has the properties of fibrin-ferment.

Reasons have already been

given for considering the fibrin-ferment and cell-globulin as identical

(see p. 241).

3. An albumin. After filtering off the globulins, the albumin remains in solution. It coagulates at 73° C. and resembles serumalbumin a in its properties. It is present in very small quantities, and may be provisionally termed cell-albumin.

In concluding this account of the proteids of lymph cells, it may be added that no substance like myosin or fibrin can be obtained from the cells; there is, however, a formation of sarkolactic acid1 after death as in muscle and if the glands be left, especially at the temperature of the body, for some hours after death, a process of self-digestion takes place, the pepsin present in the glands, as it is in most tissues (Brücke), becoming active when the reaction of the tissue becomes acid; under these circumstances there is, in addition to the proteids already enumerated, a small and varying amount of proteoses and peptones.

The nucleo-albumin was mistaken by some of the earlier observers for myosin, from which it differs markedly. Some also have mistaken it for fibrin2; the way in which it extends in strings when poured into water accounts for this; these strings subsequently contract, and here indeed is a point of resemblance between it and fibrin. But here all resemblance stops.

THE BLOOD TABLETS

In addition to the white and red corpuscles, a number of colourless discs averaging 002-003 millimetre. diameter are also seen. They exist as such in the circulating blood. By some they have been supposed to be stages in the development of red corpuscles; some3 consider them to be masses of undifferentiated protoplasm, but their origin and destiny has never been explained. The action of inert solids upon them after the blood is shed is much the same as on white blood corpuscles. It causes them to become sticky, to run together, lose contour, change shape, and in many cases undergo complete disintegration. Strands of fibrin start from collections of blood-plates, so probably one product at least of their disintegration is fibrin-ferment.

In spite of the large amount of research from the histological stand

1 It was Hirschler who showed that the variety of lactic acid formed was sarkolactic acid (Zeitsch. f. physiol. Chem. xi. 41). Berlinerblau (Chem. Centralbl. 1888, p. 757) states that lactic acid is a normal constituent of blood. But Salomon (Virchow's Archiv, exiii. 356) has shown that fresh blood contains no lactic acid, but on standing a small amount forms, no doubt from changes of a fermentative nature in the white corpuscles.

* Denis called it fibrine concrète globuline. Wooldridge also spoke of it as fibrin (Du Bois Reymond's Arch. f. Physiologie, 1881, pp. 387-411). Hammarsten was the first to show that it is not true fibrin (see p. 232).

3 Haycraft, Journ. of Anat. and Physiol. xxii. 302.

point, on these blood tablets (Blutplättchen of Bizzozero), we know virtually nothing of them chemically.

The term hæmatoblasts has been applied by some to the blood tablets; this is liable to cause confusion, as the same word is used for the nucleated red corpuscles which occur in certain stages of the formation of the non-nucleated red discs of vertebrates.

The blood tablets are found only in mammals' blood; in fishes, birds, and amphibians they are absent, and according to Eberth and Schimmelbusch, and also Hayem,3 their place is taken in these groups by certain spindle-shaped nucleated cells. Löwit on the other hand regards the spindle cell as a variety of white blood corpuscle, and the blood tablets of mammals as something peculiar to that group; according to him they consist chiefly of a globulin, and he considers they play an important part in the formation of fibrin.

THE RED BLOOD CORPUSCLES

The red or coloured corpuscles give the red appearance to the blood. They are much more numerous than the white corpuscles, there being about 5,000,000 per cubic millimetre in the human male, about 4,500,000 in the female.

The enumeration of the blood corpuscles is readily effected by the hæmacytometer of Gowers. This instrument consists of a glass slide (fig. 54 C), the centre of which is ruled into 1 millimetre squares and surrounded by a glass rim millimetre thick. It is provided with measuring pipettes (A and B), a vessel (D) for mixing the blood with a saline solution (sulphate of soda of specific gravity 1015), a glass stirrer (E), and a guarded needle (F).

The mode of proceeding is extremely simple. 995 cubic millimetres of the saline solution are measured out by means of A, and then placed in the mixing jar; 5 cubic millimetres of blood are then drawn from a puncture in the finger by means of the pipette B, and blown into the solution. The two fluids are well mixed by the stirrer, and a small drop of this diluted mixture placed in the centre of the slide C; a cover glass is gently laid on (so as to touch the drop, which thus forms a layer mm. thick between the slide and cover glass), and pressed down by two brass springs. In a few minutes the corpuscles have sunk to the bottom of the layer of fluid, and rest on 1 Bizzozero, Virchow's Archiv, xc. 261. 2 Virchow's Archiv, eviii. 366.

3 Hayem, Du sang, Paris, 1889.

Archiv f. exp. Path. u. Pharmakol. xxiv. 188.

5 Gowers, Lancet, Dec. 1, 1877. Malassez (Compt. rend. 1872) and Hayem (Du sang) have invented very similar instruments.

There are many similar saline solutions which may be employed.

the squares. The number on ten squares is then counted, and this multiplied by 10,000 gives the number in a cubic millimetre of blood. The average number of red corpuscles in each square ought, therefore, in normal human blood to be 45-50.

