862 PHYSIOLOGY 



other hand, dilution of the plasma diminishes its osmotic pressure below 

 that of the corpuscles, and water therefore passes into the latter, which swell 

 up and become spherical ; and, if the plasma be made sufficiently dilute, the 

 corpuscles burst with the liberation of the haemoglobin they contain. The 

 corpuscles of mammalian blood neither gain nor lose volume in a solution 

 containing 0-9 per cent, sodium chloride. The osmotic pressure as deter- 

 mined by the freezing-point of such a solution is identical with that of the 

 blood. For frogs' blood such a solution would be too concentrated and 

 bring about crenation. The salt solution which is normal for frogs' blood 

 contains only 0-65 per cent, sodium chloride. 



Although the average molecular concentration of blood plasma in mammals is 

 equivalent to that of a 0-9 per cent, sodium chloride solution, it may vary even in one 

 animal within fairly wide limits, as is shown by the following determinations of the 

 freezing-point of blood serum taken from animals under various circumstances : 



Man (healthy) -0-56 to -0-600 



Dog -0-55 -0-645 



Ox -0-55 -0-662 



Rabbit -0-55 -0-620 



The behaviour of the red corpuscles, when immersed in solutions of 

 sodium chloride of different concentrations, shows that its limiting mem- 

 brane or most external layer is impermeable to sodium chloride. If this 

 salt be added to defibrinated blood and the crenated corpuscles separated 

 by the centrifuge, practically the whole of the added sodium chloride 

 remains in the plasma or serum. The red corpuscle is impermeable to most 

 neutral salts as well as to cane sugar and glucose. We may therefore make 

 * normal ' solutions with sodium chloride, sodium sulphate, potassium 

 nitrate, or cane sugar, taking care that each of the solutions shall be isotonic 

 with a 0-9 per cent, solution of sodium chloride. On the other hand, a 

 solution of urea behaves towards the corpuscles like distilled water. If 

 some red corpuscles be added to a 1 per cent, solution of urea in normal 

 salt solution, they neither shrink nor swell ; and if the mixture be centrifuged, 

 and the corpuscles and supernatant fluid examined separately, the percen- 

 tage of urea in the two cases will be found identical, though there would be 

 a great preponderance of sodium chloride in the supernatant fluid. If a 

 1 or 2 per cent, solution of urea in water be added to defibrinated blood, the 

 corpuscles will swell up and burst just as if distilled water had been added. 

 There are a large number of substances to which the corpuscles are per- 

 meable, e.g. alcohol, chloroform, ether, etc. In their permeability the 

 corpuscles resemble most other vegetable and animal cells in permitting the 

 passage of all those substances that are soluble in fats and the allied sub- 

 stances, cholesterin, lecithin, and protagon, which are invariable constituents 

 of all living cells. According to Overton the external limiting pellicle of 

 the red corpuscles, as in most living cells, is formed by a lecithin-cholesterin 

 compound, whose solvent power determines the permeability of the cell by 

 foreign substances. If therefore we wish to stain the living cell, we must 



