PENEPLAINS OP FRANCE AND BRITTANY 483 



cial mechanisms, but it is entirely possible that geological processes may sometimes 

 have been complicated rather than simple. The probability of occurrence of one 

 process or another is the very question at issue, and an a priori measure of proba- 

 bility cannot be safely accepted as a guide to the solution of the problem. All that 

 can be done is to deduce certain consequences appropriate to the several theories, 

 and then to discover which of these consequences best match the observed facts. 

 Bevelled hilly regions (regions in which the hilltop height gradually decreases 

 from the interior toward the shoreline) are to my mind better explained by pene- 

 planation (or marine abrasion), uplift, and dissection than by bevelling. Maturely 

 dissected uplands with summits of subequal height (but not slanting to the seashore 

 like bevelled regions) may also be explained by peneplanation (or marine abrasion), 

 uplift, and dissection, as well as by stream spacing; but even-topped uplands, 

 bearing deep soils, extending their flat surface into protected districts aniong sub- 

 dued hills, dissected only by narrow, sharp-shouldered valleys, whose arrangement 

 exhibits an adjustment to the rock structures that underlie the deep soils, are 

 accounted for only by peneplanation. It is forms of this kind that have given • 

 effective support to the theory of peneplains, and it is particularly to forms of this 

 kind that attention should be directed when the theory is taken up for discussion. 



Remarks were made by N. S. Shaler, C. D. Walcott, H. M. Ami, and 

 G. K. Gilbert. 



AN EXCURSION TO THE COLORADO CANYON 

 BY W. M. DAVIS 



[Abstract] 



Observations made in the canyon of the Colorado and over the plateaus on the 

 north and south, during a three weeks' trip in June, 1900, add the occurrence of 

 certain landslides and migrating divides to the evidence already stated by Button 

 in favor of two cycles of erosion in the development of the Grand Canyon district, 

 the broad denudation of the plateaus being accomplished in the first cycle and the 

 incision of the narrow canyons in the second. The faults by which the plateaus 

 are divided are regarded as for the most part of greater antiquity than the canyon 

 cycle ; the antecedent origin of all the branch streams of the Colorado in this dis- 

 trict is questioned ; and the high-level floor of the Toroweap valley is explained 

 otherwise than by the failure of its former water supply through a change from a 

 humid to an arid climate. 



The paper w^as discussed by G. K. Gilbert and C. D. Walcott. The 

 full paper is printed in the Bulletin of the Museum of Comparative 

 Zoology at Harvard College, geological series, volume 5, number 4, pages 

 105-201, May, 1901. 



NOTE ON RIVER TERRACES IN NEW ENGLAND 

 BY W. M. DAVIS 



[Abstract'] 



When the title of this paper was sent in for the announcement that was issued 

 prior to the Albany meeting, it was thought that the controlling relation here 



