ORIGIN OF AUGITE ANDES ITE 411 



19 gives the result of averaging 14 analyses of augite phenocrysts 

 from basalt, dolerite, limburgite, and labradorite porphyrite, as stated 

 in the appendix to Osann's compilation. Columns 20 and 21 represent 

 typical analyses of labradorite and anorthite respectively. 



By the settling -out of olivine, augite, and magnetite the molten 

 lava or mother-liquor, must, when compared with the original basalt, 

 be poorer in iron oxides, magnesia and lime, and richer in silica, 

 alumina and the alkalies. The change in alumina might be slight, 

 provided that the anorthite phenocrysts also settled out. Crystals of 

 labradorite and andesine would slowly rise and enrich the upper part 

 of the lava-column with silica, alumina, and soda. Chemically the 

 average augite andesite appears to correspond to the mother-liquor, 

 possibly bearing up-floated plagioclase crystals, while many of the 

 ultra-basic rocks, picrite, limburgite, dunite, harzburgite, and other 

 peridotites, correspond to those magmatic types developed deep 

 within the lava-column by the settling of the phenocrysts. 



2. Observed cases of sunken and risen phenocrysts. — A few instances 

 of gravity differentiation accompanying the growth of phenocrysts in 

 magma have been observed in rocks and also in artihcial melts. Both 

 Scrope, in 1825, and Charles Darwin, in 1844, published hypotheses, 

 now classic, of such fractional crystallization in nature. Clarence 

 King, in 1878, adopted their conclusion and added new examples 

 studied by him at Kilauea. He broke asunder some of the thin, 

 tongue-like flows of once very fluid basalt and found that in every 

 case the bottom of the flow was thickly crowded with triclinic feldspars 

 and augites, while the whole upper part of the stream was of nearly 

 pure isotropic and acid glass.' He further remarked on the general 

 absence of phenocrysts in some of the great Hawaiian flows; the 

 crystals sank away into the conduit before eruption. Certain other 

 Hawaiian flows described by E. S. Dana show the complementary 

 feature of being ultra-basic and crowded with olivine crystals, which 

 make up as much as 50 per cent, of the rock.^ Neither at Kilauea 

 nor at Mauna Loa, however, would one expect to find a notable 

 differentiation of augite andesite. The extreme fluidity of lava- 

 columns in both of the great pits seems to indicate general tempera- 



Jf U. S. Geol. Explor. 40th Parallel, Sys. Geol., 1878, p. 716. 



* J. D. Dana, Characteristics of Volcanoes (New York, 1891), p. 324. 



