29 



gress, laying bare the structure of the external layer of the bubble. This portion we 

 must conceive to have been nearer the pre-existing superjacent rocks than the central and higher 

 mass, (which was probably denuded to the nucleus before it was elevated to its present level) 

 and the great variations in its structure and ingredients may have resulted from that 

 circumstance under the influences previously adverted to (1). 



The leading fact relating to the structure of the rocks is , that the principal vertical or 

 approximately vertical planes of division have a general direction approximating to NE. 

 ... SW. It is also observable that the zones of sof ter rock and the majority of the grooves 

 have the same directions. Pulo Ubin lies in the great plutonic band of elevation stretching 

 from Assam to Banca , and having , from Junkceylon southwards , a south easterly direc- 

 tion. The divisional planes are therefore nearly at right angles to the axis of elevation. 

 Is this a general geological phenomenon ? It probably is, because in a locality so far remov- 

 ed as the south west of England, it is repeated. Sir H. de la Beche informs us that the 

 divisional planes of the granite and other rocks in Cornwall and Devon are generally 

 NNW. . . . SSE. The grand conclusion which he draws is, that this direction approx- 

 imates to the present magnetic meridian of the district, and may therefore, in its origin, 

 be related to it. Pre-occupied with this view he has overlooked the fact that this direction 

 is at right angles to the direction of the principal granite masses of the district (E. 24° N. 

 W. 24° S. (2) which , beginning at Dartmoor , are continned to the Scilly Islands. Not 

 merely the south western division of England, but the general configuration of the British 

 Islands, seems to be due to axes of elevation having the same or an approximate range. 

 Thus the great body of Scotland and Ireland may be considered as one connected mass 

 upraised on such axes. Professor Phillips (3) mentions that the anticlinal axes of the High- 

 lands and Lammermuirs in Scotland prolonged to Donegal and Cavan in Ireland, and those 

 of the Cumbrian mountains, the Isle of Man, and North Wales, all range NE. and SW. 

 It also appears from Professor Paillips diagram shewing the result of his examination of the 

 joints in the mountain limestone distriets of the north of England, that the great majority 

 of the divisional planes are there in NNW. and SSE. lines (4). A coincident range 

 has been observed in the joints in other localities in England , and also in France » more 

 particularly" says Sir H. de la Beche » in granites and grauwacke" (5), I think it probable, 



(I) If the views ad vaneed in a subsequent part of this paper are correct, we must recognize in granitic fluid 

 masses a period , in the gradual diminution of their temperature to the fusing point , when the external layer , 

 having just estended itself into the superjacent rocks by melting them into its substance , was arrested in its 

 furtber extension by crystallization. Hence the external layer sliould often be variable , and partake of the chemi- 

 cal ingredients of the adjacent rocks', because time was not allowed for their thorough mixture with the general 

 mass. In fact this layer must often be merely a layer of the adjacent rock fused down and immediately crystalliz- 

 ed into a granite. ( See note, p. 33, Post. ) 



(2) De la Beche's Report p. 157. 



(3) Treatise on Gêology 2nd vol. p. 258. 



(4) ld. lst vol. p. 65. 



(5) Report p. 275. * 



