A COMPARATIVE CYTOLOGICAL STUDY OF...in man for the purpose of verifying the existence of asym-...

A COMPARATIVE CYTOLOGICAL STUDY OF THE NEOPLASMS OF ANIMALS AND PLANTS

MICHAEL LEVINE

From the Cancer Research Laboratory, Montejiore Hospital, New Yark

In the study of the cytology of human and animal cancer the work of Hansemann is the first to receive the attention of the cytologist. The older cytological studies on various normal and pathological tissues of Virchow (1851), Eberth (1876), Arnold (1879), Martin (1881), Cornil (1886), Klebs (1889), Schottlander (1888) and others have been reviewed so often that no further mention need be made here.

Hansemann (1890) calls attention to a number of cytological phenomena which he observed in epitheliomas, which he claimed to be distinctly characteristic of these overgrowths. He holds that while most of the cell divisions in the epitheliomas are normal, one finds a great number of tripartite or multipartite divisions. Hyperchromatic and hypochromatic mitoses appear in large numbers and monaster stages with a relatively small number of chromosomes are also to be found. The origin of the hyperchromatic nuclei is of interest, since it has been one of the most disputed points in the question of the cytology of cancer. Hansemann ascribes the origin of these cells to an asymmetrical division, i.e., the chromosomes are unequally distributed to the poles, so that one of the resulting daughter nuclei has more chromosomes than the other. In 1891-92- 93 and 1902 Hansemann reported further cytological data to support his first contentions. In these he reports on twenty other cancers of the human, in which he studied especially the telophase stages in cell division in order to determine the nature and degree of inequality of the distribution of the chromosomes after division. Hansemann stresses the point that these asymmetrical mitoses are to be found in no other normal tissues nor

11

12 MICHAEL LEVINE

in any other type of hyperplasia, inflammatory regenerative tissue, or even sarcoma and he concludes that this type of mitosis is diagnostic for carcinoma. In large cell, fast growing sarcoma, however, one finds hyperchromatic cells occasionally. Hanse- mann was unable to settle the question of the chromosome number in the somatic and germ cells in man. This difficulty has been shared by all students since 1891 (see Rappeport 1922, Conklin 1922). Hansemann counted in the hyperchromatic cells 18 to 40 chromosomes, while in the hypochromatic cells he found as few as seven.

In these later reports Hansemann adds to the conception of the origin of asymmetrical divisions. The hypochromatic nucleus may arise by asymmetrical division and through multipolar divisions. The nucleus with a large chromosome number may be reduced by the latter method to the normal number. Division of the nucleus is not always followed by cytokinesis, so that binucleate cells are not uncommon. These nuclei may again divide so as to form multinucleate giant cells.

Hansemann failed to study the nuclei in their resting condition. He emphasizes primarily the division stages and so failed to note the significance of cells with lobulate nuclei and the relation these cells bear to the hyperchromatic nuclei.

In 1904 Hansemann adds again to his conception of chromosome reduction by asymmetrical division. He holds that individual chromosomes may be destroyed. He contends that in hypochromatic cells the chromosomes may become doubled, quadrupled, etc., and so a cell with B chromosome number below normal may finally become hyperchromatic.

Muller, V. (1892)) stimulated by Hansemann’s (1890) re- searches, studied a number of malignant and benign growths in man for the purpose of verifying the existence of asymmetrical divisions. Miiller found these divisions and he reports also a transformation of a monaster into a diaster stage. His evidence on this point is, however, inadequate. Miiller further reports the presence of amitotic divisions in these growths. His evidence on this point is no more convincing than that offered by other cytologists in this field.

NEOPLASMS OF ANIMALS AND PLANTS 13

Stroebe (1892) studied pieces of human carcinoma obtained fresh and fixed in Flemming’s stronger solution. He agrees with Hansemann and Muller on the presence of asymmetrical divisions in the tissue, but in addition holds that this type of division is also to be found in sarcoma. He points out that pseudo-asymmetrical divisions occur due to differences in the rate at which chromosomes move to the poles of the spindle. He believes that many more asymmetrical divisions can be produced by sectioning than really occur. Stroebe showed that in the normal regenerating liver, Meibomian glands, skin, and brain, asymmetrical divisions may be found. These observations are opposed to Hansemann’s claims that asymmetrical divisions occur only in carcinoma. Stroebe (1893), in a later paper, concludes that in carcinoma and sarcoma asymmetrical divisions appear in abundance and a deviation from the normal recognized method of division is to be found.

Galeotti (1893) accepted Hansemann’s (1891) observations of hypochromatic and hyperchromatic cells in carcinoma. He describes with care the structure of the resting epithelial cells of breast cancer and points out that the spireme stages are the same in the normal, hypochromatic and hyperchromatic cells. The difference is to be found only in their sizes. In some cells as many as 60 chromosomes may be counted, while in others only six are to be seen. Galeotti claims that monasters are recognizable but not very clearly because of the large number of chromosomes.

Asymmetrical divisions are explained by Galeotti as arising from irregular divisions of the centrosomes. The number of spindles is determined by the number of parts into which the centrosome divides. It is of interest to note here, in view of my observations reported below, that the type of cell which gives rise to multipolar spindles has not been studied by him. He failed to differentiate the types of cells that may undergo multipartite division. Galeotti was unable to find simultaneous divisions of the nuclei in the multinucleate cells. He claims that direct nuclear divisions also appear in his preparations.

Galeotti (1893) in a later paper showed that irregularities in

14 MICHAEL LEVINE

nuclear divisions in the epithelial cells of the salamander may be induced by subjecting the tissue to dilute solutions of various chemicals.

yon Heukelom (1894) studied the cells of adenocarcinoma and found that the cells are not generally polymorphic but usually of larger dimensions than the normal cells. Giant cells, he claims, result from cell fusion, while constricted nuclei give evidence for direct division.

Krompecher (1895), in studying mitosis in multinucleate cells, used sarcoma and carcinoma cells of man. He points out that while large cells with many nuclei are put in the giant cell class, this group of cells may best be confined to large cells with a single or lobulate nucleus and the former group of cells be termed multinucleate cells. The multinucleate cells arise by amitosis or mitosis. The mitotic divisions represent progressive stages, while the amitotic divisions are stages of degeneration. Mitotic and amitotic divisions of the nuclei in a multinucleate cell may occur simultaneously. In uninucleate cells mitotic division of the nucleus follows amitosis. Krompecher (1902) later studied 313 tumors of which 178 were epithelial carcinoma, some adenomas and cystomas, 117 sarcomas and 18 endo- theliomas, and found in 78 per cent mitotic bipartite divisions of the cell. He concurs with Hansemann’s view of the asymmetrical type of nuclear division and claims that these divisions are not the products of sectioning, which he at first believed them to be.

Trambusti (1897) studied a rapidly growing melanosarcoma of a human and found two forms of multinucleate cells-those with nuclei of normal dimensions and with abundant chromatin material, and those of small size, poor in chromatin. He describes two kinds of giant cells-those that contain from four to six nuclei of the normal size and of normal chromatic content, and those in which there occur twenty to thirty nuclei. Tram- busti confirms Galeotti (1893) and Krompecher (1895) on the point that while some of the nuclei in multinucleate cells divide, others do not. Trambusti believes that giant cells are formed by the amitotic division of the nucleus in the uninucleate


cell. These are distinguishable from the other types of giant cells which arise by multipolar karyokinetic divisions, by their large number of small nuclei.

Nedjelsky’s (1900) investigations of sarcoma and carcinoma, which he made in fresh and autopsy material, were directed to the amitotic divisions in these growths. He claims the existence of numerous amitotic nuclear divisions followed by cell division. In his sarcoma material he failed to find any mitotic divisions. He contends that multinucleate cells are not the result of cell fusions, as held by Heukelom (1894), but that they arise by division of the nucleus. Nedjelsky contends that amitotic division is a progressive type of nuclear division rather than a division which leads to disintegration, as held by Galeotti (1893), Flernming (1891) and others. He fails, however, to demonstrate conclusively the presence of these divisions.

I n a communication in 1903 Farmer, Moore and Walker laid claim to the discovery of reduction divisions in cells of an invading and proliferating malignant tissue, which resembles the changes that go on in developing germ cells. These changes of somatic cells to reproductive cells, the authors claim, are to be found in sarcoma and carcinoma tissues. In EL later paper (1905) the same authors observed leucocytic invasions of the cytoplasm of cancerous epithelium. The simultaneous division of the epithelial cell and the leucocyte results in a mixing of the chromosomes of the two cells. Hansernann (1904-1905) refuted the statements of Farmer and his associates by pointing out the fact that many cancer cells contained fewer chromosomes than the normal somatic cells.

In 1906 Farmer, Moore and Walker further stressed the effect of the leococytic invasions in cancerous epithelium with a simultaneous nuclear division. This, they believed, resulted in changing the chromosome content of the invaded cell, making it equivalent to cell and nuclear fusions. The subsequent nuclear divisions were consequently reduction divisions. This point was independently stressed by Walker (1904). He holds that the heterotypic division, common to spermatogenesis of animals and plants, is found in cells of malignant growths and is not to be

16 MICHAEL LEVINE

found in the division of the somatic cells of the body, whether normal or pathological, for other causes than cancer. Klebs (1889) had previously noted the leucocytic invasion of cells in cancer tissue and explained the hyperchromatic cells on that basis. Walker contends that the heterotypic division is the sole criterion for the recognition of a cancerous growth.

Walker agrees with Hansemann (1890), Galeotti (1892) and others on the question of the origin of multinucleate cells and pluripolar spindles. He fails t o add any new data on this point. Walker (1905) further tried to strengthen his claim of the presence of heterotypic divisions in malignant cells by pointing out the presence of an archoplasmic body in these cells.

