326 



SCIENCE. 



[Vol. I., No. 12. 



error of result comes from an error of method 

 of but too prevalent a kind ; namely, the as- 

 sumption that things of a single geographic 

 name are to be accounted for hy a single 

 phj^sical or geological cause. Geographical 

 nomenclature is in no condition to allow such 

 an assumption ; for no science has so loose, 

 inaccurate, and insufficient a terminologj' as 

 geography'. Not a few examples could be 

 given of errors arising from this one-name, 

 one-cause idea. Until it is proved that two 

 phenomena are closelj^ alike in their several 

 characters, an explanation of the origin of 

 one will not necessarily' appl^- to the other ; 

 and for this reason, in our present ignorance of 

 the structure and form of many regions other- 

 wise comparatively well known, it is not safe 

 to extend local explanations over too broad a 

 field. 



Ldwl rejects the possibility of a river's hold- 

 ing its course across a rising mountain fold ; 

 because the several examples cUscussed in his 

 paper, chiefly those rivers on the northern 

 slope of the Alps which are temporarily' warped 

 into lakes, have failed in doing so (408, 409). 

 To this it might be answered, that these lakes 

 are perhaps formed by a local depression of 

 the vallej'-wav, rather than bj' a local uplift at 

 their outlets, and, moreover, that thej' consti- 

 tute such an ' ephemeral phase in tlie river's 

 historj' ' as hardly to constitute a serious argu- 

 ment toward a decision. The temporary forma- " 

 tion of a lake behind the growing fold, after- 

 ward drained by the victory of the river, is not 

 sufficient ground for excluding the vallej' from 

 the antecedent species, though it might serve 

 for the marking of a varietj'. But even 

 admitting the correctness of this conclusion 

 for the Swiss rivers, it proves nothing for the 

 rivers that escape from other mountain ranges. 

 The success of the river depends on the proper 

 relation of two variable factors, — the rate of 

 its erosion, and the rate of the mountain's 

 growth ; and these may have such diS'erent 

 relative values, — as determined hy rainfall, 

 drainage area, altitude, distance to the sea, 

 mountain-making force, composition and atti- 

 tude of the rocks, — • that the predetermination 

 of the result is impossible. Nothing short of 

 close local study will serve to answer the 

 question with any approach to certainty ; and 

 it therefore seems best to trust the Indian sur- 

 veyors in their explanation of the Sutlej ■" gorge, 

 and our own geologists in their reports on the 

 rivers they have examined in the western ter- 



1 Hardly recognizable in its modern Germanized form, Sat- 

 ladsc/u The German transliteration of the valuable English 

 consonant, j, is very cumbersome. Witness Udschidschi. 



ritories. Concentrated erosion can keep pace 

 with mountain folding, and antecedent vallej's 

 are often preserved. 



Reference is made to the several transverse 

 valleys of the Delaware, Potomac, and Sus- 

 quehanna in the Appalachians (407), with the 

 conclusion that the}' cannot be explained as 

 antecedent valleys.^ In spite of the many 

 observers devoted to the study of the Appala- 

 chians in the past fifty j-ears, there is j'et no 

 good topographic map of any large part of 

 them, and much remains to be done in ex- 

 plaining their geological structure. It is still 

 rather early to write their history ; but we do 

 not believe that the objections raised bj' Lowl 

 to the antecedent character of their larger 

 valleys are conclusive. The theorj' of these 

 vallej-s, so far as it can be now stated, should, 

 of course, be led bj' the facts so far as they are 

 now known ; and, in the writer's mind, the 

 facts lead directly to the theorj' that the val- 

 lej-s are antecedent. The question is made 

 clearer if we consider first the case of the 

 rivers in Tennessee and south-western Virginia 

 that rise in the archaean mountains of North 

 Carolina, — the Great Kanawha and the Ten- 

 nessee. The first of these follows the direction 

 of slope that must have prevailed through all 

 paleozoic time, in running from the old crj's- 

 talline mountains, north- westerlj', across the 

 strata derived from their waste. We must 

 conclude that the growth of the great post- 

 carboniferous folds and faults on its course 

 were insufficient to turn it into a north-east- 

 ward or south-westward channel. It flows 

 along a true antecedent vallej' ; and our notions 

 of the rates of mountain growth and river 

 erosion should conform to the fact of its exist- 

 ence. The Tennessee also finally makes its 

 waj' to the north-west ; but none of its branch- 

 es that rise in the North Carolina mountains 

 succeeded in crossing all the folds and faults 

 that grew in front of them. Although they 

 all made their way through some of these bar- 

 riers, the}' were sometimes turned to the south- 

 west ; and not until they were united in great 

 volume could they escape to the noi'th-west at 

 Chattanooga, and again at Claj'sville, Ala. 

 This shows a river greatlj' embarrassed by the 

 difficulties that arose in its way. Most of its 

 branches failed, and were turned aside into 

 consequent longitudinal valleys ; but some suc- 



1 Lowl does not detect a misquotation by Tictze, whose valu- 

 able Beinerkungen ilber die hildung von querthdlern {Jahrb. 

 qe.ol. reichsansl. 1878, 681-610) he endeavors to controvert, 

 'rietze states (600), that, according to Dana, the Appalachians 

 ^rew by addition of parallel folds on the eastern or seaward side. 

 Lijwl quotes Credner to prove the opposite order of growth, but 

 Dana also said iust the reverse. See Amer. journ. sc, iii. 1847, 

 183. 



