The Dynamics of Living Matter

We do not know which circumstances in the sexual cells determine the hereditary effects, although one would naturally think first of definite chemical compounds as the bearers of hereditary qualities. The greater part of the spermatozoon, namely, the head, consists of a salt whose acid is nuclcinic acid, whose base in some fishes and starfish is pro- tamine, in other forms histones, which latter, however, are closely related to the protamines. In order to decide whether the nucleins or the histones or the protamines are of importance for the hereditary qualities, it would be necessary to decide whether the nuclei of the eggs of one form contain always the same base as that found in the sperm of the same species. This should be expected from the fact that the hereditary influence of egg and sperm is equal in the adult offspring, at least. It seems that the base is not always identical in the egg and sper- matozoon of the same species, and this seems to indicate that the nucleic acid is of more importance for heredity than protamines and histones. Aside from the nuclein we find albumin and globulin, es- pecially in the tail, and in the latter also lecithin, cholesterin, and fat. Miescher believed that in the head of the spermatozoon an iron compound exists. It is impossible to draw any far-reaching inference concerning the nature of the substances which transmit hereditary qualities from these meager data.

The fact that the spermatozoon contributes just as much to the trans- mission of hereditary qualities in the adult as the egg, although the mass of the latter is, as a rule, many times larger than that of the spermato- zoon, makes it certain that only a small fraction of the contents of the egg has anything to do with this transmission of hereditary qualities. Since the head seems to be the more important part of the spermatozoon for the process of heredity, and this head is a homologue of the egg nucleus, Boveri expressed the idea that the nucleus, and not the proto- plasm, is the really significant part of the egg in matters of heredity. In order to test this idea he undertook a very ingenious experiment; namely, the fertilization of an enucleated fragment of the egg of one species of the sea urchin by the sperm of another species. If his view were correct, such a hybridization should produce a larva with purely paternal characteristics, as the egg only furnished the protoplasm which was not expected to influence the hereditary qualities. The execution of this experiment is extremely difficult. Boveri is inclined to believe that, according to the experiments carried out so far, the fertilization of enucleated fragments of eggs of Sphtereckinus by the sperm of Echi- nus yields plutei of a pure type of Echinus, although he does not consider the question as definitely settled.* I am inclined to believe that in the early stages the paternal influence would, at the best, be very slight.

The egg protoplasm contains more or less reserve material which is only gradually transformed into the characteristic compounds of the embryo. It is therefore obvious enough that at first the embryo must show effects of this relation. When the protoplasm of the egg possesses a striking pigment, the larva will possess the same for some time at least ; if such an egg is hybridized with the sperm of a form whose egg is unpigmented, the larva will, of course, possess a "maternal" quality which is due solely to the protoplasm (Driesch). In the eggs of birds the incubation period depends ceteris paribus upon the mass of yolk. When a species with a long incubation period is crossed with one of a short incubation period, the egg, and not the sperm, determines the incubation period, as Whitman observed in pigeons. It is obvious, then, that during the first stages of development an influence of the protoplasm upon heredity may make itself felt, which will disappear as soon as the protoplasm of the egg has been transformed into the tissues of the embryo. It does not seem to me that a discussion as to the rela- tive influence of protoplasm and nucleus upon heredity will prove very fertile, but that it is necessary to transfer this problem as soon as possible from the field of histology to that of chemistry or physical chemistry. This view is supported by investigations concerning the toxic effects of blood of one form upon not too closely related forms.

Until about thirty years ago the idea was held generally that a trans- fusion of the blood of an animal into the veins of a human being was permissible or advisable in the case of severe loss of blood. We know to-day that in such cases physiological salt solutions or human blood must be injected, and that the blood of animals is generally toxic. This important discovery was made by Landois,* who showed that blood of a foreign species generally destroys the red corpuscles of the animal into which it is infused. He investigated systematically the destructive force of foreign blood upon the red corpuscles of various animals, and made the remarkable discovery that there exists a striking relation be- tween this effect and the blood relationship of animals. I will quote the summary of this part of his investigation: "My results include a point which is of importance for the systematic order of animals; namely, that those animals which are closest to each other in regard to their anatomical qualities also possess the most homogeneous blood, inasmuch as a transfusion of blood between two closely related animals brings about the least rapid destruction of the foreign blood. The transfusion thus offers us a means of determining in questionable cases the relationship of animals. A transfusion of blood is possible between varieties of the same species ; the blood of species that are very close to each other shows hemolysis only very gradually, and the animals with- stand large quantities of foreign blood; the more distant, however, animals are, the more violent the effects of the foreign blood become'* (p. 289). The hemolysis consists in the red blood corpuscles becoming permeable for the hemoglobin they contain, which begins to diffuse out. The red blood corpuscles become in consequence pale (ghosts or shadows).

