The Reemergence of Sex

Chapter 6 of Gametes and Spores: Ideas About Sexual Reproduction 1750-1914

John Farley, 1982

Reprinted with generous permission of the author and the Johns Hopkins University Press

During the last three decades of the nineteenth century, remarkable changes took place in the methods by which microscopists examined cellular material. New fixing and staining teclmiques made possible the detailed investigation of the cell nucleus, the importance of which was becoming increasingly obvious. (1) In 1885 the first edition of Arthur Lee's influential Vade-Mecum revealed that fixatives such as chromic, osmic, nitric, and picric acid and various cammine and hematoxylin stains were in general use. (2) lmmersion objectives came into vogue, and at the end of the 1880s, cytologists began using the new Abbe-Schott apochromatic lenses, which produced images unclouded by either spherical or chromatic aberration. Together, these new techniques enabled microscopists to observe the intricate nature of nuclear division and, more to the point, the detailed cytological events surrounding fecundation.

With these discoveries, the definition and interpretation of sexual reproduction underwent a remarkable change. It became clear that sexual reproduction involved more than a simple stimulation of the egg by the entry of a fecundating principle contained within the sperm or pollen grain. Intricate structural components of the egg and sperm also were observed. Indeed, so intricate were these cytological events that not only did sexual reproduction come to be seen as a unique event, quite unlike asexual growth, but biologists began to define it in structural terms. An egg was not simply a sporelike cell, differing from the spore only in its mode of stimulation. In order to develop, an egg required the physical entry of a sperm nucleus and its chromosomes.

Thus, by the early years of the twentieth century, sex had again become a central concern of biologists; no longer to be denied, it was recognized as more than a minor variation of asexual reproduction. Such changes came about only after a period of intense controversy, however. It should not surprise us that a controversy developed, yet this aspect of scientific investigation needs to be stressed again. Scientists are brought up to believe that a sharp distinction exists between "facts" and their interpretation, and they are prone to ascribe changes in outlook solely to technological improvements. Such scientists would argue, for example, that once the new microscopes and chemicals came into use, cytologists would see exactly how the nucleus divided, how gametes and spores were really fommed, and how chromosomes behaved during fecundation. Yet this is far from what actually occurred. Just as the invention of achromatic objectives in the 1820s generated rather than solved problems, so, too, in the last decades of the nineteenth century problems were generated and not immediately solved. Biologists in both periods may have shared a common technological tool, but it did not produce any agreement on what was seen. In both periods "facts" could not simply be observed; they had to be interpreted, and these interpretations reflected a host of differing assumptions on a wide range of biological issues.

In the 1870s it was still generally believed that fecundation involved the entry and dissolution of one or more sperm cells within an egg cell. In 1869, Eduard Strasburger had postulated that only one sperm cell of the fern was involved; that it passed down the neck of the archegonium, came into contact with the egg cell, penetrated it, and dissolved there.(3) Although it was generally assumed that the dissolution of sperm within the egg guaranteed that the offspring's characters were a blending of matemal and paternal characters, in the same way that white and black paints blend together to produce gray, it was more usual to stress the stimulating influence of this dissolution. In 1874, for example, the influential physiologist Wilhelm His reiterated the standard physicochemical argument that the sperm acted as a molecular stimulant for the egg to develop.(4)

The mode by which new cells arose was still a subject of controversy. Zoologists generally believed that new cells were formed by the division of preexisting cells, but botanists, faced with the complex cellular activities within the endosperm of the flowering plants, believed also in the occurrence of free cell-formation.(5) Moreover, the role played by the nucleus in these events was not clear. Some believed that a constriction and splitting of the nucleus into two halves preceded cell division, but others, such as Strasburger, reported that the old nucleus disintegrated prior to cell division and that the daughter nuclei appeared de novo.(6)

This uncertainty provided the background for the famous papers of Oscar Hertwig. Claiming in 1876 and 1877 that the "cleavage nucleus [i.e., the zygote] arises from the conjugation of two different sexual nuclei, a female nucleus which is derived from the germinal vesicle and a male nucleus which is derived from the body of an entering spermatozoon," Hertwig posited a morphological theory of fecundation that challenged the generally held physicochemical interpretation.(7) Hertwig had been trained under Ernst Haeckel at Jena, and taught anatomy there until 1888. He had become interested in the process of fecundation in 1874, following the publication of Leopold Auerbach's findings. Auerbach had reported the de novo appearance of two "pronuclei" in the fertilized egg cell and their subsequent fusion to form the cleavage nucleus. "Suddenly the borderline between both nuclei disappears, " he wrote, "and they become united to a single mass."(8) From his work on the eggs of the sea urchin, however, Hertwig reported that "at the period of maturation the germinal vesicle undergoes a retrogressive metamorphosis and is propelled to the yolk surface.... Its membrane is dissolved, [and] its contents decompose to be finally reabsorbed again by the yolk. The Keimflecke, however, appears to remain unaltered and to continue in the yolk,where it becomes the permanent nucleus of the ripe ovum capable of fecundation." (9) Hertwig also claimed that the other pronucleus represented the head, or nucleus, of the entering sperm.

Figure 1

Figure 1   Oscar Hertwig. Reproduced by permission of Museum of Comparative Zoology, Harvard University

At the same time, Hermann Fol, who, like Hertwig, had studied under Haeckel at Jena and was conscious of the important role Haeckel had bestowed on the nucleus, investigated fertilization in the starfish. He reported that the germinal vesicle underwent two rapid divisions and that only one nucleus, the female pronucleus, remained in the egg, the others being expelled. He also described the actual penetration of a single sperm into the egg, where, he explained, it fused with some egg protoplasm to form the male pronucleus. This pronucleus then traversed the egg to unite eventually with the female pronucleus. (10)

Figure 2

Figure 2   Hermann Fol. Reproduced by permission of the Museum of Comparative Zoology, Harvard University.

Figure 3

Figure 3   A single sperm penetrating an egg. From Hermann Fol, "Recherches sur la condation et le commencement de l'hnognie chez divers animaux," Mem. Soc. Phys. Hist. Nat. Genve 26 (1879) . Reproduced by permission of the Museum of Comparative Zoology, Harvard University.

The papers by Hertwig and Fol generated considerable criticism. Much of this stemmed from continuing opposition to the idea of nuclear continuity from cell generation to cell generation and also from hostility to the morphological interpretation ot fecundation. Eduard Strasburger, like Hertwig and Fol a close associate of Haeckel's, denied nuclear continuity and argued in his text über Zellbildung und Zelltheilung that "the contents of the pollen tube penetrate the egg in a dissolved form" and that although fecundation depends on the introduction of nuclear substance from the sperm, the substance enters as "physiological elements," not as morphological units of the sperm nucleus (11)

A few years later, however, painstaking and exacting examination of the nature of cell division revealed the existence of formed nuclear elements, the chromatin, which, breaking into segments and splitting longitudinally, became distributed on the equatorial plate of the dividing cell. These pieces, or Faden, then migrated to the two poles of the cell, where they coalesced to form the nuclei of two daughter cells. Terming this process karyokinesis, or "indirect nuclear division," Walther Flemming, professor of anatomy at the University of Kiel, argued that it was the only means by which nuclei divided in cell division. "Cell division," he remarked, "may be regarded as a process of asexual reproduction. In its course the chromatin separates from the achromatin, condenses into figures of a characteristic shape, [and] divides into two parts which provide the foundation of the daughter nuclei."(12) A year later Flemming agreed with Hertwig that fecundation involved the copulation of sperm and egg nuclei, but he argued that it was the chromatin threads of the male and female nuclei which united to form the cleavage nucleus. (13) Strasburger, meanwhile, changed his position on nuclear continuity. Writing in1880 in the third edition of his über Zellbildung und Zellfheilung, he noted:

"In the plant kingdom the newly appearing cell nucleus can be traced back to an earlier existing nucleus. The new nuclei arise through the division of an older nucleus. The earlier concept in which the daughter nuclei were supposed to be newly formed owing to the dissolution of the mother nuclei, has proven to be false."(14)

At this time, however, although he accepted the continuity of nuclei, Strasburger denied the individuality and continuity of the chromatic threads (termed chromosomes by Waldeyer in 1888). Describing in detail the process of indirect nuclear division, he claimed in 1882 that the resting nucleus contained a single, long thread consisting of nucleomicrosomes embedded in a nucleohyaloplasm. By the fusion of adjacent microsomes, he argued, the thread became shorter and thicker, producing alternate thick and thin pieces of microsomes and hyaloplasm respectively. The single thread then broke within the thin areas, he reported, the number of breaks being more or less constant in any species. The microsomes then divided such that each chromatic element came to lie on the equatorial plate as a paired structure, which eventually pulled apart to form the nuclei of the two daughter cells. (15)

Figure 4

Figure 4   Eduard Strasburger. Reproduced by permission of the Museum of Comparative Zoology, Harvard University.

Strasburger's emphasis on the microsomal granules rather than on the chromosomes themselves sprang partly from his observation that cell division in Hemerocallis fulva deviated from the normal pattern: the cell sometimes divided into more than two daughter cells due to the retention on the equatorial plate of one chromatin element. This element, he noted, "forms a completely normal though very small cell nucleus," and thus "each segment of the nuclear thread shares the properties of the entire thread."(16) The role of indirect nuclear division, he therefore concluded, was to distribute equal amounts of the microsome material to each daughter nucleus:

The significance of the complicated nuclear division may above all lie in the cell nucleus splitting into two completely equal halves. In the first segmentation of the nuclear thread the pieces are apparently of very unequal size and could become distributed often in a very unequal manner on both sides of the equatorial plate. Thus by splitting longitudinally and distributing the longitudinal halves to both daughter nuclei, the halves become true and equal.... The simultaneous longitudinal division of the segments would be the surest means of distributing these substances equally to both daughter nuclei (17)

This belief in the transient nature of the chromosomes and the primacy of their microsomal elements affected Strasburger's interpretation of fecundation profoundly. Arguing in 1877 that in plants the pollen tube introduced a "formless fecundating material," while in algae copulation consisted of the "fusion of similar parts of both swarmers with each other," he denied the unique role of the nucleus in the process of fecundation. (18) In 1882, however, although he still maintained that in algae "fecundation is a matter of the union of equivalent parts of both copulatory cells," he acknowledged that with gamete differentiation in higher plants, only the nuclear substance was involved. (19) He no longer argued that a formless fecundating material was liberated from the pollen tube. Instead, he spoke of small pieces of the nuclear thread, derived from the pollen nucleus, being passed through the walls of the pollen tube into the egg cell.

Two years later, however, Strasburger admitted not only that fecundation involved the fusion of complete egg and sperm nuclei but also that the nuclear segments remained preserved in the resting nucleus (although still joined end to end in a continuous thread). Upon fecundation, he argued, the nonnuclear material of the egg and sperm fuses, but the chromatin material lies in contact without actually fusing. Then the chromatin networks of the male and female nuclei divide into segments in the normal manner, distributing an equal amount of nuclear material to the two daughter nuclei so that each daughter nucleus will contain an equal number of patemal and maternal segments. Then and only then do the individual segments fuse (verschmelzen) end to end to form a single thread, "which is formed half from the nuclear segments of the father and half from the mother."(20)

By 1884, therefore, Strasburger had added his considerable prestige to those who believed that "the specific characteristics of the organism are based in the properties of the cell nucleus," and that sexual reproduction, through the agency of the nuclear segments, or "chromosomes," was the means by which the characteristics of both parents were passed to the offspring.

With the substitution of "nuclei" for "cells" and finally "chromosomes" for "nuclei," the definition of sexual reproduction obviously changed and the role of the male gamete was considerably enhanced. Apart from that, however, views on the significance of sexual reproduction were not altered by the discovery of its chromosomal base. Those who were concerned with problems of inheritance continued to view sexual reproduction as the means whereby male characters were incorporated into a female egg, and the offspring as a blending of both parental characters. Indeed, the fact that fecundation was seen to involve a complete fusion (Verschmelzung) of male and female cells, nuclei, and chromosomes lent a cytological basis to the blending of inherited characters, by which, in the words of Strasburger, "the permanence of the species as a whole" is maintained.(21) In addition, those who supported the physicochemical interpretation of fecundation could continue to do so, particularly after François Maupas showed that fatal aging occurs in cultures of ciliates unless conjugation occurs.(22) The view of fecundation as a process of rejuvenation held that the zygotic cell, once formed, continued to divide until its original energy was exhausted, at which time gametes were produced and a new burst of energy was generated by their subsequent fecundation. Thus, this view still had its adherents.

These interpretations of blending and rejuvenation were first fundamentally challenged by August Weismann, whose speculative writings of the 1880s generated a fierce controversy over the cytological events of gamete formation and subsequent tecundation, the biological implications of sexual reproduction, and the nature of the developmental processes. These differing reactions to Weismann's speculations led cytologists to see totally different events under the microscope; the "facts" did not speak for themselves.

The Problem of Maturation Division

The commonly held view that germ cells arose in the same manner as tissue cells—in other words, by a typical, "indirect" nuclear division—began to be questioned in the 1880s. In 1884, for example, Strasburger stated that germ cells arose by a normal indirect division, but he also seemed to suggest that quite unique processes were involved. In the preparation of germ cells, he argued, it was necessary "that their idioplasm be reduced to half that mass which the embryonic nucleus possessed." But he also argued that it was necessary that the nature of the idioplasm or hereditary material change or be remodeled prior to fertilization, such remodeling perhaps being indicated by the "excretion" of polar bodies by the egg cell.(23)

It had long been known that close to the time of fertilization the egg cell buds off two minute bodies termed directive cells (Richtungskorper), or polar bodies, the significance of which had been a point of considerable controversy. One of the most popular explanations, and the one to which Strasburger took exception, was Edouard van Beneden's. At a time when the germ layers were being discussed widely with reference to the biogenetic law-specifically in terms of the embryonic ectoderm and endoderm of vertebrates constituting an embryonic recapitulation of the two-layered situation found in adult coelenterates (epidermis and gastrodermis)-van Beneden had discovered that the egg cells of coelenterates had an endodermal or gastrodermal origin, while the sperm cells were derived from the ectodermal or epidermal layer. "From the sexual point of view," he argued, "ectoderm and endoderm have an opposite significance," the ectoderm being the male germ layer, the endoderm the female.(24) Thus, since it always contains these two germ layers, the organism is hermaphroditic. In 1883 van Beneden had extended this hermaphroditic theory to include the individual cells; they, like the organism, also must be hermaphroditic. Thus, in order to produce a unisexual egg cell from a tissue cell, the male half of the nucleus must be removed as polar bodies, and to produce a male gamete the female half also must be removed. (25)

