Crreen and Blue Ultramarine. 535 



phide of hydrogen is evolved, which is recognizable by its odour. 

 The changes which ultramarine undergoes by treatment with 

 hydrogen, by ignition alone, or mixed with nitre and soda, prove 

 sufficiently the presence of polysulphide of sodium. 



From the last-named experiment, it is evident that the sulphide 

 in green ultramarine is a lower one than that in blue, for the 

 melted mass becomes first colourless and then white. So likewise 

 green passes into blue when ignited with soda and sulphur in 

 the proportion to form polysulphide of sodium, or with sulphur 

 alone, or heated for itself in the air. 



In all these three cases the polysulphide of the blue is formed 

 from the bisidphide of the green ultramarine. In the first two 

 cases the change is efi"ected by the simple taking up of sulphur ; 

 in the last case, a part of the silica of the silicate withdraws 

 sodium from the bisulphide, which is oxidized by the air, and 

 polysulphide results. 



The Silicate. — This second constituent of ultramarine consists 

 of silica, alumina, and soda. A consideration of the analyses of 

 ultramarine, shows a remarkable agreement in these bodies. I 

 calculated the quantity of oxygen contained in them, and found 

 it to be as follows : — 



Silica. Alumina. Soda. 



The quantity of oxygen in the soda, as being present in small- 

 est quantity, was taken as = 1 . 



These relations agree best with the composition of a natural 

 silicate, ncpheline, in which the oxygen ratios for silica, alumina, 

 and soda, are — 



4-5 : 30 : 1, 



for the formula of nepheline is 



(2NaO) Si03) + 2(A12 0^ SiO^). 



In order to ascertain in what equivalent relation sulphide of 

 sodium is united with the silicate, with a view to obtain the for- 

 mula for chemically pure ultramarine, the oxygen of the soda in 

 the silicate is to be divided by tlie oxygen contained in the 

 soda corresponding to the sulphide of sodium ; there is then 

 obtained : — 