It is obvious that this diffusion of the hemoglobin is rendered possible through some chemical alteration of the blood corpuscle. The experiments of Landois prove that the blood of closely related species is chemically and physicochemically more nearly identical than the blood of more distant forms. More recently the observations of Landois were taken up by Friedenthal,t Gruenbaum, and Nuttall. J These experimenters were able to avail themselves of Bordet's precipitation method. Bordet had found that after serum of a foreign species has repeatedly been injected into a rabbit, a precipitation will occur when blood of that foreign species and the blood of this rabbit are mixed. Moreover, the same reaction occurs when blood from an animal related to the one whose serum had been injected is mixed with the blood of the rabbit. Friedenthal and Nuttall used this reaction to find out the

\ G. Nuttall, Blood Immunity and Blood Relationship, Cambridge, 1904. blood relationship of animals. Nuttall found among others that if dog's serum was injected into a rabbit, the serum of this rabbit after- ward gave a precipitation with the blood of eight various canides, but with the blood of no other group of animals ! These experiments may also explain why the bastards between the sea urchin and starfish show a much greater mortality than the pure breed. In my experi- ments the hybrids between starfish and sea urchin died in large numbers after they reached the blastula or gastrula stage.* It seems as if the spermatozoon of the starfish, in addition to a developmental sub- stance, also carries something else into the sea urchin's egg which poisons the latter. It remains for further experiments to decide how far the physicochemical incompatibility of heterogeneous species which Landois and his successors discovered, restricts heterogeneous hybridization. It is, however, already obvious enough that ultimately the problem of hybridization and heredity must be transferred from the morphological to the chemical or physicochemical field.

We may now continue the discussion of the problem of heredity. The man whose work marks with that of Landois the beginning of a real theory of heredity had a fate similar to that of Landois. Gregor Mendel was a teacher of physics in Graz, and evidently the writings of Darwin induced him to investigate the laws of heredity; but he went at the problem in a spirit so entirely different from that of the biologists, and at the same time in a way which was so superior, that his discoveries were entirely overlooked for over thirty years, until De Vries discovered the same facts, "and also discovered accidentally Mendel's paper. At almost the same time Correns and Tschermak also called attention to Mendel's work. Mendel t carried on experiments on the hybridization of varieties of peas which he selected so that they differed in only one characteristic. It was his intention to find out what became of that discriminating or critical characteristic in the offspring. He found that the children of such parents - - the first generation of hybrids - - did not occupy an intermediate position between the two parents in regard to the discriminating characteristic, but were all pure breeds, inasmuch as the discriminating characteristic of one parent was transmitted to all of the children, while the characteristic of the other parent was ap- parently not transmitted. The discriminating characteristic of the peas used by Mendel for his experiments were, e.g., the difference in the shape of the ripe seeds, whether they were spherical or angular. When he crossed two forms of peas which were identical in every respect, except

t Gregor Mendel, Versuche uber Pflanzenhybride. Ostwald, Klassiker dcr Naturiaissen- schaften, Vol. 121. De Vries, Die Mutatiomtheorie, Leipzig, 1901. that the shape of one parent was angular that of the other round, all the children or hybrids of the first generation had round seeds. It was immaterial whether the female or the male was of the round seed variety. If we now assume, as we must, that there is also a corre- sponding difference in the sexual cells of the round seed variety and the angular seed variety, we must assume that in the mixing of the two the determinants of the round shape of the seed dominated, while the de- terminants for the angular shape of the seeds were prevented from manifesting themselves.

