March ii, 1892.] 



SCIENCE. 



149 



same east-noith-east direction, and are clearly analogous with the 

 lower course of the Loup, where it connects its various brandies. 

 It seems not very improhable that the channel mentioned by Pro- 

 fessor Hicks as connecting the South Loup with Wood River may 

 be of the same sort. 



3. The hills north of this compound channel, as it might be 

 called, running parallel with the Platte, are of similar height and 

 structure to those south of the Platte, but the hills south of the 

 same channel are more than 100 feet lower, and of different struc- 

 ture. Both are capped with yellow loam of almost the same 

 texture, but underneath the former have a well defined stratum 

 of northex-n drift east of the meridian of Columbus, while the 

 latter have but a faint trace of it mixed with deep stratified sand. 

 These lower hills, moreover, are less eroded, and are evidently an 

 alluvian terrace formed since the deposition of the older drift and 

 the Loess. This terrace is seventy to ninety feet above the Platte, 

 east of Columbus, and is more sandy and lower further west. The 

 ancient north bank approaches the present Platte again, near 

 Josselyn. 



4. Corresponding in level to this high terrace, is an old channel 

 crossing Saunders County along the valley of Sand Creek and in 

 direct line with the upper course of the Maple. East of this is an 

 area of higher land between it and the Platte, which has been 

 recognized as an "ancient island." It may be added, also, that 

 this high terrace seems to be easily correlated with a terrace of 



iDrainage Map of EastecQ Nebfaaka. 



similar height and structure, found at several points along the 

 Missouri, which may be referred either to the " Second Glacial 

 Epoch " or to the time of the second cluster of moraines of that 

 epoch. 



The subjoined map exhibits most of the points mentioned above, 

 as well as some knowledge of the drainage, and indirectly of the 

 topography of the surrounding region. 



These facts point strongly to the efficiency of the second influ- 

 ence mentioned by Professor Hicks, viz., ''Pliocene channel fill- 

 ing,'' as the principal and sufficient reason for the peculiar arrange- 

 ment of the Loup channels, I'ather than a secondary influence. 

 This has been already pointed out by Professor Davis. The Loups 

 did formerly flow through to the Platte, but at a time when it or 

 a portion of it occupied the north channel already described, and 

 •when it was flowing x)n a level seventy-five to a hundred feet 

 higher, relatively, than at present, somewhat as it now occupies 

 the channel north of Grand Island, and probably not long ago 

 occupied a portion of Prairie Creek. The superabundant sedi- 

 ment, the shifting of the Platte to the south in obedience to Ferrel's 

 law, — possibly reinforced by a tipping to the south, — and a deep- 

 ening of its channel, which may have been partly due to a cutting 

 through of a divide north of the "ancient island '' into the lower 

 channel of the Elkhorn, yvliich, again, may have been accelerated 

 by the recent eastward tipping of the region, are sufficient causes 

 to explain the changes of the Platte, Loup, and associated streams 

 since the disappearance of the waters which deposited the loess. 

 The exceptional course of the Platte, however, from Kearney to 



Fremont, which we conceive was first taken about that time, re- 

 mains unexplained. The causes which may be surmised are the 

 following: 1. The position of a depression in the bottom of the 

 Pliocene or Pleistocene lake, which may in some way have been 

 produced by unequal deposition of its sediment, or tlie earlier un- 

 equal ero:.ion or deposition of the subjacent formations whose 

 strike here is approximately north-east. 2. A slight fold in the 

 plains a little south of this course of the Platte. Of such no dis- 

 tinct trace has yet been found. There is a slight anticlinal axis 

 crossing the Big Blue near Milford, but it is probably quite limited 

 in extent. 3. This course may perhaps be a survival of a time 

 when this region was tipped toward the north-east, because of the 

 burden of ice which then rested upon Iowa, Minnesota, and eastern 

 Dakota. This is but a conjecture, against which several objections 

 arise, which it is needless to express. 



In this connection, it may be helpful to call attention to a 

 similar bend in the Arkansas in central Kansas, and to note that 

 in each case the exceptional direction is upon more recent beds 

 near, and parallel to their junction with, the upper Carboniferous. 

 This may be a straw which would indicate that our first surmise 

 may have some truth in it. 



Concerning the efliciency of abstraction to change lines of super- 

 ficial drainage, we may find considerable light from the study of 

 this region. The remark of Professor Davis, that this rarely oc- 

 curs where formations are nearly horizontal, seems well supported. 

 Such is the slope of the country, and the porosity of the deposits, 

 that the headwaters of the Big Blue rise a little below the level of 

 the Platte adjacent, and the tributaries of Salt Creek rise below 

 the level of the Big Blue near by, so that it is possible that water 

 may leave the Platte between Kearney and Columbus, pass into 

 the Blue, be drawn off into Salt Creek, and return to the Platte 

 through the latter stream. And yet I know of no clear case of 

 change of channel by abstraction in the whole region. The 

 abundant sand, through the water flows underground, renders an 

 open channel unnecessary. In fact, it may be argued that abun- 

 dant sand tends to prevent the formation of superficial streams, 

 unless there be first a velocity of flow sufficient to carry the sand 

 easily, which cannot occur unless the How is concentrated in some 

 way. This is frequently noted in the sunken rivers of deserts. 

 Possibly this may have had something to do with the exceptional 

 course of the Platte before considered. Dunes form an important 

 part of the divide between the Platte and the Little Blue south of 

 Kearney. 



One word further, regarding the comparative slopes of the Loup 

 and Platte, to which Professor Hicks has called attention. Do 

 we not find here examples of the law that declivity vanes in- 

 versely as the quantity of water, as pointed out by Gilbert in his 

 masterly paper on "Land Sculpture," in his report on the Henry 

 Mountains? Although the Platte is much the more important 

 river, by the time it has reached Kearney it is much reduced by 

 evaporation and abstraction; then, because of its shutting off its 

 tributaries by its abundant sediment, as before noticed, it is so 

 reduced that it is often smaller than the Loup at their junction, 

 even sometimes ceasing to flow above the surface, as I have been 

 informed, while the Loup flows with a good current. On the 

 other hand, the Loup is not so much exposed to evaporation, and 

 has numerous tributaries, which having more frequently cut 

 through the sand stratum, and on the lower side of its sloping basin, 

 are more apt to be fed by springs than lose water by seepage. 



J. E. Todd. 



Tabor, Iowa, Feb. 29. 



Estimates of Distance. 



Besides the very interesting inferences drawn by IVIr. Bostwick 

 from his experiment {Science, Feb. 26, p. 118),_one or two others 

 should be suggested, in the hope that they may lead to some fur- 

 ther investigation. 



1. Is not an effect of fatigue shown in the eight or ten percent 

 by which the average observer's " mean deviation " from his own 

 "average" is increased when the last ten of bis ttiirty estimates 

 are compared with the first ten ? Should not this effect be great- 

 est, — perhaps both appearing earliest and increasing most rapidly 



