CORRELATIVES OP WATER CONTENT IN OTHER SPECIES 



249 



tion by stomach until about 25% of Bo had been retained and con- 

 vulsions began (table 28). In a few similar tests of Gomori and 

 Molnar ('32), the rabbits weighed +34% of Bq by the time con- 

 vulsions began ; correspondingly, their tissues showed greater dilu- 

 tion. While it is also possible to include data of Gomori and 

 Molnar on other rabbits given large amounts of water with pitui- 

 tary extract until convulsions began (at -j- 31% of Bq), I judge it 

 advisable to preserve the distinction between the two types until 

 the water load with pituitrin is shown to be indistinguishable from 



TABLE 28 



Dilutions (AE) of rabtit tissues, in per cent of 'Eq, after water loads were established in 

 the whole body. Values in parentheses are those clearly not significant 



that without the pituitrin. In dog and in man the two types are 

 already known to be not identical, even aside from rates of water 

 exchange. 



Among the tissues, blood manifests larger increments of water 

 content than the average tissue. Only blood and skin appear as 

 possible known depots for water. Muscle, liver, lung, and others 

 are modified significantly less both in positive and in negative water 

 loads. This is far from supporting the supposition repeatedly 

 expressed (Underbill and Fisk, '30) that ''the tissues act as reser- 

 voirs to maintain the water content of the blood at a constant level 

 through the removal of excess fluid from the blood in hydremia and 

 the release of fluid to the blood in anhydremia. ' ' Without compu- 

 tation of AE, it is difficult to appreciate how small an increment of 



