IMAGE EVALUATION TEST TARGET (MT-3) <■:•'' -^V^ 1.0 I.I 1.25 |50 IIM m 6 IIM IM 1.4 IIIIII.6 V] .^^^ ..^m /y e ^1 0% ;v ^m ' ^y./ ^^i^^ ' A yw Photographic Sciences Corporation 23 WEST MAIN STREET WEBSTER, N.Y. 14580 (716) 872-4^03 tm CIHM/ICMH Microfiche Series. CIHM/ICMH Collection de microfiches. Canadian Institute for Historical Microreproductions / Institut canadien de microreproductions historiquas >ji Technical and Bibliographic Notes/IMotss techniques et bibliographiquas The Institute has attempted to obtain the best original copy available for filming. Features of this copy which may be bibliographically unique, which may alter any of the images in the reproduction, or which may significantly change the usual method of filming, are checked below. y Coloured covers/ Couvetture de couleur I I Covers damaged/ Couverture endommagde Covers restored and/or laminated/ Couverture restaur^e et/ou pellicu^e Cover title missing/ Le titre de couverture manque Coloured maps/ Cartes giographiques en couleur Coloured ink (i.e. other than blue or black)/ Encre de couleur (i.e. autre que bleue ou noire) Coloured plates and/or iilustrations/ Planches et/ou illustrations en couleur Bound with other material/ Relii avec d'autres documents D n Tight bindirg may cause shadows or distortion along interior marain/ Lareliure serree oeut causer de I'ombre ou de la distorsion le long da la marge intdrieure Blank leaves added viuring restoration may appear within the texv. Whenever possible, those have been omitted from filming/ II se peut que certaines pages blanches ajouties lors dune restauration apparaissent dans le texte, mais, lorsque cela dtait possible, ces pages n'ont pas iti film^es. L'Institut a microfilme le meilleur exemplaire qu'il lui a exi possible de se procurer. Les details de cet exemplaire qui sont peut-dtre uniques du point de vue bibliographique, qui peuvent modifier une image reproduite, ou qui peuvent exiger una modification dans la mAthtde normale de filmage sont indiquis ci-dessous. The to tl I I Coloured pages/ Pages de couleur Pages damaged/ Pages endommag^es □ Pages restored and/or laminated/ Pages restaurees et/ou pelliculdes E Pages discoloured, stained or foxed/ Pages d^color^es, tacheides ou piquees n Pages detached/ Pages d4tachees QShowthrough/ Transparence □ Quality of print varies/ Qualite indgale de I'impression □ Includes supplementary material/ Comprend du materiel supplement The posi of tl film Orifi begi the sion othc first sion or iil supplementaire I I Only edition available/ Seule Edition disponible n Pages wholly or pa-tially obscured by errata slips, tissues, etc., have bean refilmed to ensure the best possible image/ Les pages totalement ou partieilement obscurcies par un feuillet d'errata, une peiure, etc., cnt 6t6 film^es i nouveau de facon ^ obtenir la meilleure image possible. The shall TINl whi( Map diffe entir begii right requ meti Q Additional comments:/ Commentaires suppl^mentaires- Pagination is as follows : 175 - 274 p. This item is filmed at the reduction ratio checked below/ Ce document est film* au taux de reduction indiqu« ci-dessous. 10>i MX i8X 22X 12X 16X 26X 30X / 20X 24X 28X 32X The copy filmed here has been reproduced thanks to the generosity of: Library, Geological Survey of Canada The images appearing nere are the best quality possible considering the condition and legibility of the original copy and in keeping with the filming contract specifications. Original copies in printed paper covers are filmed beginning with the front cover and ending on the last pagA with a printed or illustrated impres- sion, or the back cover when appropriate. All other original copies are filmed beginning on the first page with a printed or illustrated impres- sion, and ending on the last page with a printed or iilusirated impression. The last recorded frame on each microfiche shall contain the symbol — »> (meaning "CON- TINUED"), or the symbol V (meaning "END"), whichever applies. Maps, plates, charts, etc.. may be filmed at different reduction ratios. Those too large to be entirely included in one exposure are filmed beginning in the upper left hand corner, left to right ano top to bottom, as many frames as required. The following diagrams illustrate the method: L'exemplaire filmi fut reproduit grAce A la g6n6rosit6 de: Bibllothdque, Commission Giolc\e ee\W ^ /B VM aeaW om REPRINTED FROM I THE QUARTERLY JOURNAL OF ! MICROSCOPICAL SCIENCE. M. i : Ji^ [RON COMPOUNDS IK ANIMAL ANU VEGETABLE CELLS. 175 On the Distribution of Assimilated Iron Com- pounds, other than Haemoglobin and Hsematins, in Animal and Vegetable Cells. By A. B. Macnlliini, ITI.B., Ph.D., Associate-rrofessor of Fliysiology, University of Toronto. With Pktes 10—12. I. Preliminary Remai.ks and References to Special Litera- ture OF THE Subject ..... II. Methods of Study ...... III. General Observations on the Distribution of Assimilated Iron in Highly SrEciALisED Animal and Vegetable Cells .... In tbe Nucleus of the Animal Cell In tl.e Nucleus of tbe Vegetable Cell In the Cytoplasm of the Animal Cell In the Cytoplasm of the Vegetable Cell IV. On the Occurrence of Assimilated Iron in Special Forms of Life In Ascaris mystax In the Larvff! of Chironomus In Protozoa In Fungi . In Bacteria In the Cyanophyccffi V. General Remarks VI. Explanation of Figures Page 175 179 205 205 211 2U 227 228 229 231 236 242 254 261 268 271 I. Preliminary Remarks. In 1891, in a communication to the Royal Society ,i I described a method by which the presence of iron in the chromatin of I " On tiie Demonstration of the Presence of Iron in Chromatin by Micro- chemical Methods," ' Proceedings Roy. Soc.,' vol. 1, p. 277. VOL. 38, PART 2. — NEW SER. M • '. • t ••' 1. .■ i 1 7t) A. r.. MACALLUM. animal and vcufetablc cells may bo demonstrated micro-cliemi- cally, and I referred to the results then obtained with it as indicatinp:, apparently, that iron is always a constituent clement of this sui)stance. The interest wliich the subject had for me led mc to continue the investigation with improved methods of research, and I am now consequently in a position to describe a much more extensive scries of observations in support of the generalisation, then somewhat tentatively advanced, that iron is a constant constituent of the nuclein substance proper. From the commencement of the investigation I have been fully aware of its difficulties, and I can, therefore, readily under- stand that view of the subject which led Gilson to remark that the solution of the question concerned is one " that seems to require more than a single man's activity." ^ The difficulties encountered in the application of the micro-chemical method are, however,, very much less formidable than those met with in the employment of the older methods. I have pointed out, in my first paper on this subject, how impossible it is to be certain that the iron revealed by macro-chemijal methods in isolated quantities of nuclein is not present through absorption from some other source, but due to a combination obtaining in unisolatcd living chromatin, and I have indicated that the ouly way in . which the question could be settled definitely is by the employ- ment of micro-chemical methods. I have shown in the suc- ceeding pages of this paper that the acid alcohol upon which Bungc relied to extract the iron of inorganic and albuminate compounds from egg-yolk and other nucleiu-holding substances, and leave intact the organic (nucleinic) iron, does not perform this function at all when the substance treated with it is in mass, while it removes the iron of all three classes of compounds from thin sections of tissues, if the time allowed for its action be prolonged. We have, consequently, in a macro-chemical investigation, no means whatever of distinguishing between organic iron on the one hand and the iron of inorganic and albuminate combinations on the other, and we are therefore ' "On the Affinity of Nuclein for Iron and otlier Substances," 'Report Britisli Association for the Advancement of Science,' 1S92, p. 778. • • ■■•« ♦. * • .■• • ••-••• • ••• • ••••••••.• « • • -• • •• • • • • •• • • • •• • •• •• • *•• •••••_• • •••• •. •< • ••• ••• i IRON COMPOUNDS IN ANIMAl, AND VEGETABLE CELLS. 177 forced, more than ever, to depend on micro- chemical methods to determine the relations of assimilated iron to the cell. Objections may be urged against these methods also, based chiefly on the facts that iron, free or combined, contaminates everything, so to speak, and that what is shown to occur in dead chromatin may not be present in the living compound ; but these objections at once lose their force when the methods are applied with all due care accompanied by such control experiments as the conditions may suggest. I have in my former communication made reference to the investigations of Bunge and Zaleski upon iron-holding nucleins. Since 1891 four other investigators have published observations on the occurrence of iron in organic compounds. jMolisch 1 endeavoured to determine the relations of iron iu the vegetable cell by means of concentrated aqueous solutions of potash. He found that when vegetable tissues were immersed in this reagent for a day or longer, they gave a reaction for iron not at all obtainable in the fresh tissues, and he explained the result as due to the removal of the iron from a firmly combined ( " maskirt " ) condition to that in which it is readily detectable by ordinary reagents. The firmly com- bined iron, as shown by this method, was sometimes in the cell wail, sometimes in the cell contents, and sometimes again in both. His results do not call for a fuller description than this, since in a later publication ^ he has stated that his solutions of potash were not free from iron, and he has con- sequently withdrawn all the conclusions which he previously based on the results obtained with this reagent. Petit,'' in investigating the occurrence of iron in barley, employed Bunge's method to separate the inorganic and albuminate from the organic iron, using for that purpose a 1 per cent, solution of hydrochloric acid in absolute alcohol. ' 'Die Pflanze iu ilireu Bcziehnngen zum Eiscn,' Jena, 1S92. - "Bemerkuiig iiber den Nacliweis von maskirterm Eisen," 'Beiichte der deutschen bot. Gescll.,' vol. xi, 1S93, p. 73. 3 "Distribution ct I'utat du fer dans I'orge," 'Comptes Rendus,' vol. cjv, p. 2iG, 1892. 178 A, Vu MArALT-UM. The dried and finely pulverised barley was put, with the acid a!eohol,in u Soxhlet extraction apparatus and heat was applied for six hours, during which time the reagent was renewed once, but the second liquid extracted no iron. The result was the same when the strength of the acid in the solution was 2-5 per cent. From his experiments he concludes that nearly all the iron is combined with nuclcin (u I'etat de nucleine) and exclusively contained in the tegmen and embryo of the barley grain. In a second publication^ he describes the separation of an iron- holding nuclcin from the malt-combs (touraillons) of barley, free from sulphur and in which the Iron amounted to 0'195 per cent. The separation was made by extracting the pul- verised matter with a 1 per cent, solution of potash at G0° C. for some minutes, and filtering off under pressure the brown liquid, which was theti neutralised with dilute hydrochloric acid. The precipitate formed was washed by decantation with water, then with alcohol and ether, and finally dried over sulphuric acid. Gilson* found iron in the nucleinic elements, not only when ammonium sulphide, according to my method of using it, was employed, but also after treatment with other reagents and in nuclei which, without such treatment, gave no rr action for iron with the ordinary methods of demonstration. He specially mentions sulphuric acid and sulphurous anhydride as giving the best results, although others, among which he includes saline solutions, produce the same effects. He is, however, inclined to regard the iron demonstrated in the nuclcin as due to a combination which is formed only after death, and similar to that which dead nuclein effects with many other substances, especially colouring matters. He showed that dead nuclein has a very strong affinity for iron compounds, the nuclei of freshly extracted cells absorbing from a 0'05 percent, solution of ferrous sulphate more iron than could be demonstrated in them when simply treated with sulphuric acid j and he maintains it is extremely difficult to ascertain whether nuclein in a living 1 " Sur une nucleine vugctale," ' Comptes Rendus,' vol. cxvi, p. 995. 1893. '■* Loc. cit. 1 I 1 IRON COMPOUNDS IN ANIMAL AND VRORTABLE OELF-S. 179 condition contains iron, or contains it only after deaths deriving it by absorption from the blood or other surrounding fluids, or even out of the reagents tliemselves, if these arc not absolntely free fr^^m iron. In his remarks upon my methods he states that liungc's lluid, upon which I relied to extract the iron of inorganic and albuminate combinations from sections of tissues, does not take away the iron artificially combined with dead nucloin even after six days. Hammarsten^ has isolated from the pancreas of the ox an iron-holding nuclco-proteid containing 1-18 per cent, of phosphorus. II. Methods or Study. In my first communication on the method of demonstrating micro-clicmieally the occurence of" masked" iron, the reaj^cnt whose use I described was called, in a general way, ammonium sulphide. This is a term that is properly appliea' '• - v tc '^e diammonium compound represented by the forr \S, but it is sometimes given to solutions which ^r ammonium hydrogen sulphide (NIIJIS), or jk ammonium, or to mixtures of diammoniuni ammonium hydrogen sulphide. At the time I was . determine which of the two latter is the most eflcctive ut. a reagent in liberating the iron from the chromatin, since either, when recently prepared, gave, with cellular elements from the same piece of tissue, reactions in which ditl'erouccs in intensity were not noticeable, and, while uncertain upon this point, I felt justified in adopting the generic term " ammonium sulphide" to designate a reagent which might be held to indicate either of the two compounds. About tvi^o years ago I gave further attention to the question whether one form uf the reagent is more ellicient than the other in this respect, and the results of a series of experiments made since then have led me to the conclusion that ammonium > " Zur Kenutniss dcr Nucleo-proteide," 'Zeit. fiir Physiol. Cliemic,' vol. xix, 1894. p. 19. I 180 A. r.. MAOAl.r.UM. liydvogcn • .ilj.liidc is more active tliati the diammonium salt, and that none of the polysulpliidcs of I'.mmonium have any action whatever on iron iu its " masked " form. These cxpen- m-nts have been controlled by others made with these reagents upon solutions of potassium ferrocyanide.i Ammonium sul- phidc, when mixed with a solution of tlic latter salt and the mix- turc kept at a teuipcrature of 30-5O''C. for one or more days, will libeiatc the iron from its combination and prcv^ipitate it as sulphide, the amount so liberated depending on the strengths of the solutions forming the raixture, on the temperature and on the time during which the reaction is all .'cd to go on. A lower temperature will sufRce when the time is prolonged. By paying due attention to all the conditions, it is possible to liberate, as sulphide, all the iron of such solutions. In tnis ammonium hydrogen sulphide is more active than diammonium sulphide, the amount of the sulphide formed being a measure of the activity of either reagent.- These experiments have, in all cases, given results which correspond with those obtained with the two sulphides upon the chromatin of isolated cells, but it was not possible in the latter case to estimate the effects as definitely. I found that of two slide preparations of isolated cells, one made with ammonium hydrogen sulphide, the other with diammonium sulphide, the former as a rule gave the 1 1 have not founJ any reference to the action of animonhini sulpliidu o:i solutions of ferrocyanides in the literature of chemistry, although, on the pre- sumption that some such reference exist?, 1 made diligent search fov it. - The results of one experiment upon this point may be mentioned. The glass.stoppered cylinder a contained 10 c.c. of a 10 per cent, solution of potassic terrocyanide and 10 c.c. of ammonium hydrogen sul|.lude made from an ammonia solution of 0'% sp. gr.. \Yhile to a similar cylinder b, with like quantities of the same solutions, 10 c.c. of dilute ammonia were added. At the end of twenty.fcur hours' stay in a warm oven with a temperature of 40" C the precipitates were filtered off with iron-froe filters, washed with water 'containing hydrogen sulphide iu solution, dissoh- ' in dilute sulphuric ac'id solutions, and, after care had been '. .1-en to redu.e all the iron to the ferrous condition, the amount of the metal in each case was estimated by titration with a standardised permanganate solution, llcsults : the precipi- tate in a contained O'OllS grm. iron, while the iroi: of the precipitate in b amounted to 00025 grm. ^ IRON COMPOUNDS IN ANIMAf AND VEGETABLE CELLS. 181 maximum reaction la about ten days, wl-.ile the latter mani- fested a reaction of moderate intensity at the end of that time, which, with a longer stay in the warm oven, did not become more marked. In the case of vegetable cells tlic leactions were more quickly obtained and the differences in intensity greater. This is illustrated in tigs. 11, 15, .'^nd IG, representing l,reparations of cells of the ovary of Erythrouium amcri- canum, in which the reagents used had been made from dilute solutions of ammonia (of OUG sp. gr.). Fig. 15 indicates the depth of the reaction with ammoruum livdrogcn sulphide at the end of twenty-four hours, the intensity attaining in mother cell in ninety hours t' degree represented in fig. IG, A\hile in fig. 1-1 is shown how far the reaction had progressed with diammonium sulphide in forty-eight hours. In the latter case the reaction did not become more marked even on the eighth day. Similar results were obtained in all the cxperi- ments of this character, demonstrating that ammonium hydrogen sulphide is more effective in liberating iron from organic combinations than is the diammoni" \ compound. In the earlier stages of the investigation the reagent was made from strong solutions of ammonia of sp. gr. 0'88; but when thus prepared it deteriorates rapidly and becomes yellow from the formation of polysulphides. Spoiled or unsuc- cessful preparations were eonseiiuently frequently obtained. Sometimes, also, difficulties were experienced in determining whether, in the preparation of the reagent, the saturation of the strong ammonia with sulphuretted liydrogeu was complete. For this reason, and also because dilute solutions of ammonium hydrogen sulphide ars less unpleasant in every way, I began to use the latter, and found that it gives results not less decided than ,hose obtained with the stronger solutions. The dilute solutions offer other advantages, for when made from pure ammonia of 0-9G sp. gr., they retain their potency for three weeks or longer, especially if kept in a bottle with a well-fitted glass stopper, and in a cool place. The smaller the amount of air in the bottle and the less frequently the stopper is removed, the longer docs the reagent retain its strength. 182 A. B. MACALLUM. During the last two years the dilute reagent has, in conse- quence of these facts, been exclusively employed. The glycerine used was chemically pure.^ It gave the best results when diluted with an equal volume of distilled water. In making the preparations, the cellular elements were teased out on the slide in a drop of the dilute glycerine, and over this, after thorough admixture with two drops of the dilute solution of ammonium hydrogen sulphide, a cover-glass of 16—33 mm. square was placed. The teasing-out process was done in each case with a clean pair of goose-quill points. Every care was taken to prevent the occurrence of impurities in the prepara- tions. The excess of the glycerine and sulphide mixture is ai- first uncovered, but if the slide be put in a warm oven with a temperature of 60° C, the mixture rapidly concentrates and in a few minutes is wholly under the cover-glass. When the solution of ammonium hydrogen sulphide is deteriorated, a deposit of sulphur forms at the edges of the cover-glass and the mixture under the latter becomes yellow through the produc- tion of polysuli.:.ides of ammonium. Such preparations never yield anything of value. On the other hand, when the fluid under the covcr-gla"- remains colourless and free sulphur does not form at the margins, the preparation, if kept at a tempera- ture of 55—60° C. for a period of from two to fifteen days is almost always successful. Sometimes at the end of one, two, or three days the mixture is further concentrated and has receded from one edge of the cover-glass. This is remedied ' Molisch ('Die I'llanze in iliren Bcziehungcii zum Eiscn,' p. 107) states that the glycerine of commerce— even tlie purest — contains traces of iron. I have not found this to be tiie case witli Price's glycerine, cpiantities of which, when mixed with ammonium hjdrof^en sulphide or diammonium sulpliidc, gave not the slightest reaction or precipitate, even after two weeks, and whenever portions of the stock supply used were evaporated at a low heat iu a platinum dish no appreciable residue was left, and not a trace of iron or lead was detected. I found that in some samples of glycerine of other manufacture the sulphide gave no immediate reaction, but at the end of a week, or later, a small precipitate, composed partly of sulphide of iron, was at the bottom of the test-tube. A similar precipitate wv. obtained in portion.^ of the stock supply of Price's glycerine only when traces of an iron salt were added, t. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 183 t. by placing at the dry side of the cover-glass a drop of a mix- ture of oue part of dilute glycerine and two of aramonii;m hydrogen sulphide, the drop so placed running under the cover, after which the preparation is replaced in the warm oven and in the end usually proves successful. I have found that when the isolated cellular elements are not very numerous and uniformly distributed under the cover-glass, evaporation rarely goes so far as to render a resort to this remedy necessary ; but when the tissues are only partially teased, and fragments tilt or elevate the cover-glass, the mixture concentrates, the pre- paration dries at one side, and the sulphide is largely converted into polysulphide. The solutions of ammonia used in the preparation of the reagents were chemically pure, and in this respect, as well as in the cleanliness of the slides and covers, I paid due regard to the suspicion that there possibly exists a ferrous sulphc-hydrate (FeS JIa), soluble to a certain extent iu solutions of ammonium hydrogen sulphide, the presence of which iu the glycerine and sulphide mixture of my preparations might, through its diffu- sion into the nuclei and precipitation therein as ferrous sulphide (FeS), give confusing results. That no such compound existed iu my reagents was shown repeatedly by allowing mixtures of tlie sulphide and glycerine to stand for weeks, when all the ammonium hydrogen sulphide was- converted into the diammo- nium salt, or into poly sulphides of ammonium, in the presence of which it would appear that the supposed existence of ferrous sulpho-hydrate is impossible. In these experiments no iron was found, nor did the mixtures in the end lose any of their trans- parency, — a result which tells against the possibility of any sucli iron compound existing in the mixtures employed upon teased- out cells. The cover- glasses and slides were cleaned in solu- tions of hydrochloric acid to remove any adherent compounds of iron, and afterwards passed through distilled water and alcohol. The bottles in which the solutions of ammonium hydrogen sulphide were kept were also, first of all, cleansed in the same way. Nothing was gained by making " stock " mixtures, in the 18-1 A. n. MACALLUM. proper proportions, of glyceriue and ammonium hydrogen sulplide, for in such the reagent is more rapidly converted into the uou-active form than when it is kept separate. Appa- rently also in " stock " mixtures the polysulphides are very rapidly formed, the fluids becoming deep yellow in twenty- four hours or less, although the sulphide used may be nearly colour- less. In summer the change of colour is rapid. That it is duo in part at least to the formation of polysulphides, appears to follow from the fact that drops of the mixture, when allowed to remain uncovered on the slide for a few minutes, quickly become milky in appearance from the precipitation of free sulphur. The mixtures retain a part of their strength during the first two or three days, after which they become useless. The tissues which were teased out for treatment were always hardened in alcohol wholly free from iron in solution. Latterly I have employed for this purpose redistilled methylated spirit. I have not used in this connection material fixed with any of the mineral hardening reagents, since the latter frequently contain iron, the presence of which in dying cells and tissues might be held to contribute, under the influence of the harden- ing leagent, to the formation of firm organic compounds of iron. Some niineral reagents, moreover — as, for example, corrosive sublimate and osmic acid— are difficult to remove from the tissues upon which they have been allowed to act, and their presence in preparations treated with p.mmonium sulphide, which forms sulphides with these metals, gives appearances obscuring, in a greater or less degree, the occur- rence of iron compounds. To facilitate the teasing-out I frequently used sections made with a clean steel knife ^ covered with absolute alcohol, the cells of such sections readily separating, and yielding sometimes a number of free nuclei. In order to determine whether iron in an inorganic or albuminate form is present, and to what ' In my eailier paper (loc. oit.) 1 pointed out that tlie knife so used gives no iron to the preparation. All my observations for the last two and a half jeais have iu uo way called iu question the correctness of this contention. .(^ inoN coiirouNDS in animal and vegetable cells. 