In another set of experiments Mendel crossed two varieties which were alike in every respect except the coloration of the albumin of the seed (endosperm). In one variety this possessed a pale yellow color, in the other it was green. The children of two such parents possessed only yellow endosperm. In a third case the discriminating characteristic between the two parents was the color of the shell of the seed : one was white, the other gray or grayish brown. The former also had white blossoms, the latter violet blossoms. The children all had seeds with gray shells and violet blossoms.

It is therefore obvious that in these cases one characteristic dominated, and Mendel called this the dominating characteristic, while the other, which was suppressed in the first generation of children, he called the recessive characteristic. Thus the spherical shape of the seed is a domi- nating, the angular shape a recessive, characteristic. The first generation of hybrids was therefore in regard to the dis- criminating characteristic not distinguishable from the pure breed of the one parent, which possessed the dominating characteristic; yet it was different in one respect, namely, its sexual cells. The child of two parents, the one of which possessed angular, the other spherical seeds, possessed two kinds of sexual cells in about equal number; namely, one half being cells possessing the determinant for the dominant, the other possessing the determinant for the recessive, characteristic. This follows from the results of Mendel's experiments when he crossed the hybrids of the first generation among themselves. In this case there was no uniform offspring, but the two distinct types, one with the recessive and one with the dominating characteristic, now reappeared. Only a fraction of the hybrids of the second generation had the dominant characteristic, the rest had the recessive characteristic. When the number of the individuals used for experimentation was sufficiently large, there existed always a definite ratio between the two kinds of offspring : the number of hybrids with the recessive characteristic was always one third of those with the dominating characteristic. This is exactly what

we should expect if the hybrids of the first generation possessed two kinds of sexual products in equal numbers; namely, those of the paternal and those of the maternal species or variety. If a large number of individuals of this kind be crossed, according to the law of probability, in one fourth of the cases an egg cell with the determinant for the reces- sive characteristic and a pollen cell of the same type would meet. This would result in pure offspring with the recessive characteristic. In one fourth of the cases egg cells with the determinant of the dominating characteristic would meet with the pollen of the same type, and the re- sult would be pure offspring with the dominant characteristic. In one fourth of the cases an egg cell with the determinant for the dominant characteristic would be met by pollen cells with the determinant for the recessive characteristic, and this would give rise to individuals with the dominant characteristic ; in the last fourth of the cases an egg cell with the determinant for the recessive characteristic would be met by pollen with the determinant for the dominant characteristic, and this again would give offspring with the dominant characteristic. In this way three fourths of the total offspring in the second generation would have the appearance of the species or variety with the dominant, and one fourth of the total offspring would have the appearance of the species with the recessive, characteristic ; both kinds would therefore be in the numerical relation of one to three, as stated.

Through many cultures, Mendel has shown that his conclusions are correct. Thus, e.g., his theory demanded that if the last-mentioned experiment be continued, and the individuals of the first-mentioned fourth of the offspring, namely, the ones that have reverted to the recessive character, be bred among themselves, only pure breeds with recessive character should be produced. The experiments proved that this is entirely correct. Mendel raised also the question as to what would happen if varieties of peas which differ in regard to two or more characteristics should be hybridized. In this case, the hybrids of the first, as well as of the later, generations behaved as if a specific hereditary substance existed for each characteristic, and as if these substances did not irfluence each other. For each one of the discriminating characteristics, the same laws hold which existed where the varieties differed only in regard to one characteristic.

The epoch-making importance of Mendel's work lies in the fact that he, for the first time, gave not a hypothesis but a theory of heredity, which made it possible to predict the results of hybridizations numeri- cally. His work forms the basis for all further work in this field which is of equal theoretical and practical importance. The observations of Mendel have since been confirmed and enlarged upon. Not only botanists like De Vries, who independently rediscovered Mendel's laws, but also zoologists like Bateson and his pupils, Cuenot, Castle, Guyer, and many others, have added to this field.

