Figs. 20-27, 30-38 and 40 in text:

'It is often totally impossible to see the nucleus of the large basophile cells in the mesentery when the specimen is stained with methylene-blue, the stained granules are crowded so closely round the nucleus as to hide it altogether. (Figs. 31 and 32.) One may however in these cases overcome the difficulty by using the Ehrlich-Biondi stain, or by cutting sections of the folded mesentery. I regard the history of the nucleus as being somewhat as follows. In the youngest cells, either in connective tissue (Figs. 24-27) or in blood (Figs. 20-23), the nucleus presents no difference from any ordinary leucocyte-nucleus; the chromatin is arranged in lumps and threads as usual; if anything, the spaces between the threads of the nuclear network are rather larger than usual. As the nucleus becomes older the nuclear reticulum becomes closer and contains relatively more chromatin (Fig. 40); if it passes into amceboid movement or becomes older still (i.e. the cell becomes larger) the staining with methylene-blue becomes more and more diffuse until at last no structure can be made out (Figs. 28, 30, 33). On analysing this-with the Ehrlich-Biondi stain, one finds that this diffuseness of staining is due to the presence of large masses of "basichromatin," staining green, while the amount of pink-stained material, whether linin or "oxychromatin" is very small, and no pink nucleoli are to be seen. I have drawn several such nuclei in Figs. 34-37, and have placed beside them as a contrast the nucleus of a young connective tissue cell (Fig. 38), from the same preparation, which shows well-marked nucleoli and a much more open arrangement of the reticulum. Reinke [note: '(24)' ['Reinke, F. Arch. f. Mikr. Anat. XLIII. 1894.']] has found mitosis once, but only once, in basophile cells. I myself have never been fortunate enough to see it. This extreme rarity of mitosis, taken along with the enormous number of basophile cells in the body, and the relative frequency of degenerated forms, goes far to disprove Kanthack and Hardy's contention that the coarsely granular basophile cells form an isolated group, and to render it almost self-evident that they must be derived from the smaller cells of the same kind, and that these again must be derived from the lymphocytes.' (404-405)

Figs. 20-32, 41-42, 44-45 in text:

'As the cells become larger the granules, as a rule, but not invariably, also grow larger, and become, generally speaking, more uniform in size. I have never seen a cell, however, in which all the granules were quite uniform, though many authors postulate this for all large basophile cells. In all the drawings I have given I have taken the greatest care not only to put the granules in the exact position in which I saw them, but to represent their relative size, and in Figs. 20-32, 41, 42, 44 and 45 (the two latter being degenerated basophiles) I have tried as far as possible to give their colour also. For while it is in these large cells that the most marked metachromatism is to be found, it is to be noted, that in very few cells are the granules stained uniformly red. Some of the very large cells in the rat's mesentery form an exception, but it is more usual to find some granules bright red, others blue, and to find between these all shades of purple. Still the prevailing effect is red.' (407)

Figs. 21-27 in text:

'I shall best be able to describe the changes through which they pass by tracing the cells from the original lymphocyte or hyaline transitionform, from which they are derived. The first change consists not so much in an enlargement of the microsomes as in the superior visibility of the mitoma. As in the eosinophile cells the cell-body becomes larger, and as there the mitoma takes up acid stains, so here it takes up basic stains (Fig. 4). As the cell increases in size, the mitoma retains its visibility and the microsomes become larger and assume the ordinary appearance of granules. These are at first few and scattered irregularly through the cell: some microsomes have increased in size, others have not; the mitoma is always most visible near the microsomes which have taken up the stain. At this stage, illustrated by Figs. 21-23, from the blood, and Figs. 24-27, from the mesentery, the granules always differ markedly in size, much more so than at later stages. They are always uniform in staining, and curiously enough, the very small ones do not often show the metachromatic reaction with methylene-blue. The cells in the blood are generally spherical, but those in connective tissue soon show an irregularity in shape suggestive of amoeboid movements; they put forth those long flat processes which are so characteristic of the larger cells. In Figs. 26 and 27, for instance, all the granules to be seen in the cell are shown, and they were all lying almost in one plane, so that hardly any focussing was required to see them. They must therefore have been lying in a very thin layer of protoplasm, and from the constancy of this appearance we may infer that the spaces through which these leucocytes have to pass between the planes of connective tissue in the mesentery must be very shallow. The rounder cells are generally found in places where, as near the larger blood vessels, the mesentery becomes thicker and allows more space; the elongated cells are generally in the thin parts of the structure.' (406)

Fig. 22, 29-31, 33 and 41-42 in text:

'It has often been remarked that the large cells show a great tendency to leave their granules behind them, thus one might come on a group of granules while the nearest cell was far away. Ballowitz [ref: '(2)' [Ballowitz, E. Anat. Anzeig. vi. 1891.']] was, I think, the first to declare that all or most of these groups of granules were attached to the cell by fine protoplasmic bridges. It is not indeed always easy to show this; in the cell shown in Fig. 31, which was so isolated that there could be no doubt that all the granules represented belonged to it, no trace could be made out of threads extending from granule to granule. They were probably stretched too much to allow them to be visible. Fig. 22, however, taken from the rat also, but not from the same individual, shows the mitoma very clearly, probably because the sheet of cell-substance has not been so greatly stretched. In the elongated forms one often sees narrow bridges, like those shown in Figs. 29, 30, 33, 41, which connect two larger masses of the cell-body; these often consist of a row of granules with a single thread running between (Figs. 29 and 42). As I shall point out when discussing the degeneration of these cells, I have little doubt that when that stage is reached these bridges are torn across, and the granules are actually left behind. This forms an exact parallel to what happens in the eosinophiles of the newt's blood.' (407)

Figs. 23-24 in text:

'Heidenhain has discussed the relation of the centrosomes to basic dyes, and has come to the conclusion that they have no very special affinity. He admits that they do stain with some basic dyes; Van der Stricht, Hermann, and Flemming have stained them with saffranin and gentian-violet, but he himself had negative results both with these and with methyl-violet, methyl-green and thionin. He does not mention methylene-blue. With this they certainly do stain, not in such a way as to constitute a specific reaction, but well enough to allow one, in favourable cases, to make out the relations. Figs. 23 and 24 show the centrosomes, one in a leucocyte from the blood, the other from the mesentery of the rabbit. Neither cell presents any difference of morphological characters from what one might see in an eosinophile cell; change the colours, and no one would know the difference. The distinction in staining between the centrosome and the granules of the cell is slight but distinct; the granules are blue-green, while the centrosomes are rather of an apple-green, of very nearly the same shade of colour as that which the hemoglobin in the red corpuscles assumes when treated with methylene-blue.' (405-406)

Figs. 24-25, 29-32 and 41 in text:

'Westphal [ref: '(4)' ['Ehrlich, P. Farbenanalytische Untersuchungen &c. Gesammelte Mittheilungen. Pt. I. 1891.']] distinguished three main forms of the large basophile cells, flat, spherical, and spindle-shaped, with two or more processes; he remarks however that this is purely a classification for convenience, as transition-forms of every kind are to be seen also. This I can fully confirm, as I can also his dicta that "in individuals of the same species and same age the distribution, number and size of the cells are similar in similar organs," and that in individuals of different species there are great differences in the number &c., of the cells in the same organs. I may add that certain species show a tendency to have cells of one or other shape or size. The mesentery of the rabbit, for instance, contains, speaking generally, small cells, mostly spherical (Figs. 4, 24, 25); of the guinea-pig, larger spindle-shaped cells (Figs. 29 and 30); of the rat, enormous round and flat cells (Figs. 31, 32); while that of the newt has huge spindle cells which send off often several secondary branches from the primary ones (Fig. 41).' (403-404)

Figs. 26-27, 29-30 and 41 in text:

'I consider, contrary to Kanthack and Hardy, that the basophile cells are amoeboid. They examined only the basophile cells of the peritoneal fluid. I find, however, that most of these cells are degenerated. The evidence for their amceboid power is the irregularity of their shape in sections of the tissues (cp. Figs. 26, 27, 29, 30, 41), the phenomenon of "Körnchenzerstreuung," and the way in which the cells are scattered throughout the body. It seems to me that such shapes can only be explained by amoeboid action. In drawing them they must of course be represented in one plane, but one does not in fact find them arranged in that way. The process of Fig. 30 was turned up into quite another direction. Fig. 29 is the most interesting; the end marked a is evidently anterior, and the leucocyte was seen to have been fixed in the act of passing through a narrow hole between two bundles of connective tissue. The nucleus shows in its anterior half that diffuse staining which Arnold long ago pointed out as accompanying amoeboid movement, and which probably depends on the squeezing together of the chromatin-containing filaments. ' (403)

Fig. 33 in text:

'I have spoken as if the elongated basophile cells extended in one plane only, but though this is the case in the membranes, it is not so when these cells occur in ordinary connective tissue. In the heart of the frog or newt, for instance, the branches run in every direction. Fig. 33 will give some slight idea of this.' (408)

Fig. 43 in text:

'In the large basophile cells I have not been able with certainty to make out the position of the microcentre. A frequent appearance is that represented in Fig. 43, where a single large granule forms a centre for a number of smaller ones.' (406)

Figs. 44-45:

'In the portion of cell-body which remains round the nucleus the granules are generally large; the threads of the mitoma lose their power of taking up stains, and the cell assumes the form of a more or less disc-shaped mass of granules, which not only stain badly, but which lose their regular outline. It is curious that in all the markedly degenerated basophile cells which I have seen the granules stained blue with methylene-blue and seemed to have lost the red metachromatism. In Fig. 45 this is well seen. Some of the granules in Fig. 44 retain the red colour to a certain extent, but the drawing rather exaggerates the amount of red in them.' (409)