185 extent, it was my practice to allow the section to lie in the glycerine and sulphide mixture for a few minutes before teasing it out, the iron of these forms of combination giving an imme- diate reaction on the penetration of the reagent. The removal of all iron of this description is necessary, since its presence may give confusing results in teased-out cells. For this purpose I have used Bunge's fluid, in which the sections were kept for about an hour with the reagent at a temperature of 55° C, the subsequent treatment with alcohol and ammoniara hydrogen sulphide in all cases showing that the inorganic and albuminate iron had been thereby removed.^ Sections so treated were teased out and mounted in the glycerine and sulphide mixture in the usual way. The disadvantages connected with the use of ammonium hydrogen sulphide to demonstrate the presence of "masked" iron are that it effects, in the animal cell at least, structural changes, that it is ^ot successful on large nuclei or on nuclei of large cells, and that it requires a great expenditure of time. In regard to the structural changes it is obvious that, however well hardened or well fixed cellular elements may be through the action of alcohol, ammonium hydrogen sulphide or diam- mouium sulphide must, when heat is applied, sometimes alter, to a greater or less degree, the structure of the cell, and especiallv of its nucleus. This is quite evident when we compare such pre'paratious with others in which the '' masked" iron has been liberated by the use of sulphuric acid alcohol, and sub- sequentlv treated with the sulphide. Figs. 23 and 21 illustrate thp differences obtained with the two methods, the former representing liver-cells of Necturus lateralis treated for ten days . . 55° C. with the glycerine and sulphide mixture, while the latter was drawn from a section of the samo material after it had been acted on by sulphuric acid alcohol for seven hours at 35" C, and then with the glycerine and sulphide mixture. The first difference to be noted between the > In regard to the capacity of Bunge's lluid for ex.-actinr iron of all forms of combination, see the description of the propcilie. of hydrochloric acid alcohol as given below. 18G A. C. MACAILUM. preparations represented is that of the iron reaction illustrated. This is partly due to the fact that in one preparation the ammonium hydrogen sulphide has not liberated all the iron of the chromatin, but partly also to the fact that the reagent has caused the delicate chromatin elements to become swollen, thereby rendering the iron reaction more diffuse and less marked. The effect on the cytoplasm is not less striking. It is, however, chiefly with concentrated solutions of ammonium hydrogen sulphide that preparations of animal nuclei exhibit this phenomenon. Solutions of the reagent made from am- monia of 0-96 specific gravity do not as readily produce this change, and in many cases none at all may be shown. When the reagent is fresh the reaction is quickly obtained, sometimes in two or three days, and then no swelling of the nuclear net- work occurs ; but when it is not fresh, or when it gives an odour of ammonia, the reaction is slowly obtained, and the prolonged application necessary in order to bring out this result, uided perhaps by the ammonia, causes a swelling of the chromatic elements. The slowness with which the reaction comes out is not wholly a disadvantage, for by this means one may determine whether the iron demonstrated is derived from other than inorganic or albuminate compounds. With the exception of hicmoglobiu, hieraatin, and the compound found in yolk- spherules, the organic combinations in which the iron is "masked" are aflected \pry slowly by ammonium hydrogen sulphide, and only when heat is applied ; whereas the reaction comes out at once, or after a few minutes at the longest, and without heat, in the case of inorganic and albuminate com- pounds. The distinction between these and the "masked" compounds is, therefore, very marked. In one of the excep- tions mentioned the distinction is not so clear, for when ammonium hydrogen sulphide is added to the fresh yolk of hen's egg it gives a greenish reaction at once, but when the yolk is hardened with alcohol or with heat the reagent gives this result only after several days' application at 50— G0° C. On the other hand, the yolk-spherules in Amphibia (Necturus IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 187 and Ambly stoma), whether hardened or fresh, yield the reaction in a few minutes. Such compounds are of too limited a range of distribution to affect the value of the reagent in making a distinction between the iron compounds. On hemoglobin and myo-hsematin (myo-hsemoglobin) the reagent has not the slightest action. I have kept mixtures of the reagent with solutions of haemoglobin and myo-hsemoglobin for more than a year at a temperature of 55° C, and in no case have I found that iron v^as liberated from these compounds as sulphide. I have, moreover, mounted in the glycerine and sulphide mixture on the slide finely powdered haemoglobin which had been coagulated in alcohol, and applied heat to the preparation for weeks without once obtaining the iron in an inorganic form. When, therefore, in preparations of animal tissues which have been hardened in alcohol one obtains with the glycerine and sulphide method after a time an iron reac- tion, it may reasonably be concluded that the iron so demon- strated is not derived from haemoglobin in the tissues. One may not, however, exclude hsematin as a possible source of iron, for although hEemoglobin in all forms will not yield its iron to ammonium hydrogen sulphide, the latter readily liberates the iron of heematin, and from a solution of hsematin in am- moniated alcohol or in dilute ammonia, into which hydrogen sulphide has been passed, part of the iron at ordinary tempera- tures, but the whole at 50° C, is precipitated as ferrous sulphide, in a few days.^ Even in a solution of hsematin in ammoniated alcohol, if kept for several days at the temperature of the room, I The compound formed from tlie lisematin in this process of liberating the iron is neither hrematoporphyrin nor bilirubin. With yellow nitric acid it gives a play of colours in which violet, faint red, and yellow successively appear, the mixture finally becoming colourless, and it yields an absorption spectrum like that of bilirubin. It is insoluble in ether, and soluble in chloroform and hot alcohol. The other properties of this compound are now under investigation. It has one special claim to interest in that it is formed from hcematin by a method very much less drastic in its effects than those in which strong sul- phuric acid or bromine in glacial acetic acid is used to form '-imatoporphyrin or bilirubin (Neucki and Sieber, ' Monatsh. fur Chemie,' vol. ix, p. 115, 188S). 188 A. B. MACALLUM. a part of the iron of the hjcmatin is precipitated as a greyish- white hydroxide, which, if filtered off, gives at once with amraonium sulphide the greenish-black sulphide reaction. Very weak solutions of hydrochloric and other acids effect the removal of the iron, and if solutions of hsematin in alcohol are kept for a week or more in contact with solutions of various salts (potassium chlorate and sulphate and sodium chloride and phosphate), decomposition of the hajmatin results, and iron is liberated as an inorganic compound. In all these respects htematiu behaves like the ferrocyanides, while it differs markedly from haemoglobin in the same points. Experiments show, however, how little, if any, of the iron demonstrated in animal cells is derived from hsematin. Sec- tions of the liver and other organs of Vertebrates, as well as of vegetable tissues, were placed in alcoholic solutions of hsematin for twenty-four hours, then washed in alcohol for a few minutes, and kept in a quantity of the glycerine and sulphide mixture at a temperature of 35° C. for twenty-four hours. At the end of the latter interval all the sections were blackened, and under the microscope the nuclei were dark green from the ferrous sulphide liberated from the haematiu absorbed by the chromatin. In order to get this result the sections do not require to be teased out at all. The rapidity with which such a strong reaction is obtained indicates that in ordinary teased- out cells mounted on the slide in the glycerine and sulphide mixture, the deep reactions obtained after several days or after a week are due to a decomposition, not of hcematin, but of some other compound or compounds. Ammonium hydrogen sulphide, then, may be regarded as a reagent of very great value in the investigation of " masked " compounds of iron, and it must constitute a final test for this purpose, whenever the accuracy of the other reagents, used also for determining the distribution of assimilated iron com- pounds in cells, is called in question. In June, 1891, Mr. F. 11. Bensley, while carrying on under my direction a research on the distribution of iron ir the ovary of Erythronium americanum, as demonstrated by i I IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 16^ the employment of ammonium sulphide, succeeded in obtain- ing some interesting results which necessitated control experi- ments based on the removal of all traces of inorganic com- pounds of iron from the tissues under investigation. For this purpose Bunge's fluid was used, and it was thought that hardened specimens of the ovary, when subjected to its action for a time, would not give, on the addition of ammonium sulphide, any immediate reaction for iron, and that further treatment with the latter reagent in a warm oven for several days would show the presence of iron in the nuclei of their cells, and possibly also in their cytoplasm. Much to our surprise, however, the treatment of the ovary of Erythronium with a quantity of Bunge's fluid for two weeks at 20° C, and the subsequent application of ammonium sulphide, resulted in the production of a marked reaction for iron, which under the microscope was found confined to th ; nuclei. I was at first inclined to believe that the iron so shown was due to diffusion into the nuclei of that present in an inorganic form in the tissues, and this would appear to be Gilsou's view; but repeated experiments have demonstrated the incorrectness of this explanation, and that Bunge's fluid liberates the iron of organic compounds.^ Experiments were also made on animal tissues and similar results were obtained. The liberation of the iron is to be attributed to the hydrochloric acid, the only active part of the reagent. This conclusion suggested a number of experiments, all based on the principle that whatever proper- > Gilson's statemeut is difficult to interpret. He dots net say wlietber he applied the reagent to sections of tissues or to the latter in mass, and at v?hat temperature it was allowed to act. lie appears to regard tlie iron absorbed by dead nuclein as combined with the latter, and he remarks, in reference to my statement that Bunge's fluid removes all inorganic and albuminate iron from sections after treatment with it for ten hours : " but I have observed that Bunge's liquid does not take away the iron artificially combined with dead nuclein after six days." I can explain his statement only on the sup- position that he used the reagent on the tissues in mass, and that he thereby obtained the same results that I did under similar circumstances; in other words, the iron "artificially combined with dead nuclein" was in reality iron liberated by Bunge's fluid from its masked condition in the chromatin and re- tained by the latter. 100 A. B. MACALLUM. ties hydrochloric acid may have in this respect are possessed ma greater or less degree, by other mineral acids, w'hethe Tn dilute aqueous solutions or in alcohol, and the results were of such a character as to induce me to employ these reagents on al species of edls in which the distribution of iron had be " dotcrmined with ammonium hydrogen sulphide The more serviceable of these were found to be sulphuric acd and n.tno acid dissolved in alcohol of 93 per cent strength. The former was prepared by adding four volumes of the strong acid to one hundred of alcohol, ;hile the late contained three volumes of the acid (of M, sp. gr ) in one hundred of alcohol. »l-bi-;mone The chemicals used in the preparation of these reagents were ree from traces of iron, and care was taken to hafe a bottles and vessels used to hold them also free from adheren iron compounds. It was, of course, impossible to pro de agains tl^ iron in the glass, but I am not certain that reagents derived any from this source, even in infinite in.^ quantities. The alcohol used contained not a trace o iron During the last eighteen months re-distilled met " ed spS was found to be in every way as serviceable as the* p Z alcohol used earlier in the investigation ^ Thealcohol of these reagents largely preverts the occurrence of digestive changes which the acids effect when, in aqueous olutions, t ey are allowed to act on tissues for severa 1;/ and e pecially at a slightly elevated temperature. Anothe; raSortiTiV' Ti^''^'''' '' '^ p--^^ ^ ^-- J trom one pa t of the tissue into another, from nucleus to cell or from cell to nucleus. Acid alcohols dissolve iron salts solutions of the acids. For evnmnlo fo„„ 11 i«o,.ue in absolute a,oo„„, ij'::'^^:^;:^:;.^ but , ,, soluble i„ these when they contain a slTaZuy dt,lled water or ,n dilute aqueous solutions of snlphu, acKl. The smaller the proportion o,' the aeid in the a eol I ">0« COMPOUNDS W AN,MAL AKD VEOETA.LE C.LLS Wl the liberation of the Xfrom \Z°" °" ° '""'' '" »"'' '•"• the neutralisation o T ho 1 "l - ' -'■"■•'ition entail, result of this neutr" , ati„„ h' ""' f '"'""' l'™'''^'- ^s a a-d wouid pasrbi ;t2 ti::rtt\T°T '-'- ^-""^ happcuing is minimise,! „r 1,;, ''°"8" "^ "'** ropert, ;,,ie„ th:'™.'::! :^s"::lz:t "' f" IS set free in itself hv fl,„ "^^ oye';ainnig the iron that speeiall.in the e^o of 'ni Tel tZ , "f '' ^"°~- of vegetable tissues are allowed to lie fo,? "i "•' ''""'"" quantity of the reagent, the™ I tt t o d oTttt "t" " '"^° as tntense a reaction for iron ,. ,i '?™'' ""'"t time give "ays. The result is e to t tt t:'t,"", '""■'" '" »hieh the ehromatiu holds the r„„ lil . , "'°.""'a<^'ty ,vitl, the extraetivo eapaeitv of the te";,,! ''"""' '" " °°""'"-" -pect The two former, whe.I.e . ^ a„ „s" TrT tissues m mass or on sections of the s,n,o I,.!v! ^ ''" "hole, in the parts in wl.ieh it is lil Z^' ,"■""' "" "'" of tissues aro'trcated »ith l^.l^l raeLut: ^VYr*"' IS cxtraetcd as nuicldv ns J u n , a'™''ol, the iron sueh preparations "trelnJi r'' """ ""*'^'""=""y a feeble" reaction for'Tt rr:i"™Ti;'i'"'"° ^"= ■hsfnetly seen „ben the temperature si^edd' if T' ■■cageutis allowed to act for two n,. tl '""■'""' 'f the conditions, no iron, organic „ i i g i "L^; ^^^ """^ "-= tioiis. When ti.p ti««np. • ' ^" *''^ pJ'epara- quantity o^lJ^^T;:^!^:: °" "^ ""'''' '-''' ^^'^ neufalised and extIC ^^ e'::!^::"? '' ^7'''' result that teased-out portions of uc. 7' '' "'''' '^'' reaction for iron limiLl ''' ^'^^ ^ "^''^'•J^^d or non, limited, as in preparations obtained with l.)'drochloric acid. ° ' ''" '"^^"""-'^ ""^ " ^^ per cent, solution of VOL. .38, I'AKT 2.— KEW SEIJ. N 192 A. 11. MACALLUM. the Other acid alcohols, to the parts iti which treatment with the glyccriuc and sulphide mixture demonstrates its occurrence. In describing the piojierties of liydrochloric acid alcohol, Buuge expressly states that while it extracts inorganic iro:i it docs not remove the iron from the nuclcin (lucmatogcn) of cgg-yolk.i This is not quite correct, for when hard-boiled yolR is treated with ammonium sulphidr; it gives only a feeble reac- tion for iron, even when kept for twenty-four hours at an elevated temperature ; but when it has been acted on by a quantity of Bunge's fluid for a day at 30—35° C, the applica- tion of ammonium sulphide, after all traces of the acid have been removed with alcohol, gives an immediate and marked reaction for iron. The iron under such conditions must be in the form of rhloride, and as an inorganic compound it should be extracted by the reagent, if Bunge's views con- cerning the properties of the latter be correct, but this happens only when the quantity of the yolk so treated is very small, and then the whole of the iron is removed in a few days, this fact demonstrating clearly that the reagent in its action makes no distinction between inorganic and organic iron. The latter is in its liberation from tiie " masked " con- dition converted into the inorganic form, and it depends on the quantity of yolk used whetlicr or not the extraction may keep pace with this conversion. If the quantity is large, the libera- tion of the iron from its organic combination eutuils a d'minu- tion of the acidity of the reagent, and at length the extraction of the liberated iron ceases. It commences again only when a fresh quantity of the reagent is substituted for the exhausted fluid. The results of its action upon the iroji-containing nucloo- albumin of yolk are therefore practically similar to those which it gives when applied to animal and vegetable tissues. The fact that a considerable diminution of the acidity of hydrociiioric acid alcohol allows the liberated iron to be retained ' " Ueber die Assimilatiou des Eiscus," ' Zeit fur Plivbiol. Cliemie ' vol ix, 1885, p, 49. • IKON OOMI'OUNDS IN ANIMAL AND VKCHTAIUJ.: CELLS. 193 in its original position in the cell has led mc to try the cfTects of solutions in which the strength of the acid was less than 1 per cent./ and they have been found, when used upon thin sections of tissues, to give very successful preparations, pirmit- tnig the iron liberated to be demonstrated as fully as after the enii^)loyincnt of cither sulphuric or nitric acid alcohol. The time during which these reagents must be allowed to act on a piece of tissue varies. I prefer to give general state- ments on this point, because specific directions arc impossible m a case where the size of the object, the quantity of the reagent, and the temperature constitute the conditions. Bunge's fluid extracts as readily as it liberates the iron in thm sections of tissue, but when the latter is in mass the reagent requires a length of time which mav vary from a week to two months, all depending on the size of the object and on the^tempcrature, which in summer may be that of the room (20°— 2'J" C), but in the colder seasons that of the warm oven (35° C). Sulphuric acid alcohol acts more slowlv, and con- sequently requires a longer time for liberating the iron in unsectioned objects, while in sections its action is complete in from one to four days, this depending also on the temperature, the most favourable being 35° C. A longer stay than is just sufficient to liberate all the organic iron results in removing from the sections some of the iron set free, the more being extracted the longer the sections lie in the reagent. When examples of the Protozoa and Protophyta were subjected to ' Tlieac diirerenccs in extractive capacity cxiiibited l)y veak and stroii- alcolioiic solutions of l.jdrociiloric acid iiave apparently not been noted by Petit (loc. cit.), xvlio used the diluted rea-ent in a Soxiiiet apparatus to remove the inorganic iron conipouiuls from barley. As the boilin-.poiut of hydrochloric acid is higher than that of alcohol, it is obvious that liltle of the former must pass from the 1 per cent, solution at the bottom of the flask as vapour to condense above and net on the substance whose iron is to be ex- tracted, while the alcohol is readily converted into vapour; in other word., the reagent in the upper part of liie apparatus must be nmcli more dilute than that in tiie flask below, and conscrpiently its extractive power must be very feeble. This method is, therefore, open to the objection that it docs not ensure the removal of inorganic iron compounds. im A. n. .MAOAI,l,UM. the notion of the acid alcohol tl-c full effect was obtained at the end of twenty-four hours at the latest, when the tempera- ture was 35" C. With nitric ueid alcohol the liberation of the organic iron was rapid, sections of ve-etablc tissue (Ery- thronium and Iris) giving, after a stay of ten hours in the reagent at 35° C, an intense reaction with the acid ferrocyanidc mixture. At a lower temperature the result was les^ marked, but the reaction was deeper than that obtained with secHons treated with sulphuric acid alcohol for the same length of time and at the same temperature. The process of liberation was usually completed in about thirty-8i>; hours. So little does nitric acid alcohol extract the iron it liberates that in sections of the ovary of Erythronium americanum kept for six weeks in it I found little diminulion in that intensity of the iron reaction which sections, placed In the same fluid at the eoramencement with the others, g.ave at the end of two days. With sections of animal tissue the intensity of the reaction was less marked with the prolonged stay in the reagent, which, after four or five days' action, sligh'tly alters the cellular structures. When nitric acid alcohol is allowed to act on a section for a longer time than is necessary to set free all its organic iron, diflusion of the iron sails thus forned IS ai)t to occur, especially in vegetable preparations, the cvto. plusni giving in such cases a reaction for iron. That the iron demonstrated after the use of acid alcohols is derived from organic compounds I have shown by numerous experiments. I have founa that when thin sections of animal or vegetable tissue are covered with a lar-e quantity of Bun-e's fluid and kept for three d.,ys ..t 35° C. or higher,' the teamed- out cells give no iron reaction whr„ mounte.; ^^itll glycerine and ammonium hydrogen .ulphide on the slide, even after two weeks and at GO- C Furthermore, sections so treated with IJunges fluid, when subsequently subjected to the action of sulphuric acid alcohol or of nitric acid alcohol, vield no iron re action whatever. Bunge's fluid, therefore, extract's the iron which the prolonged application of ammonium hydrogen sulphide and glycerine at un elevated temperature liberates and demon- IRON COMPOUNDS IN ANIMAfi AND VEOETAnr.K VF.U.S. 195 atratc8, and with this removal disappears the iroa demonstrable after treatment with either of the other acid alcohols. Tl -s shews that the iron in such cases cannot be derived from the reagent nor from the glass 0." the vessel used, and this is em- phasised by the rcsnlts of other experiment^.. 1 extracted with iJungc's finid all the iron from .1 series of sections of an ovary of Erythroninm, and then subjected the«e to the action ufa large^qnantity of suli)hnric acid alcohol for twenty^four hours at 35° C. These gave no iron reaction, while others did so which had not been t.-eated with 15ungc's fluid, and which were put in the acid alcohol at the same time. That the absence of nu iron reaction was not due to a lack of absorptive capacity on the part of the section, brought about by Bunge's fluid, was proved uhcn sucli sections were allowed to stay in sulphuric acid alcohol containing a little ferric salt in solution' for half an hour. The reaction obtained was marked, and almost wholly confined to the nuclei. These experiments were repeated again and again with sections o^ animal and vegetable tissues, and the results were always the same, proving that the iron demonstrable after acid alcohol has been used on tissues is derived from the latter, and not from the reagent or the vessel used. Tiiese experiments indicate, however, how necessary it 18, in investigating the distribution of iron in tissues, that the reagents should be absolutely free from iron, and that, in sections of tissues containing iron in an inorganic or albuminate form, there is danger, when either sulphuric acid alcohol or nitric acid alcohol is used upon them, of its redistribution, and especially of its deposition in those parts of the cell which absorb various compounds readily. In order to guard against this, I found it advisable to steep the sections in a quantity of Bunge's fluid for a time which varied with the temperat.ue at which the reagent was applied, as, for example, for one to two hours at 50°-G0° C, but for eight to ten hours at 35° C. ' This solution was made in the following way. A quantity of sulphuric acid alcohol was allowed to act on ferric oxide in powder for about a week wlien a portion passed into solution as a ferric salt. Of this solution 1 c c' was taken and added to 10 c.c. of pure sulplmric acid alcohol. 196 A. B, MAOALLUM. '/ I Bunge's fluid extracts very little or no iron from sections when the temperature is below 20° C, but at the higher temperatures stated the extraction is complete at the end of the intervals mentioned, and with a longer action more or less of the ''masked^' iron Is liberated and removed. When a tissue- as, for example, that of the spleen in some animals-contains an excess of iron in an inorganic form, the time of extraction must be prolonged, and the extracting fluid large in quantity After the inorganic and albuminate iron has been thus removed from a section-a result which may be demonstrated by treatment of the preparation with ammonium sulphide —it may be subjected to the action of either of the two other acid alcohols to liberate that portion of the ''masked" iron as vet unafltected. '' The acid alcohols do not readily attack and liberate the iron of haemoglobin and haematin except at a high temperature. Ot tins fact I have convinced myself by numerous experiments on heemoglobin, whether prepared from alcohol material or from that coagulated by heat. A quantity of it, in a powdered form, put into a flask and covered with a quantity of Bun-e^s fluid, :vas heated for twenty minutes, and the fluid then, after filtration through a filter free from iron, was neutralised and treated with ammonium sulphide. The mixture ga-e no imme- diate evidence of the presence of iron, but when the test-tube containing It was put aside for twenty-four hours, a dark-reen sediment made its appearance, and this was shown to be sul phidc of iron when it was separated on an iron-free filter and treated with a quantity of an acid ferr/cvanide fixture This iron was, in great part, derived from the huimoglobia and hjematin, as well as from organic combinations present in the leucocytes and plasma, and but little had its source in the in- organic and albuminate compounds of the same, a fact shown by further experiments on the powder which had once been acted upon by boiling Bunge's fluid. The extract made with a fresh quantity of the reagent gave, on neutralisation and on the addition of ammonium sulphide, the same evidence of the presence of iron that was obtained in the first experiment A IltON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 197 third, fourth, aud fifth extraction resulted in the same way. When, on the other hand, a quantity of crystallised h^rao- globin was acted upon by the reagent for forty-eight hours at 35 C, the filtered fluid, tested for iron in the manner described, gave a scarcely appreciable evidence of the presence of the metal. The iron, therefore, which is found in animal tissues after the use of Bunge's fluid at either 35° or 50° C. for short intervals cannot very well be supposed to be derived, in any appreciable quantity, from the hemoglobin in them, and as ammonium hydrogen sulphide does not affect the iron of the pigment, yet reveals the iron of " masked " combinations of an apparently less firm character, it follows that weak solutions of hydrochloric acid at slightly raised temperatures must attack such combinations more readily than it afl'ects hemoglobin. This was most cleaily shown by results of experiments on haemoglobin and chromatin with a quantity of Bunge's fluid for twenty.four hours at 35° C. When hemoglobin alone is thus treated, neither the powder nor the extract gives any appreciable indication of free iron, but the latter is readily demonstrable in chromatin, or in mixtures of chromatin and hemoglobin, after similar treatment. Since the iron in hemoglobin is not affected to any perceptible degree by treat- ment with the reagent for twenty-four hours at 35° C, one may postulate that it is as little aff-ected by treatment with either of the other two acid alcohols at the same temperature, and experiments with these have given results which bear out this conclusion. The substance chlorophyll, the relations of which to iron, though generally recognised, have not been definitely deter- mined, is, as is well known, an abundant constituent of the cells in many vegetable forms, aud, therefore, a brief discussion of the possibility that this substance is the source of the iron demoiistiated in vegetable cells, is necessarj. Some of the more recent investigators of this substance have made conflicting statements on the question of the presenc: f iron in the molecule. Adolph Hansen' found it to contain ' 'Die Farbstoffe dcs Cliloropliylls,' Darmstadt, 1S89, p. 58. 