208 MESSES, p. F. KENDALL AND E. G. BELL ON 



lat. 56 °V N., long. 34° 42' W., H.M.S. ' Valorous ' found a depth of 

 only 690 fathoms, and 100 miles to the west 1450 fathoms. 



Thus it will he seen that a submarine ridge, the crest of which is 

 nowhere 500 fathoms beneath the surface, extends from Scotland to 

 Greenland, while deep water occurs on each side. The inequalities 

 of the bottom we hold to be quite incompatible with the idea of the 

 permanence of marine conditions over the area from times as remote 

 as the Miocene period. 



Apart from the rather doubtful " Land of Buss," evidence is not 

 lacking on the coast of Greenland of considerable subsidence during 

 the historic period ; and this movement may be a continuation of the 

 great subsidence which depressed the great N.W. barrier. 



The former existence of such a barrier has been suggested by 

 several previous observers, who based their opinions upon widely 

 different grounds. 



The late Prof. Porbes believed that it did not extend to Scotland 

 during the Crag period, but to Norway ; we think, however, that 

 the presence of a great number of Arctic forms in the Crags entirely 

 negatives that supposition. 



Prof. Boyd Dawkins has argued in favour of a western closure 

 of the Korth Sea from the distribution of the Upper Tertiary Mam- 

 malia. 



There are in our list only four shells which offer any obstacle to 

 the acceptance of the theory which we have put forward. 



One of these, Cardium elec/antulum, Mull.,= (7. strigilUferum, 

 "Wood, is found in the Coralline Crag and the Cotentin Marnes a 

 Wassa, and its Jcnoivn recent range is strictly northern, the southern- 

 most extension being to Norway. It must be borne in mind that it 

 is a very small shell, and that our knowledge of the distribution of 

 small MoUusca, both recent and fossil, is very imperfect. 



Columhella sulcata is only known from the English and Belgian 

 Pliocenes ; but so many allied species are found in the Italian Ter- 

 tiaries that it is very questionable if it be not identical with one or 

 other of the forms described by Bellardi. 



Littorina subaperta is a very doubtful species, having a range of 

 variation which carries it even beyond generic limits. 



A much more important argument against our position is fur- 

 nished by the occurrence of Conovulus ijyramidalis, inasmuch as it 

 is only found in the East Anglian and Belgian Upper Crags, the 

 St. Erth Pliocene, and the Wexford gravels and clays (Glacial). 

 Begarding this shell, it should be noticed that there is a singular 

 paucity of evidence respecting the marine members of the genus. 



The southern connexions of the St. Erth Sea appear to have been 

 much more direct with the Mediterranean than at present. The 

 Cotentin deposits, and those on the Loire Inferieure and in the 

 Bordeaux region, are in obvious relation to the Perpignan beds, 

 and indicate a gulf extending from the Mediterranean across S.W. 

 Erance and along the northern side, then cutting off the Einisterre 

 district, throwing an arm across to the Cotentin, and thence running 

 in a westerly direction to St. Erth. 



