222 H. L. FAIRCHILD PLEISTOCENE MARINE SUBMERGENCE 



the terraces in the head section of the Connecticut Valley stimulated 

 Dana's critical interest in the study. He made surveys and reviewed 

 TJpham's data and published a second series of papers, with detailed de- 

 scription and refined discussion of the phenomena, in the Journal, vol- 

 umes 22-23, in 1881-1882. Dana confirmed Upham's measurements for 

 altitude, but disagreed as to the height of the glacial flood. Upham had 

 taken as the ^'normal" terraces, representing the summit of the flood 

 waters, the higher of the broad or more extended and continuous terraces, 

 and regarded the deltas or "delta terraces" at the mouths of the side 

 valleys and tributary streams as much above the river level. Dana 

 clearly showed that the high delta terraces were built in waters that must 

 have flooded the whole width of the valley, and that consequently they 

 are the best indication and measure of the full height of the "fl-ood" (vol- 

 ume 23, pages 94-96). 



Probably Upham realized the great difficulties involved in admitting 

 a river of sufficient depth and volume to include the high "tributary" 

 delta terraces. Dana was forced to accept the fact in spite of the conse- 

 quent difficulties, and then proceeded to wrestle with the problems. His 

 subsequent writings on the subject are mostly efforts to harmonize incon- 

 sistencies or ingenious explanations of difficulties that were unexplainable 

 under the river theory. 



Professor Dana found difficulty in accounting for the source of the 

 immense volume of water to produce and sustain a river of such great 

 size and steep gradient (volume 23, pages 367-372), the fine character 

 of the terrace material, chiefly clay and fine sand, being inconsistent with 

 a stream flow of such volume and velocity (pages 191-194) ; the necessity 

 of postulating dams in the Connecticut and adjacent valleys in order to 

 check the velocity of the flood (volume 25, pages 440-448) ; the source 

 of the enormous mass of detritus reqiiired to flll the valley to the depth 

 of 200 feet, and the lack of the coarse material that should complement 

 the fine, and particularly the disposition of the detritus derived from 

 excavating the width of the valley to the depth of 200 feet. 



The problem of the valley terraces does not appear to have been 

 seriously considered by Dana after 1883. In the Proceedings of the 

 American Association for the Advancement of Science, volume 32, page 

 198, speaking of the New Haven region, he says: "Twenty-five to thirty- 

 five feet is the greatest amount of submergence that the facts sustain." 



woAe of warren upram 



In 1878 Warren Upham published a description of the terraces in the 

 upper part of the valley, in !N'ew Hampshire and Vermont, in the third 