198 A. C. MACAI,LUM. iron, while Emich, at e request of Moliscli/ examined a quantity of pure chloropiiyll and found it free from iron. ]Molisch also made observations on tlie subject, and determined that, after every care was talcen to prevent contamination with iron salts through impure extracting fluids, the ash of chlorophyll gave not the slightest reaction for the metal. Gautier^ also claims that it docs not contain iron. Sclmnk,^ on the other hand, found ferric oxide in the ash of phylloxanthin, one of the decomposition products of chlorophyll, even after that com- pound had been treated with acids and after repeated solution of it in ether. The material from chlorophyll-holding organisms was, in all cases, thoroughly freed from that substance before the disposi- tion of the iron in it was examined. Chorophyll, however, has not in any of my preparations yielded any evidence that it contains iron, nor does its presence or absence at all affect the question of the occurrence of iron in other compounds in the cell. This is very distinctly shown when one compares the results, obtained from experiments on vegetable cells holding chlorophyll, with those determining the distribution of iron in Fungi and in Monotropa uniflora and Corallorhiza mul- tiflora, which are destitute of chlorophyll. In the two latter the disposition of the assimilated iron is as it is in the chloro- phyll-holding Phanerogamous plants, and consequently one may dismiss the objection that the pigment constitutes the source of the iron demonstrated by my methods in the nuclei of vegetable cells. It may be proved also from Monotropa- ' Op. cit., p. 87. 2 'Cheniie Biologiquc,' Paris, 1892, p. 20. 3 "Contributions to the Chemistry of Chlorophyll," No. 4, 'Proceedings Roy. Soc.,' vol. 1, 1891, p. 302. ■• The importance of Monotropa naterial for control purposes renders a short description of the methods of preparation employed upon it necessary. This plant, when hardened in alcohol, blackens more or less through the pro- duction on the part of the dying cells of a dark greenish-blue pigment, but it remains colourless when fixed in solutions of corrosive sublimate, a reagent whose use is, for reasons already mentioned, objectionable when ammonium sulphide is lo be employed. To obtain material on which this reagent may be IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 199 that none of the iron found in the nuclei is derived from the cytoplasm, for there is very little and often no cytoplasm in the cells of the coats of the ovules in this plant, and yet the nuclei of these give as intense a reaction as those of the ovary of Erythronium, Iris, Hyacinthus, or any form in which the cytoplasm is abundant. In order to get the best results with the use of the acid alcohols, I have found that the tissues must be well hardened. If the tissues are fresh or imperfectly hardened, the application of acid alcohols for a time sets free the organic iron, but the structure of the cellular elements is more or less changed in such cases by the acids — a change not at all found to occur when the tissues have been carefully hardened. Strong alcohol (90 — 95 per cent.) was used for this purpose, and it was found to present, over the other hardening reagents, a number of advantages. It can by redistillation be made free from iron, and when it is of absolute strength it neither extracts any of the iron compounds (hsematins excepted) from tissues, nor allows these to diffuse. There is the important point also, that tissues fixed with it can be subjected to all the reactions for iron, without incurring the risk of coi" plications due to the deposition of iron or other metallic salts, which occur when other hardening reagents are used. In this way one may treat pieces of a tissue with ammonium hydrogen sulphide and with the acid alcohols, and thus alio w the methods to control each other. When, on the other hand, it was not necessary to use ammo- nium hydrogen sulphide on the tissues, other hardening reagents were employed, but only such as did not, by their presence in the tissues, interfere with or obscure the demonstration of the iron. Saturated solutions of corrosive sublimate and ^ per used advantageously, parts of the fresh plant are thrown into boiling distilled water, and those which remain uncoloured at the end of ten minutes are further hardened for several days in absolute alcohol. I have often treated material so prepared with the warm glycerine and sulphide mixture for from four to ten days, and then with an acid ferrj^cyanide solution converted the ferrous sulphide demonstrated into Prussian blue. Such preparations are pro- bably the most instructive obtamable in regard to the question of the relation of iron to the vegetable cell. '/ i 200 A. B. MACALLUM. cent, solutions of osmic acid were found serviceable^ the latter reagent also having been used in the conabination known as rieraming's fluid. The corrosive sublimate solution was allowed to act on the preparations of tissue for about ten minutes, after which they were washed for a few minutes in distilled water, and then in 50 per cent, alcohol. The hardening was completed with alcohol of 70 and 90 per cent, strengths in the usual way. When rieraming's fluid was used the tissue was not allowed to lie in it for mure than half an hour, while for the osmic acid solution not more than ten minutes were given, and the fixa- tion was carried on further with alcohol of 50, 70, and 95 per cent, strengths. Preparations, whether made with corrosive sublimate or with osmic acid solutions, retain, even after careful washing, traces of the metallic salt of the reagent used, and the black or dark reaction which they give with ammo- nium hydrogen sulphide, in consequence of the presence of such metals, interferes with the proper demonstration of the di5tnbi.iion of iron by that reagent. On this material the acid alcohols only were used, and the preparations were subsequently treated with the acid ferrocyanide mixture, the Prussian blue reaction obtained not having been in the least affected by the presence of minute quantities of the metallic salts of the har- dening reagents. The latter were free from iron salts, a fact of which I convinced myself by qualitative analyses. To the use of all hardening reagents other than alcohol there are objections. Those which contain an acid may assist in the diff'usion of iron salts in the tissues, and cause tlie deposition of these in some other parts than those in which they originally were held. Farther, the acids of some of the reagents (e. g. acetic acid in Flemraing's fluid) may liberate the organic iron, which cannot in such a case be distinguished from the iron of inorganic or albuminate combinations. For these reasons I have used acid hardening reagents but occasionally, and then the time allowed for their action was short, in order to reduce to a minimum the risk of liberating organic iron, and of the diffusion of iron salts through the tissues. Against corrosive iucleolar body coloured light blue. It is chiufly in the nucui of the glandular cells that one finds these nucleolar bodies, and they are most distinctly seen iu large nuclei, as, for example, those of hepatic and renal cells and of the intestinal epithelium of Nectui'us lateralis. They are very rarely seen in the uuclei of the muscle fibre and in those of the cutaneous epithelium of the same animal, while they are never present iu those of leucocytes or lymph cells, or in those of the red blood-corpuscles. In the search for them in all these elements the greatest assistance is obtained from the employment of eosin, which, iu sections exhibiting the Prussian blue reaction, gives these bodies an ochre-red colour, while the parts showing a dark-blue reaction are unstained by it (fig. 47). In the nucleus of the glandular cell which is passing into the mitotic phase, the nucleolar body disappears, apparently by solution into the chroraatiu threads, for in the nucleus of a renal cell, iu which the meridional disposition of the chromatin filaments obtained preparatory to the formation of the loops, I saw, attached to one of the fila- ments and partly embraced by its substance, what appeared to be the remains of such a body. In later stages of mitosis not the slightest evidence of this body or of its remains can be observed. Whether the iron which these bodies contain is that of a small quantity of chromatin dissolved in them, I am unable to say. The fact that they take sometimes a very feeble stain with haimatoxylin, seems to indicate that they may contain a small amount of chromatin. The iron in them is held neither more nor less firmly than in the typical chromatin elements, since iu hepatic uuclei containing them, prolonged treatment with ammonium hydrogen sul|)hide iu the warm oven does not result in demonstrating any difference, except in the amount of iron in the one and iu the other. The substance which holds the iron does uot possess the slightest affinity for safranin, but attracts eosin as no other cellular con- stituent does, and in these properties, as well as in the VOL. 38, TART 2. — NEW SEUIES. 208 A. 1!. MACALLUM. very small amount of iron present, there vfould appear to be distinctions which separate it from chromatin. My prepara- tions were chiefly obtained from the organs of the fasting animal, and as I did not succeed in my attempts at feeding artificially some examples of Nee turns that I had, it is not possible for me to say whether the constitution of the nucleo- lar bodies is always similar to, or ever different from that described; but in preparations of the liver and other organs of specimens of Amblystoma pu net at urn killed soon after their capture, or after artificial feeding, the nucleolar bodies appeared to present the characters noted in the cells of the fasting animal, the smaller size of the elements in this case, however, not allowing as clear a view of them as was desired. In the nuclei of the liver-cells of Necturus, as illustrated in preparations made after the manner described, I frequently found a third clement, whose significance is unknown to me. It manifested itself by the red stain which eosiu gave it, the nucleolar bodies taking, in contrast, an ochre-red stain. It had no constant shape or form, in some cases being of a lila- xnentous character, in others resembling a localised granular deposit (fig. 17) ; and when the structures were filamentous, several usually appeared lU the same nucleus. Ti'e substance forming them did not contain the slightest trace of iron, and therefore appeared to have no relation to the nucleolar bodies or to the chromatin. I have not in any other organ observed similar structures. The disposition of the iron-holding compound in the nuclei of Amphibian ova deserves special mention. In the ovarian ova, whose nuclei contain no peripheral nucleoli, the ivou is distributed as represented in fig. 30, the chromatin in this case forming a fine reticulum, in the trabecula; of which large granules arc found with lateral prolongations. The iron demonstrF.ted in this preparation Mas set free by sulphuric acid alcohol, but a disposition of iron in the main like this may be found in similar nuclei when the latter arCj on removal from the ova, broken into small pieces on the slide in the glycerine and sulphide mixture, and, thus prepared and provided IRON COMPOUNDS IN ANIMAL AND VEOETAnLE CELLS. 200 with a cover-glass, kept for days in a warm oven. This method must he resorted to in order to get tlie iron reaction, since other- wise the large nuclei may he kept for a month in contact with the reagent in the warm oven without resulting in demon- strating, in the slightest, any iron reaction. In the peripheral nucleoli, when they are present, the amount of the iron, as indicated hy the depth of the reaction, is great, but in the remaining elements of such nuclei it is small. Wlien such preparations are examined with a strongly magnifying objective, the chromatin network, as revealed by the iron reaction giveu, is found to be less distinct, and instead of granules of iron- holding substance arranged at dcRuite positions along the course of the fibrilla;, as in ova much less developed, the iron is now seen to be chiefly confined to beadlets, few in number, sometimes regularly, sometimes irregularly, disposed on the fibrilla;, which, in ammonium hydrogen sulphide preparations, manifest but a feeble greenish tint. There is an inverse rela- tion between the size of the nucleoli and the amount of the chromatin in the network, and an examination of some nuclei in which the formation of the peripheral nucleoli has commenced, and of those in which the development of these bodies is much more advanced, irresistibly suggests that the latter are derived from the chromatin of the network. I have elsewhere^ pointed out that the solution of the substance of which these nucleoli are composed and its diffusion from the nucleus into the cyto- plasm of the ovum rrc connected with the formation of the yolk-sperules in Amphibia. That a solution of the ivriplicral nucleoli takes place has been noted by O. Schultze- a..n Born.'^ Sehultze found that with the solution of the nucleoli (keim- kiirpcrchen) the contents of the nucleus and the substance sur- rounding the latter were affected in the same way by reagents and staininjj- fluids, and he believed that the dissolved sub- ' 'Transactions of tlic Caniuliau Institute,' Toronto, vol. i, pari 2. - '' Unlcrsucliungcn iiber die Itt'fuug uud Bufruclituug dcs Ampiiibiuneius," 'Zeit. fiir wiss. Zool.,' vol xlv, p. 177. ' "Die Struktur dcs Keimbliisclicns im Ovariulei vou Triton tacuiatus," ' Arcli. fiir Mikr. Anat.,' vol, xiiii, p. 1. 210 A. B. MACALLUM. I stance diffused from the nucleus of the ovum into the cell- body. Born ob orved that the nucleoli are always placed as closely as possible to the cell protoplasm, while the chromatin in the development of the nucleus and ovum becomes so finely divided in the karyoplasm that it is stainable with great diffi- culty, and it is as difficult to demonstrate optically, a condition which continues till the formation or deposition of the yolk- grains (Dotterkcirner) commences. In later stages the per- sisting peripheral nucleoli lose their capacity for absorbing colouring matters. In support of these observations of Schultze and Born, I may but add that the iron in the cytoplasm of the ovum makes its appearance only after the solution of the peripheral nucleoli commences. The substance forming the peripheral nucleoli does not react with staining reagents as does the chromatin of the nuclear network, and especially with the indigo-carmine staining fluid of Shakespeare and Norris the resulting stains of each are different, the chromatin of the network being coloured red while the nucleoli are stained blue or green, the latter colour obtaining also in the yolk-spherules of such pre- parations. A further t!'fferencc is noticeable iu the effect that ammonium hydrogen sulphide exercises when applied to these structures for some time at an elevated temperature. In this case the iron of the peripheral nucleoli reacts more readily than that of the chromatin of the network, but less readily than that of the yolk-spherules, which in the ova of Necturus and Amblystoma give a green reaction in a few minutes with the reagent. It would appear as if the iron compound undergoes a change in its transference from the nucleus to the cytoplasm. The peripheral nucleoli appear to be formed at the nodal points of the chromatin network, if one may judge from pre- parations of which fig. 31 is ai illustration; but there is a possibility that these represent f patliological condition, since they are not common in tlic ovary when, if they were normal, they should be present iu larger numbers. I have, moreover, found that they were accompanied by examples of IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 211 I another condition which I regard as pathological. In the latter the nuclei were indistinct or disintegrated, their chro- matin had disappeared, and the surrounding connective tissue, ■with its blood-vessels and their red corpuscles even, gave in a few minutes, with warm ammonium hydrogen sulphide, an iron reaction, frequently so deep as to obscure largely the details, while the tissues, a little further away from such examples, and other ova under exactly the same conditions of treatment with the reagent, gave no such reaction. It may possibly be that the chromatin of such disintegrated ova furnished the iron observed thus diffused in the connective tissue and blood-vessels. In the nuclei of all the higher vegetable organisms the assimilated iron compounds are, on the whole, distributed as they are in the nuclei of the more highly developed animal forms, a fact which may be demonstrated in any Phanero- gamous plant, especially readily if its nuclei are large, as is the case in Erythrouium americanum. In many of the preparations of the latter form tlie chromatin filaments were, in the process of teasing-out, partially or almost wholly set free from the nuclei containing them, and to the parts thus set free, as well as to the remainder, the glycerine and sulphide mixture always gave a distinct reaction for iron in a few days (fig. 17). Mitotic figures in such preparations appeared very sharply defined through the iron thus revealed in the chromatin elements. In successful preparations made by this method the reaction for iron is very marked, as much so as in those made with sulphuric acid alcohol ; and in this respect there is a contrast between animal and vegetable nuclei, for in the former the glycerine and sulphide mixture brings out, after a longer application and less frequently, a reaction as intense as that which may be obtained after treatment of the nuclei with acid alcohol. Of nucleoli and nucleolar bodies there are at least three kinds. The reaction for iron given in one variety by the glycerine and sulphide mixture was weak, and it was obtained at the same time that it appeared in the chromatin network or filaments. These arc smaller, apparently, in the hardened 212 A. r.. MACALLU]\I. than Ihey are in the living cell, for, as a rule, they only partially occupy the cavity in which they lie (figs. 17 and 19), I have in some cases isolated them from their nuclei in the glycerine and sulphide mixture, and the greenish reaction which they gave could, therefore, not have heen due to the iron of a compound which diffused from the chromatin elements into them. When the sections were treated with sulphuric acid alcohol and subsequently with the acid ferro- cyanide mixture and the eosin solution, the result was usually that of which fig. 42 is a representation. These nucleoli stain intensely with eosin, which also colours very slightly the chro- matin network, the blue of the latter thereby becoming violet, but after being thoroughly washed in alcohol the bright blue colour returns ; while tlr i treatment makes no difference in the intensity of the stain in the nucleolar bodies. These effects are most distinctly observed when the diaphragm of the Abbe condenser is removed from the field, in which case it is possible to see the most minute of the nucleolar elements, a device that is necessary when the nuclei of ordinary paren- chyma cells are under examination. In the second class are those nucleolar Elements which may be found in the cells of the nuccllus, and which are composed of chromatin, sinco they give a deep iron reaction after the employment of any of the methods of treatment for liberating the element, and since, also, they stain in ever;- respect like the chromatin threads. They usually occupy cavities in the nuclei like those which contain the eosinophilous nuclei last described. o€/ I regard these as reserve masses of chromatin deposited in the nuclei engaged in the formation of chromatin, which eventu- ally is transferred to the cells of the endosperm. T( this subject I propose to refer again. Nucleoli of the third class are to be found in the nuclei of the embryo-sac (fig. li, a and b). They are not present in the mitotic nucleus, but in the retrogressive stage they appear on the course of the filaments as spherical elements enclosiug one or more refracting corpuscles and containing but a small amount of iron, which, however, in later stages, when the fila- I I / ITION COMrOUNDS IN ANIMAL AND VEGETABLE CELLS. 213 ments became thinner and les. rich in chromatin, is more abundant. These nucleoli are eventually formed chiefly of chromatin, and in stained preparations appear to contain nearly all the chromatin of the nucleus. When mitosis again com- mences the filament forms at their expense, the increase in size of the filament keeping pace, apparently, with the decrease in the quantity of chromatin which the nucleoli contain. Finally, before their disappearance, when they contain buL a minimal quantity of iron, they take the eosin stain deeply. All these forms of nucleoli take up safranin from solutions as readily as do the chromatin elements in the same nuclei, and they hold the stain as tenaciously when they are washed with alcohol. They are in this respect different from the eosino- philous nucleoli in the animal cell, which appear to be unrepre- sented in the vegetable cell. Of an exceptional character are the nucleoli in Corallo- rhiza multiflora and in Spirogyra. In these the greater portion of the chromatin in ea h nucleus forms a single large spherical element unconnected with the chromatin network, which after prolonged treatment with the glycerine and sul- phiae mixture, gives a pronounced reaction for iron. I have, on a few occasions only, in preparations illustrating the iron reaction, seen the chromatin localised at points along the course of the filament, and concluded that this was not due to faulty methods of manipulation, for hsematoxylin and other dyes just as infrequently render such a distribution visible. It was also, with the aid of the acid alcohols, found that in the loops of the mitotic nucleus of the embryo-sac the chro- matin is disposed under the membrane enclosing the filament, in such a way as to make the latter appear as a tube of chromatin. In some of the elongated oval nuclei of the nucellus and of the fibro-vascular bundle of the ovule, Mr. Bensley has observed a point of some interest. This consists in the occur- rence in the karyoplasm, amongst the trabeculse of the chromatin network in one end of such a nucleus, of an iron- holding compound with all the characters of chromatin, and, 214 A. B. MACALLUM. in some cases, in such abundance as to obscure the outlines of the trabeculjc. He has found that in the fibro-vascular bundle this end of the nucleus is directed toward the base of the ovule, and is of the opinion, as a result of some investigation of this subject, that the phenomenon in question is connected with the processes of the formation of chromatin, which he regards as taking place here. The presence of assimilated iron, apart from its occurrence in hfemoglobin and hfematin, is an exceptional feature in the cytoplasm of the cells of the higher forms of animal life, but the exceptional instances are themselves of a constant cha- racter, and comprise, in addition to yolk -holding ova, the cells of yolk-holding embryos, the hsematoblasts of Vertebrates, and the ferment-forming gland-cells of all descriptions. The iron in the yolk of Amphibian ova is held in the yolk- spherules, which manifest a strong affinity for dyes, and are usually homogeneous in composition. These give with ammo- nium hydrogen sulphide a dark green reaction, which makes its appearance sometimes in a few minutes, but at the latest in a few hours, when the preparation is kept warm. The reaction is uniform throughout cacli spherule. The enclosing cyto- plasm does not, before development of the ovum begins, con- tain any assimilated iron ; but in the developing embryo, with the multiplication of the cells and the partition of the yolk, the spherules gradually undergo solution, for they become smaller in size, and then one obtains an iron reaction in the cytoplasm of each cell. The solution of the yolk-spherules may be studied also in preparations made with the carraine-indigo- carmine fluid, the reagent giving, in the earliest stages of the embryo, a green colour to the yolk-spherules, and a red stain to the cytoplasm and nuclei ; but in later stages the red colour is rarely obtainable, and both cell and nucleus, the latter espe- cially, are coloured bluc-grccn or dark green. This result is brought about by the solution of the yolk-chromatin in each spherule and the diff'usion of the dissolved substance through