GEOLOGICAL SURVEY OF THE UNITED STATES. 



407 



source is to be found by tracing backward lines 

 of drainage. The materials of the drift occur 

 at various and often great distances from the 

 outcrops of the parent beds, and their lines of 

 transportation, which often ascend, have no 

 established relation to lines of drainage. The 

 major part of the drift has been transferred 

 from the north southward, and some of it has 

 been carried from one side to the other of the 

 basin of the Great Lakes. Another peculiar- 

 ity of the drift is found in the configuration of 

 its surface. Over some broad areas the sur- 

 face is gently undulatory, or even plain, but 

 there are other regions where it is exceedingly 

 irregular, comprising short, uneven ridges, and 

 steep- sided, roundish hills, so disposed as to 

 include cup-shaped depressions known as ket- 

 tle-holes. Many of these depressions contain 

 ponds, and these conspire with an irregular ar- 

 rangement of the streams to give the hydrog- 

 raphy a peculiar and striking character. The 

 drift-covered area is further characterized by 

 a distinctive grooving, scratching, and polishing 

 of its bed-rock, and the trend of the scratches 

 corresponds in a general way with the direc- 

 tion of transportation of the drift. 



For a long time these phenomena remained 

 without explanation ; but it was finally point- 

 ed out by Louis Agassiz that the bed of a gla- 

 cier is grooved and polished after the manner 

 of our drift area, and that the debris discharged 

 by a glacier at its end or deposited from its 

 lower surface has many of the characters of 

 the drift. In the year 1850 he broached the 

 theory that a great glacier, or sheet of ice, 

 thousands of miles in extent, had invaded the 

 country from the north, plowing up and grind- 

 ing the face of the country, and spreading its 

 debris in the form of moraines. His view was 

 widely adopted, but there were some who 

 thought that a better explanation might be 

 found by assuming that glaciers were always 

 confined to the more northerly regions, and 

 that icebergs derived from them not only 

 transported the drift but polished the rock- 

 surfaces, then submerged, by stranding as they 

 floated south ward. So wide-spread and varied 

 are the phenomena that the opinion of each 

 individual observer depended very largely on 

 his locality; and the two rival explanations, 

 known as the " glacier theory " and the " ice- 

 berg theory," were argued and discussed for 

 more than thirty years, without a satisfactory 

 conclusion. The advocates of the glacier the- 

 ory maintained and increased their numerical 

 preponderance, but as late as 1882 Prof. J. D. 

 Whitney wrote in opposition to all their more 

 important conclusions, and in 1883 Dr. J. W. 

 Dawson, in his presidential address to the 

 American Association for the Advancement of 

 Science, reiterated his opinion that " the the- 

 ory of a continental glacier must be given up." 

 During the period of this protracted discussion 

 investigation was not idle ; every year saw im- 

 portant additions to the knowledge and liter- 

 ature of the subject ; but the system of State 



surveys gave to no individual or corps the 

 opportunity for that continuous examination 

 of the whole field which was necessary for 

 the establishment of a satisfactory general 

 theory. When the work of the national sur- 

 vey was extended to the eastern portion of 

 the United States, the first geologic work in- 

 stituted was the comprehensive investigation 

 of the drift. With the aid of one or two per- 

 manent assistants Prof. Chamberlin undertook 

 to collate the abundant literature of the sub- 

 ject, and, by revisiting numerous localities 

 that had been the subject of description, to 

 interpret and utilize a body of apparently in- 

 harmonious observations. He has traced cer- 

 tain special phenomena and groups of phe- 

 nomena through the whole breadth of the 

 Northern States, and he has organized into 

 a system the local work of numerous observ- 

 ers heretofore independent. By this organi- 

 zation he has availed himself of the local in- 

 formation in the possession of the individual, 

 and has directed attention to specific questions 

 whose answers contribute to the solution of 

 certain definite problems which his compre- 

 hensive view has enabled him to formulate. 

 The investigation is by no means complete, 

 and indeed some of the broader questions in- 

 volved can not be fully answered by the phe- 

 nomena within the boundaries of the United 

 States ; but important progress has been made. 

 As the conclusions to which he has attained 

 are based not only on the work recently per- 

 formed under his direction, but on the ante- 

 cedent observations and discussions of himself 

 and others, it is impossible, in a brief state- 

 ment of results, fully to discriminate those 

 which depend solely upon the new organ- 

 ization. The following paragraphs attempt 

 merely to summarize the present condition of 

 the subject, or rather of those parts of it to 

 which the corps has given most attention. 



It is convenient to distinguish drift that 

 has a very irregular surface, characterized by 

 knobs and kettle-holes, from that with a more 

 even surface, and we will make use of the 

 terms Jcno'b-and-'basin drift and undulating 

 drift. The undulating drift has the greater 

 area. The knob-and-basin drift is arranged in 

 belts or lines, and these lines are measurably 

 independent of the configuration of the coun- 

 try they traverse, ascending and descending 

 its slopes. When they are traced out and 

 mapped, it becomes evident that they consti- 

 tute a system of loops, the convexities of 

 which are turned southward. Within each of 

 these loops there is a peculiar fan-like arrange- 

 ment of the glacial stria3. Along the center or 

 axis of the loop they are parallel to its general 

 direction, but on either side they curve away 

 from the axis, and in immediate juxtaposition 

 to the limiting line they are nearly or quite 

 normal to that line. Moreover, if each loop 

 and its associated striso are taken as a unit, it 

 is found that each such unit is associated with 

 a lake-basin or other topographic depression 



