352 MR FRANCIS J. LEWIS 



whole post-glacial history of the peat is represented in the plant strata. This is strong 

 evidence that the alternation of Sphagnum or Eriophorum beds with forest zones in the 

 upper layers of the peat is due, not to edaphic factors, but to climate. If the alternation 

 of the plant remains in the upper peat were due to soil conditions, the sequence of the 

 upper beds would scarcely be the same whether they rest almost directly upon drift 

 and a few inches of peat or upon some 3 feet of peat. 



The history of the peat over the present district is of interest, inasmuch as it is in 

 striking agreement with the general history of the peat lying upon the Spey-Findhorn 

 watershed. The oldest layer, represented in some places on the latter watershed by 

 Salix reticulata and *S. herbacea, appears to be wanting, but the dominant plants in the 

 eight superposed beds are the same. At the base of the moss Salix Arbuscula again 

 forms a thick bed of peat, mingled with traces of Empetrum. Coming 6 inches higher 

 in the peat the remains of Salix die out, and its place is taken by Empetrum, which 

 again forms a dense layer of the same character as that met with in Coire Bog and 

 elsewhere in the Highlands, mixed with capsules of Viola palustris and seeds of 

 Arctostaphylos alpina. A layer of Eriophorum-Sphagnum peat overlies the Betula 

 alba zone, which itself contains abundant remains of these plants. The wood at the 

 base of the Betula zone is of much smaller size than that near the top, although abundant 

 remains of Eriophorum and Sphagnum occur throughout ; further, there is no sharp 

 break between the Betula and Sphagnum and Eriophorum layers — one merges gradually 

 into the other. There is every indication that both these layers belong to the same 

 condition of things, the moorland having gradually changed from somewhat cold con- 

 ditions (shown by the stunted character of the birch and the close annual rings) to a 

 state of things when the growth of Sphagnum and Eriophorum became too rapid to 

 permit the further growth of birch. 



Coming into the upper forest-bed, we meet with that duplication of this bed which 

 is so common over the Highland areas. The possible explanation of this phenomenon 

 has already been discussed when dealing with the Spey-Findhorn watershed. It is 

 probable that the districts which are still to be examined in North-West Sutherland and 

 Ross may furnish more direct evidence with regard to this problem. 



Summary and General Conclusions. 



The peat deposits described in this paper may conveniently be divided into two 

 groups : — 



1. Those occurring in the Western districts, possessing no Arctic plants at the base, 

 but a basal forest-bed, overlaid by a considerable thickness of plant remains, indicative 

 of wet moorland conditions. 



2. Those in the North and North-East Highlands, possessing only one well-marked 

 forest-bed, and with an Arctic plant bed at the base of the peat. 



The peat examined in Southern and Central Skye and North Uist falls under the 

 former head, and may be considered first. 



