PABIBAULT COUNTY. 453 



Moraines.] 



to be represented by a hilly and rolling tract in the southwest part of 

 Sterling, in Blue Earth county. The first opinion of the writer, stated in 

 the ninth annual report, that this morainic belt was formed wholly as a 

 medial moraine by converging ice-currents, seems questionable. Further 

 exploration is needed to determine whether it is not instead a terminal 

 moraine, accumulated on the southwest side of this ice-lobe, after three 

 distinct times of recession from its outermost limit. This explanation is 

 strongly confirmed by comparison with the three similar morainic belts 

 beyond this toward the south and southwest, all of which are apparently 

 terminal, as shown in the report of Watonwan and Martin counties.* 



The most hilly portions of Kiester are its south side for a width of one mile, and a belt 

 through its northeast part from section 13 to sections 3 and 4, in which are the most prominent 

 of these hills, visible lifteen miles to the north and west. Their hight is from 100 to 200 feet 

 above the lowland in these directions and above Bear lake in Freeborn county; the highest points, 

 which are in the S. W. j of section 3, being about 1400 feet above the sea. These are massive 

 hills of till, of irregular outlines, but trending somewhat more from east to west than in other 

 directions. Between the hill-ranges of the north and south parts of this township, its central 

 portion for a width of two or three miles is only moderately undulating till, reaching east at the 

 head of Brush creek to the west border of the plain of modified drift in Mansfield, Freeborn 

 county. In sections 8, 17, 20 and 29, through the west part of Kiester, a series of hills of till. 60 

 to 75 feet high, connects the west ends of these ranges and forms the west border of the lower 

 tract between them, except at the gap through which Brush creek flows. 



In Foster, the township next north of Kiester, boldly rolling hills of till flf ty to seventy-five 

 feet high extend from section 28 to the north and northwest by Rice lake, where they occupy a 

 width from one-half mile to one mile on each side of the lake. Still farther northwest the same 

 contour and material border the east, north and west sides of Walnut lake, including the most 

 of sections 25 to 28, and 33 to 36, of Walnut Lake township. The land south of Walnut lake is 

 low and gently undulating till, with frequent marshes. In Barber, the township next west, a 

 prominently rolling tract is found about the little lakes in sections 14, 15, 22 and23. The mate- 

 rial here is till, and its swells or hills are thirty to fifty feet above the hollows. Through six miles 

 thence northwest a more or less rolling surface of the unmodified glacial drift continues in a belt 

 about two miles wide, to the southwest part of Lura and the east edge of Delavan. On the rail- 

 road it is crossed in the first three or four miles east of Delavan, where its swells are twenty-five 

 to forty feet high, not crowded and thickly set, but generally in long slopes, with no prevailing 

 trend. This morainic belt divides two extensive areas of till, which are characterized by a very 

 smooth and flat surface. 



In the south edge of Elmore and Pilot Grove a width from one to one 

 and a half miles is hilly or prominently rolling drift, and forms part of a 



*Iii this connection it is important to note that Prof. N. H. Winchell in 1871 and 1872 observed four terminal mo- 



Blanehard ridges (Proceedings of the Am. Assoc. for Adv. of Science, vol. xxi, H72. pp. 160177; also, Report of the geoloyicnl 

 mavey at Ohio, vol. ii, IS74) Again. Prof. T. (J. Chamberlln observed three distinct morainic belts belonging to this 

 epoch, divided bv smoother tracts, in a section between Black Brook (T. 33, K. 16) and St. Croix Falls, at the west side 

 of Wisconsin (Gcnli/ay of Wisconsin, vol iii, 1880, pp. 381 and 385). 



If this be a fourth terminal moraine, its continuation northwestward is probably traceable to the vicinity of Big 

 Stone lake. My observations of the area across which it would lie. make it certain that no very prominent accumula- 

 tions of morain c drift occur there; but suggest that this formation should be searched for in a course extending by 

 Madelin. near lake Hanska, Meepy Eye creek, and the northwest corner of Redwood county, to the southwest part ot 

 Tyro in Yellow Medicine county, nnd thence to the eastern morainic belt in township ll'J, range 4, Lac qui Parle 

 county. The glacial lake before mentioned would extend along this ice-border, through Watonwan, Brown and Ked- 

 wood counties, covering an area several mileswide in the depression between the ice-sheet and the Coteau des Prairies. 

 Its first interruption by land higher than 1 15" feet above the sea would be in Yellow Medicine county, where a fourth 

 morainic belt wa observed, with a great water-course of some former time at its west side. 