the cytoplasm and nucleus of each spherule-holding cell, for in those examples of larval Amblystomata which yield pre- lUON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 215 parations giving a dark green or blue-green colour in cell and nucleus after treatment with the reagent mentioned, the cells are found, after the prolonged application of the glycerine and sulphide mixture, to exhibit an iron reaction in the cytoplasm apart from the spherules, and a similar reaction diffused in the karyoplasm independent of that manifested by the chromatin network, the intensity of the reaction corresponding in each case to the depth of the green stain given these elements by indigo-carmine reagent. When in the advanced development of the larval Amblystomata the yolk-spherules disappear from the cytoplasm of the cells, the nuclei and all cells, except those undergoing transformation into striated muscle, lose their capacity for absorbing and retaining the indigo of the fluid of Shakespeare and Morris. This indicates that the yolk chro- matin is changed into some other compound, and the prolonged application of the glycerine and sulphide mixture confirms this, for the cytoplasm, except in secreting cells, the striated muscle- fibre, and in the hsematoblasts and red corpuscles, is destitute of iron compounds, while the nuclei give, much more slowly, and apparently with greater difficulty, a reaction for iron which is, in contrast with what is observed in the earlier stages, con- fined to the nuclear network and nucleoli. The iron containing substance is transferred to the nuclei, and with this transfer- ence the iron becomes more firmly combined— a process the very reverse apparently of that which is illustrated in the formation of the yolk-spherules, for the iron compound of the latter, though derived from the nucleus of the ovum, is less firmly combined than that of nuclear chromatin giving origin to it. The yolk-spherules of the hen's egg, as is well known, have characters difl'ering from those of Amphibian ova, but the most marked diff"erence consists in the distribution of the iron-con- taining compound. The yolk-spherule in the ova of Ambly- stoma and Necturusis homogeneous, andthe iron compound is uniformly distributed through it ; but in the lien's egg ele- ments of this character are to be found only in the constituents of the " white" yolk and in some of the " yellow" spherules in 216 A. 13. MACALLUM. the most peripheral layers of the yolk, while in all the other spherules the distinctive feature is the disposition of the iron compound in a finely granular form. This cannot be deter- mined with fresh yolk, for when treated with ammonium hydrogens 'p'le the greater part of it dissolves, and the solu- tion becomes ,M'k green owing to the formation of sulphide of iron. Under the microscope no formed elements can be ob- served in such a preparation, except those derived chiefly from the " white" region, and it is not possible to ascertain, under these conditions, the relations of the iron-holding nuclein in other parts of the yolk. Another difficulty experienced in dealing with fresh uncoagulated yolk i^ that, when removed from the egg the spherules disintegrate, the granular contents escaping and obscuring more or less the characters of the other elements. To avoid this the substance of the spherules must be coagulated, and to accomplish this satisfactorily I placed the eggs in boiling water for ten minutes. The spherules were thus fixed in polyhedral form, and, after these had lain in strong alcohol for several days, it was an easy matter to determine the distribution of the iron in them. The results obtained were according to the variety of spherules examined. In those known as "white" the reaction for iron was very distinctly obtained, but it was wholly confined to their homogeneous spherical bodies. The reaction is, imme- diately after the application of the glycerine and sulphide mixture, light green, but this becomes deeper after a few days, when the preparation is kept at a temperature of 60° C. The homogeneous elements undoubtedly contain a, quantity of nuclein, for they resist the action of artificial gastric juice and dissolve in weak alkalies, while tliey constitute the only part of the " white" spherules that possesses, like chromatin, the pro- perty of absorbing and retaining colouring matters. This was found to be the case specially when the spherules, coagulated by heat, were further treated with Flemming's chrom-osmio- acetic mixture for twenty-four hours, then with alcohol, and finally with a solution of safranin. When the excess of the stain was extracted with alcohol and the spherules mounted in IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 21! balsam, it was always found that the spherical elements ex- hibited an intense stain, while the remaining parts of the spherules were absolutely uncoloured. I found it possible to demonstrate this and the reaction for iron in the same pre- paration. When the "white" splierules, fixed with heat, were kept in slightly warm sulphuric acid alcohol for twenty-four hours, their spherical elements gave, on treatment with an acid ferrocyanide solution, a Prussian blue reaction, and, when sub- sequently stained with a safranin solution, became violet. These results show how close is the relationship between the sub- stance composing the spherical elements and chromatin. A few of the spherical elements in the " white " spherules are not of the character described, for in preparations made with Flemming's fluid one finds, now and then, a spherule ni which one or more large droplets of fat are demonstrated by the intensely black reaction of the osraic acid. Apart from the occurrence of these, there is comparatively little fat in th3 "white" spherules, a fact strikingly shown when a thin section of the hard-boiled yolk, embracing portions of the " white " and " yellow " zones, is submitted to the action of the reagent for twenty-four hours, the " white " then exhibiting a greyish appearance, while the " yellow " area is almost black. The " yellow " spherules are also richly supplied with the iron-containing compound, but this is quite differently distri- buted from what it is in the " white " zone. The appearances of these are subject to a great deal of variation. Some contain only large round granules, in others the granules have a puncti- form character, while in others again both kinds of elements may be mingled with minute fat droplets. Owing to differ- ences in the specific gravity of the constituents apparently, the granules may be found, in some cases, to be gathered in one portion of the spherule, the remainder of the contents being or ^led by a clear, non-granular substance of a firm ■ consistence, a character resulting from heat coagulation. It is in such spherules as these that one determines distinctly how the iron compound is disposed, for, in those in which the granules are uniformly distributed, it is sometimes exceedingly difficult I 218 A. n. MACALLUM. to decide whether the iron is contained in the granules or in the extra-granular stibstance, so intimately are these usually intermingled. The granules, whether of the large or of the punctiform variety, always contain an iron compound, while the substance in which they are shown is destitute of this element. In demonstrating this fact the acid alcohols are of the greatest service, the glycerine and sulphide mixture, owing to the large size of the vast majority of the spherules, ncc being as effective in liberating and demonstrating their iron, but in the smallest spherules the complete reaction may be obtained with the mixture in four or five days. In those spherules which contain, as described, granular and non- granular portions, the granules, closely aggregated as they usually are, o ^ear very prominent by reason of the reaction for iron whioh they give with both methods of demonstration. In some of the " yellow " spherules also, after treatment with sulphuric acid alcohol, vesicles '^f different sizes were observed, each of which appeared to be enveloped by an iron- containing raerabrai\e-like structure. Their position near the centre of the spherule often rendered the occurrence of iron in the envelope obscure, owing to the light passing through so many iron-holding granules above and below these vesicles. What the latter contained it is not possible to say, for although fat globules of a similar size can be demonstrated in some spherules when these are subjected to the action of the chrom- osmio-acetic mixture for twenty-four hours, it cannot be demonstrated that the two classes of structures are connected in any way.^ The difficulty lies in the fact that in order to show the occurrence of iron in the envelope, alcohol in some form must be used, and by this the fat is largely, if not wholly, removed ; while in those spherules treated only with osmic acid solutions the black reaction of the globules prevents a demon- stration by the Prussian blue reaction of any iron present. I In another paper (" On the Absorption of Iron in the Animal Body," ' Journ. of Pliysiol.,' xvi, 1894, p. 268) I expressed the viev/ that these vesicles contain fat. After a more extended study of these elements than I was able to make before that paper was published, I am doubtful of this interpretation of their structure. IKON COMroUNDS IN ANIMAL AND VEGEiAULE CELLS. 219 Apart from the question of the occurreace of fat in such elements, there may be no doubt about the intimate associa- tion of the iron-containing substance and the fat in the spherules. Owing, however, to the size of the latter, as well .'■ to the density of the coagulated material in them, the osmic acid used to demonstrate the fat penetrates but slowly, and when, as usually happens, fat droplets stud the periphery of the spherule, little or none of the reagent reaches its interior, which then has only a straw-yellow colour. If, however, a few spherules, coagulated with heat, are kept in a quantity of Flemming's fluid for twenty-four hours, the osmic acid pene- trates the spherules in some cases and causes their granules to become brownish-black, a fact which can be most distinctly observed when the cover-glass is pressed down sufficiently to disintegrate the spherules and set the granules free. If the granules are large, the occurrence of fat in them is much less readily demonstrated, possibly because the density of such elements prevents penetration on the part of the osmic acid. These granules are undoubtedly the source of the greater part of the iron-hohiing nuclein isolated by Bungc from the yolk,^ since the "white" yolk is comparatively small in amount. Miescher^ regarded the nuclein, which he separated from the yolk, as only in part localised in the homogeneous spherical elements in the " white " portion, and he 'relieved that the greater part of it was derived from the gr .ules in the " yellow " spherules, and that none of it exists in a dissolved form, a conclusion fully supported by the facts concerning the localisation of the iron. In describing the transference of the chromatin of the spherules from the cytoplasm to tlie nucleus of each cell of the larval Amblystoma, reference was made to an exception in the case of developing muscle-fibre. In the cells undergoing transformation into striated fibres, some of the chromatin dis- solved in the cytoplasm f'nds its way into the nuclei as in other I •' Ueber die Assimilation des Eiseiis," ' Zcit. (iir Physiol. Chcmie,' vol. ix, p. 49, 18S5. c, , , i>r J s "Die Kerngebilde im Dotter des Iluhuereis," ' Hoppe-Seylcr s Med.. Cliem, Uulersucliunsen,' 1S71, p. 502. 220 A. D. MACALLUM. cells generally, but the greater part appears to remain in the cytoplasm of the developing fibre, and undergoes a transforma- tion which is one of great interest in connection with the origin of haemoglobin. In the cytoplasm of the muscle-cells there is an abundance of yolk-spherules which, as in other cells, gradually undergo solution, the dissolved substance diffusing through the cytoplasm. When the striation makes its appearance at one side of the now elongated cell, the dis- solved substance passes into the striated area, for ammonium hydrogen sulphide brings out an iron reaction in this part as readily as in the undifferentiated cytoplasm and in the spherules, but con^.ned to the dim bands, the light bands giving no evidence of the presence of the compound. In the fibre from which the spherules have all but disappeared, and in which the striated area embraces nearly the whole of its width, the reac- tion with ammonium hydrogen sulphide is as distinct and as marked as in the earlier stage, and this is true also of the fibre in its final form. In this stage the iron is quickly liberated by acid alcohols, as well as by ammonium hydrogen sulphide, and its presence may be readily demonstrated by means of these reagents up to the period when all traces of yolk disappear from the cells of the larvae. After this date the iron compound becomes firmer, or, to speak more accurately, is less readily attacked by acid alcohols or the sulphide reagent, and in the muscle-fibre finally its presence may not be shown by these methods. It is not that the iron is removed from the fibre, but that the compound containing it is transformed, in red muscles, into what is called myo-hajmatin by MacMunn, or haemoglobin by Hoppe-Seyler and others. The latter com- pound can, by means of the staiuinj " "" of Shakespeare and Norris, be clearly shown to be stn^. , confined to the dim bands, which are given a grass-green colour distinctive of hajmoglobin, while the light bands and nuclei are coloured red.i ' I have pointed out the value of the reagent in this respect iu my paper entitled " Studies on the Blood of Amphibia," ' Transactions of the Canadian luslitutc,' vol. ii, 1S93. IKON COMPOUNDS IN ANIMAL AND VEGETAULE CELLS. 221 A similar con/ersiou of a compound in which the iron is easily attacked by ammonium hydrogen sulphide and by acid alcohols into one from which the liberation of the iron is more difficult, obtains in the dim bands of mxscle-fibre in Inverte- brates (Oniscus, Chironomus, Musci), but in this case the transformation does not proceed as far as the production of hajmoglobin or myo-hajmatin, if one may judge from the absence of pigment and from the fact that the liberation of the iron, though difficult, is possible, while in the case of haemo- globin the use of acid alcohols and of ammonium hydrogen sulphide is ineffective for that purpose.^ In the development of the blood-corpuscles in the larvK of Am bly stoma there is, as I have pointed out,2 a conversion of the chromatin of the ha;raatoblast into haemoglobin, a change that is analogous to that described above as occurring in muscle- fibre. In hajmatoblasts, however, the chromatin so trans- formed is not directly derived from that of the yolk-spherules, as is the case in muscle-fibre, but from that compound after it is transformed into nuclear chromatin. This is very distinctly seen in sections through the aortic arches of the larvae, which have been treated with acid alcohol to liberate the iron. In the concave side of the arches are seen hsematoblasts in all stages of division, and in these one may, by the iron reaction, diflerentiate between haematoblasts in which there is no cyto- plasmic chromatin, and those in which the cytoplasm between the chromatin loops of the mitotic figure contains dissolved chromatin to an extent varying with the example of htumato- ulast noted. This cytoplasmic chromatin does not act in the same way as - rdinary nuclear chromatin does towards staining reagents, £.s, for example, hematoxylin, eosiu, and safranin, 1 The fact that ammonium hydrogen sulphide will liberate the iron from hffimatin in solution, x/hile it docs not attack tiie iron in the compound called myo-hrematiu by MacMunn, indicates that the latter cannot belong to the hfcmatiu class. Its property in this respect shows that it is related to htemo. globin. The name given to it by MacMunu certainly appears to be a mis- leading one. Loo. cit. 222 A. B. MACAI-LUM. although it has an affiuity for them, and it persists mth ths character for a long time after the «tage of the 1-- U,blas s passed. 1 have fouud that iu a large number of H,c tuUy- Zed red eells in the spleen of the larva of 35 u.m ength the disc contains a quantity of the modified chromatin and from this the iron is readily liberated, but m later stages both the number of such corpuscles and the amount of |r°u " ^e disc which may be liberated by aeid alcohols gradua ly diminis and disappear, the haemoglobin of the disc not yielding its on on the enV ^n-ut of such methods. The nuclear chromati , however, oi all stages of the corpuscle, readily gives up its iron, even when rone can be set free in the disc. It thus appears that the haemoglobin of the red corpuscles and the analogous compound in muscle-fibre are formed in he same way, the only diflerenee obtaining between them existin. iu the fact that the pigment of musele-fibre does not m .t« evolution in the developing ovum, comprehend a stage ot nuclear chromatin. The process by which they are formed is a gradual one, and the position of the iron in the mo ecule is apparently changed. The latter result muy be partly ac- counted for if we consider the composition of chroma in and of haemoglobin. Chromatin is an iron-holdmg uuclco-albumiu in which the iron is attached to the uuclein, while m haemo- globin the iron is held in the haematin moleeu e and in the transformations which result in the formation of h^matin out of nuclcin, it is but natural to expect that the relations of the iron to the molecule should change also. In secreting cells, as, for example, those of the parotid, Lieberkiiuiar ud pancreatic glands, a certain portion of the cytoplasm gives evidence of the possession of "masked iron. AVhen the cells of the pancreas of an adult Amblystoma are, after hardening in alcohol, subjected to the action of the glycerine and sulphide mixture for six or seven days at a tem- perature of G0° C, in addition to the reaction for iron obtained In the nucleus, one is found in the cytoplasm of the so-eaued -outer zone," in some cases almost as marked as in the nuclear chromatin. The extent of the cytoplasm in^ dved in the reac- IttON COMPOUNDS \S ANIMAL AND VEGKTAULE CELLS. 223 tiou in all the specimens which 1 examined varied considerably, whether according to the stage of secretory activity could not be detcrminc(' after the use of ammonium hydrogen sulphide, for this reagent, in a day or two at an elevated temperature, causes the zymogen granules to disappear; but in sections of the pancreas from the same animal, after these had been acted on by sulphuric acid alrohol, then with the acid ferrocyanidc solution and eosin, the irou aolding area in o-ch cell was de- monstrated by the 'ulting Prussian blue, wh.lc the zymogen granules were giv •; a intense red stain, and in this case it was founr' that, apa c from the granular zone, the cytoplasm was uniformly blue. In other conditions of activity the iron- holding area was increased or decreased in correspondence with the decrease or increase in the extent of the gr.-mlar zone. In the exhausted condition of the gland-cell, that is, in which there were but few granules, arranged in the " border" fashion near the lumen of the tubule, the whole of the cyto- plasm exhibited the blue reaction, but the latter was less marked than when it was confined to a narrow zone in the neighbourhood of the nucleus. The relations of the extent of the iron-holding area to the stage of secretory activity were less easy to determine in the Lieberkuhnian and parotid glands, for it is not possible to demonstrate the mucigen in the former or the zymogen in the latter as prominently as the zymogen granules may be in the pancreas, but in these the iron-holding area appeared in all cases to correspond, in the main, with the "protoplasmic'' or "outer" zone. lu the ''mucous" cells of the submaxillary gland of the eat and dog only a narrow zone of cytoplasm about the shrunken nucleus contains iron, but in the large crescents of Gianuzzi in the eat the whole of the cytoplasm is iron-holding. In the peptic tubules of Ambly- stoma the cytoplasm in the outer half of each cell contains iron, and this is also true of the chief cells in the cardiac por- tion of the stomach in the dog and cat. In the parietal cells in these animals the cytoplasm is absolutely free from iron. The iron-holding zone in each chief cell appears to vary in extent with the stage of bccretion, but I am unable to speak VOL. 38, PART 2. — NEW SER. P 224 A. n. MACALLUM. as definitely upon this as upon the relations, in this respect, observed in the pancreatic cells of Amblystoma, for I have not been suceossfnl in my efforts to obtain, from examples of the latter animal, preparations of the gastric glands illustrating marked variations in the stages of secretory activity, and have had to rely upon those made from the cat and dog, in which the chief cells are comparatively small and less favorable for observation on this point. It is only in the mucous glands of the skin of Amphibia, and in the renal tubules of Vertel)rates generally, that I find exceptions to the lale that glandular secretion is associated with the presence of an iron-holding cytoplasm. I have not found any exceptions in Invertebrates to this generalisation, but my observations have not been comprehensive enough on this point, and I must speak with some reserve in regard to it. In the Protozoa, as I will show further on, the presence of assimilated iron in the cytoplasm seems to be a constant feature, the iron no* ')eing confined to any part of the cell, but uniformly distribu i through it, and there is a probability tliat this cytoplasmic iron-holdin-^' compound is also associated with the secretion of ferments functioning in the digestion of the ingested food. In the glands named above, which are mentioned as exceptional instances, the absence of assimilated iron from the cytoplasm may be explained on the ground that the secretory process of a renal cell is widely different from that of a pancreatic cell, the cytoplasm in the latter, but not in the former, elaborating a portion of its own constituents to furnish the secretion, whereas iu the renal cell the process is largely one of transference oi-Iy. If the exjilanation should hold in all possible cases of exception, then it would follow that the iron-holding compound is an important element in the elabora- tion of the zymogens. I have elsewhei'e ' pointed out the rela- tions that obtain between the chromatin of the nucleus and of the cytoplasm of the pancreatic cell, on the one hand, and the formative process resulting in the production of zymogen on » " CouU'ibutions to the Moiplioiogy and Physiology of the Cell," ' Trans. Canadian Institute,' vol. i, part 2, p. 217, 1S91. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 225 the other; and so intimate did these relations appear that I .las led to apply the term prozymogen to the chromatiu. I have found, as a result of experiments on the active pancreas of Amblystoma, that the zymogen granules under certain conditions give an iron reaction. When the organ, hardened in alcohol, is put in a quantity of Bunge's fluid, and the pre- paration kept at the temperature of the room (20° C.) for a week, or when it is kept for two days in a quantity of sulphuric acid alcohol, teased-out portions, after the removal of the acid and on tlie addition of ammonium hydrogen sulphide, give pre- parations of which that represented in fig. 38 is an illustration. The zymogen grauides give a greenish reaction, the colour making them more prominent than the other elements in the cells. The cytoplasm of the " outer zone" gives but a feeble iron reaction, and this appears only to a minor extent in the nuclear elements, both results being caused by the lessened action and feeble extractive capacity of the acid alcohols when used on the tissue in mass. When the reagents are used for longer periods thau those specified the iron disappears from the zymogen granules, while it becomes more strongly marked in the nuclear elements and in the cytoplasm of the " outer zone." Owing to the effect that ammonium hydrogen sulphide exer- cises on the granules, causing them to dissolve or disintegrate, an effect already referred to above, it is not possible to control the results obtained with the acid alcohob by experiments with this reagent, and one may, therefore, not regard the presence of iron iu the zymogen gr:-,nules as conclusively demonstrated, since it may be' urged that the iron reaction which they gave was due to the iron which diffused into them from that liber- ated in the other cellular elements. When one remembers, however, the fact tnat the zymogen is elaborated iu a cyto- plasm which is iron-holding aud at its expense, the occurrence of a faint reaction for iron iu the granules after the use of acid alcohols ii best explained by the view that the zymogen of the pancreis contains iron, and that its antecedent, the prozymogen, is the iron-holding constituent in the cytoplasm of the ' outer zone." I 220 A. i;. MACAFiLUM. In the rods and cones of the retina in Am bly stoma and Necturus an iron reaction was frequently obtained like that represented in fig. 37. It was always feeble and confined to tiie trabecule, which "tretched across the long axis of the rod, or which formed the network in the cones. In some cases (as in fig. 37, a) pigment-granules were observed attached to the rods, probably derived from the cells of the tapetum nigrum, and as the pigment probably contains iron, it is uncertain whether the iron demonstrated in the rods and cones was not derived by diffusion of some iron-holding substance from this source. The eleidin granules in the stratum granulosum in the human skin give, after treatment of sections of the epidermis with sulphuric acid alcohol, a dark green reaction with ammo- nium sulphide. I have not succeeded in obtaining a reaction for iron in them when the containing cells, hardened in alcohol, were simply subjected to the prolonged application of the gly- cerine and sulphide mixture in the warm oven. Since the chromatin of the nuclei in the underlying stratum raucosum is, as elsewhere, iron-holding, while the nuclei in the stratum granulosum are poor in chromatin, it is not improbable that the iron, at least of that part of the latter which disappears from the nuclei, is the source of the iron shown in the eleidin granules. The homoj^eneous substance constituting the stratum lucidum also gives a reaction for iron, Avhieh is diffuse and less marked than in the granules of the underlying layer. In my observations on preparations of the human thyroid and of that of the dog, although it was easy to demonstrate the presence of iron in the nuclear chromatin, and to a certain extent in the cytoplasm of the cells lining the alveoli, I did not succeed in finding any of it in the "colloid" matter. Under certain conditions this substance absorbs staining matters, and it also gives^ the molybdatc-pyrogallol reaction of Lilienfeld and Monti." These facts suggest that the colloid ' i\ Gourlay, " Tlie I'lotciils of the Tliyroid and tlie Si)Iecii," 'Journal of riiysiology,' vol. xvi, p. '215, ISOI. 5 "Diu niikio-clieniibclie Lokalizaliou dos riiospiiors iu deu Geweben," 'Zcit. fiir riiysiol. Clicniic,' vol. xvii, p. 410, 1893, IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 227 substance is allied to nuclein, and, according to Gourlay, the nucleo-alburain which he isolated from the thyroid was derived in large measure from the colloid matter which he, relying on the reaction of Lilieufeld and Monti, found to contain phos- phorus. If colloid matter is therefore a nucleo-albumin, its freedom from iron renders it, in contrast with the chi'omatins, a subject of oial interest. Assimilated iron is rarely found in the cytoplasm of the cells of the higher vegetable organisms, and amongst the examples illustrating its presence may be mentioned the cells of the nucellus in the ovules of Erythronium americanum, and those of the gluten layer in the wheat-grain. The cytoplasm of the cells of the nucellus, when fertilisation has taken place, and even before this occurs, gives, after treatment with sulphuric acid alcohol, a distinct reaction for iron, which, however, in respect to intensity, is not to be compared with that mani- fested in the nuclei of the same cells. The iron in the cyto- plasm in this case is not due to diffusion from the nuclei during the course of treatment with the liberating reagent, for it is also demonstrated in this situation in the glycerine and sul- phide preparations. As the nuclei of the nucellus are much richer in assimilated iron than those of other parts of the ovule, except the embryo sac, it is possible that the cyto- plasmic iron compound is intra vitam difiused from the nuclei, and, further, as the cytoplasm of the embryo-sac of this stage sometimes gives a difluse reaction for iron after it has been treated with acid alcohols, its presence here may be due to a similar diffusion from the cells of the nucellus. 