96 Proceedings of the Rojfal Irish Academy. 



borizontaliiy, wliicli is not included within llie limits of error inseparable 

 from such observations of level as it has been found possible to mate. 



The definiteness and well-marked character of the shoreline of this sheet 

 of Virater, which may conveniently be referred to as Lake Kenmare, is without 

 doubt largely due to its constancy of level. The sill of the outlet is composed 

 of very hard rock, and has not been appreciably lowered by the outtlowins: 

 waters. Tlie steep-sided pass through which the discharge took place is a 

 very remarkable feature of the district, but in no sense owes its origin to 

 glacial drainage, or even to the excavating action of ice. Its course, far 

 from being coincident with the trend of the ice-motion, is in one place directly 

 transverse to it. The floor has, nevertheless, been everywhere plucked and 

 scoured by the passing ice. The eflect of the outflowing waters of the lake 

 is only noticeable as a slight erosion and potholing of these ice-moulded 

 rocks. 



During the initial stages of Lake Kenmare, while it was still only a mile 

 or two in length, a small lake was also impounded in the passatlJerrincullig, 

 about three miles north of Kilgarvan. This small lake formed a well-marked 

 terrace on its northern shore. In the Slaheny Valley, south of Kilgarvan, 

 there was a somewhat larger lake, the surface level of wiiicli is recorded in 

 some terraces in the valley of the Gla.shagorruv Kiver, above Glanlough. 'I'he 

 outlet of this lake was eastward through the beautiful dry gap of Crumagloun 

 into Lake Kenmare. 



Connexion between the shore-rmbiintmni/s of Lalcc Kenmare and the 

 marginal drainuiji'. 



On e.\aniining the arrangement of tlie gravel terraces along the shore-line 

 of jjike Kenmare, it becomes at once apparent that the materials of which 

 they were built up were derived in tiie main from the lateral drainage of the 

 ice-sheet. As evidence of this, it is clear in the first place that, if the terraces 

 were produced by ordinary shore action and the inwash of streams, they 

 ought to be inci easingly more massive and better defined towards the eastern 

 end of the lake, which was longest in existence. A glance at the map will 

 show that this is not the case, the terraces at Caber, half way between 

 Kilgarvan and Kenmare, being just as well developed as any further east. 

 Moreover, many streams which poured into the lake have made no deltas at 

 their point of discharge. For example, the Owbeg Itiver, north of Kilgarvan 

 has built up no delta at Meelick, where it reached the level of the lake, 

 whereas a small lateral tributary of this river to the west of Meelick has an 

 immense plateiiu of gravel in the lower part of its valley. On the south side 

 the Slaheny River formed no delta, whereas its tributary on the west, the 