Walker and Debaisieux (1909) tried to throw further light on the cancer question by their study of the number, structure and function of the nucleolus, in resting cells of epithelioma in the human and the spheroidal cell carcinoma of the mouse. Their evidence, they believe, points to the fact that the nucleoli in cancer cells are not very unlike those in normal cells. The main difference, they contend, is to be found in the less compact and less homogeneous consistency of the nucleoli of the pathological cells.

Walker and Whittingham (1911) tried to show that while there are irregularities in the number of chromosomes in cells of malignant growths, as shown by Farmer, Moore and Walker (see Walker 1923), similar irregularities in the number of chromosomes in human festis are also to be found. The chromosome counts made in cancer cells show a predominance of the numbers 16 and 32, although the variations ranged from 11 to 50.

Hacker (1904) confirmed the results of Farmer, Moore and Walker on the reduction divisions in cells of malignant growths. He induced types of chromosomes similar to those found in carcinoma by subjecting the eggs of cyclops to ether fumes.

Bashford and Murray (1904),'* studied the zoological distribution, the nature of mitosis, and the transmissibility of cancer. They also confirmed the results of Farmer, Moore and Walker, They claimed that multinucleate cells result from nuclear division


without cell division. These nuclei may undergo mitotic division and give rise to pluripolar spindles, as held by Krompecher (1895), Trambusti (1897) and others. Amitosis occurs, but its full significance, these authors believe, has not yet been determined.

In 1905 Bashford and Murray completely reversed themselves on the question of reduction divisions in malignant growths and believe that heterotypic mitosis in cancer tissues may be construed as a somatic division with longitudinal split chromosomes. Bashford and Murray (1908) later stress their disagreement with the results of Farmer, Moore and Walker and show that, by careful staining, the ring and loop shape chromosomes in cancer cells, which are typical of reduction divisions, resolve themselves into short longitudinally split chromosomes. These authors accept the monaster stages described by Hansemann and explain it on the basis held by Galeotti (1892).

Ross (1905) studied the polarity of the cancer cell as compared with the normal cell. He contends that the cell becomes cancer-genetic during the early stages of division as a result of mutation.

Apolant and Embden (1905), in describing the presence of Plimmer bodies in cancer tissue, point out acidophyle centers with vacuoles in the nuclei of these cells. These centers they believe cause the nucleus to take on a lobulate form.

Delamare and Leciine (1906) studied 250 cases of epithelioma for the presence of giant cells. They found multinucleate giant cells deveIoping about such foreign bodies as a thread of a ligature. They contend that these giant cells are macrophages. They engulf and eliminate debris of generating cancer cells by a process of phagocytosis. Barratt (1907) studied mitotic divisions in artificialIy produced epithelial proliferations of the rabbit’s ear and found evidence of synapsis. He confirms the findings of Farmer and his colleagues on these artificially produced hyperplasias.

Winiwater (1907) in studying preparations of a breast with Paget’s disease describes two kinds of nuclear divisions. The

18 MICHAEL LEVINE

first is a normal type in which the cell is normal and active and in which nuclear division may or may not be followed by cell division. To the other kind of division belong those cells in which degeneration is taking place. Winiwater believes that Farmer, Moore and Walker were studying the latter type of cells. Multipolar divisions were found by Winiwater in these normal and degenerating types of cells. The cells in which asymmetrical divisions or stray chromosomes occurred were exclusively of the degenerating kind of cells. The lobulate nucleus, Winiwater contends, is no indication of an amitotic division.

Steinhaus (1909) studied cell and nuclear inclusions in malignant cancer. He considered amitosis a regressive process which always resulted in a polynucleate cell on the way to disintegration. His evidence is not convincing. Forbes (1909-10) induced foreign body giant cell formation by injecting agar, and lampblack and agar mixtures into the rabbit’s ear. He found no mitotic figures in the giant cells. He was unable to get evidence of direct cell division. Forbes concludes that nuclear divisions cease when endothelial cells fuse to form giant cells. Giant cells found in an epidermoid cancer of the tongue (Forbes 1910). are similar to those that form around foreign bodies and are derived from the same type of cell and formed in the same way, although the rate of development is faster in the foreign body giant cell.

Heiberg (1908-10) studied sizes of the nuclei in cells of Jensen’s mouse carcinoma and compared them with normal nuclear sizes of different tissue of the same animal. He found that the nuclear sizes in the pathological tissue were larger than those in the normal. Heiberg’s findings were confirmed by Borst (1910). Deton (1911) examined carefully four types of carcinoma among which a young epithelioma of the lower lip following leucoplacia was one. He finds no evidence of cell fusing in young cancer tissue, with leucocytes, although in older growths leucocytes are abundant. Deton agrees with Winiwater that the proof of amitotic divisions is difficult to demonstrate in cancer tissue. He rarely observed asym-

NEOPLASMS OF ANlMALS AND PLANTS 19

metrical division figures. He found pluripolar spindles but their origin, he holds, is not clear.

Lambert’s (1912) foreign body giant cell studies were done on tissue of the chick embryo in hanging drop cultures into which lycopodium spores were introduced. He shows that the giant cells result from fusions of wandering cells which gather about the lycopodium spores. Lambert describes two types of giant cells-those which have small deeply staining pyknotic nuclei and those which have large vesicular nuclei. In 1913 Lambert used a spindle cell sarcoma of the rat and a mixed sarcoma of the mouse in his culture work. He found no asymmetrical divisions although multipolar spindles with stray chromosomes appeared in stained preparations. Amitotic divisions were not observed even in stained material.

Ernst (1912) studied sphere crystals in cancer of the gall- bladder. He found giant cells there which he believed to be the result of cell fusions. Le Count (1902) has induced giant cell formation of a kind similar to that reported by Ernst, by injecting subcutaneously sterilized cholesterin crystals into the back of a guinea pig. Oshikawa (1920), in a paper on the changes in carcinoma cells after extirpation, reports binucleate and multinucleate giant cells. He concludes that there is no relationship between the appearance of giant cells and pluripolar divisions. Lewis and Webster (1921) worked with giant cells obtained in cultures from human lymph nodes in normal and pathological conditions. They believe that these giant cells result from the fusion of Iarge mononuclear cells and from nuclear divisions without cell division. No evidence of mitosis was found in these cultures. Unlike Lambert these authors report an abundance of amitotic divisions in stained preparations.

Drew (1922) found that the addition of certain amino com- pounds to a plasma medium, in which he grew various tissues, stimulated growth. He also contends that amitotic divisions are abundant in these cultures; the mitotic figures are relatively few. Plasma rendered hypotonic produces a similar effect. Drew contends that amitotic divisions are degenerative phenomena, since cultures die out earlier than those supplied with unaltered plasma.

20 MICHAEL LEVINE

Sokoloff (1922) 1* * studied the nucleo-cytoplasmic relation in cancer cells. He points out that the increase in size of the nucleus is not always a sign of disintegration and approaching death. Sokoloff believes that cells with multilobulate nuclei are degenerative types. His evidence is not very clear. He advances the idea that the source of energy which is manifested in malignant tumors and in embryonal tissue lies in the increase in size of the nucleus and in the very high coefficient of nucleo- cytoplasmic ratio.

MacCarty (1923) claims that our present cytological con- ceptions of 'neoplasms are based on artefacts. In fresh material, frozen and only slightly stained, he found no multipolar mitoses nor atypical asymmetrical divisions of any kind. MacCarty fails to figure the types of cells he finds by his method. He likewise fails to give the diagnostic traits of the various types of cells in the tumors he studied.

The cytology of the neoplastic tissue in plants has not yet been studied. So far the attention has been directed primarily to the parasite which calls forth the hyperplastic response in the host. In crown gall tissue, Smith, Brown, and McCulloch (1912) have made some observations. They claim that asymmetrical divisions, multinucleate cells and giant cells occur in this tissue. They offer this data as further evidence for their contention that crown gall is analogous to human and animal cancer and conclude that cancer is of parasitic origin.

In the other plant neoplasms which I have studied, namely, the potato-wart disease, Bally (1911) points out that the epidermal cell which becomes infected by the amoeba-like parasite, Synchytrium endobioticum (Schilb.) Perc., divides rapidly, so that the parasite comes to lie below the surface or sub-epidermal part of the neoplasm. Miss Curtis (1921) confirms this view in her extensive study on the life history of the parasite. She demonstrates several stages in the nuclear divisions of the host cells which appear to be normal. Artschwager (1923) holds that the potato-wart tissue consists of normal host tissue and only the peripheral regions contain parasitized cells.


MATERIALS AND METHODS

I have been studying for the last six years the cytology of neoplasms in man and in animals. I have also studied the overgrowths in plants for the purpose of comparing the types of cells there with those found in the animal to determine, if possible, their relationship.

For my human material I used a spindle cell sarcoma, two cases of rectal carcinoma, two cases of breast carcinoma, and three cases of epithelioma of the lip. The tissues studied were obtained in the living condition at operations and before the application of X-rays or radium was made. The specimens were cut into small pieces and dropped into a large number of fixing solutions, among which were the mixtures of Carnoy, Merkel, Hermann, Bouin and Flemming. The methods commonly used in pathological laboratories of fixing in form01 and alcohol and Zenker’s solution were also tried. The preparations which gave the best results were those fixed in Flemming’s stronger mixture and in Carnoy’s solution, although the latter fixative caused considerable shrinkage.

My plant studies were made on young and old crown gall tissues produced by inoculations with Bacterium tumejaciena Sm. and T. on a large variety of plants used in connection with my other studies of this disease. The stems of young Ricinus plants, the leaves and stems of tobacco, the roots of the beet, the leaves of Sedum, were used most frequently in this investi- gation. The hyperplasia due to the potato-wart organism Spchylrium endobioticum (Schilb.) Perc. was also studied as another type of overgrowth, in which the parasite may be observed. For this material I am indebted to Dr. Freeman Weiss of the U. S. Department of Agriculture, who was kind enough to give me a considerable quantity of it in fresh condition. The plant material was also fixed in a variety of fixing agents. The preparations fixed in Flemming’s strong solution gave the best results.

