BY H. I. JENSEN. *695 



From the island of Aunuu he describes trachydolerites and 

 palagonite-tuff. This is a very welcome addition to our know- 

 ledge of alkaline rocks, and fully confirms my surmise that there 

 must have been an alkaline base in the basalt which I analysed. 



The second point which I wish to touch on is, that the sub- 

 alkaline composition now established by Weber for the Samoan 

 lavas, casts some doubt on my hypothesis that the eruptions 

 ■along the Samoa-Tonga-Taupo-line depend upon an earthfolding 

 movement. In my paper on "The Distribution, Origin, and 

 Relationships of Alkaline Rocks " "^ I emphasised the point 

 that alkaline and subalkaline (mixed) magmas are erupted mainly 

 in regions where great movements along fault-planes of the 

 normal type are in progress, and not in regions of compression 

 (an adaptation of Prior's view). Whether, therefore, the sub- 

 alkaline magmas of Savaii are the result of further block-faulting 

 in the Pacific, or of a minor fold-movement subsequent upon the 

 break-up of the Fijian continent, or to the magmatic differenti- 

 ation of normal calcic magmas connected with a more extensive 

 fold-movement, is a problem that must be left in abeyance. 



Dr. Weber also draws attention to the fact that mixed magmas 

 are becoming recognised in far more localities than formerly 

 supposed. In the Pacific area they have been reported from the 

 Sandwich Islands, the Caroline Islands, Dunedin district in 

 New Zealand, the Island of New Pomerania of the Solomons, and 

 other islands, as well as from Samoa. Dr. Weber remarks that 

 it becomes increasingly difficult to draw a sharp line of demarc- 

 ation between alkaline and alkali-calcic rocks, and that our 

 interpretation of a petrological province will vary according to 

 geological time. Those are matters to which I, too, have drawn 

 attention in my thesis on the subject. I offered an explanation 

 of the origin of mixed magmas which, as yet, I see no reason to 

 alter. With regard to geological time, it appears that those 

 petrographical provinces which are almost wholly pure alkaline 



These Proceedings, 1908, p. 491, 



