662 THE. JOURNAL OF GEOLOGY. 
meager. The facts, (1) that the pre-Pleistocene gravel exists in 
the form of widely separated erosion remnants south of the drift- 
covered country ; (2) that isolated remnants of it are known to 
exist at several points beneath the drift, as many as 125 miles 
north of the southern limit of glaciation (Adams and Hancock 
counties); (3) that the glacial drift here and there at various 
points for 90 miles further north (Rock Island county) contains 
gravel which might well have come from remnants of the north- 
ern extension of the same formation; (4) that remnants of 
similar gravel occur in the driftless area, where there has been 
no chance of destruction or burial by the ice; and (5) that the 
gravel in all these situations has the same topographical habit, 
all point to the conclusion that they are parts of a once wide- 
spread and continuous gravel formation. 
The gravel on the east bluff of the Devil’s Lake is about 60 
miles east of the gravel of Crawford county. It is about 150 
miles northeast of that part of Rock Island county where similar 
gravel is a local constituent of the glacial drift. The topographic 
situation of the gravel is the same as that of all the widely dis- 
tributed remnants further west and south, to which reference has 
already been made. In its constitution and other physical 
characteristics, the Devil’s Lake bluff gravel is so similar to that 
of the region farther south (Adams county, Illinois, etc.), that it 
could hardly fail to recall the southern formation to one who 
had seen it in southern Illinois, Missouri and Arkansas. The 
geographic and topographic position of the Devil’s Lake gravel, 
as well as its constitution and physical condition, make it 
altogether rational to infer that it may have been connected with 
the isolated beds of similar material already referred to, and that, 
as originally developed, the formation of which the existing beds 
are but remnants, had a much greater extension than has hereto- 
fore been recognized. If the gravel on the bluff east of Devil's 
Lake be a remnant of an extensive preglacial gravel formation, 
it becomes a matter of much importance to fix its age. 
Limited occurrences of somewhat similar gravel or ferru- 
ginous conglomerate are known in southeastern Minnesota, and 