As a result of these concepts, van Beneden concluded that just prior to fertilization, the germ nuclei were in fact only half nuclei, and believing them to be only half nuclei, he necessarily denied that fecundation involved a complete fusion of the two complete nuclear elements. Rather, he wrote, "Fecundation consists of a replacement, in the substitution of a half-nucleus furnished by the male and introduced by the sperm, for a half-nucleus eliminated by the egg in the form of polar bodies."(26) This view was generally disregarded. His conclusions that polar bodies represented an expelled male part of the nucleus, however, seemed to receive some approval. "The female sexual character of the egg," he wrote, "appears only after the expulsion of the polar globules; in fecundation the male elements of the egg are replaced by new elements supplied by the zoosperm." (27)

Van Beneden thus disagreed with Hertwig and Strasburger that fertilization involved the activities of complete egg and sperm nuclei. "One cannot give the name 'cell nucleus' to the clear spot nor the corpuscle contained in it," he wrote in 1876, "for the element looking like a nucleus which is formed near the surface of the yolk does not become the nucleus of the first cleavage sphere until after being united to another element having also the appearance of a cell nucleus. It is for this reason that I have called the peripheral bodies by the name 'pronucleus peripherique,' and the element which is formed in the center of the yolk by the name 'pronucleus central.' "(28)

Van Beneden's claim that the pronuclei were only half nuclei was further justified, according to him, by the significant discovery that each of the pronuclei in the nematode Ascaris contained only two "anses chromatiques. "; Further, he argued, upon conjugation the anses from each pronucleus do not fuse but line up on the equatorial plate as four anses which then divide, distributing four anses to each daughter nucleus.(29) Thus, van Beneden concluded that the embryonic cell must contain twice as many "anses chromatiques" as each pronucleus and that fecundation was essentially a mechanism of replacement or substitution. "Fecundation," he wrote, "appears to consist essentially in this reconstruction of the first embryonic cell, [a] cell revivified and provided with all necessary energy in order to transform into an individual like the parent."(30) A corollary to van Beneden's concept was the implication that since the gamete mother cell must divide in such a way as to produce a half nucleus and eliminate the other half as polar bodies, the mechanism of gamete formation must differ from the exact, quantitative, normal nuclear division.

Strasburger's early suggestions that the polar bodies were an excretion related to the remodeling of the idioplasm stemmed directly from the speculations of Wilhelm Roux. In 1883 Roux, expressing his interest in the mechanics of development, posed a fundamental question. If the characteristics of the organism were based on the properties of the cell nucleus, and if indirect nuclear division resulted in an exact duplication of nuclear material, how could a series of identical nuclei produced by the repeated and exact divisions of the embryonic nucleus induce those changes seen in ontological development? Roux concluded that although indirect nuclear division did ensure an equal distribution of the nuclear material in terms of quantity, nevertheless a qualitative change in the material took place during the nuclear divisions. Thus the nuclear chromatin was not a homogeneous mass of material but an aggregate of qualitatively different bodies that were distributed unequally during cell division. From this perspective Roux concluded that in order to re-form the original embryonic nucleus containing the full complement of these qualitatively different bodies, the nuclear material must be remodeled during the formation of the germ-cell nuclei.(31) Again, this suggested that the mechanism of germ-cell formation differed from the normal method of indirect nuclear division.

In 1887 Walther Flemming provided the first cytological evidence that the cell divisions involved in the production of sperm differed from the normal type of mitosis.(32) Spermatogenesis, he reported, involved two types of cell division. The first type differed little from normal mitosis, but the second, or "heterotypic mitosis," seemed to be unique in two ways. First, the chromosomes appeared as knots or rings. Second, and more significantly, the number of such ringed chromosomes was half the number that appeared in tissue cells-twelve rather than twenty-four in the salamander. However, for reasons that need not detain us here, Flemming saw nothing of very deep significance in these differences. We owe to August Weismann the statement that germ-cell production must be fundamentally different from tissue-cell production.

In a paper of 1885 Weismann had posed Wilhelm Roux's question and, like Roux, had concluded that the nuclear material of embryonic cells must change qualitatively during ontogeny, becoming less and less complex as development proceeds.(33) The fact that the longitudinal division of chromosomes during mitosis leads to daughter nuclei with an equal quantity of nuclear material does not prove, he argued, "that the quality of the parent nucleoplasms must always be equal in the daughter nuclei."(34) However, he went on to ask, if the cell nuclei change qualitatively during development, how can germ cells be produced again, for germ cells, unlike embryonic cells, contain within their nuclei "all the hereditary tendencies of the whole organism." Denying the possibility of any gradual or sudden retransformation of embryonic nucleoplasms into germinal nucleoplasm, he argued instead "that in each ontogeny, a part of the specific germplasm contained in the parent egg cell is not used up in the construction of the body of the offspring, but is reserved unchanged for the formation of the germ cells of the following generation."(35) However, he realized that the germ cells themselves were quite complex and must therefore contain in their nuclei a specific "histogenic nucleoplasm" to control their development. In order for embryonic development to begin, he speculated, it was necessary for this histogenic nucleoplasm to be removed, leaving the germ-cell nuclei with only pure germ plasm. This histogenic nucleoplasm, he concluded, was removed as polar bodies.(36)

Figure 5

Figure 5   August Weismann and the continuation of his germ plasm: his son, Julius. Photo, taken in Freiburg in 1909, reproduced by permission of the Museum of Comparative Zoology, Harvard University.

Weismann's and van Beneden's notions on polar bodies had more in common than either man would admit. Both assumed that before fertilization could take place it was necessary to remove something physically from the germ nuclei, and that the maturation divisions leading to the production of germ nuclei were somehow related to this expulsion process. They differed, of course, in their views of the nature of this removed part, a difference that could be tested quite easily, both realized, by determining whether parthenogenetic eggs also removed polar bodies. From van Beneden's point of view, parthenogenetic eggs, since they were capable of development without fertilization, must be whole nuclei from which the male element had not been removed. In Weismann's view, however, it was still necessary for parthenogenetic eggs to have their own "histogenic nucleoplasm" and for this to be removed before development could proceed. Thus, it was something of a surprise to Weismann when he discovered later in 1885 that parthenogenetic eggs of Daphnia produced one polar body, not two. He concluded from this finding that the second polar body, expelled from normal eggs but not from parthenogenetic eggs, must have a different role than the first. The role of this second polar body provided the substance of Weismann's famous paper of 1887, "On the Number of Polar Bodies and Their Significance in Heredity, " the reaction to which stirred the cytological controversy of the 1890s.(37)

Weismann believed that originally all organisms reproduced only asexually, in which case their germ plasms were completely homogeneous. When sexual reproduction took place for the first time, however, two different germ plasms necessarily came together (Fig. 6.6). In order to keep the quantity of total germ plasm constant in this first sexually produced generation, it was obviously necessary, he argued, that each individual germ plasm be halved in quantity. In each succeeding generation, therefore, to maintain a constant total quantity of germ plasm, the quantity of each individual or ancestral germ plasm had to be halved. In other words, in each subsequent sexual generation the total number of ancestral germ plasms, or ids, doubled while the quantity of each id was halved. Obviously, since each id was a very complex entity, there was a minimal size for each id such that the quantity of each could not be halved indefinitely. Since this stage was reached soon after sexual reproduction first appeared, the question remained how sexual reproduction could continue without a corresponding doubling of total germ plasm in each generation-the halving of individual ids being no longer possible. The answer was clear: there must be a mechanism which reduced not the quantity of each id but the number of ids.

