Erosion in the Basin of the Arkansas River 337 



fore, and it is hoped that no other apology is necessary for their 

 publication. But whether the work had been thoroughly 

 comprehensive or not, it is evident from the behavior of this 

 large stream, fed from such a large and geologically diversified 

 hydrographic basin, that slight and even local changes of 

 meteorologic conditions may greatly change the results ob- 

 tained, or those that would have been obtained, had the ob- 

 servations been made daily instead of occasionally. The total 

 here given for the year 1887-8 may be twice as large or but 

 half as large as that for the next succeeding year even with the 

 same or nearly the same discharge of water. For this reason no 

 estimate of results for a longer period based upon these ob- 

 servations or estimate for any other period of time can be more 

 than approximately correct, because the relations of the 

 amount of matter carried either in solution or suspension to 

 the volume of water are not constant. A perfectly satisfactory 

 measure of the actual work done by such a stream can only 

 be determined by a series of observations covering a number 

 of years, and made in connection with careful meteorologic 

 observations during the same period over the entire hydro- 

 graphic basin. It is also evident that deductions derived 

 from observations upon the Arkansas River are not applicable 

 to the study of other streams except in a very general way. 



measurements, although that stream can scarcely be regarded as a trib- 

 utary of the Arkansas. Their results must have been considerably 

 modified by the presence of so large a body of comparatively clear 

 water. 



Palo Alto, California, 

 Aug., 1893. 