All tissues studied were carefully dehydrated by gradually exposing them to a series of graded alcohols beginning with 15 per cent. They were then cleared in xylol and infiltrated

22 MICHAEL LEVINE

with 52”-56” paraffin from 24 to 48 hours. Sections were serially cut 5 p to 12 p in thickness, The preparations fixed in Flemming solutions were bleached. in one part of hydrogen peroxide to four parts of 80 per cent alcohol for 12 to 24 hours. The sections were then stained in Flemming’s triple stain. Heidenhain’s haematoxylin and Kultschitzky’s haematoxylin were also used. Flemming’s stain, however, gave the best differentiated preparations, although Kultschitzky’s haematoxylin, counterstained with orange G or eosin, not only gave beautifully clear chromosomes but frequently made their count possible.

OBSERVATIONS ON HUMAN CANCER CELLS

The facts on the cytology of cancer, as presented by Arnold (1879), Eberth (1876), Cornil (1886), Hansemann (1891), Galeotti (1893), and Stroebe (1893), and others mentioned above, point to the existence in human cancer of a mass of polymorphic cells, which may be approximately divided into five groups.

The first group consists of small cells which are apparently normal, as shown by Hansemann, but which have assumed the power to divide. No careful study has been made of these cells but they have been mentioned incidentally in connection with the more aberrant types of cells which undergo atypical division. I have not studied these cells in great detail, but they appear to be more nearly analogous to the cells found in plant hyperplasias. The second group of cells comprise those that are perhaps somewhat larger in size than those first mentioned. The literature is not very clear on this point. The resting stages of these cells have not been carefully investigated. At- tention has been primarily fixed on the division stages. The asymmetrical mitoses in these cells (Hansemann 1890, Muller 1892 and Galeotti 1893) result in hypochromatic and hyperchromatic nuclei. The fusions of these cells with leucocytes (Klebs, 1889, Farmer, Walker & Moore, 1903) have been primarily studied. This group of cells, according to Hansemann (1891), is distinguished from all other cancer cells by their atypical divisions.


The third group of cells are readily recognizable by their large size. Krompecher (1895) has designated these as giant cells. The giant cell consists of a large cell body, many times larger than those previously described, with a single large nucleus. The life history of this type is also imperfectly known. The fourth type of cells one finds in human cancer are the multinucleate cells. The first consists of large cells, as large as the giant cells of Krompecher, and for this reason have been indiscriminately termed giant cells. They have a relatively small number of nuclei. Trambusti finds from 4 to 6 nuclei in each cell of this kind. These nuclei, it is believed by Trambusti (1897), arise by amitotic divisions, although Arnold (1879) described multiple karyokinesis and claimed that this produces multinucleate cells. A similar observation, according to Galeotti (1893) was made by Hegelman (1880) for normal plant cells. The evidence presented on this point by Trambusti and others is meagre and presumptive.

The other kind of multinucleate giant cell which is discussed in the cancer literature is the foreign body giant cell. These are large cells with many small nuclei 20 to 30 in number, according to Lambert (1913), and as many as 50, according to Reed (1902) in Hodgkin’s disease of the lymph glands. I have not met frequently with these cells in my preparations.

The fifth group of cells includes those cells of giant size with one large lobulate nucleus or several smaller lobulate nuclei. These cells are most frequently associated with amitotic divisions. The cancer literature in general is very vague on the question of the type of cells referred to when amitotic divisions are mentioned. I have assumed in this paper that when reference to amitotic division is made, these giant cells with lobulate nuclei are not specifically under consideration and that small cells with nuclei slightly constricted on one or two surfaces are implied only. These cells have been regarded as regressive stages in the life of the cell by Steinhaus (1909). According to Nedjelsky (1900), these are types of disintegrating cells and play no r61e in the further development of the growth.

I have studied several cases of rectal and breast carcinoma, a

These are of two kinds.

3

24 MICHAEL LEVINE

spindle cell sarcoma of the human body. The figures I have included in this work are from three cases of epithelioma of the iip. Yet preparations from all tissues were carefully compared. I have studied with care the giant cells, the multinucleate cells and the cells with lobulate nuclei and have examined the other types of cells in these growths, especially in the division stages, for it is in these stages that a cytological comparison between the animal and plant overgrowth may be accurately made.

THE NORMAL CELLS

In the small type of cell the resting nucleus is comparatively large and well differentiated. I t contains one or more globular or ellipsoidal nucleoli. These bodies stain a brilliant ruby red when fixed in Carnoy’s solution and treated with the safranin of Flemming’s triple stain. The chromatin material is finely divided and distributed over a dense meshwork of achromatic material, which stains a faint gentian violet. The cytoplasm is alveolar, granular, and in the late prophase stages, presents a fine fibrillar structure. No centrosomes are visible in the resting stages, as held by Deton (1911).

Very often a constriction on one surface of the nucleus (Fig. 2, P1. 1) is seen, which is similar to that described for the normal nuclei in tissues of plants. The early divisions are comparatively of long duration for, in my preparations, many spireme stages are observed (Figs. 1 and 2, P1. 1 and Fig. 11, PI. 2). Here I have not followed the transition stages in the development of the spireme band. I t appears from my preparations that the chromatin material is not derived from the nucleolus, as held by many zoologists and botanists. The bud- like projections of chromatin from the nucleolus prove in care- . fully differentiated preparations to be the result of poor fixation and staining. The spireme band breaks up into a relatively large number of chromosomes. I have not studied these cells with a view of making chromosome counts. However, the preparations of the lip and breast carcinoma were very clear and regularly showed more than forty chromosomes. The centrosome appears about the time the spireme band begins to

.


segment. Figure 10 (Pl. 2) shows the early stages in the development of the spindle. The centrosome has divided and the chromosomes are beginning to arrange themselves on the central spindle. The astral rays are abundant in this prepa- ration, and so are unlike Hansemann’s and Pianese’s figures of a similar stage. The nucleolus disappears and the chromosomes are soon found in the equatorial stage. In the anaphase stage the chromosomes appear to be evenly divided, as shown in Figure 12 (Pl. 2). In an early telophase stage, shown in Figure 3 (Pl. l ) , the masses of chromosomes appear to be fused and of equal size. The central spindle fibres are present, although not clearly visible in the picture. Furrowing of the cytoplasm is already evident. The nuclear material is soon reconstructed and the nucleoli reappear in each nucleus. The cell divides and reconstruction is then completed.

These cells in cancer tissue are apparently normal but appear to be stimulated to divide. These cells, it seems, are the only ones which are analogous t o the cells that make up the hyperplasias aasociated with crown-gall and potato-wart diseases. While the causes of the hyperplasias in the plants are known, in human and animal cancer it is one of the chief subjects of study.

Cells, which are apparently of the same size as those that go through normal divisions, described above, frequently show multipolar spindles. These are undoubtedly the cells described by all the cancer cytologists. I have also studied these cells. In Fig. 13, Plate 2, a tripolar spindle is shown, in which the chromosomes are uniformly distributed between the poles. That cell division proceeds in a normal manner is indicated by Fig. 4, Plate 1. The stage is an early telophase with part of the chromosomes at one pole appearing in the next section of the series; quadripolar spindles also appear in these cells. The distribution of the chromosomes to the four poles is equal. Simultaneous quadrupartite cell divisions also appear in these cells of normal size.

What appear to be asymmetrical divisions in my preparations are generally due to sectioning. I cannot agree with Hanse-

26 MICHAEL LEVINE

mann (1890), Galeotti (1892), Stroebe (1893) and others in their claims of the abundance of this type of division. Several carefully made sections of small pieces of lip epithelioma show very few unquestionable asymmetrical divisions. In this I am in accord with Deton (1911). Monaster stages, as recorded by Hansemann (1890) and confirmed by many other workers are rare in my materials. Those that appear there are suggestive in appearance of a section cut through a dividing cell in which the chromosomes have not yet reached the poles nor have the centrosomes disappeared; yet the constriction for cell division has been almost completed.

THE GIANT CELLS

I have studied the giant cells which have a single large nucleus to determine, if possible, their origin. I have found nothing that suggests that these cells are the result of the fusion of smaller cells. Nor does the nucleus in these large cells suggest that it arises from fusions of smaller nuclei. This does not imply that such fusions may not occur. Under special physiological conditions, such as exist in tissue cultures, as shown by Forbes (1909), Lambert (1913) and Drew (1923) cell fusions do occur. No evidence however is found for nuclear fusions, in my preparations.

It appears that these uninucleate giant cells are hypertrophied cells due possibly to increased nutrition, as suggested by Schottlander (1888) or resulting from unknown causes. Fig. 5, Plate 1, shows such a large cell with the nuclear material bounded by a uniformly smooth and unfurrowed membrane. A large nucleolus, with two smaller nucleolar bodies on either side, also appears. An abundance of chromatin granules are scattered uniformly through the delicate nuclear reticulum. The cytoplasm is granular and is bounded by coarser granules lining the cell membrane. The cell appears to be normal in all respects, except for its size.

Further observation made on these uninucleate giant cells suggests very strongly that these may develop into giant cells with lobulate nuclei such as I am describing below. In Fig. 6,


Plate 1, another uninucleate giant cell is shown, in which the nuclear membrane shows a marked tendency to furrow and to grow in the areas between the furrow. This is suggestive of an early stage in the development of the lobulate nucleus. I have not particularly stressed the study of these stages in my preparations, but I have found all gradations between the giant cell with a single, smooth and unfurrowed nuclear membrane to stages shown in Fig. 7, (Pl. 1) and Fig. 14 (PI. 2), in which the nuclear membrane becomes deeply furrowed forming compound lobes.