1 have observed in certain preparations in which the nuclei of the embryo-sac were in the stage of division, a large number of iron-containing granules interspersed amongst the fibrils of the achromatic spindles, and as in other preparations similar granules were stained with hicmatoxylin, like the chromatin loops, it would appear as if the granules were formed of chro- matin. The cytoplasm holding these granules gave no reaction for iron. The cytoplasm of the cells of the gluten, or so-called alcu- 228 A. E. MACALLUM. rone layer (Kleberscliiclit) in the Avhcat-grain isriohly supplied with a " masked" compound of iron. In some cells it is chiefly found in the large granules strewn through the cytoplasm ; in others, again, apparently it is wholly contained in the latter; while in certain instances, further^ it Mas demonstrated only in the extreme peripheral portions of the large gianules. This is most clearly shown in sections of the grain after they have been treated with sulphuric acid alcohol for twenty-four hours at a slightly raised temperature. When the individual cells of other sections are treated with the glycerine and sulphide mixture for several days the reaction for iron is readily ob- tained in their cytoplasm, but its localisation, as observed after the use of the other method, is thus less readily determined. The "masked" compound apparently belongs to the class of chromatins, for when sections are treated with the ordinary staining reagents the cytoplasm stains deeply, especially with safrani i and hscmatoxylin, and the parts which are specially affected are those which correspond Avith the iron-holding structures in preparations treated with acid alcohols. Haberlandt^ has made experiments upon the question of the site of origin of the diastase in the germinating rye-grain, and these appear to show that the ferment is elaborated in the cells of the gluten layer only. It is possible that the iron-contain- ing compound in the cytoplasm of this layer is the zymogen or prozymogen of the ferment. IV. — On the Occurrence of Assimilated Iron Compounds IN Special Forms. ^Yhat I have said in the foregoing pages with regard to the presence of iron in the chromatin of higher forms of animal and vegetable life is true also in regard to the types of lower organisation in both kingdoms. In the investigation of the less highly organised animal and vegetable forms, however, ' "Die Kleberscliiclit des Giasendosperms als Diastase aussclieidendes Drusent^ewcbc," 'Bcrichte der deutschen botan. Gesellscb.,' 1890. \i. 40. Abstract in 'Botan. Cciitralbl,,' vol. xliii, p. 39. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 229 some important variations were found in the disposition of the iron-holding substance, and it was further determined that in non-nucleated organisms the exceptional distribution of the chromophilous substance is co-extensive with that of the assimilated iron compounds observed. Such facts are worthy of an extended description, and I now propose to detail these and the more important observations allied to them. Ascaris.— In the species A. mystax the spermatozoids and ova, both before and after fertilisation, manifest special features in the distribution of the iron-containing substance. When they are hardened in alcohol, the spermatozoids are comparatively easily affected by the ammonium hydrogen sulphide, the reagent, mixed with glycerine, givijg in a couple of days, under the usual condition", a reaction for iron, which usually is confined to the " nucleus," a dense homogeneous body (fig. 31) ; but in several instances the "membrane" also contained iron. The reaction in the latter varied in intensity, and when most marked it revealed a structure in the " mem- brane " like that represented in fig. 32. The iron compound observed in such a case o])tained only in the rodlets constitut- ing the " membrane." What the occurrence of assimilated iron in this situation signifies I am unable to say, except that it possibly represents an abnormal phase of a condition normal to the spcrmatozoid after it has penetrated the ovum. When the sperniatozoid begins to penetrate the latter, its membrane frequently manifests a weak reaction for iron (fig. 29), while its cytoplasm does not give any evidence of the presence of that clement ; but in the changes it undergoes after reaching the interior, the "nucleus" becomes in part dissolved, and the chromatin, as shown by the iron reaction, diffuses into the cytoplasm and into the membrane, from which some of it passes into the cytoplasm of the ovum immediately adjacent to the sperniatozoid. The membrane in this way becomes the most prominent part of the sperniatozoid. As the transturma- tion proceeds, the membrane also dissolves, and the iron which it contains appears to pass back again into the cytoplasm of 230 A. B. MACALLUM. the spermatozoidj but what is held in the cytoplasm of the ovum apparently is retained by the latter. These observations on the diffusion of the iron-holding sub- stance from the " nucleus " of the spermatozoid into its cyto- plasm coincide with those of van Beneden upon the changes which take place in the spermatozoid of Ascaris megalo- cephala after it penetrates the ovum. He found that the protoplasm of the free spermatozoids ^ manifests no affinity for staining compounds, while its capacity for absorbing and retaining all colouring matters becomes remarkable immedi- ately after it enters the ovum. As the " nucleus " at the same time loses in part its affinity for stains, he came to the conclusion that a part of the chromatic substance (chromatin) of the " nucleus " becomes dissolved in the cellular substance (cytoplasm). O. Zacharias" has also pointed out that the proto- plasm of the free spermatozoid, apart from its " nacleus," is absolutely unstainable, but after it penetrates the ovum it at once manifests an affinity for colouring matters. Kultschitzky,^ referring to the reactions with staining fluids, suggests that possibly the " nucleus " gives off to the cytoplasm of the spermatozoid a portion of its chromatin, or that, in other words, not all of the chromatin of the " nucleus " is employed in the construction of the male pronucleus. I have found in my preparations that tlie cytoplasm and " membrane" of the spermatozoid which has penetrated the ovum, and, frequently also, that portion of the cytoplasm of the ovum in the im- mediate vicinity of the spermatozoid, have a slightly greater affinity for colouring matters than the cytoplasm of the free spermatozoid or of the unimpregnated ovum. In many of his illustrations van Beneden represents that part of the spermatozoid which I have called the "membrane " • " Rechercbes sur a maturation de I'ojuf et la fccondation," ' Archives de Biologic,' vol. iv, p. 205, 1883. " " Neue Untersuchuugcn iiber die Copulation der Geschleclitsproduktc und den Befruchtungsvorgiinge bci Ascaris mcgaloccpbala," 'Arcb. fiir Mikr, Anat.,' vol. xxx, p. Ill, 1887. '"Die Befrucbtungsvorgiiiige bei Ascaris megalocephala," 'Arch, fiir Mikr. Anat.,' vol. xxxi, p. 5G7, 18SS. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 231 as deeply stained, and in these one finds the existence of rod- lets indicated, such as those to which I have referred above; but these (les stries transversales de la queue) are more ap- parent iu the penetrating than in the free sperraatozoid. I observed only faint traces of such structures in the sperraato- zoid in the interior of the ovum, the rodlets apparently com- mencingtodisappearimmediatelyimpregnation is accomplished. The chromatin of the nucleus of the ovum gives a deep reaction for iron iu whatever stage the nucleus may be found (figs. 29 and 30). The chromatin also of the " polar globules" contains iron, and I made efforts to determine the ultimate fate of this, but these were unsuccessful. It would appear, however, as if the chromatin of the extruded elements were dissolved eventually in the cytoplasm, for it is impossible to find any traces of it after a time. Chironomus. — Balbiani,^ who was the first to call the attention of cytologists to the structure of the nuclear elements in the "salivary" glands of the larva of Chironomus, de- scribed the nuclear filament as made up of a series of dim discs or bands, each placed transversely, and separated from its neighbour on either side by a band of clear substance, the filament possessing, however, at certain points an annular swelling, and terminating at its ends either in the polymor- phous nucleolus or by an attachment to the nuclear membrane. Leydig,^ the next observer, found each dim stria to be made up of a series of elements whose separation from each other gives a composite character to the stria. The fine lines sepa- rating the elements are, according to his observation, con- tinued from one dim disc, through the light disc on either side of it, to the adjacent dim disc. In this way a series of exceed- ingly delicate longitudinal lines, in addition to the coarse trans- verse ones described by Balbiani, make their appearauce. Leydig also believes that the substance forming the dim band ' " Sur la structure du noyau des cellules salivaires cliez les larves de Cliiro- uomus," 'Zool. Anzeiger,' 1881, pp. 037 and GG3. - ' UutcrsuchuiiKcn zur Anatomie und Histologic der Tliicrc,' Bonn, 1883, p. 90. 232 A. B. MACALLDM. is situated immediately under the membrane. Korschelt's views on the structure of the filament are directly opposed to those of Leydig and Balbiani. lie regards the transverse striation of the filament as due to a folding of the surface membrane only, and explains the longitudinal striation ob- served by Leydig as caused by the action of the reagents used. In his opinion, also, the apparent diff"erentiation of the fila- ment is due to the differences in the reflected light. So far as I know, no one has hitherto observed an arrange- ment in the nuclear filament of Chironomus similar to that described by Leydig, although Carnoy has found in the salivary gland ofaNemocere larva that the dim disc is formed of a series of longitudinally disposed rodlets, but he attributed the delicate lines observed in the clear discs to folds in the mem- brane of the filament.2 The larvse of the species of Chirono- mus accessible to me oS'er preparations less favourable for study than do those of the species C. plumosus studied by Balbiani and Leydig, yet I have been able to determine, with my methods for demonstrating the presence of assimila* ■ iron, the correctness of Ley dig's observations so far as they go. The dim discs arc of difterent thicknesses, the thickest appearing to be five or six times the diameter of the narrowest. When the salivary gland, after being hardened in alcohol, is kept for several days in sulphuric acid alcoliol, treatment with an acid ferrocyaiiide solution gives all these dim bands a deep blue reaction, the intensity of the reaction coming out very markedly in the thicker bands. Under the highest magnifica- tion of service in such a case (apochromatic immersion 1-5 mm. and compensation ocular 8, Zeiss), the bands of medium thickness are resolved into a series of short rodlets disposed parallel with the filament. If the filament has, in the course of preparation, been isolated from the nucleus, one may then determine that the rodlets forming one dim l)and arc' connected by excessively delicate fibrils with the rodlets 1 Ucber die ei;;eiitliuniliclien Bilduiig in den Zellkcrneu der Speiclicldiuscu von Cliironomns plumosus," ' Zool. Anz./ vol. vii. pp. ISO, 221, 2il. 1881. 3 » Biologie CcUulaire,' p. 232. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS, 233 forming the two adjacent bands. The fibrils^ or what are in appearance such structures, have so little iron in them that frequently in a large part of an isolated filament their blue reaction may not be sufficiently deep to betray their presence, but the chances of observing them may be increased by staining such preparations carefully with safranin. Probably the ex- pression fibril is not a correct one to apply to these appear- ancesj for they may be the optical sections of the partition walls of compartments, the extreme ends of which would in that case be formed by the dim bands. What appears to sup- port the latter view is the fact that in some of the thickest dim bands the Prussian l)lue reaction reveals the presence of a single row of vesicles extending from one end of the band to the other, the vesicles sometimes having an elongated form parallel with the filament. It seemed to me that these were the initial stages in the division of one dim band into two, that the thinner bands represent those most recently formed, and that, therefore, the vesicular mode of formation would result in the production of a series of compartments the thin walls of which, in the clear bands, would appear as fibrils. The struc- tures observed are, however, so exceedingly minute that it is impossible to determine definitely anything on this point. The iron-holding substance in the filament is, therefore, dis- posed in the rodlets of the dim band and in the fibril-like elements connecting the rodlets of one dim disc with those of its neighbours. The only exception to this statement may be made in regard to the structure of the swellings which are sometimes found on the course of a filament (fig. 50). In this case the dim discs are replaced by an iron-holding reticulum disposed in the interior of the swollen portion of the filament. A comparison of this portion with the adjacent portions of the filament appears to indicate how the reticulum has arisen and what its relations arc. The iron-holding bands on either side are less regular in their disposition than else- where, and the fibril-like structures arising from them appear to be directly connected with the iron-holding substance of the reticulum referred to. The swollen portion of the filament 234 A. li. MACALLUM. varies in its size and shape, but most frequently it has the appearance represented in the figure. I have never observed the annular swellings described by Balbiiini as present in the filament in C. plumosus, but I take it that the swollen portions here described are the representa- tives of such structures. Nor have I ever determined that the filament ends by attachment to the nuclear membrane, or to the amoebiform nucleolus, through which it may pass several times in its course. The nucleolus varies not only in form and size but also in composition. It may be homogeneous, but more frequently the central portion contains vacuoles and granules and stains more deeply with eosin or safranin, while the peripheral non-granular poiuion may possess no staining capacity whatever. In many preparations made from alcohol material and stained with eosin, the nucleolar body alone is stained, and this is particularly the case when the preparation has been treated with acid alcohol and the acid ferrocyanide mixture to demonstrate the iron present. The nucleolar sub- stance, apart from its granules, contains iron, but the iron present is very small in amount compared with that observed in the filament, for, when the latter gives an intensely deep blue reaction, the colour given the nucleolus is a very pale blue, and when the nuclei are kept for a week mounted in the glycerine and sulphide mixture in the warm oven, the isolated nucleoli develop only a greenish colour, portions of the fila- ments, on the other hand, giving in the same preparations a marked dark-green reaction. Unlike the differences in staining exhibited after treatment with eosin, the faint or light blue reaction is uniform throughout the nucleolar substance. The dim bands with the excessively fine fibrils in the filament are formed of chromatin, as shown by treatment with the stain- ing reagents, when the preparations have been properly hardened. There is, howevc, a diff'erence between this chro- matin and that of the ordinary animal cell in that while acid methyl-green colours the former it leaves unaff'ected the nucleoli and the swollen portions of the filament, which stain deeply with hajmatoxylin and carmine. [RON f'OM POUNDS IN ANIMAL AND VKOKTAULE CELLS. 235 lialbiani^ concluded from such results that chromatin (sub- stance chromatiquc) is present, not only in the dim discs, but also in the annuhar swellings and the nucleoli. According to Flcmraing/- safranin colours all these clenaents, but stains the nucleoli very strongly. Flemming's observation is correct only for preparations made with the chrom-osmio-acetic reagent j but when the nuclei have been fixed with alcohol, or with corrosive sublimate, treatment with acid alcohol for two or three days affects the filament in such a way that its discs and their excessively fine fibrils absorb and retain the safranin to a very marked extent, while the nucleolus remains unstained, and the swollen jwrtions of the filament are faintly coloured. It is possible to obtain in such preparations both the safranin and the Prussian blue reactions, and then, with the exception of the faint blue in the nucleoli, both effects are co-extensive and of equal intensity. The marked difference between the substance of the discs and that of the nucleoli is thus shown, but it may be brought out in a more brilliant way by staining Prussian blue preparations with eosiu, which then affects the nucleolus only. The nucleolus thus resembles the similarly named structure obtaining in the nuclei of Vertebrates, but it differs from this in that it is amoeboid in form, and does not possess, in any case, a chromatin envelope. The presence of granules and vacuoles, moreover, appears to indicate that it is physically active, which cannot be postulated of the vast majority of the nucleoli of Vertebrate cells. V.''hatever effects may be obtained by treating the nuclei with various staining reagents, but one results in the cytoplasm of the secreting portions of the salivary gland in Chironomus. Acid methyl-green in the fresh preparations, and ha^matoxyliu and safranin in the hardened glands, demonstrate very clearly that there is a staiuable substance, in many respects like chromatin, uniformly distributed through the cytoplasm ; that it is chromatin would appear from the fact that the cytoplasm 1 Loc. cit. i* 'Zellsubslanz, Kern- mid ZellUiciluug,' pp. 112, 113. 23G A. n. MACALI.UM, holds an assiraihitcd iron compound, for if small fragments of cells, hardened iu alcohol, be subjected to the action of the warm glycerine and sulphide mixture fcr a week or more, they will manifest a dark-green reaction which, when the mixture is washed away and replaced by ai) yoid ferrocyai.de solution, ia converted into that of Prussian blue. One may more readily obtain the demonstration of the iron in these cells by allowing sulphuric acid alcohol to act on the hardened gland for two days, when the cytoplasm of the secreting cells and the sub- stance of the thread (silk ?) in the lumen give evidence of the presence of this element. Whether the iron thus demonstr-ited in the substance of the thread belongs to the latter, or is derived by diti'usion from the cytopb'sra of the secreting cells during treatment with the acid alcohol, I am unable to say, since my experiments made to determine this question, by the use of the glycerine and sulphide mixture on isolated bits of the threads, turned out to be failures.^ The substance forming the threads manifests a strong affinity for dyes, and should it eventually be ascertained that the iron demonstrated in it, after treatment with acid alcohol, is part of a " masked " compound contained L) it, the facts will then all indicate that the iron-containing substance iu the cytoplasm is the antece- dent of a*^ 'east a portion of the substance of the thread in the lumen, and one will have then also a parallel of what was pointed out as obtaining in the pancreas and other ferment- secreting cells in Vertebrates. Protozoa. — I have selected the genera Stontor, Epis- t y 1 i s, V r t i c el 1 a, and P a r a ra ce c i u m for specially illustrating the distribution of the assimilated iron in unicellular animals. Avery large number of other forms were used to conlirm the results which a study of the named organisms gave, but owing ' Gilson (loc. cit.) lias referred to the fact "that the silk of certain insects seems to possess a stronger alliuity for this metal (iron) t,!uiu nuclein itself." 1 have observed ll i peculiarity, but the -rou absorbed is at ouce demonstrated on the application of any form of ammonium sulphide, a fact which shows that the iron so revealed does not enter into a " masked " condition, and ought not to be confused with that of " masked" compounds. J'r*' i lUON COMPOUNDS L\ AMMAL AND VKOETABLE fELLS. 237 to the difficulty cxperuniccd in getting' examples of such forms in the numbers required, it was impossible to make a fully satisfactory, systematic investigation of their iron-holding character. On the other hand, examples of the genera named could be obtained at all times in abundance, and T regard the opportunities thus presented as compensating in some measure for the limited range of genera studied. One of the difficuUies encountered in attempting to study the distribution of iron-compounds in Protoza is the fact that many of the motile forms, and some also of those which are sessile or attached, have in their cytoplasm inorganic com- pounds of iron, in great part, if not wholly, derived from the food matters ingested, and when such organisms, after being hardened in alcohol, are treated with the glycerine and sulphide mixture, they give at once a deep reaction for iron which, iu many cases, obscures other details in the cytoplasm and nucleus. When, moreover, attempts are made with acid alcohols, and especially Bunge's fluid, to remove the inorganic iron, the convlitions under which the experiments are made enable the reagent to liberate the " masked " iron at the same time, in which case the liberated portion becomes indistinguish able from that present previously in an inorganic form. To avoid such difficulties it is necessary to select forms in which the amount of inorganic iron is small or infinitesimal, and by determining the amount of the reaction obtained during -he first ten minutes after the application of the glycerine and sulphide mixture, one may thus prevent confusion arising from the study of results ol)tained by the more prolonged applica- tion of the reagent. Such forms may be found in the genera above named, and one may, by attention to the character of the medium of the organisms, without any difficulty secure such examples as offer the most favourable conditions for investi- gating the distribution iu *hem of the assimilated iron. The specimens of Epistylis, for example, which were used by me for this purpose, .verc obtained from a colonial form attached to the sides and limbs of the common crayfish, and their cyto- plasm gave no immediate reaction for iron. Examples of i 238 A. B. MACALLUM. Stcntor and Paramoccium, in sufficiently large numbers, and all but completely free from inorganic iron co™Pound« were readily obtained. The cytoplasm m Vorticel a oa the other hand, usually contains such compounds, but the.e are very often in the form of granules situated in vacuoles, or at the periphery of the same., a disposition of the compounds which gives every facility for studying the distribution of the assimilated iron. i. 4.