Penfield and Ford — Stibiotantalite. 75 



What it is desired especially to bring out by the formulas 

 is that in stibiotantalite two univalent antimonyl radicals 

 (G=Sb) play the same role as a bivalent atom of Fe in colum- 

 bite, hence, chemically, the mineral is regarded as a basic anti- 

 mony tantalate, respectively, niobate. Following ordinary 

 methods, the formulas would be written (SbO) 2 (Ta,Nb) 2 6 and 

 Fe(Ta,Nb) 2 6 . It may be that the similarities in habit, cleav- 

 age and axial ratios between stibiotantalite and columbite, as 

 pointed out earlier in this paper, are mere matters of accident, 

 but, taken in connection with the chemical formulas, this does 

 not seem at ail probable. It is assumed that the antimony- 

 oxide radical, known in chemistry as antimonyl, plays the 

 part of a metal in stibiotantalite, replacing the hydrogen 

 atoms of tantalic, respectively, niobic acid. This tendency 

 may be noted in a number of compounds of the antimony- 

 bismuth group, where we have basic salts containing the 

 antimonyl (SbO) and hismuthyl (BiO) radicals. If the crys- 

 tallographic relationship to columbite is disregarded, the for- 

 mula of stibiotantalite may be simplified to (SbO)(Ta,Nb)0 3 , 

 or, looked at in another way, to Sb(Ta,ISfb)0 4 , the latter an 

 antimony salt of normal tantalic, respectively, niobic acid. 

 Against the latter assumption, however, it may be argued that 

 normal salts of tantalic and niobic acid are almost unknown, 

 and the tendency to form a basic salt* containing the anti- 

 monyl radical would probably be greater than to form an 

 antimony salt of the normal acids. Among minerals which 

 are normal salts of tantalic and niobic acids, there are only 

 two closely related species, fergusonite and sipylite, (Y,Er,Ce) 

 (Ta,Nb)0 4 and Er(Ta,Nb)0 4 , but these may as well be written 

 as basic salts, for example, (ErO) 2 (Ta,Nb) 2 6 , conforming to 

 the columbite type of formula. 



That the antimony in stibiotantalite is trivalent and not 

 pentavalent is known by the summation of the analyses and 

 also by the results of the following experiment : — Some 

 powdered mineral was dissolved in hydrofluoric acid, and the 

 precipitate thrown down by hydrogen sulphide was collected, 

 dried and tested by heating in a closed tube. Had the anti- 

 mony been pentavalent, the precipitate would have been 

 either Sb 2 S s or a mixture of Sb 2 S 3 and sulphur, and have 

 yielded an abundant sublimate of sulphur when tested in a 

 closed tube ; it gave, however, only a trace of sulphur, the 

 same as stibnite, Sb 2 S 3 , when similarly heated. 



By means of the data given in connection with the three 

 analyses which have been made, it is possible to represent the 

 relations between the specific gravity and chemical composi- 

 tion of stibiotantalite graphically, as shown in figure 18, where 

 the specific gravities are taken as abscissas and per cents as 



