414 
ON INCREMENT OF ANIMAL HEAT. 
2. Next let me explain that the direct readings of the ther¬ 
mometer which we are able to take from the most accessible 
parts of an animal—from the open cavities—are not of them¬ 
selves accurate readings of the mean temperature of the w hole 
of the body. If we take a reading from the mouth of an 
animal, and another reading from the stomach, or if we take 
a reading: from the lower third of the rectum and another read- 
O ^ 
ing from the upper third, we shall tind a difference in favour 
of the more internal part of good 2° Fahr.; or, if we take a 
reading from the mouth or rectum, and compare it with the 
reading from one of the serous cavities, or from the centre of 
the body of a large muscle, we shall again find a difference in 
favour of the more internal part of good 2,^ F. If we take our 
readings from the outside of the body, and compare those 
readings with the readings from the internal cavities, the dif¬ 
ference will be twice greater still, or even more. For these 
reasons I myself, in research, have given up readings from 
external parts altogether; and when I read from the mouth, 
rectum, or, in the bird, from the cloaca, I feel it is necessary 
from direct observation, to add 2° F. in order to express the 
mean of the body at large. When, therefore, I speak of the 
maximum of animal heat in any given case, it will be under¬ 
stood I speak of the mean degree of heat of the whole body 
of the animal. 
3. In the third place I would point out that in culminating 
the increment of heat, however produced, in any individual 
animal, it is necessary to date, not from a general rule or 
[irinciple respecting the assumed natural temperature of the 
animal, but from direct observation made immediately pre¬ 
ceding the experim.ental investigation. It is a comnion idea 
that all warm-blooded animals possess and maintain a given 
standard of animal heat under ditierent conditions. This is an 
error which repeated experiment soon puts right. Thus I find 
in pigeons kept witli every care, well fed, well protected, there 
are variations of temperature ranging from 106° to 109*^ F. 
This range of full three degrees extends to all other animals of 
warm blood which I have studied, and we may, I think, note 
as a fact that in animal bodies there is an allowance made for 
fluctuations of temperature, an allowance for expansion and 
contraction, if we like to express the matter so, of three degrees 
on Fahrenheit’s scale. Hence, in making any given research, 
we are bound to note the actual temperature of the animal for a 
period of an hour at least before we induce unnatural increment 
of heat, and to reckon from the figure thus registered. By this 
mode of procedure we shall discover in many places natural 
differences in animals of the same kind, natural differences of 
