522 



NATURE 



[March 30, 1893 



The question of the glacial origin of lakes involves many 

 separate considerations. While lakes undoubtedly abound in 

 regions now or formerly subjected to glaciation, many of these 

 are formed by the damming of valleys by moraine heaps, or by 

 extensive landslips. The school of Sir A. Ramsay afifirm that 

 glaciers are powerful excavating agents, and that there is no 

 other agent but ice competent to form a rock-basin. The last 

 argument breaks down when one considers the number of de- 

 pressions of all sizes gradually increasing from mere volcanic 

 craters to those of the Jordan Valley and the Caspian Sea, in the 

 formation of which ice could have had no part. The argument 

 that Greenland alone holds the key to the phenomena of glacia- 

 tion breaks down, for the Alps were once the seat of a vast ice- 

 sheet, which over-rode all the minor inequalities of the sur- 

 rounding country, and of which the existing glaciers are the 

 shrunken remnant. Thus the Alpine valleys should serve to 

 show the typical results of ice-action on the land. This is the 

 sum of their evidence : toothed prominences have been broken 

 or rubbed away, the rough places have been made smooth, the 

 rugged hill has been reduced to rounded slopes of rock (like the 

 backs of plunging dolphins). But the crag remains a crag, the 

 buttress a buttress, and the hill a hill ; the valley also does not 

 alter its leading outlines, the V like section so characteristic of 

 ordinary fluviatile erosion still remains ; all that the ice has done 

 has been to act like a gigantic rasp ; it has modified, not revolu- 

 tionised, it has moulded, not regenerated. No sooner do we 

 come to study in detail the effects of the ancient glaciers in the 

 upper valleys of the Alps than we are struck by their apparent 

 inefficiency as erosive agents. Here, where the ice has lingered 

 longest, just beneath the actual glacier we see that a 

 cliff continues to exist. Again and again in a valley we may 

 find that on the lee side of prominences crags still remain, some- 

 times in sufficient frequency to be marked features in the scenery. 

 The Haslithal is an excellent and representative example. 

 The result of prolonged personal study of the Alps may be 

 summed up in the words — " Valleys appear to be much older 

 than the Ice Age, and to have been but little modified during 

 the period of maximum extension of the glaciers." 



The evidence as to the erosive power of glaciers is very 

 slight. Dr. Wright showed that the great Muir Glacier in 

 Alaska covers great stretches of undisturbed gravel in which 

 upright tree-stems remain. Prof. Bonney proceeded to say : — 

 In the Alps about the year i860 the glaciers began to dwindle. 

 By 1870 considerable tracts of bare rock or debris were ex- 

 posed, which a dozen years before had been buried under the 

 ice. On none of these have I seen any basin -like hollow or 

 sign of excavation as distinguished from abrasion. The Unter 

 Grindelwald Glacier in the last stage of its descent passes over 

 three or four rocky terraces. The angles of these are not very 

 seriously worn away, nor are hollows excavated at the base of 

 the steps. The bed of the Argentiere Glacier (I made my way 

 some little distance under the ice) was rather unequal, and was 

 less uniformly abraded than I had expected. There were no signs 

 whatever of the glacier being able to break off or root up blocks 

 of the subjacent schistose rock : it seemed simply to wear away 

 prominences. This also is true of other glaciers. Prior to 

 i860, and again in 1 891, I saw glaciers which were advancing. 

 They ploughed up the turf of a meadow for a foot or two in 

 depth ; they pushed moraine-stuff in front of them, showing 

 some tendency to over-ride it, and nothing more. In 1875, ^t 

 the foot both of the Glacier des Bois and of the Argentiere 

 Glacier, was a stony plain. Both these proved to have been 

 recently uncovered by the ice ; in other words, the glacier had 

 not been able to plough up a boulder-bed even at a place where, 

 owing to the change of level, some erosive action not unreason- 

 ably might have been expected. But, further, on both these 

 plains big blocks of protogine were lying which were striated 

 on sides and top, thus showing that the ice had actually flowed 

 over them, as if it were a stream of mud. Some of the difficul- 

 ties in the way of believing in the scooping out of lake-basins 

 have now to be considered. 



First, in regard to their position : some of them, such as Con- 

 stance, Geneva, Como, Maggiore, &c., are comparatively near 

 to the lower limits of the great ice sheets, and so would be 

 covered for a relatively short time. All of them are many miles 

 from the ends of the existing glaciers, yet we are asked to admit 

 that a rock basin, in depth sometimes exceeding 1000 feet and 

 generally more than 500, has been scooped out in a time much 

 shorter than that which has proved insufficient for the obliteration 



NO. 1222, VOL. 47] 



of the original features of the upper valleys or for the deepening 

 of their beds by more than a few yards at most — indeed, as a 

 rule, the ice seems never to have been able to overtake the 

 torrent. 



