C— GEOLOGY. 47 



no recognisable bases but seem to extend indefinitely downwards so far 

 as the accessible portions of the crust are concerned. Similarly in Iceland 

 Dr. Hawkes finds that the true laccolitic form of intrusion is seldom met 

 with. This is in accordance with the generally accepted view that lacco- 

 lites are associated with mountain-building movements, whereas every- 

 thing points to the intercrustal stresses in the Tertiary province having 

 been of a tensional character. 



The main plutonic masses of Mull have been shown by the Survey to 

 be due to the rise of magma up ring-shaped or arcuate fissures caused by 

 the general subsidence of a central area. Such fissures are, as might be 

 expected, either vertical or steeply inclined in an outward direction. They 

 thus tend to bound either a cylinder or a steep angled cone. 



Subsidence of a conical block or cylindrical mass of crustal material 

 into a magma-reservoir would naturally cause a welling up of molten 

 matter into the fissure that bounds the subsiding mass. If the fissure 

 reached the surface a cauldron-subsidence similar to that of Grlencoe 

 would probably be established, and eruptions of central type would ensue. 

 If, however, the ring-fracture, instead of reaching the surface was completed 

 by a truncating cross-fracture beneath the surface, the magma would 

 not only ascend along the sides of the subsiding mass but would insinuate 

 itself in the form of a sheet between the top of the sinking mass and the 

 relatively stable crust. The thickness of this practically horizontal sheet 

 would depend upon the amount of subsidence of the block beneath, while 

 the thickness of the cylindrical part of the intrusion would be determined 

 by the inclination of the ring-fracture to the vertical together with the 

 amount of central subsidence. Repeated subsidence that would allow 

 fresh influxes of magma in the widening fissures, or the formation of fresh 

 fractures of a similar kind, would give rise laterally to a succession of ring 

 or arcuate intrusions arranged about a common vertical and generally 

 central axis. 



To such intrusions the term ring-dyke has been applied, but it will be 

 seen that in their most complete form, that is to say when the steep 

 ring-like portions of the intrusion are connected transversely, they present 

 points of similarity to stocks. In fact, in this case, the difference appears 

 to be one of degree rather than of kind, for it is quite probable that 

 many stocks, if sufficiently denuded, would reveal a ring-dyke structure in 

 depth. 



Subsidence with the quiet welling up of magma from beneath gives 

 rise to intrusions that have one great distinguishing feature from those 

 which are due to the forcible injection of magma, or which occupy regions 

 affected by mountain-building forces. They will not disturb to any extent 

 the rocks into which they are intruded but will simply replace a definite 

 block of the pre-existing crust, and at the same time transgress without 

 serious modification all pre-existing structures. Such were found to be 

 features presented by the ring-intrusions of Mull. 



A further point of interest is that within the same area of general 

 subsidence the centre of actual subsidence may, and does, shift its position, 

 so that an earlier series of ring-dykes may be intersected by another 

 series of later date, of which the axis is not coincident with the first but 

 removed some distance from it. In Mull the successive centres of subsidence 



