M. Walmstedt's Researches on the Composition of Peridot. 357 



ever, if I have misunderstood Mr. Davies, I beg to offer him 

 my best apologies, and to assure him I had no intention of 

 laying a greater stress on his language than I thought it was 

 intended to convey. p^ q_ 



LVIII. Researches on the Composition of Peridot. By 

 L. P. Walmstedt. 



CINCE the application of the electro-chemical theory and 

 ^of the doctrine of chemical proportions to mineralogy, a 

 far surer criterion of the greater or less accuracy of chemical 

 analyses has been gained ; and it has consequently been found 

 that some minerals with whose composition we seemed pre- 

 viously to have been pretty well acquainted, require in fact a 

 new examination. Among these is peridot. 



It is well known that Haiiy upon crystallographic grounds 

 united into one species the two fossils separated by Werner, 

 (chrysolite and olivine), and named them peridot. This deci- 

 sion of crystallography ought first to receive its full con- 

 firmation from chemistry ; for, with the exception of Haiiy's 

 regular bodies (as they are called), there has not yet been 

 found to exist a similaritj' of primitive form with essential 

 differences in the component parts. This confirmation is in 

 vain sought for in the specimens of peridot at present known, 

 which rather evince the contrary ; for, if we except Achard's 

 analvsis of the chrysolite, and Gmelin's examination of the 

 olivine, the results of which entitle us to suppose that the mine- 

 rals they had in their hands were very different from what they 

 named them, all the other known specimens correspond in 

 some degree with the formula p- I S for chrysolite, and j, I Sl^ 

 or J,, \ S^ for olivine. 



I undertook the following investigation, which I submit to 

 the intelligent judgement of the Royal Academy, partly on 

 account of the contradiction between theory and analysis, 

 partlv for the sake of ascertaining the real nature of the 

 mineral in the meteoric iron of Pallas which resembles olivine. 

 As the small number ot the different peridots which I had 

 at command did not always permit a repetition of the analysis, 

 I was obliged, in each experiment, to pursue the same course, 

 in order to make the results more correspondent. After the 

 washed powder of the stone was dried for about ten minutes 

 in a small ap))aratus filled with carbonic acid gas, to prevent 

 the further oxidation of the iron iiom the fire, which was 

 at the j)oint of ignition, between one and two grannnes were 

 weighed out for analvsis. The stone was fused with four times 



its 



