ON CORDIERÏTE AS CONTACT MINERAL. 323 



crystals having a very sharp outline. They are in slender prisms 

 with pyramidal ends. They are most probably Zircon-crystals. 



The peculiar forms presented by the transverse and the longitu- 

 dinal sections tell how the crystal is built up. Figure 2 is an idealized 

 representation of the structure of the crystal as constructed from the 

 forms of these sections. The horizontally shaded portion in the figure 

 is the boundary between the inner and the outer portions. These 

 planes of division, which correspond to the Brach yd ome in each twin 

 individual, together form a double pyramid, the apices meeting at the 

 middle. They are always formed by the filling in of black grains of 

 carbonaceous matter. From this figure it is evident that the relative 

 magnitude of the inner and the outer portions must differ in the trans- 

 verse sections, according to their respective distances from the centre, 

 In microscopic sections which were made at random from the slate, 

 various forms of this gradation have been observed ; it is not un- 

 common to meet the section passing very near the centre or just at 

 the centre, having a very little or no central portion (Fig. 12.) 



This peculiarity in the structure of the crystal is probably to be 

 explained by the manner in which the crystal has grown up during 

 the process of its formation. It must be assumed that this growth 

 took place in a medium not wholly suitable to the free development 

 of the crystal. Thus we often find a number of rudimentary or 

 skeletal forms of crystal, e.g., such a one as represented in Figure 6. It 

 consists of two sets of fine bundles disposed at right angles to each 

 other. By comparing the various forms which are met with (e.g., Fig. 

 4 with Fig. 5) it is easy to see that the finer set of bundles arranged 

 longitudinally correspond to the inner portion, while the other set of the 

 wedge-shaped bundles correspond to the outer portion of the crystal 

 which we have been considering. It must be inferred from this, that 

 the growth of the crystal took place in two principal directions at 



