322 



SCIENCE PROGRESS 



Comparison of the data shows that the loss in fasces observed by Kellner was 

 more than three times as much as in Armsby's experiments, but the loss as 

 methane was about 50 per cent, less, and the resultant difference in metabolisable 

 energy was largely obliterated by the difference in heat-production. The differ- 

 ence in net energy per lb. of starch digested cannot, however, be lightly dis- 

 regarded. Allowing for the difference attributed to the protein, Kellner's starch 

 equivalents multiplied by 2-25 give products nearly equal to the net energy values 

 (kt) of the foods ; and for the practical purpose of calculating rations it makes 

 but little difference which of the two systems is employed. 



From a theoretic standpoint, however, the system of net energy values has 

 much to recommend it. It is simpler, independent of any meretricious hypotheses, 

 and illuminates many points that are obscured by the starch equivalent system. 

 It leads directly to a simple law of maintenance and a law of production which are 

 applicable to all animals alike. These may require some modification, but, subject 

 to a large probable error, they may be tentatively stated as follows : 



(1) The maintenance requirements (net energy) are equal to the basal katabo- 

 lism of the animal which varies as the two-thirds power of the live weight in fat- 

 free condition, and increases faster than the live weight as condition improves. 



(2) The requirements (net energy) in excess of maintenance, as defined above, 

 are equal to the total energy of the milk or meat produced. 



The following equation, deduced by the author, 1 embodies the quantitative 

 expression of the law of maintenance for ruminants : 



V = 



_ 9-187 fM(ioo-F) 



i 



}§ x 



100 



or 



9033 l 100 J~ 100 — V2tfF 

 log V = §{l°g M + log(ioo - F)} - {log(ioo - I-247.F) + 0-3407} 



M is the observed live weight (lb.), F the percentage of body fat inferred from 

 the condition, and rj is the basal katabolism. 



For horses and pigs the constants are not necessarily the same as those for 

 cattle and sheep, but the difference, if any, is probably small. It does not follow 

 that the same kinds of food are suitable for all. The diminishing returns for food 

 consumed by fattening animals may be ascribed to three contributory causes — viz., 

 (1) increased basal katabolism ; (2) slower rate of production ; and (3) in the case 

 of animals not fully grown, the larger amounts of potential energy in each succes- 

 sive pound of increase. 



Contrary to what is commonly believed, the suitability of a food for any 

 particular purpose is not determined either by the net energy or by the starch 

 equivalent. Both of these units refer to the concentration, and afford little or no 

 information regarding the quality. A complete picture of the properties of a food 

 may be obtained if the energy values are plotted on co-ordinate axes. In fig. 1 the 

 distance along ox = the total energy, and AB = the metabolisable energy. The 



1 Jour. Agric, Sa\, vol. ix, No. 2. 



