PENEPLAINS OF FRANCE AND BRITTANY 481 



today maturely dissected to a considerable strength of relief, however perfectly 

 they may have been worn down to lowlands in a former cycle of erosion ; and 

 with these examples in mind, alternative explanations have been offered by which 

 to account for the existing forms independent of peneplanation. Tarr's theory of 

 bevelling (to which my response* did, I regret to say, insufficient justice, through 

 misunderstanding of the intended meaning) and Tangier-Smith's and Shaler's 

 theory of the subequality of mature hilltop lieights as a result of the subequality 

 of stream spacing are interesting contributions to the general subject of land sculp- 

 ture, but it does not seem to me that they touch the question in hand closelj'' 

 enough to invalidate the theory of peneplains, inasmuch as they give no sufficient 

 explanation for uplands of relatively even surface, unsympathetic with structure, 

 and only here and there dissected by streams whose valley sides rise steeply to the 

 upland level. Moreover, the forms producible by bevelling and stream spacing 

 are imitated by dissection after peneplanation. What we need is a means of dis- 

 criminating among the forms of these three origins, rather than an argument for 

 the exclusion of one origin because of the competence of the other two. 



Evenly truncated uplands are by no means rare. The four views here presented 

 are all of European localities, not because American examples are wanting, but 

 because my recent excui'sions have happened to afford illustrations from abroad. 



Plate 44, figure 1, is a part of the broad uplands that have been eroded on the crys- 

 talline and stratified rocks, all much disordered, of the central plateau of France. 

 The photograph was taken from the village of Vedrine, in the neighborhood of a 

 railway junction called Eygurande, where the upland is deeply dissected by the 

 Ohavannoux (a northern branch of the Dordogne), whose valley sides, shown in 

 plate 44, figure 1, present numerous bare ledges, while the gently rolling uplands 

 are cloaked with deep residual soils. The particular upland here shown is un- 

 usually smooth; water-worn pebbles and cobbles are plentiful at certain points 

 upon it, and it may therefore be considered to represent the floodplain of the 

 Ohavannoux in a former cycle of erosion, when the land stood much lower than 

 it now does. The adjoining uplands rise somewhat above this plain and are more 

 undulating than it is. They have deep soils, but no water-worn stones were found 

 on them. Monadnocks are numerous in certain districts, although absent here- 

 abouts. Some of them are rugged, with ledges and boulders on their slopes, almost 

 reminding one of a New Hampshire landscape, although no trace of glaciation has 

 been detected in their locality ; others are more subdued, presenting rounded out- 

 lines and smooth slopes, imitative of drumlins. Bevelling and stream spacing 

 entirely fail to account for this combination of forms, but the uplands and valleys 

 may be satisfactorily explained by assuming a former cycle of erosion well ad- 

 vanced into old age, followed by a new cycle, introduced by a broad uplift and 

 now in its youth. 



Plate 45, figure 1, is from Belle Isle, an isolated part of Brittany, today sharply 

 attacked by sea waves on all sides, so that its margin is cut by strong sea cliffs. 

 Its upland surface is somewhat dissected by valleys, one of which is here shown. 

 The striking, feature of this example is the sharpness of the shoulder between 

 the upland plain and the valley slope. The plain is broadly undulating in very 

 gentle relief, well covered with locally weathered soil; the valleys are manifestly 

 the product of a different cycle of erosion from that in which the uplands were 

 produced. Bare ledges of disordered crystalline schists are sometimes exposed on 



* The peneplain. Amer. Geologist, vol. xxiii, 1899, pp. 207-239. 



