190 iV. L. Bowen — Crystallization — 



arise an arrangement in which the upper half showed little 

 differentiation and the lower half marked differentiation. The 

 sill, however, has no such dnal character. It is a unit, the 

 bottom portion representing, in a general way, an accumula- 

 tion of the heavy crystals from the whole body and the top 

 portion the lighter residuum.* 



The Sinking of Crystals as a General Factor. 



The Palisade sill clearly illustrates, as Lewis concludes, the 

 importance that the sinking of crystals may have in differentia- 

 tion. There are many other more or less similar occurrences, 

 among them the Duluth gabbro laccolith,f the Preston gabbro 

 laccolith^ and the Sudbury sheet, which illustrate its general 

 importance.! Harker has recently expressed his belief that 

 the sinking of crystals may be of importance in deep-seated 

 magma basins, but for differentiation in place he appeals to 

 diffusion towards a cooling boundary. T In the face of so 

 many examples of differentiation in place through sinking of 

 crystals it seems wholly inadvisable to follow Harker in making 

 this distinction. 



The sinking of crystals may possibly be considered of greater 

 importance in basic than in intermediate and acid rocks on 

 account of greater viscosity of the latter. There is, however, 

 little very convincing evidence that acid rocks are really much 

 more viscous than basic rocks so long as they are kept confined 

 and retain their volatile constituents. Where they occur in 

 very large, deep-seated bodies whose cooling must be a very 

 slow process, a viscosity even several hundred times as great as 

 that of diabase magma could not prevent some differentiation 

 through the sinking of crystals. There is little reason to believe 

 that these bodies of magma usually have such great viscosity. 

 It is certain, therefore, that we cannot avoid assigning a general 

 importance to the sinking of crystals in the differentiation of 

 igneous rocks. 



/Summary. 



Experiments with artificial melts were undertaken in order 

 to observe whether sinking or floating of crystals could be 

 obtained. The sinking of olivine and pyroxene and the floating 



* Lewis, loc. cit. p. 132. 



f W, S. Bayley, Jour. Geology, vol. ii, p. 814, 1894. 



X G. F. Loughlin, Bull. 492, U. S. Geol. Survey, 1912, p. 78. 



§ A. P. Coleman, Rep't Bureau of Mines, Ontario, vol. xiv, pt. 3, 1905. 



| Daly has listed a large number of examples of gravitative differentiation. 

 Igneous Rocks and Their Origin, pp. 220-236. 



TTCongres Geologique International, Compte Rendu, Xlle Session, 1913, p. 

 208. 



