REPORT OF THE DIRECTOR I9II 



18; 



the last being new to the species. The habit is more pyritohedral 

 than either of the former types and the zone [loo . iii], ahhough 

 composed of small modifying planes, is rich in forms. The occur- 

 ring forms were identified by their zonal positions as follows : 



LETTER 



ANGLE 



NO. 



MEASURED 



CALCULATED 







Z:ne [100 



III] 





a : m 

 : m 



: n 



■-? 

 : n 



: 



100 : 311 

 : 522 

 : 211 

 : 322 



' :755 

 : III 



14 

 6 

 8 



7 

 6 



17 



25° io|' 



29 41 



35 15^ 

 43 18 

 45 23 



54 43 



25° 141' 

 29 30 



35 16 

 43 19 

 45 18 

 54 44 





Z ne [210 . 



III] 





m : e 



e : 5 



: 



311 : 210 



210 : 321 



: III 



3 

 3 



4 



19 19 

 17 6i 

 39 12 



19 17 



17 i| 

 39 14 





CHRYSOBERYL FROM ST NICHOLAS AVENUE 

 In June 1910 the writer received for identification from ^Ir James 

 J. Manchester, a small specimen which proved to be chrysoberyl 

 from a new locality on Manhattan island. The specimen in question 

 was collected by Mr IManchester from a building excavation at the 

 corner of St Nicholas avenue and 164th street. It consists of a 

 single sm.all transparent crystal of chrysoberyl embedded in I\Ian- 

 hattan schist. The crystal, which is shown in figure 5, measures 

 5 mm by 8 mm, is light yeUowish green in 

 color and is so embedded that about one-half 

 of the prismatic zone is exposed. On the 

 partly exposed end traces of terminating 

 planes were noted, but these were so rough 

 and indefinite that no terminating forms 

 could be identified. Measurements in the 

 prismatic zone showed the presence of the 

 following forms: a (100), Z?(oio), r-''(i 1.3.0), 

 w(iio), ^(120), ^''^(370) and r(i3o). Of these, t and g are new 

 to the species. Owing to the position of the matrix surrounding 

 the crystal, only one face of each of these new forms could be ob- 

 served. The planes were narrow and t yielded a fair, and g a rather 

 poor reflection of the gonioireter signal. The forms were identified 

 from the following measurements which in every case except that of 

 m corresponded to a single observation ; m furnished two readings : 



