ISLANDS. 593 



not entirely, due to the action of the sea, the irregularities being 

 for the most part produced by the alternation of hard and soft 

 rocks, the former constituting the headlands, the latter the bays. 

 The indentations of the coast are more prominent on the western 

 and southern coasts, because there the alternation of rocks of varying 

 texture is more marked than on the east. Recent elevation and 

 depression, as evidenced by Raised Beaches and Submerged 

 Forests, have naturally modified the influence of the sea. That 

 Britain once formed part of the Continent was naturally surmised 

 by early writers, on account of the similarity of the Chalk cliffs of 

 Dover and Calais; while Verstegan, in 1605, inferred the former 

 connection, from the occurrence of the Wolf, observing that no 

 man " would ever transport any of that race for the goodness of 

 that breed, out of the Continent into any Isles." ^ The English 

 Channel is due mainly to depression rather than excavation." 



Islands have been formed through denudation by sea and rivers, and by depression ; 

 but there are some areas which, although they retain the name, are no longer 

 islands, and have been united to the mainland by the accumulation of material, by 

 elevation, or by artificial barriers, while other so-called islands never have been 

 isolated. 



Sir A. C. Ramsay believes that the Isle of Anglesey was separated from the 

 mainland by a great glacier or ice-sheet which came from the north-east, although 

 for a time it was joined to the mainland by an undulating plain of Boulder Clay ; 

 but the effects of faulting and depression, combined with the action of the sea, 

 have had considerable influence on the form of the Menai Straits, whose physical 

 features, in Mr. Strahan's opinion, appear to have existed in more or less their 

 present form in Pre-Glacial times. ^ 



The Isle of Purbeck is not really an island, being only to some extent isolated 

 by the inlet of Poole Harbour, and by streams. Portland, too, is connected with 

 the mainland by the Chesil Bank, based on Kimeridge Clay. Barry and Sully 

 Islands, on the coast of Glamorganshire, are connected with the mainland at low- 

 tide ; so also are Holy Island or Lindisfarne, off the Northumberland coast, and 

 St. Michael's Mount in Cornwall.'* These instances of promontories, more or less 

 isolated from the mainland, are instructive as showing how islands may be pro- 

 duced by the denudation of the softer strata. 



Thanet, Ely, and Axholme in Lincolnshire, were islands in comparatively 

 recent times, as also the little isle of Athelney in the Somersetshire moorlands ; but 

 Glastonbury Tor — the old Isle of Avalon— is not quite surrounded by Alluvium. 

 The Isle of Ely, formed of Lower Greensand on Kimeridge Clay, is said to have 

 been an island 800 years ago, at the time of the Norman Conquest. The town of 

 Eye in Suffolk is situated on a low hill surrounded by Alluvium, proving that, as 

 its name suggests, it was once surrounded by water or marshy land. Portsmouth 

 is connected with the mainland by Alluvium ; so also is Selsea. Sheppey in Kent, 

 Hayling and Thorney Islands in Hampshire, and Walney Island in Lancashire, 

 were probably connected with the mainland in recent geological times. 



Lundy Island, and the Steep and Flat Holmes in the Bristol Channel, as well 

 as the Isle of Man, have been formed chiefly by the denudation of areas previously 

 connecting them with the mainland. 



The Isle of Wight has doubtless been formed by denudation and depression ; 

 before the Chalk ridge of the Isle of Wight was separated from that of the Isle 

 of Purbeck, the rivers draining into Southampton water, and into what is now the 



^ Restitution of Decayed Intelligence, in Antiquities concerning the English 

 Nation, etc. 



^ God win- Austen, Q. J. vi. 87; see also Ami Boue, Q. J. xii. 325. 



•^ Ramsay, Q. J. xxxii. 116. See also Sedgwick, Proc. G. S. iv. 222; 

 H. Hicks, Q. J. xxxii. 121 ; A. Strahan, Q. J. xlii. 386. 



^ See D. Pidgeon, Q. J. xli. 9. 



38 



