182(5.] Mr, Ritchie* s Experiments on Radiant Heat, 15 



the coated bulb ; and the coating being removed, the action is 

 entirely on the plain bulb. 



In one instance the iron was so placed that there was a slight 

 action on the plain bulb ; the coating being removed this action 

 was much greater. 



All the trials I made in this way tend to confirm the naked 

 fact observed by Mr. Ritchie ; but with respect to the explana^ 

 tion he gives of it, I feel obliged to differ from him : the pecuhar 

 property which he would thus ascribe to radiant heat does not 

 appear to me a necessary inference from these experiments. The 

 result, I think, may be accounted for on another principle which 

 will not require any new supposition ; this is the circumstance 

 of the expansion of the bulb by heat. When half coated, its 

 radiating power is obviously increased ; hence it will cool faster, 

 the expansion of the glass will be less, and consequently the 

 apparent initial expansion of the inclosed air greater than when 

 it is plain. But though it is easy to conceive an effect of the 

 nature described may be explained on the supposition of an 

 expansion of the bulb ; still a question may arise, and perhaps 

 some hesitation be felt, as to whether this cause could act to 

 so large an amount. In order to remove all doubt on this point, 

 and especially as this is a source of fallacy in many experiments 

 of a similar kind, I thought it advisable to give it a further 

 examination. 



This was done by observing the indications of the instrument 

 during the process of cooling, after both bulbs had been equally 

 heated by contact with a hot substance. The liquid was 

 stationary before experiment, and continued so on the application 

 of the heat ; on removing the hot body (both bulbs being plain) 

 in several repetitions, it remained stationary ; in others there 

 were slight oscillations. 



The same thing was repeated with one bulb blackened as ' 

 before, when there was always a considerable motion towards 

 the plain bulb. 



In some other trials, only one bulb was heated to a given 

 point by conduction, and then left to lose its heat by radiation, ' 

 both when plain and when coated. The following is one set of' 

 indications : — 





I 



