ON THE PLANT REMAINS OF THE SCOTTISH PEAT MOSSES. 823 



vegetable accumulations, thus causing a rise in the water level, resulting in destruction 

 of the trees. 



This theory appears to account for the facts so easily, that it has been used to ex- 

 plain the destruction of either the one or the two forests generally found in the 

 older peat deposits of N.W. Europe, without a critical examination of the whole 

 evidence. 



These buried forests, at any rate in Scotland, are very seldom associated with 

 Sphagnum peat. Further, both the Lower Forest and the Upper Forest in many cases 

 occur in districts and at altitudes beyond the present range of trees, and contain the 

 remains of a ground flora which now has a more southerly distribution. 



These features are important, for they are met with over so wide an area and are so 

 clear that they point to the conclusion that these two forests are the result of past 

 climatic changes. I think this conclusion would be fully justified even if it rested 

 solely upon the botanical evidence. 



In a most interesting passage Gunnar Andersson (7) discusses the constancy of 

 biological characters. He says : " There are whole troops of species, living under the 

 most varied ecological conditions, plants as well as animals alive in the air and in the 

 water. If they give the same result, then we have reached an empirical certainty of 

 probabilities that all of them cannot have changed their biological demands in the same 

 direction. ..." The comparison of the anatomy of these recent fossils with their re- 

 presentatives of the present day is fraught with great difficulties owing to the frag- 

 mentary condition of most remains, other than seeds and fruits. Leaves, of course, 

 would be most important, but unfortunately they are seldom sufficiently well preserved 

 to show detailed structure. The onus of proving that the ecological requirements of 

 certain species have changed during the tens of thousands of years of post-glacial time 

 lies with those who make the statement. If each stratum in the peat were distinguished 

 by only one or two plants, the possibility of such a change in ecological requirements 

 would seriously impair their value as indicators of climate ; but when numbers of species 

 are recognised in each stratum, is it not too much to assume that all these have varied 

 ecologically at the same time and in the same direction, either say as to their optimum 

 temperature for ripening of seed, or their power for living under moister or dryer 

 conditions ? 



While the observations I have made support the Blytt-Sernander view that the 

 forest beds are actual datum lines, the plant remains do not in any way support Blytt's 

 (16) original view that each forest bed was formed under dry conditions. At the same 

 time, the peat appears to have been formed at a slow rate, for layers of birch bark are 

 sometimes met with, the wood having disappeared, showing a considerable lapse of 

 time between the death of the tree and the covering of its remains by peat. While 

 the conception of the forest bed as a distinct stratum in the peat may be granted, it 

 may be well questioned whether the forest bed in one district in Scotland may be 

 contemporaneous with that in another district, i.e. whether the forest beds in different 



