9S /Mr. jR. Phiilips on Ultramarine, [July, 



should not have alluded to this circumstance, and have sug- 

 gested the nature of the combination by which iron or its oxides 

 might, with the . other constituents, produce the blue colour in 

 question. 



The analysis of MM. Clement and Desormes, (Annales de 

 Chimie, t. 57, p. 317), shows, that although lapis lazuli may yield 

 oxide of iron on account of the pyrites it contains, yet ultrama- 

 rine prepared from it is perfectly free from any ; and before I was 

 aware that they had determined this point, 1 had arrived at the 

 same conclusion, and have repeated many of their experiments ; 

 and, as far as I have gone, my results and theirs agree. 



According to the chemists just quoted, the colouring matter 

 of ultramarine is not destroyed by a moderately strong red heat, 

 remains unchanged by ammonia, and when heated in solutions 

 of potash and soda. Acids, however, destroy the colour in a 

 few minutes, and this effect is produced even by acetic acid, as 

 well as by the nitric, muriatic, and sulphuric. They also state, 

 and correctly, that solution of sulphuretted hydrogen has no effect 

 upon the colour. According to their analysis, ultramarine con- 

 sists of 



Silica 35-8 



Alumina 34*8 



Soda 23-2 



Sulphur 3*1 



Carbonate of lime 3*1 



100-0 



It is remarkable that MM. Clement and Desormes have 

 offered no conjecture as to the nature of the colouring matter ; 

 and it was the wish to ascertain this that first induced me to 

 turn my attention to it. 



Although I have been. totally unsuccessful in attaining the 

 object of my pursuit, yet I have thought it might not be useless 

 to state the experiments which I have made, more especially as 

 the colour is extremely dear, therefore likely to be adulterated ; 

 and I am enabled to point out ready methods of determining 

 its purity, and detecting the nature of any fraudulent admixture. 



I am mclined to believe from the results of the experiments of 

 MM. Clement and Desormes, as well as my own, that the 

 colouring matter of ultramarine is a peculiar substance. I must, 

 however, repeat, that I have obtamed no direct proof of it. 

 M. Thenard, alluding to the analysis of MM. Clement and 

 Desormes, observes (Traite de Chimie, t. ii. p. 205), " Comme 

 ils ont eu, dans cette analyse, une perte de 0-8, il faut en con- 

 clure que quelques principes leur ont necessairement echappes. 

 Ces principes ne joueraient-ils pas un role remarquable dans 

 la coloration du lazulite ? Cette opinion paraitra probable, si 



