PROCEEDINGS 
OF THE 
NATIONAL ACADEMY OF SCIENCES 
Volume 6 APRIL 15. 1920 Number 4 
DIFFERENTIATION BY DEFORMATION 
By N. L. Bowkn 
Department oe Mineralogy, Queen's University, Kingston 
Communicated by J. M. Clarke, February 24, 1920 
Introduction. — ^The deformation of an igneous mass during crystalliza- 
tion, with consequent separation of liquid from crystals, has frequently 
been suggested as a cause of variation of igneous rocks but the suggestion 
has apparently never received as detailed consideration as it warrants. 
That variation would result is obvious, but the full extent of its importance 
will not be realized unless one has an adequate picture of the far-reaching 
consequences that may follow from the separation of crystals and mother 
liquor. In another paper the writer has demonstrated the consequences 
of such separation in artificial melts and applied the results to natural 
magmas.^ Emphasis was there placed upon gravity as a means of separa- 
tion of crystals from liquid but the chemical consequences are identical 
whatever the means of separation. In certain respects, however, the conse- 
quences of separation by deformation may be distinctive and to these 
attention is directed in the present paper. ^ 
Discontinuous differentiation. — Discontinuous variations, in particular, 
are not normally to be expecte^d as a result of gravitative accumulation 
of crystals but discontinuity would appear to be a very likely consequence 
of deformation. This is especially true at a late stage of crystallization 
when deformation may cause a crushing of the crystal mesh with conse- 
quent closer packing of the crystals and onward movement of the inter- 
stitial liquid. At first thought it might seem that the liquid could not 
move without carrying the crystals with it but it may be pointed out that 
the break down of a crystal mesh is likely to be progressive and any area 
breaking down will in such circumstances be bordered by an area not yet 
affected. The residual liquid is, therefore, free to move into the interstices 
of the unaffected portion of the mesh and to drive ahead of it the liquid 
already there which may become a separate intrusive body or a distinc- 
tive portion of the same intrusive and of late consolidation. Elsewhere 
the writer has offered the suggestion that the common associaticn of 
gabbro and granophyre is frequently capable of interpretation in this 
159 
