W. B. Wright— The Drumlins of South Donegal. 157 



leave the median line standing up in the form of a crest. It is difficult 

 to understand, however, how this process could produce a crest on one 

 drumlin and leave another close by quite unaffected, as is often the 

 case. I have therefore come to the conclusion that this feature ia 

 original and that the form of the drumlins is approximately the same 

 as when they were iirst abandoned by the ice. 



The mere occurrence of a crest gives ground for presuming that, in 

 the building up of the drumlins, material is deposited mainly along 

 the median line, and subsequently distributed down the slopes by the 

 ice-motion. Its uneven serrated form is also in favour of thia 

 explanation as opposed to any process of ice-erosion or ice-moulding. 

 The fact that crested and uncrested drumlins occur in the same 

 area side by side is of considerable significance. It is explained by 

 supposing that in the case of the crested drumlins deposition of 

 material continued until the final exposure of the drumlin or until the 

 ice became stagnant, while in the case of the uncrested drumlins 

 deposition ceased before the ice came to rest, so that the subsequent 

 ice-motion imparted a rounded outline to the top of the drumlin. 



Fig. 4. View looking W.S.W. from a point one mile north of Ballintra 

 showing several types of drumlin, the most numerous and typical having 

 serrated crests. In the centre of the picture is a small drumlin of half- 

 cigar form, and in the foreground on the right another of rounded form. 

 Ben Bulben in the distance. The contours of the numbered drumlins are 

 shown in Fig. 5. 



Throughout the district examined there seems to be a well marked 

 prevalence of drumlins of elongated ground-plan in areas where tliere 

 is reason to suppose the fluctuation of ice-motion was least. It must 

 be admitted, however, that the number of striae recorded in these 

 areas is not sufficiently great to prove the absence of variation in 

 direction. It can, however, be very definitely stated that in the one 

 area, that lying west of Pettigo, where a distinct oscillation has been 

 proved, the drumlins assume shortly oval, rounded, triangular, and 

 crescentic forms. The reason for this fluctuation is to be found in the 

 existence of fairly lofty hills to the south of Lough Derg. During 

 the later stages of the giaciation these formed a barrier to the ice 

 coming from the north-east, so that the flows converging on the head 

 of Lough Erne from the east and south-east obtained the upper hand. 

 This consideration of the local conditions assures us therefore that in 

 this area of conflicting strife the north-westerly motion was later than 

 the south-westerly. Moreover, there is a fair presumption that in the 

 eastern part of this area the change in direction was less abrupt than 



