Manchester Memoirs, Vol. xliii. {\%g(^\ No. A. ii 



the atomic weight should be identical with the molecular 

 weight, namely 4. And if that argument holds good, there 

 is room for another element with the atomic weight 20, to 

 fill a gap which should exist between those of helium and 

 argon, as will be seen from the table which follows : — 



H= I He= 4 Li = 7 Be = 9 B =ii C =12 N = i4 O =16 

 F=i9 ? =20 Na = 23 Mg=24 h\ = 2^ 81 = 28 P=3i S =32 



= 35-5 A =40 K =39 Ca =40 Sc = 44 Ti = 48 ¥ = 51 Cr = 52 



But if, on the other hand, the atomic weights of helium 

 and argon are respectively 2 and 20, there is no room for 

 any intermediate element, for, as is evident, all places are 

 filled. I ventured, at the meeting of the British Associa- 

 tion at Toronto, in 1897, to prophesy that such an element 

 should exist ; and I there gave an account of the efforts 

 which had up to that time been made to seek for it. 

 Suffice it to say, that, in conjunction with Dr. Collie, and 

 later, with Dr. Travers, helium was diffused and re- 

 diffused, with the view of separating from it such a gas, 

 with density 10 ; so far as this object was concerned, our 

 experiments were a failure, although it was proved that 

 the minerals which yield helium also give off a small 

 quantity of argon. Mineral waters and meteorites were 

 also investigated, but with no positive results; and a great 

 number of uraniferous minerals were heated, and the 

 evolved gas collected ; in no case did the spectrum show 

 any trace of lines other than those belonging to helium 

 and to argon. There was, however, still a gleam of hope, 

 for Dr. Collie and I, before beginning our experiments on 

 the diffusion of helium, had submitted argon to systematic 

 diffusion, and had found that the portion collected at first 

 had a density somewhat smaller than the last portion. 

 But as the question regarding the homogeneity of helium 

 was at that time of interest, in consequence of the remark- 

 able work of Professors Runge and Paschen, who showed 



