86 Y. KIKUCHI 



Rhombic Pyroxene (described on pag. 7<S) sometimes run out into 

 spiny processes at their terminations, and at the breaks formed by the 

 transversal clefts (fig. 3). These Augite-bancls however become very 

 often exceedingly narrow, and then the spines appear as if they pro- 

 ceeded directly from the Rhombic Pyroxene crystals. From the ap- 

 pearances presented by the spines when looked at from the two 

 pinacoidal faces, it is evident that they all run along a common plane, 

 which would be parallel to the Macropinacoid a of the Rhombic 

 Pyroxene. This fact shows that the micro-crystals of Augite are, as 

 already indicated, flattened out along the clinopinacoid face, which, 

 when they make parallel-intergrowth with the Rhombic Pyroxene, 

 becomes parallel to or coincides with the Macropinacoid of the latter. 

 Thus we see that this phenomenon represents a phase in the develop- 

 ment of the two Pyroxenes, the molecules of which seem to have a 

 marked tendency to unite themselves so as to come into similar 

 positions. From their mode of growth it will be seen that the 

 microlithic Augite has been attached to the Rhombic Pyroxene, after 

 the latter has attained a perfect idiomorphie form, and hence the 

 former must be in general younger than the latter. Various stages 

 mnv, however, be observed under the microscope. Smaller crystals of 

 the Rhombic Pyroxene have sometimes, on both ends, symmetrical 

 pairs of simple processes as shown in fig. 19. Occasionally the 

 rudimentary Augite-crystal is found having in its centre a band 

 of the Rhombic Pyroxene as in fig. 15. These forms may again be 

 traced to those in which a very narrow median band of the Rhombic 

 Pyroxene is found between the slender form of Augite (fig. 13 b) 

 which we have already considered. 



