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



to severer climatic conditions gives us some insight into the effects 

 of reduction of temperature on the activities of the ice-sheet. 



In virtue of its discharge into the waters of the late-glacial sea 

 the outwash from the ice-margin, during its retreat across Finland, 

 has produced in that country a series of clays with well marked 

 seasonal laminae, which give an effective measure of the rate of 

 ablation. During the several halt-stages, the most marked of which 

 are those corresponding to the inner and outer wreaths of the 

 Salpausselka Moraine, ablation as recorded by these lamdnse was 

 reduced to a minimum. It is a fairly safe deduction that these halts 

 in the retreat were the result of a lowering of temperature, probably 

 of mean annual temperature, but most certainly of summer 

 temperature, since the coarse summer zone is almost wanting in the 

 laminae. Further, on mapping out the successive portions of the 

 ice-margin and plotting the rate of retreat against the corresponding 

 year, Dr. Sauramo discovered that, in almost every case, a halt was 

 followed by an acceleration in the rate of retreat, lasting for a 

 period of some fifteen years or so, after which there was a gradual 

 reversion to the normal rate. From this result, combined with the 

 fact that the laminated clays corresponding to these years of rapid 

 retreat show no evidence of abnormally great ablation, he draws the 

 conclusion that during the preceding halt-stage the margin of the ice- 

 sheet had become thinner, and consequently wasted away more 

 rapidly Avhen the normal ablation was restored. 



Now, wasting without any equivalent ablation is capable of one 

 interpretation only. It clearry must be due to evaporation from the 

 surface of the ice-sheet, that is, passage from the solid to the gaseous 

 state either directly or with so little development, of water that the 

 latter merely moistens the surface snow and forms a zone of slush, but 

 never gathers into runnels and streams. The zone of slush, according 

 to the observations of Nansen, is not, even in the case of Greenland 

 ice, which is undergoing ablation, of any great mdth. The con- 

 clusion, therefore, to be drawn from Dr. Sauramo's observation is 

 that direct evaporation from the ice may, under sufficiently low 

 temperature conditions, become an imjaortant factor in glacial 

 control. 



The Glacial Anticyclone. — It is now widely recognized that an ice- 

 sheet, once it attains sufficient magnitude, establishes its own 

 permanent meteorological conditions and wind system, the so-called 

 " glacial anticyclone ". It is nourished, not by snowfall as we know 

 it, but by frost-snow or ice-crystals precipitated from the descending 

 air of the anticyclone when it comes into proximity with and is 

 chilled by the surface of the ice. The central portions of the anti- 

 cyclone draw upon higher, colder, and drier regions of the atmosphere 

 than the peripheral. The peripheral portions of the ice-sheet are 

 therefore better nourished than the central, more especially if they 

 adjoin an ocean. This is probably the explanation of the westward 

 shifting of the centres of giaciation in the British Isles emphasized 



