^, IMAGE EVALUATION TEST TARGET (MT-3) // ^ >4 ^// ' >V^ k^' V K^ n^^ ^ ^ ^ ^ liio iil 1.1 y* i^ 12.2 III It! U 14.0 \\25 HI 1.4 12J 1.6 6" <^ y yj %^ V I^iotographic Sciences Corporation 23 WKT MAiN STRKT WS!ISTIR,N.Y. MSM (716)872-4303 CIHM/ICMH Microfiche Series. CIHIVI/iCMH Collection de microfiches. Canadian Instituta for Historical IMicroreproductiona / Institut Canadian da microraproductions historiquas Technical and Bibliographic l^otaa/Notas tachniquas at bibiiographiquaa Tha Instituta haa attamptad to obtain tha baat original copy availabia for filming. Faaturaa of thia copy which may ba bib!iographicaily uniqua, which may altar any of tha imagaa in tha raproduction, or which inay significantly changa tha usual mathod of filming, ara chackad balow. D D Colourad covars/ Couvartura da coulaur I I Covars damagad/ Couvartura andommagia Covars rastorad and/or laminatad/ Couvartura rastaurte at/ou paliiculAa Covar titia missing/ La titra da couvartura manqua Coloured maps/ Cartas gtegraphiquas an coulaur Colourad init (i.a. othar than blua or biaclc)/ Encra da coulaur (i.e. autre que bleue ou noire) I I Coloured plates and/or illustrations/ D Planches et/ou iliu3trations en couleur Bound with other material/ Reiii avac d'autras documents Tight binding may cause shadows or distortion along intarioi^ margin/ La re liure serr^e peut causer de I'ombre ou de la distortion la long de la marge IntArieure Blank leaves added during restoration may appear within the text. Whenever possible, these have been omitted from filming/ II se peut que certaines pages blanchea ajouttes lore d'una restCMration apparaissent dans la texte, mais, lorsque cela Atait possible, ces pagea n'ont pas AtA filmtes. Additional comments:/ Commentaires supplAmantairas: L'Institut a microfilm* la mailleur exemplaira qu'il lui a 4t4 poasibia da se procurer. Les ditails da cat exemplaira qui sont paut-Atre uniques du point oe vue bibliographiqua, qui peuvent modifier une image reproduite, ou qui peuvent exiger une modification dans la mithoda normale de filmage sont indiqute ei-dessous. D D D D n Coloured pagea/ Pagea da coulaur Pages damaged/ Pagea endommagAes Pages restored and/or laminated/ Pages restaurAes et/ou peliicultes Psges discoloured, stained or foxed/ Pagea dAcolortes, tachatAes ou piqutes Pagea detached/ Pages dAtachtes Showthrough/ Tranaparance Quality of print varies/ Qualit^ InAgala de Timpression includes supplementary material/ Comprend du material suppMmantaira Only edition available/ Saule Mition disponibie Pagea wholly or partially obscured by errata slips, tissuaa. etc., have been refilmed to ensure the best possible image/ Les pages totalement ou partieliement obacurciaa par un fauillet d'errata, une pelure, etc., ont *t^ fiimtes A nouveau da fapon i obtenir la meilleure image possible. This item is filmed at the reduction ratio checked below/ Ce document eat film* au taux de rMuction indlqu* ci-daaaoua. 10X 14X 18X 22X 12X 16X 20X 26X 30X A 24X 2SX 32X Th« copy ffllm«d hwr* has b««n r«produo«d thanks to tha ganarosity of: Univtrtity of Sailcatchtwin Saikatoon L'axamplaira filmA fut raproduit grAca A la ginirosit* da: Univtnity of SMkatchawan Saskatoon Tha imagas appaaring hara ara tha bast quality posslbia consldarlnj tha condition and lagiblllty of tha original copy and in kaaping with tha filming contract spaclficatlona. Las imagas suivantas ont AtA raproduitas avac l» plus grand soin, compta tanu da la condition et da la nattat« da l'axamplaira film*, at an conformity avac las conditions du contrat da fi!!maga. Original copies in printad papar covars ara fllmad beginning with tha front covar and anding on tha last paga with a printad or illustratad impraa- slon, or tha back covar whan appropriata. All othar original coplaa ara fllmad baginning on tha first paga with a printad or llluatratad impraa- sion, and anding on tha last paga with a printad or Illustratad impraaaion. Las axamplairas originaux dont la couvartura en papier est ImprimAe sent fiimAs en commen^ant par la premier plat et en terminant soit par la derniAre page qui comporte une empreinte d'impression ou d'illustration, soit par la second plat, salon le cas. Tous las autres exempiaires originaux sont filmAs en commenpent par la premiire page qui comporte une empreinte d'impreeslon ou d'illustration et en terminant par la darnlAre page qui comporte une telle empreinte. The laat recorded frame on each microfiche shall contain the symbol -^ (meaning "CON- TINUED "), or the symbol V (meaning "END"), whichever applies. Un dee symboles suivants apparattra sur la dernlAre image de cheque microfiche, selon le cas: le symbole — ► signifie "A SUiVRE", le symbols V signifie "FIN". Mapa, plates, cherts, etc., may be filmed at different reduction ratios. Thoaa too large to be entirely included in one expoeure are filmed beginning in the upper left hand corner, left to right and top to bottom, aw many frames as required. The following diegrama illustrate the method: Lea cartes, planches, tablaaux, etc., peuvent fttre filmte A dee t&ux de rMuction diff Arents. Lorsque le document est trop grand pour Atre reproduit en un seul ciichA, il est film* A partir de I'angia supArieur gauche, de gauche A droite, et de haut en bas, en prenent le nombre d'images nAcessaire. Las diagrammes suivants illustrent la mAthode. 1 2 3 1 2 3 4 5 6 ^'■ .0^' UNIVERSITY OF TORONTO STUDIES Physiological Series : edited by Professor A. B. MACALLUM ■^gf*" v^-" No. 2. ON THE CYTOLOGY OF NON-NUCLEATED OiRGANISMSy By A. B. MACALLUM % THE UNIVERSITY LIBRARY : PUBLISHED BY THE LIBRARIAN, 190c ON THE CYTOLOGY OF NON NUCLEATED ORGANISMS By a. B. Macallum, M.A., M.B., Ph.D. //ooio\ Uk'i X 197] ON THE CYTOLOGY OF NON-NUCLEATED ORGANISMS ( Keprinted hy permission from the Transactions of the Canadian Institute, iSgS-gg) In the following pajjes I have endeavoured to describe the results of observations which I have made during the last five years on the structure of certain types of non-nucleated organisms. These studies are not as complete as I would wish them to be, especially in the case of Bacteria, for only a few of the forms of the latter accessible to me were of sufficient size to enable me to employ all the micro-chemical methods used in the case of the Cyanophycea; and Saccharomyces, but the results described may prove of service to other investigators in the same line, and at the same time stimulate the employment of comprehensive methods of technique in the study of non-nucleated organisms. It is doubtful if the ordinary or more complicated methods of staining which have been used in this department give results which are of the morpho- logical value possessed by results obtained in this way in more highly specialized animal and vegetable cells. The existence of a nucleus in these low forms of life, if not denied by the great majority of observers is at least in dispute. If it is absent what takes its place ? Is chromatin present or does there exist in them an analogous substance ? As there are in many animal and vegetable cells other substances which may take up dyes, more especially of the aniline kind, it is obvious that staining methods alone ar- not sufficient to enable the observer to determine the solution of questions like these in regard to the lowest forms of life. A more satisfactory determination of such questions can be made only with methods which comprehend micro-chemical reactions of definitely ascertained values. It may be pointed out also that it is in these low forms of life that we must look for a key to the secret of the origin of the cell nucleus, as well as for data to determine the morphological character of the primal life organism. It is of course suspected by many that in Saccharomyces the [39] 4 MACALLUM ! CYTOLOGY OF NON-NUCLEATED OROANISM8 cell as it Is now found does not represent what It once was, that It Is the product of the degradation of a higher form of cell. In that case it may be contended that one cannot obtain from the present structure of the cell data which can be of service In ascertaining what the structure of the primal cell was. If the yeast cell, which, as I claim, Is non-nucleated, Is the product of the degeneration of a higher type of cell, of, for example, a nucleated cell, an examination of It may still serve a purpose In assisting in determining the origin of the cell nucleus, for we can then ascertain how in the yeast cell its surplus of chromatin Is distributed in the cytoplasm in all conditions, and this may possibly indicate to us in what manner the original non-nucleated cell disposed of the excess of the analogous chromatin compound. From this point of view, therefore, the employment, not only of methods of staining, but also and particularly of micro-chemical processes in the Investigation of these low forms of life is absolutely necessary if results of any value are to be obtained. These methods I have endeavoured to employ to a certain extent and this constitutes the excuse for adding one more contribution on the subject. CONTENTS. I.— The CvANOPHYCE.*. page. Literature 5 Methods of Study and Material Employed 16 The Living Cell in the Cyanophyceae 19 Fixed Preparations of the Cyanophyceae 21 The Granules in the Cyanophyceae 29 The Heterocyst 33 Cell Division 34 Summary 36 IL— Beogiatoa. Literature 38 Methods of Study and Forms Studied 40 General Cell Structure 41 Summary 44 III.— The Yeast Cell. Literature 45 Methods of Study and Species Studied 52 General Cell Structure 54 The Chromatin-holding Structures 56 Budding and Sporulation 62 Summary 67 i40i] 3 " 4 s "I bot. Gea MACALLUM i CVTUUXiV UP NUN-NUCLKATKU ORQANUMS THE CYANOPHYCE>E. PAGE. • • 5 . i6 .. «9 21 .. 29 .. 33 .. 34 ... 36 ... 38 ... 40 . .. 4« ... 44 ... 45 ... 52 ... 54 ... 56 , .. 6j .... 67 I.— LITERATUKK. The first to examine carefully the structure of the Cyanophyceae was Schmitz,' who found in tlie cells of Glaocapsa polydermaticn a homo- geneous central portion which stained with hitmatoxylin and represented, as he thought, the nucleus of the cell. He found al.so that there were in the cell substance a number of spherules of unknown composition, to which he applied the term " Schleimkugeln." In the cells of Oscillaria princeps, after being stained with h.-ematoxylin, he ob.served a dark, spherical, somewhat exccntrically placed body which he regarded as a nucleus, but he was unable to demonstrate its presence in all preparations of this form. Further observations,'-' however, led Schmitz to regard the cells of Cyanophycea: as non-nucleated and the supposed nuclei of GUeocapsa as simply very large chromatin granules which react with hematoxylin like the chromatin granules of the nuclei of higher organism.s. He dis- tinguished in Oscillaria princeps a peripheral, finely punctated zone in each cell which stained with h;ematoxylin less deeply than the central portion. In a later work ue denies the existence of a chromatophore and claims that the functions performed in the special parts of highly differentiated cells, are in the Cyanophyceae the property of the cell protoplasm generally. Tangl* also was unable to determine the presence of a nucleus, but he found a chromatophore in Plaxonema Oscillans. According to Wille,* however, a nucleus exists in Tolypothrix lanata. After being stained with a dilute or a concentrated haematoxylin solution, preferably the latter, the nucleus appeared faint blue, the nucleolus, on the other hand, intense blue, while the remaining cell contents were scarcely stained. In the dividing eell he observed two nuclei, each with a I "Untersuchungen aber don /ellkerne der Thallophyten," Sttzungrsb. der Niederh. Gesell. fOr Natur— und Heilkunde zu Bonn, Sitz. vom August 4, 1879, p, 555. a " Unteniuchungen ilber den Strukter der Protoplasma und der Zelikeme der Pflanzenzellen," Ibid, Sitz. am Juli la, 1880, p. 159. 3 " Die Chromatophoren der Algren," Bonn, 188a, quoted by Deinega. 4 " Zur Morphologfie der Cyanophyceen," Wien, 1883- 5 " Ueber die Zellkerne und die Poren der Wilnde bei den Phycochromacecn,'* Berichte d. deutsch, bot. Gesell., Vol. I, 1883, p. 343. [4«] MACALLCM 1 CVTOLIHIV OP NON-NI'CLKATKD OKdANIHMII (I II nucleolus, and in another staKC c)f cHvision he found one nucleus with two nucleoli. The next observer, I.agerhcim, found no nucleus in Glaucocystis Nostochinearum. The chromatophore in the younj^ cells of this form is, according to his description, in the form of a band or thread surrounded by the colourless elements of the cell, but in the older cells it is composed of a large number of small granules which form a membrane lying at some distance from the cell wall and enclosing the colourless cell substance. Reinhard^ found in Oscillnria major ij) when fixed with picric acid and .stained with haematoxylin, a large granular nucleus with large granules as nucleoli. The protoplasm contains large and small granules, the former constituting the chromatophores. Hansgirg, although main- taining the existence of a chromatophore and nucleus in Chroodactylon Wolleanuui, a unicellular blue-green form, held that in the thread-like Cyanophycea' there is neither a nucleus nor a chromatophore and that the cell protoplasm discharges the functions of both. Borzi* found neither a nucleus nor a chromatophore. He distinguished by micro-chemical methods, amongst the granules present, a kind the examples of which are partly imbedded in the protoplasm and partly applied to the cell wall. These granules are formed of a gelatin substance, which he believes replaces .starch in these forms, and which he names cyanophycin. The granules, he believes also, are .secreted in dividing cells by the young transverse septa. In Nostoc, Anabana, Spet iHosim, Cylindrospcrmum and Sphaerosyga, the protoplasm of two adjacent cells is connected by fine strands, .sometimes of plasma, at other times of cyanophycin, which pass through openings in the transverse septa. The.se openings are specially marked in heterocysts in which they closed by plugs of cellulose, protein or cyanophycin. Similar per- forations were observed in the transverse septa in Oscillnria and Borzi explains their function as that of uniting the protoplasm of all the cells to enable them to act as a unit in the case of movement. In his first publication" on the subject of a nucleus in the Cyano- phyceae, Zacharias stated that he found this organ in the terminal cells I " Ein neucB Beivpiel Jes VurkommenH von Chroinatophoren bei den Phycochromaceen," Berichte d. deut8cli. bot. Gekell.. Vol. H, 18S4. a "AlKologiacho Untcmuchungen, I. Materi.-ilen zur Morphologie und Systematik der Algen dei Schwarzen Meeres," Odessa, 1885, (Russian). Abstract given by Deinega. 3 Ein Beitrag zur Kentniss von der Verbreitung der Chroinatophoren und Zellkcrne bei den Schizophyceen (Phycochromaceen)," Berichte d. deutsch. oot Gesell., Vol. Ill, 1885, 4 Malpighia, Anno I, 1886. Abstract in Bot. Cenlralbl., Vol. XXXU, p. 35, 1887. 