430 



SCIENCE 



[N. S. Vol. XXV. No. 637 



known locally as the 'grand chasm,' and so 

 joins the drainage system, which is cut so 

 deeply into the upper level. The rapid descent 

 into the chasm is made by a series of falls and 

 rapids, known collectively as the Tallulah 

 Falls. 



In connection with this peculiar feature of 

 the drainage, is a striking drainage pattern. 

 The Chattooga River flows almost due south- 

 west until it is joined by the Tallulah, and 

 then turns abruptly at a right angle and under 

 the names Tugaloo and Savannah Hi vers 

 flows almost due southeast into the Atlantic. 

 Near the point where this sharp bend or elbow 

 occurs, a stream on the upper level. Deep 

 Creek, takes its rise and continues the south- 

 west line begun by the Chattooga, until, under 

 the names Soque and Chattahoochee Rivers, 

 it reaches the gulf. 



It is this assemblage of features which has 

 suggested that the Chattooga River formerly 

 flowed southwest into the gulf drainage, being 

 continuous with and forming the upper part 

 of the Chattahoochee; that one of the head- 

 water branches of the Tugaloo-Savannah sys- 

 tem succeeded in eating headward into the 

 escarpment so far that it undermined the 

 valley of the Chattooga-Chattahoochee some 

 500 or 600 feet above, thus diverting the upper 

 part of that system into the Atlantic drainage, 

 and causing it to flow down to the lower level 

 and so on southeast to the ocean. As a result 

 of this capture falls and rapids would be es- 

 tablished where the descent from the higher, 

 older course down to the lower, new course 

 was made. These falls would be gradually 

 worn back, leaving a young valley or gorge 

 below them, in the vicinity of the sharp bend 

 or ' elbow of capture ' where the stream 

 changed from the southwest to the southeast 

 course. From the elbow of capture the re- 

 maining portion of the beheaded stream would 

 continue its southwest course, flowing in a 

 more mature valley on the upper level. 



The features in the Tallulah district are so 

 striking and so typical that it seems difficult 

 to account for them on any other basis than 

 the theory of capture. Certainly the condi- 

 tions are eminently favorable for capture; 

 and when one sees the streams on the lower 



level actively engaged in gnawing headward 

 into the steep escarpment and thus under- 

 mining the upper level, he feels a growing 

 conviction that the streams on the upper level 

 are in imminent danger of diversion. So 

 when he witnesses those features which must 

 necessarily follow capture, he is not at all 

 surprised, but adopts the theory of capture as 

 a matter of course. There are certain fea- 

 tures of the Tallulah district which have 

 caused some doubt as to the efficiency of river 

 capture, and which, therefore, deserve special 

 attention. It would appear that if the cap- 

 ture is so recent that the stream has not yet 

 had time to wear back the falls and rapids 

 produced by capture, then there should be falls 

 in the Chattooga River as well as in the Tal- 

 lulah. The former is no larger a stream than 

 the latter, and so far as volume goes, no 

 better fitted to grade its course. Recency of 

 capture, therefore, does not seem competent 

 to explain the falls in the Tallulah, when there 

 are no corresponding falls in the Chattooga. 

 Mr. Jones concluded that there was no differ- 

 ence in the resistance of the rocks over which 

 the two streams ran, and was, therefore, led to 

 doubt the fact of capture, and to suggest some 

 other alternative. Professor Davis believed 

 that the topographic features indicated cap- 

 ture, and in order to account for the falls in 

 the Tallulah and their absence in the Chat- 

 tooga, suggested that the former courses of 

 the rivers might have been such that the Chat- 

 tooga was captured first, and the Tallulah not 

 until a later period, thus giving more time for 

 the reducing of the falls in one case than in 

 the other. 



After careful investigation it appears that 

 the difficulty lies in the interpretation of the 

 character of the rocks over which the two 

 streams run. Instead of sameness of char- 

 acter, there is seen to be the most significant 

 difference in composition and ability to with- 

 stand erosion and weathering. The Chat- 

 tooga River flows over a mica schist, which 

 in all parts of the region is seen to offer little 

 opposition to stream erosion. Occasional 

 more quartzose bands have determined the 

 location of minor falls or rapids along the 

 smaller branch streams, but even these small 



