i EXCHANGE OF MATEKIAL 5 



simplify the task, the plan has been adopted of putting the subject 

 of the experiment on as simple a diet as possible, one composed of 

 the commonest vegetable and animal foods, which are easily and 

 completely digested and have a known and constant composition. 

 The same quantities of these articles of food are given to the 

 subject of the experiment every day. In order to obtain reliable 

 results the experiment must extend over a sufficiently long period, 

 at least a week, so that in the daily average of the intake, the 

 errors inherent in the method may not accumulate but balance 

 one another, at all events partially. 



In order to form an approximate estimate of the quantity of 

 protein contained in the articles of food chosen for the experiment, 

 the nitrogen contained therein is generally ascertained by 

 Kjeldahl's method. The amount of nitrogen multiplied by 6'25 

 gives the quantity of protein contained in the food. Since the 

 nitrogenous substances in animal and vegetable foods are not 

 all found in the form of protein, it is obvious that this method 

 does not give us the exact quantity for all the nitrogenous 

 substances introduced. 



In order to ascertain the quantity of fats contained in the 

 food, a sample is extracted with ether, and thus the total quantity 

 of fats can be calculated. This method also is by no means free 

 from error; on the one hand, ether dissolves some non-fatty 

 substances, and on the other, only dissolves free fats, leaving 

 undissolved those which are more or less closely bound up with 

 protein. 



The total quantity of carbohydrate is usually estimated 

 indirectly in the following way : a sample of the food is 

 evaporated at a temperature of 100 C., leaving the dry residue, 

 which serves for the calculation of the quantity of water contained 

 in the whole amount of food taken ; by combustion of the dry residue 

 is obtained the ash, from which the quantity of salts in the food may 

 be ascertained. Subtracting the weight of the ash from that of 

 the residue, and also the weight of albumen and fat which has 

 already been ascertained, one obtains a remainder which represents 

 the weight of carbohydrate. This indirect determination is 

 obviously the least accurate, since in it are accumulated all the 

 errors of analysis inherent in the preceding direct determinations. 



In order to estimate accurately the excreta, the urine and faeces 

 evacuated in twenty-four hours must be collected, and determina- 

 tions made of the weight, dry residue, ash, and single elements 

 (N, C, H, 0, S, P) resulting from the oxidation of the three groups of 

 organic substances in the food. It is known that nitrogen, 

 sulphur, and phosphorus are derived from protein, whereas carbon, 

 hydrogen, and oxygen arise from protein as well as from fat and 

 carbohydrate. 



These researches may be very much simplified when, for the 



