1888.] On the Logarithmic Law of Atomic Weights. 



115 



one part in sixteen of absolute alcohol at 14°. No hydrate was 

 obtained by evaporating its aqueous solution, but the same crystalline 

 forms were deposited as from alcohol. With barium chloride it 

 behaves similarly to the neutral salt, a partial precipitation taking 

 place only with difficulty. It possesses much greater stability than the 

 neutral salt, but like the latter it is reduced by sulphurous acid and 

 stannous chloride, and oxidised by chlorine. Potash decomposes it 

 with evolution of ammonia, but hydrochloric, nitric, or sulphuric 

 acid has only a slight action upon it in the cold. Kept in a vacuum 

 or in a current of air, it. is appreciably volatile at ordinary tempera- 

 tures. When heated strongly, a portion of it sublimes unchanged, 

 part of it is converted into ammonium selenite, while the remainder 

 is decomposed into ammonia, water, nitrogen, and a residue of fused 

 selenium. In order to estimate the amount of ammonia precipitated 

 by platinum chloride, 0'3140 gram was taken, which yielded 

 0'6234 gram of the double chloride, equivalent to 15*26 per cent, of 

 ammonia, the amount thus precipitated being equal to 81'60 per cent, 

 of the total amount of ammonia in the salt. 



Relation of the Selenosamates to Sulphur Compounds. 



It is stated that a compound is formed by the action of ammonia on 

 sulphur dioxide, but the description of its properties shows that it 

 does not correspond with the selenosamates. The latter bodies 

 correspond more closely with the compounds which sulphur trioxide 

 forms with . ammonia. The molecule, Se0 2 , therefore, in these reac- 

 tions acts similarly to S0 3 , rather than to what is usually regarded 

 as its sulphur analogue, namely S0 2 . 



In conclusion, we are engaged at present in the production of other 

 selenosamates, and hope to give an account of them at an early 

 date. 



VI. u On the Logarithmic Law of Atomic Weights." By G. 

 Johnstone Stoney, M.A., D.Sc, F.R.S. Received April 16, 

 1888. 



(Abstract.) 

 This memoir is divided into five sections. 



Section 1. — When Newlands pointed out the dependence of the 

 atomicity and other properties of some of the chemical elements upon 

 the order in which their atomic weights succeed one another, and 

 especially when this law was extended by Mendelejefi to all the 

 elements, it became manifest that there exists a mathematical relation 

 between a series of numbers and the successive atomic weights of the 

 elements. 



