408 PRINCIPLES OF ANIMAL. NUTRITION. 
dioxide produced :per:square meter by guinea-pigs at 0° C: and at 
30° C. (critical temperature), when fasting and: after the. consump- 
tion of food ad libitum. 
PER SQUARE METER OF SURFACE. 
Fasting.* Fed. 
Live Weight, || At0°C.° | At 30°C. Live Weight, | Ato°c. At 30° C. 
Grms. . CO,, Grms.| CO, Grms. Grms. | COa, Grms. | COg, Grms. 
617 27.85 12.35 © 670 . 29.49 14.10 
568 30.30 10.53 520 | 29.08 16.19 
223 30.47 12.14 240 34.07 17.69 
206 31.56 -| .13.16 . 220 30.59 «| °18.94 
Average.. -| 30.05 | 12.05 Average... "30.81 16.73 
* Already cited on p. 366. 
Comparing the averages we see that at 0° C., considerably below 
the critical point, the consumption of food didnot materially in- 
crease the total metabolism per unit of surface. On the other hand, 
at a temperature close.to the critical point the average heat pro- 
duction was increased nearly 39 per cent. by the consumption of 
food. 
It appears also that at this higher temperature the heat produc- 
tion of the fed animals was no longer proportional to their surface, 
but was relatively greater in the smaller animals. Rubner explains 
this by the supposition that (the animals being fed ad libitum) the 
consumption of food by the animals was in proportion to their fast- 
ing metabolism; that is, to their surface. Under these circumstances 
the factor of surface enters twice, and the heat production is approx- 
imately proportional to the square of the surface. 
Rubner * has also made calorimetric determinations of the heat 
production of a dog at different temperatures with the results 
shown on the opposite page. Not’ only did the feeding increase 
the heat production, but it eliminated the effect of rising tempera- . 
ture in diminishing it; that is, it lowered the critical temperature. 
CriticaL AMount oF Foop.—The very probable hypothesis of 
a substitution of the heat produced by the work of digestion for that 
* Sitzungsber. der k, bayer. Akad. d. Wiss., Math.-phys. Classe, 15, 452. 
