432 Digestibilitij of Palm Kernel Cake, etc. 



Previous experience with palm kernel cake indicated that there 

 might be some difficulty in inducing the sheep to eat it satisfactorily, 

 and therefore for two or three weeks prior to the experiment proper 

 the two animals were fed on a ration of "seeds" hay and palm kernel 

 cake, the amount of each consumed during the last nine days preceding 

 Period I being adjusted accurately to the allowance decided upon for 

 this period. In all subsequent changes of feeding a transitional period 

 of nine days was allowed, the change of diet being effected in the first 

 two or three days of this interval. Each experimental period consisted 

 of twelve consecutive days. 



In order to obtain a direct comparison with some food of similar 

 composition widely used in farm practice a period was included in which 

 the digestibility of a sample of undecorticated cottonseed cake made 

 from Egyptian cottonseed was included. 



The general arrangement of the experiment is shown in the following 

 schedule : 



* Sheep No. 2 could not be used in this period owing to uQsatisfaetory food-consump- 

 tion, but later (July 18 — 28) a satisfactory period of 11 days was ol)tained. 



In an additional period the digestibility of a sample of dried yeast 

 was determined, but this will be dealt with in a separate communication. 



The average composition of the foodstuffs is given in Table I, and 

 of the composite samples of faeces for each period in Table II. 



Before proceeding to discuss the results in detail a word must be 

 said with regard to certain sources of error. 



In the determination of digestibility by experiments of this character 

 it is assumed that the faeces consist simply and solely of undigested 

 residues from the food. In actual fact, however, these residues are 

 always accompanied by appreciable quantities of metabolic products, 

 such as bile residues, mucus and epithelial waste, so that the recorded 



