34 president's address. 



permanent gas. There is no difficulty at all about proving the 

 presence of this gas in ordinary rock. All that is required is to 

 heat the rock in small fragments in an exhausted flask, when 

 the gas is given off and can be measured and its composition 

 ascertained. The gas is usually contained in minute cavities 

 throughout the body of the rock, and but little of it escapes even 

 on grinding. That it must be under very great pressure is 

 evident from the volume to which it expands when liberated. 

 Sir William Crookes, in his lecture on Diamonds, delivered before 

 the British Association meeting at Kimberley, mentioned that 

 diamonds frequently explode soon after reaching the surface, or 

 on being gently warmed, owing to the pressure exerted by 

 globules of inclosed gas.* Professor Tilden has thrown much 

 light on the quantity and composition of the gases occluded in 

 rocks, in a paper read before the Royal Society some years ago.f 

 His examination covered a large number of examples of granite, 

 schist, gneiss, basalt, &c., in which he found gas varying from 

 13 to 17*8, and averaging about 5 times, the volume of the con- 

 taining rock. In a general way hydrogen was found to be the 

 most abundant constituent, but carbon dioxide was also invariably 

 present in large proportion. One series of rocks gave the follow- 

 ing average figures for the composition of the contained gas : — 



Hydrogen ... 52*2 In another and larger series the 



Carbon dioxide ... 34*1 mean composition was : — 



,, monoxide ... 8-4 



Marsh Gas ... 3-2 Hydrogen, &c. ... 73-8 



Nitrogen ... 2-1 Carbon dioxide ... 26*2 



100 100-0 



At this rate it is a simple matter to show by calculation that 

 the rocks within a very moderate distance of the earth's surface 

 contain more gas than would supply several times the volume of 

 the existing atmosphere, and if the entire mass of the earth be 



* Chem. News, xcii. 1905, p. 159. 

 tChem. News, Ixxv. 1897, p. 169. 