5 " Beitrilge zur Kentniss der Zellkems und der Scxualzellen," Bot. Zeitung, 1887, Nos. 18-24. [42] MACALLIIM I CVTOI.OOV Or NON^NI'CLKATRD OROANINMN | of Tolypothrix after treatment with h digestive fluid, and extraction with alcohol and ether, and on examination in a sohition of hydrochloric acid of 0,3 \vix cent, strenj^th, Ifi this case the nucleus was demonstrated by the nuclein lustre |)roduced. The nuclein was extracted with a soda solution of 0.05 per cent, stren^^th from threads which had previously been submitted to digestion. He found also in a species cf Oscillaria, after treatment with dijjestivc fluids, a larjjc nucleus, containin^j nuclein network, in every cell. From such preparations the nuclein was removed by extraction with very ORGANISMS '5 Zacharias' confirms his earlier observations, except as to the presence of nuclein in the central body, of which he is doubtful. He regards the central body as differing in important points from the nucleus of other organisms. Fischer, in his recent publication, gives the results of a more extended series of observations on the structure of the Cyanophycese and Bacteria. His first statement is a withdrawal of the view which he held that the " central body " of Biitschli arose through plasmolytic contraction of the contents of the cell, brought about by the hardening reagents used. He finds, however, that Biitschli's description of the granules is far from exact. In many of the forms haematoxylin colours some of the granules blue, others red, in other forms all the granules may be coloured blue, while sometimes again the granules may be stained red only. In Biitschli's opinion the reserve granules, the " cyanophycin " granules of Palla, do not stain with hjematoxylin and, therefore, the granules which colour blue with this dye, do not, according to Fischer, come into Biitschli's classification of the granules. Fischer also found that all the granules stain blue with haematoxylin after treatment with soda solutions. On the other hand the " cyanophycin " granules in Oscillaria tenuis, after fixation with alcoholic iodine solutions, do not stain with haematoxylin, but they will do so if hardened with a four per cent, solution of formol. The granules do not disappear after artificial digestion (in pepsin and hydrochloric acid), nor on treatment with a ten per cent, soda solution. From all these facts he concludes that the differences in the staining power of the granules are due, not to differ- ences in chemical composition of the granules, but to their physical properties, and that the tests upon which Biitschli relied to show that the " red " granules are composed of chromatin are valueless. He regards the substance entering into the composition of all the granules as an assimilation product or a reserve material. In regard to Palla's contention that granules do not occur in the "central body," Fischer found them in this organ as well as in its peri- phery. None were observed in what he calls the chromatophore, the zone of coloured protoplasm surrounding the " central body." Granules exist outside this zone and, particularly, adjacent to the transverse septa, a fact which leads Fischer to believe in the existence of a plasma zone outside the chromatophore. The independent existence of the latter organ he claims is shown by the results of treatment with hydrofluoric 1 " On the Cells of the Cyanophyces," Report of the British Association, Liverpool Meeting, i8q6, p. 102I. 3 " Untersuchung^en tlber den Bauder Cyanophyceen und Bactcrien," Jena, 1897. [5>) wm i6 MACALLUM : CVTOLCKIV OP NON-NUCLBATRD OROANI8MS acid, which dissolves everything else in the cell. Through the chroma- tophore stretch radial strands of protoplasm from the central body to the plasma zone on the inner wall of the cell. In pepsin and hydro- chloric acid the volume of the contents is greatly decreased, but this is not due to digestion of the peripheral portions chiefly and of the central body partly as Zacharias and others believed, but to a contraction which he terms enzymatic and which can be demonstrated in Spirogyra also. In the shrunken parts the peripheral as well as the central parts are present. Digestion experiments, therefore, give no certain conclusions as to the special nature of the central body. The latter is in his view simply a portion of the protoplasm enclosed by the chromatophore and containing assimilation products. There is nothing to indicate that it is a nucleus or the homologue of a nucleus. Nor is there any other organ which may be held to represent a nucleus. II— Methods of Study and Material Employed. By hardening and staining highly organized animal and vegetable cells in the way that cytologists usually employ, one may obtain pre- parations which readily reveal the structure of these elements, but the employment of these and other simple methods in the case of the Cya- nophycex are not at all as fruitful in results. One finds in such little more than one can recognize in the living cells. It is with such and similarly meagre methods that the majority of investigators in this field of research have contented themselves when the apparently indifTeren- tiated character of the cells in these organisms should have suggested the employment of a multiplicity of methods. It is only in this way that one may obtain results which permit a generalization concerning the structure of these forms. The Cyanophyceae respond very sensitively to the conditions to which they are subjected. This fact a!so has been overlooked or not suspected. In a culture of them twenty-four hours will make a complete change in some of the more important features of their cells, and at the same time different parts of the same culture, not more than a few centimetres dis- tant from each other, may present specimens of the same species in which the cell contents are markedly unlike. It is to this fact that we may attribute discrepancies in the descriptions, by the various observers, of the structure of these organisms. To illustrate particularly how impor- tant this point is, I may refer to Palla's observations on Glceotricha pisum. In single cells of this form he found large vacuoles, and more [52] MACALLUM : CVTOLOtiV OP NON-NUCLRATRO OROANIKMIi M than one central body. The material which he used must have been in a pathological condition, for it is only when Cylindrospermum majus is ill-nourished that any one of its cells possesses similar vacuoles and a plurality of central bodies. It is obvious, therefore, that a study of the cells of the Cyanophycea: Involves the employment of a large number of methods of manipulation, and a regard for material in all conditions of nutrition. The latter requisite also entails a careful attention to the various conditions in which the organisms are usually found, and an acquaintance with the modifications that a varied environment brings about. These conditions and these modifications are, at present, not fully known, but what we do know helps in determining what is the t) ^ucal structure of the cell in this class. I have, therefore, in all cases used material from each species which was found to be in an active state of growth. This is a safeguard of great value. The material which one may collect from any locality may be in the resting stage, a stage in which also, perhaps, changes analogous to those of involution in Bacteria may manifest themselves. The only material which one can confidently regard as normal is that which grows in the laboratory, and which can be examined from hour to hour, the change in the volume of the material being an indication of its active growth. The species' used in these studies were Microcoleus terrestris Desmazieres (M. vaginatus Gomont), Oscillaria Froehlichii Kutzing {O. limosa Agardh), Oscillaria natans Kutzing (0. tenuis Agardh), Oscillaria tenerrima Kutzing {O. amphibia Agardh), Oscillaria princeps Vaucher (0. maxima Kutzing), Cylindrospermum majus KUtzing, Toly- pothrix tenuis Kiitzing, Tolypothrix rupestris Kutzing, Nostoc commune, Rivularia sp., Glceocapsa polydermatica, Lyngbia sp., Stytonema sp. The hardening reagents employed were : corrosive sublimate in saturated aqueous solution, in saturated alcoholic solution, and in con- junction with picric acid, picric acid in saturated solution, Flemming's chrom-osmio-acetic mixture in strong and weak solution, osmic acid in one per cent, solution, alcohol absolute and of ninety-five per cent, strength. I have used also aqueous and alcoholic solutions of iodine. The best preparations were made with the picric acid and corrosive I Gomont, " MonoKraphie des O&cillari^es, (Nostocactes homocysMes)," Annates dm Sciences Nat,. Tieme Sine. Botanique, Tome XV, p. 263, Tome XVI, p. qi. Also, Bornet and Flahaut, " Revision des Nostocac^es heterocysttes contenus dans les prindpaux herfaiera de France," Annates des Science Nat.. Botanique, 7ieme serie, Vols. Ill, IV, V and VII. [53I gl MACALLKM : CATULOCiV OF NON-NDCLRATRD OROANIRMd sublimate reagents. The former was allowed to act for forty-eight hours, the material was then placed in seventy per cent, alcohol, which was replaced by a fresh quantity daily for a week, after which it was transferred to alcohol of ninety-five per cent, strength. The corrosive sublimate solutions were allowed to act for an hour only, and the sub- sequent treatment with alcohol was the same as in the case of the picric acid material. Material hardened in alcohol was used only for the iron and phosphorus reactions. The .staining reagents found to be serviceable were Ehrlich's and Delafield's h.tmatoxylin solutions, Czokor's alum cochineal, safranin, eosin, picrocarmine and methylene blue. Material hardened with alcohol, picric acid or corrosive sublimate gave very valuable preparations when stained with picrocarmine solution for twenty-four hours, then for an hour with a dilute solution of hjematoxylin. The picrocarmine specially selects the granules which Horzi and Palla term " cj'anophicin," while the hematoxylin thus employed stains the " central body " and its granules particularly and the peripheral protoplasmic zone less dis- tinctly. The " central body " in corrosive sublimate preparations is less clearly differentiated from the peripheral zone. There is one advantage in the u.se of picric acid which does away with the necessity of resorting to transsections of the trichomcs.for the reagent seems to have some solvent or disintegrating effect on the substance of the membranous sheath and of the transverse septa, the trichomes breaking up, under the slightest pressure of the cover-glass, into the separate cells which very frequently then are seen by their flat or end faces. The method of obtaining the reactions for " masked " iron in the Cyanophycea; I have fully described elsewhere.' The iron, liberated by sulphuric acid alcohol, as indicated, was converted into Prussian blue, the trichomes were then stained with a picrocarmine solution for twenty- four hours when the cyanophycin granules acquired a deep red colour which contrasts markedly with the Prussian-blue tint of the ironholding granules. (Fig. 14). Instead of acid alcohol strong solutions of hydrogen peroxide which contained traces of sulphuric acid were frequently used to liberate the masked iron. The demonstration and localization of organic phosphorus were effected in the manner which I have indicated elsewhere,'^ but I may here briefly I " The Distribution of Assimilated Iron Compounds, other than HBmoglobins and Hamatini, in Animal and Vegetable Cells," Quart. Jour, Micro. Sci., Vol. XXXVIII, p. i7j, 189J. a " On the Detection and Localization of Phosphorus in Animal and Vegetable Tissues," Proceedings Roy. Soc., Vol. LXIII, p. 467. 1898. (54) ■' I MACALLUM : CYTOtOOY OP NON-NUCLKATKD ORUANIBMS tf describe the method used. The material, after thorough fixation with alcohol, was washed free from the latter »vith distilled water, then put in the nitric-molybdate solution for three to six hours at a temperature of 35° C, and finally, after (]uickly washing in water, treated for a few minutes with a one per cent, solution of phenylhydrazin hydrochloride, which converts the phospho-molybdatc into a bluish -t'rcen compound, this colour reaction thus indicatin^r the original distribution of the organic phosphorus. To provide against the phosphorus demonstrated being that of lecithin, the material was extracted with hot alcohol in a Soxhiet apparatus for five hours. The other methods resorted to included : digestion of fresh material with artificial gastric juice and with alkaline solutions of trypsin, staining of fresh material for tweity-four hours with acetic-methyl green and subsequent fixation by saturated solutions of picric acid and the treat- ment of fresh material with solutions of Khrlich's hsematoxylin. The digested material was similarly stained and all the preparations were mounted in glycerine. III. — The Living Cell in The CvANOPiivcE/t:. In the larger species of Cyanophyce the cell substance readily reveals the existence of two zones, one, peripheral, coloured blue-green, the other, central, containing granules and, but for the shimmer of blue- green of the outer zone through which it is seen, colourless. In Fig. I the division of the cytoplasm is very clearly indicated. The smaller the cell the greater is the difficulty with which the central uncoloured zone is observed and in those species in which the cell is long and narrow, e.g., Microcoleus terrestris or Oscillaria tenerrima, the central zone cannot be seen, it being of such minute dimensions that its presence is masked by the outer zone. Whenever it can be distinctly seen the central zone, or, as it is usually called, the central body, is found to contain granules which vary in size and numbers, but these most frequently appear in the outer portion of the central body and in some cases the inner portion of the latter may be completely free from them. Apart from the granules the central body is uniform in structure, appearing to possess a vesiculated structure which comes out quite distinctly in Oscillaria princeps. The vesicles in the latter are not closely aggregated and are separated from one another by coarse trabeculae of a colourless, hyaline, plasma-like substance in which excessively minute, punctiform appearances suggest a granulation [55] JO MACAI.LUM ! CVTOLOOV OP NON-NUCLRATBD ORGANIIMM .,ii ! , I I of quite another order that< thiit already referred to. In this species the demarcation of the central body from the coloured, peripheral zone is not sharp and definite, the one passing (gradually into the other. In the peripheral, coloured /one the details are more difficult of observation. One can indeed see granules, but their relation and the distinction between these and portions of the protoplasmic trabecule are difficult to determine in the fresh condition. If a fresh specimen of 0. princefis, 0. Froehlichii or 0. nutans is observed under the apochro- matic 1.5 mm. and compensation ocular 8 or 12 with best light (that from a white cloud), one can see in the very uppermost plane of the cell the membrane studded with refracting points which, when the tube of the microscope is slowly lowered, resemble granules, but at the sides and ends of the cells they appear as elongated elements which radiate from the central body. The refracting points are, therefore, really the terminals of protoplasmic trabecule which connect the central body with the membrane. At certain positions of the objective these trabeculee appear colourless, but at others they have a green tinge which may be due to the reflection of green from the colouring matter. Whether the latter is in the substance of the trabecular or in the fluid held between them it is impossible to say. There are, however, as will be seen after- wards in the description of the structure of the hardened cell, indications that the blue-green mixture is held in the cavities between the trabeculse. On the other hand I have never seen granules holding chlorophyll such as Hieronymus describes. Of the occurrence of a special chromatophore in any form there seems to be not the slightest indication found in the living cell. I regard the granules, which to Palla appeared to contain a mixture of chlorophyll and phycocyan, and the granules observed by Hieronymus as optical trar.ssections of the protoplasmic trabecule which extend from the central body to the cell wall. I have never found these trabeculse formed of other than homogeneous substance and can, there- fore, not support the view of Hieronymus that coloured granules are imbedded in numerous parallel fibres which run in a spiral fashion in the coloured zone about the long axis of the cell. I have never observed what Palla confidently asserts, namely, that the pigment granules are arranged in rows. In the outer or peripheral zone there occurs a large number of granules which do not contain any colouring matter, although they may at times be so situated in the coloured zone as to appear coloured, this being due to their reflecting the blue-green of the surrounding cytoplasm. In the spores of Cylindrospermum majus large granules imbedded in the sub- stance of the peripheral zone, and by their presence forming bays in [56] bu of to sec MACALLUM : CVTULiMiV Ot Ni>N-NtCLKArKII dHUANIBMS •I jccics the )ne is not fficult of ) and the trabeculx ecimen of apochro- ight (that ine of the the tube the sides ich radiate really the body with trabeculae ch may be hether the ;ld between seen after- indications e trabeculae. ophyll such omatophore found in the to contain a observed by leculJE which • found these d can, there- granules are ashion in the ver observed granules are ;r of granules may at times his being due asm. In the J in the sub- ming bays in central body, appear blue-grccn, but that they arc uncoloured can be determined readily by cau .ing tlu; spores tu burst, when the granules, becoming free, appear unpigincntcd. These, to a certain extent, correspond, as will lie shewn l>elow, to the "cyanophycin " granules of i'alla and Horzi. " Cyanophycin " granules arc usually abundant in the Oscittaria, being found in a row at each end of the cell adjacent to the transverse septum, but in Microcolt-ns ti-rrcstris/xn Tolypothrix, Scytofifma and Lyngbia, they are di^ributcd throughout the substance of the peripheral /one. IV.