Figure 6

Figure 6   Reduction of the quantity of each id in the early generations of the first sexually reproducing organism, according to Weismann. From Alice Baxter and John Farley, "Mendel and Meiosis," J. Hist. Biol. 12 (1979). Reproduced by permission of D. Reidel Publishing Co., Dordrecht, Holland.

Weismann visualized the ids as being arranged in rows along the nuclear threads. Clearly, then, in normal mitosis, where each thread, or chromosome, splits longitudinally, the number of ids would be the same in both daughter cells and equal to the number in the mother cell. Thus, in order for an id reduction to take place during the production of germ cells, "there must be yet another kind of karyokinesis, in which the [chromosomes] are not split longitudinally, but are separated without division into two groups, each of which forms one of the two daughter nuclei. (38) Such a mechanism is shown in Figure 6.7A. The four chromosomes do not divide longitudinally but are arranged singly on the equatorial plate of the dividing cell. Two chromosomes move to one pole to form the germ-cell nucleus; the other two are removed as thesecond polar body. Thus, Weismann concluded, "with the nucleus of the second polar body as many different kinds of idioplasm are removed from the egg as will be afterwards introduced by the sperm nucleus; thus the second division of the egg nucleus serves to keep constant the number of different kinds of idioplasm."(39) However, it was also possible for the number of ids to be reduced by a transverse division of each chromosome during maturation division (Fig. 6.7 [B]). In both cases the end result would be the same: the daughter nuclei would receive half the number of ancestral germ plasms, or ids, that were present in the mother cell. Weismann termed this division, this reduction in the number of ids, a "reduction division."

Figure 7

Figure 7   Reduction of the number of ids in modern sexually reproducing organisms (after Weismann). From Alice Baxter and John Farley, "Mendel and Meiosis, J. Hist. Biol. 12 (1979). Reproduced by permission of D. Reidel Publishing Co., Dordrecht, Holland.

At the end of his 1887 paper Weismann made the very significant remark that since it was unlikely that the polar bodies would remove the same ids each time, those retained in the germ cells also would be different. In other words, the germ nuclei were probably different from each other. This conclusion was to become very important to Weismann in his later papers.

A few years later Weismann had to alter his theory slightly when Oscar Hertwig and Theodor Boveri (a professor at Wurzburg and a former student of Richard Hertwig's) demonstrated that egg and polar-body formation was essentially the same as sperm formation. Boveri's interest in the polar bodies had been aroused by van Beneden's report of the reduced number of "anses chromatiques" in the germ cells. The express purpose of Boveri's first investigation was therefore to observe with his own eyes the process of fertilization and to verify van Beneden's description of the polar bodies. (40) Boveri did verify that the chromosome number was halved, but, disagreeing with van Beneden, he maintained that the division was a typical karyokinetic division and not simply an expulsion of the male element. In his study, Boveri also presented the first description of chromosome tetrads (Vierergruppen) and their behavior during reduction division. In Ascaris megalocephala var. bivalens, where the normal diploid number is four, Boveri observed two groups of chromosomes with four chromosomes in each group. The first division of the egg separated the tetrads into two dyads, one of which remained in the egg, the second of which entered the first polar body. The second division of the egg separated the two elements of each dyad, leaving a total of two chromosomes in the egg cell and two chromosomes in the second polar body. Meanwhile, the first polar body divided once. Thus the divisions produced a total of three polar bodies, each containing two chromosomes.

In 1890 Hertwig confirmed Boveri's findings. He also presented a complete description of sperm cell formation in which he showed conclusively that "the transformations of the nucleus which on one side take place in the transformations of sperm mother cells into sperm, and on the other in the production of polar bodies, show an extraordinary and very striking similarity."(41) Whereas in spermatogenesis one sperm mother cell divides into two spermatocytes, each of which quickly divides again to form four sperm, in oogenesis, the egg mother cell divides to form one oocyte and one polar body, and the oocyte quickly divides again to form one egg cell and the second polar body (Fig. 6.9). Thus, Hertwig concluded, the polar body does not represent a part of the germ plasm removed from the egg but has, in fact, "the morphological value of rudimentary egg cells."(42)

Figure 8

Figure 8   Theodor Boveri. Reproduced by permission of the Museum of Comparative Zoology, Harvard University.

Figure 9

Figure 9   Theodor Boveri's and Oscar Hertwig's interpretation of oogenesis and spermatogenesis. From Alice Baxter and John Farley, 'Mendel and Meiosis," J. Hist. Biol. 12 (1979) Reproduced by permission of D. Reidel Publishing Co., Dordrecht, Holland.

Thus Hertwig and Boveri confirmed that during the maturation process "there is a reduction by half of the originally existing number of chromosomes, and this numerical reduction is not therefore only a theoretical postulate but a fact."(43)

Two aspects of the Hertwig and Boveri papers need to be stressed at this stage. First, the reduction to which they referred was a reduction only in thenumber of chromosomes, not in the number of ids (i.e., chromosomal components). Second, this numerical reduction took place during two sequential cell divisions and not by the bodily removal of part of the nuclear material in the form of polar bodies. Boveri termed this interpretation of polar bodies the "egg hypothesis," or "phylogenetic hypothesis," in that polar bodies are abortive eggs, which to him were nothing but a "phylogenetic reminiscence."(44) Also, contrary to Weismann's interpretation, Boveri claimed that both polar bodies "originate in the same way and have the same size and chemical composition"; since the first polar body actually divides once, "the four sperm arising from one mother cell correspond to the egg and the three polar bodies."(45)

Weismann viewed these papers by Hertwig and Boveri as a partial vindication of his own theoretical conclusions. Agreeing with them that the number of chromosomes and thus the mass of nuclear material was reduced by half, he also claimed that since the chromosomes were not alike "but are derived from the differing germplasms of various ancestors" (a claim with which, as we shall see, Hertwig will not agree). "it follows that a reduction of the ancestral germ-plasms is admitted."(46) On the other hand, Weismann was forced to admit that his previous interpretation of the polar bodies was in error; both were intimately related to the division process, but one of them was not a carrier of histogenic nucleoplasm. In addition, Weismann was led to question why there should be two sequential divisions when one could equally well fulfill the necessary goals of reduction. The answer came in the papers of Hertwig and Boveri: "because the number of rods is doubled before the process of reduction has begun."(47) As Weismann then argued, by having two divisions rather than one, "the highest possible number of combinations of germ plasms are offered for the operation of natural selection."(48) Thus the complex cytological events leading up to the formation of the germ cells were of fundamental importance to the interpretation of sexual reproduction. They represented, according to Weismann, "the attempt to bring about as ultimate a mixture as possible of the hereditary units [the ids] of both father and mother."(49) Therefore, sexual reproduction was the means of introducing variation into the population; it provided the raw material upon which natural selection acted. Not only was sexual reproduction totally different from asexual reproduction, but it was absolutely essential to the whole evolutionary story.

Weismann's interpretation of maturation division as a reduction in ids, or characters, formed the focal point of the controversy over germ-cell formation in the 1890s. The term "reduction division" was used in descriptions of spermatogenesis and oogenesis to refer to a qualitative reduction in characters. (A division that did not result in a reduced number of qualities was usually called an "equation division.") The controversy stemmed primarily from the question whether or not such a "reduction" took place. Many cytologists interpreted maturation division as a reduction in nuclear mass only, and not as a reduction in nuclear germ plasms. In addition-and this was to become crucial a each germ cell was uniquely characterized by different combinations of germ plasms. To Strasburger, Hertwig, Leon Guignard, and others, all germ cells of an organism were identical. Sexual reproduction was not a process leading to variations in offspring as Weismann maintained; rather, it was one which, by the fusion of parental nuclei, "tends to maintain the permanence of the species as a whole."