FIG. 54.-Hæmacytometer of Dr. Gower. (Made by Hawksley & Co., Oxford Street.)

Specific gravity.-C. Schmidt gives the specific gravity of red blood corpuscles as 1.089, Welcker as 1·105.

Shape and size. They vary in size and structure in different groups of the vertebrate sub-kingdom. In Mammalia, with the exception of the Camelidæ, they are biconcave, circular discs; they have no nucleus except during embryonic life, and they have a tendency to run into rouleaux when the blood is at rest, but if it is disturbed they readily become separated. In the Camel tribe they have an elliptical outline. Their average diameter in mammals is 007-008 millimetre ! (3260 inch), and about one-fourth of that in thickness; there are very slight variations in different classes of mammals. In birds, reptiles, amphibians, and fishes, the red corpuscles are biconvex, oval discs, with a nucleus; they are largest in the amphibia.

Action of microscopic reagents.- Water causes the corpuscles to swell up, and at the same time dissolves out the hæmoglobin, leaving a globular colourless

stroma.

This is often written 7-8μ. 1 (micro-millimetre) = one-thousandth of a millimetre.

crenated on the surface. The action of water and of salt solution suggests the existence of a membrane on the surface of the corpuscle, through which osmosis takes place. The question, has the red corpuscle a membrane? was once the subject of voluminous discussion. An admirable summary of the positions held by the older writers is given in Gamgee's Physiological Chemistry' (p. 72). The matter has been finally compromised by considering the stroma to be rather denser at the surface than in the interior. The outer denser part plays the rôle of a membrane during osmotic phenomena.

Dilute alkalis (0-2 per cent. potash) slowly dissolve the corpuscles.

Dilute acids (1 per cent. acetic acid) act like water; and in nucleated red corpuscles render the nucleus distinct.

a

P

૨

FIG. 55. a-e, successive effects of water on a red blood corpuscle; f, a red corpuscle crenated by salt solution; 9, action of tannin on a red corpuscle.

Tannic acid causes a discharge of hæmoglobin from the stroma, but this is immediately altered and precipitated. It remains for a short time adherent to the stroma in the form of a round or irregular globule of a brownish tinge, consisting probably of hæmatin.

Boracic acid acts similarly, but in nucleated red corpuscles the colouring matter is partially or wholly collected around the nucleus, which may then be extruded from the corpuscle.

Nucleus. The nucleus of those red corpuscles in which one exists, has the usual reticular structure, and consists according to Lauder Brunton and Plósz 2 of nuclein. Defibrinated blood from the bird was treated with ten or twelve times its volume of 3 per cent. sodium chloride solution, and the corpuscles separated by decantation. On shaking the corpuscles with a mixture of water and ether, the nuclei alone remain undissolved and float at the junction of the two liquids.

Nuclein may, however, also be prepared from red corpuscles by Miescher's method, which consists in subjecting the corpuscles to artificial gastric digestion. The nuclei alone remain undigested.

Origin of blood corpuscles in mammals.--The following is a brief résumé of the chief ascertained facts concerning the origin of the red discs3:--

In the embryo the first formed coloured blood corpuscles are amoeboid nucleated cells. These are developed within certain mesoblastic cells which are united to form a network. The nuclei of the cells multiply, and around some of them, protoplasm coloured by hæmoglobin is aggregated. Finally the network is hollowed out and filled with fluid; thus capillaries are produced; the coloured nucleated portions of protoplasm are set free within these as the embryonic

1 L. Brunton, Journ. of Anat. and Physiol. 2nd series, vol. iii.
Plósz, Hoppe-Seyler's Med. Chem. Unters. Heft iv. p. 460.
5 Quain's Anat. vol. ii.

p. 91.

blood corpuscles. In later embryonic life these are replaced by the usual nonnucleated discs, which are moulded within connective tissue cells as before, except that the cell nuclei do not participate in the process.

Nucleated coloured corpuscles are not seen in the blood after birth; but they continue to be formed in the red marrow, and in some animals in the spleen also. Probably the nucleus disappears from them, and the coloured protoplasm is moulded into a discoid shape. Malassez, however, considers that the red discs are formed by a process of budding from these cells, which he terms globuligenic cells.' The evidence that the red corpuscles are derived from the white, or from the nuclei of the white corpuscles, or from the blood tablets, is insufficient. Composition. According to C. Schmidt, 1000 parts of moist red corpuscles contain :

According to Hoppe-Seyler and Jüdell, 100 parts of dried corpuscles contain :

The mineral constituents of the red corpuscles have been investigated by

C. Schmidt, and the following tables contrast those of the red corpuscles with those of the plasma in man.

1000 parts of moist corpuscles yield:

Mineral matter (exclusive of iron, which is contained

Hoppe-Seyler and Jiidell, Med. Chem. Untersuch. Heft iii. p. 386. P. Manasse (Zeit. physiol. Chem. xiv. 452), gives the percentage of lecithin in the red corpuscles as 1867, of cholesterin as 0.151.