2 
PHYSIOLOGY: ARMSBY, FRIES AND BRAMAN 
basal katabolism per 24 hours has been computed separately from the observed 
rate of heat production during the intervals of lying and standing, respectively, 
and also for 12 hours standing and 12 hours lying per day, assumed as repre- 
senting average conditions. 
As was to be expected, the basal katabolism increased with the size of the 
animal but with very considerable fluctuations. The graphs of the results 
indicate an equally close relation of the basal katabolism with the weight and 
with the two-thirds power of the weight (computed body surface) and this 
conclusion is confirmed by a comparison of the coefficients of correlation as 
follows: 
Coefficients of correlation 
WITH LIVE WEIGHT 
WITH 2/3 POWER OF 
LIVE WEIGHT 
0.8655 ±0.0326 
0.9032 ±0.0239 
0.8733 ±0.0308 
0.8710 ±0.0313 
0.8548 ±0.0350 
0.8250 ±0.0415 
Computing the basal katabolism per square meter of body surface as esti- 
mated by Moul ton's formulae 4 viz., 
For unfattened animals S = 0.1186 W 5/8 
For fattened animals S = 0.158 W 5/9 
the following results were obtained. 
Basal katabolism of cattle per square meter of body surface 
LYING 24 HOURS 
STANDING 12 HOURS 
STANDING 24 HOURS 
Mean, Calories 
964.0 
1173.0 
1365.0 
Probable error of mean, Calories 
± 24.0 
± 21.4 
± 25.7 
Probable error of single result, Calories . . . 
±124.8 
±110.9 
±133.6 
Standard deviation, Calories 
185.1 ± 17.0 
164.5 ± 15.1 
198.0 ± 18.2 
Coefficient of variability 
0.1920 
0.1462 
0.1451 
A positive correlation of the basal katabolism per square meter body sur- 
face with the live weight was also found as follows: 
Coefficients of correlation with live weight 
Basal katabolism per square meter 
Lying 24 hours 0.5375 ± 0.0923 
Standing 12 hours 0.3666 ± 0.1124 
Standing 24 hours 0.2405 ± 0.1223 
The results show the marked influence of standing upon the metabolism of 
cattle, the mean 24 hour basal katabolism lying, standing 12 hours and stand- 
ing 24 hours being in the proportion of 100 : 121 : 141, the differences largely 
exceeding the probable errors. Computing, from the results per square 
