822 MR FRANCIS J. LEWIS 



aiding plant immigration in this case. The relationship of the Fseroese flora affords 

 strong proof, according to Ostenfeld, that it was not introduced by wind agency, and 

 it appears to me that his case is immensely strengthened by the fact that no definite 

 evidence exists of the stocking by wind of an island that is more than 30 miles 

 from its mother country. 



These points are only mentioned here, because they indicate that, according to some 

 observers, an extent of land surface sufficient to link up Faeroe with Scotland is needed 

 to account for the peculiar relationship of the flora in those islands. If post-glacial 

 changes resulted in the formation of such a land connection, it must have taken place 

 soon after the retreat of the ice from Shetland and Fseroe, and have continued to exist 

 during the Lower Forest period. If any considerable elevation of land did take 

 place at this period, as seems certain from the geological evidence, it would diminish 

 the force of the salt-laden wind in the inland parts of Shetland, and so allow the 

 growth of forest. 



It is difficult to say what meteorological changes a land elevation in N'.W. Europe 

 would produce. But it is quite certain that the climate of this time was at least 

 as warm as the present, and continental in character. I do not regard the low 

 summer temperature alone as an important factor in limiting the growth, but in 

 conjunction with strong and occasionally dry winds its possible influence cannot be 

 disregarded. 



The extinction of this forest might be due to the inroads of animals such as 

 sheep and cattle, as in so many well known cases in which forests have been exter- 

 minated in this way ; the great age of this forest, however, lying as it does immediately 

 above the basal arctic bed and below the second arctic bed, makes it most unlikely that 

 Shetland was then inhabited at all, or if so, by inhabitants with large herds of cattle. 

 Even supposing that this was the cause of its disappearance, we should expect to find 

 traces of the forest still remaining in sheltered places inaccessible to cattle and sheep, 

 such as crannies and gullies on the hills. There are many such places in the quartzite 

 regions of Shetland, but even there not a single wild tree of the smallest dimensions 

 is to be found. 



It appears to me probable that the destruction of the forest was caused by the dis- 

 appearance of the climatological features which allowed its development. 



It is very usually stated by some authors that the growth of high moor at the 

 expense of forest may occur by the decay and falling of trees obstructing the drainage 

 and thus encouraging the growth of Sphagnum ; the production of raw humus and 

 lowering of the aeration and temperature of the peat resulting in the complete replace- 

 ment of tree growth by Sphagnum moor. That this may occur in certain basins I do 

 not deny, but it is quite inadequate to account for the widespread succession in British 

 peat mosses. The evidence that has been used to support this view originated from 

 a study of peat formed in old lake-basins in Denmark and Sweden. Such changes are 

 peculiarly likely to occur in old lake-basins, as the outlet may be temporarily dammed by 



