The restoration of the volume of the blood at the expense of the tissue fluids immediately after a haemorrhage has been already dealt with. This tissue fluid is, however, more dilute than normal plasma. Observations on the specific gravity of the plasma after hsemorrhage show that there is at first a very rapid fall, which is soon replaced by a quick, though less precipitous, rise. In the rabbit the normal figure is approximately regained in a few hours, in man in a day or two. These experiments show that after the volume is restored (rabbit). or while it is being restored (man), the body sets to work and quickly adjusts the composition of the plasma as near to normal as may be. This rapid regeneration of plasma doubtless concerns in the main the inorganic constituents ; the replacement of the specific proteins requires a longer time, though we have no very exact information on this point.
The regeneration of the haemoglobin which has been lost is a more gradual process. Haemoglobin loose in the circulation is useless to the higher animals, and is taken up by the tissues or excreted as a foreign substance ; there are reasons for believing, indeed, that it may be actually poisonous. Its rate of regeneration, therefore, may be limited, not only by the time required for its manufacture, but by the possibilities of proliferation possessed by the cells in which it must be contained when set free in the circulating blood . The rate of regeneration can be accurately ascertained only by methods which determine the total amount of haemoglobin in the body. Since, however, the total volume of the blood generally remains pretty steady, the concentration of haemoglobin usually indicates fairly well the total quantity also. The restoration is an orderly process, directed towards and generally attaining, as in other true regenerations, a definite and desirable end. The stimulus to regeneration is the deficiency of red cells to carry oxygen. In the rabbit a loss of about one-third to one-half of the total haemoglobin is made good in about three weeks, about 2 per cent, of the total normal amount of haemoglobin in the body,* and about 5 per cent, of the haemoglobin lost, being restored daily. As far as our information goes and it is at present meagre on this point regeneration in healthy persons is proportionally slower. That the rate of regeneration becomes slower as the normal level is more nearly approached is by no means clear. Assuming which may not be proper that the stimulus is greatest when the defect is greatest, a constant rate of regeneration through-out may be taken to indicate that the regenerative mechanism is working to its full capacity all the time, and that the power of growth of this mechanism is the factor which limits the rate of regeneration. The capacity of the mechanism may, however, be much increased by longer and more intense stimulation. Eepeated bleedings may more than double the rate, no doubt because they eventually lead to an hypertrophy of the regenerating apparatus. In adult life red blood-corpuscles arise in the bone marrow, either exclusively or to such a preponderating extent that other organs may be neglected. In view, however, of the importance of the liver and other organs as sites of haemopoiesis in fcetal life, the possibility of the extension of the area of growth in the adult under abnormal circumstances must not be altogether forgotten. As a rule in man there is no need for any such extension to transgress the anatomical limits of the bone marrow. Large parts of the marrow in normal adults are occupied by a tissue which is little more than fat ; this is especially the case in the shafts of the long bones. The apparatus for making red cells may therefore be much hypertrophied within the limits of the bones themselves, and, in fact, it is found that under circumstances when a call is made for this hypertrophy a large part or the whole of what is usually fatty marrow may be replaced by a red tissue in which active proliferation of red cells is proceeding. The bulk of this tissue may be further increased by the enlargement of the medullary cavity in the bones. It is clear from these considerations that, in our conception of the component parts of the tissue " blood," we must include the parts of the bone marrow which give rise to red cells, and also those in which leucocytes take origin. On the other hand, the whole of the anatomical marrow must not be included ; the supporting framework, the fat, the bone-modelling apparatus, are clearly not part of the blood. The rate of regeneration is therefore necessarily ultimately limited by the power of growth of the marrow. A sudden call for more red cells in the circulation leads to the extrusion from the marrow into the blood of immature forms of red cells. * These vary in immaturity from the large nucleated cells (megaloblasts), through the smaller nucleated cells (normoblasts) which are derived from them, to nonnucleated discs which show that they have been, as it were, put out in a hurry by their being too small or badly shaped, or by containing too little haemoglobin or obvious remnants of nucleus (polychromasia, basophilia). In man megaloblasts or the corresponding non-nucleated form (megalocytes) do not commonly appear in the circulating blood during regeneration after haemorrhage, though they may do so rather freely in smaller animals.
The recognition of the other forms of immature red cells is, however, of importance as affording evidence of the activity of the regenerative process. They are not, however, to be regarded as in themselves reparative. Regeneration involves the formation of normal tissue, in the case of blood of normal red cells. The object of the premature extrusion of ill-made red cells is to tide over the deficiency of circulating haemoglobin until enough normocytes can be made to take the place of those which have been lost. The immature forms may be less effective than normal cells, either by being bigger, and therefore having a relatively smaller surface for gaseous exchange, or by containing too little haemoglobin. Red cells with nuclei or the remnants of nuclei have also the disadvantage that they themselves use up much more oxygen than normal erythrocytes.* The haemoglobin in them, however, has a normal oxygen-carrying power. Regeneration may be interfered with if the growth of the marrow is obstructed. Severe haemorrhage may itself do this by limiting the supply of oxygen to the marrow cells ; the poisons developed in infections may have the same effect. A deficiency of iron much retards regeneration, though it is doubtful whether this factor is ever of much impoitance in man. Regeneration is less active in the adult and aged than .in young individuals. Regeneration of red cells which have been lost as oxygen-carriers by being destroyed inside the body follows the same lines. There are reasons for thinking that the presence of the remains of the cells stimulates regeneration. Almost all the iron and much of the original haemoglobin may be ready at hand in the body for the construction of new red cells. Experiments on animals show definitely that the iron-containing remnants of haemoglobin which accumulate in the spleen are used up during regeneration. The question as to whether the mode of regeneration is altered as well as the rate is discussed later.