118 J. S. Teall — Cheviot Quarts- FeJ dies and Aiigite-Granites. 



the minerals separate out in the order of their basicity ; the more 

 basic minerals, or those which do not contain silica, being the first 

 to form. This being admitted, we see at once that the tendency of 

 the progressive crystallization must be to render the magma more 

 and more acid. 



Dr. Petersen has isolated and analyzed the glassy base of the 

 Cheviot hypersthene-andesite and the devitrified base of a porphy- 

 I'ite from Allerhope Burn. His analyses appear to me to have such 

 an important bearing on the sequence of volcanic phenomena in the 

 Cheviot district that I quote them below, together with two analyses 

 of the Cheviot lavas. 



They show that the ground-mass is richer in silica and alkalies, 

 and that the ratio of potash to soda is much greater in the ground- 

 mass than in the rock as a whole. 





I. 



II. 



III. 



IV. 



SiOa 



66-25 



65-16 



61-17 



59-05 



Al,03 



13-59 



17-49 



16-87 



15-69 



FcOs 



3-11 



3-01 



2-10 



1-80 



FeO 



— 



— 



.. ' 2-94 



4-72 



CaO 



2-75 



0-84 



4-86 



1-79 



MgO 



-28 



2-34 



3-00 



4-29 



KaO 



4-95 



5-54 



1-81 



2-88 



NasO 



2-25 



3-68 



2-67 



3-97 



H2O 



5-89 



1-76 



3-09 



3-16 





99-07 



99-82 



98-51 



97-35 



Sp.Gr. 



2-437 

 I. Glassy hase. 



2-640 

 lypersthene-ande 



2-543 

 site. 





II. Devitrified base. Porphyrite. Allerhope Burn. 



III. Hypersthene-andesite. Carhops on Coquet. 



IV. Porphyrite, ~ mile above Shillraoor Farm. 



Mr. Stock ^ has shown that the ground-mass of the Cockfield dyke 

 is also richer in silica than the normal rock. Mr. Waller, of 

 Birmingham, has recently investigated certain contemporaneous 

 veins in the bronzite-diabase of Penmaenmawr. He finds that 

 they contain 6| per cent.^ more silica than the rock in which they 

 occur, and that potash is present in excess of soda, whereas the 

 reverse relation holds in the rock itself. These veins then bear 

 precisely the same relation to the parent rock as the ground-mass 

 of the Cheviot andesite does to the mass of the andesite. 



There are veins in the Rowley Eag basalt, also described by Mr. 

 Waller, which contain 9 per cent, more silica than the rock, and 

 possess as much as 11 per cent, of alkalies. 



From these and other facts which need not be here enumerated, 

 we may safely draw the general conclusion that the effect of pro- 



1 North of England Dykes, Q.J.G.S. 1884, p. 225. 



2 Midland Naturalist, 1885, p. 6. Compare the concretionary (?) patches {e.g. 

 certain inclusions in granite described by Mr. Phillips. Q.J.G.S., vol. xxxvi. p. 1, 

 and the nodules of peridotite in basalt) with the so-called contemporaneous veins. 

 The former are as a rule more basic, the latter more acid than the normal rock. 

 The former may represent the earliest, the latter certainly represent in many cases 

 the latest products of consolidation of an originally homogeneous magma. 



