1896.] Helium ami the associated Gases exist hi Minerals. 223 



that though iron absorbs helium, it absorbs it less freely than 

 hydrogen. 



Similar experiments with palladium foil gave a gas consisting 

 chiefly of hydrogen with small quantities of nitrogen and a carbon 

 compound, but no helium recognisable by the spectroscope. This, of 

 course, does not prove that no helium had been absorbed, but it 

 shows that it is taken up much less readily than hydrogen. 



Being now desirous of ascertaining whether a simple silicious 

 mineral, containing at most minute quantities of any heavy metal, 

 would absorb the helium-containing gas from cleveite, I chose a piece 

 of red Peterhead granite. In order to prepare it for the experiment, 

 the granite was powdered, placed in a bulb attached to the Spree gel 

 pump, and heated gradually by a flame till the glass began to soften. 

 Being aware that granite and other rocks commonly contain cavities 

 charged with gas, I expected to obtain a considerable quantity of 

 carbon dioxide. I was, however, rather surprised to find that the 

 gas given oif from this sample of granite contained a large propor- 

 tion of hydrogen. The following are the details of one experiment : — 

 15 2 grams of granite, in volume equal to 5 7 c.c, was heated first by 

 a Bunsen flame, and finally by a blowpipe, till the glass of the bnlb 

 softened. It gave 14 9 c.c. of gas measured at 17°, which appeared 

 to be free from water. This gas consisted of 3*7 c.c. C0 2 and 11*2 c.c. 

 H 2 , or 



Carbon dioxide 24'8 per cent. 



Hydrogen 75*2 „ 



No helium could be detected in the original gas nor in Ihe residual 

 air left after explosion. 



A section of this rock was prepared for me by Professor Judd, to 

 whom I wish to offer my best thanks, both for the specimen and for 

 his valuable assistance. Under a high power (one-eighth inch) the 

 section exhibits myriads of minute, irreguLirly shaped cavities, in 

 some of which liquid is clearly discernible. This liquid is probably 

 carbon dioxide, though, from the minuteness of the cavities, we were 

 not able to make any estimation of the critical point of the liquid. 



After this granite had been deprived of gas as completely as 

 possible, it was heated and cooled alternately in contact with the 

 helium mixture under a pressure of about 7 atmospheres, as already 

 described. The mineral, however, took up none of the gas, for, after 

 pumping out the excess of gas in the cold, the falling mercury con- 

 tinued to sound sharply, while the temperature of the mineral was 

 gradually raised to a red heat. 



From the results of the experiments recorded in this paper, it 

 seems that helium exists in the minerals in which it is found in a 

 condition comparable with that in which hydrogen is associated 



