220 Proceedings of Royal Society of Edinburgh. [sess. 
a quarter of a mile, in some parts, it is difficult to say whether 
the rocks should be described as schists traversed by veins of 
granite, or granite enveloping blocks of schist. I do not, however, 
mean to convey the idea by this that there is any lithological 
passage of the one type into the other ; for that there certainly is 
not. On the contrary, the line between the granite and the schist 
is clearly seen in hand specimens to be quite sharp and well-defined, 
and, under the microscope, the presence of crystalline felspar on 
one side of the boundary line and its absence on the other can 
also readily be made out. The field relations of these rocks, as 
seen at Torr na Sealga, is shown in fig. 15 already referred to.* 
It may be remarked, in passing, that having regard to the 
fact that a zone consisting of closely interwoven, or spliced, granite 
and schist extends for a considerable distance around the granite 
proper, one is led to speculate what the result would be were the 
whole area subjected to extensive dynamic metamorphism. The 
granite would deform into muscovite-biotite gneiss, the plexus of 
granite veins and fragments of hornfelsed greywacke, quartzite, 
and mica schist, would form a gneissoid complex of a second kind, 
while the schists themselves would form a third group, the only 
feature common to the whole being a general parallelism of the 
planes of schistosity. There cannot be much doubt that many 
older complex areas of this kind occurring in the Highlands and 
elsewhere have been affected in this manner, and it may well be 
the case that some of the anomalous groups of gneisses and 
gneissoid rocks of the Central Highlands of Scotland owe much 
of their present character to the fact that the parent rocks were 
of the type seen in the marginal zones of the Ross of Mull granite. 
But, to return to the consideration of the mode of attack 
followed by the granite in this area : what has really happened can 
oasily be made out. The granite sends forward into the schist thin 
wedges of its own material, which thicken as they advance along 
the joints or other divisional planes, and do so at the expense of 
the schist. The impression one gathers from a study of numerous 
•examples of this nature is that the whole periphery of the granitic 
magma exercised a corrosive effect wherever it came into con- 
* See a papei’ by the present author, “ On a Granite Junction in the Isle of 
Mull,” Geol. Mag., dec. iii., vol. ix. pp. 447-451 (1893). 