As a result of Weismann's papers, cytologists in the 1890s focused their attention on the problems of maturation divisions in eggs, sperm, and pollen grains. It should be stressed that the reduction-division controversy developed at the very time when the new apochromatic lenses were coming into use and at the time when a younger generation of embryologists was beginning to turn away from microscopic descriptions of embryonic stages to instead subject the embryos to experimental manipulation.

The basic position of those who opposed Weismann's hypothesis of reduction division was first clearly outlined in the 1890 papers of Hertwig and Boveri. The former worked on the two varieties of the nematode Ascaris megalocephala; the latter studied an echinoderm, a medusa, and five species of mollusk.

At that time it was known that in the normal mitotic division of Ascaris megalocephala var. bivalens, four chromatic threads appear. Each thread then divides longitudinally and four chromosomes are distributed to each daughter nucleus. During spermatogenesis, however, eight chromosomes eventually appear, not four. These are then distributed-two to each of four spermatids-by means of two nuclear divisions that follow immediately after each other without a resting stage.

The basic question, of course, was how these eight chromosomes were formed. Hertwig noted three possibilities (Fig. 6.10):

1. The chromatin network of the resting cell broke into eight segments instead of the usual four. Thus, maturation division occurred without any longitudinal division of chromosomes, and the eight chromosomes represented eight different individual threads.

2. The chromatin network of the resting cell broke into the usual four chromosomes, each of which then divided once longitudinally. Thus the eight chromosomes consisted of four identical pairs.

3. The chromatin network of the resting cell broke into only two elements, each of which then divided twice longitudinally. Thus the eight chromosomes consisted of two sets of four identical chromosomes.

On rather shaky grounds Hertwig opted for the third possibility. He argued that during the maturation process the chromatin network breaks into half as many chromatin threads as existed in a tissue cell and that each of these chromosome threads then divides twice longitudinally to form a "tetrad" of four identical chromosomes. Each member of a tetrad is then distributed to a gamete nucleus during two nuclear divisions, between which there is no resting stage. Clearly, as a result of this process, only a mass reduction of nuclear material and a numerical reduction in chromosomes occurs. There is no reduction in the number of ids nor is there any variability among the gamete cells.(50) Hertwig's position was backed by Boveri, who in Zellenstudien III reaffirmed his earlier belief that each tetrad was formed by the two longitudinal divisions of a single chromosome.

It should be noted at this point that both Hertwig and Boveri attributed thereduction of chromosome number in the gametes to the fact that only half as many chromosomal elements appeared at the initiation of the maturation divisions as appeared at the initiation of normal cell division. In Ascaris, for example, four chromosomes appeared in mitosis but, they assumed, only two appeared in gametogenesis. Each of these chromosomes then divided twice longitudinally to form two "tetrads," each containing four chromosomes. As Boveri wrote in 1890:

If we consider first of all the formation of the egg, then there is only one statement which can be put forth as certain and generally valid, that the reduction must take place, at the latest, in the primordial germ cell. Therefore, by the formation of the first maturation spindle the chromosomes already are present in the reduced number.(51)

Figure 10

Figure 10   The reduction division controversy. "R" denotes stage of reduction. Reduction (R in Fig. 6.10) took place before the cells began to divide.

A. (1) Spireme, (2) Four chromosomes appear, (3) Each chromosomes divides once longitudinally to arrange themselves on equator, (4,5) The chromosomes then move to each pole and the cell divides. Each daughter cell then contains four chromosomes. Daughter cells identical to mother cell.

B. (1) Spireme, (2) Only two cromosomes appear, (3) Each divides TWICE longitudinally to form two tetrads; all chromosomes in tetrad alike, (4,5) Two divisions without a resting stage. Each gamete receives two chromosomes, half the somatic number. All cells identical.

C. (1) Spireme, (2) Two chromosomes appear but each consists of two chromosomes joined en-to-end. "Pseudoreduction," (3) Each doubled chromosome divides ONCE longitudinally to form two tetrads, (4,5) First division: Each tetrad moves apart longitudinally. Second division: Transverse division (shown by dotted lines). All cells not identical.

It is clear that most of the more eminent cytologists of the 1890s supported the above position. Leon Guignard, professor of botany at the Paris School of Pharmacy, reported in 1891 that whereas the primordial pollen mother cells and epidermal cells of Lilium had twenty-four "batonnets chromatiques," only twelve segments, each consisting of two rows of chromatic granules, appeared in the definitive pollen mother cell. There is nothing to prove, he remarked, "that at the moment when the nucleus of the mother cell has taken birth, these 24 'batonnets chromatiques' are joined together in pairs, end to end, to give twelve." These twelve, longitudinally split chromosomes then moved apart at the first division to distribute twelve single chromosomes to each daughter cell. Each of these then split longitudinally again and twelve chromosomes passed to each of the four grains of pollen. Thus, as a result of two longitudinal splittings of each chromosome and two nuclear divisions, "the nucleus of each of the four grains of pollen is derived from a mother cell constituted of twelve segments." Apart from the fact that the cell began with twelve chromosomes rather than the typical twenty-four, Guignard could see nothing "etranger à la marche normale de la karyokinase."(52) In 1898, after describing virtually the same process in Naias major (with the exception that both of the longitudinal chromosome splittings took place before nuclear division commenced), Guignard concluded:

The second division simply has the object of distributing equally to the four sexual nuclei the chromosomes already formed during the first division. It reduces by half the quantity of nuclein which they receive, compared to the amount possessed by vegetative nuclei as a result of ordinary mitosis. But, no more than the first division, it does not reduce qualitatively, and the four nuclei are equivalent with regard to hereditary properties. (53)

In 1893 August Brauer, who like Boveri had studied under Richard Hertwig, repeated Oscar Hertwig's investigations on spermatogenesis in Ascaris. He observed that before nuclear division began, only half as many chromosomes appeared as appeared at the initiation of mitotic division, and that each chromosome then divided twice longitudinally. Thus, in his words, the only difference between normal mitosis and maturation divisions was that "one transverse division of the Faden does not occur, whereas one more division (nuclear) follows. The first causes the number of chromosomes to be reduced by half; the second division, together with the elimination of a resting stage between both divisions, produces a halving of the mass." Thus, he concluded, "a reduction division in the sense of Weismann does not occur." (54)

Similar conclusions were reached by Friedrich Meves, who was working on Salamandra maculosa, and by John Farmer, professor of botany at the Royal College of Science, who was working on the Hepaticae. Both recorded that maturation divisions of the nucleus were preceded by two longitudinal divisions of the chromosomes. Thus, Farmer concluded from his work on spore formation in the Hepaticae, "a study of these divisions affords not the slightest evidence in favour of any reduction division (in Weismann's sense) taking place.... The only reduction is a numerical one." (55)

Some cytologists interpreted maturation division in the terms demanded by Weismann's hypothesis, but as a group their prestige hardly matched that of their opponents. Indeed, three of them-Chiyomatsu Ishikawa, Otto vom Rath, and Valentin Haecker-were graduate students of Weismann's, and their goal, they stated, was to illustrate the validity of the Weismannian hypothesis rather than to test it. (56) The most impressive of those who supported Weismann's concept of reduction division was Johannes Ruckert of Munich.

