644 



MESSES. J. HOENE AND E. GEEENLY ON [Nov. 1 896,. 



Mr. Barrow ' round-grained,' whether in foliated or unfoliated rocks. 

 The micas are almost always biotite, muscovite being very rare. 



Most of the sills are full of inclusions, the smaller sills behaving 

 in every respect like the largest. There are no chilled edges 

 whatsoever. 



The junctions are of three kinds : — 



(a) ' Lit par lit ' injection of a kind more subtle still than that 



where we see parallel beds of schist and granite, for the 

 margins of a sill fade off into the gneiss through a series of 

 thinner and thinner lenticles. 



(b) The e n ds of a sill fading off into the gneiss by a dovetailing of 



biotitic folia into the granite. 



(c) True transgressive junctions, where the granite truncates th& 



folia of the biotite-gneiss. 

 All these three types can be seen in one and the same sill. 



Finally, large masses occur in which these relations are carried 

 to such a degree of intimacy as to render it very difficult to decide 

 whether to regard them as granite or as gneiss, difficult even to 

 produce a consistent map, all lines being wholly arbitrary. 



The granites are for the most part foliated, the foliation being 

 generally, as on the northern coast, parallel both to that of the 

 gneiss and to the sides of the sill. Cases, however, occur of 

 transgressive junctions where the foliation of the granite follows 

 the cheeks of the sill or dyke, and so truncates that of the gneiss 

 (fig. 3). Foliated granitoid rocks can also be seen to truncate each 

 other, as in Strathy Bay. 



Fig. 3. — Foliated granite and biotite-gneiss, southern bank of 

 Allt Tom na Bradh, Kinbrace. 



[Length of section = about 27 feet.] 



A = Biotite-gneiss. B = Granite. 



Note. — By an error of the draughtsman, the thin dyke of granite (0) at the left- 

 end of the above section has been ' shaded ' with lines parallel to the sides 

 of the dyke, instead of lines at right angles to the sides. 



The intimate relations here described do not exist in Kildonan. 

 The granites there are not only smaller and fewer in number, but 

 they have well-defined margins. As we pass north-westward they 

 increase, but even in the wavy mica-schists at Suisgill every dyke- 

 and sill is sharply separated from the country rock. The inter- 