— Fixed I'uki'akations of CvANOPUVCE/ii. When trichomes of OsciUarin }• roc h lie hit have been hardened in picric acid for several days and stained with picrocarmine and hematoxylin, the preparations contain isolated elements like those represented in Figs. 2 and 3. .Similar preparations are obtainable from this form when corrosive sublimate is used as a fixative agent, but the vesiculation of the cytoplasm, while distinct, is less clearly demonstrated than in picric acid preparations. Like preparations may be obtained with Flemming's fluid. The cytoplasm has the structure which Hiitschli claims obtains in it. The character of the vesiculation is, however, not what he describes nor does every trichome even in a picric acid preparation present a similar vesiculation. In all, on the other hand, the vesiculation of the central body appears of a finer character than that in the peripheral zone. When the spaces in the cytoplasm are, as is the case in some trichomes, very minute it would be difficult to determine whether vesiculo .on obtains, were it not that some threads show all the stages between this condition and that in which the vesiculation is distinctly pronounced. In the latter the walls separating the vesicular cavities have a mem- branous appearance with an indefinite character to their borders. Owing to the more minute character of the vesicles in the central body, the latter has a compact appearance, and as it stains with haematoxylin and other dyes some.vhat deeply anv* uniformly it is thereby brought out sharply in contrast with the cytoplasm of the peripheral zone. There is usually not a distinct line of demarcation between them, but the vesiculation of the central body may pass almost abruptly into that of the peripheral zone. The vesicles in the latter are elongated, tending to extend from the central body to the membrane of the cell. In con- sequence of the peripheral layer being thicker at the sides than it is [57] 22 MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS m 1 adjacent to the transverse septa, the membranous bands separating the vesicles in the neighbourhood of the septa are shorter than elsewhere. The bands terminate in a thin delicate layer closely applied to the ceV membrane, which can be seen with difficulty and is most evident when, as in Oscillaria Froehlichii, it is, as in that part adjacent to the transverse septa, rich in granules of the " cyanophycin " class. The latter are usually at " nodal " points of the layer and membranous bands termina- ting in it. The layer may sometimes be distinctly seen in preparations of Oscillaria Froehlichii v/hich have been hardened in alcohol and stained with haematoxylin. In these the cytoplasm has shrunken away from the membrane at one side of the cell in many trichomes and in this case the thin hyaline layer, stained somewhat more deeply than the enclosed cytoplasm, appears sharply contrasted with the latter. In Oscillaria natans, Tolypothrix tenuis, Scytonema sp., and in the vegetative cells of Cylindrospermum w^z/'«j the structure of the cytoplasm in the fixed condition is the same as in Oscillaria Froehlichii. In these forms, however, the cells are much smaller and consequently the vesiculation, which may be clearly observed in the larger form, is only rarely well seen. On the other hand the distinction in the cytoplasm of a central body and a peripheral zone is quite as readily marked as in the large forms of Oscillaria. The central body also stains in haema- toxylin more deeply than the peripheral cytoplasm, mto which it is con- tinued without any sharp transitional changes in staining or structure. In Microcoleus terrestris owing to th2 narrow transverse diameter of the trichomes the differentiation of the cytoplasm into a central b( dy and a peripheral zone is not perceptible. There is a difference between the most centrally placed cytoplasm and that of the periphery, but where one begins and the other ends it is impossible to say. The most centrally placed part stains more deeply in haematoxylin than does the peripheral part and at the same time appears denser. Vesiculation in both parts has been often observed but there is no distinction in the size of the vesicles of the two parts. In the central pa»"t more of cytoplasmic substance appears to surround each vesicle when the vesiculation is distinct. In Oscillaria tenerrima there is apparently no distinction between central part and peripheral layer, except in the deeper staining of the more central ytopiasm. Vesiculation was not observed in this form. (Fig. i6). These facts support on the whole Biitschli's claim that there are two parts, a central and a peripheral, and that the character of the latter is different from that of the former, Biitschli's view that the central [58] MACALLUM : CYTOLOGY OP NON-NUCLEATED ORGANISMS »3 iting the Isewhere. ) the ceM nt when, ransverse atter are termina- parations ohol and ken away nd in this than the nd in the cytoplasm In these ently the m, is only cytoplasm rked as in in haema- li it is con- tructure. liameter of ntral be dy :e betA/een , but where The most in does the iculation in I in the size ytoplasmic iculation is distinction )er staining rved in this lere are two the latter is the central 1 body is the homologue of a nucleus is another question which must be determined, not only on structural but also on micro-chemical grounds. There is not, as already pointed out, a distinct line of demarcation between the central body and the peripheral cytoplasm. There is also nothing to indicate the occurrence of a membrane about the central body. Further, no elements can be found in the central body which correspond to the chromatin nodules or " nucleolar" structures, although as will be shown below, granules of a chromatin-like substance obtain in the more peripheral portions of the central body, but rarely in the central portions of the latter. It must also be pointed out that no observer, with the exception of Scott, has described any structures in the CyanophyceiTc which correspond to those found during n^itosis in the higher organisms. This latter fact, perhaps, taken by itself does not count much against the view that the central body is the homologue of a nucleus, for, as Biitschli points out, the macro-nucleus of Infusoria never exhibits mitotic division, but, taken in conjunction with the other facts regarding the structure of the central body, renders it exceedingly hypothetical that the central body corresponds to the nucleus of the higher organisms. From the micro-chemical point of view the evidence for BUtschli's view would appear to be somewhat stronger. I have already pointed out elsewhere' that the substance of the central body gives indications of the presence of a compound containing " masked " iron. Further observations on a number of the larger Cyanophyceae confirm this. Trichomes of Oscillaria Froehlichii, Tolypothrix tenuis, Cylindrospermum majus, hardened in alcohol, mounted on a slide in a mixture of glycerine and ammonium hydrogen sulphide and kept in a temperature of about 6o°C, gave, after some days, preparations in which one observed a somewhat dark-green tint in the central body. Very often when the preparation had not been a success the result was due to the slow penetration of the sulphide which can only act successfully when the reagent attacking the iron -holding body can be quickly renewed. The iron reaction, when it is obtained in this way, is distinct and does not occur in the peripheral zone. The iron reaction may, however, be obtained more readily in another way. If the trichomes, hardened in alcohol, are placed for two to five hours at 35° C. in sulphuric acid alcohol (acid 4 vols., alcohol, ninety-five per cent., 100 vols.), and then after washing in ninety-five per cent, alcohol, are treated with the acid ferrocyanide mixture (potassium ferrocyanide, 1.5 per cent, solution, hydrochloric acid, I "On the Distribution of Assimilated Iron Compounds, other than Haemoglobins and HKmatins, in Animal and Vegetable Cells," Quart. Journ. Micro. Science, Vol. XXXVIII, p. 175. l59] ii-f'rr a4 MACALLUM : CYTOLOGY OP NON-NUCLEATED ORGANISMS .1 i 0.5 per cent, equal volumes) for five minutes to give the Prussian-blue reaction, or with aqueous hematoxylin (0.5 per cent, solution) for a few minutes, preparations are obtatned which show quite distinctly that the central body contains a substance in which " masked " iron is held. At the same time one species of the granules also may show the presence of " masked " iron and this may be observed when the glycerine sulphide method is used. The Prussian-blue reaction exhibited by the cential body is a uniform one and the depth of the blue colour depends on the thickness of the cell but it is unmistakable in every case. The bluish- black given by the haematoxylin is similarly marked. It is to be noticed, however, in all these cases that the blue of the Prussian-blue reaction and the bluish-black of the haematoxylin are not limited to the central body, for a lighter blue or a faint haematoxylin reaction may be found in that part of the peripheral zone next to the central body, and very often in such preparations it is not possible to say where the line of separation is. This is the case sometimes in Oscillaria Froehlichii and if the cell discs are seen from their flat surface the blue colour is marked in the centre but gradually diminishes towards the periphery and is obscure or absent in the outer part of the peripheral zone. A similar distribution of the " masked " iron in the peripheral zone is shown by the glycerine- sulphide method, but the reaction is not as decisive for minute quantities of iron. There can, therefore, be no doubt that the central body specially, and a portion of the peripheral zone immediately about it in a very much less degree, contain " masked " iron. The reaction for organic phosphorus is not less distinct. Treatment of trichomes, hardened in alcohol, with a nitric acid solution of ammonium molybdate for several hours at a temperature not exceeding 35° C, then with dilute nitric acid for a few minutes and afterwards with a two per cent, freshly prepared solution of phenylhydrazin hydrochloride for from three to five minutes, gave definite indications of the presence of phos- phorus in the production of a dull greenish-blue reaction in the central body specially and only faintly in the cytoplasm of the peripheral zone. Certain granules also give the reaction deeply as in the case of the iron reaction. The reaction in the centra' body is, as stated, a marked one, and it is uniform, shading at the margins of the central body into the feeble reaction of the peripheral zone. That the reaction is due to organic phosphorus and not to inorganic compounds of this element is indicated by the fact that the nitric-molybdate reagent brings out the full reaction after two hours at the earliest, and then only gradually. The presence of " masked " iron atid organic phosphorus in the central [60] 1 sian-blue for a few that the leld. At presence sulphide le central Is on the ,e bluish- e noticed, reaction he central be found and very le line of 'Jtckti and marked in is obscure istribution glycerine- quantities cially, and /ery much eatment of immonium 5° C. then a two per le for from :e of phos- the central heral zone, of the iron larki^d one, ly into the is due to element is figs out the .dually. the central MACALLUM : CYTOLOGY OF NON-Nl'CLEATBD ORGANISMS H body is at least an indication of the occurrence therein of an iron-holding nuclein, like, in some respects at least, the chromatin of the nuclei of higher organisms. This conclusion is supported by the results of the action of artificial digestive reagents in these organisms. When fresh trichomes of Osciltaria, Tolypothrix, Scytonema and Micro- coleus were digested with artificial gastric juice, made by adding a quantity of a strong glycerine extract of the mucosa of the gastric fundus of the pig to hydrochloric acid of 0.2 per cdnt. strength, the preparations at the end of forty-eight, seventy-two and ninety-si.x hours showed certain changes which were due to the digestive reaction- Zacharias found that the peripheral zone is in great part removed, while a portion of the central body disappears leaving two substances behind, one of the nature of plastin, th'' other having a nuclein-like character and receiving from him the name, " central substance." Fischer, on the other hand, denies that the digestive reagent has any such effect and that the diminution of the volume of the contents is due to a contracting action of the digestive ferment in the presence of hydrochloric acid, an action which he terms enzymatic. Fischer's view is incorrect for when one studies at some length the digestive action of artificial gastric juice on the cells in Cyanophyceae, not only is there a diminution in the volume of the cell contents but there is also a disappearance of a portion of its constituents. The peripheral zone is perhaps most affected but a portion of it resists digestion, and after treatment with weak potash solution (0.3 per cent.) remains. This latter substance is undoubtedly plastin. The central body also is affected, but at first sight apparently less so than the peripheral zone. It is rarely diminished in size and it has, as Zacharias points out, specially after treatment with alcohol and ether, a nuclein-like lustre. When, however, the preparation, after pro- longed treatment with alcohol and ether, is stained in a good alum- hzematoxylin solution (Ehrlich's or Delafield's) the central body, instead of appearing homogeneous or appearing uniformly stained as in good fixed preparations, gives in the majority of cases a coarse reticulai appearance in which the trabeculae are usually deeply stained. (Fig. 8). The stainable portion of the central body, the " central substance " of Zacharias, is soluble in weak alkalies, as may be proved by placing trichomes, acted on by gastric juice for forty-eight hours, in a 0.3 per cent, solution of potash for six or seven hours, when subsequent treat- ment with ether, alcohol, and Ehrlich's haematoxylin fails to indicate the occurrence of a stainable substance like that referred to. What remains of the central body and of the peripheral zone stains, but there is no differentiation both parts staining feebly but uniformly. This indicates 16.1 ^"M" a6 MACALLUM : CVTOUXiV OP NON- '.y made preparations. In the central body are frequently large lacunar spaces which in the living cell are occupied by the sulphur granules. The central body also contains minute granules which in position and staining properties correspond to the " red " granules of the Cyanophyceae. In some preparations of ^. alba the peripheral zone was very narrow, so much so that the central body appeared to come almost in contact with the membrane of the cell. \n B. mirabilis the central body is extraor- dinarily large and in it is a large vacuole in the interior of which, in the living cell, are small pale corpuscles in molecular movement. On the surface of the central body is a single layer of vesicles (Waben). Minute " red " granules are also found in the central body. Butschli holds that the central body, thus described, is the analogue of the nucleus of more highly specialized cells and that consequently the peripheral zone of cytoplasm corresponds to the ordinary cell protoplasin of higher organisms. Mitrophanow describes as nuclei of the Sulphur Bacteria structures of the most diverse form and character. In some cases, in Chromatiutn and in Rhabdochromatium for example, it is an irregular, centrally placed mass, elongated parallel to the long axis of the cell. In other cases it I " Ueber den Ba-^ der Bacterien und Tcrwandter Organismen," Leipzig, 1890. Alio, " Weitere AutfUrhruoKcn Qber den Bau der Cyanophyceen und Bacterien," Leipzig, 1896. a " £tudea aur I'organiaation dea Bacteriei." Intern. Monataicbr. fQr Anat. und Physiol., VoL X, p. 47S. 1893. 