Mendel's laws do not, however, include all the cases of hybridization. De Vries has investigated this field in a masterful way, and has shown that there are at least two types of hybridizations : one in which, as in Mendel's cases, a separation of the discriminating characteristics occurs again in the offspring, and another in which constant races are at once produced. This type of hybridization is the one which proves especially useful to plant breeders in their attempts to produce new varieties. De Vries believes that the latter type of hybrids is produced when the sexual cells of one parent have a determinant for which there is no cor- responding determinant in the sexual cells of the other parent.

The objection might be raised that such a theory of chemical deter- minants in the sex cells as the cause of heredity might find difficulty in explaining the heredity of instincts; I believe that the contrary is the case. In a paper on " Egg Structure and the Heredity of Instincts" * I have pointed out that the hereditary character of the instincts demands a chemical rather than a morphological theory of heredity. Many instincts are obviously the outcome of tropisms. For the transmission of an instinct based on heliotropism, all that is required is the presence in the sexual cells of photosensitive material, or of a substance from which such material can be formed.

The current morphological and cytological literature contains many attempts at explaining the phenomena of heredity on a purely morpho- logical or cytological basis. There is no objection to this, as long as we realize that the morphological structures can only play a role through their physical and chemical properties. Several years ago an embryologist published the hypothesis that sex could be determined by submitting the mother to a certain diet. Delage pointed out that this idea was contradicted by the fact that in about 30 per cent of the cases twins have different sex, which would be impossible if the diet of the mother Determined the sex of the offspring. There is, however, one condition under which twins have invariably the same sex ; namely, when they come from the same egg. We have seen in a former lecture that from one egg twins can arise; namely,

when the contents of the egg are cut in two during the early periods of development, e.g. when the first two cleavage spheres become separated from each other. In the case of mammalian or human twins, we possess a criterion for the fact whether they come from one or two eggs in the condition of the egg membranes. Twins coming from different eggs have as a rule separate chorions. This follows from the development of the chorion. In all cases where twins have a common chorion they have also identical sex. This indicates that the sex of the embryos was determined before the germ was split into two parts, and as this must occur in the earliest stages of development, it follows that the sex of an embryo is definitely determined very early; how early can only be guessed at in mammalians, but in certain lower forms it can be shown that the sex is already preestablished in the egg before the egg is even fertilized.

A striking example for this assertion was discovered by Korschelt in Dinophilus apatris, a worm of the group of Turbellarians. As Korschelt's paper is not accessible to me, I quote the observation after Lenhossek.* Dinophilus lays two kinds of eggs, the one large and opaque, the other small and transparent. The eggs are fertilized inside the body of the female, and are afterward deposited in the sea water. Korschelt separated the two types of eggs, and found that the large opaque eggs give rise to females, the small transparent eggs to males. If this observation is correct, there can be no doubt left that in this case sex is already determined in the egg before the egg is fertilized.

Facts of a somewhat analogous character seem to exist in a number of forms such as plant lice, Cladocera and Rotifers. The Aphides are viviparous as long as the temperature is not too low and the plant is not drying out. Under such conditions they give rise to offspring of one sex only, namely, females. These reproduce females parthenogeneti- cally which possess no receptacle. When the plant dries out, or the temperature becomes low, in addition to females, males are also produced. The females which originate at this time possess a receptacle, and hence can pair. After pairing they are not viviparous, but lay fertilized eggs, the so-called winter eggs. From such eggs parthenogenetic females invariably arise, and now the cycle may be repeated. It is obvious that at least the sex of the winter egg is determined as soon as it is formed. The same is probably true also for the sex of the embryo which pro- ceeds from the summer eggs. In Cladocera conditions are not very different. From the winter egg females invariably arise, and these give rise parthenogenetically to females until, under conditions which have not been sufficiently investigated, males and females are formed,

* Lenhossek, Das Problem der geschlechtsbestimmcnden Ursachen, Jena, 1903. which copulate. As in Aphides, only a definite type of eggs, the so- called winter eggs, require fertilization, and from the fertilized eggs a female originates in every case. In Rotifers also the winter eggs are said to require fertilization and give rise to females. We have already mentioned the fact that in bees the unfertilized eggs give rise almost, or quite exclusively to males, while females can only (or perhaps mainly) arise from fertilized eggs. This seems to indicate that the entrance of a spermatozoon may give the egg a female character, while without it, it has a male character. Lenhossek believes it is possible that in bees, as perhaps in most animals, two kinds of eggs exist, one for each sex ; that for some reason the male egg is not fertilized when it is laid, while the female is fertilized. It is, however, difficult to harmonize with such a view the fact that old queens, whose supply of sperm in the receptacle is exhausted, and virgin queens lay only male eggs.

