Hymenopteron Paracopidosomopsis, which by polyembryonic division sometimes produces both males, females and inter- mediates. That something analogous to non-disjunction here occurs is an acceptable interpretation of this anomalous instance. But there is a remarkable group of incontrovertible facts which show that sometimes the normal course' of sex- development may be disturbed. F. Lillie discovered a remarkable example. When in horned cattle twins of opposite sexes occur, the female is sometimes sterile, being called a free-martin. It might be thought that these twins arose by division of one fertilized ovum, but Lillie by study of material from the Chicago stock-yards proved that an ovum had dehisced from each ovary and that therefore the twins were originally distinct. He further showed that sometimes the twins had an actual anastomosis between their foetal circulations. The presence of a male embryo must therefore be regarded as having the power of inhibiting the development of the female embryo, poisoning it in so far as the formation of the reproductive organs is concerned.
Effects of External Conditions. Disturbance of the normal sex-ratios as a consequence of various interferences, such as starvation, high or low temperature, etc., has several times been alleged to occur, especially in Amphibia, and the evidence of R. Hertwig that delay in fertilizing the eggs of the frogs causes the production of a preponderance of males has been fully confirmed. Circumstantial details preclude the obvious suggestion that this is a result of differential mortality. Hertwig thought that inas- much as the polar bodies are excluded very late, after the eggs are laid, his result might perhaps be reconciled with the concep- tion of cytological pre-determination if the conditions of the experiment could in some way have decided which elements should be retained in the egg and which ejected in the polar body. Recently Seiler has made an observation of this kind in regard to a Psychid moth, Talaeporia. He states that if the females are kept in a high temperature while the eggs are undergoing maturation, it can be seen on cytological examination that the accessory chromosome more frequently remains in the egg, and that as a matter of observation more males are produced, the converse occurring under cold conditions. Until we know definitely which sex in Amphibia is hetero-gametic the discussion can scarcely be carried further, but the frequency with which transitional forms are found in the Ichthyopsida raises a prob- ability that the facts are of a higher order of complexity. As to the genetical composition of the Amphibian intersexes, an impor- tant observation has lately been made by Crew. From the eggs of a female fertilized by a male showing intersexual characters 774 tadpoles were reared, all females. According to Baltzer, sex in Bonellia is directly determined by the conditions of larval life. The female is of course an animal of considerable size, whereas the male is a minute creature parasitic on her. Larvae which find the proboscis of the female and attach themselves to it are said to develop into males, those which remain free- swimming becoming females. Larvae artificially detached from the proboscis become intersexes. Adequate controls, by which the hypothesis of predisposition may be excluded, would in such an instance be most difficult to institute.
Effects of Castration. Collateral evidence bearing on the nature of the distinction between the sexes has been drawn from many sources, especially from the results of castration, but though the facts thus empirically observed are of much physiolog- ical interest, they have no very direct bearing on the primary problem. The reproductive glands, acting chiefly, if not entirely, by virtue of the secretions (hormones) of their interstitial components, have often a great influence on the development and maintenance of secondary sexual characters, but the part played by the hormones must obviously be of a secondary nature. Removal of the genital glands has divers effects in various animals. In good agreement with the discovery that in the bird the female is the hetero-gametic sex, it is found that both in the fowl and the duck (Goodale) the removal of the ovary induces the plum- age and some other characteristics of the male. The female is thus an organism in which the male attributes are concealed or recessive, whereas a capon does not develop hen-feathering.
Moreover, Morgan found that in the Sebright bantam, the males of which are " henny " in plumage (not in combs or wattles), after castration the cocks acquire ordinary male plumage, which may naturally be interpreted to mean that the hen-feathering of these cocks is due to their possession of part of the female complex which has been transferred to*them. Morgan and Pun- nett also have shown that the henny character behaves as dominant in breeding, a fact which proves that the dominance proper to the whole female complex of the bird pertains also to that part of the complex which controls the plumage.
Castration performed on moths during the larval stage has not produced modification of secondary sexual characters (Oudemans). In crabs, however, the destruction of the testes by certain parasites produces very striking " feminization " of the abdomen and appendages (Geoffrey Smith), but we do not know which sex in Crustacea is hetero-gametic. We are without a satis- factory interpretation of this group of observations, some of which superficially considered seem to run counter to the facts of gametic determination already established, but the disturbances of the normal course may commonly, though perhaps not always, be conceived of as due to interruption of the chain of events by which the full effects of gametic predetermination are developed.
Hermaphrodites. Attention should be called to a remarkable lacuna in our knowledge of sex-determination. Up to the present nothing has been yet discovered either by cytological or analyti- cal methods as to the genetical relation of the hermaphrodite types among animals to those in which the sexes are distinct. In plants a little progress has, as will appear, been made, pointing to the conclusion that the hermaphrodite is dominant, containing something which the females at least do not, but as to hermaphro- dite animals nothing can be said with confidence. This is much to be regretted, as the whole subject might be greatly advanced by such knowledge.
General considerations. From the observation that the two sexes are formed by modification of a common structural plan, and from the fact that by interferences, of which some have been mentioned, individuals cytologically, or at least gametically, destined to be of .one sex may be made to assume more or less of the characteristics of the other, the conclusion has often been announced that each sex contains the other latent in it. To this expression, in the light of modern knowledge, no precise meaning can be attached, and it probably conveys nothing of essential truth. The primary sexual distinctions evidently depend on factorial elements which follow more or less closely the familiar principles of Mendelian segregation. The determining factor for sex should probably be regarded as a complex, usually transmitted in its entirety, but capable by accidental errors of cell-division of being disintegrated, so that the elements responsible for special characteristics may become detached from the rest and may even be passed over to the sex which normally does not receive them. For example, the races of fowls which do not incubate have presumably thus lost a portion of the dominant sex-complex; the hen-like cocks of the Sebright bantam may be represented as having acquired that ingredient of the female sex-factor which in the normal female inhibits the formation of the sickle-feathers and hackles of the cock, and so on. A similar representation may be applied to those cases (e.g. Phalaropes) in which the cocks are hen-like and normally incubate the eggs. It is not unreasonable to suppose that the transference of an actual fragment of critical material, presumably a portion of a chromosome, is responsible for the physiological abnormality. The literature of obstetrics and of stock-breeding abounds with nostrums for the arbitrary regulation of sex in man and the domestic animals, but from what is accurately known of the mechanism of sex-determination, nothing favourable to these claims can yet be adduced. Nor can any explanation be offered of the fairly constant departures from equality which normally occur in man and various domesticated animals. In the pig, ox and rabbit the male births are said sensibly to exceed the female, but in the horse and sheep there is a small excess of female births. The races of man show definite differences in the proportions of the sexes at birth. Of living births, taking females at 1,000, the males for England and Wales