3 UntersucLungen Uber den Bau der Cyanophyceen und Baktarien," Jena, 1897. l74] MACALLUM ; CYTOLOGY OF NON-NUCLBATBO ORGANIBMS 39 affecting uently in structure however, )rm. The ; are those icteria and ! bod/ and n pr<->:..nty ge lacunar ir granules, jsition and inophyceae. ' narrow, so ontact with is extraor- Df which, in :nt. On the jn). Minute le analogue ;quently the 1 protoplasin structures of ymatiutn and trally placed ther cases it Alio, "Weitere ind Physiol., Vol. contains a collection of granules of chromatin which stain differently from the substance in which they are held, and more deeply. Not un- frequently there may be several masses sometimes completely separate from each other, in other cases connected by narrow strands of the same substance. In some cells also the granules which Mitrophanow terms " nucleolar " are uniformly distributed throughout the cytoplasm of the cell. In such no nucleus is visible. In Ophidomonas jenensis the central elongated mass reminds one of the central body of Butschli, the peri- pheral cytoplasm corresponding to the peripheral layer of that author. This mass may appear divided into a number of smaller vesiculated irregular clumps of substance. In Beggiatoa the masses are always spherical, and, to judge from Mitrophanow's figures, homogeneous in composition. Several of these may be present in the cell, though as a rule there are not more than two large ones. Mitrophanow's terminology is very obscure. He applies the term nucleus to the large elongated, centrally placed mass and the term nucleoli to the granules which it may contain, or to those which may be distributed throughout the cytoplasm. In the case of Beggiatoa the granules found are loosely described by him as nncleoli. He has advanced nothing, except, perhaps, facts in regard to staining capacity, which justify the application of the term nucleus to these structures. In neither Chromatium nor Beggiatoa could Fischer find a differentia- tion of the cytoplasm into central body and peripheral layer, such as Butschli describes. In Chromatium there may at times be a condensation of the cytoplasm in the centre of the cell, brought about by the arrange- ment and disposition of the sulphur droplets which it contains. There may be in each cell a number of spherical grains of a deeply stainable substance which Fischer hesitates to regard as chromatin, for after the use of sotne chromatin-fixing reagents the granules formed of it are not to be seen. In Beggiatoa Delafield's haematoxylin brings out the pres- ence in each cell of " red " granules such as Butschli describes. In subsequent treatment of these preparations with safranin the central portions of the cells stain a little more deeply than their periphery, but this is due to condensation of the cytoplasm through the disposition of the sulphur granules. In cells free from sulphur this result is not obtained. There is no nucleus and the granules cannot definitely be regarded as formed of chromatin. The writer* pointed out that the "masked" iron compound is dis- I " On the Distribution of AHimilated Iron Compouiidt, other than HcmoKlobin and Hsmatin*, in Animal and VcgreUble CelU," Quart. Joum. Micro. Sci., Vol. XXXVIII, p. 158. l7S] w Hi 40 MACALLUM : CYTOLOGY OF NON-Nt'CLEATED ORGANISMS tributed uniformly throughout the cytoplasm and that this distribution corresponds with the diffuse stain given by haematoxylin. In the " comma " forms granules which stained with haematoxylin gave a re- action for " masked " iron. Material and Methods of Study. The forms used were Beggiatoa media, B. alba and B. mirabilis. . The cultures of the two former in water containing sulphuretted hydrogen in solution were always kept in the actively growing condition. These cultures could in twenty-four hours be got to yield myriads of the spirillum-like elements, the " comma " forms, and the " cocci," which, according to Zopf are different stages in the development of Beggiatoa.* One method of fixation was to place a drop of the culture on a cover glass, allow the water to evaporate, and then to float the cover, prepara- tion surface downwards, on a saturated solution of corrosive sublimate, where it was left for a couple of hours, after which it was passed succes- sively through alcohols of fifty, seventy and ninety per cent, strengths. Cover glass preparations were made with ninety-five per cent, alcohol, without the employment of any other reagent. Picric acid in saturated solution was also employed on cover preparations as well as on quantities of the material in various stages. Material in bulk was hardened in ninety-five per cent, alcohol alone. In staining, methylene-blue, safranin, eosin and Ehrlich's and Heiden- hain's haematoxylins were employed. The first three dyes, each used alone, give results of no value, but they may individually be employed with one of the haematoxylins and thereby a more marked demonstra- tion of the vesicular structure of the various forms may be obtained. If, however, the iron-alum hfcmatoxylin method alone is carefully employed it will give preparations which in distinctness leave nothing to be desired. The material from B. mirabilis was hardened, part in absolute alcohol and part in picric acid in saturated aqueous solution.' I " Znr Morphotogie der Spaltpflanzen," Leipzig:, i88j. Also " Die Spaltpiize," Breslau, 1883. a These forms are, according: to Wino^radsky ("BeitrSge zur Morphologie und Physiologie der Bacterien," Leipzig, 1888), not genetically related to Beggialoa. He was unable to find a transformation of the Btggiatoa threads into the "cocci" torms, an experience which befell Engler ("Ueber die Pilz- Vegetation des weissen oder todten Grundes in der Kieler Bucht," Vierter Bericht d. Commission zur Wiss. Untersuchung der deutschen Meere in Kiel, p. 185, Berlin, 1884). 3 For this material I am indebted through Dr. E. C. Jeffrey to Mr. Billings. [76I MACALLUM : CYTOLOGY OP NON-Nt'CLEATRD ORGANISMS 4' General Cell Structure. In fresh actively growing specimens of B. alba the cytoplasm of the filaments is crowded with minute sulphur droplets, and it is difficult to determine the presence of any other structures. One can, with very high powers and good illumination of the microscopic field, see the transverse septa marking the thread off into cells, and at the same time find the cytoplasm next these septa free from granules. In the fixed preparations which have been passed through alcohol and stained the result is different. In Fig. 6o are represented three cells of a thread of B. alba. The sulphur has been removed by the alcohol and the places occupied by the sulphur are shown as clean vacuoles. The protoplasm near the transverse septa appears denser than elsewhere, although because of the aggregation of the vacuoles around the centre sometimes the cytoplasm at the latter point gives an appearance of con- densation. The stain taken by the cytoplasm is uniform throughout the cell, but fine granules may not unfrequently be observed. I am, how- ever, unable to corrobrate Mitrophanow regarding such large granules as he illustrates. When the threads become less rich in sulphur and therefore, ill nourished, large granules may be found, not quite so often indeed, as he observed, but still much more frequently than in the preparations from actively growing cultures. I am inclined to believe that Mitrophanow's preparations were made from ill-nourished cultures, and an examination of his illustrations (Figs. 28 and 30) convinces one of this, for in them is an utter absence of such vesiculation as would be present, had the cells, when prepared, contained sulphur droplets. There is no evidence whatever of the presence of any nuclear struc- ture. The cells of well-nourished threads in no case show a differ- entiation of their cytoplasm into central and peripheral parts according to the views of Biitschli. It is rare even to get, as Fischer did with haematoxylin and safranin.a slightly deeper stain in the central part, and when this does appear it is due to the fact that the central part of the cell is seen through a greater quantity of cytoplasm than is the periphery of the cell. In B, mirabilis hardened in alcohol the cell frequently contains a slightly denser portion of cytoplasm placed adjacent to one of the transverse septa, but an examination of this mass shows that it is really a shrunken portion of the cytoplasm which filled the whole cell. When l77] 4t MACALLUM I CVTOLOOV OF NONiNUCLBATKD OROANISIM the preparations are made with picric acid the frequency of these shrunken masses is greatly reduced. The cytoplasm then usually appears vesiculated throughout and it stains uniformly with hema- toxylin, showing an absence of granules. When the threads of B. alba and B. mirabilis, after being hardened in alcohol, were treated with the nitric-molybdate reagent for two to three hours and then acted upon with a solution of phenylhydrazin hydro- chloride of two per cent, strength, in order to demonstrate the dis- tribution of organic phosphorus, the latter was found to be uniformly diffused throughout the cell. In B. mirabilis the condensed portions observed in some cells, as already described, gave a marked indication of the presence of" marked" phosphorus, but it appeared so because in the shrunken condition to which these masses are due more cytoplasm is gathered into a smaller volume than in cells with unshrunken cytoplasm. Fig. 6i illustrates this. In a and b are observed two shrunken masses of cytoplasm, in these are aggregations simulating granules, while in c the cytoplasm shows the organic phosphorus uniformly distributed through- out the cell. In B. alba the phosphorus is distributed uniformly with the cytoplasm and the method did not reveal the presence of granules. The reaction for iron derived from the " masked " condition, is in B. alba and B. mirabilis found to be uniform with the distribution of the cytoplasm. When sulphuric acid alcohol is used to set the organic iron free in the threads and the preparation is washed free from acid with abso- lute alcohol and stained with a pure aqueous solution (one per cent.) of haematoxylin, the result is decided enough to determine definitely that there are no specialized chromatin-holding structures like nuclei or like Biitschli's central body. Granules sometimes found distributed in the peripheral portion of the cytoplasm give the reaction. The distribution of the " masked " iron being then like that of the organic phosphorus, it follows that the substance containing these ele- ments, the analogue of the chromatin of more highly specialized cells, is contained, not in any nucleus however rudimentary, but diffused in the cytoplasm, and sometimes, also, localized in granules. Somev;hat different are the results of observations on the spirillum- like form, and on the " cocci," and comma-shaped organisms. Here, as little as in the thread or " leptothrix " forms, is there any evidence of the existence of a nucleus, rudimentary or otherwise. In these, the vesicles occupied by the sulphur droplets, are all crowded about the centre, or about the central axis, leaving the peripheral layer as a thin, homo- geneous structure applied to the membrane. In the "spirillum" form the [78] MACALLUM : CVTOLOOV OF NON-NUCLEATED ORGANISMS 43 vesicles are separated from each other by a thin film of cytoplasm, and here and there, and especially at the nodal points in these films, one observes, in haematoxylin preparations, granules which are more deeply stained thnn the rest of the cytoplasm. Sometimes these granules are more or less drawn out in the films between the vesicles, and then the central cytoplasm may take on the character of the central body of Biitschli. Of the actual existence of such a central body there is not the slightest evidence. In the "comma" forms the granules are very much fe^ver, and fre- quently smaller. In the " cocci " they are rarely visible, and minute lightly staip^ble granules make their appearance at the periphery of the vesicles (Fig. 59, a, b, and c). In the "spirillum," as well as in the "comma" form, the compounds of " masked " iron and organic phosphorus are, apart from the granules, faintly diffused throughout the cytoplasm,sometimes apparently less abun- dant in the central portions of the cells, and definitely indicated in the peripheral layer. The granules give a slightly more marked reaction for " masked " iron as well as for organic phosphorus, and this fact, com- bined with their capacity for taking up haematoxylin, would seem to indicate that they are formed of a compound analogous to chromatin. The difference between the structure of the threads, and that of the "spirilla" and the "comma" forms in regard to granuK.o, may be due to some inherent difference in the process of nutrition in the two different types, but it may also be explained on the view of Winogradsky that these types are not genetically connected, that they belong to different species of Sulphur Bacteria. Whether this is correct or not does not affect the question that, in all these types of structure, there is nothing to simulate a nucleus, even of the most rudimentary description. It can scarcely be contended that the granules, the affinity of whose substance to chromatin is shown by their containing, in a small degree, " masked " iron and organic phos- phorus, are morphologically of the value of nuclei, or even of nucleoli, as Mitrophanow appears to claim. [99I 44 MAOALLUM > CVTOLOOY OP NON-NUCLRATBD OROANIIIMS Summary. 1. In Beggiatoa alba and B. mirabilis there is no differentiation of the cytoplasm, as Biitschli finds, into a central body and a peripheral layer The centrally placed portion of the cytoplasm contains, in the well- nourished fresh cell, a number of sulphur droplets, and frequently the cytoplasm between the droplets may be more condensed than at the periphery of the cell, but usually both portions stain equally deeply. 2. The compounds of " masked " iron and organic phosphorus are uniformly and equally diffused throughout the cytoplasm in the threads of B. alba and in B. mirabilis, and the organic phosphorus is found uni- formly distributed in those cells which contain unshrunken cytoplasm. When granules which stain with haematoxylin occur, they are found to contain " masked " iron and organic phosphorus. 3. In the "spirilla," in the "comma" forms and in the "cocci," the cyto- plasm shows characters like the cytoplasm of the threads, but there are, in addition, granules which give slight reactions for " masked " iron and organic phosphorus, and which, therefore, are constituted of a sub- stance analogous to chromatin. 4. There is, in aP these forms, no specialized chromatin-holding struc- ture in the shape of a nucleus of any kind. (80I MACALLt'M : CYTOLOGY Of NON-Nt'CLKATRD ORGANISMS 45 THE YEAST CELL. I.— Literature. The structure of the Yeast Cell, and more especially the question whether it has a nucleus or not, has been the subject of investigation by a large number of observers since 1844, when Niigeli' affirmed the exist- ence of a nucleus in this organism. Chief amongst these were Schleiden' (1849), Briicke' (1861), Schmitz* (1879), Strasburger* (1884, 1887), Zalewski' (1885), Krasser^ (1885 and 1893), Hansen" (1886), Zacharias' (1887), Zimmermann'" (1887), Raum" (1891), Moeller" (1892 and 1893), Hieronymus" (1893), Janssens'* (1893), Dangeard'*(i893 and 1894), Macallum" (1895 and 1898), Buscalioni" (1896). Wager'* (1897 I Zeit. far Wiss. Botanik, Vol. I, p. 45, 1844. a "GrundzUge der Wins. Botanik," 1849, p. 307. .t "Die Elementar-urcfanismen." SitziinKsber. der K. Akad, d. Wins, zu Wiun, Math-Nat. ClaiM, ■861, Vol. XLIV, Ahth. a. 4 " Unterxuchungen Ober den Zellkern der Thallophyten," Sitzungsber. der Niederrhcin. Gesell. fflr Natur-und Heilkunde zu Bonn, Sitzungr am 4 Aug.. i87<)- J " Das Botanische Practicum," p. 339. 