Radiant Heat by Dry and by Moist Air. 245 



tube with a small compression-pump provided with bellows 

 which could be worked by the foot, just as in a blowpipe-appa- 

 ratus. The current of air which this pump yielded divided itself 

 at the fork and entered both tubes ; it became saturated in the 

 one with aqueous vapour, and in the other deprived by the sul- 

 phuric acid of all its moisture ; consequently moist air was con- 

 ducted to one of the tubes placed between the thermopile and the 

 source of heat, and dry air to the other. In order that these cur- 

 rents of air might traverse the tubes and not at once escape at their 

 nearest open ends, the most distant lateral apertures were con- 

 nected with each other by a forked caoutchouc tube, and put in 

 communication with an ordinary air-pump. In order to show 

 the effect to several persons at the same time, I chose the objec- 

 tive representation of the position of the magnet-mirror. As 

 soon as the magnet-mirror had quite come to rest (which always 

 followed very quickly after the water in the two cubes had com- 

 menced boiling) and the pumps had been set in action, a motion 

 of the image of the slit took place on the scale, amounting to 

 about five divisions of the latter, each division being equal to a 

 centimetre. The image remained in this position as long as the 

 pumps were allowed to work (sometimes, with very short inter- 

 ruptions, for a quarter of an hour), and it returned slowly to its 

 original position when the currents of air ceased. On inter- 

 changing the caoutchouc tubes leading from the two Wohler 

 tubes to the apparatus, so that the moist and dry air changed 

 sides relative to the thermopile, a deflection of the magnet-mirror 

 took place in the opposite direction. Through the momentary 

 introduction of a metal screen on one side between the tube 

 and thermopile, it was easy to recognize that the motion of the 

 image on the scale always indicated that the absorption of rays of 

 heat by moist air exceeded that by dry air. On repeating these 

 experiments more than twenty times in the presence of several per- 

 sons, I always obtained the same result, with very slight differ- 

 ences in the magnitude of the deflections. Professors Valentin 

 and Schwarzenbach, moreover, were obliging enough to examine 

 the direction of the deflections of the magnet-mirror, and thus to 

 corroborate the above statement. The correctness thereof was 

 likewise proved by the circumstance that a motion of the image 

 of the slit in the same direction, but far beyond the limits of 

 the scale, occurred when coal-gas instead of moist air was in- 

 troduced into one of the tubes. Again, on several occasions, 

 before the commencement or at the end of the observations, 

 the Leslie cubes were removed, and the action on the thermopile 

 was observed when, without any source of heat being employed, 

 the pumps were put in action. When this was done energeti- 

 cally, a motion of the image, amounting to about one division 



