r COLUMBIA UNIVERSITY LECTURES HEREDITY AND SEX BY THOMAS HUNT MORGAN, Ph.D PROFESSOR OF EXPERIMENTAL ZOOLOGY IN COLUMBIA UNIVERSITY SECOND (REVISED) EDITION COLUMBIA UNIVERSITY PRESS 1914 All rights reserved | I \\ COPTKIGHT, 1913, Bt COLUMBIA UNIVERSITY PRESS. Set up and electrotyped. Published November, 1913. Norbioob fates J-8-CnshIng Co—Berwick & Smith Co Norwood, Mass., U.S.A. ^ INTRODUCTION Two lines of research have developed with surpris- ing rapidity in recent years. Their development has been independent, but at many stages in their progress they have looked to each other for help. The study of the cell has furnished some fundamental facts connected with problems of heredity. The modern study of heredity has proven itself to be an instrument even more subtle in the analysis of the materials of the germ-cells than actual observations on the germ- cells themselves. In the following chapters it has been my aim to point out, wherever possible, the bearing of cytological studies on heredity, and of the study of heredity on the analysis of the germinal materials. The time has come, I think, when a failure to recog- nize the close bond between these two modern lines of advance can no longer be interpreted as a wise or cautious skepticism. It seems to me to indicate rather a failure to appreciate what is being done at present, and what has been accomplished. It may not be desir- able to accept everything that is new, but it is cer- tainly undesirable to reject what is new because of its newness, or because one has failed to keep in touch with the times. A hypercritical spirit in science does not always mean greater profundity, nor is our attitude toward science more correct because we are unduly vl INTRODUCTION skeptical toward every advance. Our usefulness will, in the long run. be proven by whether or not we have been discriminating and sympathetic in our attitude toward the importanl discoveries of our time. While everyone will probably admit such generalities, some of us may '•.•ill those who accept less than ourselves con- Bervatives ; others of us who accept more will be called rash or intemperate. To maintain the right balance is the hardest task we have to meet. In attempting to bring together, and to interpret, work that is still in the making I cannot hope to have always made the right choice, bul I may hope at least for some indulgence from those who realize the difficulties, and who think with me that it may be worth while to make the attempt to point out to those who are not specialists what specialists are thinking about and doing. What I most fear is that in thus attempting to for- mulate some of the difficult problems of present-day interest to zoologists I may appear to make at times unqualified statements in a. dogmatic spirit. I beg to remind the reader and possible critic that the writer holds all conclusions in science relative, and subject to change, for change in science does not mean so much that what has gone before was wrong as the discovery of a better strategic position than the one last held. TABLE OF CONTENTS Introduction PAGES v-vi CHAPTER I THE EVOLUTION OF SEX 1. Reproduction, a Distinctive Feature of Living Things 1-4 2. The "Meaning" of Sexual Reproduction . . . 4-15 3. The Body and the Germ-plasm .... 15-19 4. The Early Isolation of the Germ-cells . . 20-23 5. The Appearance of the Accessory Organs of Reproduction 23-26 6. The Secondary Sexual Characters . . . 26-31 7. The Sexual Instincts 31-34 CHAPTER II THE MECHANISM OF SEX-DETERMINATION 1. The Maturation of the Egg and the Sperm . 35-40 2. The Cytological Evidence 40-54 a. Protenor ........ 11-14 h. Lygaeus 41-16 c. Oncopeltus 46-48 d. Ascaris . 49-52 e. Aphids and Phylloxerans ..... 52-54 3. The Experimental Evidence 55-72 a. The Experiments on Sea-urchins' Eggs . . 55-63 b. The Evidence from Sex-linked Inheritance . 63-72 H vii Vlll TABLE OF CONTENTS CHAPTER III llli: MKXDKI.IAX PRINCIPLES OF HEREDITY \M» THEIK BEARING ON SEX ]. Ml \l>l L*8 I >l-< <>\ KIIIKS 2. 'I'm Heredity of One Pair of Characters. ;. Tin Hi i;i cditj of a Sex-linked Character . I. Tin Heredity of Two Pairs of Characters "». The Heredity of Two Sex-linked Characters 6. A Tin or-* of Linkage 7. Three Sex-linked Factors .... PAGES 73-75 75-80 80-84 84-88 88-93 93-97 98-100 CHAPTER IV SECONDARY SEXUAL CHARACTERS AND THEIR RELATION TO DARWIN'S THEORY OF SEX- UAL SELECTION L. Iiik Occurrence of Secondary Sexual Charac- rERS in the Animal Kingdom .... 2. Courtship •"•• Vl i and Secondary Sexual Characters . I. Continuous Variation as a Basis for Selection 5. l»i-« ontindoi - Variation or Mutation as a Basis I OR Si II i HON 101-112 112-120 120-121 121-125 125-131 CHAPTER V THE EFFECTS OF CASTRATION AND OF TRANS- PLANTATION ON THE SECONDARY SEXUAL CH LRACTERS Operations on Mammals .... 132-141 142-144 •_'. < >ii i: \ i iuns on Birds . I >!•! R \ I KiN- .,n \\lnill'.| \ •■ In iv RN U Si < i:i i [ONS . 1 'it i: v i [onb ON [nS] ' Parabitn Castration of Crustacea 145-146 146-147 148-155 155-158 TABLE OF CONTENTS IX CHAPTER VI GYNANDROMORPHISM, HERMAPHRODITISM, PARTHENOGENESIS, AND SEX PAGE 1. gynandromorphism 161-167 2. Hermaphroditism 167-173 3. Parthenogenesis 173-188 4. Artificial Parthenogenesis 188-193 CHAPTER VII FERTILITY 1. Inbreeding 191—199 2. Cross-breeding 200-207 3. Sexual Reproduction in Paramecium . . . 207-211 4. Theories of Fertility 211-219 CHAPTER VIII SPECIAL CASES OF SEX-INHERITANCE 1. Sex in Bees 2. A Sex-linked Lethal Factor .... 3. Non-disjunction of the Sex-chromosomes 4. The Vanishing Males of the Nematodes 5. Sex-ratios in Hybrid Birds and in Crossed Races in Man 6. Sex-ratios in Frogs 7. Sex-ratios in Man 8. The Abandoned View that External Conditions Determine Sex ...... 9. Sex-determination in Man .... BIBLIOGRHAPHY INDEX 220-221 221-223 223-224 224-225 225-227 228-229 •2-2U-S-V2 2:»2-23li 236-249 2.") 1-278 279 282 HEREDITY AND SEX CHAPTER I The Evolution of Sex Animals and plants living to-day reproduce them- selves in a great variety of ways. With a modicum of ingenuity we can arrange the different ways in series beginning with the simplest and ending with the more complex. In a word, we can construct systems of evolution, and we like to think that these systems reveal to us something about the evolutionary process that has taken place. There can be no doubt that our minds are greatly impressed by the construction of a graded series of stages connecting the simpler with the complex. It is true that such a series shows us how the simple forms might conceivably pass by almost insensible (or at least by overlapping) stages to the most complicated forms. This evidence reassures us that a process of evolution could have taken place in the imagined order. But our satisfaction is superficial if we imagine that such a survey gives much insight either into the causal processes that have produced the successive stages, or into the interpretation of these stages after they have been produced. Such a series in the present case would culminate in a process of sexual reproduction with males and 1 HKREDITY AND SEX females as the actors in the drama. But if we are asked whal advantage, if any, has resulted from the process of sexual reproduction, carried out on the two-sex scheme, we must confess to some un- certainty. The most important fact that we know about living matter is its inordinate power of increasing itself. If all the fifteen million eggs laid by the conger eel were t.> grow up, and in turn reproduce, in ten years the sea would be a wriggling mass of fish. \ single infusorian, produced in seven days 935 de- scendants. One species, stylonichia, produced in 6j/2 days a m:iss of protoplasm weighing one kilogram. At the end of 30 days, at the same rate, the number of kilograms would be 1 followed by 44 zeros, or a mass of protoplasm a million times larger than the volume of the sun. Another minute organism, hydatina, produces about ::< i eggs. At the end of a year (65 generations), if all the offspring survived, they would form a sphere whose limits would extend beyond the confines of the known universe. The omnipresent English sparrow would produce in 20 years, if none died except from old age, so many de- scendants that there would be one sparrow for every square inch of the State of Illinois. Even slow- breeding man has doubled his numbers in 25 years. At the same rate there would in 1000 years not standing room on the surface of the earth for his offspring. I have not gone into these calculations and will THE EVOLUTION OF SEX 3 not vouch for them all, but whether they are en- tirely correct or only partially so, they give a rough idea at least of the stupendous power of growth. There are three checks to this process : First, the food supply is insufficient — you starve ; second, ani- mals eat each other - - you feed ; third, substances are produced by the activity of the body itself that inter- fere with its powers of growth — you poison yourself. The laws of food supply and the appetites of enemies are as inexorable as fate. Life may be defined as a constant attempt to find the one and avoid the other. But we are concerned here with the third point, the methods that have been devised of escape from the limitations of the body itself. This is found in repro- duction. The simplest possible device is to divide. This makes dispersal possible with an increased chance of finding food, and of escaping annihilation, and at the same time by reducing the mass permits of a more ready escape of the by-products of the living machine. Reproduction by simple division is a well-known pro- cess in many of the lower animals and plants ; it is almost universal in one-celled forms, and not unknown even in many-celled organisms. Amoeba and para- mcecium are the stock cases for unicellular animals ; many plants reproduce by buds, tubers, stolons, or shoots ; hydroids and sea-anemones both divide and bud ; many planarians, and some worms, divide trans- versely to produce two new individuals. But these methods of reproduction are limited to simple structures where concentration and division of labor amongst the organs has not been carried to an extreme. In con- sequence, what each part lacks after the division can be , HEREDITY AND SEX quickly made good, for delay, if prolonged, would increase the chances of death. But i here is another method of division that is almost universal and is utilized by high and by low forms alike : individual cells, as eggs, are set free from the rest of the body. Since they represent so small a part of the body,an immense number of them may be produced on the chance thai a few will escape the dangers of the long road leading to maturity. Sometimes the eggs are protected by jelly, or by shells, or by being trans- parent, or by being hidden in the ground or under lies, or even in the body of the parent. Under these circumstances the animal ventures to produce eggs with ;i large amounl of food stored up for the young embryo. - far reaching were the benefits of reproduction by eggs that it has been followed by almost every species in the animal and plant kingdom. It is ad- hered t<. even in those cases where the animals follow other grosser methods of separation at the same time. We find, however, a strange limitation has been put upon the process of reproduction by eggs. Before the _ begins its development it must be fertilized. Cells from two individuals must come together to produce a new one. The meaning of this process has baffled biologists ever since the changes that take place during fertili- Bation were firsl discovered; in fact, long before the actual processes that take place were in the least un- derstood. There is a rather extensive and antiquated literature dealing with the part of the male and of the female in the process of . procreation. It would take us too far to attempt to deal with these questions THE EVOLUTION OF SEX 5 in their historical aspects, but some of their most modern aspects may well arrest our attention. In the simplest cases, as shown by some of the one- celled organisms, two individuals fuse into a single one (Fig. 1) ; in other related organisms the two in- dividuals that fuse may be unequal in size. Some- times we speak of these as male and female, but it is questionable whether we should apply to these unicellular types the same names that we use for the Fig. 1. — Union of two individuals (Stephanosphcera pluviaZis) to form a single individual. (After Doflein.) many-celled forms where the word sex applies to the soma or body, and not to the germ cells. One of the best known cases of conjugation is that of paramcecium. Under certain conditions two in- dividuals unite and partially fuse together. An in- terchange of certain bodies, the micronuclei, then takes place, as shown in Fig. 2, and in diagram, Fig. 3. The two conjugating paramcecia next separate, and each begins a new cycle of divisions. Here each individual may be said to have fertilized the other. The process recalls what takes place in hermaphroditic animals of higher groups in the sense that sperm from one indi- vidual fertilizes eggs of the other. We owe to Maupas the inauguration of an epoch- making series of studies based on phenomena like this in paramcecium. HEREDITY AND SEX Conjugation in Paramcecium. The micronucleus in one indi- vidual is represented in black, in the other by cross-lines. The macro- oucleua in both is stippled. A-C, division of micronucleus into 2 and 1 nuclei; '" l>. elongation of conjugation nuclei, which interchange and nbine in /•.': /•' ./, consecutive stage in one ex-conjugant to show three iaiona of new micronucleus to produce eight micronuclei (J). In lower port of diagram the first two divisions of the ex-conjugant (J) with eight micronuclei are shown, by means of which a redistribution of the eight micronuclei takes place. See also Fig. 100. THE EVOLUTION OF SEX I in E Fig. 3. — The nu lei of two individuals of paramoecium in I (homozygous in certain factors, and heterozygous in other factors), are represented as divid- ing twice (in II and III); the first division, II, is represented as reducing, i.e. segregation occurs ; the second division, III, is represented as equational, i.e. no reduction but division of factors, as in the next or conjugation division, IV, also. g HEREDITY AND SEX Maupas found by following from generation to generation the division of some of these protozoa that the division rate slowly declines and finally comes to an end. He found that if a debilitated individual conjugates with a wild individual, the death of the race is prevented, but Maupas did not claim that through conjugation the division rate was restored. On the contrary he found it is lower for a time. He also discovered that conjugation between two related individuals of these weakened strains produced in. beneficial results. Biitschli had earlier (1876) suggested that conjugation means rejuvenation or renewal of youth, and Maupas' results have sometimes been cited as supporting this view. Later work has thrown many doubts on this interpretation and has raised a number of new issues. In the first place, the question arose whether the decline that Maupas observed in the rate of division may not have been due to the uniform conditions under which his cultures were maintained, or to an insuffi- ciency in some ingredient of these cultures rather than to lack of conjugation. Probably this is true, for ( alkins has shown that by putting a declining race into a different medium the original division rate may be restored. Woodruff has used as culture media a great variety of food stuffs and has succeeded in keep- ing his lines without loss of vigor through 3000 gen- erations. Maupas records a decline in other related protozoa at the end of a few hundred generations. Biitschli's idea that by the temporary union (with interchange of micronuclei) of two weak individuals two vigorous individuals could be produced seems THE EVOLUTION OF SEX 9 mysterious ; unless it can be made more explicit, it does not seem in accord with our physico-chemical conceptions. Jennings, who has more recently studied in greater detail the process of division and conjugation in paramcecium, has found evidence on which to base a more explicit statement as to the meaning of rejuve- nescence through conjugation. Jennings' work is safeguarded at every turn by care- ful controls, and owing in large part to these controls his results make the interpretations more certain. He found in a vigorous race, that conjugated at rather definite intervals, that after conjugation the division rate was not greater than it had been before, but on the contrary was slower - - a fact known, as he points out, to Maupas and to Hertwig. Conjugation does not rejuvenate in this sense. Jennings states that, since his race was at the be- ginning vigorous, the objection might be raised that the conditions were not entirely fulfilled, for his pred- ecessors had concluded that it is a weakened race that was saved from annihilation by the process. In order to meet this objection he took some individuals from his stock and reared them in a small amount of culture fluid on a slide. After a time they became weakened and their rate of division was retarded. He then al- lowed them to conjugate, and reared the conjugants. Most of these were not benefited in the least by the process, and soon died. A few improved and began to multiply, but even then not so fast as paramcecia in the control cultures that had been prevented from con- jugating. Still others gave intermediate rates of division. 10 HEREDITY AND SEX He concludes that conjugation is not in itself bene- ficial to all conjugants, but that the essence of the pro- cesa is that a recombination of the hereditary traits occurs a- shown in the diagram, Fig. 3 and 4. Some Fig. 4. — Illustrating conjugation between two stocks, with pairs of factors .1, B, (', I), and a, b, c, d ; and union of pairs into Aa, Bb, Cc, Dd. \ftir these separate, their possible recombinations are shown in the 16 smaller circles. (After Wilson.) of these new combinations are beneficial for special conditions -others not. The offspring of those con- jugants thai have made favorable combinations will mm. ii crowd out the descendants of other conjugants that have made mediocre or injurious combinations. Hence, in a mass culture containing at all times large THE EVOLUTION OF SEX 11 numbers of individuals, the maximum division rate is kept up, because, at any one time, the majority of the individuals come from the combinations favorable to that special environment. There are certain points in this argument that call for further consideration. In a mass culture the fa- vorable combinations for that culture will soon be made, if conjugation is taking place. At least this is true if such combinations are homogeneous (homozygous, in technical language). Under such circumstances the race will become a pure strain, and further conjugation could do nothing for it even if it were transferred to a medium unsuited to it. In the ordinary division of a cell every single de- terminer divides and each of the new cells receives half of each determiner. Hence in the case of para- mcecium all the descendants of a given paramcecium that are produced by division must be exactly alike. But in preparation for conjugation a different pro- cess may be supposed to take place, as in higher animals, among the determiners. The determiners unite in pairs and then, by division, separate from each other, Fig. 4. In consequence the number of determiners is reduced to half. Each group of deter- miners will be different from the parent group, pro- vided the two determiners that united were not identical. If after this has occurred conjugation takes place, the process not only restores the total number of determiners in each conjugant, but gives new groups that differ from both of the original groups. The maintenance of the equilibrium between an 12 HEREDITY AND SEX organism and its environment must be a very delicate matter. One combination may be best suited to one environment, and another combination to another. ( Jonjugation brings about in a population a vast num- ber of combinations, some of which may be suited to the time and place where they occur. These survive and produce the next generation. Jennings' experiments show, if I understand him correctly, that the race he used was not homogeneous in its hereditary elements ; for when two individuals conjugated, new combinations of the elements were formed. It seems probable, therefore, that the chemi- cal equilibrium of paramcecium is maintained by the presence of not too much of some, or too little of other, hereditary materials. In a word, its favorable com- binations are mixed or heterozygous. The meaning of conjugation, and by implication, the meaning of fertilization in higher forms is from this point of view as follows : - - In many forms the race, as a whole, is best maintained by adapting itself to a widely varied environment. A heterozygous or hybrid con- stitution makes this possible, and is more likely to perpetuate itself in the long run than a homozygous race that is from the nature of the case suited to a more limited range of external conditions. What bearing has this conclusion on the problem of the evolution of sex and of sexual reproduction? This is a question that is certain to be asked. I am not sure that it is wise to try to answer it at present, in the first place because of the uncertainty about the conclusions themselves, and in the next place, because, personally, I think it very unfair and often very unfor- THE EVOLUTION OF SEX 13 tunate to measure the importance of every result by its relation to the theory of evolution. But with this understanding I may venture upon a few suggestions. If a variation should arise in a hermaphroditic species (already reproducing sexually) that made cross- fertilization more likely than self-fertilization, and if, as a rule, the hybrid condition (however this may be explained) is more vigorous in the sense that it leaves more offspring, such a variation would survive, other things being equal. But the establishment of the contrivance in the species by means of which it is more likely to cross- fertilize, might in another sense act as a drawback. Should weak individuals appear, they, too, may be perpetuated, for on crossing, their weakness is concealed and their offspring are vigorous owing to their hybrid condition. The race will be the loser in so far as re- cessive or weak combinations will continue to appear, as they do in many small communities that have some deficiency in their race ; but it is a question whether the vigor that comes from mixing may not more than com- pensate for the loss due to the continual appearance of weakened individuals. This argument applies to a supposed advantage within the species. But recombination of what already exists will not lead to the development of anything that is essentially new. Evolution, however, is con- cerned with the appearance and maintenance of new characters. Admitting that sexual reproduction proved an advantage to species, and especially so when com- bined with a better chance of cross-fertilization, the machinery would be at hand by means of which any I i HEREDITY AND SEX new character that appeared would be grafted, so to Bpeak, on t « i the body of the species in which it appeared. Once introduced it would be brought into combination with all the possible combinations, or races, already existing within the species. Some of the hybrid com- binations thus formed might be very vigorous and would survive. This reasoning, while hypothetical, and, per- haps not convincing, points at least to a way in which new varieties may become incorporated into the body of a species and assist in the process of evolution. It might be argued against this view that the same end would be gained, if a new advantageous variation arose in a species that propagated by non-sexual methods or in a species that propagated by self-fertili- zation. The offspring of such individuals would trans- mit their new character more directly to the offspring. Evolution may, of course, at times have come about in this way, and it is known that in many plants self- fertilization is largely or exclusively followed. But in a species in which cross-fertilization was the estab- lished means of propagation, the new character would be brought into relation with all the other variations that are found in the component races and increase thereby its chances of favorable combinations. We have in recent years come to see that a new heritable character is not lost by crossing, or even weakened by "blending," as was formerly supposed to be the case; hence no loss to the character itself will result in the union with oilier strains, or races, within the species. If then we cannot explain the origin of sexual re- production by means of the theory of evolution, we THE EVOLUTION OF SEX 15 can at least see how the process once begun might be utilized in the building up of new combinations ; and to-day evolution has come to mean not so much a study of the origination of new characters as the method by which new characters become established after they have appeared. THE BODY AND THE GERM-PLASM As I have said, it is not unusual to speak of the uni- cellular animals and plants as sexual individuals, and where one of them is larger than the other it is some- times called the female and the smaller the male. But in many-celled animals we mean by sex something different, for the term applies to the body or soma, and not to the reproductive cells at all. The reproductive cells are eggs and sperm. It leads to a good deal of confusion to speak of the reproductive cells as male and female. In the next chapter it will be pointed out that the eggs and sperm carry certain materials ; and that certain combinations of these materials, after fer- tilization has occurred, produce females ; other combi- nations produce males ; but males and females, as such, do not exist until after fertilization has taken place. The first step, then, in the evolution of sex was taken when colonies of many cells appeared. We find a division of labor in these many-celled organisms ; the germ-cells are hidden away inside and are kept apart from the wear and tear of life. Their maintenance and protection are taken over by the other cells of the colony. Even among the simplest colonial forms we find that some colonies become specialized for the pro- duction of small, active germ-cells. These colonies Hi BEREDITY AND SEX are called the males, or sperm-producing colonies. The other colonics specialize to produce larger germ-cells — the eggs. These colonies are called females or egg-pro- ducing colonics. Sex has appeared in the living world. To-day we are only beginning to appreciate the far- reaching significance of this separation into the immor- tal germ-cells and the mortal body, for there emerges the possibility of endless relations between the body on the one hand and the germ-cells on the other. What- ever the body shows in the way of new characters or new ways of reacting must somehow be represented in the germ-cells if such characters are to be perpetu- ated. The germ-cells show no visible modification to represent their potential characters. Hence the classi- cal conundrum-- whether the hen appeared before the egg, or the egg before the hen? Modern biology has answered the question with some assurance. The egg came first, the hen afterwards, we answer dogmati- cally, because we can understand how any change in the egg will show itself in the next generation — in the new hen, for instance; but despite a vast amount of arguing no one has shown how a new hen could get her newness into the old-fashioned eggs. Few biological questions have been more combated ihan this attempt to isolate the germ-tract from the influence of the body. Nussbaum was amongst the first, if not the first, to draw attention to this distinc- tion, bul the credit of 'pointing out its importance is generally given to Weismann, whose fascinating specu- lations start from this idea. For Weismann, the germ- cella are immortal- -the soma alone has the stigma of death upon it. Each generation hands to the next THE EVOLUTION OF SEX 17 one the_ immortal stream unmodified by the experience oi_the_body. What we call the individual, male or female, is the protecting husk. In a sense the body is transient -- temporary. Its chief " purpose" is not its individual life, so much as its power to support and carry to the next point the all important reproductive material. Modern research has gone far towards establishing Weismann's claims in this regard. It is true that the germ-plasm must sometimes change - - otherwise there could be no evolution. But the evidence that the germ- plasm responds directly to the experiences of the body has no substantial evidence in its support. I know, of course, that the whole Lamarckian school rests its argument on the assumption that the germ-plasm re- sponds to all profound changes in the soma ; but despite the very large literature that has grown up dealing with this matter, proof is still lacking. And there is abun- dant evidence to the contrary. On the other hand, there is evidence to show that the germ-plasm does sometimes change or is changed. Weismann's attempt to refer all such changes to recom- binations of internal factors in the germ-plasm it- self has not met with much success. Admitting that new combinations may be brought about in this way, as explained for paramcecium, yet it seems un- likely that the entire process of evolution could have resulted by recombining what already existed ; for it would mean, if taken at its face value, that by re- combination of the differences already present in the first living material, all of the higher animals and plants were foreordained. In some way, therefore, the germ- ,s BEREDITY AND SEX plasm musl have changed. We have then the alter- natives. Is there some internal, initial or driving im- pulse thai has led to the process of evolution? Or has the envir nenl brought about changes in the germ- plasm ? We can only reply that the assumption of an • Fio. .">. -Schematic representation of the processes occurring during the fertilization and subsequent segmentation of the ovum. (Boveri, from Bowell.) internal force puts the problem beyond the field of scientific explanation. On the other hand, there is a -ni.ill amonnl of evidence, very incomplete and in- sufficient Ml present , to show that changes in the en- vironment reach through the soma and modify the germinal material. THE EVOLUTION OF SEX 19 It would take us too far from our immediate subject to attempt to discuss this matter, but it has been nec- essary to refer to it in passing, for it lies at the founda- tion of all questions of heredity and even involves, as we shall see later, the question of heredity of sex. This brings us back once more to the provisional conclusion we reached in connection with the experi- ments on paramcecium. When the egg is fertilized by the sperm, Fig. 5, the result is essentially the same as that which takes place when two paramcecia fer- tilize each other. The sperm brings into the egg a nucleus that combines with the egg-nucleus. The new individual is formed by recombining the hereditary traits of its two parents. It is evident that fertilization accomplishes the same result as conjugation. If our conclusion for paramce- cium holds we can understand how animals and plants with eggs and sperm may better readjust themselves now to this, now to that environment, within certain limits. But we cannot conclude, as I have said, that this process can make any permanent contribution to evolu- tion. It is true that Weismann has advanced the hy- pothesis that such recombinations furnish the materials for evolution, but as I have said there is no evidence that supports or even makes plausible his contention. I bring up again this point to emphasize that while the conclusion we arrived at - - a provisional conclusion at best — may help us to understand how sexual repro- duction might be beneficial to a species in maintaining itself, it cannot be utilized to explain the progressive advances that we must believe to have taken place during evolution. 