CHAP, i.] BLOOD. 33 



disc by a minimal amount of stroma. Hence whatever effect the 

 stroma per se may have upon the plasma, this, in the case of 

 mammals at all events, must be insignificant : the red corpuscle 

 is practically simply a carrier of haemoglobin. 



25. The average number of red corpuscles in human blood 

 may be probably put down at about 5 millions in a cubic milli- 

 meter (the range in different mammals is said to be from 3 to 18 

 millions), but the relation of corpuscle to plasma varies a good 

 deal even in health, and very much in disease. Obviously the 

 relation may be affected (1) by an increase or decrease of the 

 plasma, (2) by an actual decrease or increase of red corpuscles. 

 Now the former must frequently take place. The blood as we 

 have already urged is always being acted upon by changes in the 

 tissues and indeed is an index of those changes ; hence the plasma 

 must be continually changing, though always striving to return to 

 the normal condition. Thus when a large quantity of water is 

 discharged by the kidney, the skin or the bowels, that water comes 

 really from the blood, and the drain of water must tend to dimin- 

 ish the bulk of the plasma, and so to increase the relative number 

 of red corpuscles, though the effect is probably soon remedied by 

 the passage of water from the tissues into the blood. So again 

 when a large quantity of water is drunk, this passes into the 

 blood and tends temporarily to dilute the plasma (and so to dimin- 

 ish the relative number of red corpuscles), though this condition 

 is in turn soon remedied by the passage of the superfluous fluid to 

 the tissues and excretory organs. The greater or less number 

 of red corpuscles then in a given bulk of blood may be simply 

 due to less or more plasma, but we have reason to think that the 

 actual number of the corpuscles in the blood does vary from time 

 to time. This is especially seen in certain forms of disease, which 

 may be spoken of under the general term of anaemia (there being 

 several kinds of anaemia), in which the number of red corpuscles 

 is distinctly diminished. 



The redness of blood may however be influenced not only by 

 the number of red corpuscles in each cubic millimeter of blood 

 but also by the amount of haemoglobin in each corpuscle, and to 

 a less -degree by the size of the corpuscles. If we compare, with 

 a common standard, the redness of two specimens of blood un- 

 equally red, and then determine the relative number of corpuscles 

 in each, we may find that the less red specimen has as many 

 corpuscles as the redder one, or at least the deficiency in redness 

 is greater than can be accounted for by the paucity of red cor- 

 puscles. Obviously in such a case the red corpuscles have too 

 little haemoglobin. In some cases of anaemia the deficiency of 

 haemoglobin in each corpuscle is more striking than the scantiness 

 of red corpuscles. 



The number of corpuscles in a specimen of blood is determined by 

 mixing a small but carefully measured quantity of the blood with a 



F. 3 



