452 ME. A. HARKER ON THE EETJPTIVE EOCKS IN THE 



majority of cases it occurs as oiDliitic plates, commonly including 

 grains, cores, or nuclei of augite, as well as felspars and magnetite. 

 The mineral is almost always deep brown in colour, with the usual 

 pleochroism, the absorption being indicated by y > /3 :>;> a. The 

 maximum extinction-angle in vertical sections is 18° or 20°. A 

 greenish-brown tint is occasionally seen with the brown. The horn- 

 blende is normally " compact " and well-cleaved, the green portions 

 sometimes fibrous. 



The relations of the hornblende and augite are worthy of notice, 

 and prove that the same brown, compact hornblende may be either 

 original or a product of amphibolization. When augite is included 

 in the hornblende, there are three cases to be distinguished. In 

 the first place, original hornblende may include augite-grains in the 

 same way as it includes felspar- crystals or magnetite. When this 

 is so, the enclosed grains are generally of rounded shape with a very 

 definite and comparatively smooth boundary ; there is no crystallo- 

 graphic relation between the two minerals, and if two or more 

 grains occur in the same plate of hornblende, they are differently 

 orientated. Rarely, as in the quarry on Mynydd Penarfynydd,'good 

 idiomorphic crystals of augite are included in the hornblende. 



Secondly, a crystal or plate of augite becomes partially converted 

 into hornblende, chiefly round its margin, the two minerals then 

 having the vertical axis and plane of symmetry common. Since 

 this amphibolization probably involves a change of chemical com- 

 position, the term paramorphism is not strictly applicable, and we 

 must call the process pseudomorphism, or where only the border is 

 affected, perimorphism. If the original augite had crystal-contours 

 the secondary character of the hornblende is obvious ; in other cases 

 it can generally be inferred from the extremely intricate and ragged 

 appearance of the boundary between the augite-core and the 

 investing hornblende. If the augite is twinned, the resulting 

 hornblende is twinned about the same plane. Where several cores 

 of augite are included in one hornblende-plate, they all have, of 

 course, one orientation. The pseudomorphic or perimorphic horn- 

 blende has precisely the same characters as the original hornblende. 

 Sometimes we may see a crystal of augite both moulded by horn- 

 blende and partially pseudomorphosed, the two kinds of hornblende 

 being undistinguishable from one another. 



There is a third way in which hornblende may include augite, 

 and this has rarely, I believe, been specifically described in 

 British rocks*. It appears that while a nucleus of augite was 

 growing the magma became so altered that after a certain point of 

 time hornblende-substance was deposited instead of augite, this 

 hornblende growing upon the augite kernel with the usual crystallo- 

 graphic relation between the two minerals. This, if I understand 

 aright, is what Eohrbachf has described in the teschenites of 

 Silesia under the name of ^^ ergdnzende" or complementary hom- 



* The picrite of Inchcolm exhibits good examples ; see, e. g., Teall, ' British 

 Petrography,' plate vii. (1888). 



t Tschermak's ' Mittheilungen,' 1886, p. 84. 