20 iii:i;ki)Itv and sex THE EARLY ISOLATION OF THE GERM-CELLS There is much evidence to show that the germ-cells appear very early in the development of the individual when they are set aside from the cells that differentiate Into the body cells. This need not mean that the germ- cell- have remained unmodified, although this is at '( ,'""""i" «*■*«*» «d origin of the germ-cells in Ascaris. THE EVOLUTION OF SEX 21 first sight the most natural interpretation. It might be said, indeed, that they are among the first cells to differentiate, but only in the sense that they specialize, as germ-cells. Fig. 7. — Origin of germ-cells in Sagitta. (From Korschelt and Heider.) In a parasitic worm, ascaris, one of the first four cells divides differently from the other three cells. As seen in Fig. 6, this cell retains at its division all of its chromatin material, while in the other three cells some of the chromatin is thrown out into the cell-plasm. The -. v "i-. ** C #, •„*.* & . 7-n v. r~ .,'- - iaJ- «; , *?- B&/ t - THf^* .* \ •v- A ' <■ • o Fig. 8. — Origin of germ-cells in Miastor. Note small black proto- plasmic area at bottom of egg into which one of the migrating segmentation nuclei moves to produce the germ-cells. (After Kahle.) 22 HEREDITY AND SEX single cell thai retains all of the chromatin in its nucleus gives rise to the germ-cells. In a marine worm-like form, sagitta, two cells can easily be distinguished from the other cells in the wall of the digestive tract (Fig. 7). They leave their first posi- tion and move into the interior of the body, where they produce the ovary and testes. LopMOMM Lepldosleua ■ \Periph. Crtf. \Vll.End. Flo. 9. Origin of germ-cells in certain vertebrates, viz. turtle, frog, gar-pike and bow-fin. The germ-cells as darker cells are seen migrating from the digestive tract (endoderm). (After Allen.) In several of the insects it has been shown that at a very early stage in the segmentation, one, or a few cells at most, lying at one end of the egg develop almost in- dependently of the rest of the embryo (Fig. 8). Later they are drawn into the interior, and take up their final location, where they give rise to the germ-cells. liven in the vertebrates, where, according to the THE EVOLUTION OF SEX 23 earlier accounts, the germ-cells were described as appear- ing late in embryonic development, it has been shown that the germ-cells can be detected at a very early stage in the walls of the digestive tract (Fig. 9). Thence they migrate to their definitive position, and give rise to the cells from which the eggs or the sperm arise. The germ-cells are in fact often the earliest cells to specialize in the sense that they are set aside from the other cells that produce the soma or body of the in- dividual. THE APPEARANCE OF THE ACCESSORY ORGANS OF REPRODUCTION As animals became larger the problem of setting free the germ-cells was a matter of great importance. Sys- tems of outlets arose — the organism became piped, as it were. In the lower animals the germ-cells are brought to the surface and set free directly, and fertilization is a question of the chance meeting of sperm and egg ; for there is practically no evidence to show that the sperm is attracted to the egg and much evidence that it is not. Later, the copulatory organs were evolved in all the higher groups of animals by means of which the sperm of the male is transferred directly to the female. This makes more certain the fertilization of the egg. In the mollusks, in the insects and crustaceans, and in the vertebrates the organs of copulation serve to hold the individuals together during the act of mating, and at the same time serve to transfer the semen of the male to the oviduct, or to special receptacles of the female. Highly elaborated systems of organs and special instincts, no less elaborate, serve to make the 24 HEREDITY AND SEX union possible. In some types mating must occur for each output of eggs, but in other cases the sperm is Btored up in special receptacles connected with the ducts of the female. From these receptacles a few sperm at ;, time may be set tree to fertilize each egg as it passes tin- opening of the receptaculum. In the queen bee enough sperm is stored up to last the queen for five or six years and enough to fertilize a million eggs. /**•- 3 ¥ ^f^y^^^ I'm. 10. — Squid : Two upper right-hand figures illustrate two methods of copulation. Lower right-hand figure dissected to show spermatophore placed in mantle cavity of female. Left-hand figure (below), spermatophore pocket behind mouth of male; upper figure, section of same. (After Drew.) There are a few cases where the transfer from the male to the female is brought about in a different way. The most striking cases are those of the squids and octopi, and of the spiders. In the squid, the male and female interlock arms Fig. 10). The male takes the packets of sperm (that are emitted at this time from the sperm-duct) by means of a special arm, and transfers the packets either to a THE EVOLUTION OF SEX 25 special receptacle within the circle of arms of the female, or plants them within the mantle chamber itself of the female. Each packet of spermatozoa is contained in a long tube. On coming in contact with sea water the tube everts at one end, and allows the sperm to escape. A .• • . as;-.. \- '7-*^ » ■ Fig. 11. — Octopus, male showing hectocotyl arm (ha). Cop- ulation (below), small male, A; large female, B. After separation the female deposits her strings of eggs, which are fertilized by the sperm escaping from the spermatophores. In octopus and its allies, one arm, that is used to transfer the spermatophores, is specially modified at the breeding season (Fig. 11). 26 HEREDITY AND SEX This arm is inserted by the male, as shown in the figure, within the mantle chamber of the female. In some species, Argonauta argo for instance (Fig. 12), the arm Fig. 12. — Argonauta Bhowing developing (A) and developed (B) hectocotyl arm, which, after being charged with spermatophores, is left in mantle <>f female. is broken off, and remains attached by its suckers in- side the mantle of the female. The eggs are later fer- tilized by sperm set free from this " hectocotylized " arm. THE SECONDARY SEXUAL CHARACTERS In the most highly evolved stages in the evolution of sex a new kind of character makes its appearance. This is the secondary sexual character. In most cases such characters are more elaborate in the male, but occasionally in the female. They are .the most aston- ishing thing that nature has done: brilliant colors, plumes, combs, wattles, and spurs, scent glands (pleas- :nii and unpleasant); red spots, yellow spots, green spots, topknots and tails, horns, lanterns for the dark, songs, howlings, dances and tourneys -- a medley of odds and ends. The mosl familiar examples of these characters are found in vertebrates and insects, while in lower forms THE EVOLUTION OF SEX 27 they are rare or absent altogether. In mammals the horns of the male stag are excellent examples of second- ary sexual characters. The male sea cow is much greater in size than the female, and possesses long tusks. The mane of the lion is absent in the lioness. Fig. 13. — Great bird of Paradise, male and female. (After Elliot.) In birds there are many cases in which the sexes differ in color (Figs. 13 and 14). The male is often more brilliantly colored than the female and in other cases the nuptial plumage of the male is quite different from the plumage of the female. For example, the black and yellow colors of the male bobolink are in striking contrast with the brown-streaked female (Fig. 15). The male scarlet tanager has a fiery red plumage with black wings, while the female is olive green. The male 28 HEREDITY AM) SEX of tne mallard duck has a green head and a reddish breasl Fig. L6), while the female is streaked with brown. h, insects the males of some species of beetles have hornsOD the head that are lacking in the female (Fig. 17 . The males of many species of butterflies are col- ored differently from the females. FlO. 11. — White-booted humming bird, two males and (me female. (After Gould.) The phosphorescent organ of our common firefly, Phot in us pyralis, is a beautiful illustration of a second- ary sexual character. On the under surface of the male there arc two bands and of the female there is a single band that can be illuminated (Fig. 18). At night the males leave their concealment and fly about. A little later the females ascend to the tops of blades of grass THE EVOLUTION OF SEX 29 f Fig. 15. — Male and female bobolink. (From " Bird Lore.") Fig. 16. — Male and female mallard duck. (From " Bird Lore.") :*o HEREDITY AM) SEX and remain there without glowing. A male passes by and flashes his light ; the female flashes back. In- Btantly litudy the body cells of the female protenor, we find fourteen chromosomes (Fig. 22, A). Twelve of these are the ordinary chromosomes, and two, larger than the rest, are the sex chromosomes. At the synap- sis stage all of the chromosomes unite in pairs, including the two aex chromosomes. When the process is finished, there are seven double chromosomes (Fig. 22, B). THE MECHANISM OF SEX-DETERMINATION 43 When the egg sends off its two polar bodies, the chro- mosomes divide or separate. At the first division seven chromosomes pass out (C), and seven remain in the egg. At the next division the seven chromosomes in the egg divide again, seven pass out and seven remain Pro/en or 9 X ••• • • A •3. in the egg (D). Of these seven, one chromosome, recognizable by its large size, is the sex chromosome. All the eggs are alike (E). There is only one kind of egg, but there are two kinds of sperm. Any egg that is fertilized by a sperm carrying six chromosomes pro- duces an individual with thirteen chromosomes. This individual is a male. Any egg that is fertilized by a sperm carrying seven I! HEREDITY AM) SEX chromosomes produces an individual with fourteen chromosomes. This individual is a female. In another species of insect, Lygaeus bicrucis, the male differs from the female, not in having a different LyOa.etis 3 ••••• KS V 3 Fig. 23. number of chromosomes as in protenor, but by the occurrence of a pair of different-sized chromosomes. The body cells of the male have twelve ordinary chromosomes and two sex chromosomes -- one larger, .V. than the other, Y (Fig. 23, A). After synapsis there are six double chromosomes* and the two sex chromosomes, called X and Y (Fig. 23, I)). THE MECHANISM OF SEX-DETERMINATION 45 At the first spermatocyte division all the chromosomes divide (C). The two resulting cells have eight chro- mosomes, including X and Y. At the second division (D) the double chromosomes again divide, but X and Y do not divide. They approach and touch each other, and are carried into the spindle, where they separate from each other when the other ordinary chromosomes Lygaetts' ••• ••oV A divide. Consequently there are formed two kinds of spermatozoa — one containing X and the other Y (Fig. 23, E). In the body cells and early germ-tract of the female of lygseus (Fig. 24, A), there are twelve ordinary chromosomes and two sex chromosomes, X and X. After reduction there are seven double chromosomes, the two X's having united when the other chromosomes n; HEREDITY AND SEX united \H). Two divisions lake place (C, D), when the two polar bodies are formed, leaving seven chromosomes in the egg E). Each egg contains as a result only one .V chromosome. Any egg of lygseus fertilized by a sperm carrying an A' chromosome produces a female that contains two Fig. 25. - or XX. Any egg fertilized by a sperm containing chromosome produces a male that contains one a X and one )', or A')'. Another insect, Oncopeltus fasciatus, represents a thud type m which the chromosome groups in the male and in the female are numerically alike and alike as to visible size relations. THE MECHANISM OF SEX-DETERMINATION 47 In the body cells of the male there are sixteen chro- mosomes (Fig. 25, A). After reduction there are nine chromosomes - - seven in a ring and two in the middle (B). The seven are the fused pairs or double chro- mosomes ; the two in the middle are the sex chromo- somes that have not fused. Onevpettuj 9 B The evidence for this interpretation is circumstan- tial but sufficient. At the first reduction division all nine chromosomes divide (C). Just before the second division the two central chromosomes come together and remain in contact (DDr). All the double chromosomes then divide, while the two sex chromosomes simply sepa- rate from each other, so that there are eight chromo- somes at each pole (DE). 18 HEREDITY AND SEX h, this case all of the spermatozoa {EE') contain eighl chromosomes. There is no visible difference between them. Nevertheless, there is reason for be- lieving that here also there are two kinds of sperm. The principal reason is that there are all connecting stages between forms in which there is an unequal pair, ^scares S hi #n 4»- Fig. 27. &1 as in lygffius, and forms with an equal pair, as in oncopel- tus. Another reason is that the two sex chromosomes behave during the synapsis stages as do the XFchromo- somes in related species. Moreover, the experimental evidence, of which I shall speak later, leads us to con- clude that the determination of sex is not due only to THE MECHANISM OF SEX-DETERMINATION 49 a difference in size of X and Y. The sex chromosomes must carry a host of factors other than those that de- termine sex. Consequently it is not surprising that in many species the sex chromosomes appear equal or nearly equal in size. It is a fortunate circumstance for us that in some species there is a difference in size or ^Asca.ris , two with 6 chromosomes; 17, spermatids; 18, mature sperm ; 19, living sperm. (After Mulsow.) In these insects a study of the chromosomes shows thai the male has one less chromosome than the female. At the first maturation division in the male (Fig. 30), all the chromosome- divide except one, the X chromo- Bome, and this passes to one cell only. This cell is also larger than the sister cell. The small cell lacking the X degenerates, and does not produce spermato- THE MECHANISM OF SEX-DETERMINATION 53 zoa. The large cell divides again, all of the chromo- somes dividing. Two functional spermatozoa are produced, each carrying one sex chromosome. These spermatozoa correspond to the female-producing sper- matozoa of other insects. In the sexual female there is an even number of chro- JL*r • Jk • * s. •.' _ v. vs. •*ft Fig. 29a. — 20 and 21, oogonia (equatorial plate); 22, growth period; 23, before fertilization; 24-25, entrance of sperm; 26-31, prophases of first division ; 32-33, formation of first polar body ; 34-36, extrusion of same and formation of second polar body ; 37, two pronuclei ; 38-41, union of pronuclei ; 42-45, cleavage. (After Mulsow.) mosomes - - one more than in the male. They unite in pairs. When the two polar bodies of the sexual egg are formed, all the chromosomes divide twice, so that each egg is left with one sex chromosome. It is now evident why only females are produced after fertilization. The female-producing sperm alone is functional. 54 HEREDITY AND SEX Fio. 30. — Diagram of chromosomes in Phylloxera carycecaulis. Top line, somatic cell of female with 6 chromosomes and somatic cell of male with 5 chromosomes. Second line, stages in first spermatocyte division producing a rudimentary cell (below) with two chromosomes. Third line, second spermatocyte division into two equal cells. Fourth line, sexual egg (3 chromosomes) and two polar bodies; and two functional, female- producing sperm with three chromosomes each. THE MECHANISM OF SEX-DETERMIXATION 55 THE EXPERIMENTAL EVIDENCE The experimental evidence, indicating that there is an internal mechanism for sex determination, is derived from two sources - - from experimental embryology, and from a study of the heredity of sex-linked characters. The evidence from embryology shows that the chro- mosomes are the bearers of materials essential for the production of characters. The evidence from hered- ity shows that certain characters follow the sex chromosomes. It has long been taught that the hereditary factors are carried by the nucleus. The evidence for this was found in fertilization. When the spermatozoon enters the egg, it carries in, as a rule, only the head of the sper- matozoon, which consists almost entirely of the nucleus of the original cell from which it comes. Since the male transmits his characters equally with the female, it follows that the nucleus is the source of this inheritance. The argument has not been regarded as entirely conclusive, because the sperm may also bring in some of the protoplasm of the original cell — at least that part lying immediately around the nucleus. In addition a small body lying at the base of the sperm head seems also to be brought in by the male, and according to some observers it becomes the center about which the entire division system or karyokinetic spindle develops. The most convincing evidence that the chromosomes are the most important elements in heredity is found in some experimental work, especially that of Boveri, Baltzer, and Herbst. Under certain circumstances in 56 HEREDITY AND SEX the sea-urchin two spermatozoa may enter a single egg. They both unite with the egg nucleus (Fig. 31). Each brings in 18 chromosomes. The egg contributes is chromosomes. There are in all 54, instead of 36 chromosomes, as in normal fertilization. Fig. 31. — Dispermy and its effects in egg of sea urchin. (After Boveri.) Around these chromosomes a double system of threads develops with four poles. The chromosomes become unequally distributed on the four spindles that develop. Each chromosome then divides, and half of each goes to the nearest pole. To some of the poles many chromosomes may pass, to other poles fewer. THE MECHANISM OF SEX-DETERMINATION 57 In order to simplify the case let us imagine that each sperm has only four chromosomes and the egg nucleus only four. Let us represent these by the letters as shown in Fig. 32. Any one of the four cells that is (it* ©© \ ©© ©© , [ © J i ©@ ® \ i © © \ © © J \®©@ @@ / Fig. 32. — Diagram illustrating the irregular distribution of the chro- mosomes in dispermic eggs in an imaginary case with only four kinds of chromosomes, a, b, c, d. There are here three sets of each of these in each egg. The stippled cells are those that fail to receive one of each kind of chromosome. (After Boveri.) produced at the first division of these dispermic eggs may contain a full complement of the chromosomes, or only some of them. The possibilities for four chromosomes are shown in the diagram. Any cell that does not contain at least these four chromosomes is shaded. One case is present in which all the four ;,s HEREDITY AND SEX cells contain a complete assortment. If normal devel- opment depends on an embryo containing in every cell at least one of each kind of chromosome, then in our simple case only one group of four cells has this possibility. Boveri found that such dispermic eggs produce normal embryos very rarely. He calculated what the chance would be when three times 18 chromosomes are involved. The chance for normal development is probably not once in 10,000 times. He isolated many dispermic eggs and found that only one in 1,500 of the tetrad type developed normally. Boveri went still further in his analysis of the prob- lem. It had been shown for normal eggs that if at the two-celled stage the cells are separated, each forms a perfect embryo. This is also true for each of the first four cells of the normal egg. Boveri separated the four cells of dispermic eggs and found that the quadrants not infrequently developed normally. This is what we should anticipate if those cells can develop that contain one of each kind of chromosome. The evidence furnishes strong support of the view that the chromosomes are different from each other, and that one of each kind is necessary if development is to take place normally. The evidence that Baltzer has brought forward is also derived from a study of sea-urchin eggs. It is possible to fertilize the eggs of one species with sperm of another species. The hybridizing is greatly helped by the addition of a little alkali to the sea water. Baltzer made combinations between four species of THE MECHANISM OF SEX-DETERMINATION 59 sea-urchins. We may take one cross as typical. When eggs of strongylocentrotus are fertilized with sperm of sphaerechinus, it is found at the first division of the egg that, while some of the chromosomes divide and pass normally to the two poles, other chromosomes remain in place, or become scattered irregularly between the two poles, as shown in Fig. 33. When the division It ^'""/iV- » v'/^' Fig. 33. — 1 and la, chromosomes in the normal first cleavage spindle of Sphaerechinus ; 2, equatorial plates of two-cell stage of same ; 3-3a, hybrid, Sphaerechinus by Strongylocentrotus, spindle at two-cell stage ; 4-4a, same equatorial plates; 5-5o, hybrid, Strong, by Sphaer., cleavage spindle in telo- phase ; 6, next stage of last ; 7, same, two-cell stage ; 8, same, later ; 9, same, four-cell stage ; 10, same, equatorial plate in two-cell stage (12 chromosomes) ; 11, same, from later stage, 24 chromosomes. (After Baltzer.) is completed, some of these chromosomes are found outside of the two main nuclei. They often appear as irregular granules, and show signs of degeneration. They are still present as definite masses after the next division, but seem to take no further part in the de- velopment. Baltzer has attempted to count the number of chro- mosomes in the nuclei of these hybrid embryos. The 60 HEREDITY AND SKX nun 1 1 xt is found to be about twenty-one. The maternal egg nucleus contains eighteen chromosomes. It appears that only three of the paternal chromosomes have succeeded in getting into the regular cycle -- fifteen of them have degenerated. Baltzer thinks that the egg acts injuriously in this case on the chromosomes of foreign origin, especially on the fifteen that degenerate, so that they are elim- inated from the normal process. The embryos that develop from these eggs are often abnormal. A few develop as far as the pluteus stage, when a skeleton appears that is very characteristic for each species of sea-urchin. The plutei of these hybrids are entirely maternal. This means that they are exactly like the plutei of the species to which the mother belongs. The conclusion is obvious. The sperm of sphserechi- nus has started the process of development, but has produced no other effect, or has at most only slightly affected the character of the offspring. It is reason- able to suppose that this is because of the elimination of the paternal chromosomes, although the evidence is not absolutely convincing. Let us now examine the reciprocal cross. When the eggs of sphterechinus are fertilized by the sperm of strongylocentrotus, the division of the egg and of the chromosomes is entirely normal. All the chromosomes divide and pass to the poles of the spindle. The total number (36) must, therefore, exist in each cell, although in this case they were not actually counted. The pluteus that develops has peculiarities of both maternal and paternal types. It is hybrid in structure. THE MECHANISM OF SEX-DETERMINATION 61 Both parents have contributed to its formation. It is not going far from the evidence to infer that the hybrid character is due to both sets of chromosomes being present in all of the cells. Fig. 34. — 1. The chromosomes of the egg lie in the equator of the spindle, the chromosomes of the sperm lie at one side. 2. A later stage, showing all the paternal chromosomes passing to one pole. 3 (to the right). A later stage, a condition like the last. There is also a supernumerary sperm in the egg (to left, in another section.) 4. Same condition as last. 5. Plu- teus larva that is purely maternal on one side and hybrid on the other. (After Herbst.) The evidence that Herbst has brought forward is of a somewhat different kind. It supplements Baltzer's evidence and makes more probable the view that the chromosomes are essential for the development of the characters of the individual. 