116 JF. B. WrUjlit — Submerged Fure-st-s in Donegal Lakes. 



instability of the sea-level, their presence is little to be wondered at. 

 In all bodies of water which have no outlet — and the sea is only the 

 largest of these — the surface-level is determined by the balance between 

 supply and evaporation, and oscillation is inevitable. In the case 

 of freshwater lakes, however, the level is determined within small 

 limits by the overflow, and the occui-rence of tree stumps in the 

 position of growth beneath their waters presents therefore a problem 

 of considerable interest. 



The Swedish cases already described may be classified under three 

 heads — 



(1) Trees which have become submerged owing to the transgression of the 

 waters at one side of the lake as a consequence of the post-glacial tilting of the 

 country. 



(2) Trees which are submerged in lakes the outlet of which is through 

 a marsh or peat moss. These cases admit of a simple explanation through 

 choking up and raising of the outlet by the growth of vegetation. 



(3) Cases where the outlet is over rock or drift, in which there is either no 

 tilting, or the tilting is towards the outlet, or towards the opposite shore from 

 that on which the submerged trees occur. 



Several examples of class 3 have been discovered, and it is these 

 Avhich have a more special interest for us, for they are held by certain 

 of the Swedish geologists to be capable of only one explanation, 

 namely, that the lake during some period of its history was without 

 outlet, and either entirely dried up or was considerably below its 

 normal level. 



JTow in Ireland cases referable to class 2 are not uncommon,^ but 

 I am not aware of any case having, up to the present, been either 

 described or observed which is without question referable to class 3. 



In the summer of 1910, however, while engaged in geological 

 mapping in the mountain and moorland district north-west of Pettigo, 

 Co. Donegal, I came across several cases of this nature, having made 

 a systematic search for evidence of change of level along the shores 

 of "all the lakes in the district. An area of 22 square miles of 

 moorland was examined, and this was found to contain twenty-six 

 small lakes. In seven of these submerged stumps of fir-trees can 



p§. norra delen af^det Snialjindska hoglandet " : Sveriges. Geol. Unders., 

 ser. C, No. 204 (= Arsbok i, No. 1), 1907. 



Kutger Sernander, " Hornborgasjons nivaforandringar " : Geol. Foren. i 

 Stockholm Forh., Bd. xxx, p. 70, 1908. 



' ' Hornborgasjons nivS.f6randringar och vara hogmossars bildningssatt ' ' : 



ibid., Bd. xxxi, 1909. 



" Sjon Hedervikens vegetation och utveklings historia " : Sv. Bot. 



Tidskr., Stockholm, Bd. iv, 1910. 



L. von Post, " Skarbysjo-komplexet och dess draneringsomrSdes postglaciala 

 utvekling " : Geol. Foren. i Stockholm Forh., Bd. xxxi, 1909. 



Eutger Sernander, " Die schwedischen Torfmoore als Zeugen postglazialer 

 Klimaschwankungen." Die Verdnderungen des Klimas : Geol. Congress, Stock- 

 holm, 1911, p. 197. 



Gunnar Andersson, "Swedish Climate in the Late Quaternary Period" : 

 ibid., p. 247. 



^ There is a submerged forest in Loch Fada in the Island of Colonsay, off the 

 west coast of Scotland, to which attention was first called by Symington Grieve. 

 The outlet of this lake is, however, through a peat moss ; see W. B. Wright and 

 E. B. Bailey, Geology of Colonsay and Oronsay (Mem. Geol. Surv. Great 

 Britain, 1911). 



