378 



On the Heat in the Sun's Rays. 



portion or nearly so to the difference between its temperature 

 and that of the medium in which it is placed. 



I regard it therefore as well established that the differential 

 thermometer affords a correct measurement of the differences be- 

 tween the heat of the canister and that of the surrounding air. 

 These differences may evidently be varied in two ways: by 

 changing either — 



1st. The heat of the canister ; or — 



2dly. The temperature of the air. 



An increase or diminution in the heat of the canister would 

 directly increase or diminish the differences ; whilst an increase 

 in the temperature of the air would diminish the difference until 

 an equalit}?- between the two was obtained. If the temperature 

 of the air were uniform and the changes were those of the canis- 

 ter alone, the instrument measuring the differences would cor- 

 rectly indicate those changes. But if the heat of the canister 

 were uniform and that of the air were varied, then would the in- 

 strument equally indicate those changes, but in a contrary direc- 

 tion. In case the heat of both the canister and the air was varied 

 at the same 'time, if we knew the change in one and its effects 

 upon the instrument, we could easily deduce the changes in 

 the other. Suppose, for example, an increase of ten degrees on 

 the scale of the instrument and an elevation of five degrees in 

 the temperature of the air ; the effect of the latter having been 

 to depress the thermometer five degrees, and the canister having 

 not only overcome that effect but increased the indications ten 

 degrees, the sum of the two or fifteen degrees would be the real 

 change which had taken place in the heat of the canister. Had 

 there been a depression in the temperature of the air, it obvi- 

 ously should be subtracted from the indications of the instru- 

 ment to obtain the desired measurement. 



»It is upon these principles that I have applied the differential 

 thermometer to measure the comparative heat in the sun's rays. 

 One of its bulbs received their direct action in the same way 

 that it received the rays proceeding from the canister. The tem- 

 perature of the air was at the same time obtained by a common 

 thermometer. An increase was added to, and a diminution sub- 

 tracted from, the indications of the instrument to obtain the real 

 changes in the heat of the rays proceeding from the sun. 



My first experiment was of the simplest kind. It was a win- 

 ter's day. The differential thermometer was placed on the out- 

 side of a window where the temperature was below the freezing 

 point. The effect measured by the scale (which merely divided 

 the stem into equal parts) was 53°. It was then placed on 

 the inside of the window where the temperature was about 

 70°, and to my surprise the effect rose to 115°. The experiment 

 was many times repeated with similar results, although varying 