62 HEREDITY AND SEX Herbst put the eggs of sphaerechinus into sea water to which a little valerianic acid had been added. This is "iif of the many methods that Loeb has discovered by which the egg may be induced to develop parthe- nogenetically, i.e. without the intervention of the sper- matozoon. Alter live minutes the eggs were removed to pure sea water and the sperm of another species, stron- gylocentrotus, was added. The sperm penetrated some of the egg3. The eggs had already begun to undergo the changes that lead to division of the cell — the sperm entered ten minutes late. The egg proceeded to divide, the sperm failed to keep pace, and fell behind. Consequently, as shown in Fig. 34, the paternal chromosomes fail to reach the poles when the nuclei arc re-formed there. The paternal chromosomes form a nucleus of their own which comes to lie in one of the two cells. In consequence one cell has a nucleus that contains only the maternal chromosomes; the other cell contains two nuclei, one maternal and the other paternal. In later development the paternal nucleus becomes incorporated with the maternal nucleus of its cell. There is produced an embryo which is maternal on one side and hybrid on the other. Herbst found in tact that half-and-half plutei were not rare under the conditions of his experiment. This evidence is almost convincing, I think, in favor of the view that the chromosomes are the es- sential hearers of the hereditary qualities. For in this case, whether the protoplasm of the embryo comes from the egg or the sperm also, the fact re- mains that the half with double nuclei is hybrid. Even if the spermatozoon brought in some proto- THE MECHANISM OF SEX-DETERMINATION 63 plasm, there is no reason to suppose that it would be distributed in the same way as are the paternal chromosomes. EVIDENCE FROM SEX-LINKED INHERITANCE The experimental evidence based on sex-linked in- heritance may be illustrated by the following examples. The eyes of the wild fruit-fly, Drosophila ampe- lophila, are red. In my cultures a male appeared that had white eyes. He was mated to a red-eyed female. The offspring were all red-eyed — both males and females (Fig. 35). These were inbred and produced in the next generation red-eyed females, red-eyed males, and white-eyed males (Fig. 35). There were no white- eyed females.. The white-eyed grandfather had trans- mitted white eyes to half of his grandsons but to none of his granddaughters. Equally important are the numerical proportions in which the colors appear in the grandchildren. There are as many-females as the two classes of males taken together ; half of the males have red eyes and half have white eyes. The proportions are therefore 50 % red females, 25 % red males, 25 % white males. Only white-eyed males had appeared at this time. / It may seem that the eye color is confined to the male * sex. Hence the origin of the term sex-limited inheri- v tance for cases like this. But white-eyed females may be produced easily. If certain of the red-eyed grand-/ daughters are bred to these white-eyed males, both white-eyed females and males, and red-eyed females and males, appear (Fig. 37). The white eye is there- fore not sex-limited but sex-linked. I.I HEREDITY AND SEX With these white-eyed females it is possible to make the reciprocal cross (Fig. 36). A white-eyed female was mated fee a red-eyed male. All of the daughters had red eves and all the sons had white eyes. These were then inbred and gave red-eyed males and females, XX ■w XX XIX In;. :i."). — Sex-linked inheritance of white and red eyes in Drosophila. Parents, white-eyed $ and red-eyed 9 ; ^li red-eyed $ and 9 ! F2 red- eyed 9, red-eyed $ and white-eyed $ . To right of flies the history of the 91 i chromosomes -V.V is shown. The black X carries the factor for red the "pen A" the factor for white ej es ; the circle stands for no X. and white-eved males and females in equal numbers (Fig. 36). The heredity of this eye color takes place with the utmosl regularity, and the results show that in some way the mechanism that is involved is closely bound up with the mechanism that produces sex. THE MECHANISM OF SEX-DETERMINATION 65 Other combinations give results that are predictable from those just cited. For instance, if the Fi red-eyed female from either of the preceding crosses is mated to a white-eyed male, she produces red-eyed males and females, and white-eyed males and females, as shown in ct X XX XI If Fig. 36. — Reciprocal cross of Fig. 35. Parents, white-eyed ? and red-eyed $, (criss-cross inheritance). Fu red-eyed s>. Mated to long-winged females only Long-winged offspring were produced. When these were mated to each other, there were produced X Y XVI X Y :^XY Fig. -i~. — Upper series, back crass of Fi 9 to white $. Lower series back cross of l-\ red-#yed $ to white ?. long-winged females (50%), long-winged males (25%) and miniature-winged males (25%). It is possible to produce, in the way described for eye color, miniature-winged females. When such miniature-winged females are mated to long-winged males, all the daughters have long wings, and all the sons have miniature wings (Fig. 39). If THE MECHANISM OF SEX-DETERMINATION 67 these are now mated, they produce, in equal numbers, long-winged males and females and miniature-winged males and females. The same relations may again be illustrated by body color. XX XXXIX Fig. 38. — Sex-linked inheritance of short (" miniature") and long wings in Drosophila. Parents, short-winged $ , long-winged 9 . Fi long-winged $ and 9 . F2 long-winged 9 and $ and short-winged $ . Sex chromo- somes to right. Open X carries short wings. A male appeared with yellow wings and body. Mated to wild gray females he produced gray males and females. These mated to each other gave gray females (50%), gray males (25%), and yellow males (25%). As before, yellow females were made up. Mated to gray males they gave gray females and yellow males. 68 HEREDITY AND SEX These inbred gave gray males and females and yellow males and females, in equal numbers. These cases serve bo illustrate the regularity of this type of inheritance and its relation to sex. In the fruit fly we have found as many as twenty-five sex-linked X XI X Fig. 39. - Reciprocal cross of Fig. 38. Parents, long-winged $ and shortrwinged 9. Fi long-winged 9, short-winged $. F2 long-winged 9 and J , short-winged 9 and $. Sex chromosomes as in last. factors. There are other kinds of inheritance found in these flies, and at another time I shall speak of some of these : hut the group of sex-linked factors is of special importance because through them we get an insight into the heredity of sex. In the next chapter, when we take up in detail Men- dcliaii heredity, I shall try to go further into the ex- THE MECHANISM OF SEX-DETERMINATION 69 planation of these facts. For the present it will suffice to point out that the cases just described, and all like them, may be accounted for by means of a very simple hypothesis. We have traced the history of the X chromosomes. If the factors that produce white eyes, short (miniature) wings, and yellow body color are carried by the X chromosomes, we can account for these results that seem at first sight so extraordinary. The history of the sex chromosomes is accurately known. Their distribution in the two sexes is not a matter of conjecture but a fact. Our hypothesis rests therefore on a stable foundation. At the risk of confusion I feel bound to present here another type of sex-linked inheritance. In principle it is like the last, but the actual mechanism, as we shall see, is somewhat different. Again I shall make use of an illustration. If a black Langshan hen is mated to a barred Plymouth Rock cock, all the offspring will be barred (Fig. 40). If these are inbred, there are pro- duced barred females and males, and black females. The numerical proportion is 50 per cent barred males, 25 per cent barred females, and 25 per cent black females. The black hen has transmitted her character to half of her granddaughters and to none of her grandsons. The resemblance to the case of the red-eyed, white- eyed flies is obvious, but here black appears as a sex- linked character in the females. The converse cross is also suggestive. When a barred hen is mated to a black cock, all the daughters are black and all the sons are barred (Fig. 41). When these are inbred, there are produced black males and females and barred males and females in equal num- 7() HEREDITY AND SEX bere. Again, the resemblance of the reciprocal cross to one of the combinations for eye color is apparent. In fact, this case can be explained on the same prin- ciple as that used for the flies, except that in birds it is £11 '■^zfc-i Parents T' Fig. 40 c-linked inheritance in fowls. Upper line black Langshan ben and barred Plymouth Rock cock. Second line, b\, barred cock and hen. Third line, /•'■. three barred (rock, hen, cock) and one black (hen). (Cute fro,,, " Reliable Poultry Journal.") Fx and F2 for color only. THE MECHANISM OF SEX-DETERMINATION 71 the female that produces two kinds of eggs ; she is heterozygous for a sex factor while the male produces only one kind of spermatozoon. ; ,sk Fig. 41. — Reciprocal cross of Fig. 40. Upper line, black cock and barred hen. Second line, fr\, barred cock and black hen. Third line, Fi, barred hen and cock, black cock and hen. (Cuts from " Reliable Poultry Journal.") F\ and F2 correct for color pattern only. 72 HEREDITY AND SEX We lack here the certain evidence from cytology that we have in the case of the insects. Indeed, there is some cytological evidence to show that the male bird is heterozygous for the sex chromosome. But the evidence does not seem to me well established ; while the experimental evidence is definite and has been independently obtained by Bateson, Pearl, Sturtevant, Davenport, Goodale and myself . However this may be, the results show very clearly that here also sex is con- nected with an internal mechanism that is concerned with other characters also. This is the mechanism of Mcndelian heredity. Whether the chromosomes suffice or do not suffice to explain Mendelian heredity, the fact remains that sex follows the same route taken by characters that are recognized as Mendelian. To sum up : The facts that we have considered furnish, I believe, demonstrative evidence in favor of the view that sex is regulated by an internal mech- anism. The mechanism appears, moreover, to be the same mechanism that regulates the distribution of cer- tain characters that follow Mendel's law of inheritance. CHAPTER III The Mendelian Principles of Heredity and Their Bearing on Sex The modern study of heredity dates from the year 1865, when Gregor Mendel made his famous discoveries in the garden of the monastery of Brunn. For 35 years his paper, embodying the splendid results of his work, remained unnoticed. It suffered the fate that other fundamental discoveries have sometimes met. In the present case there was no opposition to the principles involved in Mendel's discovery, for Darwin's great work on "Animals and Plants" (1868), that dealt largely with problems of heredity, was widely read and appreciated. True, Mendel's paper was printed in the journal of a little known society — the Natural History Society of Brunn, - - but we have documentary evidence that his results were known to one at least of the leading botanists of the time. It was during these years that the great battle for evolution was being fought. Darwin's famous book on "The Origin of Species" (1859) overshadowed all else. Two systems were in deadly conflict — the long-ac- cepted doctrine of special creation had been challenged. To substitute for that doctrine the theory of evolution seemed to many men of science, and to the world at large, like a revolution in human thought. It was in fact a great revolution. The problems that bore on the 73 j I BEREDITY AND SEX questioD of how the higher animals and plants, and man himself, arose from the lower forms seemed the chief goal of biolo^cal work and thought. The out- come was i" establish the theory of evolution. The circumstantial evidence that was gathered seemed so fully in accord with the theory of evolution that the theory became widely accepted. The acute stage was passed, and biologists found themselves in a position to examine with less haste and heat many other phe- nomena of the living world equally as important as evolution. It gradually became clear, when the clouds of con- troversy had passed, that what I have ventured to call the "circumstantial evidence" on which the theory of evolution so largely rested, would not suffice as a direct proof of evolution. Investigation began to turn once more to that field of observation where Darwin had found his inspiration. The causes of variations and the modes of inheritance of these variations, the very foundations of the theory of evolution, were again studied in the same spirit in which Darwin himself had Mudied them. The return to Darwin's method rather than to Darwin's opinions marks the beginning of the new era. In 1900 three botanists were studying the problem of heredity. Each obtained evidence of the sort -Mendel had found. Happily, Mendel's paper was remembered. The significance of his discovery now became apparent. De Vries, Correns, and Tschermak brought forward their evidence in the same year (1900). Which of the three first found Mendel cannot be stated, and is of less importance than the fact that they ap- TF TENDELIAN PRINCIPLES OF HEREDITY 75 j he significance of his work, and realized tl ound the key to the discoveries that they tc .xiade. From this time on the recognition of Mendel's discovery as of fundamental importance was assured. Bateson's translation of his paper made Mendel's work accessible to English biologists, and Bateson's own studies showed that Mendel's principles apply to animals as well as to plants. THE HEREDITY OF ONE PAIR OF CHARACTERS Mendel's discovery is sometimes spoken of as Men- del's Principles of Heredity and sometimes as Mendel's Law. The former phrase gives a better idea perhaps of what Mendel really accomplished, for it is not a little difficult to put his conclusions in the form of a law. Stated concisely his main discovery is this : - - in the germ-cells of hybrids there is a free separation of the elements derived from the two parents without regard to which parent supplied them. An example will make this more obvious. Mendel crossed an edible pea belonging to a race with yellow seeds to a pea belonging to a race with green seeds (Fig. 42). The offspring or first filial generation (Fi) had seeds all of which were yellow. When the plants that bore these seeds were self-fertilized, there were obtained in the next generation, F2, both yellow and green peas in the proportion of 3 yellows to 1 green (Fig. 42). This is the well-known Mendelian ratio of 3:1. The clue to the meaning of this ratio was found when the plants of the second generation (F2) were selfbred. The green peas bred true ; but the yellows were of two 76 HEREDITY AND SEX kinds some produced yellow and green seeds again in the ratio of 3 : 1 ; others produced only yellow peas. Now. it the yellows that bred true were counted, it was found thai the Dumber was but one-third of all the yellows. I "• 12 Illustrating Mendel's cross of yellow (lighter color) and green (dark color) peas. THE MENDELIAN PRINCIPLES OF HEREDITY 77 In other words, it was shown that the ratio of 3 yel- lows to 1 green was made up of 1 pure yellow, 2 hy- brid yellows, 1 pure green. This gave a clue to the principles that lay behind the observed results. Mendel's chief claim to fame is found in the discovery of a simple principle by means of which the entire series of events could be explained. He pointed out that if the original parent with yellow (Pi) carried something in the germ that made the seed yellow, and the original parent with green seeds (Pi) carried some- thing that made the seed green, the hybrid should con- tain both things. If both being present one domi- nates the other or gives color to the pea, all the peas in the hybrid generation will be of one color - - yellow in this case. Mendel assumed that in the germ-cells of these hybrids the two factors that make yellow and green separate, so that half of the germ-cells contain yellow-producing material, and half contain green- producing material. This is Mendel's principle of separation or segregation. It is supposed tv o.ccur both in the male germ-cells of the hybrid flower, i.e. in the anthers, and also in the ovules. If self-fertili- zation occurs in such a plant, the following combina- tions are possible : A yellow-bearing pollen grain may fertilize a "yellow" ovule or it may fertilize a " green" ovule. The chances are equal. If the former occurs, a pure yellow-seeded plant will result ; if the latter a hybrid yellow-seeded plant. The possible combina- tions for the green-producing pollen are as follows : A "green" pollen grain may fertilize a "yellow" ovule, and produce a hybrid, yellow-seeded plant, or it may fertilize a "green" ovule, and produce a green-seeded BEREDITY AM) SEX plant. If those meetings are random, the general or average outcome will be: 1 pure yellow, 2 hybrid yellows, and 1 pure green. h is now apparent why the pure yellows will always breed true, why the yellow-greens will split again into yellows and greens (or 1:2:1), and why the pure greens breed true. By this extremely simple assump- tion the entire outcome finds a rational explanation. P\H£NTS t 8 % • 9 1 i' .. \'i. — " Checker " diagram to show segregation and recombination of factors. In upper line, a black bearing gamete is supposed to unite with a white bearing gamete to give the zygotes shown in Fi, each of which is heterozygous for black-white here represented as allelomorphs, etc. The same scheme may be represented by means of the above "checker" diagram (Fig. 43). Black crossed to white gives hybrid black, Fh whose germ-cells are black or white after segregation. The possible com- bination of these on random meeting at the time of fertilization is shown by the arrows in Fx and the results are shown in the line marked F2. There will be one pure black, to two black-and- whites, to one pure white. THE MENDELIAN PRINCIPLES OF HEREDITY 79 The first and the last will breed true, if self-fertilized, because they are pure races, while the black-and-whites will give once again, if inbred, the proportions 1:2: 1. A better illustration of Mendel's principles is shown in the case of the white and red Mirabilis jalapa de- scribed by Correns. This case is illustrated in Fig. 44, PARENTS Fig. 44. — Cross between white and red races of Mirabilis Jalapa, giving a pink hybrid in F\ which when inbred gives, in F-i, 1 white, 2 pink, 1 red. in which the red flower is represented in black and the pink is in gray. The hybrid, Fi, out of white by red, has pink flowers, i.e. it is intermediate in color. When these pink flowers are self-fertilized they produce 1 white, 2 pink, and 1 red-flowered plant again. The history of the germ-cells is shown in Fig. 45. The germ- SO HEREDITY AND SEX eel] of the /•', pink flower segregates into "white" and "red," which by chance combination give the white pink, and red flowers of Fi. The white and red flowers are pure ; the pink heterozygous, i.e. hybrid or mixed. In this case ncit her red nor white dominates, so that the hybrid can be distinguished from both its parents. -O O o o o Fig. 45.— Illustrating history of gametes in cross shown in Fig. 44. A white and a red bearing gamete unite to form the pink zygote in F\, whose gametes, l).v segregation, are red and white, which by random combinations txiv«' the Fi zygotes, etc. Mendel tested his hypotheses in numerous ways, that I need not now discuss, and found in every case that the results coincided with expectation. THE HEREDITY OF A SEX-LINKED CHARACTER We are now in a position to see how Mendel's funda- THE MENDELIAN PRINCIPLES OF HEREDITY 81 mental principle of segregation applies to a certain class of characters that in the last chapter I called " sex- linked" characters. Diagram 35 (page 64) will recall the mode of trans- mission of one of these characters, viz. white eyes. Let us suppose that the determiner for white eyes is carried by the sex chromosome. The white-eyed male has one sex chromosome of this kind. This sex chromosome passes into the female-producing spermato- zoon. Such a spermatozoon fertilizing an egg of the red- eyed fly gives a female with two sex chromosomes - one capable of producing red, one capable of producing white. The presence of one red-producing chromosome suffices to produce a red-eyed individual. When the Fx female produces her eggs, the two sex chromosomes separate at one of the two maturation divisions. Half of the eggs on an average will contain the "white" sex chromosome, half the "red." There are, then, two classes of eggs. When the Fi male produces his sperm, there are also two classes of sperm — one with the "red" sex chromosome (the female-producing sperm), and one without a sex chromosome (the male-producing sperm) . Chance meeting between eggs and sperm will give the classes of individuals that appear in the second filial generation (F2) . It will be observed that the Mendelian ratio of 3 red to 1 white appears here also. Segregation gives this result. The explanation that has just been given rests on the assumption that the mechanism that brings about vj IIi;i!i:i)lTV AND SEX the distribution of the red- and the white-producing factors is the same mechanism that is involved in sex determination. On this assumption we can readily understand that any character that is dependent on the sex chromosomes for its realization will show sex-linked inheritance. The reciprocal cross (Fig. 36) is equally instructive. If a white-eyed female is mated to a red-eyed male, all the daughters are red-eyed like the father, and all i he sons are white-eyed like the mother. When these, Fh flies are bred to each other there are produced red- eyed females (25%), white-eyed females (25%), red- eyed males (25%), and white-eyed males (25%). The explanation (Fig. 36; page 65) is in principle the same as for the other cross. If, for instance, we assume that the two X chromosomes in the white-eyed female carry the factors for white, all the eggs will carry one white-producing X. The red-eyed male will contain one X chromosome which is red-producing and passes into the female-producing sperm. The other sperm will not contain any sex chromosome, and hence lacks the factors for these eye colors. When the female-producing sperm, that carries the factor for red, fertilizes a "white" egg, the egg will give rise to a female with red eyes, because of the presence of one red-producing chromosome. When the male-produc- ing sperm fertilizes any egg, a white-eyed son will be produced, because the single sex chromosome which he gets from his mother is white-producing. The production of four classes of individuals in the second generation works out on the same scheme, as shown in the diagram. The inheritance of white and THE MENDELIAN PRINCIPLES OF HEREDITY 83 red eyes in these cases is typical of all sex-linked in- heritance. In man, for instance, color blindness, so common in males and rare in females, follows the same rules. It appears that hemophilia in man and night-blindness are also examples of sex-linked in- heritance. These cases, as already stated, were formerly included under the term "sex-limited inheritance," that implies that a character is limited to one sex, but we now know that characters such as these may be trans- ferred to the females, hence the term is misleading. Their chief peculiarity is that in transmission they ap-\ pear as though linked to the factor for sex contained in ! the sex chromosome, hence I prefer to speak of them as sex-linked characters. If our explanation is well founded, each sex-linked .character is represented by some substance — some material particle that we call a factor in the sex chromosome. There may be hundreds of such materials present that are essential for the development of sex- linked characters in the organism. The sex chromosomes must contain, therefore, a large amount of material that has nothing whatever to do with sex determination ; for the characters in question are not limited to any particular sex, although in certain combinations they may appear in one sex and not in the other. What then, have the sex chromosomes to do with sex ? The answer is that sex, like any other character, is due to some factor or determiner contained in these chro- mosomes. It is a differential factor of such a kind that when present in duplex, as when both sex chromo- somes are present, it turns the scale so that a female si HEREDITY AND SEX is produced --when present in simplex, the result is to produce a male. . In other words, it is not the sex chromosomes as a whole that determine sex, but only a part of these chro- mosomes. Hence we can understand how sex is deter- mined when an unequal pair of chromosomes is pres- ent, as in lygaeus. The smaller chromosome lacks the sex differential, and probably a certain number of other materials, so that sex-linked inheritance is pos- sible here also. Moreover, in a type like oncopeltus, where the two sex chromosomes are alike in size, we infer that they too differ by the sex differential. If all the other factors are present, as their size suggests, sex-linked inheritance of the same kind would not be expected, but the size difference observable by the microscope is obviously too gross to make any such inference certain. We have come to see that it was a fortunate coincidence only that made possible the dis- covery of sex determination through the sex chromo- somes, because the absence of the sex factor alone in the Y chromosome might have left that chromosome in the male so nearly the same size as the X in the female that their relation to sex might never have been suspected. When, however, one of the sex chromosomes began to lose other materials, it' became possible to identify it and discover that sex is dependent upon its distribution. THE HEREDITY OF TWO PAIRS OF CHARACTERS Mendel observed that his principles of heredity apply not only to pairs of characters taken singly, but to cases where two or more pairs of characters are involved. THE MENDELIAN PRINCIPLES OF HEREDITY 85 An illustration will make this clear. There are races of edible peas in which the surface is round ; other races in which the surface is wrinkled. Mendel crossed a pea that produces yellow round seeds with one that pro- duces wrinkled green seeds. The result of this cross was a plant that produced yellow round peas (Fig. 46). Both yellow and round are therefore dominant characters. When these Fi plants were self-fertilized, there were produced plants some of which bore yellow round peas, some yellow wrinkled peas, some green round peas and some green wrinkled peas. These were in the proportion of 9:3:3:1. The explanation of the result is as follows : One of the original plants produced germ-cells all of which bore determiners for yellow and for round peas, YR ; the other parent produced gametes all of which bore deter- miners for green and for wrinkled, GW (Fig. 47). Their combination may be represented : YR by GW = YRGW The germ-cells of the hybrid plant YRGW produced germ-cells (eggs and pollen) that have either Y or G, and R or W. Expressed graphically the pairs, the so-called allelomorphs, are : Y- ' R_ G * W and the only possible combinations are YR, YW, GR, GW. When pollen grains of these four kinds fall on the stigma of the same kind of hybrid plant whose ovules are also of the four kinds the following chance combinations are possible : 86 HKRKIHTY AND SKX YR YR YR YW YR (ill YR GW YW YR GR YR YW YW YW GR YW GW GR YW GR GR GR GW GW YR GW YW GW GR GW GW ceo o G OO © 3 3 1 Fig. 4(i. - - Illustrating Mendel's cross of yellow-round with green-wrinkled peas. The figures show the peas of Fi and F2 in the latter in the charac- teristie ratio of 9 : 3 : 3 : 1. THE MENDELIAN PRINCIPLES OF HEREDITY 87 O PARENTS G YR GW ;' (yrJ |jrwj O GR GR YW GR GW i v Gtf}si* YR \ GW Fig. 47. — Illustrating the