46 



Journal of Agricultural Research 



Vol. Xin. No. I 



The results of Table I are shown graphically in figures 1,2, and 3 in 

 which the crosses represent the individual results, the straight lines the 

 mean katabolism computed in proportion to the live weight, and the 

 curves the mean katabolism computed in proportion to the body surface 



as recorded in Table III. 

 Both the tabulated results 

 and the graphs show rather 

 wide variations, perhaps 

 due in part to the fact that 

 the computation is one by 

 difference. As would be 

 expected, the basal kata- 

 bolism increased in general 

 with the size of the animal 

 but with very considerable 

 fluctuations. According to 

 Rubner's surface law, the 

 basal katabolism within 

 the same species is ap- 

 proximately proportional 

 to the two-thirds power of 

 the live weight. To test 

 the extent to which this 

 was true in these experi- 

 ments, the coefficient of 

 correlation ^ of the basal 

 katabolism with the live 



eOO 300 -lOd 500 



LIVE VV£IGHT(kII.0GRAM5) 



Fig. I. — Graph of the basal katabolism of cattle per 24 hours' 

 lying. 



weight and with the two-thirds power of the live weight has been 

 computed, with the results shown in Table II. 



Table II. — Coefficients of correlation 



Basal katabolism, lying 24 hours. .. . 

 Basal katabolism, standing 12 hours 

 Basal katabolism, standing 24 hours 



With live weight. 



o. 8655 ±0. 0326 



•8733± -0308 

 .8548± .0350 



With two-thirds 

 power of live weight. 



O. 9032 lb O. 0239 

 .87IO± .0313 

 .8250± .0415 



Rather high coefficients were naturally to be expected, but the results 

 fail to show any closer correlation with the two-thirds power of the 

 weight than with the weight itself, a fact which is in harmony with 

 Benedict's results upon man cited on a subsequent page. 



> The statistical computations throughout this paper follow the methods described by C. B. Davenport 

 in Chapter II of his "Statistical Methods with Special Reference to Biological Variation." (Davenport, 

 C. B. STATISTICAL METHODS. WITH SPSCIAX, RBPSRBNCS TO BtOWGICAI, VARIATION, ed. 3, p. IO-18, 



fig. 4 New York, iKJndon, 1914.) 



