392 



University of California. 



[Vol. 3. 



olivine. The orientation of the feldspar crystals is entirely 

 irregular, and although between the olivine rods, as seen in thin 

 section, the areas occupied by the feldspar may appear also radial, 

 this is due to the fact that the olivine controls the structure. It is 

 not radially disposed, in the sense that the olivine is, with a 

 definite orientation for each crystal. 



In the crystallization of the orbules, then, it is the disposition, 

 regular orientation, and intermittent growth of the olivine to 

 which they owe their peculiar structure. 



There are some minor mineralogical variations which may be 

 briefly mentioned. Hypersthene has been mentioned as a sub- 

 ordinate constituent. This mineral is, however, chiefly found 

 toward the central parts of the orbules, becoming scarcer toward 

 the periphery, and is of rare occurrence in the interorbicular 

 matrix. In the orbules magnetite is very sparingly represented 

 by occasional small grains, while in the interorbicular matrix of 

 granular structure it is quite abundant. It is, however, not 

 abundant in that phase of the interorbicular structure in which 

 the olivine is disposed in diversely oriented rods. In such por- 

 tions of the matrix the crystallization of the olivine has been the 

 same as in the orbules, but has not been controlled by the same 

 orienting force. 



The Olivine. — The olivine which constitutes the rods of the 

 orbules required some considerable study to place its determina- 

 tion beyond question. It is in nearly all the slides quite fresh 

 and nearly colorless. The rods are traversed by strongly pro- 

 nounced cracks, which fall into three categories. The first of 

 these comprises those cracks which are perpendicular to the rods 

 or nearly so, the second those which are approximately parallel 

 to the direction of elongation of the rods, and the third those 

 which are strongly oblique to the direction of elongation but at 

 variable angles. Most of the rods have only the transverse 

 and oblique cracks. The longitudinal cracks are usually found 

 only in those cases where the rods broaden out locally and have 

 the character, in section, of elongated plates. They are usually 

 crossed by transverse cracks. These occurrences are rather 

 exceptional, and the fact that the rods are not sections of plates 

 in general was proved by cutting sections parallel to a radius of 