history of the gametes ot the cross represented in Fig. 46. The composition of the parents YR and GW and of the Fi hybrid YRGW is given above. The four classes of ovules and of pollen are given in the middle of the figure. These by random combinations give the kinds of zygotes represented in the squares below. ss HEREDIT AM) SEX In each combination in the table the character of the plant is determined by the dominant factors, in this case yellow and round, hence: 9 YR : 3 YW :ZGR:1 GW This result works out on the assumption that there is independent assortment of the original determiners thai entered into the combination. The determiner for yellow and the determiner for round peas are assumed to act independently and to segregate from green and wrinkled that entered from the other parent. The 9:3:3:1 ratio rests on this assumption and is the actual ratio realized whenever two pairs of characters freely Mendelize. THE HEREDITY OF TWO SEX-LINKED CHARACTERS The inheritance of two sex-linked characters may be illustrated by an imaginary case in which the linkage of the factors to each other is ignored. Then the same case may be given in which the actual results obtained, involving linkage, are discussed. The factors in the fruit fly for gray color, G, and for red eye, R, are both sex-linked, i.e. contained in the X chromosome. Their allelomorphs, viz., yellow color, Y, and white eye, W, are also sex-linked. When a gray red-eyed female is mated to a yellow white-eyed male, the daughters and sons are gray-red, GR. Their origin is indicated in the following scheme : Gray-reds G R X — GR X Yellow-white $ Y W X— ... F \G R X Y W X Gray-red 5 ' ' I G R X . . . Gray-red $ THE MENDELIAN PRINCIPLES OF HEREDITY 89 In the gray-red Fx female there will be the possibility of interchange of the G and Y, and of the W and R factors, so that gametes of four kinds will be formed, namely, GRX - GWX YRX YWX. For the moment we may assume free interchange of factors ; and therefore these four classes of eggs will exist in equal numbers. In the gray-red Fi male there is but one X chromo- some that contains the factors G and R. There will be then only one kind of female-producing sperm, viz., GRX; and one kind of male-producing sperm, the latter containing no X, and therefore none of the factors in question. The chance meeting of these two classes of sperm and the four classes of eggs gives the following results : Fi eggs Fi sperm GRX GRX GWX — YRX - - YWX Males. GRX gray-red. GWX gray-white. YRX yellow-red. YWX yellow-white. Females. GRXGRX gray-red. - GRXGWX gray-red. - GRX YRX gray-red. — GRXYWX gray-red. - All the females are gray with red eyes, since these are the dominant characters. There are four classes of males. These males give a measure of the kinds of eggs produced by the females, since the male-producing sperms, having no sex chromosomes, do not affect the sex-linked characters derived through the sex chromosome of the F\ female. The expected proportion is therefore : 90 HEREDITY AM) SKX GR9 GRS GWS YR$ YW & 111 11 These results are illustrated by means of Fig. 48, in which the yellow color of the fly is indicated by stippling the body and wings, and the red eyes by black. The X chromosome is also marked and colored *2 xx n x© X' Fig. 48. — Inheritance of yellow-white ($) arid gray-red (9) of Dro- Bophila. In Fx both sexes are gray-red. In F» are produced 4 GR 9 — i <;r $ — i aw $ — i yr $ — l yw S. THE MENDELIAN PRINCIPLES OF HEREDITY 91 in the same way as the flies ; thus the two X's in the red-eyed gray female are half black (for red) and half gray ; the single X in the white-eyed yellow male is half white and half stippled. In the Fi generation the X chromosomes are first represented as they came in (second line), i.e. with their original composition. The next line gives the three large classes that result, viz., 2 GR9 - 1 GRS — 1 YW $ . But if free interchange takes place in the female, some of the eggs will have chromosomes like those in the fourth line, viz. YR and GW. Such eggs will give the classes represented in the lowest line, viz., 2 GR9— 1 GW$— 1 YRS . Thus, as already explained, there results one kind of female and four kinds of males. I said that the proportion 4:1:1:1:1 is the ideal result in the cross between the yellow-white and the gray-red flies. This ideal scheme is not realized because of a complication that comes in. The complication is due to linkage or a tendency to hang together of the characters that go in together. We must now take up this question. It is one of the most modern develop- ments of the Mendelian theory - - one that at first seemed to throw doubt on the fundamental idea of random assortment that gives Mendel's proportion 9:3:3:1. But I believe we can now offer a reasonable explanation, which shows that we have to do here with an extension of Mendelism that in no sense invalidates Mendel's principle of segregation. It not only extends that principle, as I have said, but gives us an oppor- tunity to analyze the constitution of the germ-plasm in a way scarcely dreamed of two or three years ago. 92 BEREDITY AND SEX The actual Dumbers obtained in the GR by YW crosa arc as follows. These are the figures that Dexter has obtained : GR9 GRS G\Y$ YRs YWS 6080 2870 36 34 2373 The apparent discrepancy between the expected and the realized ratios is due to the linkage of the factors that went into the cross. For instance, the factors for gray and red that went in with one chromosome are linked ; likewise their allelomorphs, yellow and white. AlS shown by the analysis, the F\ female offspring will have two sex chromosomes, one of each sort — one from the father, the other from the mother. But the male will have but one sex chromosome derived from the mother. If in the germ-cells of the Fi females there were random assortment of the allelomorphs in the sex chromosomes, the ideal ratio of 4:1:1:1:1 would, as has been said, be realized. But if the red and gray factors tend to remain together since they go in together in the one chromosome, and the yellow and white in the other chromosome tend to remain together, and if the chances are about 84 to 1 that the factors that enter together remain together, the realized ratio of 170 : 84 : 1 : 1 : 84 will be found. Experiments show that, for these two factors, the chances are about 84 to 1 that the factors that go in together remain together; hence the departure from Menders ratios for these two pairs of characters. We may make a general statement or hypothesis that covers cases like these, and in fact all cases where THE MENDELIAN PRINCIPLES OF HEREDITY 93 linkage occurs : viz. that when factors lie in different chromosomes they freely assort and give the Mendelian expectation ; but when factors lie in the same chromo- some, they may be said to be linked and they give departures from the Mendelian ratios. The extent to which they depart from expectation will vary with different factors. I have suggested that the departures may be interpreted as the distance between the factors in question. A THEORY OF LINKAGE In order to understand more fully what is meant by linkage on the interpretation that I have here followed, it will be necessary to consider certain changes that take place in synapsis. The sex chromosomes Ox g? ~tt± -K^ Ck <&* f 4* 19 32 g® <&=> 5§f U, L M«w). Bn* Fig. 49. — Illustrating chiasma-type theory. 1 and 2, from Triton cristatus, 3-46, chromosomes of Batracoseps attenuatus. Note especially the chiasma shown in 13. (After Janssens). f'l HEREDITY AND SEX when two axe present as in the female), like all other chromosomes, unite in pairs at the synaptic period. A recognized method of uniting is for like chromosomes to come to lie side by side. Before they separate, as they do at one of the two maturation divisions, each chromosome may be seen to be split throughout the length. Thus there are FlG. 50. — Chromatin filaments in the amphitene stage from spermato- cytes of Batracoseps. (After Janssens.) formed four parallel strands each equivalent to a chromosome -- the tetrad group. At this time Jans- sens has found that cross unions between the strands are sometimes present (Fig. 49). In consequence a strand is made up of a part of one chromosome and a pari of another. Whether this cross union can be referred to an earlier stage — at the time when the two like chromosomes come together, when they can be THE MENDELIAN PRINCIPLES OF HEREDITY 95 seen to twist around each other (Fig. 50) - - is not certain ; but the fact of the existence of cross connec- tions is the important point. A consequence of this condition is that the chromosomes that come out of the tetrad may represent different combinations of those that united to form the group. On the basis of this observation we can explain the results of associ- ated inheritance. For, to the same extent to which the chromosomes that unite remain intact, the factors are linked, and to the extent to which crossings occur the exchange of factors takes place. On the basis of the assumption of the linear arrangement of the factors in the chromosomes the distance apart of the factors is a matter of importance. If two factors lie near together, the chance of a break occurring be- tween them is small in proportion to their nearness. We have found that some factors cross over not once in a hundred times. I interpret this to mean that they lie very near together in the chromosome. Other factors cross over to various degrees ; in the extreme cases the chance is one to one that they cross over. In such cases the factors lie far apart — perhaps near the ends of the chromosome. The strongest evidence in favor of this view is found when the constant relation of the factors to each other is considered. If, for instance, we know the distance from A to B (calculated on the basis of crossing over) and from B to C, we can predict what A and C will do when they are brought into the hybrid from two parents. If a fourth factor, D, is discovered and its distance from A is made out, we can predict before the experiment is made what will take place when D and 96 HEREDITY AM) SEX B or D and C are combined. The prediction has been fulfilled so many times and in so many ways that we feel some assurance thai we have discovered here a working hypothesis of considerable interest. If the hypothesis becomes established, it will enable us to analyze the structure of the chromosomes themselves in the sense that we can determine the relative location of factors in the chromosomes. If, as seems not improbable, the chromosomes are the most important element in Mendelian inheritance, the determination of the linear series of factors contained in them becomes a matter of great theoretical interest; for we gain further insight into the composition of the material on which heredity itself depends. There is a corollary to this explanation of crossing over that has a very direct bearing on the results. In the male there is only one X chromosome present. Hence crossing over is impossible. The experimental results show that no crossing over takes place for sex-linked factors in the male of drosophila. Other factors, however, lie in other chromosomes. In these cases the chromosomes exist in pairs in the male as well as in the female. Does crossing over occur here in both sexes? Let me illustrate this by an example. In drosophila the factor for black body color and the factor that gives vestigial wings lie in the same chromosome, which we may call the second chromosome. If a black, long-winged female is crossed with a gray vestigial male, all the offspring will be gray in color and have long wings, since these are the dominant characters. If these Fi flies are inbred, the following classes will appear; THE MENDELIAN PRINCIPLES OF HEREDITY 97 Gray Long Black Long Gray Vestigial 2316 1146 737 It will be noted that there are no black vestigial flies. Their absence can be explained on the assump- tion that no crossing over in the male, between the factors in the second chromosome, has taken place. But if another generation (Fs) is raised, some black vestigial flies appear. With these, it is possible to test the hypothesis just stated. If, for instance, some of the long, gray, Fx females are mated to black vestigial males, the following classes are produced : GL BL GV BV 578 1413 1117 307 The results are explicable on the view that crossing over takes place in the germ-cell of the Fi female, and that the chance that such will occur is as 1 to 3. But if the long-winged, gray, Fi males are crossed to black vestigial females, only the following classes are produced : BL GV 992 721 These results are in accord with the hypothesis that no crossing over takes place between the second chromosomes in the Fx male. Why crossing over should occur in the Fx female, and not in the Fi male, we do not know at present ; and until the synaptic stages in the males and females have been carefully studied, we must wait for an answer to the question. 98 HEREDITY AND SEX THREE SEX-LINKED FACTORS When three sex-linked factors exist in the same chromosomes, we have a method by means of which the " crossing-over" hypothesis may be put to a further test. Sturtevant has recently worked over the evidence I I vfl E i 9/?' ' H' FlG. 51. — A-D, YW and G/2 that enter (.-1), crossing over to give YR and GW as seen in Z). E-E\, no crossing over. F-Fj, crossing between WM and /SL. G-Gi, crossing between YW and G/2. H-H\, double crossing over of YWM and GRL, to give FfiJlf and GWL. for a case of this kind. He analyzed the data of the cross between a fly having gray color, red eyes, long wings, mated to a fly with yellow color, white eyes, and miniature wings. The relative location of these three factors is shown in the above diagram (Fig. 51, THE MENDELIAN PRINCIPLES OF HEREDITY 99 E, F, G, H). The Fi flies gave the expected re- sults. These inbred gave the following F2 significant classes : 1 GRL YWM GWM YRL GRM YWL GWL YRM 2089 1361 17 23 887 817 5 0 In these results the classes where single crossing over is shown are GWM (17) and YRL (23) (Fig. 51, G, G') and GRM (887) and YWL (817) (Fig. 51, F, F'). There are two classes, namely, GWL (5) and YRM (0) (Fig. 51, H, H'), which involve double crossing over. In order that they may take place, the two sex chromo- somes in the female must break twice and reunite between the factors involved, as shown in the diagram. Such a redistribution. of the parts of the homologous chromosomes would be expected to occur rarely, and the small number of double crossovers recorded in the results is in accord with this expectation. In these questions of linkage we have considered some of the most recent and difficult questions in the modern study of heredity. We owe to Bateson and his collaborators the discovery of this new departure. In plants they have recorded several cases of linkage, and other authors, notably Correns, Baur, Emerson, East, and Trow have described further cases of the same kind. Bateson has offered an interpretation that is quite different from the one that I have here followed. His view rests on the assumption that separation of factors may take place at different times, or periods, in the development of the germinal tissues. 1 The classes omitted are those that do not bear on the question in hand. 100 HEREDITY AND SEX In a word, he assumes that assortment is not confined to the final stages in the ripening of the germ-cells, but may take place at any time in the germ-tract. It seems to me, however, if the results can be brought into line with the known changes that take place in the germ-cells at the time when the maternal and paternal chromosome unite, that we have not only a simpler method of dealing with these questions, but it is one that rests on a mechanism that can be studied by actual observation. Moreover, on purely a priori grounds the assumptions that I have made seem much simpler and more tangible than the assumptions of " reduplication" to which Bateson resorts. But leaving these more theoretical matters aside, the evidence from a study of sex-linked characters shows in the clearest manner that they, while following Men- del's principle of segregation, are also undeniably asso- ciated with the mechanism of sex. There is little doubt that sex itself is inherited in much the same way, since we can explain both in terms of the same mechanism. This mechanism is the behavior of the chromosomes at the time of the formation of the germ- cells. CHAPTER IV Secondary Sexual Characters and their Rela- tion to Darwin's Theory of Sexual Selection In his "Origin of Species" Darwin has defined Sexual Selection as depending "on a struggle between the individuals of one sex, generally the males, for the possession of the other sex. The result is not death to the unsuccessful competitor, but few or no offspring. Sexual selection is, therefore, less rigorous than natural selection. Generally, the most vigorous males, those which are best fitted for their places in nature, will leave most progeny. But in many cases, victory depends not so much on general vigor, as on having special weapons, confined to the male sex. A hornless stag or spurless cock would have a poor chance of leav- ing numerous offspring. Sexual selection, by always allowing the victor to breed, might surely give indomi- table courage, length to the spur, and strength to the wing to strike in the spurred leg, in nearly the same manner as does the brutal cock-fighter by the careful selection of his best cocks." Darwin continues: "Amongst birds, the contest is often of a more peaceful character. All those who have attended to the subject, believe that there is the severest rivalry between the males of many species to attract, by singing, the females. The rock-thrush of Guiana, birds of paradise, and some others, con- 101 L02 HEREDITY AND SEX gregatc, and successive males display, with the most elaborate rare, and show off in the best manner, their gorgeous plumage; they likewise perform strange antics before the females, which, standing by as spec- tators, at last choose the most attractive partner." Here we have two different pictures, each of which attempts to give an account of how certain differences lict ween the sexes have arisen — differences that we call "secondary sexual characters." On the one hand we deal with a contest between the males; on the other with choice by the female. The modus operandi is also different. After battle the successful male takes his pick of the females. If the scheme is to work, he must choose one that will leave the most offspring. On the other hand, we have the tourney of love. The males "show off"; the females stand by spell- bound and "at last choose the most attractive partner." Now, concerning this display of the males, I beg leave to quote a paragraph from Wallace's "Natural Selection and Tropical Nature" : "It is a well-known fact that when male birds possess any unusual ornaments, they take such positions or perform such evolutions as to exhibit them to the best advantage while endeavoring to attract or charm the females, or in rivalry with other males. It is therefore probable that the wonderfully varied decora- t ions of humming-birds, whether burnished breast- shields, resplendent tail, crested head, or glittering back, are thus exhibited ; but almost the only actual ob- servation of this kind is that of Mr. Belt, who describes how two males of the Florisuga mellivora displayed SECONDARY SEXUAL CHARACTERS 103 their ornaments before a female bird. One would shoot up like a rocket, then, suddenly expanding the snow-white tail like an inverted parachute, slowly descend in front of her, turning around gradually to show off both back and front. The expanded white tail covered more space than all the rest of the bird, and was evidently the grand feature of the performance. Whilst one was descending the other would shoot up and come slowly down expanded." There is just a suspicion in my mind that these males were otherwise engaged, for while I know nothing about the habits of these humming birds I find on the next page of " Tropical Nature " this statement : "Mr. Gosse also remarks: 'All the humming- birds have more or less the habit, when in flight, of pausing in the air and throwing the body and tail into rapid and odd contortions. This is most observable in Polytmus, from the effect that such motions have on the long feathers of the tail. That the object of these quick turns is the capture of insects, I am sure, having watched one thus engaged.' " If what I have just said implies that I take a light- hearted or even facetious attitude toward Darwin's theory, I trust that my position will not be misunder- stood. Darwin brought together in his book on the "Descent of Man" a mass of interesting observations for which he suggested a new theory. No one can read his wonderful book without the keenest interest, or leave it without high admiration for the thorough- ness with which the subject is treated ; for the ingenuity and skill with which the theory is applied to the facts, and, above all, admiration for the moderation, modesty, 101 HEREDITY AND SEX and honesty with which objections to the theory are considered. I will let do one admire Darwin more than I admire Darwin. But while affection and respect and honor are the finest fruits of our relation to each other, we cannot let our admiration for the man and an ever ready recognition of what he has done for you and for me prejudice us one whit in favor of any scientific theory that he proposes. For in Science there is no authority ! We should of course give serious considera- t ion to any theory proposed by a man of such wide expe- rience and trained judgment as Darwin ; but he himself, who broke all the traditions of his race, would be the first to disclaim the value of evidence accepted on authority. From the definition of sexual selection with which we started it may be said that Competition and Courtship stand for the two ways in which Darwin supposes the secondary sexual characters to have arisen. Competition amongst the males is only a form of nat- ural selection, as Darwin himself recognized (if we leave out of account the further assumption that the victor chooses his spoils). We may dismiss this side of the problem as belonging to the larger field of natural selec- tion, and give our attention mainly to those secondary sexual characters that Darwin supposes to have arisen by the female choosing the more ornamented suitor. I shall first bring forward some of the more striking examples of secondary sexual characters in the animal kingdom. These characters are confined almost ex- clusively to three great groups of animals -- Insects, SECONDARY SEXUAL CHARACTERS 105 Spiders, and Vertebrates. There are a few scattered instances found in other groups, but they are rare. In the lowest groups they are entirely absent, and are -*v Fig. 52. — ■ Four species of beetles in which the male (to the left) has horns which are absent in the female (to the right). (After Darwin.) not found at all in plants ; or rather, if character- istic differences exist in plants, they are not called by this name — for plants cannot see or move and there- fore cannot court each other. KM) HEREDITY AND SEX In fact, sighl in the sense of forming visual pictures can occur only when eyes are well developed. This \ "v pK?B Fig. 53. — Male (to left) with long eye stalks and female (to right) of a fly, Achia longividcns. (After AVood.) may be taken to score a point in favor of Darwin's hypothesis. In the group of insects the most noticeable differences occur in the butterflies and moths, and in flies. A few cases are found in the beetles and bugs. The male cicada's shrill call is supposed to attract the Fig. 54. - Male to left with horns and female to right without horns of a fly, Elaphomyia. (After Wood.) females. The males of certain beetles have horns — the female lacks them (Fig. 52). In a genus of flies the eyes are stalked, and the SECONDARY SEXUAL CHARACTERS 107 eyes of the male have stalks longer than those of the female (Fig. 53). In another genus of flies there are horns on the head like the antlers of the stag (Fig. 54). In the spiders the adult males are sometimes very small in comparison with the females (Fig. 55). The size difference may be regarded as a secondary sexual Fig. 55. — Male (to left) and female (to right) of a spider, Argiope aurelia. (From " Cambridge Natural History.") character. Darwin points out, since the male is some- times devoured by the female (if his attentions are not desired) , that his small size may be an adaptation in order that he may more readily escape. But the point may be raised as to whether he is small in order to escape ; or whether he is eaten because he is small. In one of our native spiders, Habrocestum splendida, the adult males and females are conspicuously different 108 BEREDITY AND SEX i,, color the male more highly colored than the female. In another native species, Maevia vittata, there arc two kinds of males, both colored differently from the female. Passing over the groups of fishes and reptiles in which some striking cases of differences between the 3exes occur, we come to the birds, where we find the besl examples of secondary sexual characters. Fig. 56. — Super!) bird of paradise. (After Elliot.) In the white-booted humming bird (Fig. 14) two of the tail feathers of the male are drawn out, their shafts denuded of the vanes except at the tip where the feather ends in a broad expansion. In the great bird of paradise, of the Aru Islands (Fig. 13), the male has wonderful plumes arising from the aides that can be erected to produce a gorgeous display. SECONDARY SEXUAL CHARACTERS 109 The female is modestly clothed. In the male of the superb bird of paradise (Fig. 56), the mantle behind the neck, when erected, forms a striking ornament ; and on the breast there is a brilliant metallic shield. In the six-shafted bird of paradise (Fig. 57) the male has on its head six feathers with wiry shafts, Fig. 57. — Six-shafted bird of paradise. (After Elliot.) ornaments that occur in no other birds. In the king bird of paradise there are remarkable fans at the sides of the body of the male that can be expanded. The feathers of the fan are emerald-tipped. The two middle feathers of the tail are drawn out into "wires" with a green web at one side of the tip. In mammals, secondary sexual differences are very 110 BEREDITY AND SEX common, although startling differences in color are rather rare. In the male the coat of fur is often darker than thai of the female. In many deer the antlers are present in the male alone. In Steller's sea-lion the male is much larger and stronger than the female. In a race of the Asiatic elephant the male has tusks much larger than those of the female. If we fix our attention exclusively on these remarkable Fig. 58. - Wilson's phalarope, female (in center), male (to right and behind). A bird in winter plumage is at the left. (From Eaton, "Birds of New York.") cases where differences between the sexes exist, we get a one-sided impression of the development of ornamentation and color differences in animals. We must not forget that in many cases males and females are both highly colored and exactly alike. We forget the parrots, the cockatoos, the kingfishers, the crowned pigeons, toucans, lories, and some of the starlings; the "brilliant todies" and the "sluggish jacamars" whose brilliant metallic golden-green breasts rivaJ SECONDARY SEXUAL CHARACTERS 111 those of the humming-birds ; we forget the zebras, the leopards ; the iridescent interiors of the shells of many mollusks ; the bright reds and purples of starfish, worms, corals, sea anemones, the red, yellow, and green sponges, and the kaleidoscopic effect of the microscopic radiolarians ; - - a brilliant array of color. ■.jpLkiM Fig. 59. — (.A), female of a copepod, Calocalanus plumalnsus. (B), a female of Calocalanus parvus. (C), male of last. (After Giesbreeht.) In the egret both males and females have remark- able nuptial plumes, which, had they been present in one sex alone, would have been classified as secondary sexual characters. It does not appear that selection had anything to do with their creation. Our common