Giant cells, with what appears to be a series of nuclei arranged in the shape of a ring (Fig. 8 P1. l), are of common occurrence in the carcinoma tissues I studied. This arrangement of the nuclei in these giant cells has been used as evidence of the presence of amitotic divisions. The part of the nucleus to the left in the photograph shows the occurrence of a constriction. There is no evidence, however, for the complete severance of this part of the nucleus. The part of the nucleus to the right in the photograph appears to be separate and distinct from the part to the left. The facts made clear by my preparations are that there is only one nucleus in the cell, and that this is only another type of a lobulate nucleus. It appears that this is a type of nuclear growth and lobulation which arises from one or two opposite points on the nucleus. In the other types of lobulate nuclei growth and constriction occur at a number of points, thus giving rise to the multilobulate effect such as is shown in the early stages in Figs. 6 and 7 (Pl. 1).

The multinucleate giant cells which occur in the specimens of carcinoma on which I am reporting, show large cells in which two to six nuclei are found. In Fig. 9 (Pl. 1) one of a series of three sections of such a cell with five nuclei is shown. They appear to be grouped in the center of the cell partially over- lapping each other. The non-nucleolate body is a part of a nucleus and not an inclusion. The nuclei are well differentiated. The nucleoli appear as spherical bodies with small vacuoles. The whole body of the nucleolus stains brilliantly with safranin. The chromatin material appears to enmesh the nucleoli. It is

28 MICHAEL LEVINB1

made quite distinct in my preparations, however, by the gentian violet stain it takes.

The multinucleate cells according to Bashford and Murray (1905) originate from cells in which nuclear division is not followed by cell division. Krompecher (1895) earlier figured one of a number of such nuclei in a division stage and contended that the multinucleate cells arise in this manner. The multipolar spindles reported by Hansemann (1891), Galeotti (1893),' Krompecher (1895), it is claimed, are due to the simultaneous division of these nuclei. It appears to me that the hyperchromatic cells may possibly be due to the simultaneous division of these nuclei. However, neither my preparations nor the reported results of the earlier workers are entirely conclusive on this point. The prophase stages in the division of these multinucleate giant cells are lacking. My observations on the mitotic divisions of giant cells with lobulate nuclei, reported below, give more definite and convincing evidence that the hyperchromatic cells with multipolar spindles arise from these cells.

LOBULATE NUCLEI IN GIANT CELLS

The question of the significance of nuclei, with constrictions on opposite surfaces, has not been definitely answered; much less has the meaning of numerous furrows and constrictions in the giant nuclei in cancer been carefully considered. The latter type of cell has been observed by the zoologists and animal pathologists, Werner (1886), Pfitzner (1886), Hess (1890), Stricht (1891), Flemming (1892) and others in normal and pathological tissues. The interpretation put on these constrictions and furrows has been amitosis. In the giant cells of cancer these lobulations have been generally accepted as evidence of degeneration.

The more recent zoologists, who have studied these types of cells in the various tissues of the invertebrate snimals, have associated the lobed nucleus with increased metabolic processes. Child (1907) contends that amitoses occur frequently in rapid growths and mitoses are to be found more frequently in slower growths. This author, from the study of Monezia believes that


mitotic divisions follow after long periods of direct divisions. Child further advances the view that amitotic divisions are associated with metabolic processes, such as are involved in the secretion and the storage of materials.

Hargitt (1905) in his studies of the egg of Clam Zeptostyla found no evidence of mitoses up to the sixteen cell stage. He assumes that the mitotic division is disguised and possibly lacking. Conklin (1912) pointed out in Crepidulu that many abnormal mitoses appear superficialIy to be amitoses. In 1917 Conklin further contends that the occurrence of amitoses in pathological and degenerating cells is undoubted. Yet the number of cases in which amitosis is known to occur is too great to believe that it is always a case of a pathological phenomenon.

Glaser (1908) studied the entoderm of Fasciolaria and claims he found that nuclear nests arise by amitotic divisions of a multilobulate nucleus. Yet, he states that the mere lobulation of the nucleus does not indicate division. Glaser maintains that the indirect divisions are really the important ones, while the direct divisions are physiological. Macklin (1916) concludes that the multilobulation of the nucleus is a pathological condition due to the unfavorable conditions of the nucleus. Macklin holds that fragmentation of the nucleus has been confused with amitosis and this, in his opinion, accounts for the view held by many observers that amitosis is evidence of degeneration.

Beer and Arber (1919) cite Meckel (1846), Zaddach (1854) and Korschelt (1891) as having shown that cells of silk glands of various insect larvae, characterized by great secretory activity, are associated with much lobed and branched nuclei.

Nakahara (1918) studied the physiological relation of amitosis in the adipose cells of insects. He contends that amitosis occurring in secreting or reserve forming cells secures an increase of the nuclear surface to meet the physiological require- ments of an active metabolic interchange between the nucleus and cytoplasm. Nakahara maintains, however, that amitosis is not a method of multiplication nor a sign of degeneration or senescence of cells but indicates, wherever it occurs, an intense activity in the vegetative functions of the cell.

30 MICHAEL LEVINE

While no evidence is at present available on the physioIogical behavior of the cells with lobed nuclei in cancer, my preparations indicate quite clearly that these cells are not disintegrating but are taking part in some dynamic process associated with growth and building of the neoplasm. The evidence shows further that these cells with lobed nuclei are not dividing by amitosis, but indirect or mitotic divisions undoubtedly occur here. I am in accord with Winiwater (1907) and Deton (1911) on the view that the evidence of amitosis in cancer is presumptive. The large number of hyperchromatic giant cells, with pluripolar spindles, are in the main the division stages of these cells and not the result of asymmetrical divisions as maintained by Hanseniann, Muller, Stroebe and others. It appears further that any cell in carcinoma tissue may undergo disintegration, as it does in any pathological condition in animal or plant tissue. The cells with lobulate nuclei are not specially consigned to this fate. There is no evidence in my preparations nor have I been able to find in the literature evidence which shows that all cells in cancer tissue must pass through the stage in which they show lobulate or constricted nuclei before they disintegrate.

I have studied very carefully, as mentioned above, the giant cells with lobulate nuclei. It appears that these nuclei arise by growth and constriction of the nuclei in a uninucleate or multinucleate giant cell. One or several areas in the nucleus and nuclear membrane may grow while other portions remain unchanged. This forms a one or more lobed nucleus depending upon the number of growing areas. Constriction with little or no growth brings about similar effects. The compounding of the lobes results from growths and constrictions in the first lobes. Such a compound lobed nucleus is represented sche- matically in Fig. 14 (Pl. 2). It represents one of a series of four sections. The cell is one of many that I found in a very actively growing part of a carcinoma of the lower lip. This tissue, fixed and stained as mentioned above, showed in section brilliantly colored chromatic and nucleolar material. The nucleoli have one or more small vacuoles. Each nuclear lobe has a nucleolus. The nucleoli appear to develop de novo in


each lobe. The chromatin is evenly distributed through the lobe in a fine delicate network. The cytoplasm consists of a fine granular substance with a number of congregates of granules sparsely distributed through the cell. I have not been able to identify these with centrosomes, although occasionally ex- tremely delicate streams of cytoplasm seem to come from them. Small vacuolar areas are found towards the periphery of the cell. The entire cytoplwm beoomes a faint orange with Flemming's triple stain. Within the cell membrane masses of long fibrillar bodies are observed. This is undoubtedly indi- cative of cytoplasmic activity in this portion of the cell.

I am basing my contention that these cells divide indirectly on stages such as are shown in Figs. 15 and 16 (Pl. 2). These are critical stages and while they are not present in abundance, a sufficient number of them may be found. Fig. 15 (PI. 2) represents one of a series of four sections in which the chromatin material consists of densely colored granules arranged on a faintly stained linin reticulum. Here and there the chromatin granules have fused to form a portion of the spireme. The disappearance of the nuclear membrane has progressed con- siderably but traces of it may still be seen, which point un- deniably to a n u c l e u ~ ; ? f , f i t @ ~ ~ t $ e ~ . The nuclear material seems to have cori%i.d&dz sQ G*tb'di;h a concentrated nuclear body. A $i&@$ .df klger &knh&; n@y*:@3ear in the cytoplasm, whk$:&*&l$ ' s ' r i g g e s ~ i ~ ~ d ' f . c ~ ~ t ~ ~ 6 m ~ s . The nucleoli do not appear in thig s@Qf$@. >.I#:B& 16 (Pl. 2) several phases of the spireme band:form'atiwn md:segmentation may be seen. The chromatin granules have congregated toward the periphery of the nuclear mass. Toward the center large portions of the spireme band are seen, while scattered in between are smaller segments of it. These are undoubtedly chromosomes as they appear on the spindle shown in the succeeding stages. The nucleolar bodies are less heavily stained and are in a regressive state. The nuclear membrane has entirely disappeared. A fibrillar zone similar to the one shown in the resting stage, Fig. 14 (Pl. Z), is present. Here it is much wider and appears to bound the nuclear mass. The cytoplasm is granular in the part near

. . * :. ... . .

32 MICHAEL LEVINE

the cell membrane, while it takes the form of a fine, delicate network near the fibrillar zone.

Fig. 17 (Pl. 2) represents the next stage in the mitotic division of the multilobulate nucleus. This figure is as far as I have been able to determine similar to those generally spoken of in the literature as the cells with multipolar spindles and the hyperchromatic cells. Their derivation has been inadequately shown. As I have pointed out above, there is no other type of cell equal in size to the giant cell with lobulate nucleus that could possibly give rise to this stage, excepting the uninucleate and multinucleate giant cells. But in the latter cases no prophase stages have been shown which indicate that these cells divide mitotically. While this does not preclude these types of cells from dividing mitotically, it appears quite definitely that the giant cells with lobulate nuclei undergo indirect division. Fig. 17 (Pl. 2) shows three groups of chromosomes distinctly oriented about three faintly indicated centers. Four other groups of chromosomes not visibly oriented in this section form parts of chromosome masses in the other sections. While the stage is not altogether clear, it is apparently an anaphase in which the chromosomes are moving toward the poles. The spindle fibres are not y@y -pr~@@!y thained and remain invisible, except at the ;f;de:;h'he: clfrdz&dmes do not move

These cells seem:irb.I&+hv6dt&by bhi&'*m*&Hheb almost as prominent as the cell walt*.)lt:dAnfi$'.:;4I+g the inner wall of this membrane a number 6f* vacuole&.ap&w of various sizes. The fibrillar zone is also present but it is now much thinner with relatively fewer fibrils.

