W. B. Wright— A Theory of the Marginal Drift. 555 



of a state of balance between precipitation and evaporation. 

 In districts where there was a copious supply of moisture from the 

 adjoining ocean, this balance was probably deferred to a very late 

 stage or perhaps never reached at all. If the amelioration of climate 

 which brought about the active motion and effective ablation of the 

 later stages of the retreat supervened before this state of balance 

 was reached, or even before the line of maximum extension was 

 abandoned, then there could be no marginal featureless drift and no 

 feather-edge in such a district. If there was ever any dead ice stage 

 in these portions of the periphery, it did not outlast the maximum. 

 Summary. — Reviewing the argument once more so as to bring out 

 its salient characters, we find that it may be briefly stated as follows. 

 The investigation of Sauramo has established the efficiency of direct 

 evaj)oration as an agent of glacial control during a cold spell. It is 

 reasonable to suppose that during the much greater rigour of the 

 maximum it was still more effective. Moreover, owing to the anti- 

 cyclonic circulation the growth of the ice-sheets tended to reduce their 

 nourishment and increase the depletion by evaporation." A balance 

 may thus have been established between nourishment and depletion 

 without the glacier front reaching the zone of ablation. When such 

 a balance was established, however, motion continued owing to the 

 already established gradient. In virtue of this residual motion, 

 combined with the radial snow-drift, the ice-margin was advanced 

 to its limit and subsequently withdrawn therefrom as a result of the 

 still further increased evaporation, until it once more took up a 

 position at the line of balance. This line in Europe is probably 

 the northern limit of the loess, a short distance south of the " Young 

 End-moraine". Here the ice lingered until an amelioration of 

 climate gradually introduced marginal ablation as a new factor. 

 This change was accompanied by greater precipitation, an inevitable 

 result of increased evaporation, due to the rise of temperature, in the 

 extra-glacial region. The effects were sUglit at first, and the 

 beginning of the withdrawal unmarked by any recognizable moraine. 

 After a little, however, a new gradient was established, and the 

 ablation becoming effective, moraine formation was initiated. The 

 first of the moraines produced by this combination of active motion 

 and rapid decay was that known as the Young End-moraine, and 

 from this in to the centre a progressively greater development of 

 moraines, eskers, and outwash gravels characterized the retreat. 



