16 



PROCEEDINGS OP THE BOSTON MEETING 



top, whic^h are often clotted together, were largely formed before the lava cuine to 

 rest. Apparently the early formed oligoclase rose to the top and asodiferous magma 

 was thus formed, which had not so corrosive action on the olivine as the calcareous 

 magma left below. 



We find the same kind of relation that exists between the top and bottom of the 

 same flow on comparison of different flows. This would suggest that a similar dif- 

 ferentiation went on before eruption ; but among the set of dikes which are by all 

 authorities supposed to be feeders to this same set of fiows we find a variation (C 1 

 and D 1) like that described by Lawson and Shutt between the center ( D 2 and D 3) 

 and margin (C2 and Co) of dikes of the same family on the north shore of lake 

 Superior. Indications of analogous variation may be found elsewhere (D4and 

 C4). In these analyses the decrease of lime is concomitant not only with the in- 

 crease of alkalies, which is characteristic in both variations, but with an increase 

 of silica. Lawson ascribes this difference to a concentration of the water glass 

 soluble in superlieated water toward the last solidified center.* 



This explanation seems reasonable, and in that case it is natural that the differ- 

 entiation in eftlisives should go another gait, for the SiOs is no longer highly solu- 

 ble, but quite the contrary. We might by analogy supj^ose that the differentiation 

 of the calcareous magma of the ophite which (as the ophitic texture shows 

 must be formed in a state of rest) is almost surely the last to solidify, is due to a 

 concentration of the more fusible (soluble in caloric) part. Augite is much more 

 fusible than olivine, and labradorite is probably the most fusible of the plagioclase 

 series. It is worth noting, too, that this effusive differentiation tends toward Bun- 

 sen's normal basaltic magma (A 3). 



Group A — Effusives {loiver Part of Flow). 



Components. 



1 



2 



3 



Si02 



46.13 



46.32 

 2.78 



15.95 

 0.86 

 8.92 

 4.08 



10.23 



3.56 



1.23 



3 25 

 0..0 



} 



I 

 J 





Ti02 



48.47 



AI2O3 



19.79 

 7.24 

 3.79 

 7.27 



11.43 



2.55 



0.52 



1.S3 



COo 0.29 





Fe203 



30.16 



FeO 



MgO 



6.89 



CaO 



11 87 



NaoO 



1 96 



K2O 



65 



HjO 







MnO 89 













Sum 



100.84 

 2.877 







Sp. Wt 













*See Becker on Fractional Crystallization, in Am. Jour. Sci., fourth series, 1S97, vol. iii, p. 21. I 

 cannot understand why Becker should consider fractional crystallization antipodal to magmatic 

 differentiation, for it seems to me a most effective way of promoting it. However, it is only a 

 question of names. 