In 1893 and 1894 he produced an objective summary of the whole problem together with factual evidence in support of Weismann's hypothesis.(57) Admitting in his 1893 paper that previous work on Ascaris denied Weismann's law of reduction, Ruckert pointed out that the work of Weismann's students on the maturation division of arthropods had provided factual proof of Weismann's theoretical concepts. Both schools of thought, he pointed out, generally agreed that prior to the first nuclear division the number of chromosomes seemed to be halved. They also agreed that each of these chromosomes eventually formed a tetrad, which was constituted of four parts. Then, pointing to agreement on a third issue, Ruckert concluded: "Through two sequential mitotic divisions without a resting phase, the tetrad is divided in such a way that in each spermatid or ripe egg cell a single rod of each group arrivesas a chromosome."(58) The questions at issue, of course, were the derivation ofthe reduced number of initial chromosomes and the derivation of the tetrad. To simplify matters, each question will be dealt with separately below.

Derivation of the Reduced Number of Initial Chromosomes

This question presented no problem to Guignard, Brauer, and Hertwig. To them, chromosomes were merely transitory bodies that dissolved into the nuclear sap at the end of each cell division; they simply assumed that prior to the onset of maturation divisions half the usual number of chromosomes segmented out.

However, to those who accepted that, in the words of Boveri, "the chromatic elements are independent individuals that retain this independence even in the resting condition,"(59) this decrease in chromosome number presented a difficult problem. Rather reluntantly, Boveri accepted the position that half the chromosomes degenerated sometime before maturation division commenced. In 1890 he observed, in addition to the tetrads, two small, darkly stained bodies. "One could regard these two bodies," he wrote, "as degenerate chromosomes, and if this interpretation be correct, then the reduction would be produced when half the chromosomes atrophied." (60) Strasburger, on the other hand, argued that "the reduction of the number of chromosomes by half is due to the fusion into one of two chromatic individuals." (61)

Although, as we shall see, this position seemed to be identical to the view of Weismann's supporters, it was, in fact, fundamentally different. To Strasburger this union of two chromosomes into one was a complete fusion, or Verschmelung, such that in reality a completely new set of chromosomes appeared. The Weismann group assumed that the initial reduction in number was merely a "pseudoreduction," which appeared because two chromosomes temporarily attached to each other end to end. Thus, each of these so-called chromosomes was in fact a pair of dissimilar chromosomes. Weismann's students believed that to ensure a combination of ids during the sexual process, it was necessary to retain the individuality of each chromosome. As Haecker remarked of Strasburger's opinion, "I believe that by assuming such a fusion the process of reduction is robbed of all theoretical significance, as far as such significance bears upon the theory of heredity." (62)

Derivation of the Tetrad

Most of the more eminent cytologists of the day believed that each chromosome divided twice longitudinally to form a tetrad with four identical parts. "If it really exists," wrote Ruckert, "then one must see the essential of reduction to be the reduction of chromatin mass." The Weismann group, however, believed that only one longitudinal division took place (Fig. 6.10). Each original chromosome, which they assumed consisted of two chromosomes joined end to end, divided once longitudinally to form the tetrad. Then, during the first nuclear division the two longitudinal halves of each tetrad were distributed to each daughter nucleus and during the second division the two chromosomes that had joined end to end separated by means of a transverse division. "Such a division," Ruckert remarked, "would not only answer the requirements of all those who previously wished to see explained only the empirically established numerical reduction of chromosomes, but also would accomplish the reduction of ancestral plasms postulated theoretically by Weismann. (63) The second nuclear division had the effect of distributing half of the original and permanent chromosomes to one daughter cell and half to the other, such that a reduction in the number of ids took place. Moreover, unlike Hertwig and Strasburger, Weismann's supporters denied that any fusion took place during fertilization. To them the embryonic nucleus was not a newly generated entity but one which contained a combination of permanent chromosomes, half from each parent.

An Overview of the Controversy

At this point it may be helpful to summarize the different positions taken in the reduction-division controversy. There were two points on which all the investigators agreed: (1) prior to the first nuclear division the number of chromosomes appeared to be halved such that the number of tetrads present was always half the normal somatic number; (2) through two sequential divisions, without a resting phase, the tetrad was divided such that each sperm or ripe egg cell received a single rod from each tetrad. There was no agreement, however, on the manner in which the reduced number of chromosomes initially appeared or on the manner in which the tetrad was formed and divided.

Derivation of the reduced number. Brauer and Hertwig, who believed the chromatin rods were temporary aggregates of granules, simply assumed that the spireme segmented into half the normal number of rods. Strasburger, on the other hand, believed that the reduction in number was caused by a process of fusion, or Verschmelzung. Boveri reluctantly set forth the hypothesis that half the chromosomes degenerated before the tetrads were formed.

Weismann's group assumed that the initial reduction was a "pseudoreduction" caused by the attachment of two individual, dissimilar chromosomesend to end. This interpretation differed from that of Hertwig and Brauer in that it stressed the temporary union of permanent individual chromosomes as opposed to the random grouping of chromatin granules into transitory rods. The Weismann group also differed from the majority by denying that any actual Verschmelzung took place during the process of fecundation.

Division of the tetrad. The Hertwig group, which included Strasburger and Boveri, believed that each chromatin rod divided twice longitudinally to form a tetrad with four identical parts. All the germ cells formed from the separation of such a tetrad would be identical since the reduction was only a reduction in mass.

The Weismann group believed that only one longitudinal division took place. Each rod, which was, in fact, two chromosomes joined end to end, divided longitudinally to form the tetrad. The tetrad was thus composed not of four identical parts but of two pairs of chromosomes. During the first nuclear division the two longitudinal halves were distributed to each daughter nucleus and during the second division the two chromosomes that had joined end to end were separated by a transverse division. The germ cells did not receive identical sets of chromosomes; half the permanent chromosomes went to one daughter cell and half to the other. Thus, a reduction in the number of ids occurred.

During the 1890s no special evidence was brought forward by either of the debating groups that would compel one to assent to its view. One could assume that the process of germ-cell formation varied from one species to the next, but this would be an exception to the general uniformity of cellular structure and processes throughout the plant and animal kingdoms. The different interpretations that were put forward reflected two facts: first, that even with improved techniques, it was difficult to evaluate what one saw through the microscope; second, that the investigators differed in their fundamental assumptions about the nature of the hereditary material and about the nature of development. In the 1880s and 1890s the old issue of preformation versus epigenesis reemerged with the questions, "Is embryonic development epigenesis or evolution? Is it the new formation of complexity, or is it the becoming visible of complexity previously invisible to us?" (64) The fact that the two groups also differed over this question illustrates again the major philosophical difference between them.

