M. R. Bunsen on the Law of Absorption of Gases. 125 



The value of the absorption-coefficient, as obtained from these 

 elements by equation No. 1, is 



"=^(W- V ')= 001448 - 



The accuracy of these determinations depends in a great mea- 

 sure upon the water employed being perfectly free from air. 

 This is best accomplished by boiling the water briskly for several 

 hours, and then causing it to pass while still boiling into a bottle, 

 the neck of which has been drawn out to a fine point before the 

 blowpipe. The water is then again boiled for half an hour in 

 the bottle, and the end of the tube hermetically sealed, access of 

 air being prevented during the closing by pressing a caoutchouc 

 tube attached to the extremity. When the water thus freed 

 from air, which should strike against the glass like a water- 

 hammer, is required for experiment, the end of the drawn-out 

 tube moistened with solution of corrosive sublimate is broken 

 under mercury, and the water admitted directly into the absorp- 

 tion-tube. Before admitting the water, care must be taken that 

 no bubble of air appears in the bottle, which would show imper- 

 fect boiling. The following coefficients of absorption have been 

 thus determined from experiments made in my laboratory, chiefly 

 by Dr. Pauli*. 



1. Nitrogen in Water. 



The gas was prepared by passing dry ammonia and air, freed 

 from carbonic acid, over glowing copper turnings. 



By combination of the experiments 1, 2, 3 ; 2, 3, 4; and 

 3, 4, 5, we obtain the interpolation formula, 



c= 0-20346-0-00053887* + 0-00001 1156/ 2 . . (19) 



By the help of this formula the following Table is calculated : — 



* Tliis talented young chemist, who most zealously assisted me in the 

 Experimental part of this research, has since, to my deep regret, been 

 removed by death from science and from his friends. 



