130 BULLETIN OF THE UNIVERSITY OF WISCONSIN 



The face y is certainly (0.25.4) and not the known form 

 (061), as the face is a large one and quite perfect. Both 

 y and & reflect finely, giving single sharp images of the 

 signal. The calculated angle & : t (010 : 061), is 12 59', 

 which differs from the value obtained by over 18', whereas 

 it is impossible that an error of more than a very few min- 

 utes should be made in the reading. With the exception 

 of r, which is small and gives only a faint reflection of the 

 signal, all the faces afford excellent readings. 



Type 2. Crystals with pseudo-hexagonal habit due to the 

 nearly equal development oj the pyramid and brachydome: The 

 individuals of this type (plate 7, fig. 2) are generally larger 

 than the crystals of the first type and have the habit of the 

 crystals from Berezov figured by Dana, ' but they are more 

 modified. Their habit is conditioned chiefly by the forms 

 p, -i, and &. They may be easily confused with stellate 

 twins, which occur with them and from which they can 

 often only be distinguished by a measurement of the angles 

 in the prismatic zone, the angle m : m being either 54 28' 

 or 62 46', while that of m : b is 58 37'. Most of the 

 crystals are twinned at least once, the twinning line taking 

 usually a somewhat irregular course over the faces and 

 frequently bringing p and i nearly into coincidence. On 

 single individuals the faces are all remarkably perfect and 

 give each a single sharp bright image of the signal. The 

 forms are as follows: * 



P, P (HI)- 

 m, ooP (110). 

 a, ooPoo (100). 



&, ooPoo (010). 

 i, 2Pao (021). 

 k, Poo (Oil). 

 x t tPao (012). 



' System, 6th ed., p. 287, fig. 8. 



