546 IF. T. Schaller — Composition of Hulsite and Paigeite. 



less than 4 per cent CaO which may belong to the borate min- 

 eral. Assuming this to be so, the ratio of (FeO, MgO, CaO) 

 to Fe 2 3 would become 11-9:1 or nearly 12:1. From the 

 analysis of sample No. 3, given above, we obtain for the for- 

 mula of hulsite, then, either 10£KO.2Fe 9 O,.SnO a .3B a O,+?H 9 O, 

 or regarding the 4 per cent of CaO as belonging to the mineral, 

 the formula 12R0.2Fe 2 3 .Sn0 2 .3B 2 3 + ?H 2 0. 



Discussion of Formulae of Hulsite. 



From the results of the analyses of the three samples of hul- 

 site, as just given, there are derived the following partial 

 formulae for the mineral. 



Partial Formulas for Hulsite. 



EO Fe 2 3 Sn0 2 B 2 O s H 2 



From Sample No. 1 13-2 2 



" 2 12-4 2-0 .. 3-0 1-9 



" 3 ) ... 10-5 2-0 1-0 2 8 .. or 



" 3 \ ... 12*0 2'0 1-0 2-8 



From this compilation, incomplete as it is, the most probable 

 formula derivable from the results obtained is 12R0.2Fe o 3 . 

 lSn0 2 .3B 2 3 .2H 2 or (Fe,Mg) 12 Fe 4 SnB 6 H 4 31 . In consider- 

 ing the relations of the various elements, it becomes almost 

 impossible to consider hulsite as a borate of iron (ferrous 

 and ferric), tin and magnesia, as the bases are present in 

 overwhelming amount. Therefore, the ferric iron may be con- 

 sidered as present as an acid, playing the same, role as the 

 borates. On this assumption the above formula mav be writ- 

 ten : B // 9 (B0 3 ) 6 .[R // 3 [Sn(OH)] /// H 3 (Fe0 3 )J. The tin in hulsite 

 is in a condition which is so easily soluble in dilute acid that 

 it seems more probable to regard it as present as some form of 

 stannic acid. The presence, however, of stannic acid, in com- 

 bination with boric and possibly some iron acid, leads to a 

 complicated subject on which it is best to, at present, pass by. 



Analyses of Paigeite. 



For paigeite, three different samples were prepared and 

 analyzed. The amount of gangue varied slightly, but no such 

 large variations in the amount were encountered as with hul- 

 site. Sample No. 1 was the sample on which the first analyses 

 were made. Tin was not determined in it, and the value given 

 for B 2 3 is the one determined by Wherry and Chapin. The 

 average of the analyses is : 



FeO 44-48 



MgO 1-44 



Fe 2 3 ... 16-72 



H.O 2-03 



B 2 b 3 .._ 9-83 



Insol. 15-65 



