134 



J. H. Pratt — Northupite, Pirssonite^ 



4*6 : 1 : '53. The analysis is also incomplete since the bases are 

 calculated wholly as soda. 



An analysis has also been made by Penfield * on material 

 from a large crystal presented to the Brush collection by the 

 late Prof. J. S. Newberry. It was quite impure, apparently 

 owing to included clay, the analysis giving 441 per cent, of 

 insoluble material and 1*32 percent, loss on ignition. In addi- 

 tion to sodium, 2*33 per cent, of potassium was determined, 

 which is just sufficient to unite with the 2*13 per cent, of 

 chlorine to form potassium chloride. An examination of a 

 section of this crystal by Prof. E. S. Dana f showed numer- 

 ous rectangular inclusions, supposed to be either sodium or 

 potassium chloride. The material was regarded as too impure 

 to warrant the establishment of a complicated formula, and the 

 results of the analysis^ after deducting the insoluble material, 

 loss on ignition and KCl, approximated to the formula 4:^^^ 

 SO4 -h Na^COj, suggested by Mackintosh. 



In making the optical examination of the hanksite it was 

 observed by the author that although the sections, when exam- 

 ined with the microscope, showed trifling impurities, nothing 

 of an isometric character could be detected. Moreover, on 

 testing numerous crystals for chlorine, it was found to be 

 invariably present, and since the results by Penfield and Mack- 

 intosh have shown that the mineral contains an amount of 

 chlorine corresponding to over 4 per cent, of sodium or potas- 

 sium chloride, it is not possible that either of these latter 

 compounds could be present to such an extent, as an impurity, 

 without being detected with the microscope. It was sug- 

 gested, therefore, by Prof. Penfield, that new analyses, made 

 on the exceptionally pure material now at hand, might indi- 

 cate that chlorine is an essential constituent of the mineral. 



Some flat tabular crystals were, therefore, selected, and in 

 order to free them as far as possible from any impurities they 

 might contain, they were crushed and sifted to a uniform 

 grain and separated by n^eans of methylene iodide. Most of 

 the material varied in specific gravity between the narrow lim- 

 its 2*567 and 2*553, and this was used for the analysis. 



The prismatic crystals, fig. 10, derived from the 

 stratum of mud fifty feet below the surface, ap- 

 peared even purer than the above, and, fortunately, 

 enough of these had been supplied by Mr. Northup 

 for an analysis. The specific gravity was found to 

 be 2*545. 



The results of the analyses of the two samples 

 are as follows : 



* This Journal, XXX, p. 137, 1885. 

 f Loc. cit. 



