of the Clyde Area. 313 



of the crystalline residuum as compared with the possible holo- 

 crystalline development. Glassy rocks are probably best named in 

 such a way as to connect them with their nearest holocrystalline 

 equivalents, and consequently, for accurate naming, the chemical 

 composition will have to be determined in many cases. In general, 

 however, petrographers will only find this necessary when the rock 

 is of an abundant and widespread type not directly associated with 

 its holocrystalline equivalent, as is the case, for example, with the 

 rocks under discussion and the pitchstones of Arran. 



The use of the dominant mineral constituents and texture as 

 qualifying terms, and of the prefix hijalo-, in naming glassy rocks, 

 results in some unwieldy designations, which may be illustrated as 

 follows : — 



Cumbraite = Anorthite-enstatite-augite-hyaloandesite. 



Inninmorite = Anorthite-(uniaxial) augite-hyaloandesite. 



Leidleite = Subvariolitic-augite-hyaloandesite. 

 In such cases the short locality-names, such as those given above, 

 are much the more convenient for use when the rocks of a single 

 petrographic province are being described, and for the Cumbrae rock 

 I therefore propose the term cumbraite, which is defined according to 

 the mineralogical and chemical composition given above. A new 

 name should only be given to a glass v rock when it is an abundant, 

 widespread, and homogeneous type, different either mineralogically 

 or chemically from previously described glassy rocks, not when it is 

 a mere hyaline facies of an associated holocrystalline rock. Whether 

 these terras should obtain a circulation outside the discussion of the 

 British Tertiary petrographic province is a question beyond the scope 

 of this paper. My own opinion is that they should not, unless 

 future work should show that they are widespread types in other 

 petrographical periods and provinces. 



Other Occurrences of Cumbraite. 



Two dykes closely resembling the typical cumbraite occur on 

 the Ayrshire mainland. One forms a N.W. dyke, 5 miles long, 

 cutting across the River Ayr and its tributary the Coyle Water, near 

 Sundrum House, about 6 miles east of Ayr. This rock is somewhat 

 decomposed, but the anorthite phenocrysts remain fresh, and the 

 enstatite is clearly recognizable by its pseudomorphs. The other 

 occurrence is in the Burnock Water near Ochiltree, and appears to be 

 quite a short dyke. In thin section this rock is seen to contain 

 rather less glass than the typical cumbraite, and is richer in felspar 

 and pyroxenes, especially the latter, thus providing a transition to 

 the true tholeiites. The enstatite is pseudomorphed by a green 

 fibrous mineral. 



In the Cowal peninsula of Argyllshire a perfectly typical cumbraite 

 occurs at Toward (Clough, 1897, pp. 131-2, 136). 1 This dyke is 

 said to have a glossy band in the interior, running parallel with 

 the length of the dyke, a feature connecting it with the Eerie Port 

 dyke of the Cumbrae. A slightly different type which may, however, 



1 I am obliged to the Geological Survey for the loan of the Cowal slides on 

 which these remarks are based, and also for the Eskdalemuir slides. 



