8 Ramsay, The Newly Discovei^ed Elements. 



I need not remind you how Lord Rayleigh's remark- 

 able observation that the density of atmospheric nitrogen 

 is somewhat higher than that of nitrogen prepared from 

 purely chemical sources, such as ammonia, or nitric acid, 

 led to the discovery of argon. Determinations of its 

 density gave the number I9"94, compared with that of 

 oxygen equal to i6. I may remark in passing that this is 

 the best standard to take for the density of gases, because, 

 if air be adopted, as used to be customary, the standard 

 will vary, inasmuch as air is a mixture of fluctuating 

 composition ; if hydrogen be chosen, there is still uncer- 

 tainty, because the true weight of a litre of hydrogen is 

 still unknown. And add to these reasons the recent 

 decision of the Committee of the German Chemical 

 Society to adopt i6 as the atomic weight of oxygen, and 

 to refer all other atomic weights to the sixteenth part of 

 this number, and it will, I think, be regarded as tending 

 towards accuracy and uniformity. 



The atomic weight of argon, however, has given rise 

 to some dispute. Its molecular weight admits of no 

 doubt ; it must of necessity be twice its density, namely, 

 approximately 40. But argon, like mercury, shows the 

 ratio between its specific heats of i '6 ; and it was argued 

 in the memoir on " Argon, a new constituent of the 

 Atmosphere," that this could point only to one conclusion, 

 namely, that its molecule must be identical with its 

 atom ; and its atomic weight therefore identical with its 

 molecular weight, 40. But to this many objections were 

 raised, chiefly on the ground that if 40 were the atomic 

 weight, then there would be no place for the element in 

 the Periodic Table, seeing that it would follow, instead of 

 preceding potassium, the atomic weight of which is 39' i. 

 The hypothesis was thrown out at the time that argon 

 might be a mixture, and contain some heavier gas of 