1887. Also edition of 1884, 6 "On Spore Formation in Yeast Cells." Trans-ictions of the Scientific Academy of Cracow. (Polish). Vol. XIII, 1885. Abstract in Bot. Centralbl., Vol. XXV, p. 1, 1886. 7 " Ueber das angebliche Vorkommen eines Zellkerns in den Hefezellcn." Oestereich, Bot. Zeits., 1885; also: "Ueber den Zellkern der Hefe : " ibid., 1893, p. 14. 8 " Recherches sur la Physiologie et la Morphologie des Ferments Alcooliques." Mcddelser tra Carlsbergr Laboratoriet, Vol. II, p. 15a, 1886. 9 " Beitrilge zur Kentniss des Zellkerns und der Sexualzellen." Botanische Zeitung, 1887, Nos. i8-a4 10 "Die Morphologie und Physiologie der Pflanzenzelle." Breslau, 1887, p. aj. II "Zur Morphologie und Physiologic der Sprosspilze." Zeitschr. fur Hygiene, Vol. X, p. i, i8qi. la " Ueber den Zellkern und die Sporen der Hefe." Centralbl. fUr Bnkt. und Parisitenkunde. Vol. XII, p. S37, 189a; also: " Weitere Mittheilungen tiber den Zellern und die Sprosse der Hefe:" ibid., Vol. XIV, p. 3j8. 1893. 13 " Ueber die Organization der Hefezellen." Ber. d. deutsch. Bot. Gesell., Vol. XI, p. 176, 1893. 14 " Beitrage zu der Frage uber Kern der Hefezelle." Centralbl. fUr Bakt, und Parasitenkunde, Vol. XIII, p. 639. 1893. 15 " Sur la structure histologique des levures et leur diveloppement." Comptes Rendus, Acad. d. Sciences, Vol. CXVII, p. 68, 1893 ; also: "La structure des levures et leur diveloppement." Le Botaniste, 1894, p. a8a. 16 " On the Distribution of Assimilated Iron Compounds, other than the Haemoglobin and Haematius, in Animal and Vegetable Cells," Quart. Journ. Micro. Sci., Vol. XXXVIII, p. a43, 189s; also: "On the Detection and Localization of Phosphorus in Animal and Vegetable Tissues." Proc. Roy. Soc., Vol. 63, p. 467. 1898. 17 "II Saccharomyces guttulatus Rob." Malpighia, Vol. X, 1896. 18 " The Nucleus of the Yeast Plant." Report British Association, Toronto Meeting, 1897, p. 860 ; alto: ibid., Bristol Meeting, 1898, p. 1,069, and Annals of Botany, Vol. XII, p. 499, 1898. [8.1 MACALLl'M I CVTOUXtV OP NON-NUCLBATKD ORGANIBMS t ' 1: and 1898), Janssens and Leblanc' (1898), and Bouin' (1898), a fairly full account of whose observations, as of those of others, is given by Wager (1898) to which the reader may be referred. Many of the observations, especially those of more than ten years ago were made with imperfect means and methods, and although there is amongst the older observers almost an unanimous agreement on some points, as for example, the presence of a nucleus, it must be now recognized that with the employment of the simple methods alone used by them, no observer can maintain with the same degree of certainty that he can see the structures which they found, or can admit that what they found is what they claimed it to be. I propose here, therefore, to limit any discussions of the observations to those made within the last seven years, and particularly to those of Moeller, Buscalioni, Wager, Janssens and Leblanc and Bouin, for these observers not only gave thorough attention to the structure of the yeast cell, but also used very much improved methods of fixation and staining. I may add in justification of my not giving an account of the earlier observations that in 1895 ^ discussed these at r.ome length.' Moeller in his first paper claimed that the yeast cell contains a nucleus which is homogeneous and without a membrane. This nucleus changes its shape readily, and, therefore, its position in the cell varies. Owing to this property it may assume a thread-like form when budding occurs, a portion of which is thus enabled to pass into the protoplasm of the bud through the narrow connecting tube. The part which projects into the bud breaks off and separates with the bud, a.ssuming finally the rounded form of the mother nucleus. In the spores, how- ever, Moeller could not find any evidence of the presence of a nucleus, but in his later communication he states that he found the nucleus in the spore element, and he describes its character. The nucleus of the cell, at the beginning of spore formation, enlarges and becomes elongated and constricted at the middle. The constriction deepens, the ends separate to the opposite poles of the cell, and the fine thread joining the two parts breaks, two daughter nuclei being thus formed. A second and similar division follows. The division is a direct one. Janssens found in S. cerevisicB and in S. Ludwigii a nucleus pro- vided with a homogeneous nucleolus and membrane, the diameter of the nucleolus being one-third that of the nucleus. The nucleus divides 1898. Recherche* cytologiquet sur la cellule de levure," La Cellule, Vol. XIV, p. aoj, iSqS. a "Contribution k I'itude du noyan des levurea.' Arch. d'Anat. Microscopique. Vol. I, p. 435, 3 guart. Journ. Micro. ScL, Vol. XXXVIII, p. MS. [81] MACALLUM t CYTOLOOV OP NON-NUCLKATIO OROANIIMS 47 B in the mitotic fashion, and Janssens claims to have observed the equa- torial plate and dyaster stages. In spore formation the nuclear mem- brane disappears, and the plane of the second division is at right angles to that of the first, the two spindles also being at right angles to each other. Each spore is provided with a nucleus. Later Janssens, in conjunction with Leblanc, published a fuller and somewhat different account of the structure of the cell. They found in 5. cerevisice and 5. Ludwigii a nucleus with a distinct nuclear mem- brane, caryoplasm, and a nucleolus constituted of nuclein. The caryo- plasm is formed of a fine network of fibrils intimately connected with each other and applied to the nucleolus. When, however, the cells are put in a fresh culture medium, the nucleus becomes vacuolized, but the nucleolus maintains its shape and central position, and the protoplasm remains homogeneous. The vacuolated condition ceases at about the thirteenth hour. In a longer stay in the medium the protoplasm becomes granular. Ordinarily the cytoplasm is formed of a typical reticular structure, the meshes of which contain granules. Both the contents of the meshes as well as the reticulum and its nodal points, in some cases, manifest a strong affinity for colouring matters, and it is evi- dent that this is due to a nucleo-albuminous substance dissolved in this structure. When the cells are grown on plaster blocks the granules may become very large and refracting, but when sporulation begins these granules disappear, presumably contributing a portion of the material which constitutes the spores. Glycogen in the cell ordinarily is dissolved in the enchylema, but when it is very abundant it localizes itself in vacuoles which may fill the cell. When the cell buds, the nucleolus elongates and divides, but the two parts remain united by a strand of substance apparently like fibrils, which may slightly resemble a spindle. The two nucleoli pass toward that part of the cell which is giving rise to the bud. The nuclear membrane and the caryoplasm disappear, leaving the nucleoli nude. A structure which resembles, to a certain extent, a cell plate, makes its appearance in the cell between the two nucleoli. The process up to this point is a rudimentary form of kinesis. One of two nucleoli slips into the bud, and now, if not before, the nuclear membrane re-forms about both. The bud thus constituted is, in all respects, like the mother cell. In the formation of the spores, the authors find that early in the spore mother cell the nucleolus divides, as already described, but the nuclear membrane does not disappear. This is not accompanied by any divi- sion of the cell. After some hours, however, only one nucleus with a large nucleolus is observed. The authors believe that the two nucleoli 183] T 48 MACALLUM I CVTOLOOV OF NON-NUCLKATIO OROANUMB originally present have fused or conjugated, that is, fecundation haa taken place. This is followed by division of the large nucleolus, the elongation taking place in the long axis of cell, In each ol the two nucleoli thus formed a second division occurs, immediately following the first, the elongation of the two daughtf nucleoli being at right angles to each other. When the four nucleoli are fully formed the membrane develops about each, and around each of the nuclei thus constituted the protoplasm collects, a membrane finally forming about each mass, which becomes a spore. ■■t According to Buscalioni, whose observations were made on S. guttu- latus, the yeast cell contains a nucleus which ordinarily is homo- geneous, and its division is by constriction. This obtains when budding occurs, the two daughter nuclei remaining connected by a thin fibril until one of them enters the bud. This is simple fragmentation. In the formation of spores, however, the nucleus undergoes a slightly different species of division which may be looked upon as a rudiment- ary kind of kinesis. Bouin found a sharply defined nucleus which, during fermentation sends prolongations into the cytoplasm. The latter are less sharply ned the further they proceed from the centre of the nucleus. The js ordinarily may be granular, or may contain in its interior irregular, deeply stainable masses. In some cells a nucleus would appear to be absent. In these an intense stain may serve to show a nucleus poor in chromatin ; but cells which do not appear to have a nucleus have lost it by its transference, without division, into the bud. The nucleus may, under certain conditions, divide and re-divide, while the cell may remain undivided, and Bouin holds that this multiple division of the nucleus accounts for the number of chromatin granules found in some yeast cells. The granules are, in this case, nuclei, and the cell containing them, multinucleate. The division of the nucleus ordinarily is amitotic, that is, there is elongation of the nucleus with constriction, the thread uniting the two ends becoming more and more delicate till rupture occurs. One of the two daughter nuclei thus formed passes through the canal between mother cell and bud, and into the latter, where it becomes spherical. No striation of the cytoplasm between the two daughter nuclei was observed, nor was any evidence of an equatorial plate and of chromosomes found. In the formation of spores the nucleus dl /ides into two chromatin masses, which separate, then become rounded, arid constitute the nuclei of the spores. This mode of division is intermediary between mitosis and amitosis. I MACAIXl-M I CVTOLOOV Or NON-NICLRATKD ORUANICMS 49 Wa|:cr, from his first observations, claimcet with small ones, and the homogeneous nucleus is centrally placed. In it, granules, however, soon make their appearance, which accumulate in the centre, and look like a nucleolus. In division, the outline of the nucleus becomes irregular, and the granules arrange them.selves in the form of a short rod, surrounded by other portions of the nucleus, which stain differently, and appear to form a structure like a spindle. The granules form two groups, each of which constitutes a nucleus, and each of the two nuclei (' ides in the same way, forming thus the nuclei of the spores. Around each of these nuclei protoplasm accumulates and a membrane forms. This form of nuclear division Wager regards as a simple form of karyokinesis. In S. Ludwigii, he found a nucleus with membrane, and a nuclear network and nucleolus the latter containing all the chromatin. In division, the nucleolus increases in size, and divides, each part becoming a nucleus. In the paper detailing his later observations. Wager gives a fuller, and, in some respects, a considerably different account of the yeast cell. In the fresh, actively growing organism the cell contents are clear and homogeneous, with .sometimes one or more bright refracting granules. In this condition a vacuole or vacuoles can be seen, and in each occurs at least one refracting particle which is in a state of movement. The vacuoles disappear in a later stage of fermentation, and the protoplasm then appears homogeneous and clear ; but, when the culture medium becomes exhausted, the contents become granular and possess fat globules, the protoplasm shrinks from the cell wall, and the cell presents an appearance of disintegration. In compressed yeast, on the other hand, the cells are rich in refracting granules, which are sometimes uniformly distributed through the protoplasm, sometimes located around the vacuoles, or grouped together at one side of the cell. In regard to the nuclear apparatus, Wager distinguishes two struc- [8sl so MACALLUM ; CYTOLOGY OF NON-NUCLBATED OROANISMS tures: one of which he calls the nuclear body, while he terms the other the nuclear vacuole. The nuclear body, which is the nucleus of Moeller, and of his own earlier observations, is homogeneous, but surrounded more or less completely by granules, and which, with low powers of magnification, give it a granular appearance. It is, in actively growing cells, usually in close contact with the cell wall, but it may, in a few cells, be more centrally placed. In relation frequently to this nuclear body there are granules, first described by Hieronymus, some of them of an oily nature, others of a proteid character, sometimes grouped about the nuclear body, sometimes in its immediate neighbor- hood or distributed throughout the cell. At times these form a coiled thread. The nuclear vacuole, which is in contact with the nuclear body in growing cells, contains chromatin, sometimes in the form of granules, sometimes in the form of a network, sometimes as an irregularly shaped mass attached to the wall of the vacuole by fine threads. In some cells all the chromatin substance appears to reside in the vacuole ; in others it is diffused through the protoplasm, and in some cells again it appears in the nuclear body. The vacuole Wager regards as the nucleus of Janssens and Leblanc, and the nucleolus of these observers con- stitutes his nuclear body. The nuclear vacuole may persist but for a short time. After fermentation has proceeded for some hours it dis- appears, and its place is occupied by a granular network in contact with the nuclear body. The vacuoles seem to arise by fusion of minute vacuoles which develop in connection with what appears to be chroma- tin granules. In regard to division, Wager found that the nucleolus (the nuclear body) is separated from the bud by the vacuole, which, as the bud deve- lops, begins to pass into it. At the same time, the nucleolus makes its way to the base of the opening, and there, or in the neck, at once begins to elongate and constrict for division. The vacuole at this time divides, but not completely or equally, the smaller portion being found in the daughter cell, both parts remaining connected by a granular thread. The divisions of the nuclear body are nearly or quite equal, and one of them m»^kes its way into the daughter cell. When the nucleolus is in the neck, the constriction takes place with the ends in the mother and daughter cells. When there is no vacuole, the granular network in contact with the nuclear body undergoes division into two more or less equal portions, either in the mother cell or in the neck of the bud. The granules which the young bud thus receives seem to develop, in some way, the vacuoles which form the single large nuclear vacuole. In sporulation the nuclear vacuole disappears, or its place is taken by [86] w| ni MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS 5« two or more smaller ones which in turn are replaced by many others and, as a consequence, the protoplasm acquires the foam structure of Butschli. The nuclear body moves towards the centre, the protoplasm condenses about it, and on the periphery of this condensed zone deeply stainable granules collect. The nuclear body now appears to take up from the surrounding^ cytoplasm