1903.] Specific Heats, especially at Low Temperatures. 



181 



that the difference between these numbers and the previous numbers 

 is not, or is not in the main, due to variation in the time of trans- 

 ference. The effect of a variation in the time of transference is shown 

 in Table I. This table gives the results of fourteen experiments on 

 a copper cylinder, the time of transference being varied from experi- 

 ment to experiment. The cylinder here used was not the same as 

 that to which the previous numbers refer ; it was however cut from 

 the same casting of pure copper. It was made with a conical upper 

 end, so that it might be the less able to carry liquid air on its upper 

 surface ; its edges also were rounded off. It was always slightly 

 shaken during the transference. 



On graphical examination of the numbers in Table I, it appeared 

 that if the relation between time and apparent specific heat in Experi- 

 ments 8 to 14 be assumed linear, then the apparent specific heat in 

 Experiments 1 to 7 is on the average slightly greater than that given 



Table I. 



Apparent 

 specific heat 



0-08025 



0-0798 



0-0806 



0-0794 



0-0799 



0-0797 



0-0797 



0-0795 



0-0784 



0-0783 



0-0774 



0-0762 



0-0742 



0-0724 



Table II. 



Number 

 of experiment 



1 

 2 

 3 

 4 

 5 

 6 

 7 



X 



1 



2 

 3 

 4 

 5 



? 



10 



10,000 i/ 



800-7 

 800-0 

 799-3 

 798-7 

 798-0 

 796-3 

 794-6 



10,000 A 



+ 1-8 

 -2-0 

 + 6-7 

 - 4-7 

 + 1-0 

 + 0-7 

 + 2-4 



O 2 



