318 



NATURE 



[February 4, 1892 



Department of the Interior, United States Geological Survey, 

 Washington, D.C., December \2, 1891. 



Prof. G. Frederick Wright, Oberlin, Ohio. 



My DEAR Sir, — Your letter, calling attention to Mr. Bonney's 

 remarks on the nature of cirques in a review of your "Ice 

 Age in North America" (Nature, vol. xliv. p. 537), led me to 

 think that possibly all who have written on their character and 

 origin were not considering the same phenomena. This 

 suggestion was strengthened on referring once more to several 

 of the essays which have appeared on the subject. In Bonney's 

 paper on the formation of cirques (Quart. Journ. Geol. Soc, 

 London, vol. xxvii., 1871, pp. 312-324), three conditions are 

 mentioned, which are stated to be most favourable to their 

 production ; these are : — 



" (i) Upland glens, combes, or terraces, so shaped as to give 

 rise to and to maintain many small streams. 



" (2) Strata, moderately horizontal, over which these streams 

 fall, which, by their constitution, yield considerably to the other 

 forms of meteoric denudation." 



" (3) These strata must nevertheless allow of the formation of 

 cliflFs ; and thus perhaps the most favourable structure is thick 

 beds of limestone, with occasional alternating bands of softer 

 rock." 



In the High Sierra of California — where the cirques are 

 situated, my description of which you quote, and which called 

 forth Mr. Bonney's criticism in Nature — none of the condi- 

 tions mentioned above, except that the rocks are sufficiently 

 durable to stand in cliffs, have perceptibly influenced the forma- 

 tion of the topographic features under consideration. 



The cirques are not situated "in upland glens, combes, or 

 terraces, so shaped as to give rise to and maintain many small 

 streams," but are in the very crests of the mountains, and 

 receive practically no drainage from above. They are not in 

 horizontal strata, but in various locks some of which are highly 

 inclined. The most typical examples occur in granite, which 

 is broken by well-defined vertical joints. 



The rocks are sufficiently hard to allow of the formation of 

 cliffs, but this property is requisite for the production of other 

 similar topographic forms, and not an exception peculiar to 

 those in which cirques occur. 



Not only the cirques of the Sierra Nevada, but hundreds of 

 others in the basin ranges of Nevada and Utah, in the Rocky 

 Mountains, and in the mountains of Southern Alaska, have been 

 formed under conditions which are the reverse of those stated 

 by Mr. Bonney. 



Another class of topographical forms resembling cirques are 

 well known ; these are the "alcoves" formed in the edges of 

 mesas and table-lands where the rocks are essentially horizontal 

 and usually heterogeneous. They are common in the Catskill 

 Mountains, and all along the borders of the plateau on the west 

 of the Appalachians from New York to Alabama. They occur 

 in the great plateau regions of New Mexico and Arizona, and 

 are common in the caiion walls along the Green and the 

 Colorado Rivers, where they were studied by Powell. Modified 

 examples occur in the sediments of Lake Lahontan, and in the 

 bad lands of Dakota, but in these instances the strata are mostly 

 too soft to stand in vertical walls. I have seen good examples 

 at Table Mountain, Cape of Good Hope, and it is safe to say 

 they will be found wherever nearly horizontal rocks, sufficiently 

 durable to form cliffs, have been caiioned or eroded into mesas. 



The mode of formation of alcoves is too well known to 

 require detailed description. They are formed by the action of 

 streams cutting niches in precipices, which are enlarged by the 

 dashes of spray from water-falls, by solution, and by the grind- 

 ing of o'/f^r/i- in the pools into which waters cascade. Alcoves 

 are most abundant in comparatively low regions, for the reason 

 that when strata are greatly elevated, they usually become in- 

 clined or folded ; but there is no reason why they should not be 

 found at any altitude, provided the rock conditions favour their 

 formation. 



It seems to me that alcoves, and not cirques, were formed 

 under the conditions postulated by Mr. Bonney, and that the 

 two are genetically distinct. 



Since writing the account of the cirques of the High Sierra, 

 quoted by you in "The Ice Age in North America," I have visited 

 the Southern Appalachians and the high mountains of Southern 

 Alaska, always having the various problems of mountain sculp- 

 ture in mind. The Southern Appalachians are without evidence 

 of former glaciation, while the mountains of Southern Alaska 

 are now undergoing intense glaciation. In the former, cirques 



NO. TT62, VOL. 45] 



are absent, but in the latter they are abundant, especially on the 

 northern sides of the crests of the mountains, and are the sources 

 from which numerous glaciers flow. 



