C— GEOLOGY. 51 



In these last-named regions they exist as the usual sheet-like intrusions, 

 directed towards the apex of an inverted cone which is situated deep down 

 in the plutonic centre. They constitute a very definite and important 

 stage in the igneous history of the respective regions. Although of great 

 thickness, they exhibit a marked constancy of type, but have been intruded 

 at more than one period. 



The most interesting feature connected with their distribution in Mull 

 is that each successive centre of subsidence has its own suite of cone- 

 sheets, a fact that even more than their central inclination seems to connect 

 them with the respective ring-dyke centres. The same is true in the 

 case of Ardnamurchan, for here also each of the two centres has its related 

 cone-sheets. 



Cone-sheet intrusion, therefore, is definitely connected with the 

 establishment of a local magma reservoir beneath an area of central 

 subsidence. It is a phenomenon that can be repeated on several occasions 

 during the formation of a ring-dyke complex, but it generally comes to 

 an end before the last plutonic members of the complex have established 

 themselves. In Mull the Loch Ba ring-dyke is later than all the cone- 

 sheets, and in Ardnamurchan the inner and later members of the complex 

 appear to be cpiite free from such intrusions. 



The relation of cone-sheet to ring-dyke intrusion has been discussed 

 theoretically by Mr. Anderson, who attributes the conical fracturing and 

 the subsequent injection of magma to the development of excess pressure 

 in the upper portion of a cylindrical or parabolic magma-reservoir. He 

 proves that in the event of an increase of pressure in the magma-basin 

 the crust above would have superimposed upon it a system of tensions 

 acting across surfaces which near the basin are roughly conical. At 

 the same time a system of upward pressures would act across surfaces that 

 were parallel to the roof of the magma-basin. The effect of such forces 

 would be the opening of conical fissures followed by the uprise of magma 

 from the reservoir beneath. 



Expressed in other words, the production of cone-sheets may be 

 described as the result of an effort on the part of a deep-seated magma 

 to raise its roof by fracture rather than by flexure. The uplift of a roof 

 by such means is purely differential, but in the aggregate the uplift, as 

 gauged by the cone-sheets, must be very considerable. In Mull the 

 combined thickness of the cone-sheets reaches several thousands of feet, 

 and Mr. Bailey considers that the central elevation which they denote 

 is quite comparable to the doming of the roof of the Arran granite. 



A highly characteristic feature of certain of the plutonic centres, such 

 as Mull, Arran, and the Mourne Mountains, is the occurrence of swarms of 

 basic dykes. These dykes, which show a definite orientation, are crowded 

 together in the region of the plutonic intrusions but extend far beyond the 

 limits of the areas of local subsidence. It is clear, therefore, that their 

 intrusion was influenced but not wholly controlled by these centres. The 

 Mull swarm has a width of some 8 miles and is over 100 miles in length. 

 Such dyke-swarms, related to older plutonic intrusions, are well known in 

 the regions of the Ettive and Ben Nevis granites. 



In the Tertiary province intrusion of the dyke-swarms extended over a 

 considerable period, but was determined by a renewal, possibly in a less 



E 2 



