22 
It 
June 9, 1891, a mountain reservoir broke and discharged its waters 
through the canon, with the effect of carrying away the instrument 
house, and slightly altering the cross section of the river, so that 
more water is now carried for the same depth of water. 
A self-recording instrument has been at the gauging station for 
most of the time, which kept record of the rise and fall of the river. 
Until June, 1891, the instrument was one furnished by the office of 
the State Engineer. After the flood already mentioned, which re¬ 
sulted in the destruction of the instrument there, one of the Richards 
instruments was put in place by this section, and from this the read¬ 
ings for 1892 were made until the river became too low to move the 
float. The method of reduction, which w^as done by Mr. Trimble, 
is shown by the table, page 31. 
The general character of the stream, which is typical of the 
streams flowing down the eastern slope of the Rockies, is shown by 
Fig. 9, which shows the flow of the river during the years 1891 and 
1892. With no living tributaries from the plains, these rivers are 
fed by the melting snows of the mountains. Low in winter and 
early spring, they begin to rise as the snows melt, or with the 
spring rains, and increase with increase of heaf, or until there is a 
lessening in the snow fields, and then decrease until the low stage is 
reached in the fall and remain low until the rise in the following 
spring. An understanding of the character of the streams is neces¬ 
sary to understand some of the peculiarities of Colorado practice. 
The river is usually high in June, sometimes earlier, depending upon 
the character of the season, and whether a large part of the snow 
is on the lower or the higher mountains. 
The character of the flow of the stream affects both the char¬ 
acter of the crops raised and the nominal duty of the valley as a 
whole. With an abundance of water in June, and little in August, 
