TBANSACTIONS OF SECTION A. 



475 



WEDNESDAY, AUGUST ZO. ' 



The following Papers were read : — 

 1. On a Method of investigating experimentally the Absorption of Radiant 



Heat by Gases. By Professor Tait, Sec. B.8.E. (From a letter to 



SirW. Thomson). 



There are grave objections, which have been only partially overcome, to almost 

 iill the processes hitherto employed for testing the diathermancy of vapoui-s. These 

 arise chiefly from condensation on some part of the apparatus. Thus when rock- 

 salt is used, an absorbent surface-layer may be formed ; and, when the pile is used 

 without a plate of salt, the effect of radiant heat may be to cool it (the pile) by 

 the evaporation of such a surface-film. 



Some time ago it occurred to me that this part of the difficidty might be got rid 

 of by dispensing with the pile, and measuring the amount of absorption by its 

 effects on the volume and pressure of the gas or vapour itself. 



Only preliminary trials have as yet been made. They were carried out for me 

 by Professor MacGregor and Mr. Lindsay. Their object _was,7??-sf, to find whether 

 the method would work well ; second (when this was satisfactorily proved), to find 

 the best form and dimensions for the apparatus. 



The rough apparatus is merely a double cylinder, placed vertically. Cold water 

 circulates iu the jacket surrounding the curved surface and 

 bottom, and steam can be blowni into the double top. The = 

 changes in the pressure of the gas are shown by a manometer = 

 U-tube at the bottom, which contains a liquid which will not 

 attack the contents. 



This apparatus was foiu- feet long, with two inches' radius. 

 The results of a number of experiments show that it shoidd be 

 shorter and much wider. The former idea I was not quite 

 prepared for ; the latter is obvious. 



The effects on the manometer are due to five chief 



STEAM 





causes: — = 



1. Heating of the upper layer of gas by contact with lid. 



2. Cooling of the upper layer of gas by contact with sides. 

 ?>. Heating of more or less of the column by aljsorption. 

 4. Cooling of more or less of the column by radiation. 

 .5. Cooling of more or less of the coliuun by contact. 

 1 and 2 only are present in a perfectly diathermanous 



gas, as well as in a perfectly adiathermanous gas or vapour ; 



all five take place simultaneously in a partially diathermanous gas or vapour. 



The preliminary experiments show that the manometer efi'ect is only very slifiltthj 

 less for dry defiant gas than for dry air, while moist air shows a markedly smaller 

 efiect than either of the others. 



This is conclusive as to the absorption of low radiant heat by aqueous vapour, 

 but it shows also that the absorption is so small as to take place tln-oughout the 

 whole colvmin. 



Even with the present rude apparatus I hope soon to get a very accurate de- 

 termination of the absorbing power of aqueoiis vapour, by finding in what pro- 

 portions olefiant gas must be mixed with dry air to form an absorbing medium 

 equivalent to saturated air at diflerent temperatures. 



I have to acknowledge valuable hints from Professor Stokes, wlio, before I told 

 him the results I had obtained (thus knowing merely the nature of the experiments), 

 made something much higher than a guess, though somewhat short of a prediction, 

 of the truth. 



In these preliminary trials no precaution was taken to exclude dud. The results, 

 therefore, are still hable to a certain amount of doubt, as Mr. Aitken's beautiful 

 experiments have shown. 



The point of the method is that there can be no question of surfoce-layers, pro- 

 vided the operation be carried on long enough. 



