WATSON AND WHERRY: PYROLUSITE FROM VIRGINIA 557 



different from that at the free end," although practically the same 

 forms are present at both ends, the difference being in their rela- 

 tive sizes. The mineral is therefore probably not to be regarded 

 as belonging to a hemihedral class, but to be pseudo-hemimorphic 

 along the 6-axis; the rate of accretion of material was of course 

 greater at the free ends, and less where the crystals are crowded 

 together, and this may account for the different development 

 at the two ends. It may be noted that hemihedral (in part 

 sphenoidal) arrangement of faces has been observed in manga- 

 nite from many localities, so it is evidently a subtance that is 

 especially sensitive to differences in rate of growth. This is prob- 

 ably to be explained by the presence in its space-lattice of some 

 hemihedral features, which in the structure as a whole compensate 

 each other and make the class holohedral, just as in diamond, 

 which possesses a balanced tetrahedral arrangement of atoms and 

 accordingly often simulates tetrahedral habit, although thought 

 to be fundamentally holohedral. 



Twelve crystals were measured on a Goldschmidt two-circle 

 goniometer. Most of the faces present are more or less curved 

 and striated, so that perfect reflections of the signal are rarely 

 obtained. The unit prism sometimes yields, excellent images, 

 however, and one of the side domes fairly good ones. Variations 

 in angle of half a degree or more occur from one crystal to an- 

 other, even with the faces giving the best reflections, because 

 of the subparallel intergrowth. In particular the wedge-shaped 

 crystals tend to show a somewhat greater value for axis a than 

 do the tabular ones. The measurements here accepted as char- 

 acteristic of the occurrence are those obtained from small and 

 apparently single tabular crystals with two or more faces yield- 

 ing sharp images of the signal. The prism and dome zones gen- 

 erally show long trains of reflections, in which bright nodes in- 

 dicate the positions of faces, and as these nodes reappear from 

 one crystal to the next with considerable regularity, they un- 

 doubtedly represent actual forms. 



" This habit has apparently not been noted in normal manganite heretofore, 

 but appears in a diagram of pyrolusite given by Dana (System of Mineralogy, 

 6th ed., fig. 3, p. 244.) 