The initiation of rain and rill sculpture is well illustrated on 

 the sides of many of the valleys of Southern Alaska ; and, as 

 you are well aware, there are many valleys in that region which 

 were formerly filled with glaciers to a depth of 2000 feet or 

 more, but from which the ice has now disappeared. All the 

 minor features due to aqueous erosion on the sides of the 

 valleys, previous to their occupation by glaciers, were removed by 

 the ice. The surfaces were cleared of their old records as the 

 sculpture on ancient monuments was sometimes removed to 

 make room for a new bas-relief. When the ice melted, it left 

 the steep slopes without vertical lines, but more or less deeply 

 scored in horizontal bands. Through analogy with the pene- 

 planes described by Davis, we might call these "pene-slopes." 

 The rain falling on these slopes initiated new lines crossing the 

 glacial grooves at high angles, and forming miniature caiions, 

 which receive numerous branches in their upper courses. Many 

 times, several small stream channels coming together have cut 

 radiating channels separated by buttresses, while below, the 

 stream plunges over cascades or runs through deeply-cut gorges. 

 This miniature sculpture is characteristic of immature drainage 

 on inclined surfaces. The combined excavations made by several 

 tributaries near the heads of these new drainage channels have a 

 somewhat basin-like form, butthe interiors of thebasins are rough, 

 and divided by ridges, and their walls, although steep, do not ap- 

 proach the vertical. These topographic forms in miniature are 

 clearly the work of rain and rills, and have no complicating 

 conditions, except the nature of the strata in which they are 

 cut, and in the best examples this is homogeneous. 



Let us renew our studies in the Southern Appalachians, 

 where, as already stated, no evidence of former glaciation has 

 been detected. About many of the higher summits in that 

 region there are large conical depressions, on the sides of 

 which are many rill channels that converge towards a common 

 outlet. These depressions are many times strikingly crater- 

 like in form, but their sides are sloping, and usually deeply 

 covered with debris, but in no instance are they vertical cliffs. 

 So far as can be judged, these depressions are not dependent on 

 rock structure, and are certainly not confined to horizontally 

 bedded rocks, but in the best examples have been excavated in 

 strata that are highly inclined. They occur near the summits 

 of mountains and on the sides of precipitous ridges, and receive 

 little if any drainage from above. They belong genetically to 

 the same class as the miniature excavations on the pene-slopes 

 of Alaska, but are of far greater size ; and owing to the dis- 

 integration and decay of the rocks, and probably, also, to their 

 fuller development, are much more uniform in contour. 



The depressions about the summits of the Southern Appa- 

 lachians are not cirques, but in my opinion, are of the nature of 

 {hG prototypes of cirques. The general outline of the model has 

 been secured ; but for the final sculpturing, in order to form 

 typical cirques, another tool is required. That tool is ice. 



Should a climatic change occur which would admit of the 

 accumulation of perennial snow about the summits of the 

 Appalachians, the depressions referred to would be the first 

 to be filled, and would give origin to ice streams of the Alpine 

 type. The neve snow accumulating in these depressions would 

 become compacted at the bottom and form glacial ice. As the 

 ice flowed outward, it would remove the debris encumbering the 

 j sides of the basins and attack the rocks beneath, probably in 

 j some such manner as described by Lorange, and quoted by you 



on pp. 244-45 of " The Ice Age in North America." 

 I I do not feel that we can completely analyze the action of 

 I snow and ice in depressions like those in which many glaciers 

 I originate, but there are certain considerations which may be 

 I suggested in this connection. 



I It is ;,well known that ice may be moulded by pressure and 

 made to flow like a viscous body, while it yields but very 

 slightly to tension. When an attempt is made to stretch it, 

 fracture results. Crevasses therefore indicate tension in the 

 glaciers where they occur. 



The crevasses near the upper margins of neves, known as 

 bergschrunds, are really faults, formed by the subsidence of the 

 neve oxi the lower side of a break. They indicate, as has been 

 suggested to me by Mr, W. J. McGee, that the neve where 

 they occur is in a state of tension, and the tendency is for the 

 snow to pull away from the cliffs, and thus tear out portions of 

 the rock to which it adheres. More than this, the crevasses 



