ORIGIN OF LAKE NICARAGUA 341 



it formed a barrier across the bay of Nicaragua. This barrier was built 

 gradually higher by successive eruptions, and since in the area behind 

 it precipitation was greater than evaporation, the waters rose above sea- 

 level and doubtless escaped westward over the barrier during periods of 

 quiescence in the volcanic activity. As the surface of the barrier was 

 raised by successive additions of volcanic ejecta, the sui'face of the im- 

 pounded waters was raised to a hight probably somewhat above the 

 present elevation of lake Nicaragua. The lake thus formed occupied 

 not only the position of the former bay, but flooded the basins of the 

 tributary streams. Its surface finally reached the lowest point in the 

 Continental divide, where a west-flowing stream headed against one 

 wliich occupied the present position of the San Juan. When this point 

 was reached the intermittent escape of the impounded waters across the 

 volcanic dam to the westward was changed for a permanent outlet to the 

 eastward. 



The gap when first discovered and overtopped by the rising waters 

 was doubtless of deeply weathered rock and residual clay. This must 

 have been very rapidly cut down by the escaping waters until the under- 

 lying hard rock was reached, when the permanent level of the lake was 

 established, which it has retained practically unchanged to the present 

 time. 



It is quite possible that the gaps through the Continental divide to the 

 east and through the divide across the west strip of land between the 

 former bay and the Pacific ocean were so near the same level that the 

 lake had for a short time an outlet both to the Atlantic and to the Pacific. 

 An examination of a portion of the Rio Grande gorge possibly throws 

 some light on this question. From the point where the Rio Grande turns 

 abruptly to the northwest in the reversed channel of the stream which 

 formerly flowed east, for a distance of 4 or 5 miles to the point where the 

 gorge opens out to the alluvial plain bordering the lower river, there is 

 an old channel which has been partially silted up by the present river. 

 The stream only occasionally touches the rock walls of the gorge on the 

 convex sides of its meanders. At the same time it nowhere departs 

 wholly from the old channel — that is, it nowhere has the character of a 

 superposed stream. It is evident that the present stream is smaller than 

 one which excavated and formerly occupied this valley. There are three 

 ways in which the present conditions might have been brought about : 

 1. The present valley might have been occupied by a stream which 

 was once larger than at present, but which has suffered a partial diversion 

 of its headwaters by capture through the encroachment of a neighboring 

 stream. This possible explanation, however, is not applicable in this 

 case, since the Rio Grande is itself a growing stream, and is constantly 



XLIX— Bull. Geol. Soc. Am., Vol. 10, 1898 



