208 Prof. Powell on the alleged Polarization of Heat. 



the whole account given of a portion of the research at once 

 (according to my view of the matter) the most essentially im- 

 portant, and certainly the most difficult and delicate to car- 

 ry on. 



Of the nature of the results in general, so long as any light 

 is present, I have already expressed my opinion, and in these 

 cases the causes of fallacy do not act in the same proportion ; 

 but in the case of the simple radiant heat, I think it will be 

 allowed that a vertical arrangement of the apparatus must leave 

 the results open to very serious objections, and the adoption of 

 the air-thermometer will by some be thought still more ques- 

 tionable. 



In my former communication, I mentioned that my experi- 

 ments, made with a mercurial thermometer, though unsatisfac- 

 tory from the minute effects produced, appeared to confirm 

 M. Berard's as to all cases of heat accompanying light, but 

 that they failed to do so in the case of simple radiant heat. 

 With the information now before me relative to M. Berard's 

 researches, I resolved again to repeat the experiment in the 

 case of simple heat from metal below visible redness, making 

 use of an air-thermometer as he had done. I arranged my 

 apparatus as before, horizontally, the iron ball, two inches 

 diameter, was placed in the focus of a tin reflector, six inches 

 diameter, so situated that the rays would fall on the first glass 

 at the proper angle. The blackened bulb of an air-thermo- 

 meter was fixed in the focus of the small reflector attached to 

 the second glass ; the adjustment being made by the light of 

 a candle. The bulb of the thermometer was three-fourths of an 

 inch in diameter, and the air was confined by a bubble of 

 coloured liquid in the tube, as in M. Berard's experiments. 

 The tube of the apparatus passed through a thick screen, so 

 that all communication of heat to the thermometer, except 

 along the tube, was completely intercepted. Every precaution 

 being duly taken, and the second glass reflector, with the ther- 

 mometer, being in azimuth 90°, on admitting the radiant heat, 

 a small effect toas produced, and the reflector being then turn- 

 ed into azimuth 0°, and back again to 90° several times, no 

 sensible difference could be observed between, the effects in the 

 respecti vc post tio ns. 