The size of the cell makes it impossible to trace its division with certainty. That cell division occurs in the giant cells, described above, cannot be doubted. The preparations of freshly fixed carcinoma of the breast, and especially the lip material studied, show that the thickened membranes which bound large masses of cells, such as described above for Fig. 17 (Pl. 2) circumscribe an isolated mass of cancer tissue. It appears that such dividing cells as shown in Fig. 17 (Pl. 2)

. . , a * . . , :.."' ** .,. . ....:. ..* .: to the poles unifdf.)aiy.:{ f ;;**.***; : .-, .. ... ... ..


give rise simultaneously to such clusters of cells as shown in Fig. 18 (Pl. 2) and the clusters a, b and c have had an im- mediate common origin. This may explain the rapid growth that takes place in certain types of carcinoma.

NEOPLASM8 IN PLANTS

Crown Gall Smith

and his associates (1912), on the basis of a histological study of crown gall contend that multinucleate cells, giant cells and asymmetrical divisions are found here as in animal cancer. I have studied sections of a large number of preparations made in connection with other investigations on crown gall (Levine, 1919 to 1924).

Crown gall on the tobacco, beet, Ricinus, Sedum, geranium, tomato, nasturtium, and bean were especially used for the present study. Drawings from the cells of the beet, tobacco, Ricinus and Sedum are, however, only used in this work. Special emphasis has been placed on the study of the nuclear divisions to prove, if possible the existence or unequal or asymmetrical divisions. The so called embryomata, described by Smith (1917), which appeared in my cultures (Levine 1919 to 1924) have also been studied cytologically to determine the presence of aberrant nuclear phenomena there. Embryonic tissue of growing tips of the plants studied were used as controls.

The method of inducing crown gall formation has been stated too often to need mention here. I have pointed out elsewhere that crown gall in the early stages of its development consists of small cells, embryonic in appearance. These cells do not continue to proliferate indefinitely. The crown gall has only a relatively short period of progressive proliferation of its un- differentiated cells. Differentiation soon ensues and adult cells appear, This is followed by regression and very often death before the host’s destruction is complete. In malignant neoplasia the proliferation is progressive and has no finality. Its death occurs with the death of the host.

No cytological data on crown gall tissue are available.

34 MICHAEL LEVINE

My drawings are made from cells in the actively growing portions of the overgrowths. Large mature cells with resting nuclei are found in young crown galls induced by inoculating B. tumefuciens into the common red beet. The nucleus is well differentiated and the distinct chromatin bodies are constantly present in the resting nucleus. They resemble the prochromo- somes of Rosenberg (1904) and Overton (1905), and those that appear in my own studies of the American Droseru intermedia (Levine 1918). Their arrangement is quite suggestive of that shown by Stout (1912) for Curez. I have not followed these chromatin masses, however, in their development. A large nucleolus of a homogeneous consistency is found in the cell. Two nucleoli in each nucleus are not uncommon in these preparations. The cytoplasm is granular and has several large vacuoles in it. The appearance of the cell is quite normal.

The prophases which appear here do not vary from the somatic divisions found in other normally dividing plant cells. Segmented spireme stages occur frequently. A slightly later stage shown in Fig. 22 (Pl. 3) is drawn from a crown gall on Ricinus. The two dense conical areas on opposite sides of the nuclear membrane in this figure indicate the anlagen of the polar caps. These consist of a granular substance in which fibrillar bodies are beginning to form.

In the crown gall of the beet and the tobacco, spindles figures with chromosomes in the equatorial plate stage are frequently found. These figures were carefully studied in serial sections and there is no evidence to show any aberrant condition. As the chromosomes move toward the poles their distribution appears normal and in no way suggests an asymmetrical division. Belated chromosomes on the spindle occasionally appear. This phenomenon has been observed in a great variety of tissues other than crown gall and has been discussed by me elsewhere (Levine, 1918).

Fig. 19 (Pl. 3) represents an early telophase stage in the division of a cell in the crown gall of the beet, Here the chromosomes have reached the poles. The chromosomes are still distinct and their masses are approximately equal in size. In

NEOPLASMS O F ANIMALS AND PLANTS 35

the crown gall on the leaf of the Sedum plant I have found division figures in the early telophase stage in young and old cells. Its cytoplasm is of a very dense granular consistency. It has a few small vacuoles and represents a typical embryonic cell. The two chromosome masses shown a t the poles here, and in many other cases of the same type, are normal. Large cells from the same crown gall show similar stages in nuclear division. The increase in age of the large cells is marked by their size and their vacuoles. While I have not studied particularly the nuclear cytoplasmic relation, it appears to be the same here as in the smaller cells similar to that shown in Fig. 24 (Pl. 3).

Fig. 20 (Pl. 3) represents a stage in the reconstruction of the daughter nuclei and early stages of cell division. These cells are drawn from crown galls of the beet. Similar stages have been found in abundance in Ricinus. In the above figure the nuclear reconstruction has progressed to the point where the chromatic bands and nucleoli are formed. In Fig. 20 the young cell plate has not yet completely divided the cell body. The cell plate formation fails frequently of completion, as I shall describe below. Binucleate cells result from this. The figures described above are found to be identical with those found in the apical or embryonic portions of normal plants.

Fig. 24 (PI. 3) represents a very young cell.

BINUCLEATE CELLS IN PLANTS

The cytological evidence, as found in my preparations of crown gall tissues, shows conclusively that the young, actively growing cells there are normal, and that all division stages in this tissue are identical with those found in normal somatic tissue. Cell plate formation may occasionally fail of completion and thus give rise to binucleate cells. Even this phenomenon has been observed in normal plant cells, as shown by Beer and Arber (1919), although the method of origin of these cells is a mooted question. A review of this literature on the multinucleate cells is given by Beer and Arber and need not be repeated here. I am only giving that phase of the question which may have a possible bearing on the present problem.

36 YICHAlL LEVINE

While Nageli was the first to formulate the rule of uninucleate cells, Schmitl;, Treub and Strasburger were among the first to claim the presence of multinucleate cells in certain species of plants, McLean (1914) found that amitosis occurs commonly in the cortical parenchyma of aquatic angiosperms, such as Myriophyllum and Hippuris and suggests that this is a characteristic of both dicotyledons and monocotyledons. Cell divisions follow later on the nuclear divisions. Prankerd (1915) considers vegetative cells as multinucleate. This author has observed two nuclei in the somatic cells of thirty-six species of plants. Some plants contain three nuclei while in Morw there are many and may best be described as a “nuclear complex.’’ Prankerd also maintains the view that these nuclei arise by amitosis. Cell walls are formed later, so that ultimately the cell contains a single nucleus.

Beer and Arber (1915) contend that in the case of cortical and medullary parenchyma of stems, a stage in which each cell characteristically contains more than one nucleus intervenes as a normal phase of development between meristematic and mature conditions. The multinucleate cells generally arise by mitosis. However, in some cases amitosis may occur followed by cell division. The view is maintained that these processes are a means of tissue formation in rapidly growing organs. In 1919 Beer and Arber stress the presence of mitotic divisions in these somatic tissues and claim that the cell plate fails to develop. The spindle becomes reabsorbed and the cell becomes binucleate.

These authors describe lobed nuclei in Helminthostachys. These are not nuclear fusions nor amitotic divisions. They represent a change of form which is the expression of increased metabolic activity. Unfortunately, the figures of Beer and Arber on the study are inadequate. Their description leads one to infer a marked analogy between these lobulate cells in this normal tissue of Helminthostachy8 and those described by Nakahara (1918) for Fasciolama.

Arber in 1920 reinvestigated the nuclear phenomena in the pith and cortical cells in Morus and found that the so called “nuclear complex” of Prankerd waa due to mitotic divisions.


My observations on the crown gall tissue are in accord with the later observations of Beer and Arber (1919). Cell divisions in the crown gall tissue frequently fail to occur. Fig. 21 (Pl. 3) represents a binucleate cell in the crown gall on Ricinus. The method of development of these binucleate cells is not by amitotic divisions, as maintained by Prankerd (1915), but as mentioned above by the failure of the cell plate to complete its development. The part that is formed is slowly absorbed and disappears. This process is very suggestive of Farr’s (1918) observation on the spermatogenesis of a large number of dicotyledons. In these preparations the cell plate pro- gresses further in its development than those in the germ cells described by Farr. I have not seen cell constrictions occur after nuclear divisions, as maintained by McLean. Bi- nucleate cells are of frequent occurrence.

When the cell plate is in the process of formation, the nuclei approach each other. Frequently, two nuclei lying in the center of the cell, are flattened out against each other and the remnant spindle fibres may be seen on both sides of them. Very often these remains also disappear and the two nuclei, pressed against each other, are seen in the cytoplasm. That the nuclear membranes at the points of contact may be absorbed, as in the nuclear fusions in the higher fungi, is very suggestive (Levine, 1913). A more plausible interpretation of Fig. 23 (Pl. 3), drawn from the crown gall of the Sedum leaf, is that the nuclear membranes between the two touching nuclei has been somewhat understained. Such figures are not common but have led to the belief that fusion and amitotic division occur here. Quadri- nucleate cells occasionally appear in the crown gall tissues studied. The nuclei lie in groups of two. I have not found more than four nuclei in these cells. A large majority of these cases appear to be questionable.

LOBED NUCLEI

Cells in older portions of crown gall tissue not infrequently show nuclei with constrictions on one surface such as appear in normal animal and plant tissues. From my preparations of

38 MICHAEL LEVINE

crown gall no evidence is found to show they are dividing amitotically. In carcinoma Fig. 2 (Pl. 1) and Fig. 11 (Pl. 2) I have shown that these cells divide mitotically. These cells are commonly found in all types of tissues, animal and plant, and they are not to be associated with any one pathological process. Cells with multilobed nuclei, as reported by Beer and Arber (1919), do not appear in the crown gall or in the normal tissues of the plants I studied.