screech owl exists in two colored types sharply separated. No one is likely to ascribe these differences to sexual selection, yet if one sex had been 112 HEREDITY AND SEX red and the other gray, the difference would have been put down to such selection. There are also cases like the phalarope, shown in Fig. 58, where the female is more highly ornamented than the male. In fact, for these cases, 1 )arwin supposed that the males select the females ; and in support of this view he points out thai the females are more active, while the male con- cerns himself with the brooding of the eggs. In some of the marine copepods female ornamentation is car- ried to even a higher point. In Calocalanus plumosus the female has one of the tail setsB drawn out into a long leather-like structure (Fig. 59). In another species, ( '. parva, all eight seta? of the tail of the female are feather-like (Fig. 59, B), while the male (Fig. 59, C) lacks entirely these "ornaments." In some butterflies also, two, three, or more types of females are known, but only one male type. I shall have occasion later to consider this case. COURTSHIP The theory of sexual selection hinges in the first place on whether the female chooses amongst her suitors. It has been objected that the theory is anthropo- morphic it ascribes to beetles, butterflies, and birds the highly developed esthetic sense of man. It has been objected that the theory leaves unexplained the development of this esthetic sense itself, for unless the female kepi in advance of the male it is not self-evident why she should go on selecting the more highly orna- ment ed. If she has advanced esthetically, what has broughl it about ? In answer to this last question SECONDARY SEXUAL CHARACTERS 113 Allen suggests that if the word conspicuousness is sub- stituted for the word beauty, the objection may to some extent be met. The more conspicuous male would be more likely to attract attention and be selected. It has been pointed out that there is more than a suspicion that the contests of the males for the females are sham affairs. They are like certain duels. There is seldom any one hurt. There are very few records of injured males, but many accounts of tremendous battles. And he who fights and runs away will live to mate another day. It is clear, I think, that the case against the theory must rest its claims on actual evidence rather than on ar- guments or poetry pro or con. Darwin admitted that the evidence was meager. Since his time something more has been done. Let us consider some of this new evidence. It will be conceded, I think, that Alfred Wallace, through his wide experience with animals in their native haunts, is in a position to give weighty evidence concerning the behavior of animals. He was with Darwin a co-discoverer of the theory of Natural Se- lection and cannot be supposed to be prejudiced against the selection principle. Yet Wallace has from the beginning strongly opposed the theory of sexual se- lection. Let me quote him : Referring to Darwin's theory of Sexual Selection — "I have long held this portion of Darwin's theory to be erroneous — and have held that the primary cause of sexual diversity of color was the need of protection, repressing in the female those bright colors which are normally produced in both sexes by general laws." I 1 1 BEREDITY AND SEX Again, Wallace Bays: "To conscious sexual selec- tion thai is, the actual choice by the females of the more brilliantly colored males or the rejection of those less gaily colored- -I believe very little if any effect is directly due. It is undoubtedly proved that in birds the females do sometimes exert a choice; but the evidence of this fact, collected by Mr. Darwin ('Descent of Man,' chap, xiv), does not prove that color determines that choice, while much of the strongest evidence is directly opposed to this view." Again, Wallace says: "Amid the copious mass of facts and opinions collected by Mr. Darwin as to the display of color and ornaments by the male birds, there is a total absence of any evidence that the females, as a rule, admire or even notice this display. The hen, th<' turkey, and the peafowl go on feeding, while the male is displaying his finery; and there is reason to believe that it is his persistency and energy rather than his beauty which wins the day." Hudson, who has studied the habits of birds in the field, asks some very pertinent questions in connec- tion with their performances of different kinds. "What relation to the passion of love and to the business of courtship have these dancing and vocal performances in nine cases out of ten ? In such cases, for instance, as that of the scissor-tail tyrant-bird, and its pyro- technic displays, when a number of couples leave their nests containing eggs and young to join in a wild aerial dance; the mad exhibition of grouped wings; the triplet dances of the spur-winged lapwing, to perform which two birds already mated are compelled to call in a third bird to complete the set; the harmonious SECONDARY SEXUAL CHARACTERS 115 duets of the oven-birds and the duets and choruses of nearly all the wood-hewers, and the wing-slapping aerial displays of the whistling widgeons, - - will it be seriously contended that the female of this species makes choice of the male able to administer the most vigorous and artistic slaps ? " He continues : ' ' How unfair the argument is, based on these carefully selected cases, gathered from all regions of the globe, and often not properly reported, is seen when we turn to the book of nature and closely consider the habits and actions of all the species in- habiting any one district." Hudson concludes that he is convinced that anybody who will note the actions of animals for himself will reach the conviction, that "conscious sexual selection on the part of the female is not the cause of music and dancing performances in birds, nor of the brighter colors and ornaments that distinguish the male." In the spiders Mr. and Mrs. Peckham have described in detail the courtship of the males. They believe that his antics are specifically intended to attract the female. They point out that his contortions are of such a sort that his brightest spots are turned toward the female. But, as he makes in any case a hundred twists and turns, there is some danger of misinterpret- ing his poses. Montgomery, who has studied spiders of other groups, reaches the conclusion that here the male is contorted through fear of the female. The male goes through some of the same turns if approached by another male. The courtship of the male spider is, he thinks, a motley of fear, desire, and general excitement. 1 16 HEREDITY AND SEX The evidence thai the Peckhams have given, even if taken to mean that the motions of the male attract the attention of the female, -- and lean see no reason whythismay not he the case, — fails nevertheless to show thai the female selects, when she has a chance, the more highly colored male. Mayer,and Mayer and Soule have made many ex- periments with moths. The moth promethea, Callo- % < Fig. 60. -Above, Callosamia promethia, male to left, female to right. Below Porthetria dispar, mule to left, female to right. samia promethea, is distinctly sexually dimorphic, as shown in Fig. 60. Mayer's experiments show that the male finds the female entirely by the sense of smell. The wings of some 300 males were painted with scarlet or green. They mated as often as did the normal male with which they competed. Where the wings of males were stuck on the female in place of her own wings, no disturbance in the mating was observed. (Vmversely, normal females accepted SECONDARY SEXUAL CHARACTERS 117 males with female wings as readily as they accepted normal males. In the gipsy moth (Porthetria dispar), the male is brown and the female white (Fig. 60). Here again it was found that the males are guided solely by the odor of the female. The silkworm moth is also sexually dimorphic. Kel- logg has shown that males with blackened eyes find a female with as much precision as does a moth with normal eyes. If the antennae are cut off, however, the male can not find the female unless by accident he touches her. He then mates. The female has scent glands whose odor excites the male with normal antennae even at some dis- tance. Chemotaxis and contact are the active agents in mating. The eyes do little or nothing. Andrews has found that touch determines mating in the crayfish. Pearse has obtained similar results. Chidester has shown the same thing for crabs. Holmes found this kind of behavior in Amphipoda. Fielde and Wheeler have also found that in ants sex-discrimination is through smell or by what Forel calls contact-odors. Montgomery and Porter recognize touch as the most important factor in mating in spiders. Petrunke- witsch has shown that in the hunting spider vision also helps the sexes to find each other. Tower has found that contact or odor rather than sight is the important condition in mating in leptinotarsa. I am able to give the unpublished results of A. H. Sturtevant on the mating of the fruit fly, drosophila. The male carries on an elaborate courtship in the sense that he circles around the female, throws out one MS III III IMTY AM) SEX wring, then the other, and shows other signs of exfcite^ incut. The male has sex combs on his fore legs, the female lacks them. Lutz cut them off and gave the female a choice between such a male and a normal male. One was chosen as often as the other. The wings of the male and female are wonderfully irides- cent. Sturtevanl cut off the wings of a male and matched him against a normal male. The female showed no marked preference. The converse experi- ment, when a clipped female competed with a normal female, showed no selection on the part of the males. If instead of allowing two males (a normal and a clipped) to compete for one female, a female is given to each male separately, and the interval before mating is noted, it is found that on an average this interval is 18 minutes for t he normal and 40 minutes for the clipped. If any such difference existed in the first case, when the two males were competing, we should expect a much greater selection in favor of the normal male than was actually found. This would seem to mean that the female is more quickly aroused by the normal male, and hence when both males are present she will accept the clipped male more quickly than when he alone is present. This suggests thai normal courtship precipitates copulation. In the following experiments the female was offered a choice between a new type (mutant) with white eyes am! a normal male. Conversely, the white-eyed fe- male had a like alternative. The evidence shows that the more vigorous male -- the red-eyed male — is more successful. Since vision itself is here involved, for the white- eyed flies are probably partly blind, the observations SECONDARY SEXUAL CHARACTERS 119 RED VERSUS WHITE EYES. Red tf Red ? White 9 Red ? -54 White 9-82 w. .4 , J Red 9 - 40 Whlte^ | White 9 -93 Red d" - 53 White d - 14 Red c? - 62 White cf - 19 GRAY VERSUS YELLOW COLOR. _ „ f Gray 9 - 25 GTayd (Yellow 9 "31 Yellow do not support any such interpretation. The Peckhams easily overturn his argument, as applied to spiders. Second, in birds, to which Wallace mainly refers, the sex glands of the male do not affect the secondary SECONDARY SEXUAL CHARACTERS 121 sexual characters of the male, while the sex glands of the female suppress these characters. Wallace's theory leaves out of account the hereditary factor that is also present and which acts quite apart from the physiological effects of the sex glands. Cunningham, who has more recently written on the same subject, accepts the hormone hypothesis as the basis for all cases of secondary sexual characters. But he fails to make good his view when it is applied to insects, for reasons that we shall take up later. He is especially concerned, however, in the attempt to make plausible his own hypothesis that secondary sexual characters have arisen through the use of the parts, or through special nervous or blood supplies to certain lo- calities of the body which become suffused during sexual excitement. In both cases he thinks the increased local activity will cause the cells to produce hormones that will be dispersed throughout the body, and absorbed by other cells. The germ-cells will in this way get their share and carry over the hormone to the next generation. Cunningham forgets one important point. If these imaginary hormones can get out of cells and into germ- cells, they can get out of the germ-cells again. Hence in the long period of embryonic and juvenile existence through which the individual passes before the second- ary sexual characters appear they would surely be lost from the body like any other ordinary hormone. CONTINUOUS VARIATION AS A BASIS FOR SELECTION And now let us turn to an entirely different aspect of the matter. What could selection do, admitting that selection may take place. For fifty years it has been I 22 HEREDITY AND SEX n A-E Fig. 61. — I. Diagram of five pure lines of beans (A, B, C, D, and E) and a population formed by their union, A-E. II. Diagrams illustrating a pure line of beans and two new biotypes derived from it. The upper diagram indicates the original biotype ; the second and third diagram in- dicate the elongated (narrower) and shorter (broader) type of beans. X indicates the average class of the original biotype. (After Johannsen.) SECONDARY SEXUAL CHARACTERS 123 taken for granted that by selecting a particular kind of individual the species will move in the direction of selection. A few examples will bring the matter before us. If we take a peck of beans and put all of those of one size in one cylinder and those of other sizes in other cyl- inders, and place the cylinders in a row, we get a result like that in Fig. 61, A-E. If we imagine a line joining t3(5 .Fig. 62. — The normal binomial curve or the "ideal curve" of distribu- tion. At the base line, the directions from the average value (o) are indicated with the standard deviation ("f Fig. 66. — Colias philodice, showing two female forms above and one male form below. view, and unaccountable on the other theory. In fact polymorphic forms, if they appear, would be expected to persist if they are not harmful to the species. We have in this country several species of butter- flies in which polymorphism exists. In the north the species Papilio turnus (Fig. 65) is alike in the male and in the female. But in the south two types of females exist — one like the male and the other a black type. L30 HEREDITY AND SEX Id the Eastern States there is a butterfly, Colias philodice, in which two types of female exist (Fig. 66). Gerould lias studied the mode of inheritance of these two types and finds that they conform to a scheme in which t he t wo females differ by a single factor. The evi- dence is strongly in favor of the view that one of these forms has arisen as a mutation. There is no need to suppose that sexual selection has had anything to do with its origin, and no evidence that it owes its exist- ence to mimicry of any other species. Finally, I should like to speak of a case that has come under my own observation. One of the mutants that appeared in a culture of drosophila had a new eye color that was called eosin. In the female the eye is much deeper in color than in the male. The race main- tains itself as a bicolor type without any selection. CONCLUSIONS In conclusion let me try to bring together the main considerations that seem to me to throw serious doubts on Darwin's theory of sexual selection. First Its fundamental assumption that the evolution of these characters has come about through the "will," "choice," or selection of the female is questionable, because of want of evidence to show that the females make their choice of mates on this basis. There is also some positive evidence to show that other conditions than selection of the more ornamented individual (because he is the more ornamental) are responsible for the mating. Second. We have come to have a different concep- tion of what selection can do than the sliding scale SECONDARY SEXUAL CHARACTERS 131 assumption that has been current, at least by implica- tion, in much of the post-Darwinian writings. Third. Recent advances in the study of variations have given us a new point of view concerning the na- ture of variation and the 'origin of variations. If we are justified in applying this new view to secondary sexual characters, the problem appears greatly sim- plified. CHAPTER V The Effects of Castration and of Transplan- tation on the Secondary Sexual Characters In several of the preceding chapters I have spoken in some detail of sex-linked inheritance. In sex-linked inheritance we deal with a class of characters that are transmitted to one sex alone in certain combinations, and have for this reason often been called sex-limited characters ; but these same characters can be trans- ferred by other combinations, as we have seen, to the other sex, and are therefore not sex-limited. In contrast to these characters secondary sexual char- acters appear in one sex only and are not transferable to the other sex without an operation. For instance, the horns of the stag and the colors and structures of certain male birds are in nature associated with one sex alone. It has long been recognized in mammals and birds that there is a close connection between sexual maturity and the full development of the secondary sexual char- acters. This relation suggests some intimate correla- tion between the two. It has been shown, in fact, in some mammals at least, that the development of the secondary sexual characters does not take place, or th.it they develop imperfectly, if the sex glands are removed. It may appear, therefore, that we are deal- ing here with a purely physiological process, and that 132 THE EFFECTS OF CASTRATION 133 the development of these structures and colors is a by- product of sex itself, and calls for no further explana- tion. But the question cannot be so hastily dismissed. This can best be shown by taking up at once the ma- terial at hand. OPERATIONS ON MAMMALS In the deer, the facts are very simple. If the very young male is castrated before the knobs of the antlers have appeared, the antlers never develop. Fig. 67. — Merino; male (horned) and female (hornless). If the operation is performed at the time when the antlers have already begun to develop, incomplete development takes place. The antlers remain covered by the velvet and are never thrown off. They are called peruke antlers. If the adult stag is castrated when the horns are fully developed, they are precociously i;;i EEREDITY AND SEX dropped, and are replaced, if at all, by imperfect ant- lers, and these are never renewed. These tacts make it clear that there is an intimate relation between the orderly sequence of development of the horns in the deer and the prese. ce of the male sexual glands. In the case of sheep, the evidence is more explicit. Here we have carefully planned experiments in which both sexes have been studied ; and there are breeding Fig. 68. Dorset; male (horned) and female (horned). experiments also, in which the heredity of horns has been examined. In some breeds of sheep, as in the Merinos and Herdwicks, horns are present in the males, absent in the females (Fig. 67). In other breeds of sheep, as in Dorsets, both males and females have horns (Fig. 68). In still other breeds both sexes lack horns, as in some of the fat-tailed sheep of Africa and Asia (Fig. 69). Marshall has made experiments with Herdwicks — a race oT sheep in which the rams have large, coiled horns and the ewes are hornless. Three young rams (3, 4, and 5 months old) were castrated. The horns had begun to grow (3, 4J/£, and 6 inches long) at the time of operating. They ceased to grow after the operation. THE EFFECTS OF CASTRATION 135 A similar operation was also carried out on females. Three Herdwick ewe lambs (about 3 months old) were operated upon. After ovariotomy, the animals were kept for 17 months, but no horns appeared, although in one, small scurs developed, in the other two scarcely even these. It is cle^r that the removal of the ovaries does not lead to the development of horns like those in the male. Now, the interpretation of this case can be made only when taken in connection with experiments in heredity. There is a crucial experiment that bears on this question. Arkell found when a Merino ewe (a race with horned males and hornless females) was bred to a ram of a hornless breed, that the sons had horns. In this case the factor for horns must have come from the hornless mother, while the development of the horns was made possible by the presence of the male glands. It is evident therefore in the castration experiment that a factor for horns is inherited by both sexes, but in order that the horns may develop fully, the male glands must be present and functional. In the Dorset, both sexes are horned, the horns of the females are lighter and smaller than the horns of the ram (Fig. 68). In the castrated males the horns are like those of the females. In this case we must sup- pose that the hereditary factor for horns suffices to carry them to the point in development reached by the females. To carry them further the presence of the sex glands of the male is necessary. In the case of the hornless breeds I do not know of any evidence from castration or ovariotomy. We may suppose, either that the factor for horns is absent ; or, 136 HEREDITY AND SEX if present, some inhibitory factor must bring about sup- pression of the horns. The former assumption seems more probable, for, as I shall point out, certain experi- ments in heredity indicate that no inhibitor is present in hornless breeds. The series is completed by cases like that of the eland and the reindeer. Both males and females rf ■ Fig. 69. — Fat-tailed hornless sheep (Ovis aries steatopyga persicci). have well-developed horns. In this case the hereditary factors suffice in themselves for the complete develop- ment of horns, for even after castration the horns de- velop. We have anticipated to some extent the conclusions arrived at by breeding experiments in these races of sheep. The best-known case is that of Wood, who crossed homed Dorsets and hornless Suffolks. As THE EFFECTS OF CASTRATION 137 shown in the picture (Fig. 70) the sons had horns — the daughters lacked them. When these are inbred, their offspring are of four kinds, horned males, hornless males, horned females, hornless females. It seems probable that these four classes appear in the following proportions : Horned 6 Hornless 6 Horned $ 3 1 1 Hornless $ 3 The explanation that Bateson and Punnett offer for this case is as follows : The germ-cells of the horned race • • 7. » Fig. 70. — 1, Suffolk (ram), hornless in both sexes; 2, Dorset (ewe), horned in both sexes ; 3, F\ ram, horned ; 4, F\ ewe, hornless ; 5-8, the four types of F2 ; 5 and 6 are rams. 7 and 8 are ewes. The hornless rams are pure for absence of horns, and the horned ewes are pure for the presence of horns. Figs. 5 and 6 represent lambs. (Bateson, after Wood.) (both male and female) carry the factor for horns (H) ; the germ-cells of the hornless race lack the factor for horns (h). The female is assumed to be homozygous for the sex factor, i.e. two sex chromosomes (X) are present ; while the male has only one sex chromosome i;;s BEREDITY AND SEX carried by the female-producing sperm. The analysis is then as follows: One "dose" of horns (H) in the male produces horns, bui two doses are necessary for the female. Hornless ? hX — hX Horned 6 H X — H Fr HXhX hornless $ II h X horned 6 Gametes f Eggs H X -hX of F, Sperm HX -hX — H — h F-2 Females F2 Males H X II X horned H HX horned // X h X hornless HhX horned h X H X hornless h H X horned h X h X hornless h h X hornless As pointed out by Punnett a test of the correctness of this interpretation is found by breeding the F\ hornless female to a hornless male (of a hornless breed). It is assumed that such a female carries the factors for horns in a heterozygous condition ; if so, then half of her sons should have horns, as the following analysis -hows : Fl Hornless $ HX — hX Hornless S h X — h h X HX hornless $ hXhX hornless $ h H X horned S h h X hornless S THE EFFECTS OF CASTRATION 139 \\h . JSP Fig. 71. — Upper figure normal male guinea pig (from below), to show mammary glands. Lower figure, a feminized male ; i.e. castrated when three weeks old and pieces of ovaries transplanted beneath the skin, at Ov. 1 jo HEREDITY AND SEX The actual result conforms to the expectation. The results of both of the experiments are consistent with the view that one factor for horns in the male produces horns, which we may attribute to the combined action of the inherited factor and a secretion from the testes which reenforces the action of the latter. This, how- ever, should be tested by castrating the Fi males. In the females, one factor for horns fails to produce horns, while two factors for horns cause their development. Aside from some of the domesticated animals (horses, cattle, dogs, cats, pigs), the only other mammals on which critical experiments have been made - - if we exclude man -are the rat and the guinea pig. The next case is unique in that the ovary was transplanted to a male. Steinach removed the sex glands from the male guinea pig and rat and transplanted into the same animals the ovaries of the female, which established themselves. Their presence brought about remarkable effects on the castrated male. The mammary glands, that are in a rudimentary condition in the male, be- come greatly enlarged (Fig. 71). In the rat the hair assumes the texture of that of the female. The skele- ton is also more like that of the female than the male. The size of the feminized rats and guinea pigs is less than that of normal (or of castrated) males and like that of the female (Fig. 72). Finally, in their sexual behavior, the feminized rats were more like females than like males. These cases are important because they are the only ones in which success- ful transplanting of the ovary into a male has been accomplished in vertebrates. THE EFFECTS OF CASTRATION 141 Fig. 72. — Two upper figures, normal male guinea pig to left, M, and his brother, F, to right — a feminized male. Two middle and two lower figures, a normal male at M , and his feminized brother, F. (After Steinach.) 1 l_> BEREDITY AND SEX OPERATIONS ON BIRDS In striking contrast to these results with mammals are those with birds, where in recent years we have gained sonic definite information concerning the devel- opment of secondary sexual characters. I am fortunate in being able to refer to several cases - - the most successful on record — carried out by my friend, H. D. Goodale, at the Carnegie Lab- oratory at Cold Spring Harbor. One "case" is that of a female Mallard duck from which the ovary was completely removed when she was a very young bird. Figure 10 illustrates the striking difference between the normal male and the female Mallard. In the spayed female the plumage is like that of the male. Darwin records a case in which a female duck in her old age assumed the characteristics of a male, and similar cases are recorded for pheasants and fowls. Goodale also removed the ovary from very young chicks. He found that the female developed the secondary sexual plumage of the cock. How shall we interpret these cases ? It is clear that the female has the potentiality of producing the full plumage of the male, but she does not do so as long as the ovary is present. The ovary must therefore be supposed to prevent, or inhibit, the development of secondary sexual characters that appear therefore only in the male. The converse operation - - the removal of the male glands from the male - - is an operation that is very conn non among poultry men. The birds grow larger and fatter. They are known as capons. In this case THE EFFECTS OF CASTRATION 143 the male assumes his full normal plumage with all of his secondary male sexual characters. It is said that the comb and wattles and to some extent the spurs are less developed in the capon than in the normal male. But aside from this it is quite certain that the de- velopment of the secondary sexual plumage in the U? HE LIABLE l>oun Kf yJOU ti N At, Fig. 73. — Male and female Seabright. Note short neck feathers and incomplete tail cover in male. In the Seabright cock the sickle feathers on back at base of tail are like those of the hen. (After " Reliable Poultry- Journal. ") male is largely independent of the presence of the sex glands. The method of inheritance of the secondary sexual characters in birds has been little studied. Daven- port has reported one case, but I am not sure of his in- terpretation.1 I have begun to study the question by using Seabright bantams, in which the male lacks some 1 Because it is not evident whether the secondary sexual char- acters as such are involved or only certain general features of coloration. 