„;i „« in In the examples of Epistylis there were, as stated, no u - organic compounds of iron, at least none I'^lf'^'^'f.'l^' in the glycerine and sulphide mixture within the first hour after the application of the reagen. but on the third and fourth day both cvtoplasm and nucleus gave a marked reaction lor iron The latter was, of course, most prominent in the nucleus, in which was revealed, by the dark-green colour, in some exam- pies a granular .trueture, in others a fibrillar arrangement The reaction of the cytoplasm was a diffuse one, with here and there large granules in which it had developed more markedly. The membrane and stalk were, in these cases, free from iron. All these points were more readily observed in preparations treated with sulphuric acid alcohol or with P ange s fluid for twenty-four hours (fig. 28). ., ^ j • In Vorticella a similar distribution of the assimilated iron was observed in both cytoplusm and nucleus, and a diftuse reaction for iron was also obtained in the central or axial por- tion of the stalk, after the preparatioti had been kept in the warm glycerine and sulphide mixture for several days. Ihe reactions are represented in fig. 27, drawn from a preparation which contained inorganic iron compounds disposed in vacuoles In this the central portion of the stalk is shown to be contur ed into a funnel-shaped organ at tne base, which also contauu "masked "iron. I was unable to determine how this organ .vas connected with the cytoplasm. I found no d. Ihculty m obtaining the complete reaction in all the parts at he end of a five days' application of the warm glycerine and sulphide 'Samples of Stentor poiymorphus, free from inorganic IRON COMl'OUNDS IN ANIMAL AXI) Vl^riETAnM'] CELLS. 239 iron compounds, were, after l)eiug hardened in alcohol and after treatment witli ammonium sulphide, isolated from those more or less impregnated with iron salts, the large size of the organisms enahling one to do this readily. One of such, after treatment for fourteen days with the warm glycerine and sul- phide r^ixture, is represented in fig. 25. la this no distinct reaction was obtained during the first two days, definitely showing that no inov^anic iron was present. In the interior of the spherical ek icnts constituting the nucleus there appeared eventually a diffuse iron reaction, as well as one localised in granules, and the cytoplasm gave a diffuse reaction like that given by the cytoplasm in E pis ty lis and V or micella. I do not think that in this case the reaction had developed to the fullest extent of which it was capable, for I found other examples in which the nuclear and cytoplasmic elements gave a moi'e intense one ; but it is usually difficult in such large cells to obtain the best effects of the reagent, since in two weeks' time it is apt to undergo decomposition, when the development of the iron reaction ceases. In order to ascer- tain how abundant the assimilated iron is, I employed acid alcohols to liberate it, and, after the remoA'al of the acid, treated the preparation with ammonium sulphide. Sul- phuric acid alcohol is the lx;st reagent for the purpose, since with it there is less iron diffused from the parts in which it is liberated; but, in order to get the most exact results, the examples of S ten tor used should be h-ec from inorganic iron compounds, a point of which one may be certain by putting the hardened examples in ammoiiium sulphi !br a few minutes, when, if they pass this test, they may be >Fashed in alcohol to remove all traces of the reagent and placed in the acid alcohol for one or two days. I have represented in fig. 20 an example of S. polymorphus, in the wall of the funnel-shaped oeso- phagus of which was found the only inorganic iron compound present, and in this, after it had been treated as described, ti.e ribbon-like nucleus appeared intensely greenish-black, while the cytoplasm gave a deeper reaction than was obtained in any specimen simply by prolonged treatment of it with the warm VOL. 38, I'AKT 2. — NEW SEIi. *4 A. 11. MACALLUM. I glycerine and sulphide mixture. In examples absolutely free from inorganic iron compounds the reaction in the cytoplasm and nucleus was as marked as that represented in the figure. The method is, of course, open to the objection that it may permit a diffusion of the liberated iron from the nucleus to the cytoplasm, but that the latter contains assimilated iron is shown by prolonged treatment with the warm glycerine and sulphide reagent. In examples of different species of Paramoecium, the cytoplasm, which gave no reaction for inorganic iron, mani- fested with the warm glycerine and sulphide reagent after ten days a reaction as distinct as that obtained under similar conditions in the cytoplasm of Stentor, Vorticella, and Epistylis. These organisms were the only ones in which the micro-nucleus was revealed by the iron reaction, and the latter appeared to me to develop more slowly than that in the macro-nucleus; but the explanation for this may be that the large quantity of chromatin in the latter renders a reaction of any degree of intensity obtaining in it much more proh.i- nent than a reaction of a similar intensity w:)uld a[)pear in the mieru-nucleus. In both the reaction •• ..s almost wholly confined to the granules and fib'-illar elements. All the forms of Protozoa studied illustrated the fact so prominently indicated in the organisms referred to above, that ixa assimilated compound of iron is a constant element in their cytoplasm. It is probable that this compound belongs to the chromatin class, for the cytoplasm in Protozoan orsianisms geiierallv stains much more readilv, and bolds the dyes more tenaciously, than the cytoplasm in higher organisms does. In support of this may be urged other facts. I pointed out, when dealing with the relations of assimilated iron compounds to the ferment-forming cells in Vertebrates, that the sul)stance which elaborates the ferment, or out of which it is prepared, contains iron and acts towards staining reagents like chromatin. Digestion in Protuzoa is, in all probability, effected by ferments derived, as in higher forms, from the eytoplasri!, and it is only reasonable to suppose tluit IRON COMPOUNDS IN ANI^rAL AND Vl'^GETABI^K CELLS. 241 the antecedent of the ferments is, in this class also, an iron- holding chromatin.^ Euglena viridis is a form whose position, whether as a vegetable or as an animal organism, has not by any means been definitely determined, but the distribution of assimilated iron in its interior appears to indicate that if it does not be- long to the animal kingdom, its physiological processes pos- sibly resemble those of the Protozoan cell, and it is for this reason that I deal with it in this place. Examples of this organism free from inorganic compounds of iron may be ob- tained readily, and when hardened in alco':.oi, they may be subjected to the action of the glycerine and sulphide mixture for twenty-four hours, without manifesting a react'on for iron, but when the application is extended for three day.^ or longer, a reaction for iron is obtained in the nucleus and cytoplasm. The chromatin network is usually so afl'ected by the reagent that its nodal points only manifest the reaction, while the nucleolus exhibits a less intense dark-green colour. The cytoplasmic trabccuUe separating the " amylaceous " corpuscles from each other develop a dark-green reaction, which is found to be most intense at the nodal points. All these features are more clearly seen in specimens whicli have been hardened in alcohol, then treated for two days with sulphuric acid alcohol, and finally, after being acted on with the acid ferrocyanide mixture to produce the Prussian blue reaction, mounted in balsam (fig. 41)). In these preparations the iron revealed in the cytoplasm is most abundant in its nodal points, which, with the reticulum ol the nucleus, are thereby rendered most pro- minent. The nucleolus, separated from the other elements by a clear zone, in which the light I)lue observed is derived from tlie nuclear elements above and below the foca' plane, gives a less intense reaction than one of the niucli smaller nodal points of the nuclear network. If the piopu- Uon has also been stained with eosin tlie nucleolus alone appears to be ' The ferment or ferments, acconling to M. Greenwood ( ' .Tournal of Tliysio- logy,' vol. viii, 1887, p. 2G3), jjuss into the UuIlI surrounding the ingested mutter. I 242 A. T,. MACALLU^r. ■ markedly affected by it, exhibiting an ochre-red colour so characteristic of the nucleoli in the hepatic cells of Necturus after similar treatment. Safranin leaves the nucleolus un- affected, but colours deeply the chromatin network and the iron-holding portions of the cytoplasm. When, however, the organism has been hardened in picric acid, the nucleolus exhibits no affinity for eosin, while it colours as deeply as the chromatin network does with hpcmatoxyliu and picro-carmine. From this it would appear as if the nucleolus were intermediate in composition between the nucleolus of higher animal cells and the chromatin of the nuclear reticulum. The occurrence of assimilated iron in the cytoplasm of Euglena viridis, it it is not cliemically associated with the chlorophyll present, appears to indicate that the organism is closely related to the Protozoa, in common with which it has other characters. 1 If the view, that the assimilated iron in the cytoplasm of Protozoa is part of the antecedents of the zymogenic compounds of these organisms, is correct, it would explain the phenoironon in Euglena in which the presence of a short digestive " tract" also postulates, to a certain extent, the occurrence of processes of nutrition belonging to the animal type. Fungi. — The presence of nuclei has not yet been demon- strated in a largo number of the Fungi, nor has the occurrence of a subctance similar to the chromatin of other organisms been determined with any degree of ccrtiiinty, except in a few forms ; and, therefore, the question of the occurrence and dis- tributicn of assimilated compounds of iron in the cells of this class is not quite as easy of solution as that dealing with the ' G. Klcbs, who has given special attention to the Eui,'Ieiiac«c (" Organiza- tion einii,'cr Flagcliaten-Gi-niipon und ilire ISezielnuigen zu Al^'en unci Infn- sorien," ' Unfersncli. aus dem Bot. Inst, zu Tiibingen,' lSSl-85), is of the opinion that this gronp should be classed amongst the Protozoa. Khawkinc (" llecherclics biologiques sur I'Astasia occllata, u.s., et I'Euglena viridis. Seconde Pa.tic, L'Euglena viridis." 'Ann. des Sciences Nat., Zoologie,' Serie 7, vol. i, 188(3, p. 319) came to the conclusion, i : a result of experiments, that Euglena takes in organic compounds in the dark, but in daylight assimilates only inorganic coin))ounds. f IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 243 presence of these compounds in higher organisms. I have, however, endeavoured to solve it by the investigation of a few widely different forn^s, and the results now to be described show the presence of " masked " iron compounds similar to those found in all the higher organisms. These forms com- prise : Saccharomyces cerevisise, S. Ludwigii, Hy- phelia terrestris Fries^a leucosporous AgaricinCj Cystopus candidus, and Aspergillus glaucus. The question of the occurrence of a nucleus in Saccha- romyces bears upon that relating to the presence of iron- containing chromatin-like substances in this genus ; and, con- sequently, it is necessary to give an account of the various observations that have been made on this subject. The earlier botanists, Nageli^ and Schleiden,- claimed that they had found a nucleus in the yeast-cell, and the later observers, Schmitz,"' Strasburgcr,* Zalewski,^ and Zachariao," have maintained that it exists, while Zimmerman 7 speaks reservedly on the question. Raum*^ found in yeast-cells which had been fixed on the cover-glass by heat or by solutions of corrosive sublimate, and stained, first with warm methylene- blue and then with bismarck brown, black spherical granules, varying in number from one to fifteen, in a more or less brown-tinted protoplasm. These were usually arranged in the » ' Zeit. fih- wiss. Botaiiik,' vol. i, p. 45. lleferonce in Raum's paper. 2 'Giuudziige der wiss. Botanik,' 1849, p. 207. llcfened to by llaum. 3 " Uutcrsucliungcn iibcr dtn ZcUkeni dor Tliallopiiyten," ' Sitzuiigsber. der Niedcnlu'in, Gcscll. fiii- Natur- und Ilcilkundc zu Bouii,' Sitzuiig. am 4 Auj?., 1879. ■• 'Das Botaiiisclie Tract icuni,' p. 3119, 1SS7. * "On Spore Formation in Yeast Cells," 'Transactions of the Scientific Academy of Cracow' (Bolisb), ISSG. Abstract in 'Bot. Cenlralb!.,' vol. XXV, p. 1. « "Beitriige zur Kenntniss des Zcllkcrns und der Sexualzellcn," 'Bot. Zeitnnfr,' 1SS7, Nos. 18-24. 7 "Die Morpbologie und riiysiologie der Pflanzenzelle," Breslau, 1887, p, 25. 1 have not had access to this publication, and my attention was first called to it by a reference made by liaum. 8 " Zur Morphologic uud Physiologic der Spross-^ilze," ' Zeit. fiir Hygiene,' 1891, vol. X, p. 1. 244 A. B. MACALLUM. form of a circle or of a segment of a circle at either pole of the oval cell, and there was no relation between their size and that of the cell containing them, although they appeared to have some connection with the budding process, since he observed them undergoing transference to the protoplasm of the bud. What the nature of these granules is Raum does not say, but the ?'osults of his experiments would seem to indicate that they are not formed of nuclein, for on submitting the yeast-cells to digestion witli an artificial gastric fluid at a temperature of 40° C. for one or two days, and afterwards on washing with ether and alcohol, every trace of the granules had vanished. Nuclein is undoubtedly present in yeast-cells, and Baum prepared some of it from this source, which he mounted in egg-albumen on a cover-glass, and stained, first with methylene blue and afterwards with bismarck brown, when he found that the nuclein particles took a brownish stain while the albumen appeared light yellow, a reaction in marked contrast with that obtained in the granules of the hardened yeast-cells after the employment of the same staining methods. Raum appears to be doubtful concerning the existence of anything resembling a nucleus in the yeast-cell. The more recent observers who claim to have found a nucleus in the yeast-cell are Moller aud Janssens. The former^ found in older yeast-cells a spherical corpuscle which he regards as a nucleus, but without a membrane or nucleolus. This changes its shape readily, and therefore its position in the cell varies. Owing to this property, it is capable of assuming a thread-like form when budding occurs, a portion of it being thus enabled to pass into the protoplasm of the bud through the narrow tube which connects the mother and daughter elements. The part in the latter eventually breaks off, and both portions become splierical. Janssens,^ who used ' "Ueberden Zellkern und die Sporen der Hefe," 'Centralbl. fiir Bakf. und Parasitenkunde,' vol. xii, IS'J-i, p. 537; also " Weitere Mittlieilungen iiber den Zellkeru und die Spiosse der llefe," ibid., 1893, vol. xiv, p. 358. - " Beitriige zu der Fragc iiber den Kern der Hefezelle," 'Centralbl. fiir Bakt. und I'arasitenkunde,' vol. xiii, 1893, p. C39. IRON f^OMPOUNDS IN ANIMAL AND VEGKTABLE CELLS. 245 in hi!^^ investigations the species S. cerevisise, S. Ludwigii, and S. PastorianuSj states that he found in tlic two former a nuclc'is provided with a membrane and a nucleolus^ the latter spherical and homogeneous and of a diameter one third that of the nucleus. The remaining portion of each cell is occupied by a cytoplasmic network with fine meshes, whose nodal points readily absorb colouring matters, and, in the opinion of Janssens, constitute the granules of llauin. He claims to have observed mitotic stages of the nucleus, which obtain when budding commences and when spore formation occurs. Two observers only, Briicke^ and Krasser,^ have denied the existence of a nucleus in the ycast-cell. Krasser in his later publication asserts that the body described by Mijller as a nucleu-i is not such an organ, and he found, after employing Mci -r s methods on beer yeast-cells, that the latter possessed no body like the one described by that observer. He further observed that the bodies described by Mdller as nuclei, after being submitted to digestion with artificial gastric juice, gave no evidence of the presence of nuclein. The occurrence of the latter substance in yeast-cells, which is readily demonstrable in a macro-chemical way, Krasser attempted to show micro- chemically, and, after many failures, succeeded in finding it in a few specimens in the form of granules at the side of the body regarded by Miiller as a nucleus. I have followed the methods of hardening and staining adopted by Mciller, for the purpose of ascertaining the nature of the body considered by him to be a nucleus^ and have compared the results thus obtained with those found in yeast- cells after hardening the latter in satui'ated solutions of cor- rosive sublimate and staining them with hoematoxylin and eosin. I have also used Flemming's fluid for hardening, and stained preparations so made with safranin. Holler's methods certainly do reveal, now and then, a structure like that which ' "Die Elementarorganismen," 'Sitzungsber. tier K. Akad. d. Wiss. zu Wien, Mat!i..Nat. Classe,' ISGl, vol. xliv, Abth. 2. ' "Ueber das angebliclie Vorkommen eiiies Zellkerns in den Hefezelleii," 'Ocsterreicb. Bot. Zeits.,' 1885, No. 11; also " Ueber den Zellkern der Ilcfc," ibid., 1893, p. It. 246 A. n. MACALLUM. he took to be a nucleus, but this body, when hardened with eorrosive sublimate, stains with eosin but not with hsema- toxylin, wliile after fixation with Flemming's fluid it appears to have no particular affinity for any dye. On the other hand, in S. Ludwigii, as it usually develops in the sap of the iron-wood tree (Ostrya virginica), there is in the great majority of cells a corpuscle which corresponds with the " nucleus " of Moller. This structure is round, homogeneous, and in diameter sometimes more, sometimes less, than half the length of the shorter axis of the cell, in the centre of which it is usually placed, and after being hardened with corrosive sublimate it exhibits a special affinity for eosin, but none for hseraatoxylin, while it acts like the cytoplasm towards safranin. In preparations madr with Flemming's fluid the results were practically the same, and therefore not indicating on the part of the body in question the possession of a substance in all points like chromatin. A substance like chromatin appears to be distributed through the cytoplasm. In S. cerevisise, after being hard- ened with corrosive sublimate, the cytoplasm takes, when treated with hscmatoxylin (Delafield's and Ehrlich's), a blue- violet tinge. With favourable illumination and apochromatic objectives, the stain is found to be localised in the trabeculai of the cytoplasmic network, and, v. here the vesicular character of the cytoplasm appears pronounced, all the cytoplasm, except the contents of the vesicles, is coloured. In some of the cells granules were observed with a stain slightly deeper than that of the eytonlasm, and similar elements were found in cells liardened with Flcmming's fluid and stained with safranin. These, possibly, are those described by Ilaum. In S. Ludwigii the cell is usually very much larger, and the structure and staining reactions are, therefore, much more distinct. In this form, when hardened with corrosive sublimate and stained with hnematoxylin, the vesicular structure of the cytoplasm comes out quite markedly through its blue-violet stain, which also is found now and then to characterise prominently gran- ules in the cytoplasm between the vesicles. The granules of f IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 247 Raum are, however, much more common elements than these, and are to all appearances quite different structures, as is apparent in ordinary cover-glass preparations made after Raum's methods. Tlic larger examples of the granules of Raum seem to be less abundant in corrosive sublimate prepa- rations stained with hasmatoxylin and eosin. From these results I am inclined to regard the existence of a nucleus in the yeast-cell, in its usual conditio:', as extremely doubtful, and, on the other hand, to support Krasser's con- tention that nuclein is disseminated through the cytoplasm. Whether, in other stages, as, for example, those in which spore formation occurs, there is a nucleus I cannot say, but there appears in the ordinary stages of the organism to be nothing which may be looked upon as a specialised chroraatin-holding structure. These conclusions are, on the whole, confirmed by the results of experiments made to determine the distribution of assimilated compounds of iron in these organisms. When specimens of S. cerevisia;, hardened in alcohol, are subjected to the action of the glycerine and sulphide mixture at a tem- perature of G0° C. for several days, their cytoplasm acquires a greenish tint. Sometimes, however, the latter reaction may not appear except in a few grannies scattered through the cytoplasm (fig. 4). On account of the small size of the cells and of the alteration produced in them by the reagent, one cannot definitely determine whether the granules correspond to those described by Raum. When the cells have been sub- jected to the action of sulphuric acid alcohol, the subsequent application of an acid fcrrocyanide solution gives their cyto- plasm a faint blue colour, which is more distinct and deeper when the light transmitted passes through several cells in succession. Blue granules are sometimes observed in such preparations. It is in specimens of S. Ludwigii that one obtains the clearest evidence of the occurrence of an assimilated iron compound. In these, after being hardened in alcohol, the glycerine and sulphide miyture eventually gives results like 248 A. n. AFAnALlJJM. I those represented in fig. 5. The differences observed appear to depend on the cytoplasmic structure in the specimen ex- amined. Wlien tlicre are a few large vesicles in the cell, the iron-holding substance seems to be, in great part, at their peri- pheries. This disposition also obtains in the buds. The remaining portion of the cytoplasm in each element is very slightly coloured greenish, but whether that is due to ferrous sulphide is uncertain. When, on the other hand, the cells are markedly vcsiculated, the glycerine and sulphide mixture gives the cytoplasm between the vesicles a distinct reaction for iron. In the majority of such cells there are one or more large spherical .elements, which, in the glycerine and sulphide mix- ture, after the third or fourth day appear dark green, much more so than does the surrounding cytoplasm. Tliey are homogeneous, manifesting a uniform reaction throughout their substance, and their position is, if not in the centre of the cell, at least in that neighbourhood ; but smaller granules of the same character may be more remotely situated. From their position, size, and shape, they would appear to be the bodies which, in preparations made with corrosive snblimate, hocmatoxylin, and eosin, stain exclusively with the latter reagent. In cells which are treated with acid alcohol, then with an acid ferrocyanide solution, and finally, after being stained with eosin, mounted in balsam, similar bodies are given a violet tint, while the cytoplasm is coloured bluish, the violet being undoubtedly due to a combination of the Prussian blue colour with the eosin stain. As the granules of Raum are not specially selected by eosin, it would appear that the iron-containing body observed does not belong to that class. It is thus seen that in S. cerevisijc the assimilated iron is, like the substance which absorbs hsematoxylin, distributed through the cytoplasm and sometimes also in the latter in the form of granules, but in S. Ludwigii it may be chiefly found at the periphery of each large vesicle when only a kw vesicles are present, while in those cells in which the whole of the cytoplasm is vesieulated, the latter gives a uniform reaction for iron corresponding in its depth with that given by ha;ma- IRON COMPOUNDS IN ANIMAL AND VEGETAnr,E CELLS. 249 1 toxylin. Furtlicr, there is a substance which constitutes cor- puscles of a nucleolar character in cells of this form, which stains with eosin and gives a marked reaction for iron, but diflcring from the chromatin substance in remaining unstained after treatment with hreraatoxylin. There is no nucleusj although such an organ may occur in other stages, especially in S. Ludwigii,^ When the mycelial threads and hyphse of Ilyphclia tcr- restris, Fries, are hardened in alcohol and stained with hacma- toxylin, the cytoplasm generally is coloured, but it is specially affocted by the stain in the terminal portions of the hypliEC on which the elements of fructification arc developing. One can find also, in such preparations, deeply-stained granules scattered in the cytoplasm of the hyphse, and at times also a vesicular cavity and a membrane enclosing these granules, which then simulate nucleoli. Sometimes such structures strongly rc- serble nuclei, and mitotic conditions are suggested by the presence of pairs of rows of deci)ly-coloured granules placed opposite, and at a very short distance from, each other. In the fully-formed fructification these vesicular cavities and their deeply-stained granules may be most readily seen. Whether such structures are nuclei in the proper sense of the term it is difficult to say, but if they are, they contain only a small por- tion of what may be considered as the chromatin, which is diffused in the cytoplasm of the mycelial threads in the younger stages, but appears to be transferred to the hypliai Avhen the fructification of the latter commences. "When the latter stage is fully attained the mycelia and I' .ver portions of the hyphse are found to have little or no cytoplasm and to stain very feebly, a result quite different from that obtained in the fructifi- cation. The distribution of the " masked " iron in this form is found to coincide very closely with the distribution of the stainable substance. In the simplest form of the hypha, the glycerine ' Ludwig ('Lehrbuch der iiiederen Kryptogamen,' 1892, p, 201) appears to regard S. Ludwigii as merely a stage in the development of Endomyces Magiuisii. 250 A. n. MAOALLUM. and sulphide mixture gives iu twe.ity-four hours a reaction like that represented in fig. 13 a, while iu the slightly more developed structure the reaction is deeper with large dark- green granules (fig. 13 b). A similar result is obtained in the hyphse which terminate iu two, three, or more pear-ihaped outgrowths (fig. 12). In the hyphte below the fructification the cytoplasm is of a vesicular character, the walla of the vesicles being formed of an iron-holding substance, and as the terminal element develops, the vesicular character becomes less marked and the iron reaction less distinct, so that, finally, no iron may be found in this part of the filament. At the same time the granules in the fructification become moio numerous, larger, and manifest a deep reaction for iron (fig 11). These granules are then found to be situated iu small vesicles very much like the vesicles which, in hajma- toxylin preparations, resemble nuclei. The granules revealed by the iron reactirn are the same as those indicated by the ha3matox;'lin sta.u. This is also true of the granules in the younger hypliae. The cytoplasm of the mycelial threads is, at this stage, free from "masked" compounds of iron, but in the earliest stages the mycelial threads give at once, on the applica- tion of the glyceri '0 and sulphide mixture, a slight reaction for iron, which, however, becomes deeper at the end of twenty- four hours if heat be applied, this indicating the presence of "masked" iron. Granules iu the cytoplasm along the course of the threads give a marked reaction for the metal like that manifested in the hypha;. It is probable that the absence of iron in the later stages of the threads may be due to the transference of the iron-holding compound to the hyphse. The question concerning the occurrence of nuclei in the Hymenomycetes has been dealt with by Strasburger,^ Rosen- vii'ge,^ and Wager.