FORMATION OF LEUCITE IN IGNEOUS ROCKS 387 
common than cases where the norm and mode correspond in many 
particulars. 
The readjustments shown to be possible in the case of the leucitic 
rocks indicate that the mineral molecules probably do not exist as 
such in the molten magma, but that they are dissociated into simpler 
molecular groupings, or their constituent oxides or ions, capable of 
reacting differently among themselves according to the conditions of 
solidification. 
The facts brought out in the preceding discussion do not seem to 
favor the theory of eutectic mixtures in explaining the crystallization 
of rocks, at least so far as the leucitic rocks are concerned. 
There is indicated the possibility that pseudo-leucite is not a 
pseudomorph after original leucite, but that it replaces a distinct, 
soda-potash mineral, as yet unknown, which was not stable under the 
conditions following its period of crystallization. The evidence in 
regard to this will be discussed subsequently, in connection with 
some Sardinian leucitic lavas. 
