JrxE '28, 1895.] 



SCIENCE. 



■15 



than at present. The upland surface is l)e- 

 lieved to be an elevated peneplain. 



It is now moderately dissected by valU'vs 

 which along the larger rivers are from luo 

 to 200 feet deep. In comparison with the 

 width of the gently undulating interstreani 

 surfaces these valleys are not very wide. 

 The slopes are quite steep and locally form 

 bluffs, but towards the top they pass by a 

 graceful curve into the almost level upland. 

 The present Hood-plains along the bottoms 

 of the valleys are generally from an eightli 

 to a quarter of a mile in width. In terms 

 of development the present valleys are well 

 on towards maturity. The sharp narrow 

 valleys of extreme youth are entirely ab- 

 sent. The rivers have made considerable 

 progi-ess in the present cycle in reducing the 

 land mass to the level dependent on the 

 grade of their channels, l)ut the amount of 

 work still to be done is vastly in excess of 

 what has already been accomplished. 



The three topographic features mentioned, 

 namel}% the broad undulating upland, with 

 an elevation of from 850 to 1,000 feet: 

 the few monadnocks rising above it, and 

 the vallej's cut into it, give a clue to the 

 stages of geogi'aphic development of this 

 region. The upland peneplain is a surface of 

 denudation produced by long continued 

 erosion, when the land mass stood lower 

 than at pi-esent. This cycle of erosion 

 lasted a long time and the baseleveling was 

 almost completed. Very few monaduocks 

 rose above the general plain. The cycle 

 was ended by an uplift, which quickened 

 the streams, restored to them their cutting 

 powers, and compelled them to erode new 

 valleys in the old peneplain. They have 

 now cut down their channels until their 

 ability to transport material is just about 

 equal to the material which they have to 

 carry. Rivers, the profiles of whose stream- 

 channels are in this condition of equilibrium, 

 have been called by Davis (Science, X. S.. 

 Vol. I., p. 17(1) gi-aded rivers. The differ- 



ence in the slope of the valley sides aud the 

 upland plain indicates a change of level 

 before the excavation of the valleys and 

 after the formation of the upland plain. 

 The process by which the valleys are being 

 formed is not a direct continuation of the 

 process by which the gentle upland slopes 

 were fashioned. The valleys were cut in 

 the upland surface after it was elevated 

 from the low position which it had during 

 its foi-mation. 



Confirmatoi-y evidence for this hypoth- 

 esis is found in the winding courses of 

 the valleys which now dissect this upland. 

 Fever river was studied in the field, and 

 the topographical atlas of the Wisconsin 

 Geological Survey shows that the Platte, 

 Little Platte, Grant and Pecatonica rivers 

 have this same habit. If the geographical 

 development of this region was as outlined 

 above, the streams at the close of the 

 earlier cj-cle must have possessed wide, flat 

 valleys, with broad flood-plains, in which 

 they meandered freely. The elevation of 

 the land would have caused the streams to 

 degrade their channels rapidly. In many 

 cases the meanders on the flood-plain 

 would have been superimposed upon the 

 rock below, as the river bed was lowered. 

 The valleys cut in the elevated peneplain 

 would thus come to preserve, and, as 

 pointed out Ijy Winslow, also increase the 

 meandei-s of the earlier cycle. 



Such seems to have been the case with 

 the Fever river. Its meandei-s have an 

 average radius of a little less than half a 

 mile, but they are by no means constant. 

 Rock spurs of the upland project into each 

 curve. The slopes on these spurs are gen- 

 erally gentler than on the outside of the 

 curves, where the stream is often undercut- 

 ting the base of the slope and increasing 

 the meanders. Both open and close ox- 

 bows occur. The most marked of the close 

 type of mcan{lers was noted near Benton, 

 where the river makes an almost closed sig- 



