256 RESPIRATION 



combinations with a number of the constituents of the blood, both 

 inorganic and organic, and our knowledge of these combinations, 

 especially of the compounds formed with organic substances, is far 

 i'rom complete. The inquiry is complicated by the circumstance 

 that the proportion of the total combined carbon dioxide united 

 with a given constituent or bound by plasma and corpuscles respec- 

 tively is not constant, but varies with the varying tension of the gas, 

 while the total amount of carbon dioxide is itself dependent upon 

 the varying ' titratable alkalinity ' (p. 25). There is no doubt that 

 some of the carbon dioxide in blood is combined with alkali, but 

 the amount of alkali available is not nearly sufficient to unite with 

 all the carbon dioxide even in the form of bicarbonate. Some of 

 the dissociable carbon dioxide must therefore be combined with 

 organic substances. The relations, even of that portion which 

 exists as bicarbonate, are peculiar. This is sufficiently indicated by 

 the fact that from defibrinated blood the whole of the carbon 

 dioxide can in time be pumped out without the addition of an acid 

 to displace it from the bases with which it is united. On the other 

 hand, from a bicarbonate solution whose concentration corresponds 

 to that of the blood, not much more than half of the loosely bound 

 carbon dioxide (that is, the carbon dioxide which comes off accord- 

 ing to the equation 2HNaCO 3 = Na 2 CO 3 +CO a + H 2 O) can be ob- 

 tained even when the evacuation is kept up for days. This is only 

 about one- fourth of the total carbon dioxide in the bicarbonate; 

 yet, when sodium bicarbonate is added to blood, even in consider- 

 able amount, all the carbon dioxide in it can be obtained by the 

 pump. From serum a great deal, but not the whole, of the carbon 

 dioxide can be likewise pumped out, and the liberation of the gas 

 does not stop, as in the case of the bicarbonate solution, when all 

 the bicarbonate has been changed into carbonate. The residue 

 (from 10 to 18 per cent, of the whole) is set free on the addition of 

 an acid e.g., phosphoric acid. 



The most satisfactory explanation is that in the serum there exist 

 substances which can act as weak acids in gradually driving out the 

 carbon dioxide, when its escape is rendered easier by the vacuum. 

 The quantity of these, however, is not so large but that a portion of 

 the carbon dioxide remains in the serum. The proteins of the serum, 

 such as serum-globulin, behave in certain respects like weak acids, 

 and may contribute to the driving out of the carbon dioxide. When 

 defibrinated blood is pumped out, the whole of the carbon dioxide can 

 be removed, apparently because substances of acid nature pass from 

 the corpuscles into the serum and help to break up the carbonates. 

 The haemoglobin in the corpuscles acts as a weak acid, and as some 

 corpuscles are haemolyzed during the evacuation, the haemoglobin may 

 exert this action in the serum as well as in the interior of the corpuscles. 



The quantity of carbon dioxide combined with alkali (as bicar- 

 bonate) has not been exactly determined. Bohr estimated it by 

 shaking blood with atmospheric air, which was supposed to leave 



