S. L. Penfield — Crystallization of Herderite. 333 



very well with the value given above, the second varies 0*004: 

 from /9 derived from the prism and the mean of the two deter- 

 minations 1/630 may be accepted as very nearly correct. The 

 dispersion was p>v and was decidedly inclined, the borders of 

 one hyperbola being much more brilliantly colored than the 

 othei*. 



Herderite from Hebron, Maine. 



The occurrence of this hydro-herderite has already been 

 referred to and the analysis given. The habit of the crystals 

 is represented in fig. 4. Only a very little of this material was 

 found, and the crystals were not well adapted for accurate 

 measurement. As stated in the original description, one of 

 the best measurements that was made was n^n, 331^331 = 

 102° 22' which does not agree well with the angle calculated 

 by Dana, 103° 24/, but does, however, compare very favorably 

 with the angle calculated from the hydro-herderite from Paris. 

 If the crystals are twinned about the basal plane, according to 

 a method to be described later, the calculated angle should be 

 102° 25' and if they are simple, 102° 41/. 



Herderite from Stoneham, Maine. 



On the crystals from this locality the following forms have 

 been identified : 



b, c, m, e, s, v, t, q, n and n 



corresponding to the herderite from Paris, and in addition : 



I, 120, i-2 e, 302, \-l p, 111, -1 x, 362, 3-2 



d, 101,-1-* u, Oil, 11 q, 332, f 



Of these all except d, which is mentioned by Hidden,* were 

 observed by Dana, while the prism jul, 130 and the pyramid 

 y, 3-3 observed by him have not been found by the writer, 

 and the position of y in the monoclinic system cannot there- 

 fore be determined. Figures 5 and 6, which, except for the 

 lettering, have been copied from Dana, represent the prevail- 

 ing combinations. The crystals are penetration twins and imi- 

 tate orthorhombic symmetry, similar to the crystals of stilbite. 

 The twinning plane can be either the base or the orthopina- 

 coid ; /? being very near 90°, and the character of the faces 

 such that sufficiently accurate measurements cannot be made 

 to determine this point. Considering the faces in the upper 

 front and lower rear segments as in normal and parallel posi- 

 tion, the lettering on the remaining faces has been underlined 



*This Journal, III, xxxii, p. 209, 1886. 