Large hypertrophied cells are occasionally found in sections of tissue of the tobacco. They differ from other cells only in size. There is no evidence that these cells are aberrant and unlike large cells in normal tissue. They are suggestive of the uninucleate giant cells in animal cancer. I have also found what appears to be an abnormality in the cells of the crown gall on the beet. It appears that fragmentation of the chromatic or nuclear material takes place. Each fragment of deeply staining material is surrounded by a hyaline area. These figures are suggestive of the abnormalities in apparently normal tissue described by Beer and Arber (1919). I am not at all sure that these are not found in the normal root of the beet.

It appears from these studies that only those types of cells that are found in the normal tissue appear in the pathological plant tissues. Crown gall is a protective reaction of the plant against the invading parasite. In the early stages the tissue is uniformly embryonic in appearance; as it grows older the characteristic cell structures are produced.

POTATO-WART

The quarantine of potatoes infected with potato-wart limits, to a certain extent, the quantity of available material for this study. However, on the basis of this material, Levin and Levine (1922) have pointed out that this tissue which de- velops about the parasitized cells in this disease, constitutes a real plant neoplasm. It is in these cells that I have studied the types of division to detect any abnormalities and to compare them with human cancer cells. The cells and the division of these cells under the influence of the parasite are normal and

NEOPLASMS O F ANIMALS AND PLANTS 39

compare favorably with the studies of Ntimec (1899) on the cytology of the normal potato, Solanum tuberosum. The cell behavior of the potato wart from my observations are like those reported by Miss Curtis.

I have stressed in my studies of this disease the spindle figures and the prophase stages. In these cells I have found nothing that is suggestive of the cells of the malignant neoplasms of animals. Unlike animal cancer, these stimulated cells are limited in their proliferating powers and are thus similar to those of the crown gall. The epidermal cells that become stimulated divide rapidly for one or more generations. They then become large and develop into mature parenchymatous cells and may even serve for a time as storage organs.

The cytoplasm of the old cells has a number of large vacuoles which indicates its approaching maturity. A number of starch grains are also present. In young cells the cytoplasm is made of a granular reticulum with few vacuoles and no starch grains. The division and distribution of the chromosomes are normal and not unlike the stages found in normal and crown gall tissue, mentioned above. The division of the parasitized cell is also normal. In Fig. 26 (PI. 3) an epidermal cell is shown infected by two parasites. The host nucleus is in late telophase. The masses of chromosomes at each pole are equal. The nuclear division is followed by cell division, thus removing the parasitized cell from the surface of the tissue. I have seen a number of divisions of the host cell, but have found nothing that suggests any aberrant condition.

*

SUMMARY AND CONCLUSIONS

1. The normal type of cell that appears in human cancer is distinguished from the other cells on the basis of size and normal type of karyokinetic division.

2. Atypical divisions, such as tripolar and quadripolar spindles are found in what appear to be normal sized cells. I have found no evidence of asymmetrical divisions in these cells and no indication that these cells give rise to hyperchromatic or hypochromatic daughter cells, as maintained by Hansemann and others.

4

40 MICHAEL LEVINE

3. Uninucleate or multinucleate giant cells constitute an important part of cancer tissues. That these cells may give rise to pluripolar spindles and hyperchromatic cells is likely, although no adequate evidence appears in the literature, nor have I found conclusive evidence for these conditions in my preparations.

4. The giant cells with lobulate nuclei make up one of the most important groups of cells in malignant cancer. These nuclei appear to develop from nuclei in single and many nucleate giant cells.

5. I am contending that the cells with lobulate nuclei have not been adequately studied and their significance has not been understood.

6. These cells are not degenerating cells, as claimed by earlier workers, nor is the evidence conclusive that they are dividing by constriction or amitosis.

7. These cells are found in actively growing parts of the carcinomas reported above, and their nuclear structure appears to be associated with some increased metabolic process involved in the growth of the malignant neoplasm.

8, The giant cell with the multilobulate nucleus appears to play an important r81e in the growth and development of the malignant neoplasm.

9. The giant cell with the multilobulate nucleus divides mitotically. The mitotic divisions in these giant cells with lobed nuclei may be entirely responsible for hyperchromatic cells and pluripolar spindles in them.

10. Malignant neoplasms in man are characterized by rapidly dividing cells with aberrant types of division. The cells that make up the growth have developed in situ or have come through the blood and lymph channels from distant parts of the body. These cells are endowed with a progressive proliferating power and die with the host.

11. Plant overgrowths resulting from infection with Bacterium tumejaciens Sm. and T. and Synchytrium endobioticum (Schilb.) Perc., as reported above, are due to rapid but normal division of cells.


12. Atypical divisions are not found in these tissues. 13. Binucleate and multinucleate cells may appear in crown

gall tissue due to the failure of cell plate formation. 14. No type of cell or cell divisions, e.g., giant cell, multi-

nucleate cell, tripolar or quadripolar divisions, appear in these overgrowths, which is not present in the normal tissue in the region of inoculation.

15. All cells in infectious plant neoplasms differentiate, reach maturity, become senile, and die independently of the host.

I wish to express my gratitude to Dr. I. Levin, former Chief of the Cancer Division, for the human material used in this work.

1905. APOLANT, H., AND G. EMDEN:

LITERATURE CITED

Ueber die Entstehung Plimmer'scher Korperchen aus Kerndegenera- tionen. Zeitschr. f. Krebsforsch., 3: 579-580.

1820. ARBBIR, Aams: Studies on the binucleate phase in the plant cell. Jour. Roy. Micr. SOC.,

1-21, figs. 1-2, pl. 1. 1879. ARNOLD, J.:

Beobachtungen tiber Kerntheilung in den Zellen der Geschwulste. Virch. Arch. 78: 279-301, pls. 5-6.

1884. ARNOLD, J.: Weitere Beobachtungen uber TheilungsvorgtLnge an den Knochenmark-

zellen und Weissen BlutkGrperchen. Virch. Arch., 97: 107-131, pl. 4.

Anatomical studies on potato-wart. Jour. Agr. Res., 23: 963-967, pl. 1-5.

Cytologische Studien an Chytridineen. Jahrb. f. Wiss. Bot., 50: 95-156,

1923. ARTSCHWAOER, ERNST F. :

1911. BALLY, W.:

figs. 1-6, p l ~ . 1-5. 1907. BARRATT, J. 0. W.:

1904.1 BASHFORD, E. F., AND J. A. MURRAY: On mitosis in proliferating epithelioma. Proc. Roy. SOC., L., 79B, 372-377.

The significance of the zoological distribution, the nature of the mitoses and the transmissibility of cancer. Brit. Med. Jour., L., 269-271, 413-416; Proc. Roy. SOC., L., 73: 66-77, Figs. 1-8! pl. 2.

Comparative cytological characters of malignant new growths. Sci. 1904.2 BASHFORD, E. F., AND J. A. MURRAY:

Reports, Imperial Cancer Res. Fund, London, 1: 16-36, figs. 6-35. 1905-6. BASEFORD, E. F., AND J. A. MURRAY:

On the occurrence of heterotypical mitoses in cancer. Proc. Roy. SOC., L., Ser. B., 77: 226-232, pls. 5-6.

1908. BASEFORD, E. F., AND J. A. MURRAY: On the occurrence of heterotypical mitoses in Cancer. Sci. Repork,

Imperial, Cancer Res. Fund, London, 3: 61-68, Figs. 1-15.

42 MICHAEL LEVINE

1915.

1919.

1920.

1910.

1907.

1907.

1912.

1917.

1922.

1886.

1921.

1906.

1911.

1922.

1876.

1912.

BEER, R., AND A. ARBER: On the occurrence of binucleated and multinucleated cells in growing

tissue. Ann. Bot., 29: 597-8. BEER, R., AND AGNES ARBER:

On the occurrence of multinucleated' cells in vegetative tissues. Proc. Roy. Soc., L., Ser. B, 91: 1-17, pl. 1.

BEER, R., AND AGNES ARBER: On multinucleate cells; an historical study. (1872-1919). Jour. Roy.

Micr. ,Sot., L., 1: 23-31. BORST, M.:

CHILD, C. M.: Die Kerngrhseder Krebszellen. Sitz. d. Phys. Med. Gesel. 1;. Wiirzb., 24-26.

Studies in the relation between mitosis and amitosis. I: Development of ovaries and oogenesis in M o w z k . Biol. Bull.,

11: Development of the testes and spermatogenesis in Monezia. 12: 89-114, fig. 1-35.

Biol. Bull., 12: 191-212, pls. 11-16.

Amitosis as a factor in normal and regulatory growth. Anat. Anz., CHILD, C. M.:

30: 271-297, figs. 1-12. . CONKLIN, E. G.:

Experimental studies on nuclear and cell division in the egge of Crepidukz. Jour. Acad. Natl. Sci., Phila., 15: 503-591, pls. 43-59.

CONKLIN, E. G.:

CONKLIN, E. G.:

CORNIL, V.:

Mitosis and amitosis. Biol. Bull., 33: 396-436, pls. 1-10.

Heredity and environment in the development of men. N. Y., p. 162.

Sur le prodd6 do division indirect des noyaux et des cellules epithdliales dans lee tumeurs. Arch. de phys. norm. et path., Ser. 3,s: 310-324, pls. 10-11.

CURTIS, K. M.: The life history and cytology of Synchytrium endobioticum (Schilb.)

Perc., The cause of the wart disease in potato. Phil. Trans. Roy. SOC., L., Ser. B., 210: 409478, pls. 12-16.

DELAMARE, G., AND P. LEC~NE: Sur la pr6sence de cellules &antes dans les cancers 6pithbliaux. Arch. de

MBd., Exp. et d'Anat. Path., 18: 102-108, pl. 1. DETON, W.:

Contribution A I'btude cytologique du cancer. La Cellule, 27: 27-52, $8. 1-3.

