42 W. UPHAM — giants' kettles eroded by MOULIN TORRENTS 



gravel, sand, and clay of the modified drift, which must have soon 

 choked up the passages wherever these drift-laden streams found cre- 

 vasses, causing them to flow in superficial channels walled and under- 

 lain by ice, until, near their mouths, the ice was melted through to the 

 ground and kames and eskers there received the coarser part of the 

 river's burden. 



In some places, however, we ma}'' better ascribe the moulin torrents 

 forming giants' kettles to the closing stage of glaciation, at the time of 

 final melting of that part of the ice sheet, because this appears to be 

 strongly indicated b}^ a general lack of drift deposits to fill the potholes, 

 as in the Interstate park, most particularly described in this paper. If 

 we can affirm for such tracts of the ice margin, while it was finally melt- 

 ing away, a nearly or quite stagnant condition, allowing a moulin to 

 remain without advance during a series of years, it seems to account for 

 these deep but mainl}^ empt}^ potholes more satisfactorily. Therefore I 

 am inclined to refer the giants' kettles of this park in the Saint Croix 

 valley to the latest recession of the ice-sheet from this area. 



The rock gorges of the Lower and U})per Dalles seem to me, as noted 

 in the preceding paper, to have been eroded by the river, even to a great 

 depth beneath its present level, during the long Buchanan interglacial 

 stage or epoch of the Ice age. At the same time a wide reach of the 

 valle}'', now partl}^ occupied by Thaxter lake, was eroded between these 

 narrow gorges, and a gradually descending ravine was channeled at the 

 east side of the Upper Dalles, slightly north of the angle. 



During the later and long envelopment of this area for a second time 

 by ice, drift may have been deposited in this valley, on an average, to 

 a nearly similar amount as on the adjoining higher ground — that is, to 

 depths of 10 to 20 feet, more or less — but its scantiness on the trap out- 

 crops, and the many empty potholes, suggest that there very little drift 

 had accumulated. When the ice was melted back so far as to leave the 

 valley open above the Upper Dalles, its next mile northward to the 

 Saint Croix rapids became filled, to the height of about 60 feet above 

 the present river, by a sand and gravel plain, now forming a terrace on 

 each side, and in part, especiall}'' on the east, by coarse boulder drift. 

 Upon these deposits the river ran during many years, smoothly wearing 

 the boulders of its bed, which, as the river cut down its central channel 

 to the present level, have fallen in great numbers from the upper part 

 of the river bank to the water's edge. 



Glacial Streams of Giants' Kettles and Eskers compared 



The rate of erosion of the giants' kettles, referred in most localities to 

 a stage of incij)ient glaciation, and the rate of formation of kame knolls 