all the stainable material, and to deposit it in its centre .is a granular mass. The division of the nuclear body occurs as Wager describes it in his earlier paper. In regard to the nature of the nuclear apparatus, Wager regards it as a simple form of a nucleus, although he admits the possibility of it being either a primitive structure representing an early stage in the organo- geny of the nucleus, or a degenerated form of nucleus. The division of the nuclear body, he thinks, may be regarded as a case of direct division, but, in his opinion, it may also be a very simple case of kary- okinesis. In my own studies on the distribution of assimilated compounds of iron, I pointed out that, in S. cerevisice and S. Ludivigti, chromatin is to be found distributed throughout the cytoplasm of the cells and, some- times, also in the latter in the form of granules ; but, in S. Ludwigii, it may be found chiefly at the periphery of each large vesicle, when only a few large vesicles are present in the cell. In this form also there is a substance constituting corpuscles of a nucleolar character, the nuclei of Moeller, which stains with eosin, and gives a marked reaction for iron, but c'iffers from chromatin in remaining unstained after treatment with haematoxylin. My conclusion was that there is no nucleus, although such an organ may occur in other stages of this organism. In a later contribution embodying the results of observations made to determine the distribution of organic phosphorus in animal and vegetable cells, I pointed out that in the yeast cell the phosphorus-holding substance, or nucleo-proteid, although sometimes in the form of granules or spherules which have been taken for nuclei, is frequently dissolved in the cyto- plasm. It will be seen from this abstract of the more recent literature on the yeast cell, that there are some discrepancies in the views of various investigators of the subject. An agreement is indeed found as regards the division of the nuclear body or nucleolus, but not as to the manner of this division. The most radical difference, perhaps, exists between Wager, on the one hand, and Janssens and Leblanc on the other, as to what constitutes the nucleus, and as to its structure apart from the nuclear body or nucleolus. [87] mam i» MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS II. — Method of Study, and Species Studied. ... ■• I The fixing reagents which I used were corrosive sublimate and picric acid, each in saturated aqueous solutions, the chrom-osmio-acetic mixture of Flemming, and a one per cent, solution of potassic iodide saturated with iodine. Undoubtedly for yeast cells the best of these, in my experience, are corrosive sublimate and Flemming's fluid. They produce less alteration in structure than any other, and they do not, ivhen properly used, alter the staining power of any part of the yeast cell. The most satisfactory method of employing them was to allow them to act in bulk on a large number of yeast cells, separated from the actively growing cultures by centrifuging the latter. After the reagents had acted sufficiently long the fluid was decanted, distilled water was poured upon the cells, which were immediately subjected to centrifuge action and thus separated, when they were treated with alcohols of gradually increasing strengths. With these reagents also I obtained reliable cover-glass preparations, in which drying of the cells was not a factor, by spreading a film of the yeast culture on the cover-glass, and then, with this face downward, placing it floating on a quantity of the solution, to remain there for from one to twenty-four hours. The fluid, at the moment of touching, removed very many of the elements, but enough were left adherent to make a good preparation. Afterwards the cover-glass so treated was placed in alcohol of 30, 50, 70, a.id 90 per cent, strengths successively. The cover-glass method of preparation could not be employed with the other reagents except to some extent in the case of iodine, as these remove from the cover all the cells. The only safe way of fixing with these solutions was to allow them to act on the yeast cells in the test tube. They were separated completely from the fluid, after the required time for complete fixation, by centrifugalizing the fluid. The hardening in these cases was completed by the use of alcohols of 30, 50, 70, and 90 per cent, strengths successively. The cells were completely separated in these cases by gravity. In the case of the iodine solution the method employed by some of allowing a film of yeast cells to dry on a cover- glass, and then placing it in the reagent, has no advantage over the one described, and I am not sure that it is free from objection. It is dif- ficult to believe that yeast cells can be unaffected when the fluid about them is completely removed by evaporation. I have, therefore, avoided this method of preparation, as well as that in which heat alone is used for the purpose of fixation. [88] the MACALLVM : CYTOLOGY OF NON-NUCLEATED ORGANISMS 53 and icetic odide se, in They 3 not, yeast allow im the agents er was trifuge lols of )tained s not a iS, and of the le fluid, Its, but rwards a.id 90 ed with IS these ng with the test required irdening , and 90 iparated method a cover- the one It is dif- jid about , avoided le is used With regard to iodine solutions more particularly, it is necessary to use a word of caution. The prolonged action which is required with this reagent affects the proteids of the cell, and must change consequently the reactions of the cellular structures to staining solutions. I have found that it alters and changes the staining capacity of the cell in Spiro- gyra, as well as of various animal tissues, and it can scarcely be admit- ted that it has no effect on the cytoplasm of the yeast cell. I find that it greatly diminishes the affinity of the chromatin distributed through- out the cell for haematoxylin and other dyes which, in the case of cor- rosive sublimate preparations, select only the chromatin. The reagent does, indeed, assist in fixing the yeast cells in such a way as to single out for special demonstration a spherical, more or less homogeneous body, known to certain investigators as the nucleus, and to others as the nucleolus or nuclear body ; but this property depends on the power of the iodine to change the chemical character of the cytoplasm in a greater degree than that of the nuclear body which is homogeneous and dense. Amongst the staining reagents which were used were hsematoxylin, safranin, eosin, and acetic-methyl green and methylene-blue. The solu- tions of haematoxylin gave the best results and, more particularly, Delafield's, Ehrlich's, and Meyer's (haimalum). The solutions were made very dilute, so much so that it required always from sixteen to eighteen hours' application of the fluid to bring out the full stain. The iron- alum haematoxylin of Heidenhain was also employed, and it is of value in revealing the structure of the cytoplasm of the yeast cell, but it is of no value as a micro-chemical reagent, and it does not show any sharp distinction between chromatin-holding and chromatin-free cytoplasm. Eosin was used as a counter stain to haematoxylin. Acetic-methyl green was employed on the fresh cells, and methylene-blue on the cover- glass preparations. The organic iron and phosphorus compounds were demonstrated in the manner described in the case of the Cyanophyceae. The yeast cells were, for this purpose, always hardened in alcohol. To reveal the dis- tribution of organic iron the cells were mounted on a slide under a cover-glass, in a mixture of glycerine and fresh ammonium hydrogen sulphide, and the preparation kept at a temperature of 6o°C. for a week. To demonstrate the organic phosphorus, the cells were kept in a solu- tion of ammonium molybdate in nitric acid for five hours, after which they were washed in distilled water for a few minutes, and then sub- jected to the action of a 2 per cent, solution of phenylhydrazi , hydro- chloride, which converted the molybdic portion of the phospho-molyb- I89] 54 MACALLUM : CYTOLOOY OF NON-NUCLEATED ORGANISMS date into a greenish-blue compound. The cells, washed with distilled water and dehydrated, were mounted in balsam. The species studied were Saccharo myces cerevisice, S.Lud'wigii,AnA two found in cultures obtained from the throat in suspected cases of diph- theria. The specimens of S. cerevisice and S. Ludwigii, employed for the purposes of preparation, were in the actively growing condition in Pasteur solutions, from which a quantity of the cells, separated by centrifugalization, was taken hourly, starting with the commencement of growth, up to the twentieth hour, and treated as indicated above. Cul- tures of 5. Ludwigii, in the sap of the iron-wood tree, Ostrya virginica, and of the maple, gave very valuable and instructive preparations. For the study of sporulation 5. cerevisice was used, the sporulation having been brought about by cultivation in a 5 per cent, solution of sugar, as indicated by Wager. III. — General Cell Structure. In the fresh yeast cell at the beginning of fermentation, even with the highest powers of magnification, very little can be made out, except the occurrence of granules and vacuoles. These are grouped irregularly in the cell and their number and character may vary, although as a rule there is but one large vacuole. In some young, actively growing yeast- cells, that is in those which are observed two hours after the commence- ment of fermentation, there may be no vacuoles observable. In such, however, one or more granules may be found. This condition may be observed also in cultures of from sixteen to twenty hours. Many of these granules appear to possess a fatty nature. When the yeast cells of this stage are hardened in Flemming's fluid for twenty-four hours the granules take a dark tinge, due apparently to the reduction of the osmic acid derived from the solution. They are not nearly as numerous in preparations hardened in alcohol as they appear to be in the fresh cell. At a later stage of fermentation the granules present are of a different composition. They do not react, or at most react but slightly, with the osmic acid of Flemming's fluid. They seem to be of a purely proteid character, for when the hardened cells are heated with a solution of potassic plumbate the granules acquire a light brown colour, this fact indicating the presence of organic sulphur. These granules were found to react also with freshly made Millon's reagent in from eight to ten hours without the application of heat. I90I me MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS S5 tilled dtwo diph- ;d for ion in :d by ent of Cul- ■gitttca, . For having igar, as ivith the cept the ilarly in ,s a rule g yeast- nmence- In such, I may be /hen the enty-four uction of nearly as r to be in resent are react but to be of a ed with a ivn colour, e granules in from It is of course possible that granules which give the reaction with osmic acid do not consist wholly of fat, for the reaction in all cases is not intense enough to suggest a purely fatty compo!,ition. The basis of apparently all the granules seems to be a proteid substance which may vary in its particular character from stage to stage in the process of fermentation, and in these granules the fat which may be demonstrated is deposited. The structure of the cytoplasm varies. When the cells of 5. Ludwigii of early stages of fermentation are hardened in Flemming's fluid and stained in Heidenhain's haematoxylin, we get an appearance like that illustrated in Figs, 36, n, and 38. In these the cytoplasm is shown to contain a reticulum whose meshes are delicate and whose nodal points are thickened. In some cells, as for example in Fig. 38, the reticulum forms a ring around a cospuscle, whose nature will be discussed below. In other cases the reticulum is in intimate connection with the corpu.scle. In corrosive sublimate preparations a reticulum is not readily observ- able, and this is due to the property the reagent has of fixing not only the reticular portion of the cytoplasm, but all the proteids in its meshes, whereas the acetic acid of Flemming's fluid dissolves out some of these. Indications, however, of a reticulum can be found if the stain of the iron-alum haematoxylin is carefully decolourized with weak iron-alum solutions. When the cells have been prepared with iodine solution and stained with iron-alum haematoxylin the reticulum shown is coarser as a rule than in Flemming's fluid preparations, the meshes are larger and the trabeculae thicker. I am inclined to regard this result as due to the iodine reagent which fixes the cytoplasm slowly, and consequently may permit plasmolytic alterations. The presence of vacuoles affects only slightly the reticular structure, merely condensing the cytoplasm in their immediate neighborhood. In corrosive sublimate preparations which have been carefully stained with Delafield's or Ehrlich's hjematoxylin or with Meyer's haemalum, the cytoplasm gives unmistakable evidence of the presence of chromatin diffused through it as well as localized at particular points. This difl"use distribution causes the whole cell to be deeply stained when one employs the ordinary staining reagents, in the concentrated form which is usual in the case of other cytological preparations. It is the most striking point that one finds when one for the first time makes preparations of yeast cells, and this being so it is surprising that little attention has been given to it in the literature of the subject. (9'1 S6 MACALLUM : CVTOLOOV OF NON-NUCLEATED ORGANISMS In the cytoplasm as Errera' has shown, may be found glycogen, and it occurs in abundance in the cells of the later stages of fermentation. When iodine solution is applied to the cells from Pasteur solutions in the first nine hours after fermentation begins, very rarely only does it show the presence of glycogen, and then only in the form of minute granules. The slight brown tint which the cytoplasm in general at this time gives is due to the absorption of iodine by the cytoplasmic chromatin, and is not due to dissolved glycogen. In cells of from ten to thirteen hours of cultivation in Pasteur solutions the glycogen occurs in small masses in the cell. These masses, of which there is usually one to each cell, vary in size, and are more or less irregular in outline and placed adjacent to the cell membrane. In later stages the glycogenic mass may be so large as to occupy the greater part of the cell. .:>.. ... ). '■""■■■ 1 IV.