246 H. Wild on the Absorption of 



of the scale, always occurred in the opposite direction to the 

 former one; that is to say, this motion indicated that the side 

 of the thermopile upon which moist air was driven through the 

 tube was most heated. It is doubtless this phenomenon which 

 Professor Magnus also observed, and which has already been so 

 well explained by him ; nevertheless it appears to me to be 

 more than probable that, in his experiments conducted accord- 

 ing to this method, some action of the above kind prevailed over 

 the absorption and produced the result opposed to the state- 

 ments of Tyndall. The above experiment, however, removes 

 another objection which perhaps might be raised against Tyn- 

 dalPs method. It might be supposed, for instance, that in the 

 tube containing sulphuric acid, a development of heat takes place 

 in consequence of the chemical action which ensues when the 

 moist air gives up its vapour to the acid, and that this heat 

 serves to warm slightly the dry air which issues ; that, further, 

 on the other hand, the evaporation of the water in the other tube, 

 through which a current of air is passing, is accelerated, aud that 

 in cod sequence thereof an appreciable cooling takes place of the 

 moist air which issues therefrom. According to this view, the 

 observed deflection of the galvanometer-magnet would merely be 

 a consequence of the lower temperature of the moist, and the 

 higher temperature of the dry air. The same deflections, how- 

 ever, ought also to occur when, after removing the sources of 

 heat, the forcing of air through the tubes is continued; but 

 this does not take place. Nevertheless, in order to clear 

 up this point quite satisfactorily, the gases were conducted 

 in some experiments through pretty long thin metal tubes, in- 

 terposed between the apparatus and the Wohler tubes, and 

 immersed in one and the same water-bath ; this disposition of 

 the experiment caused no change, however, in the above-described 

 results. Since, lastly, in all the experiments the apparatus de- 

 scribed was placed quite freely on a table in the middle of the 

 room, all disturbing reflections of air-currents proceeding from 

 lateral screens, and so forth, are out of the question. 



After these more qualitative experiments I proceeded to exact 

 measurements. 



Second Series of Experiments. 



The apparatus was, in the next place, arranged in precisely the 

 same manner as before, with the exception that the air was in- 

 troduced into the tubes by the lateral apertures furthest from 

 the thermopile, and the nearest apertures were connected with 

 the air-pump. By this modification the disturbing action of 

 air-currents upon the ends of the thermopile was quite avoided, 

 as the following will show. The observation of the position of 



