THE PROBLEM OF THE ANORTHOSITES 91 



and their ancient age. The Canadian, North American, South 

 Russian(?), and African(?) anorthosites are of pre-Cambrian 

 age, and the Norwegian are Silurian. Of course the question of 

 age is not to be valued too highly. New data must be expected, 

 for we must not forget that granites were formerly also considered 

 as exclusively pre-Cambrian, while now we know not only Paleozoic 

 but also Mesozoic, and even Cenozoic representatives of this group. 



During recent years especial attention has been paid to gravi- 

 tational differentiation, a phenomenon produced by the difference 

 between the densities of the crystallizing minerals and the still 

 liquid portion of the magma. This conception has been especially 

 favored by Bowen and Daly.^ 



The idea of gravitational differentiation is not new, having been 

 introduced many years ago by Darwin. I have myself made use 

 of this idea and have applied my own considerations and qualita- 

 tive experiments to it. As before, I continue to consider gravita- 

 tional distribution of phenocrysts as a factor in the mechanism of 

 crystallization and differentiation of igneous rocks and in the 

 formation of magmatic ore bodies; but I must confess that some- 

 times the significance of this factor is valued much too highly, and 

 the limits of its application are made too broad. In order to prove 

 this assertion and to show to what degree this factor is considered 

 applicable by the writer, let us see briefly how different authors 

 have applied the factor of gravitative adjustment of crystals to 

 the theory of differentiation. 



According to Sartorius von Waltershausen, even during the 

 period of the formation of the earth's crust the liquid mass was 

 spHt into layers of different densities, so that the deeper layers con- 

 sisted of denser liquid masses and the upper of the lighter ones. 

 The different volcanic centers were fed by different layers. This 

 general speculative conception is rather far from the phenomenon 

 of differentiation and cannot explain in a satisfactory way the 

 diversity of volcanoes and lavas. It is useless to enter here into 

 further details of this hypothesis. 



Gouy and Chaperon expressed the idea that in a deep layer of 

 a solution there must be a gravitational concentration of the dis- 



^ R. Daly, Igneous Rocks and Their Origin, 1914. 