DREW, A. H.: A comparative study of normal and malignant tissues grown in artificial

culture. Brit. Jour. Exp. Path., 3: 20-27, pls. 1-5. EBERTH, C. J.:

ERNST, P.: Ucber Kern und Zelltheilung. Virch. Arch., 67: 523-541, pls. 18-20.

Sphkoide und Spharokristalle in Krebs und Riesenaellen. Beitr. z. path. Anat. u. z. allg. Path., 53: 429-445, pls. 18-21.


1903. FARMER, J. B., MOORE, J. E. S., AND C. E. WALKER: On the resemblance exhibited between cells of malignant growths in

man and those of n o d reproductive tissue. Proc. Roy. SOC., Lon-

1905. FARMER, J. B., MOORE, J. E. S., AND C. E. WALKER: don, Series B. 72: 499-504.

On the behavior of leucocytes in malignant growths. Tr. Path. Soc., London, 56: 377-380, pl. 3.

1905-6. FARMER, J. B., MOORE, J. E. S., AND C. E. WALKER: On the cytelogy of malignant growths. Proc. Roy. Soc., London, Ser.

B., 77: 336-353, pls. 8-12. 1916. FARR, C. H.:

Cytokinesis of the pollen-mother-cells of certain dicotyledons. Mem. N. Y. Bot. Garden, 6: 253-317, pls. 27-29.

1879. FLEMMINO, W.: Ueber dm Verhalten des Kerns bei den Zelltheilung und iiber die Bedeu-

tung mehrkerniger Zellen. Virch. Arch., 77: 1-29, pl. 1. 1891. FLEMMING, W.:

Ueber Theilung und Kernformen bei Leukocyten, und Uber deren Attractionssphiiren. Arch. f. Mikros, Anat.. 37: 249-298, pls. 13-14.

1892. FLEMMINQ, W.: Entwicklung und Stand der Kenntnisse iiber Amitose. Erg. der Anat.

u. Entwick. Merkel u. Bonnet, 2: 37-82. 1898. FLEMMINQ, W.:

tfber die chrosomenaahl beim Menschen. Anat. Anz. 14: 171-174, fig. 1.

1909. FORBES, A.: The origin and development of foreign body Siant cells. J. M. Res.,

15 n.8. 20: 45-52, pls. 1-2. 1910. FORBES, A.:

The origin and development of giant cells in an epidermoid carcinoma of the tongue. J. Med. Res., 23: 107-110, pls. 1-2.

1893.’ GALEOTTI, G.: Beitrag zum Studium des Chromatins in den Epithelaellen der Carcinome.

Beitr. path. Anat. u. z. allg. Path., 14: 244-271, ple. 12-13. 1893.¶ GALEOITI, G.:

Ueber experimentelle Erzeugung von Unregelmassigkeiten des karyo- kinetischen Processes. Beitr. a. path. Anat. u. a. allg. Path., 14: 288-316, pl. 15.

1908. GLABER, 0.: A statistical study of mitosis and amitosis in the entoderm of Fascioluiia

tulip var. disfans. Bio. Bull., 14: 219-248, figs. 1-9. 1904. HHCKER, v., V.:

Ueber die in malignen Neubildungen auftretenden heterotypischen Teilungsbilder. Biol. Centralbl., 24: 787-797, figs. 1-11.

1907. HALLION, L. H.: Theorie karyogamique du cancer. Arch. d’elect. medic. Paris, 15:

713-727.

44 MICHAEL LEVlNE

1890. HANEEMA", v., D.:

Bedeutung. Virah. Arch., 119: 2094326, pl. 9, Ueber aaymmetrisahe Zellhilung in Epithelktebn und deren biologisohe

1891.' H A N ~ u " , v., D.: Karyokinesa und Zellulas pathologie. Ber. Klin. W o a h n ~ ~ b . , 28:

1039-1042. 1891'. BANBEMA", v., D.:

1892. HANSHIMA", v., D.:

1893. HANEEMA", v., D.:

Ueber patholo&che Mitosen. Virch. Arch., 123: 366-370, pls. 10-11.

Ueber die Anaplasie der Geschwulatzellen und die aaymmetrische Mitoae. Virch. Arch., 129: 436-449.

Die studien Uber die Spaifisitat der alixuiamue und die Anaplaaie der Zellen, mit besonderer Beruclcaichtigung der Gemhwtilste Ber., 1893.

1902. HANSEU", v., D.: Die mikroscopische Diagnose der B 6 d g e n Geachwlllaten. 2nd ed.,

Ber., p. 91. 1904. HANSEMA", v., D.:

24: 189-192. 1904. HANSEMANN, v., D.:

1905. HANSEMANN, v., D.:

Ueber Kernbilungsfiguren in b W g e n Geschwulsten. Biol. Zentralbl.,

Malignant growths and normal reproductive tiesum. Lancet, 251.

Einige Bemerkingen fiber die augebliah heterotyphen Zellteilungen in Weartigen Geschwlllsten. Biol. Zentralbl., 25: 151-156.

19065. HARQXTT, C. W.: The organization and early development of the egg of C h a Zepbatyla Ag.

Biol. Bull., 10: 207-231, pl. 9. 1808. HEIBERQ, K. A,:

Ueber die Erklbung einer Verschiedenheit der Krebamllen von anderen Zellen. Nord. Med. Ark., 41; Abt. 2: 1-20.

1910. HEIBERG, K. A. Ein paar Bemerkungen Uber die Zellkerne und die Granula experimentder

Karzinome wie auch Uber Abstrich prilparate aua diesen Tumoren. Bio. Zentralbl., 30: 4oo-po2.

1890. HESS, K.: Ueber Vermehrunga und Zerfallsvorgilnge an den grossen Zellen in der

akut hyperplaatischen Mile der w e h n Maua. Beitr. z. path. Anat. u. I. allg. Path., 8: 221-244, ple. 13-16.

1894. v. HEUKELOM, 8.: Daa Adenocarcinom der Leber mit Cirrhoee. Beitr. xu path. Anat. u.

dlg. Path., 16: 341-387, pls. 7-8. 1909. JOHNSON, T.:

1889. KXBBS, E.:

Chtyeophlydie endobiociccr Bahilb. (potato wart or black scab) and other Chytridiaceae. Sci. Proo. Roy. SOC., Dublin, 12: 131-144, pls. 9-11.

Allgemeine Pathologie, 11, Jena.


1895. K!wMPECBER, E.:Ueber die Mitose mehrkcrniger Zellen und die Beziehung zwischen Mitose

and Amitose. Virch. Arch., 142: 447-473, pl, 9-10.1902. KROMPECHER, E.:

Ueber Zelltheilung. Zentralbl. f. Path. u. Path. Anat., 13: 273-292,pl. 1-2.

1912., LAMBERT, R. A.:The production of foreign body giant cells in vitro. Jour. Exp. Med.,

15: 510-515, pis. 71-74.1913. LAMBERT, R. A.:

Comparative studies upon cancer cells and normal cells. II. The character of growth in vitro with special reference to cell division.Jour. Exp. Med., 17: 499-510, pis. 98-100.

1920. LEVIN, I., AND M. LEVINE:

Malignancy of the crown gall and its analogy to animal cancer. Jour.Cancer Res., 1920, 5: 243-260, figs. 1-15.

1922. LEVIN, I., AND M. LEVINE:

The rMe of neoplasia in parasitic diseases of plants. Jour. Cancer Res.,7: 171-178, figs. 1-10.

1913. LEVINE, MICHAEL:

Studies on the cytology of the hymenomycetes, especially the Boleti.Bull. Torrey Bot. Club, 40: 137-181, pls, 4-8.


Somatic and reduction divisions in certain species of Drosera. Mem.N. Y. Bot. Gard., 6: 125-147, figs. 1-2. pIs. 16-19.


Studies on plant cancers: 1. The mechanism of the formation of theleafy crown gall. Bull. Torrey Bot. Club, 46: 447-452, pls, 17-18.


Studies on plant cancers: II. The behavior of crown gall on the rubberplant (Ftcus elastica). Proo, Soc Exp. Biol, and Med., 17: 157-158,M~cologllt, 13 1-11, pl. 1-2.


Studies on plant cancers: TIL The nature of the soil as a determiningfactor in the health of the beet (Beta Vulgaris), and its relation tothe size and weight of the crown gall produced by inoculation withBacterium tumefaciens. Amer. Jour. Bot., 8: 507-525, figs. 1-9.


The origin and development of lamellae in Agaricus campestris and incertain species of Coprinus. Amer, Jour. Bot., 9: 509-533, pls,28-35, figs. 1-12.

1923. LEVINE, MICHAEL:Studies on plant cancers: IV. The effects of inoculating various quantities

of different dilutions of Bacterium tumefaciens into the tobaccoplant. Bull. Torrey Bot. Club, 50: 231-243.

1923. LEVINE, MICHAEL:Studies on plant cancers: V. Leafy crown galls on tobacco plants re

sulting from Bacterium tumefaciens inoculations. Phytopath., 13:107-116. pis. 5-7.

46 MICHAEL LEVINE

1924.

1921.

1923.

1914.

1916.

1881.

1892.

1917.

1918.

1900.

1899.

1920.

1905.

1886.

1896.

LEVINE, MICHAEL: Studies on plant cancers: VI. Further studies on the behavior of crown

gall on the rubber plant ( F h elastic&). Mycologia, 16: 24-29( pl. 6.

Giant cells in cultures from human lymph nodes. Jour. Exp. Med., LEWIS, W. H., AND L. T. WEBSTER:

33: 349-360, pls. 27-31. MCCARTHY, W. C.:

The cytological diagnosis of neoplaams. Jour. Amer. Med. h., 81: 519-522.

