It seems to be generally true that where sexual reproduc­tion leads only to the forma­tion of females the case finds its explana­tion in the fact that the male-producing spermatozoa perish and only the female-producing spermatozoa survive. Such an observa­tion was made by Morgan on a certain species of phylloxerans.

The slight preponderance in the number of one sex which is occasionally found—an excess of six per cent. males over females in the human race—may well find its explana­tion on the assump­tion of a slightly greater mortality of the female-determining spermatozoa.

In certain forms partheno­genetic and sexual reproduc­tion may alternate in a cycle, e. g., in plant lice, Daphnia, and rotifers. In plant lice it has been observed for a long time that when the plant is normal and the weather warm the aphides remain wingless, reproduce partheno­genetically, and only females exist, and this may last for years and for more than fifty genera­tions; but that when the plant is allowed to dry out both sexes appear.

Here we are dealing with a limited determina­tion of sex inasmuch as the experi­menter has it in his power to prevent or allow the produc­tion of males. The facts do not in all probability contradict the statements made concerning the rôle of the X chromo­somes in the determina­tion of sex. We have seen that where sex is determined by two types of spermatozoa one type of eggs is produced which possesses only one X chromo­some. Such eggs might produce males if not fertilized (as they do in bees), but they cannot produce females because for that purpose they must have two X chromo­somes. It has been shown for certain cases, and it may be true generally, that if eggs of this type give rise to partheno­genetic females they may do so because they have for some reason two X chromo­somes. Usually such an egg loses one of the X chromo­somes in a process of nuclear division (the so-called reduc­tion division) which usually precedes fertiliza­tion. If this reduc­tion division is omitted the egg has two X chromo­somes and if such an egg develops partheno­genetically it gives rise to a female. These cases do not, therefore, contradict the connec­tion between X chromo­somes and sex determina­tion established by cytological observa­tions and breeding experi­ments, on the contrary, they confirm it. The ques­tion remains: How can external condi­tions bring it about that the reduc­tion division is omitted? To this ques­tion no definite answer can be given at present.

We may in passing mention the well-known observa­tion that twins which originate from the same egg always have the same sex; while twins arising from different eggs show the usual varia­tion as to sex. Twins coming from one egg have the same chorion and can thereby be diagnosed as such. They can be produced as we have stated in Chapter V by a separa­tion of the first two cleavage cells of the egg, each one giving rise to a full embryo. It harmonizes with all that has been said above that the sex of two such individuals must be the same since they have the same number of X chromo­somes, the latter being determined in the human race by the nature of the spermato­zoön which enters the egg.

4. While thus far all the facts agree with the dominating influence of certain chromo­somes upon sex determina­tion, one group of facts has not yet been explained: namely, hermaph­ro­ditism. By hermaph­ro­ditism is meant the existence of complete and separate sets of female and male gonads in the same individual. This condi­tion exists regularly not only in definite groups of animals, e. g., certain snails, leeches, tape-worms, but also, as everybody knows, in flowering plants. While in some forms both kinds of sex cells, male and female, are formed and mature simultaneously, as, e. g., in the Ascidian Ciona (see Chapter IV), in others they are formed successively, very often the spermatozoa appearing first (protandric hermaph­ro­ditism). In the long tapeworm Tænia each ring has testes and ovaries, but the young rings are only male while in the older rings the testes disappear and the ovaries are formed. The same ring is in succession male and female. How can we reconcile the facts of hermaph­ro­ditism with the chromo­some theory of sex determina­tion? Rhabdo­nema nigro­venosum, a parasite living in the lungs of the frog, is hermaph­ro­ditic, but its eggs produce not a hermaph­ro­ditic genera­tion but one with the two separate sexes; this genera­tion is not parasitic and lives in the soil. The genera­tion produced by these separate males and females gives rise again to a hermaph­ro­dite which migrates into the lungs of the frogs. According to Boveri and Schleip[188] the cells of the hermaph­ro­dite have twelve chromo­somes. It produces two types of spermatozoa with six and five chromo­somes respectively (one-half of the cells losing one chromo­some which is left at the line of cleavage between the two cells); and one type with six chromo­somes. In this way separate males and females are produced by the hermaph­ro­dite, females with twelve and males with eleven chromo­somes.

The males produce again two kinds of spermatozoa, male and female producing, but the male-producing spermatozoa become func­tionless. This fusion of the other spermato­zoön containing six chromo­somes with an egg having six chromo­somes leads again to the forma­tion of the herm­aph­rodite with twelve chromo­somes. It is obvious that in this case the cause for the herm­aph­roditism is not disclosed. If chromo­somes have anything to do with hermaph­ro­ditism there must be an undiscovered element in the chromo­somes which may explain why the female as well as the herm­aph­rodite have the same chromo­some constitu­tion; or we are forced to look for another determinant outside the X chromo­somes or the chromo­somes altogether. This seems to be the only cytological work on the problem of hermaph­ro­ditism. Experimental work has been begun by Correns[189] and by Shull on the determina­tion of hermaph­ro­ditism in plants but lack of space forbids us to give details.

II. The Physiological Basis of Sex Determina­tion

5. As stated at the beginning of this chapter, the chromo­some theory of sex determina­tion explained only one feature of the problem, namely, the relative numbers in which both sexes or only one sex, as the case may be, are produced; and in this respect the evidence is so complete that we must accept it. But with all this, the problem of sex determina­tion is not exhausted, since a physio­logical solu­tion of the problem of sex determina­tion demands an account of how the sex chromo­somes can induce the forma­tion not only of ovaries and testes but also of the other sex characters. For the solu­tion of this problem biology will have to depend largely on experi­ments in which it is possible to influence the forma­tion of sex characters and of the sex glands themselves.

The most striking observa­tions in this direc­tion were made by Baltzer on a marine worm, Bonellia. In this animal the two sexes are very different, the male being a tiny parasite, a few millimetres in length, which spends its life in the uterus of the female, whose size is about five centimetres. A female carries as a rule several and often a large number of the male parasites in its uterus, which indicates that the males prevail numerically. The fertilized eggs of the animals are laid in the sea water where the larvæ hatch. At the time of hatching all larvæ are alike. The differentia­tion of the larvæ into the dwarf males and the giant females can be determined at will. The larvæ have a tendency to attach themselves to the proboscis of the female as soon as they hatch. If given a chance to do so and if they stick to the proboscis for more than three days they will develop into males, which soon afterwards creep into the female where they continue their parasitic existence. If, however, no adult female Bonellia is put into the aquarium in which the larvæ hatch, about ninety per cent. of the larvæ will, after a period of rest, develop into females; the rest develop into males. Those which develop into females will often show a primary maleness which may manifest itself in the produc­tion of sperm or of other secondary male sexual characters. This tendency is stronger the longer the period of rest lasts. If the larvæ are allowed to settle on the proboscis of the adult female but are removed too early hermaph­ro­dites are produced having male and female characters mixed.