OF LITTLE KNOTT (crMBERLAND). 521 



the augite appears to assume the form of diallage * before either 

 uralitic or actinolitic change. AYhat then is the history of the 

 perfectly clear, rather silky, well-deiBned augite crystals described 

 above as included in brown and green hornblende and perhaps also in 

 enstatite ? Are they parts of the original light augite left unchanged ? 

 or are they secondary products ? or are they original enclosures ? 

 Their definite crystalline form, whether within or without the horn- 

 blende, their sharp demarcation from the strong-coloured hornblende, 

 are unfavourable to the first or second supposition and favourable to 

 the third (Plate XYI. figs. 4, 5, 6). But still their perfect preser- 

 vation in rocks where the other constituents appear more or less 

 altered is strange. Also what variety of augite are they? Their 

 general appearance suggests an augite allied to diopside, which, 

 indeed, appears to be a rather stable mineral. At present I must 

 content myself with stating the facts and the inferences to which 

 they appear to lead, and must leave it to further work or other 

 workers to resolve the difficulty. 



It results, however, that a group of rocks, characterized by the 

 predominance of magnesianbisilicatesandunisilicates, by the frequent 

 enclosure of olivine, perhaps also of a magnesian bisilicate, in the 

 larger augite or hornblende crystals, and by the infrequency of 

 felspar, of which we may take the Schriesheim, Gippsland, and 

 Inchcolm picrites as characteristic types, besides that of Hain — rocks 

 which, on the whole, exhibit a tendency to graduate into normal 

 olivine-diabase rather than into true peridotites, though occasionally 

 they come very near to the latter — occur not only in boulders, but 

 also in situ at Little Knott, and at two localities both in Anglese}^ and 

 in the Lleyn peninsula. Even if no others be found, which is very 

 likely, at any rate in South-west Wales (the more probable home of 

 the St. Davids boulder), this is a fair list of localities for a rock the 

 occurrence of which in England and Wales was only chronicled 

 in 1881. 



EXPLANATION OF PLATE XVI. 



Fig. 1. Group of crystalline grains, external form not generally perfect, of a 

 pale-coloured augite, associated with a dark staining. (S. Dyke, 

 Caemawr, page 516.) 



2. Part of a slide of the Little-Knott rock, illustrating the rapid change 



(with crystalline continuity) from a dark brown hornblende to an 

 almost colourless greenish slightly fibrous variety. (Slide No. V., 

 page 513.) 



3. Grain of a serpen tinous mineral (a) showing one well-marked cleaxage, 



and probably an altered enstatite, with brown hornblende {b), au- 

 gites &c. in a serpentinous ground-mass (c). (Boulder, Porthlisky, 

 page 519.) 



4. Grains of somewhat blackened olivine (a) associated with grains of a 



yellowish serpentinous mineral (b) and colourless augite (c) in a 

 ground-mass of similar minerals, hornblende, a little decomposed 

 felspar, &c. (Boulder, Porthlisky, page 519.) 



5. Grains of two varieties of augite («, b) enclosed in a crystal of brown 



hornblende. (Boulder, Porthlisky, page 519.) 



6. Crystals of colourless augite in a pale greenish serpentinous ground-mass. 



(Boulder, Porthlisky, page 519.) 



* I have seen this also exhibit " schillerization." 



