ETIOLOGY OF BERIBERI. 417 



now thought desirable to ascertain whether polyneuritis would 

 develop in fowls fed on a balanced ration consisting of food- 

 stuffs which did not contain this principle. To achieve this 

 purpose, it was first necessary to test these food principles 

 separately. 



Exveriment 23. — Cottonseed oil was chosen as a digestible and 

 readily obtainable fat. Four fowls were fed on polished rice 

 and were given a daily addition of 5 cubic centimeters of cotton- 

 seed oil. 



All 4 fowls developed neuritis in twenty-two, twenty-six, 

 twenty-seven, and thirty-four days respectively. 



Egg albumin was chosen to represent the proteid element to 

 be added to the rice. Four fowls were fed on polished rice and 

 given a daily addition of 1 cubic centimeter of eg^ albumin taken 

 from fresh eggs. 



Three fowls developed neuritis after twenty-three, twenty- 

 five, and thirty-five days respectively. 



Sugar, asparagin, and inorganic salts had already been tested 

 in experiments previously reported. Having completed these 

 preliminary experiments, 4 fowls were fed on a diet consisting 

 of the following components which were given daily: 100 grams 

 of polished rice, 5 cubic centimeters of cottonseed oil, 1 cubic 

 centimeter of eg^ albumin, and 10 cubic centimeters of a solution 

 which was prepared as follows: twenty grams of saccharose, 5 

 grams of sodium chloride, 5 grams of potassium phosphate, and 

 5 grams of asparagin were dissolved in 1,000 cubic centimeters 

 of distilled water. Five grams of magnesium phosphate were 

 added to this solution which was well shaken before adminis- 

 tration. 



All 4 fowls developed polyneuritis after twenty-three, twenty- 

 six, twenty-eight, and twenty-nine days respectively. 



Before any conclusions can be drawn from this experiment, 

 it is necessary to determine whether this was a balanced ration. 

 Voit's standard for a man of 150 pounds was 118 grams of 

 proteid, 56 grams of fat, 500 grams of carbohydrate, producing 

 3,054 calories. Chittenden thought that equilibrium of meta- 

 bolism was maintained on a diet containing only 60 grams of 

 proteid with fats and carbohydrates sufficient to produce 2,800 

 calories. The fowls on which we experimented averaged 3 

 pounds in weight. Let us assume that their food requirements 

 are, proportionately to their weight, the same as that of a man. 



