250 «/. F. Kemp — Extension of the Cortlandt Series. 



blende. which may go so far that (as in 102 from A, 7) no horn- 

 blende remains and the rock is a pnre mica-diorite. 



In the less simple mineralogical aggregates augite frequently 

 appears and at times in considerable amount, forming as much 

 of the rock as any other component. It is light, transparent 

 green in color, not at all pleochroic and exhibits well marked 

 cleavage and characteristic optical properties. Some large de- 

 tached masses (in B, 8) evidently not far from place, were found 

 to consist almost wholly of brown hornblende and green augite, 

 a very curious and interesting mixture. (Spec. 207.) The 

 hornblende is in large irregular masses perhaps 10 mm . X 15 mm . 

 of rude, corroded prismatic outline. These are surrounded by 

 smaller, often idiomorphic crystals of augite and frequently 

 contain abundant masses of the same scattered throughout. 

 This association suggested the interesting question of the para- 

 morphosis of these two minerals.* As the most reliable basis 

 from which to draw any inference, enough of each mineral was 

 isolated by careful sorting for analysis. 



III. IV. V. 



SiO„ 43-60 46-00 43'35 



Fe 3 ]4-08 11-20 5'61 



Al 2 (> 3 _• 20-02 14-80 29-75 



CaO 12-19 ]5'52 12-46 



M^O 4-03 4-75 2-03 



K o 2 06 4-70 5-93 



Xa 2 3-50 3-20 tr. 



Loss on ignition. 0-73 



99-48 100-17 99-86 



(Xumbers III and IV are the hornblende and augite, re- 

 spectively, analyzed by the writer. No. Y is an analysis of a 

 somewhat altered piece of the same rock kindly made by Mr. 

 L. M. Dennis, instructor in chemistry in this university.) 



These analyses show that the hornblende is more basic than 

 the augite and corroborate the view first taken from the slide, 

 that the augite has been formed subsequent to the hornblende 

 by the corroding action of the more acid magma which re- 

 mained after its differentiation. A similar process with the 

 formation of magnetite has been described by Oebbeke in the 

 andesites of the Phillipines ;f but elsewhere in the older 

 rocks;}; and especially in those of the main Cortlandt,§ the pro- 

 cess of change in these minerals seems to have occurred in the 



* Gr. H. Williams, this Journal, III. vol. xxviii. p. 259. Full references and a 

 review of the literature are here given. 



f Xeues Jahrbuch, Beil. Band i. 1881, p. 474. 

 X Rosenbusch, Mikros. Phys.. Band i. p. 468. 

 § G. H. Williams, this Journal, vol. xxxi, p. 33. Jan.. 1886. 



