■o6s7 



.0654 



.0656 



.0649 



.0651 



57 



52 



54 



53 



47 



63 



52 



48 



57 



5o 



56 



52 



56 



55 



46 



55 



52 



5i 



5i 



47 



60 Mr. C. H. Lees on the 



•0653 

 61 

 60 

 54 

 53 



This is a case in which the agreement is about an 

 average one. 



The value of n found for the three bars used was 1*2 1 ; it 

 was found to vary with the nature of the surface and the 

 form of the section, but to be independent of the material 

 of the bar. It seems probable that its value is determined 

 by the nature of the surface of the bar, and by the form of 

 the stream lines of the air flowing past it. 



We can then write equation (2) with a very close degree 

 of approximation, in the form 



e *u 



This equation is evidently not satisfied by Angstrom's value 



oc -M-nX f x \ 



v = (j)(x)+ SA n e cos2?i7r( t - r- + a„ J • 



which on account of the summation can only apply to a 

 linear equation. The value of k v can be best deduced by 

 means of an experiment on the steady flow, i.e., when 



— - = 0. We then have 



ct 



which gives 



*1 *2 



qK~*~/ = / phv n dx. 



Xl X* 



if there is no " source or sink " of heat between the limits 

 x x and Xi. 