An epigenetic interpretation of the developmental process had become firmly established in the nineteenth century following the collapse of preexis tence theories. As everyone realized, however, epigenesis was basically a statement of fact. It described what was seen to occur during ontogeny: a progressive and gradual appearance of parts. "There must be no mistake," wrote the Oxford zoologist Gilbert Bourne: "epigenesis is a fact, not a theory." (65) This was the basic problem, however. By the 1880s, mere formal descriptions of embryonic events no longer seemed adequate to many biologists; but causal explanations brought a return to preformationist-like doctrines. Thorough-going epigenesists simply had great difficulty explaining the causes of embryonic development. Such epigenesists, among them Hertwig and Strasburger, assumed that the embryonic nucleus consisted of basically homogeneous nuclear material produced each time afresh by the complete fusion of male and female germ-cell nuclei. According to Strasburger, Weismann's ids, visible "as discoid segments of the chromosomes," each contained all the hereditary characters-a view to which Strasburger had subscribed following his earlier work on Hemerocallis. Thus, he wrote, "the serially arranged ids in the chromosome are, in my opinion, repetitions of each other. " The hereditary particles were not, as was later believed, bearers of discrete and different characters, and neither were they, as Weismann believed, bearers of different ancestral germ plasms. Instead, Strasburger argued, "the neoplasm of many and different ancestors enters into the formation of each individual id", and their entry is brought about by "the fusion in pairs of the ids and therefore also of the chromosomes. " "I do not," Strasburger concluded, "consider that these ancestral plasms exist isolated in the id; I regard them as completely fused into one." (66) "Neopreformationists" like Weismann and Roux, however, assumed that differentiation occurred because of the gradual appearance of a latent differentiation that was already present in the embryonic nucleus and in the germ-cell nuclei. which did not fuse. As we have seen, Weismann assumed that the embryonic nuclei divided qualitatively such that the nucleus of the cells became progressively dissimilar. Cellular differentiation, then, was caused basically by this nuclear unlikeness. To Hertwig and Strasburger, however, the embryonic nuclei always divided quantitatively to produce cells with identical nuclei. In that case, differentiation could not be attributed to nuclear differentiation but came about through the interaction of cells with each other and with their environment.

Both Hertwig (who was a Lamarckian) and Strasburger attacked the preformationist position from the same perspective. "Theoretical speculation, which transcends the limits of experience," Strasburger argued, "must start from definitely ascertained facts."

Minute investigation of the longitudinal splitting of the chromosomes can but produce the impression of equal division; there is absolutely no foundation in fact for the assumption of unequal division. Hence, from the very beginning, I have taken the standpoint of epigenesis in forming my theoretical interpretation of the facts of development. The only conception of development that I am able to form is that it is a succession of stages, such that each stage determines the conditions for the succeeding stage and inevitably leads on to it. (67)

Hertwig, in his famous paper entitled "The Biological Problem of Today: Preformation or Epigenesis?," vigorously attacked Weismann's position:

To satisfy our craving for causality, biologists transform the visible complexity of the adult organism into a latent complexity of the germ, and try to express this by imaginary tokens.... Thus craftily, they prepare for our craving after causality a slumbrous pillow.... But their pillow of sleep is dangerous for biological research; he who builds such castles in the air easily mistakes his imaginary bricks, invented to explain the complexity, for real stones. He entangles himself in the cobwebs of his own thoughts, which seem to him so logical, that finally he trusts the labour of his mind more than nature herself. (68)

By 1900 there was not enough evidence either to settle the preformation-versus-epigenesis controversy or to resolve the question of chromosome individuality. Experiments on the former question yielded contradictory results for different species, and the arguments for and against individuality were based primarily on cytological observations-those of Rabl and Boveri, for example. Such observations were no more convincing than the observations of maturation division had been.

It is against this background that Mendel's work was rediscovered in 1900. It is obvious that the favored interpretation of germ-cell formation—two longitudinal divisions—would not have led anyone to see the connection between the chromosomes and Mendel's factors. If indeed chromosomes behaved in the way the majority of cytologists described, and if Mendel's factors lay on those chromosomes, then all gametes would be alike and there would be no segregation of factors, no 3:1 or 9:3:3:1 ratios.

Nevertheless, Carl Correns, one of the codiscoverers of Mendel's work, did see the connection between Mendel's results and Weismann's interpretation of reduction division. To explain such results, Correns remarked,

one must assume with Mendel, that after the union of sexual nuclei the Anlage for the "recessive" character, in our case for the green, becomes suppressed by the other "dominant" character, the gold. The embryo becomes all gold. The Anlage, however, remains; it is only "latent," and before the definitive formation of the sexual nuclei there always occurs a separation of both Anlage in such a manner that half of the sexual nuclei contain the Anlage for the recessive character, for green, and half the Anlage for the dominant, for gold. The separation takes place at the earliest in the formation of sperm and pollen Anlage. The 1:1 ratio very much supports nuclear division taking place by Weismann's reduction division. (69)

It should be emphasized again, however, that Weismann's hypothesis was little supported in 1900 and that those who did support it had worked almost exclusively on insects. Mendel's botanical colleagues were unanimously opposed to any form of reduction division. "There is no reduction division in the plant kingdom nor anywhere else," wrote Strasburger in 1894, and this was a view to which botanists subscribed even after 1900. (70)

By the turn of the century, therefore, sexual reproduction had become an issue of major importance to biologists. Although its definition had changed and now emphasized the behavior of male and female chromosomes, just how these chromosomes arose and how they acted during fecundation remained points of controversy. As a result, the significance of the sexual act remained obscure. Cytologists had moved away from ascribing the phenomenon to an act of sperm stimulation and toward a concern with heredity. Still, it was not clear whether sexual reproduction involved simply the conservative act of blending male and female characters, thereby conserving the characters of the species, or whether it was an act by which variation was introduced into the population so as to ensure continuous change or evolution. Sex had reemerged, but what it involved and what it signified remained unclear.

Literature Cited, Including Footnotes

1. For reviews of this period, see William Coleman, "Cell, nucleus, and inheritance: An historical study," Proc. Amer. Phil. Soc. 109 (1965): 124-58. Much of this and the following chapter is derived from Alice Baxter and John Farley, "Mendel and Meiosis," J. Hist. Biol. 12 (1979): 137-73

2. Arthur Lee, The Microtomist's Vade-Mecum (Philadelphia, 1885).

3. Eduard Strasburger, "Die Befruchtung bei den Famkrautem," Jahr. wiss. Bot. 7 (1869): 390-408.

4. Wilhelm His, Unsere Korperform und das physiologische Problem ihrer Entstehung (Leipzig, 1874), p. 152.

5. By the end of the nineteenth century, free cell-formation no longer referred exclusively to endogeny or exogeny but referred as well to the nucleus of a cell dividing several times in the absence of a corresponding division of the cell cytoplasm.

6. Eduard Strasburger, über Zellbildung und Zelltheilung (Jena, 1875), p. 309

7. Oscar Hertwig, "Beitrage zur Kenntniss der Bildung, Befruchtung und Theilung des thierischen Eies," Morphol. Jahr. I (1876): 347-452; 3 (1877): 1-86. Quotation from 3: 30.

8. Leopold Auerbach, Zur Charakteristik und Lebensgeschichte der Zelikerne (Breslau, 1874), p. 213.

9. Hertwig, "Beitrage zur Kenntniss der Bildung," p. 357.

10. Hermann Fol, "Sur le commencement de l'hénogénie chez divers animaux," Arch. Zool. Exper. Gen. 6 (1877): 145-69. A more extensive treatment appeared in idem, "Recherches sur la fécondation et le commencement de l'hénogénie chez divers animaux, " Mem. Soc. Phys. Hist. Nat. Geneve 26 (1879): 92-397.

11. Strasburger, über Zellbildung und Zelltheilung (1875), pp. 308-9.

12. Walther Flemming, "Beitrage zur Kenntniss der Zelle und ihrer Lebenserschemungen, Theil II," Arch. mikro. Anat. 18 (1880): 151-259. English translation by L. Pitemick, J. Cell. Biol. 25 (1965): 1-69.

13. Walther Flemming, "Beitrage . . . Theil III," Arch. mikro. Anat. 20 (1881): 1-86.

14. Eduard Strasburger, über Zellbildung und Zelltheilung, 3rd ed. (Jena, 1880), p. 321.

15. Eduard Strasburger, "über den Theilungsvorgang der Zelikerne und das Verhaltniss der Kemtheilung zur Zelltheilung, " Arch . mikro . Anat. 21 (1882): 476-590.