— The Chromatin-holdixg Structures. In yeast cells which have been hardened in Flemming's fluid or in corrosive sublimate solutions, and stained with very dilute solutions of Ehrlich's or Delafield's haematoxylin applied for from sixteen to twenty hours, one finds, as already pointed out, a slight colour, due to the presence of chromatin in the cytoplasm generally, and a very deep stain at one or more points in the cell. The latter may also be demonstrated by employing the iron-alum haematoxylin method for staining, but as it is not selective its action is less clearly indicative of the presence of chromatin, or of substances allied to chromatin, than that of the staining reagents mentioned. The diffuse stain which is given to the cytoplasm may serve to obscure the presence of a structure or structures which may be present. One frequent type of this structure is, ordinarily, a spherical mass like that represented in Figs. 36, 37, and 38, and, as in these cases, varying somewhat in size. This body, which I may term, for the sake of brevity, the corpuscle, is, in the great majority of cells, homogeneous and dense, and it stains much more deeply than the cytoplasm generally. It is not, however, always present, for it appears to be absent in cells in the different stages of fermentation, and no method of hardei Ing and staining the cells will demonstrate its presence in all. This has been admitted by Bouin and Buscalioni. The former observer tried, by deeply restaining cells which at first appeared to be free from corpuscles, to bring out the presence of the latter, but succeeded only in I Report Britiih Auuci'ation, Briitol Meeting, iSgS, p, i,a68. [92] MACALLUM : CVTOLOOV OF NON-Nl'CLEATKD ORGANISMS 57 a few of such cells, in which the corpuscles were found to be deficient in chromatin. Buscalioni believes that yeast cells exist which are free from these bodies. Sometimes, on the other hand, one finds yeast cells also which contain not only one but several corpuscles, each, however, smaller than the single corpuscle of other cells. Very rarely in preparations of S.cerevisia but very frequently in those of 5. Ludwigii as cultivated in the sap of the iron wood tree {Ostrya virginica), the corpuscle instead of being spherical and homogeneous, is irregular in contour and consists of one or more deeply stainable, dense granules, imbedded in a substance which constitutes the greater part of the corpuscle, and which is less markedly affected by dyes. Sometimes the irregularities in the contour may be so great as to give the corpuscle a stellate appearance. Bouin observed corpuscles of similar shape and structure in 5. cerevisicB and 5. pastorianus. The corpuscle is the nuclear body of Errera and Wager, and the nucleus of Moeller, Bou'n, Buscalioni and others. As these authors describe it, it divides by a process which is a simple form of karyokin- esis, and, therefore, it is in their view a fully developed chromatin- holding organ. According to Janssens and Leblanc the corpuscle is a nucleolus of a nucleus which can, by appropriate methods, be revealed as surrounding and containing the nucleolus. This nucleus further is provided with a membrane which becomes invisible when division of the nucleolus takes place, the caryoplasm also disappearing. ' In preparations hardened with iodine solution and stained with dilute haematoxylin, a body like the nucleus of Janssens and Leblahc can be observed surrounding the " nucleolus," but it is found only in a small number of cells, whereas in the greater number the " nucleolus " or corpuscle lies free in the cytoplasm. In preparations ahso made with Flemming's fluid and stained with iron-alum haematoxylin, the corpuscle is rarely found to be included by a structure like that described by these authors, and when the latter is observed it contains no chromatin and does not give any evidence of structure in its interior. It is as such quite different from the corpuscle already referred to, to be found in 5. Ludwigii when cultivated in sap. What it is I am not certain, but I am inclined to regard it as a vacuole which, placed above or below the corpuscle, may with the latter strongly resemble a nucleus and nucleolus. In iodine preparations stained with iron-alum haematoxylin, the structure in question may be absolutely unstained while the " nucleolus " or corpuscle, and the cytoplasm are deeply coloured. On the other hand, one may find a vacuole, whose wall is rich in stainable material, overlie [93] 8> MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS or underlie a corpuscle, giving exactly one of the conditions described and illustrated by Janssens and Leblanc. The corpuscle does not in the fresh cell react with acetic-methyl green like a chromatin body. Preparations of this reagent, which would bring out clearly in fresh animal and vegetable cells the chromatin-holding structures, left the corpuscles of the yeast cell, even after hours, unaf- fected, while the cytoplasm stained readily. Another point of contrast is to be found in its affinity for eosin, which is more readily absorbed and retained by it than is hematoxylin. In the throat yeast the corpuscle was, in the great majority of instances, more or less irregular in outline, always excentrically placed, and in the great majority of cells, in close contact with a vacuole. Some Mnes it was crescentic in outline, the vacuole fitting into its con- cavity. In this form very little chromatin was found in the cytoplasm and as a consequence the corpuscle stood out quite clearly. In S. Ludwigii as it grew in the sap of the iron-wood tree, fixation with Flemming's fluid and staining with iron-alum haematoxylin and eosin demonstrated the corpuscle frequently as a reddish body, having at times a tint of blue-violet and surrounded by a garland-like structure formed of deep blue-violet-stained chromatin. This structure, on closer analysis, is found to be formed of granules and elongated masses of chromatin. Sometimes it appears open or discontinued at one side, and, especially if the corpuscle is pear-shaped, its prolongation extends beyond the limits of the chromatin structure. Usually, however, in these preparations the structure is closely applied to the corpuscle. When there are two corpuscles in the cell there is a structure in question about each of them. Rarely the corpuscle appears separated from the cyto- plasm, and, consequently, from the garland-like structure, by a zone of clear space. (Fig. 28). I regard this garland-like structure formed of chromatin as quite the same as the membrane of fine granules closely applied to, and com- pletely surrounding, the " nucleolus," as described by Wager, who found that the granules and nucleolus stained difierently. In my preparations of S. cerevisioe the membrane does not appear as uniform and as regular as Wager figures it ; and its irregularity, and sometimes the absence of close contact between it and the corpuscle, remind one strongly of the conditions in S. Ludwigii. The point to note specially is the contrast in staining presented by the corpuscle and the structures surrounding it. The latter has a marked affinity for haematoxylin, while the former absorbs eosin readily. Sometimes, in corrosive sublimate preparations I94J MACALLl'M : CYTOLOGY OF NUN-Nl'CLBATED ORGANISMS 59 Stained with very dilute solutions of h.-ematoxylin, the corpuscles may be unstained, or stained no more deeply than the cytoplasm generally. (Fig. 28). In ordinary cultures of S. Ludwigii the garland-like structure may be so infrequently present in its typical form as to be overlooked. Then in many cells granules like those found by Wager and Bouin in S. cerevtsicB may be observed. This illustrates how much, as regards structure, is dependent on the mode of cultivation of the yeast cell. When yeast cells, hardened in alcohol, were submitted at 37°C. to digestion in artificial gastric juice, made by dissolving some glycerine extract of pepsin in a 0.2 per cent, solution of hydrochloric acid, after forty-eight, seventy-two, and ninety-six hours there were very few evidences of the occurrence of corpuscles remaining, nor was there any stainable substance left in the cytoplasm. The corpuscles seem to be affected very much by the digestive solution, for even when the cells were deeply stained with the iron-alum hsematoxylin, cor- puscles were only rarely found, and then they appeared with a large vacuole in their interior (Figs. 39 and 41), or to have lost their stainable substance (Fig. 40). These traces of the corpuscle also disappeared after treatment of the preparations for 24 hours with a 0.1 per cent, solution of potassic hydrate. The results of the treat- ment with artificial gastric juice seem to indicate that the stain- able substance in the cytoplasm and in the corpuscles is different from the chromatin of the nuclei of other organisms, which is unaffected by this fluid. Similar results were obtained when fresh cultures of S. cerevisia, in Pasteur solution, were digested for forty-eight hours or more in artifi. cial gastric juice, after which the cells were fixed with iodine solutions and hardened in alcohol. Only very rarely in these, even after the most careful apf i:v.ation of the various methods of staining, and especially of the iron-alum haematoxylin process, did I succeed in finding traces of a corpuscle. The cells of such preparations seem to have lost their power to stain with Ehrlich's and Delafield's haematoxylin solutions, but to have increased their affinity for eosin. The micro-chemical reactions are quite decisive as regards the rela- tionship of the stainable substance. When the glycerine-ammonium sulphide method is employed to demonstrate the organic iron in cells of 5". Ludwigii and 5. cerevisia, the results obtained after ten days at the latest distinctly demonstrated the presence of " masked " iron in the one or more corpuscles which may be present in each cell in the walls [95] 6o MACALLUM : CVTOLOUV OF NON-NUCLKATKU OKOANIKMN of the vacuoles, and in the cytoplasm generally (Fi^s. 42-49). The reaction is usually intense in the corpuscles, so much so that they may appear not dark [;rcen but black from the excess of ferrous sulphide developed in them by the reagent. The reaction in the cytoplasm is much less marked but distinct. It may be diffuse, but if tiie prepara- tion, though successful, has been only a few days in the reagent, one may observe, in a majority of the cells, that the reaction is limited to the cytoplasmic network (Figs. 50 and 51). In .some cells a series of granules constituting a membrane about the corpuscles like those already described, gives a very distinct reaction (Fig. 46). Sometimes the number and size of the corpuscles thus demonstrated suggest the character of granules, but there is, in the majority of such cases, one at least which is of the usual size (Figs. 45, 46, 47, and 49). The reaction in the wall of the vacuole m?y be very distinct, and especially in granules located in it. (Figs. 42, 43, 4.1., and 48). On applying the method to demonstrate the presence of organic phosphorus, the later is found to be localized in the same manner as the organic iron. The corpuscle is rich in it, and the wall of the vacuole in contact with the corpuscle gives a distinct reaction for it, and at times specially in the granule.^ found in it. Portions of the cyto- plasm, which appear to correspond to the nodal points of the reticulum, give a deep reaction (Fig. 56). It is evident from the presence of "masked" iron and organic phos- phorus, both distributed in the yeast cell to an extent parallel with the distribution of the substance which stains with haematoxylin, that the organism contains in its corpuscle or corpuscles, in its cytoplasm, as sometimes in the wall of its vacuole, a substance closely related to the chromatin of higher organisms, but differing from the latter in the effect exercised on it by artificial gastric juice. The stainable substance found in the corpuscles further differs from ordinary chromatin in that it has no affinity for acetic-methyl green. There remain now to be described structures which are, so far as my observations go, to be found only in S. Ludwigii, when cultivated in the sap of the iron-wodd tree. The character of these structures is seen by an examination of Figs. 21-27, and 29, in which they are illustrated. They are not all of the same type. As in Figs. 21 and 29 one may observe a large nucleus-like body in which a network, somewhat like that found in a fully typical nucleus, exists. There may also be a corpuscle in this structure which simulates a nucleus. In some cases the structure in question may resemble a nucleus in the stage preparatory to the for- [96] MACALLl'M I CVTOLUOV OF NON-Nl'CLKATRU ORUANIHMB bl mation of the chromosomes (Fig. 23). In other preparations, one may obtain cells in which a scries of vacuoles is found in close contact with each other, and all surrounded by a substance which stains deeply in haematoxylin (Fif^s. 22 and 24). Rarely one sees structures like those illustrated in F'ifjs. 26 and 27. In Fig. 25, there are in mother and daughter cells structures allied in general form to those found in Figs. 22 and 24. A somewhat similar instance of vacuolation of chromatin-like masses may be observed in mycelium-like threads developing sometimes with the cells of 5. Ltidwigii in sap cultures. In these the vacuoles vary from an almost infinitesimal size to that of extraordinarj' dimen- sions (Fig. 58). The largest ones appear to be formed of a large num- ber of vacuoles fused through the rupture of the more centrally placed partitions (Fig. 58^). In a few cases the fused vesicles may form a very large structure, presenting some resemblance to a nucleus (F"ig. 57). Whether these structures are formed of that variety of chromatin which is to be found in yeast cells cannot be decided as yet. The cells containing them are so few in any preparation made with the glycerine- ammonium sulphide method to show the distribution of organic iron, or with the nitric-molybdate reagent to determine the occurrence of organic phosphorus, that they must only very rarely be observed. In only one sulphide preparation did I see a cell which appeared to contain a structure like one of tho'jc in question (F^ig. 52). In this case the iron reaction of the substance forming the structure was quite marked. It would seem to indicate that these structures are formed of yeast chromatin. There can be no question about the nature of these structures. They certainly are not nuclei, either normal or degenerated. They owe their form and arrangement to a property of chromatin which, I believe, has not hitherto been regarded as characteristic of it. In the Cyanophyceae, as already described, the chromatin-like substance not dissolved in the " central body " forms spherules, in the centre of each of which may be found a vacuole (Figs. 4, 6, and ii). In the F'oraminifer, Calcituba poly- morplta Roboz,' the nuclear chromatin before division, at first homo- geneous, undergoes extensive vacuolation, and upon this process division depends. Indeed the structure of the nucleus ordinarily would appear to depend on the inherent power of chromatin to produce vesi- culation or vacuolation, with the formation ultimately of a reticular I F. Schaudinn, " Untersuchungen an Foraminiferen. I. Calcituba polymorpha Rcboz," Zeit. fur Wiss Zool,, Vol. LIX, p. 191, i8i)S. [97I 6t MACALLl'M I CVTOI.OUV OF NON-NCCLRATKt) OHOANINMH arrangement. This vacuolation may sometimes be observed, in a marked degree, in the masses of chromatin from chromatoiysed nuclei in higher animal and vegetable organisms, and it is manifested even in chromatin nucleoli in normal nuclei. This indicates that chromatin secretes fluid in certain conditions, and that viicuoles are formed bv this secretion. In the cells of Saccharomyces under consideration, it is not certain, as was pointed out, that the structures which simulate nuclei are formed of a chromatin-likc substance, but it is probable that they are constituted of it, and this would explain their appearance, although in plasmolysed chromatin a like richness of vacuolation has not yet been observed. I would regard these structures as caused by extensive vacuolation of chromatin-like masses formed in the cytoplasm. v.