/. W. Mallet on the Atomic Weight of Lithium. 355 



excess of nitrate of silver, the solution gently heated to con- 

 dense the precipitate, and the latter washed with very dilute 

 nitric acid and then with pure water, first by decantation, and 

 afterwards upon a filter through which all the decanted liquid 

 had been passed. When perfectly washed, the filter with chlo- 

 rid of silver was carefully dried, and the chlorid of silver 

 transferred as a single lump to a counterpoised porcelain cruci- 

 ble, upon the lid of which the filter was burned, the ashes being 

 moistened with nitric and then with hydrochloric acid. The lid 

 was placed upon the crucible, the latter was heated until the 

 chlorid of silver began to fuse, and was then cooled over oil of 

 vitriol, and weighed. 



Two experiments made by the above method gave the follow- 

 ing results : — 



(1.) 71885 grm. of Li CI gave 24-3086 grm. of Ag CI. 



(2.) 8-594:7 grm. of Li CI gave 29*0621 grm. of AgCl. 

 Now — 



24-3086 : 71885 : : 1792-94 (equiv. of AgCl) : x 



x =530-21 (equiv. of Li CI). 

 530-21-443-28 (equiv. of Cl)=86'93 = equiv. of Li and 

 29-0621 : 8'5947 : : 1792*94 : x 



x =530*24 



530-24-443-28=86-96=equiv. of Ii, • ■ 



the two numbers thus obtained for Lithium agreeing with re- 

 markable closeness. 



The difference between these numbers and those of Berze- 

 lius and Hagen is however considerable ; and as it seemed possi- 

 ble that a little chlorid of sodium still retained in spite of the 

 purification by ether-alcohol might be the cause of this differ- 

 ence, I resolved to precipitate a solution of this supposed pure 

 chlorid of lithium with carbonate of ammonia, to redissolve the 

 carefully washed carbonate of lithia in hydrochloric acid, and, 

 again evaporating to dryness and fusing, to redetermine the 

 chlorine by a slightly different method — namely, that of analy- 

 sis by measure, as applied by Pelouze to the examination of the 

 atomic weights of sodium and barium. 



3*9942 grm. of the chlorid of lithium thus prepared from the 

 carbonate were dissolved in water. 10*1278 grm. of chemically 

 pure silver (the quantity necessary for the precipitation of the 

 chlorine, if Li -=89*, and therefore not quite sufficient for the 

 amount of CI actually present) were dissolved in pure nitric acid, 

 and the two solutions were mixed in a white glass flask. The 

 mixture was gently heated, and shaken until the chlorid of sil- 

 ver had completely separated, leaving the fluid clear. A solu- 

 tion of 1 grm. of pure silver in nitric acid had been prepared, 

 and diluted until the volume=1000 cubic centimeters; 1 c. c. 

 therefore containing *001 grm. of silver. This solution was now 