MCLEAN, R. C.: Amitoses in the parenchyma of water plants. Proc. Camb. Phil. Soc.,

17 (pt. 5): 380-382, fig. 1. MACKLIN, C. C.:

MARTIN, W. A.: Amitoses in cella grown in vitro. Biol. Bull., 30: 445-461, pls. 1-3.

Zur Kenntnis der indirekten Kerntheilung. Virch. Arch., 86: 57-67, pl. 4.

MOLLER, V.: Ueber cellulare Vorgange in Geschdsten. Virch. Arch., 130: 512-

528, pl. 14-15. NAKAHARA, W. :

Preliminary note on the nuclear division in the adipose cells of inaects. Anat. Record, Phila., 13: 81-86, figa. 1-11.

NARAHARA, W.: Studies of amitoais, its physiological relations in the adipose cells of

insects, and its probable significance. ,Jour. Morp., Phila., 30: 483-513, pls. l-$'i:'.::: , ~ ~ ' ~ ~ ' ~ ! *:!*.::

NEDJELSKY, W.: 3 :*.**.*::..: : :..:*.: Ueber die aWit~$kche .Theilwg ip patholagi&c+ JS'&ildmgen, haupt-

slichlichi&&ebe$ I.@ Eagcinomiq.'.;. Be$!?. j~*&&. Anat. u. Z. allg. P;tll::*m432~3: p1e: 15-18:' 0:''

*'.*** N ~ E c , B.: '. : :a: ='= :-: :':: :

Ueber Kern und Zellteil;Fhg*&j @&'i&&b&ooeurn. Flora, 86: 214-227, pl. 13.

OSHIXAWA, K.: Observations concerning the pathological changes of tumor cells, espe-

cially of the nucleus aegmentation after extirpation of carcinoma tissue. Gann., 14: No. 2; 9-10,

OVERTON, J. B.: Ueber Reduktionsteilung in den Pollenmuttereellen einiger Dikotylen.

Jahrb. Wisa. Bot., 42: 121-153, pls. 6-7.

pl. 5. PIANESE, G.:

PFITZNER, W.: Zur pathologischen Anatomie des Zellkerns. Virch. Arch., 103: 275-300,

Beitrage zur Histologie und Xtiologie des Carcinoma. Beitr. zur path. Anat. u. z. allg. Path., (Supp. I) 20: 1-193, pls. 1-8 (colored).


1915.

19n.

1902.

1904.

1905.

1888.

1912.

1917.

1922.

1922.

1909.

1912.

1891.

1893.

1892.

PRANKERD, T. L.: Notes on the occurrence of multinucleated cells. Ann. Bot., 29: 599-

604, figs. 1-8. RAPPEPORT, T.:

Ueber die somatische Mitose des Menschen. Arch. f. Zellforsch., 16: 371-382, figs. 1-2, pl. 18.

REED, D. M.: On the pathological changes in Hodgkin’s Disease with special reference

to the relation to tuberculosis. Johns Hopkina Hosp. Record, 10: 133-196, pls. 5-7.

ROBENBERG, 0. : Ueber die Tetradenteilung eines Drosera-Bastardes. Ber. Deuta. Bot.

Gesells., 22: 47-53, pl. 4. Ross, F. W. F.:

cancer. SCH~LXNDER:

Observations on certain features exhibited by cells in their relation to Brit. Med. Jour., London, 2: 1101-1102.

Ueber Kern und Zelltheilungs Vorgange in dem Endothel der entrtindeten Hornhaut. Arch. f. Mikr. Anat., 31: 426-482, figs. 1-6, pl. 22.

SMITE, E. F., BROWN, A. NELLIE, AND LUCIA MCCULLOCH: The structure and development of crown gall. U. S. Dept. Agr., Bur.

P1. Ind. Bull., 255: 1-60, pls. 1-109. SMITE, E. F.:

Embryomas in plants. Johns Hopkins Hosp. Bull., 28: 277-294, pls. 26-53.

SOKOLOFF, B.: Relations entre le noyau et le cytoplasme dans la cellule maligne. Compt.

rend. SOC. de biol., 87: 1200-1202.

The nucleo-cytoplasmic ratio and cancer. Jour. Cancer Res., 7: 395- SOKOLOFF, B. :

415, pls. 1-10. STEINHAUS, J. :

La cellule canohreuse considhrhe principalement au point de vue de la g e n h et de la nature des inclusions nuclhaires et plasmatiques. Jour. med. de Brux. 14: 369-375, 386-394, 408-410, 421-423, 437- 439, 489-490, 505-507, 518-524, 531-540, 550-551, pl. 1.

STOUT, A. B.: The individuality of the chromosomes and their serial arrangement in

Curix aqmtilis. Arch. f. Zellfors., 9: 114-140, pls. 11-12. STRICHT, VAN DER OMER:

Le dhveloppement du sang dans le foie embryonnaire. Arch. de. Biol., 11: 19-113, p1~. 1-2.

STROEBE, H.: Ueber Vorkommen und Bedeutung der asymmetrischen Karyokinese,

nebst Bemerkungen iiber die “Schlummerzellen” in den verletzten Cornea. Beitr. path. Anat. u. z. allg. Path., 14: 154-173, pl. 8.

STROEBE, H.: Zur Kenntnis verschiedener Celluliirer Vorgange und Erscheinungen in

Beitr. z. path. Anat. u. 1;. Allg. Path., 11: 1-38, pl. 1. Geschwiilsten.

48 MICHAEL LEVINE

1897. TRAMBUSTI, A,: Ueber den Bau und die Theilung der Sarkomsellen. Beitr. s. path.

Anat. u. s. allg. Path., 22: 88-104, ph. 4-5. 1851. VIRCHOW, R.:

pl. 2.

444.467, pl. 9.

Die endogene Zellenbildung beim Kwbs. Virchows Arch., 3: 197-227,

1904. W-R, c. E.: On the cytology of malignant growths. Tr. Path. Soc. London, 65:

1906. WALKER, c. E.: On the nature of the vesicular structures found in malignant growth

and sometimes known as Plimmer’s bodies. Tr. Path. 800. Lon- don, 56: 372-377.

1923. WALKER, C. E.: Theories and problems of cancer. London.

1809. WALKER, C. E., AND G. DBIBAISIBIUX: On the behavior of nucleoli in the cells of malignant growths. h c .

Roy. SOC. Med., London, 23 (Path. sect.): 128-133. 1911-1912. WALKBR, C. E., AND H. E. WEITTINQHAM:

Further observations upon the resemblance between the cells of malignant growths and those of normal gametogenic t ime. Jour. Path. and Bact., 16: 185-198.

1886. WERNER, W.: Ueber Theilungworgtinge in den Rieaenrellen des Knochenmarkes.

Virch. Arch., 108: 354-377, pl. 4. 1907. WINIWATER, v., HANS:

Ueber Pagetsohe Krankheit. Arch, Derm. u. Syph., 85: 239-262, pl. 15.

EXPLANATION OF PLATES

All photographs were made with the Bausch and Lomb D and L type camera and Zeiea 1.6 mm. apochromatic objective and Nos. 8 and 12 compensating oculars. The length of the bellows was 26 cm. The drawings were made with the aid of a camera lucida and the eame microscopical equipment as was used in making the photographs. All the illustrations were made from preparations stained with Flemming’s triple stain.

PLATE I Photographs of cells from an epithelioma of the lip

FIQ. 1. RQ. 2.

FIQ. 4. FIQ. 5.

FIG 3.

FIQ. 6. FIQ. 7. FXO. 8. WO. 9.

Normal cell in the spireme stage. Normal cell with constriction on one surface, showing segmented spireme. Nuclear division of normal cell followed by cytokinesis. Tripolar spindle in the telophase stage. Cytokinesis is also in progress. Uninucleate giant cell with uniform nuclear membrane. Uninucleate giant cell with furrowed nuclear membrane. Simple bilobed nucleue in a giant oell. Ring-shaped nucleus in a giant cell. Multinucleate giant cell.

PLATE I

PLATE I1

18a

. .

PLATE 111

26 22


PLATE3 I1 This plate waa reduced X %

Represents figures made from an epithelioma of the lip FIQ. 10. Early stage in the division of normal cell. Spindle fibres developing

FIG. 11. Simple lobe nucleus in spireme stage. X 1600. FIG. 12. Normal division of cell with equal distribution of chromosomes.

FIQ. 13. Tripolar spindle in a amall cell. X 1200.

FIQ. 14. Giant cell with multilobulate nucleus. X 1600. FIQ. 15. Early prophase stage in a giant cell with multilobulate nucleus.

FIQ. 16. Segmented spireme in a giant cell. X 1600. FIQ. 17. Anaphaae stage in a pluripolar spindle in a giant cell. FIQ. 18. Three clusters of cells which have arisen from the division of a giant

cell. x1m.

between two centrosomes and chromosomes. X 1200.

x 1800.

Figures lk-17 represent one of a series of four sections

x 1600.

X 1600.

PLATE3 I11

Figures 19 and 20 represent cells of crown gall on Betu d g a d (common red beet). Figures 21 and 22 represent cells of crown gall on Ricinwr eanguineus. The magnifi-

Figures 23 and 24 represent cells from the crown gall on a Sedum leaf. FIG. 19. Early telophase with equal chromosome a t both poles. FIG. 20. Cell division, nuclei in reconstruction stage. FIG. 21. Binucleate cell. FIG. 22. Segmented spireme, polar caps. FIG. 23. Binucleate cell arising from mitotic division without cell plate formation. FIQ. 24. Early telophase in embryonic cell. FIQ. 25. Young cell in telophase stage; the chromosomes are equally divided;

FIG. 26. Parasitized cell of a potato wart; host nucleus in late telophase with

This plate was reduced X

cation is X 1600.

from a crown gall on tobacco.

cell plate formation.

A COMPARATIVE CYTOLOGICAL STUDY OF...in man for the purpose of verifying the existence of asym-...

Documents

Transcript of A COMPARATIVE CYTOLOGICAL STUDY OF...in man for the purpose of verifying the existence of asym-...