^ The two former describe them as obtaiu- > 'Das Botanische Practicnm,' pp. 301 and 433, 1887. ^ " Sur les noyaux des liymenomycctes," « Annales des Sciences Nat., Bot.,' 188G, Serie 7, vol. iii, p. 75. 3 "Oil the Nuclei of tlie Hymenomycetes," 'Annals of Botany,' 1892, vol. vi, p. 140. IRON COMPOUNDS IN ANIMAL AND VRGETAnLK OKLLS. 251 ing in the liyplia', in the busidia, and in the spores .f the various species, in the form of small elements wiiiclj are brought into view only wlien alcoholic mat ,ial is acted on by very dilute solutions of hipmatoxylin. Their number in a hypha varies, but in each basidium there is at first only one, which, when the sterigmata are being formed, divides, the daughter nuclei undergoing division also, sometimes a second time, each of the four or eight t'lUs resulting passing through the tubes of the sterigmata into the spores at the end of the latter. When the spores are mature they thus contain, according to the species, one or two very minute nuclei, while the basidia at this stage contain none. Wager also found nuclei in the basidia, but maintains that the spores do not contain any until after the formation of the thick spore- men^brane. It i? an easy matter to demonstrate in the hyphfc and some- times in the Lasidia and in the matu spores of leucosporous Hymenomycetes,^ the structures regarded by Strasburger and Jlosenvinge as nuclei, but, as was the case in Ilyphelia ter- restris, such elements contain only a small portion of the ehro- mophilous substance, for when preparations are made, as recom- mended by StrasL rger, with very dilute solutions of hsema- toxyliii, the cytoplasm also stains though not quite so deeply as the minute nuclei, especially in young hypha;. This and other staining reactions indicate that chromatin is dissolved in the cytoplasm, a conclusion borne out by the results of experi- ments with the glycerine and sulphide mixture and with acid alcohols, in which case the hyphal elements of a very young stage of growth give a reaction for iron diffused throughout the cytoplasm, but when the spores are formed the hyphal cells and their shrunken nuclei rarely give a reaction for iron. At this stage also, in sections of the lainella3, a reaction for iron is obtained in the hymenium and in the spores, while the hyphal elements of the " trama '" appear free from the metal. If the spores and the basidia are teased out and mounted in the The pigment la the spores of the other divisions of the Hymenon-.ycetes greatly obscure the reaction obtained with the glycerine and sulphide mixture, 252 MACALLUiM. I glyceilne and sulphide mixture, the applicatiou of heat to t!;e preparation for a week will bring out appearances in the isolated elements like those represented in fig. 10. The most prominent feature in these is that the cytoplasm in both classes of structures contains " masked " iron. When the bodies regarded by Strasburger and llosenvinge as nuclei were observed, they manifested a slightly deeper reaction for iron than the cytoplasm generally, but no structure was detected in them end they appeared as large granules rather than nuclei. The most marked icaction for iron was obtained in tlie spores in which a cytoplasmic re'iculum was thus deraonstrated. When, however, the spores are provided with a thick mem- brane, a reaction with the glycerine and sulphide mixture does not a])pear, but is obtained after the use of acid alcohols. As a rule, the reaction is uniform throughout the cytoplasm of the basidia. There are, however, constituents of the hynienium occasicually observed in which no iron was found. They possessed no stcrigmata or spores, and from their association with the basidia I was inclined to regard them ai paraphyses, but from the comparative scarcity of such elements free from, or poor in iron, they can scarcely be Inni- li upon as belonging necessarily to that class, which in stained preparations is abundantly represented. The subhymcnial cells also give a faint reaction for iron. It thus aj)pcars that in the leucosporous Ilymcuomycetcs the cytoplasm of the hyplue in the early stages of the fungus con- tains iron, which is also present in the minute " nuclei," and that in later stages this cytoplasm gives a faint reaction or none at all for iron, while the cytoplasm of the basidia and i,pores contains enough "masked" iron to give a marked reaction. This distribution of the iron corresponds with the distribution of the stainable substance, and it may, therefore, be fairly concluded that the chromatin is here also iron- holding. In my earlier communication reference vas made to the occurrence of an iron-containing substance in the gonidia of Cyst op us candid us, and I stated that the iron compound lUON COMPOUNDS IN ANIMAL AND VEGKTABI^E CELLS. 253 was found to be localised in spherical eleraenta of I'Gfi dia- meter, corresponding to the nuclei of the zoogonidia. I have, since that date, investigated the cytological charactei of this organism, and have found that though there are, as Fiscli,^ Wager," and others have observed, nuclei in the mycelia and in the gonidia, the whole of the protoplasm, except in the mature gonidia, is chroraophilous, that is, it contains chro- matin. The nuclei are, indeed, of the more regular form in the mature gonidia, but in the mycelia amongst the cells of the liost (Capsella bursa-pastoris) they are chiefly, if not wholly, small masses of chromatin, like those forming the " nucleoli "' in the abjointing gonidia. I have not succeeded in finding the mitotic jjhase either in the mycelia or in tlie developing gonidia, although I have carefully looked for such in a large number of preparations. The disposition of the assimilated iron corresponds closely with the distribution of the chromophilous substance in this form. The cytoplasm of the haustoria and of the mycelia ijtive a marked reaction for iron in all the methods of demon- stration.3 The mycelial membr;ine gave no evidence of the presence of the element. The small masses of chromatin were found to be rich in organic iron. In the terminal enlarged, sometimes club-shaped, sometimes truncated, portion of eacli liyplia the iron was found to be in a localised as well as in a diffuse form. The " nuclc li " gave abundant evidence of its presence, these structures thus appearing in marked contrast with the remaining portions of the nucleus, which contain relatively less iron than the surrounding cytoplasm in this stage. Tit the subsequent development of the abjointed gonidia, the nuclei appear to take u). tVom the cytoplasm all, or nearly • "Ueber das Verhalten der Zcllkernein fusionireiuleii Piizi-elien," 'Ver- sainmluu|T dcutsclier Naturfoischer uiul Acizte in Slrassburg,' 1885. This paper I have not seen, and tlin only references to it tiiat 1 can fnul arc tliose made by Waj^tr and Dangeard ('Coniptcs J^endns,' cxi, 1890, p. 382), '' "Observations on tlie Structure of Cystopus candidus," 'Rep. Brit. Ass. for the Adv. of Science,' 1892, p, 777. ^ Tlie material was hardened in alcoiiol, wiiicii was renewed until every trace of cliloropliyll was removed fioni the tissues. 254 A. B. MACALI.TJM. I all, of the iron-holding substcince, and with this the character of the nuclei seems to change. Tlie " nucleoli," first of all, are converted into fine granules distributed through the nuclear cavity, and, finally, in the mature gonidia the nuclei appear, in the glycerine and sulphide preparations, to be simply more or less homogeneous masses of iron-holding sub- stance, while the cytoplasm does not contain a trace of the metal (fig. G/). In Aspergillus glaucus the cytoplasm of the young mycelia and the gonidiopi.orcs, especially their globular ends, absorbs staining matters readily, but it contains also, scat- tered through it, granules of a nucleolar character, which, in very dilute solutions of heeraatoxylin, applied for twenty- four hours or more, stain deeply. The cytoplasm of the stcrigmata and of the immature gonidia is similarly aflected. In the mature gonidia hasmatoxylin selects large granules which are distributed through the cytoplasm. In what appear to i)e old mycelial threads, the cytoplasm is stained with diffi- culty, while the membrane may be deeply coloired. These results correspond in the main with those obtained in regard to the "masked" iron present. When the warm glycerine and sulphide mixture is applied for about a week, the cyto- plasm of young mycelia gives a difl'use reaction for iron, while a deeper one appears in the large granules referred to as affected by haematoxylin. In the cytoplasm and granules of the gonidiophores a relatively deeper reaction makes its appearance, and a marked one is obtained in the sterigmata. In the immature gonidia the reaction is diffuse, a special one at the same time obtaining in gi'annlcs collected or scattered in the cytoplasm. In mature gonidia the granules are larger, and give a deeper reaction for iron, the cytoplasm otherwise shu'.ving no trace of its presence (fig. 7). The same results are obtained, but more readily, when sulphuric acid alcohol has been employed to liberate the iron present. Bacteria. — The question of the occurrence in bacteria of a substance like the chromatin of more highly developed organisms has been investigated to a certain cxtciit by IRON- COMPOUNDS IN ANIMAL ANTD VEr.ETABLE ORI.LS. 255 Ernst.i Babes,3 Wahrlich,'' Butsohli/ Trambusti and Galeotti.*- Ernst found in a large number of species of bacteria granules which stain with hsematoxylin and other dyes, while the sur- rounding protoplasm is coloured faintly or not at all. These, which on account of their direct transformation into spores he termed sporogeuous, undergo in their earlier stages solution in artificial gastric juice, but in the more advanced condition resist digestion. From Babes' observations, which agree in the main with those of Ernst, it would appear that the granules which absorb and retain colouring matters and take part ui spore formation, also stand in some relation to the division of the bacterial cell. According to Wahrlich, the protoplasm is formed of two constituents at least, a ground substance of reticular structure resembling linin, and one forming granules distributed in this reticulum, and, owing to its capacity for absorbing and retaining dyes, regarded by him as chromatin. In Bacillus pseudoanthracis the small granules which appear before the spores are formed are constituted of chromatin, and from them is derived the main portion of each spore, while the plastin serves apparently for the construction of the spore membrane. Biitschli found in species of Beggiatoa, Chro- raatium, in Spirochsete serpens, Spirillum undula, Bacterium lineola, and in some CyanophycetC, a faintly stain- able peripheral portion, and a central body readily stainable, iu which a honey-comb structure (Wabenbau) was distinctly ' "Ueber den Bacillus xerosis und soiue Sporenbildung," 'Zcit. fiir Hygiene/ vol. iv, p. 25, 188S; riso "Ueber Kciu- un-^ ;:poreiibiidung in Buctcrien," ibid., vol. v, p. 428, 1889. - "Ueber isolirte, filibbarc AuUieile von BaLterieu." • .d., vol. v. p. 173, 1889. ^ "Bacteriological Stu-'ics." Reprinted fron ' ccupta Botanica,' vol. iii, St. Petersburg, 1890-91. 1 have not seen tlii;: v/cvk, and the repiTscntation of Wahrlich's observations and views i.s taken from 'Bot. Centra!.,' vol. xlix, 1892. * 'Ueber den Bau der Lukljrien und verwuudtcr Organisnicn,' Lei'/zig, 1890. ^ "Neuer Beitrag zuni Studium der innercn Slruklur dor Baktcricn," ' Centralbl. fiir Bakt. und rurasitenkundo,' \A. xi, p. 71", 1892. VOL. 38, .;aRt 2 'kw sfr. b 256 A. B. MACALLUM. seen. The central body is, in Biitschli's opinion, a nucleus. In or on this organ were observed granules which became red after treatment with hsematoxylin, and were identified with the granules described by Ernst. Trambusti and Galeotti found in one stage of a very large bacillus isolated from drinking water, that the whole of the protoplasm stained uni- formly and deeply with safrauiu, while in a later stage of the same the stainable substance was converted into granules, dis- posed at the periphery and arranged in the form of a garland of oval outline. The granules eventually fused to form a homogeneous garland out of which arose from three to four elliptical rings, at lirat connected by their ends, but afterwards independent of each other, and in this condition became free. These changes the observers regard as analogous to those of mitosis in tlie cells of more highly specialised organisms. Schottelius^ and Ilkewicz- have described structures in the bacterial cell which they regard as nuclei, and Sjobring ^ claims to have found many of the phenomena of mitosis, as it obtains in the cells of higher organisms, exemplified in bacteria. The results of these observor'^ njipear to me to have been due to defective methods of tec .. \ I find that in Bacillus ^ is, B. anthracis, B. mega- therium, B. tuberculosis, . in the root bacillus, there are granules like Lhose described by Ernst and Babes, and which stain with hajmatoxylin, and in B. pseudosubtilis (?), in which there is only one granule to each rodlet, each granule is developed into a spore, the remaining protoplasm at the sani'^ time losing all its affinity for colouring matters. The struc- tures observed are the same whether alcohol, corrosive subli- mate, or heat has been employed for their fixation. Ernst iV)und, as already stated, that the granules, except in the later stages, undergo solution in artificial gastric juice. 1 " Bcobaclitung Kcrnartiger Krirpcr im Inncni von Spaitpili'.cu," ' Centralbl. fur Bald, uud rarasiteiikumie,' vol. iv, 1888, p. 705. - " Ucbcr die Kcnii; dor Milzbraiidt.poreu," ibid., vol. xv, p. 201, ]S94. » "Ucber Kerne uud Tlicilungeu bui den Bakteritn," ibid,, vol.'xi, n 05 1893. ' IKON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 257 This would seem to indicate that they are not constituted of typical chromatin.' I have endeavoured to determine whether they contain iron in a " masked " form ; but the results of ray experiments; except in the case of B. megatherium, have not been decided enough to permit a general conclusion on this point. The organisms are very small, and their size would postulate the occurrence of a very small amount of iron in them., and even in the larger spores. "When, therefore, a cover-glass preparation of B. megatherium is treated wiUi sulphuric acid alcohol for twenty-four hours, it is not surpris- ing that the subsequent treatment with an acid ferroeyanide solution should give but a very faint blue reaction. When the granules referred to wero under observation they manifested themselves by a blue colour slightly deeper than that apparent in the rest of the protoplasm of the organism. In B. sul)tilis the granules are the only parts of the bacillus which appear to contain iron, the reaction for which is very faint. I have in none of these form? obtained a reaction with the glycerine and sulphide mixture distinct enough to permit certainty of opinion in regard to this. Sulphate- of iron, when present in very minute quantities in preparations, appears loss distinct than the same amount of iron when revealed by the Prussian blue reaction, and on this account the uppai'cnt absence of the sulphide reaction determines nothing. In some preparations of B. pseudosubtilis the largest granules and the spores gave, after treatment with acid alcohols, a blue reaction with the acid ferroeyanide mixture. The root bacillus gave fre- quently a diffuse and faint blue reaction under the same condi- tions. It is obvious that these organisms are too minute to furnish results which would allow the question, whether they contain " masked " iron, and how it is distributed, to be definitely and decisively answered, and I had to employ other forms, of such a size that no difficulty would be experienced in this respect. ' Vandevelde (" Studieii zum Clieuiie dcs Bacillus subtilis," 'Zeit. fiir Pliysiol. Cheniie,' vol. viii, \i. 307, ISSl) btatcs, that he has isolated nucleia from B. subtilis. ids-, 258 A. n. MACALLUAI. The most readily accessible form was Bcggiatoa alba. This organism, as is well kuown, manifests itself in five different conditions : long threads composed of cells of varying lengths, shorter filaments also formed of cells usuuUy free and motile, spirillum-like elements, comma-shaped, two-celled, swarming bodies, and simple "cocci." Cover-glass preparations of all these forms, fixed first with heat and subsequently with alcohol, were subjected for about two weeks to the action of the glycerine and sulphide mixture at 00° C, while like prepa- rations were treated with sulphuric acid alcohol for about two hours at a temperature of 30° C. The results of both methods agreed. In the long threads the abundance of sulphur granules causes the cytoplasm to have a reticular, or more properly speaking, a vesicular appearance, brought out very prominently when the glycerine and sulphide mixture has dissolved out the sulphur and at the same time given the cytoplasm a greenish colour, developing into a faint blue on treatment with an acid ferr^cyanide mixture. At times the greenish or the blue re- action appears most prominently in some of the nodal points of the " network," but this is doubtless due to the fact that more of the cytoplasm is condensed at such points. The shorter free, motile filaments, which contain, as a rule, very many fewer sulphur granules, have a more homogeneous cyto- plasm, and in these the reaction for " masked" iron obtained was a diffuse one. A similar result was obtained in the examples of the spirillum form. In the comma-shaped forms the reaction obtained was, as a rule, slightly deeper, and it frequently appeared most markedly in the central portions of each of the two cells. In some examples a granule in this central mass gave a marked reaction for iron. I did not suc- ceed in determining the relations of the iron in the " cocci.'' So far as these results go they correspond with those ob- tained M'hen cover-glass preparations of Beggiatoa alba are stained with ha;matoxylin, which colours diffus'jly the cytoplasm in all the forms, but rarely reveals the existence of sjiecial chromatin elements. I have been unable to determine, except iu a few comma-shaped elements, the occurrence of die denser IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 259 central portion described by Biitschli, and I am inclined to re- gard the structure observed in the exceptional cases as due to shrinkage caused by the method of preparation. In some of the comma-sha'^ed elements the hsematoxyliu stain demon- strates granules like those which were observed to manifest an iron reaction in the glycerine and sulphide preparations. The use of Lciffler's solution of methylene blue, followed by tliat of a saturated solution of bisraarck brown, as recommended bv Ernst, stains similar granules in the " comma " elementSj and in a few of the spirilla only ; but it is doubtful if these may be classed with the " sporogenous " granules of other bacteria revealed in the same way. I have not found that there are any granules in the spirilla which contain " masked" iron, although there is the possibility that spirilla, containing granules, were not present in the preparations made with the glyce.'ine and sulphide mixture or witli acid alcohol. The diffusion of the "masked" iron throughout the cyto- plasm of Beggiatoa corresponds, on the whole, with what was observed in the other bacteria, but the interpretation of the results in the latter has an element of obscurity in it. It is evident that the iron-holding compound is not, as a rule, localised in granules or in special structures; and although the distribution of this compound, in IBcggiatoa alba at least, corresponds with that of the substance which stains with hicmatoxylin and other dyes, it is uncertain whetlier the two compounds are identical. It is possible that experiments with some of the larger forms, as, e. g., Beggiatoa mirabilis or Crenothrix Kiihniana, may result in determining a solu- tion of the question. Unfortunately I had no opportunity of studying the distribution of iron in such large forms. I did, however, obtain a few preparations of a form which is possibly allied to Crenothrix, and whose size (2-8— 3-2 ju x 6-4, — 8 ft) rendered it favourable for such observations as I had an opportunity of making. This organism grew on the surface of some sewage water collected in '''C fal' of 1893, in which also myriads of examples of Euglena viridis throv3. It multiplied by fission. Some of them exhibited rounded 260 A. B. MACALLCJM. ends, while others had an oval shape, but the majority were cylindrical with flat end-surfaces. Several cover-glass prepara- tions of the organisms, fixed by heat and subsequently placed in alcohol, were made, but no cultivations were attempted, since before its value for the purpose of this investigation was ascertained, the original culture fluid had been thrown away. One of the cover preparations was subjected to the prolonged action of the glycerine and sulphide mixture, but, as sometimes happened in other cases, no result was obtained. The other two were placed in sulphuric acid alcohol for about eight hours at a temperature of about 25° C, and then treated in the usual way with the acid fcrroeyanide mixture. One of the prepara- tions was also stained with cosin, and both were, after beinir washed in water, dehydrated with alcohol and mounted in balsam. Examples of the organism exhibiting the Prussian blue reaction and the eosin stain are represented in fig. 53. The eosin reveals a large central body, sometimes of irregular shape, and always lying free in a cavity in the markedly reticular cytoplasm. The body in question contains no iron, but in other respects resembles the large body present in Saccharomyces Ludwigii. The iron demonstrated appears to be in a granular form distributed in the trabeculse of the cytoplasm, though sometimes a very large granule, richly supplied with iron, Avas found adjacent to, or in contact with, the large central body destitute of iron. As inorganic iron is a constituent of the sheath and other parts in species of Crenothrix and allied forms (C Kuhni- ana, Leptothrix ochracea and Cladothrix diehotoma), it is possible that all of the iron observed in the form described, and whose relationship to Crenothrix has been suggested, was not derived from a "masked" compound. The amount of inorganic iron must, however, have been very little, for, in the cover preparations subjected to the prolonged action of the glycerine and sulphide mixture, but a few of the forms gave an immediate reaction for iron. The chief difficulty lies in the fact that through the failure of the last-mentioned IRON COiiruUNDS IN ANIMAL AND VEGETABLE CELLS. 261 metliod of liberating the " masked " iron in this organism, it is uncertain whether the iron demonstrated after the use of sulphuric acid alcohol had the distribution it obtained in the living organism, or in the cytoplasm before it was treated with acid alcohol. Apart from these matters it seems to me quite certain that the results indicate the presence of iron in a " masked " form in this organism. Cyanophycea3. — These organisms, which are generally regarded as closely related to bacteria, offer, on account of their much larger size, fewer and less formidable difficulties to an investigation of the morphological and micro-chemical characters of their cells, and I have, therefore, endeavoured to give a careful attention to the question of the presence of assimilated iron in them. The determination of the relations of the iron compounds in these organisms has entailed also an investigation of the morphology of their cells, and I have, in consequence, obtained a very large number of results, the description of which is beyond the scope of the present paper. These, and a fuller account of the literature of the subject, I propose to detail on a future occasion, and I now deal with the ascertained facts relating to the iron compounds and, in 80 far as morphological characters are associated with these, M'ith the structure of the cells themselves. The literature on the subject of the Cyanophyceae has grown considerably in the last ten years, but as it is only within the last six that improved technical methods have been employed in the investigation of their structure, a short sketch of the more important publications, which have appeared in the latter period, will suffice for present purposes. Zacharias found that the cell is constituted of a coloured peripheral part, and an uucoloured central portion of a reticu- lated or granular structure. In the central portion he observed two substances, one exhibiting the characters of a plastin, the other, which he termed the " central substance " (Central- substanz), varying in amount in the different cells, and re- sembling nuclein in its chemical reactions. In the central portion he found granules destitute of nuclein, and related in 2G2 A. B. MACALLUM, many of their characters to th. .ucleoli of higlily developed vegetable organisms.i lie considers that the central portion differs very greatly from a nucleus, but whether it performs the functions of the latter he is not prepared to say. Butschli,2 whose observations on the structure of bacteria have been already referred to, found one type of structure prevail in both these and the Cyanophyccfe. The cytoplasm IS, according to his view, formed of a faintly stainable peri- pheral zone, and of a denser, deeply staiii.-Me central portion which, in the living Cyanophycese, is ahvuNs uncoloured. Both parts are vesiculated. He found that h.