BODY TEMPERATURE OF MICE 323 



deducible in many cases from the notes made at the time.^^ jn 

 some of the later series, on the other hand, it was regularly 

 recorded. 



While the maximmn figure, in exceptional instances, was more 

 than a degree higher than the thirty-second figure, in a large pro- 

 portion of cases the two were equal, while the mean difference 

 between them, based upon 140 examples, was only 0.24°. In the 

 ensuing pages, the thirty-second figures will be used so far as pos- 

 sible, but for some purposes the use of the maximum figures will 

 be unavoidable. Such of the latter as are believed to be unreli- 

 able,^^ are designated by an asterisk. These have been rejected 

 in the computations, unless the contrary is stated. 



The number of temperature readings represented in tables 7 

 to 12 is large, and the total number of mice is likewise considerable. 

 If it were permissible to average all of the warm-room figures 

 on the one hand, and all of the cold-room figures on the other, a 

 comparison of these averages would be statistically highly con- 

 vincing. Unfortunately, however, this cannot be done. Several 

 factors influencing body temperature manifested themselves, 

 which had no relation to the temperature of the atmosphere 

 and were therefore quite ii'relevant to the chief problem at hand. 

 Such factors were the degree of excitement or activity of the 

 animal, its sex and apparently its age as well. It is possible, 

 too, that the time of day must be considered in this relation, 

 though I am not convinced that this factor had any influence 

 apart from the degree of activity of the animals. 

 . It is evident, therefore, that in order to arrive at a fair com- 

 parison of the body temperatures of the cold-room and the warm- 

 room mice, we must eluninate these irrelevant factors. In other 

 words, we can compare only mice of the same sex and roughly 

 of the same age, and — temperature aside — we must compare 

 them under conditions as far as possible identical for the two 

 lots. 



1- Where this figure has merely been calculated or inferred, it has been desig- 

 nated by a dagger (f) . Such figures are believed to be accurate, in all cases, within 

 a very few tenths of a degree. 



" For example, when they differ from the thirty-second figure by as much as 

 0.5°, or when doubt is cast upon their accuracy by the records of the experiment. 



