314" J. V. Elsden — Origin of Pegmatite Veins. 



easily form such streaks as would, on crystallisation, form pegmatite 

 veins. Slowness of diffusion in a viscous mass would prevent any 

 appreciable mingling of the solutions, except perhaps at their 

 margins. They would form thin bands, like a mixture of syrup 

 and water, and they would suffer deformation and contortion in 

 consequence of any magmatic movements to which they might be 

 exposed. This would explain the formation of contorted bands in 

 the midst of an uncontorted rock. 



We have next to account for their coarsely crystalline texture, 

 a feature usually taken to indicate slow cooling. Slowness of growth 

 undoubtedly tends to the development of large crystals ; for owing 

 to the influence of surface tension it can be shown that in any 

 solution in which crystals are forming, while the solution is super- 

 saturated with regard to the larger crystals it may be unsaturated 

 with regard to the smaller ones. The latter, therefore, may be 

 resorbed and deposited upon the former, which will then continue 

 to grow at the expense of their smaller brethren. But it may be 

 taken as an established fact that the formation of large crystals 

 depends upon other factors than the rate at which cooling takes 

 place. Thus, at Ofvra Trensum, in the neighbourhood of Carlshamn, 

 may be seen a well-marked irregular junction between a very fine- 

 grained blue granite and a coarse porphyritic pink rock. These 

 were evidently part of one and the same cooling mass, but the 

 junction shows no relation to any conceivable isothermal planes. 

 Variations in texture are, indeed, a common phenomenon in plutonic 

 masses. It may be safely assumed that the pegmatites crystallised 

 at a later stage than the parent rock. Whether we look upon them 

 as true eutectic mixtures or not, the circumstances of their origin, 

 as assumed above, would indicate a freezing-point lower than that 

 of the main mass. Pressure probably plays an important part in 

 crystallisation, but we are only beginning gradually to realise the 

 direction of its influence. The experiments of Oetling ^ seem to 

 show that the chief role of pressure is to promote superfusion. 

 Now Taramann - has shown that the velocity of crystallisation 

 increases with degree of superfusion up to a certain maximum, and 

 then diminishes. The larger and more perfect crystals are formed 

 near the minimum of superfusion. Another factor which comes into 

 play is the number of crystalline nuclei present, upon which the 

 influence of certain catalysers or ' agents mineralisateurs ' must 

 be considered. I am aware that the action often ascribed to these 

 mysterious agencies has been called in question by some recent 

 observers, notably by Lagorio and Morozevvicz ; but our knowledge 

 of the ionic dissociation of fused salts is still so incomplete that for 

 the present it would seem advisable to retain as a useful hypothesis 

 the conception of catalytic action by such agents under certain 

 conditions. 



1 " Vergleichende Experimente iiber Verfestigun;? geschmolzener Gesteinsmassen 

 unter Erhohtera und normalem Druck ": Tschermak's Min. und Pet. Mittheilungen, 

 xvii, p. 331. 



* Zeit. fur physikalische Chemie, xxvi (1898), p. 307. 



