ATOMIC WEIGHTS. 539 



The result deduced from between forty and fifty experi- 

 ments is that 100,000 volumes of oxygen unite with 

 200,245 volumes of hydrogen to form water ; or in other 

 words, equal volumes of these gases contain at ordinary 

 temperatures and pressures not exactly an equal number of 

 molecules but in the ratio of 200,245 of oxygen to 200,000 

 of hydrogen. Applying this to the density of oxygen found 

 by Lord Rayleigh to be 15*882 we get for the atomic weight 



of oxygen-^ = is "862; Dittmar and Henderson found 



/s 100-1225 J 



1 5 *866, Cooke and Richards found 1 5 '869. Leduc ( 1 6) found 



for the ratio of the densities 15*905, and for that of the 



volumes 1 : 2*0037, giving, therefore, the atomic weight of 



oxygen a number a little higher, viz., 15*876. 



Professor Cooke (18) in 1S89 found the density of 

 hydrogen referred to air "06958, and comparing this with 

 Regnault's density of oxygen 1*1056, deduced the value 

 15*891 for the density of oxygen referred to hydrogen, 

 and this gives 15*8715 for the atomic weight of oxygen. 

 It is much to be regretted that Professor Cooke did not 

 at the same time redetermine the density of oxygen by his 

 method. 



With regard to the composition of the salts of ammo- 

 nium and their evidence as to the relative atomic weights of 

 hydrogen and oxygen it was pointed out by Stas at the end 

 of his paper (19) of i860 that if O = 16, NH 4 = 18*06 

 and N = 14*041, that is, as he says, we have 4*02 in- 

 stead of 4*00 for H 4 . He continues : " L'ensemble de 

 mes recherches me porte a croire que l'erreur existe plutot 

 sur le poids atomique de l'hydrogene que sur celui de 

 l'azote. 



" Si ce fait, sur lequel j'attire, sous toute reserve, l'atten- 

 tion des chimistes, est exact, ce que je compte rechercher 

 bientot, en reprenant la synthese de l'eau par une methode 

 nouvelle, il en resultera que la base sur laquelle le docteur 

 Prout avait etabli sa loi, manque elle-meme de fonde- 

 ment." 



Stas in his paper of 1876 gives 100 parts of silver as 

 equivalent to 49*5987 parts of ammonium chloride and 



