Composition of Staurolite. 89 



were probably quite firm and solid while the staurolite was 

 forming. The crystals, therefore, must have exerted great 

 force in crowding away the surrounding rock material in order 

 to make room for their growth, and we must take into con- 

 sideration their inability to exclude foreign matter under these 

 conditions, as well as their tendency to take it up. Large 

 crystals have surely resulted from a growth about smaller ones 

 and the beginnings of the crystals under consideration were 

 undoubtedly at the centers, where the apices of the pyramids 

 P, fig. 5, join. In the development of a large crystal from a 

 small one it is imagined that at various points on the crystal 

 faces the growth commences. The addition of particles or of 

 crystal molecules must then advance, forcing foreign matter to 

 one side until the crystal surfaces are complete. The particles, 

 however, which meet to form the edges of the crystals may 

 come together in such a way that they cannot exclude certain 

 foreign materials. It would, moreover, seem reasonable to 

 expect that the more obtuse the angle at which the faces, or 

 the crystal molecules forming the faces, meet to form an edge, 

 the less tendency there would be to hold impurities, while the 

 more acute the edge the greater this tendency would become. 

 If these conclusions are correct, then inclusions would be 

 taken up hj the edges, and being largely of carbonaceous 

 material, as in the staurolite under consideration, the result 

 would be that, in the development of a large crystal from a 

 smaller one, the inner prism I, fig. 6, as it enlarged to form II, 

 III and IY, would leave a dark deposit along the paths de- 

 scribed by its advancing edges, corresponding to the planes A, 

 B and P of fig. 5. In examining many basal sections it has, 

 moreover, been generally observed that the bars running par- 

 allel to the macro-axis, representing the impurities taken up at 

 the acute edges of the prism, are the heaviest, those parallel 

 to the brachy axis are the lightest and in some sections prac- 

 tically fail, while the outlines of the inner rhomb, represent- 

 ing the impurities taken up along the edges of 90° between 

 prism and base, are intermediate as regards the quantity of 

 included matter. Also the inclination of the phantom pyra- 

 mid P, fig. 5, seems to be wholly dependent upon the relative 

 development of the prism and base during the growth of the 

 staurolite crystal and to be in no way connected with the 

 length of the vertical axis as expressed by the axial ratio 



a '• h : c. 



The considerations given above seem sufficient to account 



for the curious arrangements of the impurities in the crystals 



under consideration and doubtless by a similar explanation the 



impurities in some andalusite crystals could be accounted for. 



Laboratory of Mineralogy and Petrography, 

 Sheffield Scientific School, November, 1893. 



