DETERMINATION OP BURIED TOPOGRAPHY 297 



valley of the San Juan river beyond the Boca San Carlos, and also some 

 distance up its tributaries. It is probable, however, that the land sank 

 very slowly, so that the estuaries were never deep, but were filled by al- 

 luvium almost as fast as formed. The depth of these old valleys havino- 

 been determined by borings at various points on the trunk stream and 

 some of its tributaries, it is possible to reconstruct the former surface and 

 determine approximately the depth of the alluvial filling in an}^ part of 

 the drainage system. It is found that the erosion of the hills has been 

 inconsiderable, since the submergence for the slopes above the margin of 

 the floodplains are practically the same as the old slope beneath the al- 

 luvial cover. The valleys of the lower San Juan and its tributaries have 

 been filled in such a manner that the present streams follow very nearly 

 the same course as the streams which formed the valleys. In some cases 

 their meanders have carried them to one side or the other of the old 

 valley, where they are now cutting against the bordering hills of residual 

 clay. The form of the old San Juan valley is shown on plate 31. 



The original form of the surface concealed by the floodplain of the 

 upper San Juan is much more difficult to make out. This plain was 

 formed by deposition in quiet water, the river valley being entirely 

 drowned. Hence the present channel was not determined by the deepest 

 portion of the old valley, but by tlie relative amounts of sediment brought 

 into this portion of the lake by tributaries on either side. It is evident 

 that the stream bearing the largest amount of sediment is the Rio Frio, 

 and the delta of this stream has pushed the outlet of the lake northward 

 away from the deeper portion of the old valley and against the hilJs 

 which formed its margin. The same thing is seen at various points be- 

 tween the lake and the Toro rapids. At numerous points the meanders 

 of the river carry it away from the deeper portions of the old valley and 

 against the marginal hills. In most cases these meanders are not acci- 

 dental, but are determined by the entrance of a tributary on the opposite 

 side. It is therefore impossible to determine the position of the stream 

 which formerly occupied this valley from the present position of the San 

 Juan. Sufficient boring has been done in this portion of the river chan- 

 nel, however, to determine the fact that the rock or residual clay slopes 

 of the hills which at present rise above the alluvial plain continue prac- 

 tically unchanged beneath the alluvium. The importance of this fact 

 in the location of the canal line is at once apparent. The line in general 

 follows the channel of the river, but if this were strictl}^ followed con- 

 siderable rock excavation would be necessary where the channel swings 

 against one of the marginal hills. It is evident, however, that by shift- 

 ing the line away from the hill the rocky slope will pass below the 



