135 



streams usually drop to a low stage and remain low through 

 the heated term, evaporation and absorption being so great as 

 to dispose of nearly all the rainfall. In the autumn, about the 

 autumnal equinox or a little later, heavy rains occur, which 

 cause the streams' to become swollen for a few days, or even 

 wrecks, but which seldom cause them to overflow their banks. 

 In some years these seasonal variations are slight, and the 

 streams show but little change in volume, but such years are 

 exceptional. The rainfall is seldom sufficient to cause freshets 

 to last more than a few days. The moderate and low stages 

 are estimated to generally cover ten months of the year, and 

 occasionally eleven months." 



In the Annual Reports of the Chief of Engineers of the U. S. 

 Army for the years 1890 and following. Captain Marshall has 

 published the readings of the river gages located at the govern- 

 ment dams in the Illinois River. The readings at Copperas 

 Creek, 16.8 miles above Havana, begin in 1879, and those at 

 LaGrange, 42.7 miles below, in 1883. The dam at Copperas 

 Creek was completed in 1877 and the one at LaGrange in 1889. 

 On Plate VII. will be found hydrographs plotted from the daily 

 readings at the gages below these dams from 1879 and 1883, re- 

 spectively, to 1900. The heavy sinuous lines represent the 

 fluctuations of the river, referred in the plot to the low-water 

 level of 1879. The mean annual curve was plotted from the 

 mean monthly readings at the above-named localities from 

 1879 and 1883, respectively, to 1900. Deficiencies in the records 

 at the two points named have been supplied, by estimate, from 

 records at nearest point of observation, in a few cases from our 

 Havana records. 



A comparison of the hydrographs of the gages at Copperas 

 Creek, Havana, and LaGrange reveals a close correspondence 

 in the fluctuations, and suggests that these main movements 

 are due to general rains or spring thaws coincident in the 

 greater part of the drainage basin. The differences are of a 

 minor character, and many of them are consequent upon local 

 conditions of rainfall. On closer inspection the records at Cop- 



