266 , N. L. BOW EN 



that have approximately the degree of contrast shown between 

 crystals and residual Hquid at the time the action takes place. It 

 must be understood, of course, that the crystal mesh is very weak 

 and not capable of sustaining open fissures, but that local crushing 

 of the mesh, formation of "fissures," and their filling with liquid 

 are absolutely synchronous. 



Such action, oft repeated as the warping continued, would 

 seem to be thoroughly competent to produce the banded structure; 

 nor does it fall behind in ability to produce fluxion structures. The 

 flow of liquid through the pores of the adjacent mesh would not 

 be particularly directional, but in its action of filling the "fissure" 

 it would spread laterally and crystals would be correspondingly 

 oriented. At first thought one might ask: "What crystals?" 

 but it can scarcely be doubted that the thrusting action pictured 

 would result in tearing a multitude of crystals from their relatively 

 insecure moorings along the walls of the thrust planes and in their 

 distribution by the liquid filling the "fissures." 



Both the banding and fluxion seem, then, to be readily accounted 

 for by the warping of the mesh as described, but the full conse- 

 quences of such action are not yet exhausted. The liquid filling 

 the lenticular openings developed carries, as we have seen, a cer- 

 tain amount of crystals, and the alignment of these marks the 

 fluxion structure, particularly where the lenses are thin. In 

 thicker lenses settling of these already large crystals will immedi- 

 ately take place when they are heavy, and there may develop from 

 their accumulation the most extreme of monominerahc layers. 

 Moreover, the Hquid, thus purged of its suspended crystals, begins 

 to crystallize itself under conditions that could scarcely be more 

 favorable for differentiation by crystal settling. The temperature 

 gradient is exceptionally low for a liquid mass just beginning to 

 crystallize. Slow cooling and freedom from convection, the 

 arch enemy of differentiation, are thus assured and the extent to 

 which sorting of crystals is carried as these quiet pools crystallize 

 is not likely to be matched in any ordinary type of intrusion. Not 

 only the normal bands of moderate contrast may be produced by 

 this warping action, therefore, but also bands of the most extreme 

 types as an ultimate consequence. 