144 HEREDITY AM) SEX of the secondary sexual characters of the domestic races, aotably the saddle feathers, as shown in Fig. 73. \ male Seabright was mated to a black-breasted game female. The son was hen-feathered and like the Sea- brighl father in this respect. Evidently in this case the secondary sexual character in question is dominant and is transmitted from father to son. In the reciprocal cross one hen was obtained which was back-crossed to a recessive male. She produced both hen-feathered and normally feathered sons. The character appears therefore to be sex-limited but not sex-linked. If hen-feathering in the Seabright be rep- M'-ented by S and its normal allelomorph by s, the first cross would be as follows : — Game $ sF — s Seabright S S — S F1 SsF female Ss hen-feathered male Eggs of Fi SF~sF--S Sperm of Fx S — s F2 Females F2 Males SSF SS hen-feathered SsF Ss hen-feathered sSF sS hen-feathered ssF ss cock-feathered In conclusion, then, in mammals the secondary sexual characters owe their development to the testes. The testes add something to the common inheritance. Bui in birds the ovary takes something away. THE EFFECTS OF CASTRATION 145 OPERATIONS ON AMPHIBIA The male triton develops each year a peculiar fin or comb on the back and tail. Bresca has found that after castration the comb does not develop. If present at the time of castration, the comb is arrested, but only after several months. Certain color marks pe- culiar to the male are not lost after castration. If the comb is removed in normal males, it regenerates, but less perfectly in castrated males. If a piece of the dorsal fin of the female is transplanted to a normal male in normal position, it may later produce the comb under the influence of the testes. In the frog there appears at the breeding season a thickening of the thumb. Castrated males do not produce this thickening. If it is present in a male at the time of castration it is thrown off, according to Nussbaum, but according to Smith and Shuster its further progress only is arrested. According to Nussbaum and Meisenheimer injection of pieces of testes beneath the skin of a castrated male cause the thumb development to take place, or to continue, but Smith and Shuster question this con- clusion. Such are the remarkable relations that these experi- ments have brought to light. How, we may ask, do the sex glands produce their effect, in the one case to add something, in the other to suppress something? It has often been suggested these glands produce their effects through the nervous system by means of the nerves to or from the reproductive organs. This has been disproved in several cases by cutting the 1 iti HEREDITY AND SEX nerves and isolating the glands. The results are the same as when they are left intact. This brings us to one of the most interesting chapters of modern physiology, the production and influence of Internal Secretions. INTERNAL SECRETIONS It has become more and more probable that the effects in question are largely brought about by internal se- cretions of the reproductive organs. These secretions are now called "hormones" or " exciters." They are produced not only by glands that have ducts or outlets, but by many, perhaps by all, organs of the body. Some of these secretions have been shown to have very re- markable effects. A few instances may be mentioned by way of example. The pituitary body produces a substance that has an iin port ant influence on growth. If the pituitary body becomes destroyed in man, a condition called gigan- tism appears. The bones, especially of the hands and feet and jaws, become enlarged. The disease runs a short course, and leads finally to a fatal issue. The thyroid and parathyroid bodies play an im- portant role in the economy of the human body through their internal secretions. Removal leads to death. A diseased condition of the glands is asso- ciated with at least six serious diseases, amongst them cretinism. The thymus secretion is in some way connected with the reproductive organs. Vincent suggests that "the thymus ministers to certain needs of the body before the reproductive organs are fully developed." THE EFFECTS OF CASTRATION 147 Extirpation of the adrenal bodies, another ductless gland, leads to death. Injury to these bodies causes Addison's disease. Finally, the reproductive glands themselves produce internal secretions. In the case of the male mammal it has been shown with great probability that it is the supporting tissues of the glands, and not the germ-cells, that produce the secretion. Likewise, in the case of the ovary, it appears that the follicle cells of the corpus luteum give rise to an important internal secretion. If the sac-like glands are removed, the embryo fails to become attached to the wall of the uterus of the mother. If the ovary itself is removed from a young animal, before corpora lutea are formed, the uterus remains in an infantile condition. From a zoological point of view the recent experi- ments of Gudernatsch are important. He fed young frog tadpoles with fresh thyroid glands. " They began very soon to change into frogs, but ceased to grow in size. The tadpoles might begin their metamorphosis in a few days after the first application of the thyroid, and weeks before the control animals did so." In contrast to these effects Gudernatsch found that tadpoles fed on thymus grew rapidly and postponed metamorphosis. They might even, in fact, fail to change into frogs and remain permanently in the tad- pole condition. If thyroid extracts produce dwarfs; thymus extracts make giant tadpoles that never become adults. These examples will suffice to show some of the im- portant effects on growth that these internal secretions may bring about. 1 is HEREDITY AND SEX OPERATIONS ON INSECTS The Ensects constitute the third great group in which secondary sexual characters are common. The firsl operations on the reproductive organs were caiiicd out by Oudemans on the gipsy moth, Ocneria Parthekria i dispar. The male and female are strik- ingly different. Oudemans removed the testes from Fig. i 1. i»\arii> of Lymantria (.Porthetria) dispar transplanted to male. They have established connection with the sperm ducts. (After Kopec.) young caterpillars and found no change in the color, or size, of the male. He also removed the ovaries from young caterpillars, and again found no effect in the fe- male. The same experiments were later carried out on a large scale by Meisenheimer, who obtained similar results. Meisenheimer went further, however, and per- formed another operation of great interest. He removed the male glands from a male and implanted in their THE EFFECTS OF CASTRATION 149 place the ovary of a female, while it was still in a very immature condition. The caterpillar underwent its usual growth, changed to a chrysalid, and then to a moth. The moth showed the characters of the male. The presence of the ovary had produced no effect what- ever on the body character of the individual. When this individual was dissected, Meisenheimer found that the ovary had completely developed. It contained mature eggs, and the ovary had often established con- nection with the outlets of the male organs that had Fig. 75. — Testes of Lymantria (Porthetria) dispar transplanted to female. They have connected with the oviducts. (After Kopec.) been left behind, as seen in Fig. 74, taken from Kopec's description. The converse experiment was also made. The ovaries were removed from young caterpillars, and in their place were implanted the male sex glands from a young male caterpillar. Again no effects were produced on the moth, which showed the characteristic female size and color. On dissection the testes were also found to have grown to full size and to have produced spermatozoa (Fig. 75). These remarkable results, confirmed by Kopec, show L50 HEREDITY AND SEX that in these insects the essential organs of reproduc- tion have no influence on the secondary sexual char- acters of the individual. They show furthermore that the male generative organs will develop as well in the female as in the body of the male itself, and vice versa. It is evident, then, in insects (there is a similar, but less complete, series of experiments on the cricket), Fig. 7fi. — Papilio Memnon. 1, male; 2, '■], 4, three types of females. (After Meijere.) that the heredity of the secondary sexual characters ••an be studied quite apart from the influence of the sex glands. How, then, are the}- inherited so that they appear in one sex and not in the other sex? Within t he last two or three years the inheritance of the second- ary sexual differences in insects has been studied. First, there is the case of the clover butterfly, Colzas phUodice, that Gerould has worked out, where there THE EFFECTS OF CASTRATION 151 are two types of females and one kind of male (Fig. 66). Without giving the analysis of this case I may say that the results can be explained on a Mendelian basis. The peculiar feature of GeroukTs explanation is that two doses of the yellow-producing determiner in the female give yellow color — one dose gives white. In the male, on the other hand, one dose of yellow gives yellow. The second case is that of Papilio memnon, worked out by de Meijere from the experiments of Jacobson. There is one male type and three female types, Fig. 76. De Meijere accounts for the results of matings in this species recorded by Jacobson on the assumption of three factors, one for each type of female. The three factors are treated as allelomorphs, and therefore only two of them can be present in any one individual, and since they are allelomorphs they will pass into different gametes. The order of dominance is Achates, Agenor, Laomedon. The male carries these same factors, but they are not effective in him. Baur accounts for the results in a somewhat different way, but involving or- dinary Mendelian conceptions. An interesting case is that reported by Foot and Strobell. They crossed a female of a bug, Euschistus variolarius, the male of which has a black spot on the end of the body (the female lacking the spot), with a male of Euschistus servus that lacks the spot both in the males and the females (Fig. 77). The daughters had no spot ; the sons had a faint spot, less developed than in variolarius. When these {F\) offspring were inbred, they obtained 249 females without a spot, 1 52 HEREDITY AND SEX 107 males with a spot (developed to different degrees), and Si males without a spot. The authors give no explanation of their results --but they use the re- 31 - I'm. 77. — To left, in 1, is male of Euschistus variolarius, to right male of E. semis. 2 and 3 show eight F2 males; 4 shows seven F2 males from another mating. (After Foot and Strobell.) suits to discredit some of the explanations, that rest on the assumption that the chromosomes are the chief factors in Mendelian heredity. I venture, neverthe- less, to suggest the explanation shown on the accom- THE EFFECTS OF CASTRATION 153 panying diagram (Fig. 78). The analysis rests on the assumption that neither one, nor two doses of S in the female is able to produce a spot, while in the male one dose of S suffices. E. variolarius $ SX — SX E. servus $ sX — s Fi sXSX spotless $ sXS spotted $ Gametes (Eggs sX — SX of Fl [ Sperm sX — SX - - s — S sXsX sXSX SXsX ■ spotless $ SXSX , F2 sXs spotless 6 sXS spotted $ SXs spotted S SXS spotted S Fig. 78. — Diagram to show inheritance of spot when E. variolarius (?) is mated to E. servus ( origin of the two kinds of parthenogenetic females is unexplained. If we were justified in assuming that two classes of female-producing sperm are made in the male, even this point would be cleared up, for in this '/ I lliqiant. )/"< 'f/in-i t I Fig. 92. — Life cycle of Phylloxera carycecaulis. way the two classes of parthenogenetic females could be explained. In another group of insects, the aphids and phyllox- erans, the situation is different. In the phylloxerans of the hickories there emerges in the spring, from a fertilized egg, a female known as the stem mother (Fig. 92). She pierces a young leaf PARTHENOGENESIS 179 with her proboscis, which causes a proliferation of the cells of the leaf. Eventually the leaf cells grow so fast that the stem mother is overarched in the gall that she has called forth. Inside the gall she begins to lay her eggs. From these eggs emerge young individuals that remain in the gall until they pass their last molt, when they become winged migrants. Externally all the migrants are alike ; but if they are dissected, it will be found that some of them have large eggs, some small eggs. But all the offspring of the same mother are of one or of the other sort. The migrants crawl out of the opening in the gall and fly away. Alighting on other hickories, they quickly deposit their eggs. From the large eggs the sexual females emerge. They never grow any bigger than the egg from which they hatched. In fact, they have no means of feeding, and contain only one large egg with a thick coat — an egg almost as large as the female herself. From the small eggs of the migrants, minute males are produced — ripe at their birth. They fertilize the sexual female. She then deposits her single egg on the bark of the hickory tree. From this egg (that lies dormant throughout the entire summer and following winter) there emerges next spring a female, the stem mother of a new line. Here we find three generations in the cycle — - two of which reproduce by parthenogenesis. The first parthenogenetic generation gives rise to two kinds of individuals — one makes large eggs, the other small eggs. The large eggs produce sexual females, the small eggs males. ISO HEREDITY AND SEX A study of t ho chromosomes has explained how some of these changes in successive generations are brought about. It has explained, for instance, how males are produced by parthenogenesis, and why the sexual egg produces only females. Let us take up the last point first . When the spermatocytes are produced, we find, as in many other insects, that at one division a sex chromo- some passes to one cell only (Fig. 93). Two classes of cells are produced --one with three, one with two, chromosomes. The latter degenerates, and in conse- quence only the female-producing spermatozoa become functional. All fertilized eggs give rise therefore to females. The second point that has been made out concerns the production of the male. When the small egg produces its single polar body, all of the chromosomes divide, except one, which passes out entire into the polar body. In consequence the number of chromo- somes left in the egg is one less than the total number. In a word, there are five chromosomes in the male, while there are six chromosomes in the female (Fig. 93). By throwing out one chromosome, the change is effected. The chromosome is the mate of the sex chromosome, that appeared as a lagging chromosome in the spermato- genesis. In the large egg no such diminution takes place, consequently the diploid number of chromosomes is present in the female. These unite in pairs and are reduced to three when the two polar bodies of the sexual egg are produced. We see that by means of the chromosomes we can PARTHENOGENESIS 181 bring this case into line with the rest of our informa- tion bearing on the relation of the chromosomes to sex. One important point still remains to be explained. What causes some of the migrants to produce large PHYLLOXERA CAKYJECAULJS O Q 0 (J -eaa "> J TvtsC Sfizvmatcci/tt StccneL Fig. 93. — Chromosomal cycle of P. carycecaulis. lv_> HEREDITY AND SEX eggs and ot hers small eggs ? There must be either two kinds of stem mothers or one kind with double po- tentiality. Inasmuch as in other parthenogenetic types there is experimental evidence to prove that environ- mental conditions determine which alternative state, whether male-producing or female-producing individ- ual, is realized, so here we may, provisionally, follow the same interpretation. Once the course is deter- mined the subsequent internal events follow for two generations in a definite order. If the stem mother has been affected in one way, all of her daughters produce large eggs ; if in the other way, small eggs. In another group of animals, the daphnians, parthen- ogenetic species occur, that, in certains respects, are like the phylloxerans ; but these species illustrate also another relation of general interest. The fertilized winter egg produces always a female, the stem mother, which gives rise by parthenogenesis to offspring like herself, and the process may continue a long time. Each female produces one brood, then another and another. The last broods fail to develop, and this is a sign that the female has nearly reached the end of her life. But a parthenogenetic female may produce one or two large resting eggs instead of parthenogenetic females, and the same female may at another time produce a brood of males. The large resting eggs are inclosed in a thick outer protecting case. They must be fer- tilized in order to develop, yet they do not develop at once, but pass through an enforced, or a resting, stage that may be shortened, if the egg is dried and then returned to water. PARTHENOGENESIS 183 fi-icedtt / z 3 ^ 3 4, 7 r 9 so // /Z /3 ^ /3 i / + ?• Z O © O © o o © o °© © 9=) © o • 3 9. O O © O © o °© © °o v © • A a 1 o 0 o O °© © °<* 9® °9 V • S °« '?■ © 0 °0 % °»® © °3 °9 © • 6 a 1 -1 o O © °3 9P °9G °3 © ®. • o O © O °a 9» © 9® © * o © 0» o3® °» c^ © G#3 © • + 9-. o O °3 o® • © °» °ae °3 © $» 0 1 o © o °© °o Q3 © • ^// i o I /z o 1 13 £ ?, W i o °3 O® 3 °3 o® • /S k* ^ °0 °3 °» ® 03 • • /6 o °3 °3 s? • // O. °« °3W 3# • // o °o °3 © • /? fe 9, 3# • zo © • • A JDeaeruwaZe, (T) ttrUcdlcd> female- Q\ ?HaZe> C°S 7yavri&&ncf'7&rt-et£C' female (\ Tiwia-tC' Fig. 94. — Life cycle of Simocephalus ; successive broods in horizontal lines, successive generations in vertical lines. (After Papanicolau.) I s| BEREDITY AM) SEX In this life history wo do not know what changes take place in the chromosomes. It has, however, often been claimed in this case that the transition from par- t benogenesis to sexual reproduction is due to changes in the environment. In fact, this is one of the stock cases cited in the older lit nature to show that sex is determined by external agents. It was said, that if the environment causes males to appear, then sex is determined by the environ- ment. But as a matter of fact, in so far as changes in the environment affect this animal, they cause it to cease reproducing by parthenogenesis, and induce sexual reproduction instead. The evidence is consistent in showing that any external change that affects the mode of reproduction at all calls forth either sexual eggs or males. The machinery of parthenogenesis is switched off, and that for sexual reproduction is turned on. The discrepancies that appear in the older accounts are probably due, as Papanicolau suggests, to dif- ferent observers using females that belong to different phases of the parthenogenetic cycle. Papanicolau, starting in each case with a winter egg, finds that as successive broods are produced the color of the par- thenogenetic eggs can be seen to undergo a progressive change from blue to violet. As the change progresses the chance that males and sexual eggs (" females ") will appear is greater. Until towards the end of the life of the individual the males and females come, as it were, of themselves (Fig. 94). If, however, individuals of successive broods are subjected to cold, it is found that while earlier broods do not respond, later ones respond PARTHENOGENESIS 185 more and more easily and change over to the sexual phase of the cycle. What has just been said about the successive broods might be said equally of the first-born offspring of the successive generations, as Papanicolau's table shows (Fig. 94). Later born offspring respond more readily than do those that are historically nearer to the fer- tilized egg. More recently Grosvenor and Smith have found for Moina that if females are reared alone (at 25°-30° C.) no sexual broods appear, while parallel cultures of females crowded together give 30 or higher per cents of males. Agar has carried isolated females through 46 parthenogenetic generations and has found that mothers from late brood also give only parthenogenetic offspring in a suitable environment. A third type, Hydatina senta (Fig. 95), an almost microscopic wormlike animal belonging to the rotifers, reproduces by parthenogenesis. The resting egg always gives rise to a parthenogenetic female, i.e. she also reproduces by parthenogenesis. Whitney has obtained 500 generations produced in this way. But from time to time another kind of individual appears. She is externally like the parthenogenetic female, but has entirely different capacities. Her eggs may be fertilized, and if they are they become resting eggs inclosed in a hard case. The sperm enters when the eggs are immature and still in the ovary of the mother. The presence of a spermatozoon in an egg determines that the egg goes on to enlarge and to pro- duce its thick coat. But if perchance no males are there to fertilize the eggs, this same female produces a ISC, BEREDITY AM) SEX crop of male eggs that develop into males without being fertilized at all. There are several facts of unusual interest in the Hydat/na senta /// r'/,t_ nt r$ (Zero o! I, I, (z — u (Zcro)d F. L,(,9 Low F.(%(^ Z«,ro9 L^cf (Low) L2f .£ (Low)<3 F.L,t2 -Wz Low $ Lzl, Lc(, (Low) <5 F. L, in l~zlf ■Htqh? L, ^2 L2€f (Hi<>h)cf 9 F— L* d ^ft — -t2 -Hi«(v 9 (Low) d* r 9 F*, U>w 9 (-Hi$fi) c? J* F l% «**i^> L- • £ TiA Um-9 -ixH (Loxjd Fig. 102. — Illustrating Pearl's hypothesis. F = female factor present in half of the eggs and determining sex. L\ = factor for low egg produc- tion; k, its allelomorph for zero production of winter eggs. L2 = factor for high winter production; h, its allelomorph. the high productiveness of the mother. It is her sons that inherit the character, although they cannot show it except in their offspring. Aside from whatever practical interest these results may have, the facts are important in showing that such a thing as a factor for fertility itself may be present, without otherwise being apparent, and that this factor 214 HEREDITY AND SEX taken in connection with another (or others) gives high productivity. The other point to which I wish to call attention relates to a different matter. We have met with some cases where lowered fertility was due to eggs failing Fig. 103. — Normal mule of Drosophila (on Left) and male with "rudi- mentary" wings (on right). Note sex comb (lower left). to a greater or less degree to be fertilized by sperm of the same strain. A striking case of this kind is found in a mutant of the fruit fly that appeared in my cultures. The mu- tant has rudimentary wings (Fig. 103). The females are absolutely infertile with males of the same kind. FERTILITY 215 If they are mated to any other male of a different strain, they are fertilized. The males, too, are capable of fer- tilizing the eggs of other strains, in fact, are quite fertile. The factor that makes the rudimentary winged fly is of such a sort that it carries infertility along with it — in the sense of self-infertility. This result has nothing to do with inbreeding, and the stigma cannot be removed by crossing out and extracting. A somewhat similar factor, though less marked, is found by Hyde in certain of his inbred stock to which I have referred. As his experiments show, the infer- tility in this case is not due to lack of eggs or sperm, but to a sort of incompatibility between them so that not more than 20 per cent of the eggs can be fertilized by males of the same strain. In the flowering plants where the two sexes are often combined in the same individual, it has long been known that there are cases in which self-fertilization will not take place. The pollen of a flower of this kind if placed on the stigma of the same flower or of any other flower on the same plant will not fertilize the ovules. Yet the pollen will fertilize other plants and the ovules may be fertilized by foreign pollen. Correns has recently studied that problem and has arrived at some important conclusions. He worked with a common plant, Cardamine pratensis. In this plant self-fertilization is ineffectual. He crossed plant B with plant G, and reared their offspring. He tested these with each other and also crossed each of them back to its parents that had been kept alive for this pur- pose. The latter experiment is simple and more in- 216 IIKKKDITY AND SK\ structive. His results and his theory can best be given together. Correns assumes that each plant contains some factor that produces a secretion on the stigma of the flowers. This secretion inhibits the pollen of the same plant from extending its pollen tube. He found, in fact, that the pollen grains do not grow when placed on the stigma of the same plant. All plants will be hybrid tttck (haw in Ozidamin^ putemts (CntinJ.) 6 B I J! i BC B I 9 #6 Fig. 104. — Illustrating the crossing of the types Bb and Gg to give four classes : BG, Bg, bG, bg. Each of these is then back-crossed either to B or to G with the positive ( + ) or negative ( — ) results indicated in the diagram. for these factors, hence plant B will produce two kinds of germ-cells, B and b. Similarly, plant G will produce two kinds of germ-cells, G-g. If these two plants are crossed, four types will be produced. When these are back-crossed to the parents, the expectation is shown in the diagram (Fig. 104). Half the combination should be sterile and half should be fertile. This is, in fact, what occurs, as shown in the same diagram. The — signs indicate that fertilization does not occur, while the + signs indicate successful fertilization. Correns' theory is also in accord with other com- FERTILITY 217 binations that he made. There can be little doubt that he has pointed out the direction in which a solu- tion is to be found. There is a somewhat similar case in animals. In one of the Ascidians, Ciona intestinalis, an hermaphrodite, the sperm will not fertilize the eggs of the same indi- vidual. But the sperm will fertilize eggs of other individuals, and vice versa. Castle first found out this fact, and I have studied it on a large scale. The diagram (Fig. 105) gives an example of one such ex- periment made recently by W. S. Adkins. Five individuals are here used. The eggs of one individual, A, were placed in five dishes (horizontal line) ; likewise those of B, C, D, E. The sperm of A, designated by a (vertical lines) was used to fertilize the eggs, A, B, C, D, E; likewise the sperm b, c, d, e. The self-fertilized sets form the diagonal line in the diagram and show no fertilization. The other sets show various degrees of success, as indicated by the percentage figures. These results can best be under- stood, I think, by means of the following hypoth- esis. The failure to self-fertilize, which is the main problem, would seem to be due to the similarity in the hereditary factors carried by eggs and sperm ; but in the sperm, at least, reduction division has taken place prior to fertilization, and therefore unless each animal was homozygous (which from the nature of the case cannot be assumed possible) the failure to fertilize cannot be due to homozygosity. But both sperm and eggs have developed under the influence of the total or duplex number of hereditary factors ; hence they are alike, i.e. their protoplasmic substance has been 218 HEREDITY AND SEX unt lor the same influences. In this sense, the case is like that of stock that has long been inbred, and has jSetf and Cross ttrfi/izaT/on //? Gona. Aa o A* #7 A° £>Z Ad 84^ Ae ^>6 Ba 38 Bb o Bc 35 Bd 9# Be ^7 93 Cb 96 cc o cd 97 90 Da 9/ 98 D° // Da o D€ Ea 9$ V 9Z Ec eo Ea 7* Ee 6> Fig. 105. — The oblique line of letters .