164 TEE AO OLINALEROFIGTS OTRO EN 
Colombia and Ecuador in their chemical composition, their 
structure, their mineral composition and in their segregations, 
even down to small details. It will hence be instructive to com- 
pare the diagram of the molecular variation of the present rocks 
with those of Colombia and the Andes volcanoes (especially 
the former), which Iddings has already published on pages 173 
and 174 of Vol. I. of this journal. 
On comparing the diagrams great differences are at once 
apparent. In the first place in our region the AI,O, line, while 
on the whole the highest, is crossed at the basic end by both 
CaO and FeO, whereas in Iddings’ diagrams it is far above all 
the others, and moreover drops notably more toward the acidic 
end than in the present diagram. In the South American rocks 
it runs much more parallel on the whole to CaO than in ours 
and is also markedly parallel to FeO, while in the Greek rocks 
the two are inverse throughout almost their entire course. 
Al,O, and Na,O are decidedly inverse in both regions, but in 
the South American rocks the two alkalies are inverse (espec- 
ially im: the’ Colombian rocks), while im) (Greece they );ane 
parallel. Another striking point of difference is the behavior of 
the CaO, MgO and FeO molecules which in both of Iddings’ 
diagrams run close together and in the Andes rocks especially 
are much intertwined; in the Greek rocks on the other hand 
they run quite far apart from each other. 
Further study of the diagrams will reveal many more points 
of dissimilarity, but those mentioned are sufficient to show that 
the rocks of the two regions are differentiation products of widely 
different magmas; in other words that the two petrographical 
provinces are fundamentally distinct. The same differences are 
also seen on comparing our diagram with that of the rocks of the 
Yellowstone National Park given by Iddings,’ and which he 
remarks show a great similarity to those of the Andes rocks. 
The diagram here given may also be instructively compared with 
those rocks from other localities given in the same paper. 
‘Orig. Ign. Rocks, Pl. Il. I was unfortunately unable to consult the paper by 
Dakyns and TEALL (Q. J. G. S. 48, 1892, p. 104) quoted by him in Jour. OF GEOL. 
Te L7O. 
