EXPERIMENT STATION BULLETINS. 559^ 



The above table adds materially to the evidence of great variation and 

 also demonstrates that this variation may be very sudden. Here we are 

 inclined to criticise operators who have made a practice of taking only 

 one or two temperatures. These sudden and decided changes cannot be 

 taken into account when only a single temperature is considered, and 

 it is difficult to understand how a good judgment may be formed from 

 so limited data. The greater the number of normal temperatures we 

 have of an animal the better may we be able to discover any influence of 

 tuberculin. Of course, there is a limit to practicability. The expense of 

 a test must be taken into account, accordingly the number of normal 

 temperatures must be confined to a definite period. The rule should 

 be made, take as many normal temperatures as it is feasible to take. 

 Keep all the records and compare them each time. It may be that 

 the different seasons of the year exert an influence or that a cold day or 

 warm day has its power in producing a change. For these reasons we shall 

 make a comparative study of seasons as to their effect in modifying tem- 

 perature in animals. 



INFLUENCE OF SEASONS OR ATMOSPHERIC TEMPERATURE UPON TEMPERA- 

 TURE OF ANIMALS. 



All the material in the tables will not be used to demonstrate this point,, 

 but will confine my attention to Table IX of normal temperatures taken 

 September 3d and 4th, 1897, during very warm weather, and the normal 

 temperatures of the same animals taken March 28th and 29th, 1898, when 

 the air was quite cool but not cold. While the following table does not 

 comprise all the data, it will give sufficient to establish this point, 

 whether atmospheric temperature has any influence or not upon animal' 

 heat. A further study of other tables conveys the same testimony, but 

 too much space would be required to bring all the facts together at this* 

 place. We shall, therefore, content ourselves with a single table. 