-la, Bb, Cc, Dd, Ee, gives the self- fertilized sets of eggs; the rest Ab, Ac, etc., the cross-fertilized sets. A, B, C, D, E = eggs; a, b, c, d, e, = sperm of same individuals. (From unpub- lished work of W. S. Adkins.) come to have nearly the same hereditary complex. If this similarity decreases the chances of combination be- tween sperm and eggs, we can interpret the results. Cor- rens' results may come under the same interpretation. FERTILITY 219 I have tried to bring together the modern evidence that bears on the problems of fertility and sterility. It is evident that there are many obscure relations that need to be explained. I fear that, owing to the diffi- culty of summarizing this scattered and diverse ma- terial, I have failed to make evident how much labor and thought and patience has been expended in ob- taining these results, meager though they may appear. But while it is going to take a long time and many heads and hands to work out fully these problems, there can be little doubt that the modern method is the only one by which we can hope to reach a safe conclusion. CHAPTER VIII Special Cases of Sex-Inheritance The mechanism of sex-determination that we have examined gives equal numbers of males and females. But there are known certain special cases where equality does not hold. I have selected six such cases for discussion. Each of these illustrates how the mechan- ism of sex-determination has changed to give a different result ; or how, the mechanism remaining the same, some outside condition has come in that affects the sex ratio. It is so important at the outset to clearly recognize the distinction between sex-determination and sex ratio, that I shall take this opportunity to try to make clear the meaning of this distinction. The failure to recognize the distinction has been an unfailing source of misunderstanding in the literature of sex. (1) A hive of bees consists of a queen, thousands of workers, and at certain seasons a few hundred drones or males. The workers are potentially females, and these with the queen give an enormous preponderance of females. In this case the explanation of the sex ratio is clear. Most of the eggs laid by the queen are fertilized, and in the bee all fertilized eggs become fe- males, because as we have seen there is only one class of spermatozoa produced, and not two as in other insects. There is a parallel and interesting case in one of the wasps described by Fabre. The female lays her eggs 220 SPECIAL CASES OF SEX-INHERITANCE 221 as a rule in the hollow stems of plants, each egg in a separate compartment. In some of the compartments she stores away much more food than in others. From these compartments large females hatch. From com- partments where less food is stored the smaller males are produced. It may seem that the amount of food stored up determines the sex of the bee. To test this Fabre took out the excess of food from the large compartments. The wasp that emerged, although small for want of food, was in every case a female. Fabre enlarged the smaller compartments and added food. The wasp that came out was a male, larger than the normal male. It is evident that food does not determine the sex, but the mother wasp must fertilize the eggs that she lays in chambers where she has stored up more food, and not fertilize those eggs that she deposits in com- partments where she has accumulated less food. (2) A curious sex ratio appeared in one race of fruit flies. Some of the females persisted in producing twice as many females as males. This was first discovered by Miss Rawls. In order to study what was taking place, I bred one of these females that had red eyes to a white-eyed male of another stock. All the offspring had red eyes, as was to be expected. I then bred these daughters individually to white-eyed males again (Fig. 106). Half of the daughters gave a normal ratio ; the other half gave the following ratio : Red Red White White 9 $ 9 $ 50 0 50 50 — *- *- HEREDITY AND SEX It is evident that one class of males has failed to ap- pear - - the red males. If we trace their history through these two generations, we find that the single sex chro- | )0 9 o1 9 9 & d1 (ramchs <3 Fig. 106. — Diagram to show the heredity of the lethal factor (carried by black X). A, red-eyed female, carrying the factor in one X, is bred to norma) white-eyed male. B, her red-eyed daughter, is bred again to a normal white-eyed male, giving theoretically the four classes shown in C, but one of the classes fails to appear, viz. the red-eyed male (colored black in the dia- gram). The analysis (to right) shows that this male has the fatal X. One of his sisters has it also, but is saved by the other X. She is the red-eyed female. If she is bred to a white-eyed male, she gives the results shown in D, in which ""<■ class of males is again absent, viz. the red-eyed male. In this diagram the black X represents red eyes and lethal (as though completely linked). SPECIAL CASES OF SEX-INHERITANCE 223 mosome that each red male contains is one of the two chromosomes present in the original red-eyed grand- mother. If this chromosome contains a factor which if present causes the death of the male that contains it, and this factor is closely linked to the red factor, the results are explained. All the females escape the fatality, because all females contain two sex chromo- somes. If a female should contain the fatal factor, her life is saved by the other, normal, sex chromosome. The hypothesis has been tested in numerous ways and has been verified. We keep this stock going by mat- ing the red females to white males. This gives con- tinually the 2 : 1 ratio. The white sisters, on the other hand, are normal and give normal sex ratios. (3) Another aberrant result, discovered by C. B. Bridges, is shown by a different race of these same fruit flies. It will be recalled that when an ordinary white- eyed female is bred to a red-eyed male all the sons have white eyes. But in the race in question a different re- sult follows, as shown by the diagram. From 90 to 95 per cent of the offspring are regular, but 5 per cent of the females and 5 per cent of the males are uncon- formable, yet persistently appear in this stock. The results can be explained if we suppose that the egg contains two X's and a Y chromosome and in con- sequence the two X's may pass out into one of the polar bodies, in which case the red-eyed males will develop if the egg is fertilized by a female-producing sperm; or the two X-chromosomes will both stay in the egg, and give a kind of female with three sex chro- mosomes. Here also numerous tests can be made. They verify 224 HEREDITY AND SEX the expectation. Thus by utilizing sex chromosomes that carry other sex-linked characters than white eyes, it can be shown that the results are really due to the whole sex chromosome being involved, and not to parts of it. The result is of unusual interest in another direction ; for it shows that the female-producing M M -» v,V' >]/„.', Q X y( /ir« :• ) y-5 )Y(i »*<* (S A\ ' | /Ay{i .'><> ■y/ L/AV>\ X'; /.-/ ; k«i .'' .; 30CDH' 0) Fig. 107. — Non-disjunction. A non-disjunction female produces four types of eggs, viz., XY — X — XX — Y. Such a female will give, with a normal male, XY, the classes indicated on the diagram. Y should be sub- stituted for O in the diagram. The non-disjunction 9 is XXY. sperm will make a male if it enters an egg from which both sex chromosomes have been removed. It is therefore not the female-producing sperm, as such, that gives a female under normal conditions, but this sperm plus the sex chromosome already present in the egg that gives an additive result — a female. (4) In the group of nematode worms belonging es- pecially to the genus Rhabditis, there are some extraor- SPECIAL CASES OF SEX-INHERITANCE 225 dinary perversions of the sex ratios. The table gives the ratios that Maupas discovered. Not only are the Diplogaster robustus 0. 13 male Rhabditis guignardi 0.15 male to 1000 females Rhabditis duliehura 0.7 male Rhabditis caussaneli 1.4 males Rhabditis elyaus 1.5 males Rhabditis coronata 5.0 males Rhabditis perrieri 7.0 males Rhabditis marionii 7.6 males Rhabditis duthiersi 20.0 males Rhabditis viguieri 45.0 males males extremely rare — almost reaching a vanishing point in certain cases — but they have lost the instinct to fertilize the female. The females, on the other hand, have acquired the power of producing sperm, so that they have passed over into the hermaphroditic state. The behavior and history of the sperm that the females produce has only recently been made out by Miss Eva Kruger. It is found that a spermatozoon enters each egg and starts the development, but takes no further part in the development (Fig. 108). The egg may be said to be half fertilized. It is a parthenogenetic egg and produces a female. (5) Some very high male ratios have been reported by Guyer in cases where birds of very different families have been crossed - - the common fowl by the guinea hen, individuals of different genera of pheasants bred to each other and to fowls, etc. Hybrids between different genera gave 74 $ - 13 9 . Hybrids between different species of the same genus 72 $ — 18 9. In most of these cases, as Guyer points out, the sex is 22G HEREDITY AND SEX recorded from the mounted museum specimen which has the male plumage. But it is known that the re- productive organs of hybrids, extreme as these, are gen- erally imperfect and the birds are sterile. It has been Fig. 1. Kig. 2. Fig 3 •••ft < * •' FiB •' Fig. 6 Kg. 4 Fig. U Kg 7. Kg. 8. Fig. 9. Kg- 10. Fig. 11. ^.. 'tat •/{• ,' >rrq £*«* -^: •'1' '*• V**'" Fig. 108. — Oogenesis and spermatogenesis of Rhabditia aberrans. 1-5, stages in oogenesis, including incomplete attempt to form one polar body. Eighteen chromosomes in 1 and again in 4 and 5. In 3 the entering sperm seen at right. 6, prophase of first spermatocyte with 8 double and two single chromosomes (sex chromosomes). At the first division (7) the double chromosomes separate, and the two sex chromosomes divide, giving ten chromosomes to each daughter cell (8). At the next division the two sex chromosomes move to opposite poles, giving two female-producing sperm (9 and 10). Rarely one of them may be left at the division plane and lost, so that a male-producing sperm results that accounts for the rare occurrence of males. (After E. Kriiger.) shown that if the ovary of the female bird is removed or deficient, she assumes the plumage of the male. Possibly, therefore, some of these cases may fall under this heading, but it is improbable that they can all be explained in this way. In the cases examined by Guyer himself the hybrids were dissected and all four were found to be males. SPECIAL CASES OF SEX-INHERITANCE 227 Pearl has recently pointed out that the sex ratio in the Argentine Republic varies somewhat accord- ing to whether individuals of the same race, or of dif- ferent races, are the parents. As seen in the following table, the sex ratio of Italian by Italian is 100.77 ; Comparison op the Sex Ratios of the Offspring of Pure and Cross Matings Sex Ratio Difference P.E. of Difference Italian $ Argentine 9 Italian $ Italian 9 105.72 ±.46 100.77 ± .20 Difference 4.95 ±.50 9.9 Italian $ Argentine 9 Argentine $ Argentine 9 105.72 ±.46 103.26 ±.34 Difference 2.46 ±.57 4.3 Spanish $ Argentine 9 Spanish $ Spanish 9 106.69 ±.74 105.55 ±.36 Difference 1.14 ±.82 1.4 Spanish $ Argentine 9 Argentine $ Argentine 9 106.69 ±.74 103.26 ±.34 Difference 3.43 ±.81 4.2 Argentine by Argentine, 103.26 ; but Italian by Argen- tine, 105.72. If, as has so often been found to be the case, a hybrid combination gives a more vigorous progeny, the higher sex ratio of the cross-breed may account for the observed differences, since other data show that the male infant is less viable and the in- creased vigor of a hybrid combination may increase the chance of survival of the male. 228 HKRKDITY AND SEX (6) We come now to the most perplexing case on record. In frogs the normal sex ratio is approximate equality. Professor Richard Hertwig has found that it' t he deposition of the eggs is prevented for two to three days (after the eggs have reached the uterus) the proportion of males is enormously increased — in the extreme case all the offspring may be males. By critical experiments Hertwig has shown that the results are not due to the age of the spermatozoa, al- though in general he is inclined to attribute certain differences in sex-determination to the sperm as well as to the eggs. The evidence obtained by his pupil, Kuschakewitsch, goes clearly to show that the high male sex ratio is not due to a differential mortality of one sex. In the following table four experiments (a, b, c, d) are summarized. The interval between each record /72\ a) 47 9 : 32 2 0 9: 97 2 /°\ /18\ /3(l\ b) 349:472 659:772 1569:1942 79:482 /36\ /18\ e) 64 9 : 61 2 1019: 139 2 1 15 9 : 169 2 /,8\ /24\ /22\ d) 559:522 1489:872 719:702 179:1292 is written above in hours. In all cases an excess of males is found if the eggs have been kept for several hours before fertilization. In the first (a), second (b), and fourth (d) cases the excess of males is very great. Hertwig attempts to bring his results into line with SPECIAL CASES OF SEX-INHERITANCE 229 his general hypothesis of nucleo-plasm relation. He holds, for instance, that sex may be determined by the relation between the size of the nucleus and the proto- plasm of the cell. As the value of the evidence has been seriously called into question in general, and as there is practically no evidence of any weight in its favor in the present case, I shall not dwell further on the question here. But the excessively high male ratio is evident and positive. How to explain it is difficult to say. It is just possible, I think, that delay may have injured the egg to such an extent that the sperm may start the development but fail to fuse with the egg nucleus. Under these circumstances there is the possi- bility that all the frogs would be males. Miss King has also carried out extensive sets of ex- periments with the common toad. She has studied the eggs and the sperm under many different conditions, such as presence of salt solutions, acids, sugar solutions, cold, and heat. Her results are important, but their inter- pretation is uncertain. In sugar solutions and in dry fertilization the males decreased, in the latter from 114.10 to 29.41 per 100 $ $ . The normal sex ratio for the toad is 90 $ to 100 $ . Whether the solutions have in any sense affected the determination of sex, or acted to favor one class of sperm at the expense of the other re- mains to be shown, as Miss King herself frankly admits. In the case of man there are extensive statistics concerning the birth rate. The accompanying tables give some of the results. There are in all parts of the world more males born than females. The excessively high ratios reported from the Balkans (not given here) may be explained on psychological grounds, as failure 230 HEREDITY AND SEX Males Italy 105.8 France 104.6 England . 103.6 Germany 105.2 Austria 105.8 Hungary 105.0 Switzerland 104.5 Belgium 104.5 Holland 105.5 Spain 108.3 Russia 105.4 to 100 females to report the birth of a boy is more likely to lead to the imposition of a fine on account of the conscription. There can be no doubt, however, that slightly more males than females are born. Moreover, if the still- born infants alone are recorded, surprisingly large ratios occur, as shown in the next table. to 100 females Males Italy 131.1 France 142.2 Germany 128.3 Austria 132.1 Hungary 130.0 Switzerland 135.0 Belgium 132.1 Holland 127.7 Sweden 135.0 Norway 124.6 Denmark 132.2 And if abortive births are also taken into account, the ratio is still higher. It seems that the male embryo is not so strong as the female, or else less likely, from other causes, to be born alive. In many of the domesticated animals also, especially SPECIAL CASES OF SEX-INHERITANCE 231 the mammals, there is an excess of males at birth, as the next table shows. Males Females Horse 98.31 100 (Diising) Cattle 107.3 100 (Wilckens) Sheep 97.7 100 (Darwin) Pig 111.8 100 (Wilckens) Rat 105.0 100 (Cuenot) Dove 105.0 100 (Cuenot) Hen 94.7 100 (Darwin) A little later I shall bring forward the evidence that makes probable the view that in man the mechanism for sex-determination is like that in other animals, where two classes of sperm are produced, male- and female-producing. How then can we account in the human race for the excess of eggs that are fertilized by male-producing spermatozoa ? At present we do not know, but we can at least offer certain suggestions that seem plausible. In mammals the fertilization occurs in the upper parts of the oviduct. In order to reach these parts the sperm by their own activity must traverse a dis- tance relatively great for such small organisms. If the rate of travel is ever so slightly different for the two classes of sperm, a differential sex ratio will occur. Again, if from any cause, such as disease or alcoholism, one class of sperm is more affected than the other, a disturbance in the sex ratio would be expected. At present these are only conjectures, but I see no ground for seizing upon any disturbance of the ratio in order to formulate far-reaching conclusions in regard to sex-determination itself. As I pointed out in the beginning of this chapter, we may go 232 HEREDITY AND SEX wide of the mark if we attempt to draw conclusions concerning the determination of sex itself from devia- tions such as these in the sex ratio, yet it is the mistake that has been made over and over again. We must look to other methods to give us sufficient evidence as to sex-determination. Fortunately we are now in a position to point to this other evidence with some assurance. With the mechanism itself worked out, we are in a better position to explain slight variations or variables that modify the combinations in this way or in that. THE ABANDONED VIEW THAT EXTERNAL CONDITIONS DETERMINE SEX But before taking up the evidence for sex-determina- tion in man I must briefly consider what I have been bold enough to call the abandoned view that external conditions determine sex. Let us dismiss at once many of the guesses that have been made. Drelincourt recorded 262 such guesses, and Geddes and Thomson think that this number has since been doubled. Naturally we cannot consider them all, and must confine ourselves to a few that seem to have some basis in fact or experiment. The supposed influence of food has been utilized in a large number of theories. The early casual evidence of Landois, of Mrs. Treat, and of Gentry has been entirely set aside by the careful observations of Riley, Kellogg and Bell, and Cuenot. In the latter cases the experiments were carried through two or even three generations, and no evidence of any influence of nourishment was found. SPECIAL CASES OF SEX-INHERITANCE 233 The influence of food in sex-determination in man has often been exploited. It is an ever recurrring episode in the ephemeral literature of every period. The most noted case is that of Schenk. In his first book he said starvation produced more females ; in his second book he changed his view and supposed that starvation produces more males. Perhaps the most fertile source from which this view springs is found in some of the earlier statistical works, especially that of Diising. Diising tried to show that more girls are born in the better-fed classes of the com- munity, in the poorer classes more boys. The effective difference between these two classes is supposedly one of food ! For instance, he states that the birth-rate for the Swedish nobility is 98 boys to 100 girls, while in the Swedish clergy the birth-rate is 108.6 boys to 100 girls. Other statistics give exactly opposite results. Pun- nett found for London (1901) more girls born amongst the poor than the rich. So many elements enter into these data that it is doubtful if they have much value even in pointing out causes that affect the sex ratio, and it is quite certain that they throw no light on the causes that determine sex. In other mammals where a sex ratio not dissimilar to that in man exists, extensive experiments on feeding have absolutely failed to produce any influence on the ratio. We have, for instance, Cu&iot's experi- ments with rats, and Schultze's experiments with mice. The conditions of feeding and starvation were much more extreme in some cases than is likely to occur ordinarily, yet the sex ratio remained the same. Why in the face of this clear evidence do we find 234 HEREDITY AND SEX zoologists, physicians, and laymen alike perpetually discovering some new relation between food and sex? It is hard to say. Only recently an Italian zoologist, Russo, put forward the view that by feeding animals on lecithin more females were produced. He claimed that he could actually detect the two kinds of eggs in the ovary - - the female- and the male-producing. It has been shown that his data were selected and not complete ; that repetition of his experiments gave no confirmative results, and probably that one of the two kinds of eggs that he distinguished were eggs about to degenerate and become absorbed. But the food theories will go on for many years to come — as long as credulity lasts. Temperature also has been appealed to as a sex fac- tor in one sense or another. R. Hertwig concluded that a lower temperature at the time of fertilization gave more male frogs, but Miss King's observations failed to confirm this. There is the earlier work of Maupas on hydatina and the more recent work of von Malsen on Dinophilus apatris. I have already pointed out that Maupas' results have not been con- firmed by any of his successors. Even if they had been confirmed they would only have shown that tempera- ture might have an effect in bringing parthenogenesis to an end and instituting sexual reproduction in its stead. In hydatina the sexual female and the male producing individual are one and the same. A more striking case could not be found to show that the en- vironment does not determine sex but may at least change one method of reproduction into another. There remain von Malsen's results for dinophilus, SPECIAL CASES OF SEX-INHERITANCE 235 where large and small eggs are produced by the same female (Fig. 109). The female lays her eggs in clus- ters, from three to six eggs, as a rule, in each cluster. The large eggs produce females ; the small eggs pro- Fig. 109. — Dinophilus gyrociliatus. Females (above and to left) and males (below and to right). Two kinds of eggs shown in middle of lower row. (After Shearer.) duce rudimentary males that fertilize the young fe- males as soon as they hatch and before they have left the jelly capsule. Von Malsen kept the mother at different tempera- tures, with the results shown in the table. The ratio of small eggs to large eggs changes. But the result Temperature No. OF Broods d 9 Sex Ratio Eggs per Brood Room temp. 19° C. . Cold, 13° C. . . . Heat, 26° C. . . . 202 925 383 327 973 507 813 2975 886 1:2,4 1:3,5 1:1,7 5,6 4,2 3,6 236 HEREDITY AND SEX obviously may only mean that more of the large eggs are likely to be laid at one temperature than at another. In fact, temperature seemed to act so promptly accord- ing to Von Malsen's observations that it is very un- likely that it could have had any influence in deter- mining the kind of egg produced, but rather the kind of egg that was more likely to be laid. We may dis- miss this case also, I believe, as not showing that sex is determined by temperature. SEX-DETERMINATION IN MAN Let us now proceed to examine the evidence that bears on the determination of sex in man. I shall draw on three sources of evidence : 1. Double embryos and identical twins. 2. Sex-linked inheritance in man. 3. Direct observations on the chromosomes. The familiar case of the Siamese twins is an example of two individuals organically united. A large series of such dual forms is known to pathologists. There are hundreds of recorded cases. In all of these both individuals are of the same sex, i.e. both are males or both are females. There is good evidence to show that these double types have come from a single fer- tilized egg. They are united in various degrees (Fig. 110) ; only those that have a small connecting region are capable of living. These cases lead directly to the formation of separate individuals, the so-called identical twins. Galton was one of the first, if not the first, to recognize that there are two kinds of twins — identical twins and ordinary or fraternal twins. SPECIAL CASES OF SEX-INHERITANCE 237 Identical twins are, as the name implies, extremely alike. They are always of the same sex. There is every presumption and some collateral evidence to show that they come from one egg after fer- tilization. On the other hand, amongst ordinary twins a boy and a girl, or two boys and two girls, occur in the ratio expected, i.e. on the basis that their sex is mm ¥¥f° •#£<> DIAGRAM SHOWING THE INTERRELATIONS OF THE VARIOUS SORTS OF DIPLOPAGI AND DUPLICATE TWINS, ILLUSTRATIVE OF THE THEORY. ADVANCED IN THIS PAPER. FURTHER EX- PLANATION IN THE TEXT. Fig. 110. — Diagram showing different types of union of double monster (After Wilder.) not determined by a common external or internal cause. Since fraternal twins and identical twins show these relations at birth and from the fact that they have been in both cases subjected to the same condi- tions, it follows with great probability that sex in such cases is determined before or at the time of fertilization. This conclusion finds strong support from the condi- 238 HEREDITY AND SEX t ions that have been made out in the armadillo. Jehring first reported that all the young of a single litter are of the same sex (Fig. 111). The statement has been verified by Newman and by Patterson on a large scale. In addition they have found, first, that only one egg leaves the ovary at each gestative period ; and second, that from the egg four embryos are pro- Fig. 111. — Nine-banded Armadillo. Four identical twins with a common placenta. (After Newman and Patterson.) duced (Fig. 112). The material out of which they develop separates from the rest of the embryonic tissue at a very early stage. The four embryos are identical quadruplets in the sense that they are more like each other than like the embryos of any other litter, or even more like each other than they are to their own mother. The second source of evidence concerning sex-deter- SPECIAL CASES OF SEX-INHERITANCE 239 urination in man is found in the heredity of sex-linked characters. The following cases may well serve to illustrate some of the better ascertained characters. The tables are taken from Davenport's book on " Heredity in Relation to Eugenics." The squares indicate males, affected males are black squares ; the heavy circles indi- cate females, that are supposed to carry the factors, but i: "■■■ Wk Fig. 112. — Nine banded Armadillo. Embryonic blastocyst that has four embryos on it, two of which are seen in figure. (After Newman and Patterson.) such females do not exhibit the character themselves. Solid black circles stand for affected females. Haemophilia appears in affected stocks almost ex- clusively in males (Fig. 113). Such males, mating with normal females, give only normal offspring, but the daughters of such unions if they marry normal males will transmit the disease to half of their sons. Affected females can arise only when a haemophilious male marries a female carrying haemophilia. If we 240 m:i:i:i)iTY and sex SPECIAL CASES OF SEX-INHERITANCE 241 9 a X XX 9 icr iCT X( Fig. 114. — Diagram to indicate heredity of color blindness through male. A color-blind male (here black) transmits his defect to his grandsons only. XX )CXX &m><>mm XX X© X 9 9 cr cr Fig. 115. — Diagram to indicate heredity of color blindness through female. A color-blind female transmits color blindness to all of her sons, to half of her granddaughters and to half of her grandsons. 