42 PHYSIOLOGY CHAP. 



so as to form an adequate mixed diet ; the dogs wasted steadily 

 away ; one of them was at the point of death from exhaustion on the 

 thirty-sixth day, the other reached the same stage ten days sooner ; 

 yet we know that dogs which are entirely deprived of food can 

 live considerably longer. Forster also fed three pigeons on starch 

 and casein, which contains a very small quantity of salts ; on this 

 diet they lived only thirteen, twenty-five, and twenty-nine days 

 respectively. 



Bunge considers that the brief duration of life, when there are 

 not enough salts in the food, is due to the formation from the 

 proteins of sulphuric acid which, failing to find in the blood 

 the bases with which to saturate itself, attacks the bases of the 

 tissues and thus hastens their consumption. 



Lunin (1881) showed by special researches the probability of 

 this explanation. He fed a number of mice on food almost 

 entirely devoid of salts. He precipitated the casein and fat from 

 cow's milk with acetic acid and then washed it until it contained 

 only 0-05-0-08 per cent of salts, i.e. 10 per cent less than in 

 Forster's casein ; he added to this casein cane sugar and distilled 

 water, and found that the mice on this diet lived only eleven to 

 twenty-one days. When he added to this diet enough carbonate of 

 soda to neutralise the sulphuric acid formed, the mice lived longer, 

 sixteen to thirty days. When, on the other hand, he added sodium 

 chloride, i.e. a neutral salt incapable of neutralising the acid, the 

 mice lived about the same length of time as those in the first 

 series. But even when the sulphuric acid formed was neutralised 

 the mice did not live long on a diet devoid of salts. Further, when 

 Lunin added the ashes of milk to the purified casein, the mice did 

 not live longer than when only carbonate of soda was added, 

 whereas it is well known that mice can live for an indefinite period 

 on nothing but milk. The most probable explanation of these 

 results is that salts taken as such are not perfectly assimilated ; 

 apparently they must be combined with the protein of the food. 



This hypothesis is not, however, applicable to salts of iron, 

 which can be assimilated as such, although in common articles of 

 food they are found chemically associated with protein molecules. 

 This was proved by the interesting researches made by Coppola in 

 1890. He fed four cocks with a diet almost entirely devoid of 

 iron, consisting of a paste composed of 30 grms. of starch, 2 of 

 white of egg, 5 of gelatine, 4 of cane sugar, 0'70 of sodium 

 carbonate, 0'40 of sodium chloride, 0'30 of potassium phosphate, 

 0'15 of calcium phosphate, and 0*15 grms. of magnesium phosphate. 

 He found that on this diet, which was almost devoid of iron (0*11 

 milligrms. of iron per day contained in the gelatine), the cocks 

 continually eliminated iron in the urine and faeces ; the erythro- 

 cytes in the blood gradually became paler owing to the decreased 

 quantity of haemoglobin without, however, decreasing in number, 