16. Ibid., p. 498.

17. Eduard Strasburger, "Die Controversen der indirecten Kemtheilung," ibid. 23 (1884): 301.

18. Eduard Strasburger, "über Befruchtung und Zelltheilung," Jena. Zeit. Naturwiss. 11 (1877): 508.

19. Eduard Strasburger, "über den Befruchtungsvorgang," Sitz. nied. Gesell. Nat. Heil. Bonn, in Verhandlung nat. Vereines Preuss. Rhein. Westfalens 39 (1882): 185.

20. Eduard Strasburger, Neue Untersuchungen uber den Befruchtungsvorgang bei den Phanerogamen als Grunglagefur eine Theorie der Zeugung (Jena, 1884), p. 86.

21. Eduard Strasburger, et al., Lehrbuch der Botanik (Jena, 1898), English trans. H. C. Porter (London, 1898), p. 277.

22. Francois Maupas, "Le rejeunissement caryogamique chez les cilies," Arch. Zool. Expr. Gén. 7 (1889): 149-517.

23. Strasburger, Neue Untersuchungen, p. 133.

24. Edouard van Beneden, "De la distinction originelle du testicule et de l'ovaire, caractare sexuel des deus feuillets pnmordiaux de l'embryon, hermaphroditism morphologique de toute individualite animale: Essai d'une theorie de la fécondation," Bull. Acad. Roy. Sci. Belg. 37 (1874): 530-95 25. Edouard van Beneden, "Recherches sur la maturation de l'oeuf et la fécondation," Arch. Biol. 4 (1883): 610-20.

26. Edouard van Beneden and Adolphe Neyt, "Nouvelles recherches sur la fécondation et la division mitosique chez l 'Ascaride mégalocephale, " Bull. Acad. Roy. Sci. Belg., n.s. 14 (1887): 238.

27. Van Beneden, "Recherches sur la maturation, " p. 611. 28. Edouard van Beneden, "Contributions à l'histoire de la vésicule germinative et du premier noyau embryonnaire," Bull. Acad. Roy. Sci. Belg. 41 (1876): 79.

29. Van Beneden, "Recherches sur la maturation, " p. 526.

30. Ibid., p. 527.

31. Wilhelm Roux, über die Bedeutung der Kerntheilungsfiguren: Eine hypothetische Erörterung (Leipzig, 1883).

32. Walther Flemming, "Neue Beiträge zur Kenntniss der Zelle," Arch. mikro. Anat. 29 (1887): 389-463.

33. August Weismann, "The continuity of the germ-plasm as the foundation of a theory of heredity," in Essays upon heredity and kindred biological problems (Oxford, 1889), pp. 162248.

34. Ibid., p. 188.

35. Ibid., p. 168.

36. Ibid., p. 214.

37. August Weismann, "On the number of polar bodies and their significance in heredity," ibid., pp. 335-84.

38. Ibid., p. 360.

39. Ibid., p. 355.

40. Theodor Boveri, Zellenstudien I: Die Bildung der Richtungskörper bei Ascaris megalocephala und Ascaris lumbricoides (Jena, 1887).

41. Oscar Hertwig, "Vergleich der Ei-und Samenbildung bei Nematoden: Eine Grundlage für celluläre Streitfragen," Arch. mikro. Anat. 36 (1890): 61.

42. Ibid., p. 71.

43. Theodor Boveri, "Zellenstudien III: über das Verhalten der chromatischen Kemsubstanz bei der Bildung der Richtungskorper und bei der Befruchtung," Jena. Zeit. Naturwiss. 24 (1890): p. 374.

44. Ibid., p. 385.

45. Hertwig, "Ei-und Samenbildung," p. 92.

46. August Weismann, "Amphimixis, or the essential meaning of conjugation and sexual reproduction" (1891), in Essays upon heredity, 2: 121.

47. Ibid., p. 122.

48. Ibid., p. 136.

49. Ibid., p. 132. Frederick Churchill has discussed the genesis of these ideas in "August Weismann and a break from tradition, " J. Hist. Biol. I (1968): 91-112. One break involved the reemergence of sex: from an "ill-defined blending of two stuffs" to a "unique combination of material worthy of investigation in its own nght. " This aspect of sex was elaborated on in much greater depth in Churchill's "Sex and the single organism: Biological theories of sexuality in mid-nineteenth century," Stud. Hist. Biol. 3 (1979): 139-78, and fomms the major focus of my work .

50. Hertwig, "Ei-und Samenbildung," pp. 62-71.

51. Boveri, "Zellenstudien III," p. 62.

52. Leon Guignard, "Nouvelles etudes sur la fécondation," Ann. Sci. Nat. (Bot.) 14 (1891): 170-76, 255.

53. Leon Guignard, "Le developpement du pollen et la reduction chromatique dans le Naias major," Arch. Anat. Micro. 2 (1898): 505.

54. August Brauer, "Zur Kenntniss der Spermatogenese von Ascaris megalocephala," Arch. mikro. Anat. 42 (1893): 207.

55. John Farmer, "On spore formation and nuclear division in the Hepaticae, " Ann. Bot. 9 (1895): 518; Friedrich Meves, "über die Entwicklung der männlichen Geschlectszellen von Salamandra maculosa," Arch. mikro. Anat. 48 (1897): 1-83.

56. Otto vom Rath, "Zur Kenntniss der Spemmatogenese von Gryllotalpa vulgaris mit besonderer Berucksichtigung der Frage der Reductionstheilung," Arch. mikro. Anat. 40 (1892): 102-32; Valentin Haecker, "Die Vorstadien der Eireifung," ibid. 45 (1895): 200-273.

57. Johannes Ruckert, "Die Chromatinreduktion bei der Reifung der Sexualzellen," Ergebnisse der Anatomie und Entwicklungsgeschichte 3 (1893): 517-83; idem, "Zur Eireifung bei Copeopoden," Anat. Hefte 4 (1894): 263-351.

58. Ruckert, "Chromatinreduktion," p. 576.

59. Theodor Boveri, Zellenstudien II. Die Befruchtung und Theilung des Eies von Ascaris megalocephala (Jena, 1888), p. 5.

60. Boveri, "Zellenstudien III," p. 63.

61. Eduard Strasburger, "The periodic reduction of the number of chromosomes in the life-history of living organisms," Ann. Bot. 8 (1894): 302.

62. Valentin Haecker, "The reduction of the chromosomes in the sexual cells as described by botanists: A reply to Prof. Strasburger," ibid. 9 (1895): 96.

63. Ruckert, "Chromatinreduktion," p. 580.

64. Wilhelm Roux, "Beiträge zur Entwicklungsmechanik des Embryo," Zeit. Biol. 21 (1885): 414.

65. Gilbert Boume, "Epigenesis or evolution," Sci. Prog. 1 (1894): 112. 66. Strasburger, "Periodic reduction," pp. 306-7.

67. Ibid., p. 304.

68. Oscar Hertwig, The biological problem of today: Preformation or epigenesis? trans. P. C. Mitchell (New York, 1894), p. 11. The English edition has recently been reprinted by Dabor Science Publications, Dabor Services, New York.

69. Carl Correns, "G. Mendel's Regel über das Verhalten der Nachkommenschaft der Rassenbastarde," Deut. Bot. Gesell. Berlin 18 (1900): 163.

70. Strasburger, "Periodic reduction," p. 310. This English translation omits the words "nor anywhere else," which appear in the original German article in Biol. Centralblatt 14 (1894): 851.