438 PHYSICAL GEOGBAPHY OF THE COUNTKY. Book III. 
terranean water-courses, many of which never find 
an outlet to the surface. Sudden as well as gradual 
changes appear to take place in these water- courses, the 
water in one spot removing an obstruction, and in another 
creating some new one. Mr. P., an engineer of San 
Antonio, told me the following fact: — The Rio Yerde, 
a river now entirely dependent on the atmosphere for its 
supply of water, and which falls into the Rio Hondo, 
was formerly a permanent, clear, and broad stream, 
several feet deep ; people settled on its bank ; when, 
it all at once disappeared, nor did it appear again. 
A deep well, dug near it, at first yielded some water, 
but afterwards became dry. Suddenly, one night, a 
regular torrent broke forth from it with a tremendous 
noise, but it flowed only for a few hours ; in the morning 
the well was again empty. At noon the torrent gushed 
forth anew. This continued for several hours, since which 
time it supplies very little permanent water. Of course I 
cannot vouch for the accuracy of these statements. 
From numerous intelligent friends at San Antonio I 
learned many interesting facts relative to the nature and 
features of this country, with which — being differently en- 
gaged — I should not otherwise have become acquainted. 
One of my friends had been recently to the Atascoso, a 
small river about fifty miles S.S.W. of San Antonio, and 
had made some observations which complete my former 
remarks on the changes of the vegetation and climate in 
Texas. The Atascoso flows into the Rio Trio, a tributary 
of the Nueces. The country there is — or was at that 
time — quite wild and uncultivated, mostly covered with 
wood, consisting of scattered very old trees and a thick 
growth of young ones. The soil is formed of an upper 
stratum of sand, beneath which is a layer of black humus. 