— Budding and Spokulation. In the process of budding a portion of the cytoplasm is forced into a diverticulum of the cell membrane, the quantity at first forced out being small, but eventually the bud may contain from one-third to one-half of the cell contents. It is only in this way that we can explain the " streaming out " appearance of the cytoplasm in the neck of the bud. We may see vacuoles elongated and extending into the bud (Figs. 30, 32, 42, 44 and 48), or one or more of Raum's granules having extended dumb-bell shapes, the extremities of which lie in the mother and daughter cells. Sometimes also one may find one of the peculiar reti- culated, chromatin-like masses, described above, occupying, as a dumb- bell shaped figure, the neck of the bud. The pressure to which the cytoplasm of the mother cell is subjected and the narrow passage of the neck of the bud tends to draw out and elongate all the structures which are forced through the narrow neck. These conditions are responsible for the elongation and constriction of the corpuscle as described by Bouin, Janssens and Leblanc, Wager and others, who regard these phenomena as constituting evidence of nuclear division. I have found in many instances that the corpuscle is thus divided between the mother and daughter cells. When the corpuscle is found in the neighbourhood of the commencing bud, it is, with the cytoplasm surrounding it, forced to the opening, which is rarely large enough to permit its passage, and if the diameter is large enough the cytoplasm that is driven with it prevents the corpuscle from passing through the opening. The corpuscle, being plastic like the cytoplasm, may completely fill the passage and project into the interior of the bud, its dumb-bell form being then quite marked. The constriction may I98] MACALLl'M I CVTOLOOY Of MUN-NVCLiATBP URUANMMI <3 deepen until only a fine strand connects the two terminal spheres, and when the bud further develops there may lie a complete separation of the two parts, one remaining in the mother c'.l, the other forming; the corpuscle of the daughter cell. Sometimes, however, the whole corpuscle is forced through the neck into the bud. This is to be found not rarely in the sap cultures of S. Ludivifrii and rarely in S. a'trvisiie. Hus- calioni believes that this happens sometimes in S. guitulatus, and Houin found that it docs occur in Mycoderma cerevisur. The latter author would thus explain the absence of a nucleus from some cells. it is in this way, I believe, that the complete absence of a corpuscle in the mother cell and the presence of a large one in the daughter cell may be ex- plained. I have also founrl the bud in a few chills of S. Lmhvigii ^xo\\x\ in sap to contain two small corpuscles, while the mother cell gave not the slightest evidence of the presence of a corpuscle. In the.scca.ses one of the daughter corpuscles, after their formation by constriction of the parent structure in the manner described, which should, as is usually the case, remain in the mother cell, is carried with the cytoplasm into the bud. In ^\ Luihvigii buds may develop and separate without the con- striction and division of the corpuscle of the mother cell. This is specially the case when the corpuscle is in a part of the cell remote from the commencing bud. Wager states that in this case the nuclear body (the corpuscle) makes its way to the opening of the mother cell into the bud and then begins to divide. This may happen, but I have found in a number of instances the bud full grown, while the corpuscle remained undivided in the remote part of the cell. There can be but one interpretation of these facts. The elongation and constriction of the corpuscle are the results of purely physical forces and conditions, such as operate on the cytoplasm in the neighbourhood of the bud, and the constriction and resulting division of the corpuscle are not absolutely necessary factors in the development of the bud. The formation of two corpuscles out of one in this way can scarcely be regarded as a case of direct division or simple karyokinesis, as some observers have claimed it to be. In S. Z-w^TV/^/V many ofthe buds contain cytoplasm richer in "masked" or organic iron than that in the parent cell. This would indicate that the cytoplasm of the bud is richer in chromatin, and the results of staining with haematoxylin seem to support this view. It is not infre- quently found that the cytoplasm lining that part of the membrane of the bud remote from the neck stains deeply and gives a deep reaction [99] I 04 MACALLUM : CYTOLOGY OP NON-NUCLEATED ORGANISMS for organic iron with the glycerine-sulphide method. Perhaps this dis- tribution may be explained by supposing that the first cytoplasm to pass out into the developing bud is richer in chromatin-like substance. Quite different is the action of the corpuscle in sporulation. Here, however, the cytoplasm also acts differently. When the cell is ready to sporulate the chromatin dissolved in the cytoplasm begins to concen- trate in a zone about the corpuscle, the diameter of the zone diminish- ing as this stage advances, whilo the corpuscle appears to lose its distinctness. The concentration advances until all of the cytoplasmic chromatin is collected in a very narrow zone about the corpuscle which, in some cases, may appear very finely granular. At this stage occurs an elongation of the corpuscle and its enclosing body of cytoplasmic chromatin, the elongation rarely being to the full length of the cell. The central portion constricts or becomes more and more slender, until separation of the more or less rounded extremities occurs. Thus two corpuscles are formed, each witii a very narrow enclosing zone of cytoplasmic chromatin. Each of these now elongates and divides as the parent structure does. In the second division, however, the cyto- plasmi chromatin seems to disappear, or perhaps is taken up into the corpuscle. My observations on the whole agree with those of Wager, but I have never been able to find the granules which form in the corpuscle or nucleolus immediately before and during its elongation, as described and illustrated by tnat observer. Nor can I corroborate his view that the nuclear vacuole, which exists in the cell previous to sporulation divides and redivides many times, thus distributing the chromatin through the cytoplasm, which has, in consequence, a delicate foam-like ; tructure. So far as my observafions go, the chromatin in the cyto- plasm before the stage of sporulation commences, is not different in its character or distribution from that found in the ordinary yeast cell, for example, during budding. According to Janssens and Leblanc, the act of sporulation is preceded by a division of the nucleus and its nucleolus, followed by a fusion of the two nuclei thus produced. These authors believe that this fusion or conjugation constitutes sexual fertilization. They find that the two nuclei formed disappear, and in their place, one only, whose nucleolus is large and distinct, is observed. It is rathe: difficult to accept this interpretation. What they claim to have observed as constituting a nucleus may, as I h-we pointed oul, be found in some cells only, and their nucleolus is the corpuscle, two or more examples of which may some- [loo] MACALLUM : CYTOLOGY OP NON-NUCLEATRD ORGANISMS 65 times be found in the cell. It is, therefore, not Impossible to find frequently in cells in cultures beginning the sporulation stage two small corpuscles. It is, however, another matter to prove that these corpuscles f':se to constitute a sinp;le large corpuscle, such as may be found in .jther cells of the same preparation in which no evidence of iusion aving occurred can be observed. According to these authors, a spinHle formed of very fine parallel threads constitutes the connecting strand in the division of what they term the nuclei in sporulation. I have never been able to observe such a structure, but it is possible that what is found in the varieties of yeast used for observing it by Janssens and Leblanc, may permit the demonstration of such a spindle more readily than in the forms I em- ployed. I must say, however, that in 5. Ludwigii, employed by them also for this object, I was unable to find anything resemble a spindle of fine threads. I found in a number of cells of S. cerevisice, a structure which resembles very much that described by Janssens and Leblanc, and compared by them to a cell plate. It was a line formed of delicate, closely placed granules running transversely to the strand connecting the two developing corpuscles and completely dividing the cell into two halves. The dotted line was in the majority of cases so fine and so difficult to see properly that it required the best illumination and the highest available magnification to bring it out. Whether it is to be regarded as a cell plate cannot at present be determined. The division found in sporulation can scarcely be described as a simple form of karyokinesis. It is rather to be compared to the division of a chromatic filament in the formation of two chromosomes. A more analogous case is that of the division of the nucleolus in Euglena viridis, as described by Blochman' and Keuten'-. In this form the nucleolus, at the commencement of division, does not disappear as it ordinarily does in other cells, but remains in all the stages. When the chromosomes which are formed in the normal way begin to constitute the dyaster stage, the nucleolus elongates into a dumb-bell figure, the constriction first observed deepening, until complete separation of the spherical ends takes place. Each of these passes into the corresponding daughter nucleus. In this case, while the ordinary chromatin of the nucleus undergoes division by the karyokinetic method, the nucleolu- undergoes direct division. In sporulation in Saccharomyces it is difficult I "Ueber die Kerntheilunff bei Euglena." Biol. Centralbl., Vol. XIV, p., 194, 1894. a " Die Kerntheilung von Euglena viridis Ehrenberg." Zeit. fUr Wiss. Zool., Vol. LX, p. 115, 1895. (.0.] 66 MACALLUM : CYTOLOGY OF NON-NUCLEATED ORGANISMS to believe that anything more complex occurs than this nucleolar division. If, on the other hand, the division in sporulation is regarded as kary- okinetic, it must be also held to be an exceedingly rudimentary type of that process. It would in fact have to be admitted as differing so little from what is called direct division as to be almost indistinguishable from it. [102] m- MACALLCM : CYTOLOGY OF NON-NUCLEATED ORGANISMS 67 Summary. 1. In Saccharomyces the cytoplasm is usually finely reticulated, and contains one or more vacuoles. It takes a diffuse stain with haema- toxylin, and gives a diffuse reaction for " masked " iron and organic phosphorus. 2. In addition to the chromatin-like substance diffused throughout the cytoplasm, there is usually a more or less homogeneous, spherical body in the cell, the corpuscle, the " nucleus," " nucleolus," and " nuclein body " of various observers, which stains specially with haematoxylin. and gives the reactions for " masked " iron and organic phosphorus, but does not stain with acetic-methyl green. This body is neither a nucleus nor a nucleolus. Several examples of it, though of small size, may be present in a cell. On the other hand, cells are found without a trace of a corpuscle. 3. The chromatin-like substance differs from the chromatin of higher animal and vegetable cells in being soluble in artificial gastric juice. 4. When budding begins, the corpuscle, if placed adjacent to the point where the bud is developing, becomes elongated and constricted in its middle portion. One end of the elongated structure may be forced into the neck of the bud, and when the constriction is completed by separation of the two halves, the daughter cell may thus receive a corpuscle. Both daughter corpuscles may pass into the bud, leaving the mother cell without a corpuscle. A complete bud may be formed with- out such a division of the corpuscle taking place, and thus the ''augh- ter cell may commence independent life without a corpuscle. 5. In sporulation the cytopla mic chromatin collects in the immediate neighborhood of the corpuscle, which also undergoes certain granular changes, then elor '^ates with constriction in its central part. Each of the two daughter c .rpuscles thus formed repeats this process of division one or more times, th daughter corpuscles resulting ultimately forming the corpuscles of the spores. 6. The division of the corpuscles in budding is a purely mechanical result, and is not essential to the formation of the bud. The division preparatory to sporulation is apparently a functional act. It is not of the nature of true karyokinesis, and it may be compared to the division of the nucleolus in Eugkna viridis. [103] t§ MACALLUM : CYTOLOGY OF KON-NUCLEATED ORGANISMS 7. In cells of Saccharomyces Ludwigii, from a culture in sap, one finds, rarely, structures which strikingly remind one at times of a nuclear organ. These structures are apparently formed of a chromatin-Hke substance, and their existence is due to a property which chromatin possesses of forming vacuoles in itself. When the vacuoles are fully formed, the portions of chromatin-like substance separating the vacuoles may be so delicate as to suggest the occurrence of a network like that of a nucleus. Since the foregoing paper was written, the results of an investigation by Ascoli,* conducted in Kossel's laboratory on plasmic acid, a variety of nucleic acid, have been published. The preparation of plasmic acid examined was obtained from yeast cells, and was found to contain about I per cent, of iron in a " masked " or organic form. This con- firms what I have hitherto advanced, and what 1 have described in this paper, regarding the presence of " masked " iron in combination with a nucleic or chromatin compound in yeast cells. I " Ueber die Plaminsaure." Zeit, fUr Physiol. Chemie, Vol. XXVIII. p. 426, 1899. t«'-4] MACALU:M : CYTOLOGY OF NON-NUCLEATED ORGANISMS C<» EXPLANATION OF PLATE. All the Fifpires were outlined from the objects with an Abbe camera Iticida and as viewed with a 3mm., 2mm., or 1.5mm. apochromatic oil immersion objective, and an 8, 12 or 18 compensation ocular (Zeiss). Fig. I. — Cylindrospermnm majus. Fresh, 48 hour's culture in the laboratory, x 2,000. Fig. 2. — Oscillaria Froehlichii, A cell viewed from the flat surface. Picric acid, picro- carmine, haematoxylin. x 3,000. The transition from the central body to the peripheral zone, as represented in the figure is, unfortunately, rendered too abrupt. Fig. 3. — Oscillaria Froehlichii, View of two cells from the side, optical section. Picric acid, picrocarmine, haematoxylin. x 3,000. Fig. 4. — Tolypothrix sp. Corrosive sublimate, hasmatoxylin. x 1,500. Fig. 5. — Tolypothrix sp. Digestion with artificial gastric juice three days, extracted with alcohol and ether, Ehrlich's haematoxylin (dilute) 24 hours, x 1,500. Fig. 6. — Tolypothrix sp. Strong Flemming's fluid 24 hours, Delafield's haematoxylin. x 1,500. Fig. 7. — Tolypothrix sp. Corrosive sublimate, Delafield's hsematoxylin. x 1,500. Fig. 8. — Oscillaria natans. Digested with artificial gastric juice 6 days, alcohol 1 day, alcohol and ether 8 hours, haematoxylin. x 1,500. Fig. 9. — Oscillatra Froehlichii. Digested with artificial gastric juice 6 weeks, alcohol and ether, KHO 0.3 per cent. 24 hours, picrocarmine. x i,j6o. Fig. 10. — Microcoleus terrestris. Digested with artificial gastric juice 3 days, haema- toxylin, glycerine, x i ,500. The dye has given a greenish-blue tinge to the cytoplasm in the central part. Fig. II a and b. — Microcoleus terrestris. Alcohol, a, hydrogen proxide 3 hours, acid ferrocyanide solution, balsam ; h, haematoxylin, glycerine, x 1,500. Figs. 12 a.nA \j,. — Tolypothrix sp. Alcohol, hydrogen peroxide 3 hours, acid ferro- cyanide solution, glycerine, x 3,000. Fig. i^.— Oscillaria Froehlichii (?). Alcohol, hydrogen peroxide 3 hours, acid ferro- cyanide solution, picrocarmine, balsam, x 3,000. Fig. 15. — Lyngbiasp. Picric acid, picrocarmine. x 1,000, Fig. 16. — Oscillaria ttnerrima. Picric acid, picrocarmine, haematoxylin. x 3,000. Fig. x-j.— Tolypothrix sp. Picric acid, haematoxylin, balsam, x 1,334. Fig. 18 a and b. — Cylindrospermum majus. Old culture, acetic-methyl green. x 3,000. Fig. iq.— Oscillaria natans. Fresh culture, artificial gastric juice 48 hours, picrocarmine, glycerine, x 2,250. Fig. ao. — Tolypothrix sp. Alcohol, nitric-molybdate 10 hours. Phenylhydrazin hydro- chloride, balsam, x 1,334. liosl 7« MACALLUM : CYTOLOGY OF WOM-HWCLIATID OmOANISMR Figs, h -ig-Sacckaromycts Ludwigii. Corrosive sublimate 8 hours, Hicohol, DelafielcT* hsematoxylin (very dilute) 17 hours, x 3,000. FlOS. ya-ja-—Saccharomycts Ludwigii. Alcohol, Delafields hematoxylin (very dilute) 48 hours, balsam, x 2,400. Kios. 3^.39,,— Saccharomyces Ludwigii. Corrosive sublimate, alcohol, Delafield's hasma- toxylin (very dilute) 18 hours, x 2,400. FiOS. ^-2fi.—Saccha>omyces Ludwigii. Flemming's fluid, iron-alum hematoxylin, balsam, x 2,00a Fios. iQ-^i.—Saecharomyces Ludwigii. Artificial gastric juice, 96 hours, iron-alum haematoxylin, eosin, balsam, x 2,000. F108. ^2-^q.—Saccharomy(es Ludwigii. Alcohol, glycerine and ammonium hydrogen sulphide 10 days, x 3,000. FlOS. si>-$i.—Sai(Aaromyc*j Ludwigii. Alcohol, glycerine and ammonhim hydrogen sulphide 6 days, x 2,250. Pigs. 52-55.— SaccAaromyces Ludwigii. Alcohol, glycerine and ammonium hydrogen sulphide 10 days, x 2,250. Fig. 56.—Scucharomyces Ludwigii. Alcohol, nitric-molybdate 5 hours, phenylhydrazin hydrochloride, balsam, x 3,000. Figs. 57-58 «-