ASSOCIATED IGNEOUS ROCKS OP THE AYRSHIRE COAST. 779 



is a rough ground-plan of what was found on the north side of the 

 gabbro dyke. Here the diallage rock contains fragments of serpen- 

 tine from 2 or 3 inches to 12 or 15 inches in diameter, and is cut 

 by a vein of gabbro corresponding with that in the main dyke. 

 This vein, in greatest width, is about 4 inches ; it runs roughly 

 parallel with the main dyke, splits into two strings, and so dies 

 away. Hence the gabbro is the latest of the three rocks. A little 

 of the diallage rock is also exposed near the northern gabbro dyke, 

 but its relations to the latter are less distinct. Here the gabbro 

 cuts serpentine, which is intrusive into an altered " greywacke." 



The following is Mr. Houghton's analysis of a specimen of the 

 diallage in the above rock. It varies rather in amount of decompo- 

 sition. An analysis of the least altered gave : — 



Dried at 100° 0. 



H 2 9-10 



Si0 2 47-22 



A1 2 3 2-76 



FeQ 10-02 



CaO 6-51 



MgO 25-59 



101-20 



I have examined microscopically each of the above rocks. In the 

 former the saussuritic mineral is partly translucent, but for the 

 most part is rendered opaque by a fine earthy dust with a some- 

 what fibrous arrangement. "With crossed Mcols the field is gene- 

 rally dark, but exhibits occasional specks and streaks of brighter 

 colours and indistinct traces of aggregate polarization. In one or 

 two parts, however, some very characteristic plagioclase felspar still 

 remains (very transparent and free from all enclosures, except some 

 minute dark acicular microliths), thus indicating the origin of the 

 saussuritic constituent*. 



The other constituent of the rock exists in three forms — (a) 

 crystalline grains of diallage (and, in one or two cases, of normal 

 augite) comparatively little altered ; (b) uralite wholly or partly 

 replacing diallage, green- coloured, strongly dichroic, exhibiting 

 occasionally very characteristic hornblende cleavage; (c) a trans- 

 parent nearly colourless and non-dichroic, rather fibrous mineral, 

 which sometimes is associated with either of the above, sometimes 

 occurs alone. Part of this exhibits brilliant colours with crossing 

 Nicols ; part changes from a light bluish white to a blue-black, like 

 some fibrous serpentines : the former mineral is probably actinolite ; 

 the latter may be only a variety of serpentine. This rock then 

 appears to have been originally a coarsely crystalline gabbro, which, 

 perhaps, contained a little olivine. 



The second rock is practically a mass of diallage crystals. Some 

 of these are very fairly preserved, others are traversed by cracks 

 with dusty borders showing incipient decomposition ; while others 



* Cf. Zirkel, Geol. Explor. of Fortieth Parallel, vol. vi. p. 107. 