-cmatoxylin colours the cytoplasm blue, while it gives a red stain to granules situated in the central portion, and, in the nodal points of the vesiculated structures, more especially of those of the peripheral zone. These disappear after subjecting the cells to the action of artificial gastric juice, but he nevertheless regards them as chromatin elements, and he looks upon the central portion as a nucleus. Bcides these granules, he found in certain Oscillarije, in the extreme peripheral portions of the cell, and especially adjacent to the transverse cell walls, others which did not stain with hasraatoxyliu, but which exhibited a strong affinity for eosiii. Deinega'' could formulate no conclusion in regard to the presence or absence of a nucleus in these organisms, and also in regard to the nature of the granules, although he is disposed to regard the latter as formed of an isomer of starch. These, of which he found but one species, stain specially with picro' carmine, and dissolve in weak hydrochloric solutions (0-3 per cent.). Passing over the observations of Zukal,^ who appears to ' "Ucber die Zellcu der Cyauophjccen," ' Botanisclic Zeitune.' 1890 JNos. 1 — 5. n> > - Op. cit. J •' Dor gegenwiirtige Zustand unserer Kenntnisse iiber den Zcllinhalt der ri,ycoel.romaceen," ' Bulletin dc la Soc. imper. des Naturalistes de Moscow • aniiec 18'Jl, p. 481. ' ^ "Ueber den Zellinhalt der Schizophylcn." ' Sitzungsber. der K. Acad der Wiss. Wien, 1892, Matb.-Nat. Classe, vol. ci, p. 301. ^. ^, IRON COMPOUNDS IN ANIMAL AND VEGETAULK CELLS. 2G3 regard all the granules as nu ici, the next investigator of this subject is Ilieronyraus,^ who found in these cells a thin hya- line membrane externally, a chromatophore, and a central body consisting of a single much-wound fibril, comprehending in its turns all the granules in the cell. The granules he looks upon as crystals belonging to the regular system, and composed of a substance " cy nophycin," which, though not identical with uuclein in its reactions, he regards as related to the chromatin and pyrenin of highly specialised vegetable cells. The central body is, in his opinion, an " open nucleus." According to Palla- the cells in the Cyanophycere consist of a chroraatophore with a vesiculated structure, of a central homogeneous body, and of granules of different composition always outside the latter. The central body is affected, like a nucleus, by staining reagents. In preparations fixed with corrosive sublimate and stained with Bohmcr's lijcmatoxylin the granules adjacent to, or in contact with, the central bodv are stained reddish-violet, while those scattered in the ehroma- tophore are coloured blue. He finds that those which thus become blue dissolve in dilute solutions of hydrochloric acid (0-3 per cent.) and do not stain intra vitara when treated with solutions of methylene blue. The substance constituting these, and which he calls " eyanophyein," he regards as the first assimilation product of the activity of the chromatophore. Those which stain reddish-violet with lu-cmatoxylin are com- posed of a viscid substance, are not soluble in dilute acids, and in the living cell manifest a strong affinity for methylene blue. To such structures he has applied the name " mucous sphr ■ rules," first given them by Schmitz. They correspond -li a the granules which, in Butschli's preparations, stainetl red with haematoxylin, but, in opposition to the views of that observer, Palla regards it as extremely doubtful if they contain any compound comparable to chromatin. ' " Beitriige zur Morpliologie und Biologie der Algen," Cohn's ' Beitrii V ^ #/ v f Photographic Sciences Corporation d ^ ^^ V ^\ *% V h \ ^ '*%" 33 WEST MAIN STREET WEBSTER, N.Y. 14S80 (716) 872-4503 V^^ 264 A. B. MACALLUM. From all tliis it may be gathered that nuclei, in the strict sense of the terra, are not present in the cells of the Cyano- phycese, and that if any structure performs the functions of such an organ, it must be the colourless central body. In regard, however, to the composition, the position, and the number of varieties of the granules, there is loss of concord- ance. All the observations quoted would appear to indicate that a typical chromatin substance is absent. If a " masked " iron compound is present in these organisms, with what part of the cell is it associated? The forms which I used, in endeavouring to determine an answer to this question, were : OscillariaFroelichii, Oscil- laria princeps, Oscillaria sp., Tolypothrix sp., Scy- tonema sp., Microcoleus terrestris, Cylindrosper- mum majus, Anabsena (Spherozyga) oscillarioides, and Nostoc commune. The fixative reagents used were alcohol, corrosive sublimate, the stronger Flemmiug's fluid, and saturated solutions of picric acid ; while the staining fluids employed were hfematoxyliu (Ehrlich's and Delafield's), alum cochineal, picro-carmine, safranin, and eosin. In determining the presence of iron compounds, material hardened by alcohol only was used. The results of my experiments, so far as they affect the question of the relations of iron to the cytoplasm of these cells, may be summarised as follows : 1. The cytoplasm consists of a dense central portion and of a vesiculated peripheral zone, the former staining with hsema- toxylin, alum cochineal, and safranin more deeply than the latter when it is free from granules or vesicles, but when vesicles are "resent they stain deeply, while the remhinder of the central portion acquires a faint colour only slightlv more marked than that of the peripheral portion. The size of these vesicles of the central portion varies from that, in vhich they appear as scarcely larger than granules, to that observed in Tolypothrix sp., in which they measured in diameter a third of that of the cell. The stainable substance of these torms a thick raembraue enclosing an apparently inert sub- IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 265 stance, and when subjected to the action of artificial gastric juice for two or three days it lessens slightly in volume, but its presence is quite as readily demonstrable then as it was previously. In this case the central portion of the cell also diminishes in volume slightly, the diminution entailing a shrinkage of the peripheral portion away fiom the original limits of the cell. Digestion does not affect the capacity, on the part of the central substance or of the membrane of the vesicles referred to, of absorbing staining matters, but on subsequently treating such preparations with a solution of potassium hydrate of 0"1 per cent, strength for twenty-four hours, the vesicles disappear and the central body, now some- what swollen, has lost its capacity for fixing colouring matters in itself. Evidently there is here a substance which has the characters of nuclein. This is confirmed by the results of experiments to determine the presence of "masked" iron. The central body always gives, with the glycerine and sul- phide mixture, in an interval of from two or three days to two weeks in length, depending apparently on the size of the cell, a diffuse greenish reaction which is changed to light blue on the addition of a drop of an acid ferrdcyauide solution. When granules and vesicles stainable with hseraatoxylin are present, they also give a reaction for iron, but it does not always manifest the same intensity. The iron in them is most readily demonstrated after they have been treated with sul- phuric acid alcohol (fig. 51). 2. In the peri' heral portions of the cytoplasm, in well- nourished forms only, are granules not ;- readily stainable with hsematoxylin, but which are intensely coloured with picro- carmine. These are dissolved out of the fresh cell with dilute hydrochloric ac id, and even in preparations thoroughly hard- ened in alcohol they are but slightly less soluble in the same reagent. In Oscillariae they are placed in a row at each end of the cell and adjacent to the transverse walls, but in Micro- coleuo terrestris and Cylindrospermum majus they are disposed in all the peripheral portions of the cytoplasm. In the spores of the latter some of them appear as if embedded I 266 A. B. MACALLUM. iu the central body. These are the " cyanophycin " granules of Palla and such as Biitsohli found in Oscillariae to be un- affected by hseraatoxylin bui markedly stained by eoain. They may give a reaction for iron, but not always or -^ of the same intensity, for in Oscillariae it was very slight, and in one pre- paration of Microcoleus terrestris none was obtained, while in preparations from the same specimen of fresh material made a few days later than the other, the reaction v as quite distinct. In two preparations of Scy tonema sp. the granules gave no reaction, a result which I attribute to a deterioration of the solution of the sulphide reagent then used. In Cylin- drospermum majus these granules give an intense reaction for iron (fig. 8). The iron is not less firmly combined in the substance of these granules than it is in the chromatin, for in the last mentioned species the glycerine and sulphide mix- ture brought out its complete reaction only after an applica- tion often days or more. Within twenty-foar hours after the addition of the mixture, they gave, in all the species in which they were iron-holding, a slight greenish reaction. I have not succeeded in demonstrating the presence of iron in them after the use of sulphuric acid alcohol, and the explanation for this is that the latter reagent liberates, but at the same time wholly extracts the iron in these granules, the substance of which, unlike chromatin, is incapable of retaining it. 3. Beyond the fact that the " cyanophycin " granules may contain iron, there is nothing to show a relationship, chemical or physiological, between them and the vesicles. From their situation the " cyanophycin " granules would, as Palla sug- gested, appear to be the assimilation product of the activity of the chromatophore, while the chromatin vesicles and granules might be regarded as due to processes of elaboration on the part of the central body. In Cylindrospermura majus, which grows on soft mud, the former are usually extremely abundant, but in twenty-four hours after placing the thallus in water, the granules diminish very much in number, and on the third day they may be wholly absent in very many of the filaments. Central vesicles, on the other hand, are iu this form extremely IRON OOMPOUNDS IN ANIMAL AND VEGETAULli: CELLS. 2(i7 few in number^ and the conditiousi which greatly influence the number of the " cyanophycin " gi'anules have apparently no effect upon them. In Oscillaria Froelichii a filament may contain large numbers of both elements, another may contain " cyanophycin " granules only^ while a third may be free from the latter but contain a large number of vesicles, and all in the same preparation. In Cylindrosperraum majus the "cyanophycin" granules of the spore diminish somewhat in number and volume during the formation of the episporium, and in the spore which is undergoing its initial division their number is very greatly reduced, the central body appearing at the same time increased in volume. 4. In the heterocysts of Nostuc commune, Cylindro- spermum majus, and Scytonema sp., picro-carmine demon- strates the presence of" cyanophycin" substance in a button- shaped body at one or both ends of tlie cell, according as the heterocyst is terminal or intercalary. A strand of " cyano- phycin " connects this body with the contents of the neigh- bouring cell.^ In the heterocyst the " cyanophycin " body is quite unconnected with the homogeneous cytoplasm which occu- pies the remainder of the cavity, and stains but faintly with hsematoxylin and not at all with picro-carmine. When sub- jected to the prolonged action of the glycerine and sulphide mixture the "cyanophycin," both of the button and of the strand, gave a deep reaction for iron, and a feebler reaction was obtained in the cytoplasm (fig. 8). It thus appears that in the Cyanophycca} there is a substance, containing " masked " iron, in many respects like the chromatin of more highly organized cells, and that the " cyanophycin," a compound of undetermined nature, may, in some forms at least, a'so give evidence of the presence of the element in a firmly combined condition. ' This connection lias already been described by Ilansgirg, ' Pliysiologische uud Aiyologische Studien,' Prague, 1887, pp. 125, 12G. The description is quoted in full by Deincga. I A. P.. MACALLUM, General Remarks. The facts described in the preceding pages appear to indi- cate that a substance, in which iron is firmly held, is a constant constituent of the nucleus, animal and vegetable, of the cyto- plasm of non-nucleated organisms and those possessed of apparently rudimentary nuclei, and that, further, a similar iron-containing substance obtains in tlie cytoplasm of ferment- forming cells. This substance, to which cytologists apply the term chromatin, cannot, on theoretical grounds, be regarded as constant in its molecular structure, even in the same organism, and its most marked characteristic, apart from the iron in its com- position, is the occurrence in it of nuclein or nucleiriic acid. Beyond the fact that the iron is firmly held, it is difficult to say how it is disposed in the molecular structure of the nuclein or nucleinic acid. It is, possibly, united directly to the carbon of the latter. The acid alcohols liberate it as a ferric salt, but this fact cannot be held to indicate that it is combined in the nuclein or nucleinic acid in a ferric state, since from solutions of potassium ferrocyanide, in which the iron is contained in a ferrous state, acids liberate the iron in a ferric condition, ^ as evidenced by the formation of ferric ferro- cyanide or Prussian blue. It is also difficult to say whetlier there is, in the way in which the iron ii leld in the animal cell, anything diflFcrent from that obtaining in the vegetable organism. I have, as a rule, found it easier, in the case of the vegetable cell than in that of the animal cell, to liberate the iron with ammonium hydrogen sulphide ; but upon this no conclusion may be founded, since the same reagent liberates the iron of free hsematin readily, while it does not affect the iron of hajmatiu in haemoglobin, and it is possible that in the animal cell the ' The iron immediately on liberation may be in Uie ferrous state, but it quickly assumes the ferric form. Similarly, the iron liberated in the chro- matin may at first be a ferrous compound which, with the continued action of the liberating reagent and under the conditions obtaining in the hardened tissues, may further undergo a conversion mto a ferric salt. IRON COMPOUNDS IN ANIMAL AND VEGETABLE CELLS. 209 proteid molecules attached to the iron-containing nuclein or nucleinic acid may more greatly aflfect the activity of the reagent tlian those of the vegetable cell are capable of doing. Since, on the other hand, haemoglobin, which, as I have pointed out, is derived, in Amblysioraa, from chromatin, occurs in a large number of animal forms, but is present in no vegetable organism, it would appear to follow that the iron is combined in animal chromatin in a way unlike that in which it is held in the vegetable cell.^ The apparently universal occurrence of such iron com- pounds renders intelligible the fate of the iron salts absorbed by plants from the soil, and of the iron compounds found by Ilaulin^ and Molisch' to be necessary for the growth of Aspergillus nig^-r. Chromatin, to the formation of which the iron absorbed contributes, is, as the results of cytological investigatlnns show, a substance of primary importance to the cell, and a diminution in, or a cessation of, the supply of iron to the vegetable organism, whicli produce.^ the condition known as chlorosis, instead of affecting only the formation of its chlorophyll, as generally supposed, strikes at its very life. The conditions known as ansemia and chlorosis in the higlicr Vertebrates have been hitherto explained as caused by a diminished production of ha3moglobin directly from organic or inorganic iron compounds absorbed by the intes- tine from the food matters ; but they must now be referred to a deficient supply of the primary iron-containing com- ' Compounds which appear to resemble, somewhat remotely, the Iiscmatins of animal organisms have been found in Palmella cruenta (Pliipson, "Sur la matiere colorante du Palmella cruenta," 'Comptcs Rendus,' vol. Ixxxix, p. 316, 1879), and in Aspergillus niger (Linossier, "Sur une hematine vegctale ; I'aspergilline, pigment des spores de I'Aspergillus niger," 'Comptes Rendus,' vol. cxii, p. 489, 1891). The colouring matter of the lavier is, as I have found, held in the membrane, but not. in crypto- plasm of the spore, and it would, therefore, appear to be simply a degeneration product.. » " l^tudes chimiques sur la vegetation," ' Anuales des Sc. Nat.,' Bot,, Serie 5, vol. xi, 1869, p. 93. 3 Op. cit., pp, 97—117. I 270 A. B. MACALLUM. pound, chromatin, not only in the haematoblasts, but in all the cells of the body. The cousequenu'y lessened proliferation of cell and tissue would explain the hypoplasia of the imper- fectly developed vascular system observed by Virchowi in chlorotic human subjects. Accepting this explanation of the nature of chlorosis, one may infer that this condition is not limited to animal organisms in which haemoglobin is found, although its occurrence in others may be difficult to detect because of the total absence of this pigment. From this point of view animal chlorosis IS fundamentally similar to the chlorosis of the vegetable kingdom. The oxygen-carrying property of haemoglobin and of hioma- tin IS generally attributed to the iron present in these, because when ha3raatin is deprived of its iron, the resulting compound, whether haematoporphyrin or bilirubin, manifests no affinity for oxygen. The proof may not be quite conclusive, for we cannot be certain that either compound represents the un- changed remainder of the hsematin less its iron, but assuming that It is correct, it follows, as I have pointed out in my pre- vious communication, that the antecedent of hemoglobin chromatin, has the capacity of absorbing and retaining oxygen,' and that one may attribute the processes grouped under the term " vital,'' to an alternation of the conditions of oxidation and reduction in the iron-holding nuclear constituent. This hypothesis, reasonable as it now appears to me to be. I do not regard as free from difficulties, since in vegetable cells the two processes of respiration and assimilation, involving two activities of different natures, so far as the oxygen is concerned appear to postulate the existence of two ditferent iron com' pounds in the same nucleus.^ There are no facts to indicate ' " Ueber die Clilorose uud die daniit zusammenliiingenden Anomalien in, beiassapparate, nisbesondere iiber Endocarditis puerperalis." 'Vortidi? ' Berlin, 1872. ' vorucg., » On the relations of the vegetable nucleus to the processes of assimilation see btrasburger, 'Ueber Kern- und Zelltheilung in, Pflanzenreiche,' ISSs' pp. 194—204. ' TRON COMrOTTNDFt IN ANIMAL AND VKGETAHLE CELLS. 271 the occurrence of such, and it is scarcely possible to explain away the objection without advancing some hypotheses regard- ing the action of the sulphar and the phosphorus in the nuclein. I propose to detail these on another occasion. EXPLANATION OF PLATES 10-12, Illustrating Dr. A. B. Macallum's paper "On the Distribution of Assimilated Iron Compounds, other than Ilsemoglobin and Ilsematins, in Animal and Vegetable Cells." Explanation of Figures. Note.— In the preparation of all llie figures Abbe's camera luciila was em- ployed when the size of the objects represented permitted its use, and all except 25, 26, 35, and 36 are illustrated as they were seen with an apocliro- matic immersion objective (Zeiss 3 mm., 2 mm., or 1'5 mm.). The excep- tions are represented as they appeared under a Zeiss D. Figs. 1—40 show the distribution of assimilated iron as it was demonstrated by the dark green colour of ferrous sulphide, but in Figs. 41—53 the disposition of iro;i com- pounds of this kind is indicated by the colour of the Prussian blue reaction. Fig. 1.— a nucleus and a cell from the testicle of Necturus lateralis. Alcohol, the glycerine and sulphide mixture eleven days. X 620. This and tlie two succeeding illustrations were drawn from the very first preparations made witii ti)is reagent. yig. 2.— Testicular elements of another example of N. lateralis. Alcohol, the glycerine and sulphide mixture eleven days. X G20. Fig. 3.— «, a leucocyte, *, a red corpuscle, of N. lateralis. Alcohol, the glycerine and sulphide mixture six days. X 500. Fig. 4.— Two yeast-cells, Saccharomyces cerevisiee. Alcohol, the glycerine and sulphide mixture ten days. X 1500. Fig. 5.— Four yeast-cells, Saccharomyces Ludwigii. Alcohol, the glycerine and sulphide mixture four days. X 1 640. Fig. 6.— The developing and fully-formed spores of Cystopus candidus. a, b, c, d, e, alcohol, sulphuric acid alcohol two days, ammonium hydrogen sulphide in glycerine. X 750. /, alcohol, the glycerine and sulphide mixture ten days. X 680. VOL. 38, PART 2.— NKW SER. S 272 A. B. MACALLUM. n/'rr ^— ^P"'" °f A.spcrgiilus glaucus, a, in the unripe, *. in the ripe condition. Alcohol, the glycerine and sulphide mixture three days, x 1640. rJjl: 8--f^^;l"«. I'cterocyst (/}.). and spore (.,;,.) of Cylindrospermum majus, Alcohol, tht; glycerine and sulphide mixture fourteen days, x IGIO FiG.O.-Threecellsof afilamentofMicrocoleus terrestris. Alcohol, the glycerine and sulphide mixture four days, x 2n00. Tllr ^^'"r ^^''"'■''^ C''"'^"'"^''). *. ^> and d, basidia of a leucosporous I yn en myeete. d. Basidium «ith sterigmata and one attached spore tyt X sto"" '" *■ '^'""''°'' "'' *'''^""""' ""^ '"''''"'*^ '"''^'"^ ^'Sl'* Figs 11-13 -Portions of hypha; of Ilyphelia terrestris Erics illus ra ing the development of the fructiCeation. 13 a and b representing the simplest form. Alcohol, the glycerine and sulphide mixture three day" Figs. It-lS.-Prom the ovary of a specimen of Erythronium ameri- canun hardened in alcohol. Fig. 11 illustrate, the effect produced by diam- milium sulphide and glycerine in two days ; Figs. 15, 17. and 18 represent hat produced by ammonium hydrogen sulphide and glycerine in the same time ; and n tig. 10 is shown how intense the reaction appeared after treat- nient for four days with thn same reagent, x 1210. Figs. 19-22.-From the ovary of a specimen of Erythronium ameri- can urn ardened in alcohol. Sections treated for thirty hours with sulphuric sulphide ;<"l2.rr'"'" '" ' '""'"" °' ^"^"""^ ""^^ ""^"^""'"^ "^'^-«- Fig. 23.-Four hepatic cells from a specimen of Necturus lateralis. Alcohol, the glycerine and sulphide mixture eight days, x G20. Fig. 21-Two hepatic cells from the same animal, illustrating the distri- bution of the iron and the nuclear structure after they were treated with sul- phuric acid alcohol for twerty-four hours, and mooted in a mixture of glycerine and ammonium hydrogen sulphide, x G20. nnrVrT"^" f ^"f'^^f S'^"*"-- Polymorphus. Alcohol, the glycerine and sulphide mixture two weeks, x 305, fl.,mi'-f~^" ."""'''' °^ ^''"*" P°'.vmorph„s. Alcohol. Bunge's fluid thirty-seven hours, ammonium hydrogen sulphide and glycerine, x 305 Fig 27.-Exainples of Vorticella sp. Alcohol, the glycerine and sulphide mixture seven days, x GOO. fou^'hou!!";;'^" '""''^i "^ ^^'■''^^'' '^- ^'•=°''°'' ^'^S^'' fl»id twent;, lour hours, glyceriue and ammonium hydrogen sulphide, x 600. Fig 29.-An ovum of Ascaris my stax. Oxed during impregnation. Only a portion the ovum is represented. Alcohol, the glycerine and sulphS^I mixture eight days, x 820. "uipmue 1 IRON COMPOUNDS IN ANIMAL ANi: VKOETABLE CKLLS. 273 ■^ ! Fio. 30. — An impregnated ovum of Ascaris mystax, showing Uie division of its nucleus («.) and the condition of the spermatozoid (sp.). Alcoiiol, the glycerine and sulphide mixture ten days, x 750. Figs. 31 and 32, — Spermatozoids of Ascaris mystax. Alcoiiol, the glycerine and sulphide mixture nine days. 31, X S20; 32, x ICIO. Fios. c"^ — 3G.— Ovarian ova of the lake-lizard, Necturus lateralis, illustrating diirurenccs in the diistribution of the "masked" iron. In 5 is shown the iron-containing peripheral nucleoli, and a represents a more highly magnified (= X 1210) portion of the nuclear structure. lu 30 is seen an earlier stage with a, a portion of its nuclear network, more highly magnilied (X 1210). Alcohol, sulphuric acid alcohol thirty-six hours, glycerine and ammonium hydrogen sulphide. X 305. Fig. 37.— lU'tinal rods and cones from a larva of Amblystoma. Alcohol, vfhole of retina i;i Bunge's fluid two days, glycerine and ammonium hydrogen sulphide. X C20. Fio. 38.— Cells from the pancreas of a larva of Amblystoma. Alcohol, Buiigc's fluid (on the whole of the organ) two days, glycerine and ammonium hydrogen sulphide. X G20. Fig. 39.— a portion of a section of the human epidermis, illustrating the occurrence of "masked" (?) iron in the granules (cleidin) of the stratum granulosum and in the stratum lucidum. AIco' ' ■ ' liuri' cid alcohol two days, glycerine and ammonium hydrogen sulpl.:' Fig. 40.— Strands of fibrils from the muse! ' stoma. Alcohol, sulphuric acid alcohol two liydrogen sulphide. X 750. Figs. 41 a and i.— From the ovary of a speeime. ricanum; i represents an isolated nucleus. Alcohol, . thirty hours, acid ferrocyanide mixture, balsam. X 1240. Fig. 42.— a cell from a section of the ovary of the same specimen, with the iron demonstrated as in last case, but the preparation, before being mounted in balsam, was stained with eosin. X 1240. Figs. 43 and 44 « and A.— Nuclei of the embryo sac of a specimen of E. americanum. Alcohol, sulphuric acid alcohol thirty-six hours, acid ferro- cyanide mixture, balsam. X 020. Figs. 45 a and A.— Nuclei from the liver of a specimen of Necturus lateralis, n. Nucleoli. Alcohol, sulphuric acid alcohol thirty-six hours, acid ferrocyanide mixture, balsam. X 1240. Figs. 40 »—. r s ,^i.V' -^■*.^- fw ">!>■ s fx % m m ^ .^'-A # ii^^ 4. ^V I • -V^ '1 h r f ^. IQ. '\ 11 r: ■t.f'Ay.- yt^ *■'«* ,<*»:y»^*'; ..'i^ :/ IS. ^ 17. 19. ^ / ■^' A.' >r .">>■ V," " 18. 20. "ii-tUuiiv ic; .UUC'C. < ^CH^'/l. yUx"."' ~JC<^. ^ uO-Od, :v.sm.. i"?-!' I: vc ) I t U. 20. m 12. 13 25. I t Hati.,.-'.th' Kan: I I 0^-- .. vv:- J" 28 30. 29. /'r "'-yr^ ^-Uviii /'i. > ■'«■ 'tA*-^^^' \ ^ V .\.. X, /-<■■>- ^, ^ X / / / 37. .i^ 38. '•*«■■'•■ •■-'■: • ■ ,7 i A B IVlaca'ilum del SJ> MUl^..^'fi>tO''n t..V:-U-r: oCu ^ :::(.CA,.,.^.C'l: //. « 31. ^^'"N .a 3.'. I V f.v. 23. 30. ^"^ J M. 39. 35. m 40. f »:.:v<-.:-/... •••„-, u 41. 43 V - 42 F Huth.Luh'' Etiin' I A" 44 .-^''^v. 'VE: .- -"'7 /•>*'- -*fe .^^>> \ 46. -^-^ 46 / ; .L ...1 A, V ^ .^, V T 47 'ti' >'i{: fir: ^;r/- t * 53. — ^r*^ U.- SO iR*l<*j^v. ■"5*«?,V r M»i aiium