1820.] Br. Thomson's Analysis of Isle edlest one. 411 



small portion of white precipitate appeared. It weighed when 

 dry 0'3 gr. and was phosphate of magnesia. It is equivalent to 

 about 0-1 gr. of magnesia. 



From the preceding analysis, it follows that the Dumbarton 

 needlestone is composed of the following constituents : 



SiUca 18'4 or per cent. 36*8 



Alumina 15-68 31-36 



Lime 7-7 15-4 



Magnesia O'l 0*2 



Peroxide of iron . . 0*3 0'6 



Water 6-5 13-0 



48-68 97-36 



I suppose the sulphate of lime which made its appearance 

 durino- the analysis to be derived from a small trace of sulphuric 

 acid contained in the nitric acid employed. This acid was not 

 free from muriatic acid. Hence doubtless the origin of the 

 muriates of lime and iron held in solution by the ammoniacal 

 liquid. 



It is obvious enough from the preceding analysis that the 

 Dumbarton needlestone can neither belong to the species natro- 

 lite nor skolezite. It contains no soda, which is an essential 

 constituent of natrohte. The lime is nearly the same in it as 

 exists in skolezite ; but its proportion of silica is nearly a fourth 

 less, while its alumina is somewhat more. It is obviously a 

 combination of three atoms of silicate of alumina and one atom 

 of silicate of lime, together with 24- atoms of water ; so that its 

 symbol is 



3 A S + C S + 2i Aq. 



This is a much simpler and more satisfactory combination than 

 either natrolite or skolezite. 



It seems from the observations in the preceding part of this 

 paper to follow that the mesotype of Haiiy must hereafter be 

 divided into at least three species ; namely, natrolite, skolezite, 

 and thomsonite, the constitution of which is- as follows : 



Natrolite 3 Al S + S S^ + 2 Aq. 



Skolezite 3 Al S + C S' -}- 3 Aq. 



Thomsonite 3 Al S + C S + 24. Aq. 



The difference between the two last species consists chiefly in 

 the silicate of lime. In skolezite there exists one atom of trisili- 

 cate of lime; but in thomsonite the combination is a simple 

 silicate. As to the very minute quantity of iron and magnesia 

 present, it can only be considered as mechanically mixed, and 

 not chemically combined. 



As for the mesolitc, it seems still doubtful whether it is to be 

 considered as a distinct species, or only a mechanical mixture of 



