1895.] Constituent of certain Minerals. 85 



total quantity of gas employed in this way was 920 c.c., and the 

 volume of the residue, unaltered further by copper oxide, was 183 c.c. 

 This large-scale experiment shows the gas to contain about 80 per 

 cent, of hydrogen. 



The density of the partially purified gas was then determined 



Volume of gas 162*843 c.c. 



Pressure (corrected) 757*6 mm. 



Temperature 2ri. 



Weight 0-0652 gram. 



Weight of 1 litre at and 760 mm. 0*4337 

 Density (0 = 16) 4'84 



A vacuum - tube filled with this sample of gas showed strong 

 flutings of nitrogen, and it was therefore further purified by repeated 

 passage over red-hot magnesium, copper, copper oxide, soda-lime, 

 and phosphorus pentoxide. This sample gave the result : 



Volume of gas 162-843 c.c. 



Pressure 704'7 mm.* 



Temperature 17'62 



Weight 0-0510 gram. 



Weight of 1 litre at and 760 mm. 0'3472 

 Density (0 = 16) 3'89 



This gas, however, when examined spectroscopically, was seen 

 still to contain nitrogen, but obviously the amount was diminished. 

 I must reserve to a later date a more accurate determination. But 

 judging by a preliminary experiment, in which a small quantity of 

 this gas was sparked with oxygen until no further contraction 

 occurred, and until it gave a spectrum absolutely free from all 

 trace of nitrogen, no great reduction in the density is to be expected. 

 The number 3'9 may, therefore, be taken as a maximum number for 

 the density of helium. 



5. Wave-length of Sound in the Gas Purified from Hydrogen, and 

 Ratio of Specific Heats. 



The tube used and the method employed for this experiment were 

 the same as those used for argon by Lord Bayleigh and myself, t 

 The half- wave-length of air in the tube at 15 was 34*45 mm. ; that 

 of the new gas (sample of density 4*84) as a mean of twelve readings 

 was 69*33 at 13*6. This gives a ratio of 1*863, which is obviously 

 too high. But it must be remarked that the heaps of lycopodium, 



* As the stock of gas was too small to fill the globe and permit the requisite 

 manipulation, a gauge was connected with the globe, and it was filled under 

 reduced pressure. 



t ' Phil. Trans.,' A, 1895. 



