76 



differences between the olivine-basalts of Hawaii and many other 

 localities, and the non-olivine bearing- basalts of Carnmoney Hill 

 and Franz Joseph Land are due to differences in the rate of 

 cooling. 



" We have now considered to some extent two groups of 

 volcanic rocks — the andesites and the basalts — and we have seen 

 that the principle of progressive or fractional crystallisation, 

 coupled with the separation of the earlier-formed crystals from 

 the mother-liquor, will account for many of the variations seen 

 in each of the two great groups, and also for some rocks, such 

 as quartz-felsite and olivine-rocks, which do not belong to either. 

 Penological work for some time past has had in view the con- 

 struction of a genealogical tree in which the leaves shall corres- 

 pond to the rocks as we see them, the twigs and branches to 

 the various magmas that have been formed during the evolu- 

 tionarv process, and the trunk to some primordial magma. By 

 starting with andesitic and basaltic magmas we have endeav- 

 oured to construct some of the leaves of this genealogical tree, 

 and, if it is ever to be completed, I think it will have to grow 

 mainly in this way. Dr. Bowen, of the Geophysical Laboratory 

 ; n Washington, in a most valuable paper, has suggested that 

 basalt may be the primary magma ; but into that question I do 

 not propose to enter. Even if we accepted his view the inevi- 

 table question would arise, ' what is the origin of this magma?' 

 and we should thus reach the border-land of geology and cos- 

 mical physics — a somewhat perilous country for the geologist to 

 travel in, and one in which he may easily be led astray, as, in- 

 deed, has happened more than once in the past. Still, I will 

 trespass in it for a moment. Dr. Emmerson Reynolds, five years 

 ago, placed a stepping stone on which it is safe to stand for a 

 moment and look ahead. 



" It is perfectly legitimate to regard igneous rocks as being 

 mainly composed of oxides ; but there are many reasons for 

 believing that oxygen is absent from the interior of the earth, 

 which may consist of concentric shells of varying composition, 

 with heavy metals, such as iron and nickel, forming a large por- 

 tion of the interior, and elements of lower atomic weight, such 

 as silicon, aluminium and calcium, forming outer shells. From 

 certain theoretical considerations Dr. Reynolds was led to ask 

 himself the question whether some at least of the alumino-sih- 

 cates which enter so largelv into the composition of igneous 

 rocks may not be fully oxidised products of silicides, formed at 

 high temperatures below the oxidised crust of the earth. In 

 the attempt to throw light on this question he effected the syn- 

 thesis of the lime-felspar, anorthite, by an entirelv new and most 

 suggestive method. By first fusing together silicon and 

 aluminium and then adding calcium, he obtained a compound 

 or alloy in which these elements were associated in the same 

 proportions as in anorthite, and then by passing moist oxygen 