242 HEREDITY AND SEX substitute white eyes for haemophilia, the scheme already given for white versus red eyes in flies applies to this case. If, for instance, the mother with normal eyes has two X chromosomes (Fig. 114), and the fac- tor for haemophilia is carried by the single X in the male (black X of diagram), the daughter will have one affected X (and in consequence will transmit the factor), but also one normal X which gives normal OrD OtD O OtD 6 6 6 6 6 Fig. 116. — Pedigree of Ichthyosis from Bramwell. (After Davenport.) vision. The sons will all be normal, since they get the X chromosomes from their mother. In the next generation, as shown in the diagram (third line), four classes arise, normal females, hybrid females, normal males, and haemophilious males. Color blindness fol- lows the same scheme, as the above diagrams illustrate (Figs. 114 and 115). In the first diagram the color- blind male is represented by a black eye ; the normal female by an eye without color. The offspring from SPECIAL CASES OF SEX-INHERITANCE 243 244 HEREDITY AND SEX two such individuals are normal, but the color blindness reappears in one-fourth of the grandchildren, and in these only in the males. The reverse mating is shown in the next diagram in which the female is color-blind. She will have color-blind sons and normal daughters (criss-cross inheritance), and all four kinds of grand- children. Other cases in man that are said to show sex-linked inheritance are atrophy of the optic nerve, multiple f , Up ■ ■ B ■ ■■ Fig. 118. — Pedigree of night blindness in a negro family, from Bordley. (After Davenport.) sclerosis, myopia, ichthyosis (Fig. 116), muscular atrophy (Fig. 117). Night blindness is described in certain cases as sex-linked ; in other cases, however, a disease by the same name appears to be a simple domi- nant and not sex-linked (Fig. 118). All these cases of sex-linked inheritance in man are explained by the assumption that the factor that produces these characters is carried by the sex chromo- some, which is duplex (XX) in the female and simplex (X) in the male. A simpler assumption has not yet been found. If one is fastidious and objects to the SPECIAL CASES OF SEX-INHERITANCE 245 statement of factors being carried by chromosomes, he has only to say, that if the factors for the characters follow the known distribution of the sex chromosome, the results can be accounted for. The culmination of the evidence of sex-determina- tion in man is found in a study of the cell structure of the human race itself. Strange as it may seem, we have been longer in doubt concerning the number of chromosomes in man than in any other animal as extensively studied. Four conditions are responsible : (1) The large number of chromosomes present in man. (2) The clumping or sticking together of the chromo- somes. (3) The difficulty of obtaining fresh material. (4) The possibility that the negro race has half as many chromosomes as the white race. Two years ago Guyer announced the discovery that in all probability there exist in man two unpaired chromosomes in the male (Fig. 119) that behave in all respects like that in the typical cases of the sort in insects, where, as we have seen, there are two classes of spermatozoa, differing by the addition of one more chromosome in one class. These produce females ; the lacking class produces males. But Guyer's evidence was not convincing. He found in all 12 chromosomes in one class of sperm and 10 in the other. Mont- gomery has also studied the same problem, but his account, while confirming the number, is in disagree- ment in regard to the accessory. Jordan has gone over a number of other mammals, in some of which he thinks that he has found indica- tions at least of two classes of sperm. Still more recently another investigator, von Wini- 246 HEREDITY AND SEX warter, has attacked the problem (Fig. 120). His material and his methods appear to have been superior to those of his predecessors. His results, while stated with caution and reserve, seem to put the whole question on a safer basis. His main results are illustrated in the diagram m ■ -•■' MA: Ml J» <* •t * »«* ■ ■' • - ▼;. \-*r>. $5 4 - TO|r Fig. 119. — Human spermatogenesis according to Guyer. The sex chromosomes are seen in 6-9. (Fig. 120). In the male he finds 47 chromosomes. Of these 46 unite at reduction to give 23 double chromosomes — one remains without a mate. At the first reduction division the pairs separate, 23 going to each pole, the unpaired chromosome into one cell only. SPECIAL CASES OF SEX-INHERITANCE 247 At the next division all the chromosomes in the 23 group divide, likewise all in the 24 group divide. There are produced two spermatozoa containing 24 /Tt« /«j W.1 -% fi • ? h. t 5 7? ,'v • > * * j k 1 m * Fig. 120. — Human spermatogenesis according to von Winiwarter, a, spermatogonial cell with duplex number; b, synapsis ; c, d, e,f, first spermato- cytes with haploid number of chromosomes ; g, first spermatocyte division, sex chromosomes (below) in advance of others ; h, two polar plates of later stage ; i, first division completed ; j, second spermatocyte with 23 chromo- somes ; k, second spermatocyte with 24 chromosomes ; /, second spermato- cyte division; m, two polar plates of later stage. 248 HEREDITY AND SEX chromosomes, and two containing 23 chromosomes; all four sperms having come from the same spermato- gonial cell (Fig. 121). In the female von Winiwarter had difficulty in deter- mining the number of chromosomes present. His Sex tleternu nation in /linn (Wtnitrax&r.) t!\ ..•• A A S,«. /■iff I '<',' ""l,,,'' 2 4 "•Hmmjwf*^ : •• m>y a*x \\o B £3 D Wife .... S3 Fig. 121. — Diagram of human spermatogenesis. A, spermatogonial cell with 47 chromosomes; B, first spermatocyte with reduced haploid number and sex chromosome (open circle) ; C, first division ; D, two resulting cells = second spermatocytes; E, division of second spermatocytes; F, four resulting spermatozoa, two female-producing (above), two male-produc- ing (below). best counts gave 48 chromosomes for the full or duplex number. These observations fit in with the results from the male. If these observations are confirmed, they show that in man, as in so many other animals, an internal mechanism exists by which sex is determined. It is futile then to search for environmental changes that SPECIAL CASES OF SEX-INHERITANCE 249 might determine sex. At best the environment may slightly disturb the regular working out of the two possible combinations that give male or female. Such disturbances may affect the sex ratio but have nothing to do with sex-determination. BIBLIOGRAPHY Andrews, E. A., 1895. Conjugation in an American Crayfish. Am. Nat., XXIX. Andrews, E. A., 1904. Breeding Habits of Crayfish. Am. Nat. XXXVIII. Andrews, E. A., 1910. Conjugation in the Crayfish Cambarus affinis. Jour. Exp. Zool., IX. Arkell, T. R., 1912. Some Data on the Inheritance of Horns in Sheep. N. H. Agr. Exp. Sta. Bulletin, 160. Arkell, T. R., and C. B. Davenport, 1912. The Nature of the Inheritance of Horns in Sheep. Science, N. S., XXXV. Arkell, T. R., and C. B. Davenport, 1912. Horns in Sheep as a Typical Sex-limited Character. Science, N. S., XXXV. von Baehr, W. B., 1908. Ueber die Bildung der Sexualzellen bei d. Aphididte. Zool. Anz., XXXIII. von Baehr, W. B., 1909. Die Oogenese bei einigen viviparen Aphiden und die Spermatogenese von Aphis saliceti. Arch. f. Zellf., III. von Baehr, W. B., 1912. Contribution a l'etude de la caryo- kinese somatique, de la pseudoreduction et de la reduction. La Cellule, XXVII. Baitsell, G. A., 1912. Experiments on the Reproduction of the Hypotrichous Infusoria. Jour. Exp. Zool., XIII. Baltzer, F., 1908. Uber die Grosse und Form der Chromosomen von Seeigeleiern. Verh. d. deutsch. zool. Ges., 1908. Baltzer, F., 1909. Die Chromosomen von Strongylocentrotus lividus und Echinus microtuberculatus. Arch. f. Zellf., II. Baltzer, F., 1910. Ueber die Beziehung zwischen dem Chromatin und der Entwicklung und Vererbungsrichtung bei Echinoder- menbastarden. Arch. f. Zellf., V. Baltzer, F., 1911. Zur Kentnis der Mechanik der Kernteilungs- figuren. Arch. Entw.-mech. f. d. Org., XXXIII. Bateson, W., 1909. Mendel's Principles of Heredity. Cam- bridge. 251 252 BIBLIOGRAPHY Bateson, W., 1912. Biological Fact and the Structure of Society. Cambridge. Bateson, \Y., 1913. Menders Principles of Heredity. 3d im- pression. Cambridge. Bateson, \\\, and R. C. Punnett, 1905. Rep. Evol. Corrtm., II. Bateson. \V., and R. C. Punnett, 1908. The Heredity of Sex. Science, XXVII. Bateson, W., and R. C. Punnett, 1911. On the Interrelations of Genetic Factors. Proc. Roy. Soc, LXXXIV. Bateson, W., and R. C. Punnett, 1911. The Inheritance of the Peculiar Pigmentation of the Silky Fowl. Jour. Gen., I. Bateson, W., and R. C. Punnett, 1911. On Gametic Series In- volving Reduplication of Certain Terms. Jour. Gen., I. Baumgartner, W. J., 1912. Observations on the Gryllidse — IV. Copulation. Kans. Univ. Sci. Bull., V. Baur, E., 1911. Ein Fall von Faktorenkoppelung bei Antir- rhinum ma jus. Verh. naturf. Ver. Briinn, XLIX. Baur, E., 1912. Vererbungs- und Bastardierungsversuche mit Antirrhinum — II. Faktorenkoppelung. Zeits. Abst. u. Ver- erb., VI. Baur, E., 1912. Ein Fall von geschlechtsbegrenzter Vererbung bei Melandrium album. Zeits. Abst. u. Vererb., VIII. Beard, J., 1902. The Determination of Sex in Animal Develop- ment. Zool. Jahrb. Abt. Anat., XVI. Beddard, F. E., 1906. Proc. Zool. Soc. London, 1906, II, p. 533. van Beneden, Ed., and Ch. Julin, 1884. .La spermatogenese chez 1'ascaride megalocephale. Bull, de Vacad. roy. de Bel- gique. Annee 53, 3me Ser., T. VII. Berry, E. H., 1906. The "Accessory Chromosome" in Epeira. Biol. Bull, XL Blackman, M. W., 1905. The Spermatogenesis of Myriapods. III. Bull. Mus. Comp. Zool, Harvard College, XLVHL Blakeslee, A. F., 1904. Sexual Reproduction in the Mucorineae. Proc. Amer. Acad. Arts and Sci., XL. Blakeslee, A. F., 1904. Zj'gospore Formation a Sexual Process. Science, XIX. Blakeslee, A. F., 1906. Differentiation of Sex in Thallus Game- tophyte and Sporophyte. Bot. Gaz., XLII. BIBLIOGRAPHY 253 Blakeslee, A. F., 1908. Sexual Condition in Fegatella. Bot. Gaz., XLVI. Boring, A. M., 1909. A Small Chromosome in Ascaris megalo- cephala. Arch. f. Zellf., IV. Boveri, Th., 1907. Zellen Studien. Die Entwicklung dispermer Seeigel-Eier, etc. Jena. Boveri, Th., 1908. Ueber die Beziehung des Chromatins zur Geschlechtbestimmung. Sitz. Phys.-Med. Gesell. Wiirzbwg. Dez., 1908. Boveri, Th., 1909. Die Blastomerenkerne von Ascaris megalo- cephala und die Theorie der Chromosomen-Individualitat. Arch, Zellf., III. Boveri, Th., 1909. Uber " Geschlechtschromosomen " bei Ne- matoden. Arch, Zellf., IV. Boveri, Th., 1911. Ueber das Verhalten der Geschechtschro- mosomen bei Hermaphroditismus. Verh. Phys.-Med. Gesell. Wurzburg, XLI. Brauer, A., 1893. Zur Kenntniss der Spermatogenese von Ascaris megalocephala. Arch, mikr. Anat., XLII. Bresca, G., 1910. Experimentelle Untersuchungen iiber die sekundaren Sexualcharaktere der Tritonen. Arch. Entw.- mech., XXIX. Brewster, E. T., 1906. Note on the Determination of Sex in Man. Am. Anthropol, VIII. Bridges, C. B., 1913. Partial Sex-Linkage in the Pigeon. Science, XXXVII. Buchner, Paul, 1911. Ueber hermaphrodite Seesterne. Zool Anz., XXXVIII. Bugnion, E., 1910. Les Cellules Sexuelles et la Determination du Sexe. Bull. Soc. Vaudoise Sci. Nat., LXVI. Burgeff, H., 1913. Ueber Sexualitat, Variability und Vererbung bei Phycomyces nit ens. Ber. deutsch. Bot. Gesell, X. Calkins, G. N., 1902. Studies on the Life-History of Protozoa. III. Biol. Bull, III. Calkins, G. N., 1904. Studies on the Life-History of Protozoa. IV. Jour. Exp. Zool, I. Calkins, G. N., 191 1 . Protozoon Germ Plasm. Pop. Sci. Monthly. Calkins, G. N., 1913. Further Light on the Conjugation of Para- mecium. Proc. Soc. Exp. Biol. Med., X. 254 BIBLIOGRAPHY ( ' a i. ki\s, (1. X., and C. C. Lieb, 1902. Studies on the Life-History of Protozoa, II. Arch. Protist,, 1902. Carmichael, E. S., and F. H. A. Marshall, 1908. On the Oc- currence of Compensatory Hypertrophy in the Ovary. Jour. Physiol. XXXVI. Castle, W. E., 1903. The Heredity of Sex. Bull, Mus. Comp. Zool,, Harvard College, XL. ( ' \st:.e, \Y. E., 1909. A Mendelian View of Sex Heredity. Science, XXIX. Castle, W. E., 1910. The Effect of Selection upon Mendelian Characters Manifested in One Sex only. Jour. Exp. Zool., VIII. Castle, W. E., 1911. Double-mating of Silkworm Moths. Science, XXXIII. Castle, W. E., 1911. Heredity in Relation to Evolution and Animal Breeding. Xew York. Castle, W. E., 1912. Are Horns in Sheep a Sex-limited Character ? Science, XXXV. Castle, W. E., 1913. Simplification of Mendelian Formulae. Am. Nat., XLVII. Castle, W. E., F. W. Carpenter, A. H. Clark. S. O. Mast, and W. M. Barrows, 1906. The Effects of Inbreeding, Cross- breeding, and Selection upon the Fertility and Variability of Drosophila. Proc. Am, Acad. Arts and Sci., XLI. Castle, W. E., and J. C. Phillips, 1909. A Successful Ovarian Transplantation in the Guinea-pig, etc. Science, XXX. Chidester, F. E., 1911. The Mating Habits of Four Species of the Brachyura. Biol. Bull., XXI. Chopard, L., 1912. XTote sur un cas de gynandromorphisme chez Forficula auricularia. Arch. Zool, Exp. et Gen., Ser. 5, T. X. Cole, L. J., 1912. A Case of Sex-linked Inheritance in the Do- mestic Pigeon. Science, XXXVI. Collins, G. N., 1912. Gametic Coupling as a Cause of Correla- tions. Am. Nat., XLVI. Colton, H. S., 1912. Lymnsea columella and Self-fertilization. Proc. Acad. Nat, Sci, Phil. Conklin, E. G., 1906. Sex-differentiation in Dinophilus. Science, XXIV. BIBLIOGRAPHY 255 Cook, M. H., 1910. Spermatogenesis in Lepidoptera. Proc. Acad. Nat. Sci. Phil. Correns, C, 1907. Zur Kenntnis der Geschlechtsformcn poly- gamer Blutenpflanzen und ihrer Beeinflussbarkeit. Jahrb. f. Wiss. Bot., XLIV. Correns, C, 1907. Die Bestimmung und Vererbung des Ge- schlechts. Berlin. Correns, C, 1908. Weitere Untersuchungen uber die Geschlechts- formen polygamer Blutenpflanzen und ihre Beeinflussbarkeit. Jahrb. f. wiss. Bot., XLV. Correns, C, 1908. Die Rolle der Mannlichen Keimzellen bei der Geschlechtsbestimmung der gynodioecischen Pflanzen. Ber. deutsch. Bot. Ges., XXVI. Correns, C, 1908. Weitere Untersuchungen ueber die Ge- schlechtsformen polygamer Blutenpflanzen und ihre Beein- flussbarkeit. Jahrb. f. wiss. Bot., XLV. Correns, C, 1912. Selbststerilitat und Individualstoffe. Fests. Med. Nat. Gesell. 84. Versam. dent. Natur. Aerzte. Correns, C, and R. Goldschmidt, 1913. Die Vererbung und Bestimmung des Geschlechts. Gesamtsitz. \ naturw. mediz. Hauptg. 84. Vers, deuts. Naturf. u. Aerzte in Munster. Cue'not, L., 1899. Sur la determination du sexe chez les animaux. Bull. Scient. France Belg., XXXII. Cunningham, J. T., 1908. The Heredity of Secondary Sexual Characters in Relation to Hormones. Arch. f. Entw.-mech., XXVI. Darling, C. A., 1909. Sex in Dioecious Plants. Bull. Tor. Bot. Club, XXXVI. Darwin, C, 1886. The Descent of Man, and Selection in Relation to Sex. New York. Darwin, C, 1889. Different Forms of Flowers on Plants of the Same Species. New York. Darwin, C, 1895. Cross and Self-Fertilization in the Vegetable Kingdom. New York. Davenport, C. B., 1906. Inheritance in Poultry. Carnegie Inst. Wash., pub. 53. Davenport, C. B., 1909. Inheritance of Characteristics in Do- mestic Fowl. Carnegie Inst. Wash., pub. 121. 250 BIBLIOGRAPHY 1 ).\\ export, C. B., 1911. Heredity in Relation to Eugenics. New- York. Davenport, C. B., 1911. Another Case of Sex-limited Heredity in Poultry. Proc. Soc. Exp. Biol. Med., IX. Davenport, C. B., 1912. Sex-limited Inheritance in Poultry. Jour. Exp. Zool., XIII. Dederer, P. H., 1907. Spermatogenesis in Philosamia cynthia. Biol. Bull., XIII. Delage, Y., 1908. Les ldees nouvelles sur la Parthenogenese experimentale. La Revue des ldees, 50. Delcourt, A., and E. Guyenot, 1911. Genetique et Milieu. Bull. Scient. France Belg., XLV. Della Valle, P., 1913. Die Morphologie des Zellkerns und die Physik der Kolloide. Zeits. Chem. Industr. Kolloide, XII. Dexter, J. S., 1912. On Coupling of Certain Sex-linked Char- acters in Drosophila. Biol. Bull., XXIII. Dodds, G. S., 1910. Segregation of the Germ-Cells of the Teleost, Lophius. Jour. Morph., XXI. Doncaster, L., 1907. Inheritance and Sex in Abraxas grossu- lariata. Nature, LXXVI. Doncaster, L., 1907. Garnet ogenesis and Fertilization in Nema- tus ribesii. Q. J. M. S., LI. Doncaster, L., 1908. On Sex-inheritance in the Moth, Abraxas grossulariata and its var. lacticolor. 4th Rep. Evol. Comm., R. Soc. Lond. Doncaster, L., 1908. Animal Parthenogenesis. Sci. Progr. Doncaster, L., 1909. Recent Work on the Determination of Sex. Sci. Progr., XIII. Doncaster, L., 1910. Gametogenesis of the Gall-fly, Neuroterus lenticularis (Spathegaster baccarum). Part I. Proc. Roy. Soc, B, LXXXII. Doncaster, L., 1911. Gametogenesis of the Gall-fly, Neuroterus lenticularis. Part II. Proc. Roy. Soc, LXXXIII. Doncaster, L., 1911. Note on the Inheritance of Characters in which Dominance appears to be Influenced by Sex. Jour. Genet., I. Doncaster, L., 1911. Some Stages in the Spermatogenesis of Abraxas grossulariata and its var. lacticolor. Jour. Genet., I. BIBLIOGRAPHY 257 Doncaster, L., 1912. The Chromosomes in the Oogenesis and Spermatogenesis of Pieris brassicse, and in the Oogenesis of Abraxas grossulariata. Jour. Genet, II. Doncaster, L., 1912. Note on the Chromosomes in Oogenesis and Spermatogenesis of the White Butterfly, Pieris brassicse. Proc. Camb. Phil. Soc, XVI. Doncaster, L., and G. Gray, 1911. Cytological Observations on Cross-fertilized Echinoderm Eggs. Proc. Camb. Phil. Soc, XVI. Doncaster, L., and F. H. A. Marshall, 1910. The Effects of One-sided Ovariotomy on the Sex of the Offspring. Jour. Genet., I. Doncaster, L., and G. H. Raynor, 1906. Breeding Experiments with Lepidoptera. Proc. Zool. Soc. Lond. Drew, G. A., 1911. Sexual Activities of the Squid, Loligo Pealii (Hes.). I. Copulation, Egg-laying and Fertilization. Jour. Morph., XXII. Durham, F. M., 1911. Further Experiments on the Inheritance of Coat Colour in Mice. Jour. Genet., I. Durham, F. M., and D. E. C. Marryat, 1908. Note on the In- heritance of Sex in Canaries. Ifih Rep. Evol. Cornm., Roy. Soc. Lond. East, E. M., 1910. Notes on an Experiment concerning the Na- ture of Unit Characters. Science, XXXII. East, E. M., 1910. The Role of Hybridization in Plant Breeding. Pop. Sci. Mon., Oct., 1910. East, E. M., 1911. The Genotype Hypothesis and Hybridization, Am. Nat., XLV. East, E. M., 1912. A Study of Hybrids Between Nicotiana bigelovii and N. quadrivalvis. Bot. Gaz., LIII. East, E. M., and H. K. Hayes, 1911. Inheritance in Maize. Conn. Exp. Sta, Bull, 167. East, E. M., and H. K. Hayes, 1912. Heterozygosis in Evolution and in Plant Breeding. U. S. Dept. Agric, Bureau Plant Ind. Bull. 243. Edwards, C. L., 1910. The Sex-determining Chromosomes in Ascaris. Science, XXXI. Emerson, R. A., 1909. Inheritance of Color in the Seeds of the 258 BIBLIOGRAPHY Common Bean, Phaseolus vulgaris. Ann. Rep. Nebr. Agr. Exp. Sta., XXII. Emerson, R. A., 1911. Morgan on Coupling vs. Random Segrega- tion. Science, XXXI V. Emerson, R. A., 1911. Genetic Correlation and Spurious Alle- lomorphism in Maize. Ann. Rep. Arebr. Agr. Exp. Sta., XXIV. Federley, H., 1911. Yererbungsstudien an der Lepidopteren- Gattung Pygaera. Arch. Rass. Gesell., III. Fere, C, 1898. Experiences relatives aux rapports homosexuels chez les hammetons. C. R. Soc. Biol., V. Fere, C, 1898. Experiences relatives a l'instinct sexuel chez le bombyx du murier. C. R. Soc. Biol., V. Fick, R., 1907. Uber die Vererbungssubstanz. Arch. Anat. u. Phys., Anat. Abt. Foot, K., and E. C. Strobell, 1913. Preliminary Note on the Results of Crossing Two Hemipterous Species, etc. Biol. Bull., XXIV. Fowler, G. H., 1894. Notes on some Specimens of Antlers of the Fallow Deer, etc. Proc. Zool. Soc. Lond., 1894. Frolowa, S., 1912. Idiochromosomen bei Ascaris megalocephala. Arch. Zellf., IX. Geddes, P., and J. A. Thomson, 1889. The Evolution of Sex. London. Gerould, J. H., 1911. The Inheritance of Polymorphism and Sex in Colias philodice. Am. Nat., XLV. Goldschmidt, R., 1911. Einfiihrung in die Vererbungswissen- schaft. Leipzig. Goldschmidt, R., 1912. Erblichkeitsstudien an Schmetterlingen. I, 1. Zeits. Abst. Vererb., VII. Goldschmidt, R., 1912. Bemerkungen zur Vererbung des Ge- schlechtspolymorphismus. Zeits. Abst. Vererb., VIII. Goldschmidt, R.. 1913. Zuchtversuche mit Enten. I. Zeits. Abst. Vererb., III. Goodale, H. D., 1909. Sex and its Relation to the Barring Factor in Poultry. Science, XXIX. Goodale, H. D., 1910. Breeding Experiments in Poultry. Proc. Soc. Exp. Biol. Med., VII. BIBLIOGRAPHY 259 Goodale, H. D., 1910. Some Results of Castration in Ducks. Biol, Bull, XX. Goodale, H. D., 1911. Studies on Hybrid Ducks. Jour. Exp. Zool., X. Goodale, H. D., 1911. Sex-limited Inheritance and Sexual Di- morphism in Poultry. Science, XXXIII. Goodale, H. D., 1913. Castration in Relation to the Secondary Sexual Characters of Brown Leghorns. Am. Nat., XLVII. Gregory, R. P., 1911. Experiments with Primula sinensis. Jour. Genet., I. Gregory, R. P., 1911. On Gametic Coupling and Repulsion in Primula sinensis. Proc. Roy. Soc, LXXXIV. Gross, J., 1912. Heterochromosomen und Geschlechtsbestimmung bei Insekten. Zool. J ahrb., XXXII. Gruenberg, B. C, 1912. Physiological Sex-determination. Science, N. S., XXXV. Gudernatsch, J. F., 1911. Hermaphroditismus verus in Man. Am. Jour. Anat., XL Gudernatsch, J. F., 1912. Feeding Experiments on Tadpoles. I. Arch. Entw.-mech., XXXV. Gudernatsch, J. F., 1912. Futterungsversuche an Amphibien- larven. Zentrbl. f. Physiol., XXVI. Gulick, A., 1911. Ueber die Geschlechtschromosomen bei einigen Nematoden. Arch. f. Zellf., VI. Guyer, M. F., 1902. Hybridism and the Germ Cell. Univ. of Cincinnati Bull., No. 21. Guyer, M. F., 1903. The Germ Cell and the Results of Mendel. Cincinnati Lancet-Clinic. Guyer, M. F., 1905. Guinea-Chicken Hybrids. Science, N. S., XXL Guyer, M. F., 1905. Notes on Cross-bred Chickens. Science, N. S., XXI. Guyer, M. F., 1907. Do Offspring Inherit Equally from Each Parent ? Science, N. S., XXV. Guyer, M. F., 1909. Atavism in Guinea-Chicken Hybrids. Jour. Exp. Zool, VII. Guyer, M. F., 1909. La Livree du Plumage chez les Hybrides de Pintade et de Poule. Bull. Mus. d'hist. nat. 260 BIBLIOGRAPHY (.Iyer, M. P., 1909. The Spermatogenesis of the Domestic Chicken. Anat. Am., XXXIV. Guyer, M. F., 1909. The Spermatogenesis of the Domestic Guinea. Anat. Am., XXXIV. Guyer, M. F., 1909. Deficiencies of the Chromosome Theory of Heredity. University Studies (Cincinnati), Series II, Vol. V. Guyer, M. F., 1909. On the Sex of Hybrid Birds. Biol. Bull, XVI. Guyer, M. F., 1910. Accessory Chromosomes in Man. Biol. Bull, XIX. Guyer, M. F., 1911. Nucleus and Cytoplasm in Heredity. Am. Nat., XLV. Hadley, P. B., 1910. Sex-limited Inheritance. Science, XXXII. Hagedoorn, A. L., 1909. Mendelian Inheritance of Sex. Arch. f. Entw.-mech., XXVIII. Hagedoorn, A. L., 1912. The Genetic Factors in the Develop- ment of the House Mouse, etc. Zeits. Abst. Vererb., VI. Hartley, C. P., E. B., Brown, C. H., Kyle, and L. L. Zook, 1912. Cross-breeding Corn. U. S. Dept. Ayr. Bull., 218. Hartman, F. A., 1913. Variations in the Size of Chromosomes. Biol. Bull, XXIV. Hasper, M., 1911. Zur Entwicklung der Geschlechtsorgane von Chironomus. Zool. Jahrb., Abt. Anal, XXXI. Heape, W., 1897. Further Note on the Transplantation and Growth of Mammalian Ova within a Uterine Foster-mother. Proc. Roy. Soc, LXII. Heape, W., 1899. Note on the Fertility of Different Breeds of Sheep. Proc. Roy. Soc, LXV. Heape, W., 1905. Ovulation and Degeneration of Ova in the Rabbit. Proc. Roy. Soc, B, LXXVI. Heape, W., 1907. Notes on the Proportion of the Sexes in Dogs. Proc Camb. Phil Soc, XIV. Heape, W., 1907. Note on the Influence of Extraneous Forces upon the Proportion of the Sexes produced by Canaries. Proc Camb. Phil Soc, XIV. Heape, W., 1908. Note on Russo's Attempt to show Differen- tiation of Sex in the Ovarian Ova of the Rabbit. Proc. Camb. Phil Soc, XIV. BIBLIOGRAPHY 261 Heape, W., 1908. The Proportions of the Sexes produced by Whites and Colored Peoples in Cuba. Phil. Trans. Roy. Soc. Lorul., Series B; Vol. 200. Heider, K., 1906. Vererbung und Chromosomen. Jena. Henking, H., 1891. Untersuchungen ueber die ersten Entwick- lungsvorgange in den Eiern der Insekten. Zeits. f. wiss. Zool., LI. Herbst, C, 1909. Vererbungstudien VI. Arch. Entw.-mech., XXVII. Hertwig, O., 1890. Vergleich der Ei und Samenbildung bei Nematoden. Arch. mikr. Anat., XXXVI. Hertwig, R., 1906. Ueber Knospung und Geschlechtsentwicke- lung von Hydra fusca. Festschr. f. J. Rosenthal. Leipzig. Hertwig, R., 1907. Uber den Chromidialapparat und den Dualismus der Kernsubstanzen. Sitz. her. Gesell. Morph. und Phijs. in Munchen. 1907. Hertwig, R., 1907. Untersuchungen tiber das Sexualitatsproblem. III. Teil. Verh. Deutsch. Zool. Gesell. Hertwig, R., 1912. Uber den derzeitigen Stand des Sexualitats- problems nebst eigenen Untersuchungen. Biol. Centr., XXXII. Holding, R. E., 1905. Exhibition of, and Remarks upon, a Series of the First Year Antlers of Certain Deer. Proc. Zool. Soc. Lond., II. Holmes, S. J., 1903. Sex Recognition among Amphipods. Biol. Bull, V. Janda, V., 1912. Die Regeneration der Geschlechtsorgane bei Criodrilus lacuum. Arch. Entw.-mech., XXXIV. Jannsens, F. A., 1901. La spermatogenese chez les tritons. La Cellule, XIX. Jannsens, F. A., 1905. Evolution des Auxocytes males du Batra- coseps attenuatus. La Cellule, XXII. Jannsens, F. A., 1909. La theorie de la chiasmatypie. Nou- velle interpretation des cineses de maturation. La Cellule, XXV. Jannsens, F. A., and R. Dumez, 1903. L'element nuclemien pendant les cineses de maturation des spermatocytes chez Batracoseps attenuatus et Plethedon cinereus. La Cellule, XX. 262 BIBLIOGRAPHY Jennings, II. S., 1908. Eeredity, Variation and Evolution in Protozoa, II. Proc. Am. Phil. N«c, XLVII. Jennings, II. S., 1909. Heredity and Variation in the Simplest Organisms. Amer. X Pearl, R., 72, 212-213, 227 Pearse, 117 Peckham, 115-116, 120 Pellew. 212 Peltogaster, 158 Petrunkewitsch, 117 Phalarope, 112 Pheasants, 225 Phidippus, 34 Photinus, 28 Phylloxerans, 52, 54, 178, 179, 180, 181, 189 Pigeons, 32 Pituitary body, 146 Plutei, 60 Plymouth rock, 69-71, 212 Polar bodies, 37 Polytmus, 103 Porter, 117 Porthetria, 117, 148 Primula, 201, 202 Promethea, 116 Protenor, 40 Punnett, 127, 128, 138, 233 Rawls, 221 Rat, 140, 233 Reduplication, 100 Reindeer, 136 Rhabditis, 169, 224, 226 Riley, 232 Ritzema-Bos, 195 Rotifers, 185-189 Rudimentary wing, 214, 215 Russo, 234 Sacculina, 155 Sagitta, 21, 22 Schenk, 233 Schleip, 170, 171 Schultze, 233 Sclerosis, 242 Seabright, 143-144 Sea cow, 27 Sea-lion, Steller's, 110 Sea-urchin, 56-62 Segregation, 81, 100 Sex, 83, 84 Sex chromosome, 50, 80, 83, 84 Sex determination, 84 Sex-limited, 83 Sex-linked, 81, 83, 84, 132 Sheep, 134-138 Shull, A. F., 187, 197, 205 Shull, G. H., 173, 202, 204, 211 Shuster, 145 Siamese twins, 236 Silkworm, 117, 165 Sinety, 50 28 I INDEX Skeleton, rat, 1 10 Smith. <;.. 145, 155 Soule, 1 16 Sparrow, 2 Spermatophores, 25 Kpluerechiuus, 59-60 Spiders, 34, 107, 115, 117 Squid, 24 Stag, 133 Steinach, 1 10 Stephanosphsera, 5 Stevens, 51 Strobell, 151 Strongyloeentrotus, 59, 60, 62 Sturtevant, 72, 98, 117, 118 Stylonichia, 2 Suffolks, 136-138 Synapsis, 93 Tadpoles, 147 Tanager, scarlet, 27 Thomson, 232 Thymus, 146-147 Thyroid, 146-147 Toad, 229 Tower, 117 Toyama, 165 Treat, 232 Triton, 145 Trow, 99 Tschermak, 74 Vermilion eye, 119 Vestigial wing, 96-97 J Vigor, 120 Vincent, 146 de Vries, 74, 125 Wallace, 102, 113-114, 120, 125, 127 Wasp, 220 Weismann, 16, 17, 40, 194, 195 Wheeler, 117 White eye, 62-65, 81, 82, 88-92, 118, 119, 221-223 Whitney, 185, 187, 197, 205 Wilder, 237 Wilson, 51 Winiwarter, 245-248 Wood, 136 Woodruff, 8, 198 X-chromosome, 51, 82, 84, 242 Y-ehromosome, 51, 84 Yellow body color, 67, 88-92, 119 j ' BINDING LIZT JUl \ jc QH 431 M757 19U Morgan, Thomas Hunt Heredity and sex BioMed PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY BH m i ■ m H BHPH HMtlia m ■ Hi Hj ■J 11 ^ HI Hi una ■ IS ^H