MINERALOGICAL NOTES: XO. III. — ANDERSON. 139 



otherwise we must suppose that the mineral contains 20 "/„ of 

 free chlorine. 



When this dithculty was pointed out to Mr. Petterd, whose 

 enthusiasm in mineralogical research is well known, he generously 

 ^ve me permission to make a new analysis of his material. 

 The result has been to show that, by some mischance, the 

 percentages for chlorine and arsenic pentoxide have been trans- 

 posed. To this accidental circumstance, and the somewhat 

 unusual crystal habit, we owe the name petterdite, which must 

 now lapse into the synonomy of mimptite, where it may perhaps be 

 retained for crystals of a habit similar to this (PI. xxxi., fig. 4). 



Two partial analyses were made. In I, -5589 gram was taken, 

 dissolved in cold, dilute nitric acid and the chlorine precipitated 

 by silver nitrate. The silver chloride obtained weighed -0557 

 gram, giving 246 "/o chlorine. From the filtrate silver was 

 removed by hydrochloric acid in slight excess and lead thrown 

 down by sulphuric acid in alcoholic solution, the precipitate being 

 subsequently treated in the crucible with sulphuric acid to 

 convert any lead chloride into sulphate. The lead sulphate 

 weighed -5861 gram, equivalent to 77-16 "/„ oxide. Alcohol was 

 evaporated from the filtrate by gentle heat and arsenic and 

 phosphoric acids precipitated together by "magnesia mixture." 

 The precipitate was approximately weighed after gentle ignition 

 (reckoned as pure \)yvo-arsenate it was equivalent to about 22 "/,, 

 arsenic pentoxide), redissolved in hydrochloric acid and a current 

 of sulphuretted hydrogen passed through the solution. A bulky 

 yellow precipitate of sulphide of arsenic came down and was 

 filtered off. Part of this was accidentally lost in oxidising to 

 arsenic acid ; the remainder was thrown down by " magnesia 

 mixture," and ignited with addition of some ammonium nitrate 

 crystals; it weighed -0803 gram, giving 10-63 "/o arsenic pentoxide. 

 The filtrate from sulphide of arsenic was concentrated and 

 phosphoric acid estimated by precipitation with " magnesia 

 mixture " ; magnesium pyro-phosphate weighed -0222 gram, equal 

 to 2-53 "/„ phosphorus pentoxide. Hence we may put arsenic 

 pentoxide as comprising 18 or 19 "/,,. 



For analysis II, -6391 gram was dissolved in nitric acid, the 

 solution evaporated on the water bath, lead thrown down by 

 sulphuric acid and filtered off. From the filtrate, arsenic and 

 the remaining traces of lead were removed by sulphuretted 

 hydrogen, treated with sodium sulphide solution, the small 

 residue of lead sulphide filtered off, converted into sulphate and 

 added to the main portion, the total weight being -6699 

 gram, making 77-12 "/„ oxide of lead. In the filtrate from lead 



