324 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



rectly upon this beach; elsewhere in the northwest Highlands, the 

 raised beach, which is developed at the seaward ends of the lochs, 

 may be absent at tlie upper ends, possibly because the ice then reached 

 the sea. The snow line at this period stood at 2,400 to 2,500 feet. 

 The submergence was in fact associated with a cold period which, 

 in the peat bogs outside the limits of the glaciers, finds its expres- 

 sion, in artic beds witli Salix herhacea^ S. reticulata^ Betula n/ina, 

 and EmpetruTn nigrum. The Mollusca of the 45 to 50 foot beach 

 are mostly of local species, with some northern forms. This sub- 

 mergence was followed by a renewed elevation above the present 

 level, accompanied by a drier and warmer climate, which permitted 

 great pines to grow more than 500 feet above the present limits of 

 tree growth in the Highlands. This forest period extended even 

 to the Orkneys, w^here conifers are not now indigenous, for in the 

 Bay of Skaill occurs a submarine forest with roots of small firs, 15 

 feet below high water. This upper forest layer is again overlain 

 by Sphagnum peat, indicating a return of moist conditions. 



In the mountains there appears to have been a very slight re- 

 crudence of glacial conditions after the valley glaciation of the 50- 

 foot beach period ; Geikie was unable to find any direct evidence as 

 to the horizon of these glaciers, but correlated them with this upper 

 peat layer. 



On the coasts there was a renewed submergence of 25 to 30 feet 

 decreasing northward, the beaches of which occasionally overlie a 

 forest layer, which may be either the upper or lower forest layer. 

 These beaches were correlated by Geikie with the upper Sphagnum 

 peat and the corrie glaciers, but this seems unlikely, as the beach 

 nowhere contains any suggestion of a fauna of northern origin, and 

 in 1865 its Mollusca were described by Jamieson (100) as seeming 

 " to have more relations to the south than to the north, indicating a 

 climate if anything milder than the present ; " the corresponding 

 beach in, northeast Ireland has a decidedly warm fauna, but that of 

 Lancashire gives no definite indication of a climate differing from 

 the present. According to J. Geikie, neolithic pine dugout canoes 

 have been found in the Carse clays of the 50- foot beach, but this re- 

 quires elucidation, for they are stated by Jamieson (100) to have 

 occurred in the deposits of the 25-foot beach. 



The succession of events outlined by Geikie is confirmed with this 

 exception of the 25-foot beach by Lewis and G. Samuelsson (101). 



The occurrence of the arctic bed overlying the lower forest bed 

 in regions which were not occupied by the glaciers during the valley 

 glaciation (50-foot beach) is considered by Lewis as proof that that 

 glaciation was due to a return of cold conditions after the climate 

 had once become temperate. Lewis has also found the arctic bed be- 



