Origin of Pecpnafite. — Crosby and Fuller. 177 
ly, solidification being due only in part to cooling and in part 
to spontaneous dehydration, — a general principle, the appli- 
cation of which to the pegmatite magma is indicated by the 
absence of hydrous species among the component minerals. 
The experiments of Daubree and and others favor the view 
that the proportion of water essential to the constitution of 
the pegmatite magma is small; and this is indicated, also, by 
tlie infrequent and relatively small pockets; although the 
enormous pressure prevailing at these great depths may, per- 
haps, tend to prevent the formation of pockets. 
The origin of pegmatites at great depths is suggested by 
the regional metamorphism of the schists and the consequent 
general absence of characteristic or significant igneous con- 
tact phenomena ; and a high temperature may be inferred from 
the usually jagged outlines of the veins, indicating a tear- 
ing or rending, rather than a -breaking, of the rocks, as if 
they had been softened and toughened by heat. The minute, 
thread-like veins and apophyses, both of pegmatite and com- 
mon granite, and the very common lack of gradation along 
the walls for both rocks, and other facts point to the same 
general conclusions. The pegmatite magma is probably very 
liquid only in pockets; while for the development of the nor- 
mal pegmatite a much thicker jelly than the normal granite 
magma, combined, as in the latter, with extremely slow cool- 
ing and dehydration, seems to be required. The order of 
crystallization shows that pegmatite magma is only a more 
extreme form of the granite magma. In a true igneous or 
granite magma, however, the size of the crystals is definitely 
limited by the viscosity; and beyond a certain point it is 
practically independent of slowness or rate of cooling. It 
may be noted, also, that while in pegmatite the tabular and 
prismatic crystals tend to be perpendicular to the walls, in 
true magmatic consolidation the orientation of the crystals is 
essentially indifferent to the walls. 
The essential continuity of fusion and aqueous solution is' 
not readily demonstrated by experiment. Under the relative- 
ly moderate pressures which have been employed, the addition 
of heat to a solution, beyond a certain low maximum, vapor- 
izes and expels the solvent; and. in like manner, although the 
addition of water to a magma promotes liquidity, the amount 