Among certain insects, e.g. the Hemiptera and Orthoptera, two' kinds of spermatozoa have been found, but one kind of eggs. These two kinds of spermatozoa differ in regard to a single chromosome, which is found only in one half of the spermatozoa, while it is lacking in the other; or which is larger in one half of the spermatozoa than in the other half. The first one to recognize the existence of two kinds of spermatozoa was Henking, who stated that in Pyrrhocoris (a Hemipteron) one half of the spermatozoa of each male possessed a nucleolus while the other half did not. Montgomery afterwards showed that Henking's nucleolus was an accessory chromosome. To McClung * of the Uni- versity of Kansas belongs the credit of having first recognized the importance of this fact for the problem of sex determination. He observed an accessory chromosome in one half of the spermatozoa of two forms of Orthoptera, Brachystola and Hippiscus, and traced their history. His conclusion may be quoted in full: "A most significant fact, and one upon which almost all investigators are united in opinion, is that the element is apportioned to but one half of the spermatozoa. Assuming it to be true that the chromatin is the important part of the cell in the matter of heredity, then it follows that we have two kinds of spermatozoa that differ from each other in a vital matter. We expect, therefore, to find in the offspring two sorts of individuals in approximately equal numbers, under normal conditions, that exhibit marked differences in structure. A careful consideration will suggest' that nothing but sexual characters thus divides the members of a species

into two well-defined groups, and we are logically forced to the con- clusion that the peculiar chromosome has some bearing upon the arrangement. "I must here also point out a fact that does not seem to have the recognition it deserves ; viz. that if there is a cross division of the chro- mosomes in the maturation mitoses, there must be two kinds of sper- matozoa regardless of the presence of the accessory chromosome. It is thus possible that even in the absence of any specialized element a preponderant maleness would attach to one half of the spermatozoa, due to the ' qualitative division of the tetrads '

McClung was inclined to believe that that half of the spermatozoa which contains the accessory chromosome gives rise to male offspring, while the other half gives rise to female offspring. E. B. Wilson f has recently investigated the chromosomes of the sex cells in a number of Hemiptera, and ascertained that the occur- rence of two kinds of spermatozoa is a constant phenomenon in this group. While in some forms the two kinds differ by an accessory chro- mosome occurring in one half of the spermatozoa of a male, in other forms they differ in regard to the size of one of their chromosomes. In the latter forms this specific " idiochromosome " is large in one half of the spermatozoa of a male and small in the other half of the sperma- tozoa. While McClung assumed that the accessory chromosome is a determinant for the male sex, Wilson shows that it is more probable that it gives rise to the female offspring.

If we sum up all these data concerning determination of sex, we therefore come to the conclusion that sex is, in all probability, already predetermined in the sex cells. In some forms, e.g. Aphides and other parthenogenetic forms, sex can unquestionably be determined by the eggs alone, and consequently we must assume here the existence of two kinds of eggs. In other organisms, like the Hemiptera and Or- thoptera, we have two kinds of spermatozoa and apparently one kind of eggs, and in these cases it is the spermatozoon which determines the sex. There exists apparently a third type of forms, e.g. bees, ants and social wasps, in which both eggs and spermatozoa share in the deter- mination of sex, inasmuch as the eggs alone determine the male, while the spermatozoon determines the female sex; if both are united, the influence of the spermatozoon predominates. If this is correct in this group of animals, only one kind of eggs and only one kind of sperma- tozoa will be found .|

j It follows from these data that the female egg, or spermatozoon, predetermines, also a different group of instincts from the male egg. Morphologically as well as in regard A good deal of mysticism arose from the fact that the number of young males and females is so approximately equal in many forms, e.g. in the human race. The fact, discovered by McClung, that through the process of chromosome division two kinds of sex cells must be formed in equal numbers in the male of Hemiptera and Orthoptera removes this source of mysticism.

