Green and Blue Ultramarine. 537 



Tliat the values as found by analysis do not agree better with 

 the values as calculated from the formulae, that is, that the consti- 

 tution of the pure blue and green comj^ound in the ultramarines 

 analysed is not quite the same, arises from the mode of forma- 

 tion, m which the temperature and the quaUty of the clay exer- 

 cise great influence. Perhaps the mode in which the analyses 

 were executed is partly at fault. It is possible that the clays 

 used for the preparation exhibit various degrees of resistance to 

 the action of acid and alkali. 



The most beautiful blue, that is, the pure compound, will 

 probably be obtained by heating a silicate of the constitution of 

 nepheliue mixed with polysulphide of sodium in given propor- 

 tions, to the right temperatm-e. Strangely enough, according 

 to present experience, green ultramarine is alone formed. If 

 this be finely powdered, and freed by washing from substances 

 soluble in water, as sulphide of sodium, sulphite and hyposul- 

 phite of soda, and then heated under access of air, and with the 

 addition of 2 equivs. of sulphur to 2 equivs. of the green com- 

 pound, 1 equiv. of the blue compound is obtained, and 1 equiv. 

 of sulphate of soda, which is separated by washing. Experience 

 can alone teach at what temperature the green compound is 

 formed, and at what it passes into the blue. In the last opera- 

 tion, access of air, as theory and experiment sufficiently show, is 

 essential. The blue compound of ultramarine is, in all its pro- 

 perties and analyses, the same as that in lapis-lazuli, and in the 

 related minerals, hauyne, nosean, and sodalite. 



Unfortunately, of lapis-lazuli no analysis exists capable of a 

 probable interpretation ; the last one made by Field corresponds 

 in some measure to the formula 



3[3(NaO SiO^) + (AP 0^ + 2Si03)] + NaS. 



But the sulphui' has certainly not been all found. 

 The formulae for nosean, hauyne, and sodalite are, — 

 Nosean [(3RO)Si03] + 3{A12 0^ SiO^) + NaO SO^ 

 Hauyne ... ... +2CaOS03 



Sodalite ... ... +NaCl 



On comparing the silicates of these three minerals, a consider- 

 able similarity is not to be disputed; for the relation of the 

 oxygen of the soda to that of the alumina and silica is as 1 : 3 : 4, 

 while the average relation in ultramarine is as 1 : 3 : 4-5. We 

 might hence assume, that in ultramarine the same silicate is' 

 contained as in nosean, &c., but in that case wc must suppose 

 that the caustic alkali used in the separation of silica and clay 

 had extracted more silica from the residual clay than arose 

 from the decomposed blue compound. 



Phil. Mag. S. 4. No. 75. Suppl, Vol. 11. 2 N 



