544 IF. T. Schaller — Composition of Hnlsite and Paigeite. 



necessary, and for the complete elucidation of the problem 

 considerable material must be available. 



The present investigation was carried out under two great 

 disadvantages, namely, (1) paucity of material, and (2) the 

 unavoidable presence of considerable gangue in the samples. 

 For the hulsite samples, free from magnetite, less than a gram 

 was available, while several grams of paigeite, as pure as possible, 

 had to suffice. Yarious questions developed only after some of 

 the analyses had been completed, so that additional material was 

 often needed but could not be had. It was found, too, that 

 the different samples of each mineral varied somewhat in com- 

 position due to the different quality and quantity of the 

 gangue and also to isomorphous replacements in the minerals 

 themselves. All these points, developed as the analyses pro- 

 gressed, served to render the results obtained unsatisfactory to 

 the analyst. 



Analyses of hulsite. 



Three different samples of hulsite were obtained and anal- 

 yzed, the first two being the ones given in the original paper. 

 Sample No. 1 consisted of about half a gram of magnetite-free 

 hulsite, which was partially analyzed, only ferrous and ferric 

 iron, magnesia, and the insoluble matter being determined. 

 Sample No. 2 was about a gram of material, consisting of 

 a mixture of hulsite and magnetite with but a few per cent of 

 insoluble gangue. Tin was not determined in either of these 

 samples and the value for boric acid given below for sample 

 No. 2 was furnished by Wherry and Chapin.* Sample No. 3 

 was obtained by reexamining all the available material and 

 picking out a small amount of non-magnetic hulsite mixed 

 with considerable gangue. The relatively large amount of the 

 gangue that was soluble in acid complicated the discussion of 

 the results. 



Only the average of the analyses of samples Nos. 1 and 2 

 are repeated below, the fuller data being given in the original 

 paper. 



Average analyses of samples Nos. 1 and 2, hulsite. 



No. 1. No. 2. 



FeO _ -_ 33-27 34'44 



MgO 10-17 8-48 



Fe 2 3 17-83 27'64 



B 2 3 --.- 10-44 



H 2 -. .-.. 1-66 



Insol 10-00 2-24 



Sn0 2 not det. not det. 



From the analysis of. sample No. 3, the ratio of B 2 3 to Fe 2 3 

 is found to be 3 : 2. From this ratio the amount of magnetite 



* Loc. cit. 