The fact that all attempts to influence the sex of a developing embryo have thus far failed, harmonizes with the data given above. Born and others maintained that it was possible to influence the sex of tadpoles or frogs, or of the larvae of flies, by the food on which the larvae were fed. These statements have proved to be untenable. Maupas and Nussbaum have tried to determine sex, not by any influence upon the developing embryo directly, but upon its offspring. They experimented on a Rotifer, Hydatina senta. Nussbaum states that the mode of nutrition of the female embryo after it leaves the egg determines whether it will later give rise to large female eggs, or to small eggs for both sexes. Maupas had stated that the temperature determines the sex; but Nussbaum disagrees with him, believing that temperature has no direct effect upon the determination of sex.*

The two sexes differ also in regard to the so-called secondary sex- ual characters, e.g. the shape of the antennas in male and female butterflies, etc. The question now arises, Are these secondary char- acters already predetermined in the egg, or are they secondarily determined by the maturing or mature sexual glands? If the former were the case, the castration of the larvae before sexual maturity is reached should not prevent the development of the secondary sexual characteristics. Oudemans extirpated the sexual glands in caterpillars of Ocneria dispar, yet the butterfly showed all the secondary sexual characters. Professor Kellogg told me that he found that the castra- tion of the young caterpillars of the silkworms has no effect upon the formation of the secondary sexual characters. These observations also agree with the idea that the secondary sexual characters are pre- determined in the egg, and some of them possibly at as early a stage as the primary sexual characters.

The idea that the sexual glands determine, e.g. by internal secretion, to instincts, man and woman represent different species, and inasmuch as for a normal and happy life the instincts must act as a guide, it would seem erroneous to attempt to make life for both sexes absolutely identical. It would be equally preposterous, however, to insist that, for this reason, man and woman should not have equal rights. The traditional barriers to the rights of women are based, not on physiological grounds, but on the survival of the savage's idea, who made woman his slave. The adjustment of the sphere of action of woman should be left to her own instincts and judgment, and not to the dictation of lawyers and politicians.

* For further information, see Herbst, Formative Reize in der thierischen Ontogenese, Leipzig, 1901. '"the formation of all the secondary sexual characters, is also refuted by the following observations. There are cases of hermaphroditism known in which the one side of the body contained a testicle, the other an ovary. In hermaphroditic insects of this kind, it has been observed that the secondary sexual characters differed also on the two sides, the side with the ovary having a female, the other side a male antenna. This would be impossible if a substance produced by the sexual glands, and circulating in the blood, were the cause of the secondary sexual characters. If, however, the primary as well as certain secondary characters are already preformed in the egg, it might be well possible that an egg was already female on one side and male on the other side. Crampton grafted the heads of one sex upon the bodies of the other sex in pupae of butterflies, in trying to find out whether the sexual glands could influence the secondary sexual characters on the head, but this was not the case. There are then certain secondary sexual characters which seem to be determined before the sexual glands reach the mature stage. Certain secondary sexual characters, of course, such as develop at the period of sexual maturity, are determined by the development of the sexual glands and fail to appear when these glands are removed before the time of sexual maturity.

The form of the body as well as the instincts of the animal are trans- mitted through the sexual cells. We are forced to assume that the egg or the spermatozoon must possess a structure of a degree of complexity equal to that of the adult, or that the development occurs in a manner which renders such an assumption unnecessary. It is hardly necessary to mention that we must choose the latter alternative. If we examine the living egg of a sea urchin or a starfish, we find that its contents are chiefly liquid. If such eggs are exposed to a slight one-sided pressure, e.g. under the cover glass, the surface film or mem- brane bursts, and the liquid contents can be seen streaming slowly into the surrounding sea water. In this liquid minute granules are notice- able, which may be solid, but the main mass of the egg is liquid. The nucleus is surrounded by a solid film. It is possible or probable that the chromosomes are, in a certain phase of cell division, solid, or possess a high degree of viscosity. This follows from the fact that the form of each individual chromosome remains constant through all cell divisions. It is obvious that a mass which is to a large extent liquid cannot possess a structure of such a degree of complexity as the adult starfish or sea urchin. Moreover, we can observe directly