S56 Mr. WiLLLiAM Phillips on tht OxydofTin. 



the planes composing the pyramid of fig. 11. It will hereafter be 

 shewn in describing the apparently dodecahedral made, fig. 188. 

 that it results from a section of the prism, both in the direction 

 described by L'hermina, and in the opposite direction. Let these 

 sections be described by the dotted lines, b g dh zxv^fg c /?, fig. 11. 



Now, it may be noticed that by a practicable cleavage each way 

 through the centre of a crystal similar to fig. 11. but parallel with 

 the planes of the prism, it is divisible into four parts, similar in form 

 to the fracture described by fig. 9. On one of these portions 

 similar to that figure let the sections given on fig. 11. be repre- 

 sented by the lines bgdh and eg dh, fig. 12. and it will be seen 

 that a b c d on that figure will represent a fracture similar to fig. 4. 

 If this be pursued still further it may be observed by representing 

 the lines of section bgdh fig. 12. on fig. 13. that by the parallel 

 section b eg, a tetrahedron a b e g h obtainable. 



The fragments represented by fig. 8. were obtained by a cleavage 

 of others represented by fig. 4. in the direction of its diagonal e i f. 

 If therefore a section of fig. 9. be made in the direction of that 

 diagonal, one portion of that figure so divided, will agree in form 

 with fig. 14. which figure exactly corresponds with one fourth 

 part of a crystal represented by fig. 15. by a section along the 

 edges both of the prism and the pyramids, the planes P P and b 

 resembling each other. The planes PP and b fig. 15. also cor- 

 respond with those of PP ^, fig. 16. which planes are usually 

 supposed to arise from a decrement on the edges of a crystal 

 similar to fig. 27. PI. 16. 



I presume it has been satisfactorily demonstrated, that by the 

 fractures represented by figs. 4, 5, 6, 7, 8, and 9, PI. 15. a 

 mechanical division of the oxyd of tin is unquestionably obtainable, 

 parallel with the planes of the prism, as well as, by figs. 5, 7, and 8, 