The radiating arms of the Lakes of Lucerne, Lugano, and 

 Como are insuperable difficulties in the way of accepting a glacial 

 theory of the origin of these lakes, and the configuration of the 

 Lake of Geneva and the other lakesin France recently minutely 

 surveyed, lends no countenance to the theory of excavation. 



One fact to which Prof. J. Geikie has called attention seems 

 at first sight strongly to support Sir A. Ramsay's hypothesis, 

 and is the only real addition, in my opinion, which has been 

 made to the original reasons. It is that many of the Scottish 

 lochs are true rock basins, and that similar basins frequently 

 occur outside their mouths. This also often holds of the fjords 

 in Norway, New Zealand, and elsewhere. Prof. Geikie points 

 out that several of these basins occur just when the ice might be 

 expected to obtain an increased scooping power. His map at 

 first sight appears very convincing ; but a study of the larger 

 charts reveals many anomalies. Loch Linnhe, for example, from 

 below the entry of Loch Leven, maintains a general depth of 

 from 34 to 50 fathoms ; then, below Loch Corrie, a channel 

 may be traced whici varies in depth from 50 to 60 fathoms, after 

 which, in the Lynn of Morven, we find it deepen to 70 fathoms, 

 then to 90 fathoms ; and at last, a little north-east of the line 

 joining Barony Point with Lismore Point, it expands into a basin 

 with a maximum depth of no fathoms. But outside, in the 

 Sound of Mull (to the north-west) the depths become very ir- 

 regular, varying from about 35 to 70 fathoms. Barony Point 

 appears to be connected with Mull by a submerged isthmus, 

 generally less than 20 fathoms below the surface. But here, if 

 the glacier were stopped by impinging on Mull, it ought in 

 splitting to be pushing hard upon its bed. In all this region the 

 irregularities of the bed are very perplexing, whatever 

 hypothesis be adopted ; but I will restrict myself to a single in- 

 stance. OfTthe west coast of Scarba, under the lee of the "Islands 

 of the Sea," and where the opening towards Colonsay makes it 

 improbable that the ice can have forced into a narrower space, 

 an elongated basin occurs in which the soundings — outside about 

 60 fathoms — deepen to 100, and at one place to 137 fathoms. 

 The sea-bed about Arran presents similar difficulties. In short, 

 here, at Loch Etive, Loch Lomond, and in other places, all goes 

 well only so long as we restrict ourselves to generalities and 

 abstain from details. 



The theory of the origin of rock-basins, which I brought 

 forward full twenty years ago, is now supported by much addi- 

 tional evidence. It is that the lake beds are ordinary valleys 

 of sub-aerial erosion, affected by differential earth-movements. 

 This has been very strongly confirmed by the surveys of the old 

 beaches of the great lakes of North America, the Iroquois 

 beach being full 600 feet higher at the north-eastern part than 

 it is at the western end of Lake Ontario. 



To conclude, glaciers, when the paths which they have 

 traversed are carefully studied, appear to have acted, as a rule, 

 as agents of the abrasion rather than of erosion. Even in the 

 former capacity they have geneially failed to obliterate the 

 more marked pre-existent features due to ordinary fluviatile and 

 sub-aerial sculpture. In the latter capacity they seem to have 

 been impotent, except under very special circumstances ; thus, 

 while we may venture to ascribe to glaciers certain shallow 

 tarns and rock basins in situations exceptionally favourable, we 

 cannot assign to their agency either the greater Alpine lakes or 

 any other important lakes in regions which were overflowed by 

 the ice only during the period when it attained to an abnormal 

 development. In the discussion which followed the paper. Dr. 

 Blanford, Sir Henry Howorth, Mr. Freshfield, and Mr. Conway 

 took part. 



FURTHER STUDIES ON HYDRAZINE. 



A FURTHER contribution to the chemistry of hydrazine, 

 '^*- N2H4, is communicated by Prof. Curtius to the current 

 number of the Berichte. The first portion of the memoir deals 

 with the preparation and properties of substituted hydrazines 

 containing the radicles of the organic acids. In the latter portion 

 a number of inorganic salts containing hydrazine are described. 

 When hydrazine hydrate is brought in contact with the amides, 



