494 A MANUAL OF PHYSIOLOGY 



logical experiment; but it is very easy to determine by a separate 

 observation the water-equivalent of the calorimeter that is, the 

 quantity of water whose temperature will be raised i by a quantity 

 of heat which just suffices to raise the temperature of the metal by 

 i (p. 528). Then the water-equivalent is added to the quantity of 

 water actually present, and the sum is multiplied by the rise of 

 temperature. If the temperature of the room is constant, as will be 

 approximately the case in a cellar, any error due to interchange of 

 heat between the calorimeter and its surroundings may be eliminated 

 by making the initial temperature of the water as much less than that 

 of the air as the final temperature exceeds it. Then if the loss of 

 heat by the animal is uniform, as much heat is gained during the first 

 half of the experiment by the calorimeter from the air as is lost by it 

 to the air during the last half. Or, without lowering the temperature 

 of the water, the amount of heat lost by the calorimeter during an ex- 

 periment may be previously determined by a special observation, 

 and added to the quantity calculated from the observed rise of 

 temperature. Or, finally, two similar calorimeters may be used, one 

 containing the animal and the other a hydrogen flame, or a coil of 

 wire traversed by a voltaic current, which is regulated so as to keep 

 the temperature the same in the two calorimeters. From the quantity 

 of hydrogen burnt, or electricity passed, the heat-production of the 

 animal can be calculated. 



Of late years air calorimeters have come into vogue for physio 

 logical purposes. A diagram of one is shown in Fig. 139. Such 

 calorimeters are really thermometers with an immense radiating 

 surface, for only a small proportion of the heat given off by the 

 animal goes to heat the measuring substance. The heat required to 

 raise the temperature of a litre of air by one degree is very small in 

 comparison with that required to raise the temperature of a litre of 

 water by the same amount. Hence a given quantity of heat raises 

 the temperature of an air calorimeter much more than that of a 

 water calorimeter of the same dimensions ; and the loss of heat to 

 the surroundings being proportional to the elevation of temperature, 

 in the water calorimeter the chief part of the heat is actually retained 

 in the water, while in an air calorimeter the greater portion passes 

 through the air space, and is radiated away. When the amount of 

 heat lost by the calorimeter becomes equal to that gained from the 

 animal, the ' steady ' reading of the instrument is taken, and from 

 this the heat production can be deduced by an experimental 

 graduation of the apparatus. One advantage of an air calorimeter 

 is that it follows more closely rapid variations in the heat production 

 of the animal, or, to speak more correctly, in the heat loss. It 

 should be carefully noted that in calorimetry what is directly 

 measured is the quantity of heat given out by the animal, not the 

 quantity produced. The two quantities are identical only when the 

 temperature of the ani.nal has remained unchanged throughout the 

 experiment. If the temperature has fallen, the quantity of heat 

 produced is equal to the quantity measured by the calorimeter minus 

 the difference between the quantity in the animal at the beginning 



