number of children in a family has also been correlated with the sex-ratio; Geissler (1889) found that in families with seven or more children there is a greater proportion of sons than in families with 2 to 7 children. Punnett (1903) reached just the opposite conclusion. The former result was confirmed by several other observers (von Korosy, Janke, and Nichols) ; Nichols considers it very probable that in large families the higher proportion of sons is due to a smaller number of abortions, leaving a larger number of children to be born alive, and thus their sex-ratio more closely approaches the primary sex-ratio.


Besides the above-mentioned causes for the variations found in sex-ratio, many more have been discussed in the literature in an effort to throw light on the question of sex determination. Only a few, if any, of these factors actually exert any influence upon the primary sex-ratio. The changes have all been found in the secondary sex-ratio and the probability is great that the factors causing them affected only the intrauterine death-rate. This is especially true in regard to


changes resulting from locality, from age, nutrition, and health of the mother. Sormani (1883) reported that in Italy the sex-ratio of births in urban districts was 104.9 and in rural districts 107.0. Nichols, in contrast to this, states that in Paris the sex-ratio of new-born was 103.7 and in the remainder of France 104.3; while in Paris the mortality in utero was 7.7 per cent of births, in other parts of France only 4.4 per cent. In regard to the alleged influence exerted by the age of the mother on the sex-ratio of her children, the finding of Dempsey (1919) may serve as proof that this influence does not concern the primary sex-ratio. This author finds that still-births in women of 30 years of age and over were more than four times as numerous as in women under 30.


To briefly sum up the results of this study, the author believes he has succeeded in correcting two errors frequently found in the literature:


First, that the relation of the sexes at conception does not, as frequently stated, show an extremely high preponderance of males, but a surplus of 10 per cent at most.


Second, that a great number of factors claimed to influence the sex-ratio at conception, if playing any role at all, are only sex-eliminating during intrauterine life and have no effect upon sex determination.


Further results of interest are the marked fluctuations in the sex-incidence of abortions and still-births in different periods of development and also the great changes in the death-rate during intrauterine life. These facts may serve as a helpful guide in the search for the cause of the greater mortality among male fetuses. 2

1 Schenk (1898) asserts that diabetic mothers bear more female than male children. 1 This is not restricted to man, but has been found also for the rat. (King, 1921.)



  1. ??
  2. I use the term cyemata as a collective name for fertilized ova, embryos and fetuses. (See Meyer, 1919.)
  3. The writer had an opportunity to compare the death certificates of a large number of abortions and still-births with his own findings on the specimens themselves, and found a surprisingly high percentage of errors in sex determination in the certificates, sometimes even on full-term fetuses.
  4. The relative number of fertilized ova, estimated by Rauber to be 100 to 76 living-born or, calculated as above, 131.6 to 100 living-born, is somewhat larger than that obtained by the author.
  5. The most common method in use for representing the sex-ratio is to determine a number which indicates the proportion of males to every 100 females. Unless otherwise stated, this ia the method used herein. The sex-ratio is also frequently termed masculinity.
  6. These papers could not be obtained by the author, but are discussed by Nilsson.
  7. Von Winckel found among 1,000 new-born, of over 4,000 grams weight and 52 cm. length, a sex-ratio of 226.
  8. According to Diising (1884), the longer the intervals between births the higher is the sex-ratio.
  9. Hansen (1913) reports that in Denmark the first born weigh on an average 3,457 grams; the second born 3,607 grams; the third born 3,698 grams, the difference between the first and second being much greater than between subsequent ones. The figures for weight at birth, given by Heiberg (1911), show an analogous relation.




Embryology - 21 Oct 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Mall FP. and Meyer AW. Studies on abortuses: a survey of pathologic ova in the Carnegie Embryological Collection. (1921) Contrib. Embryol., Carnegie Inst. Wash. Publ. 275, 12: 1-364.

In this historic 1921 pathology paper, figures and plates of abnormal embryos are not suitable for young students.

1921 Carnegie Collection - Abnormal: Preface | 1 Collection origin | 2 Care and utilization | 3 Classification | 4 Pathologic analysis | 5 Size | 6 Sex incidence | 7 Localized anomalies | 8 Hydatiform uterine | 9 Hydatiform tubal | Chapter 10 Alleged superfetation | 11 Ovarian Pregnancy | 12 Lysis and resorption | 13 Postmortem intrauterine | 14 Hofbauer cells | 15 Villi | 16 Villous nodules | 17 Syphilitic changes | 18 Aspects | Bibliography | Figures | Contribution No.56 | Contributions Series | Embryology History

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)