546 



the oxygen of the river water after the rains, followed presently by 

 a return to the original mean. 



This effect is graphically illustrated by Figure 2, Plate LXXX, 

 where two of the lines, representing the first and last trips, run 

 closely parallel from Marseilles onward, while the line, for the inter- 

 mediate trip diverges from the others to an increasing degree from 

 Marseilles to Chillicothe. These interesting facts may be taken as 

 suggestive of a general fouling of the water of the stream by heavy 

 midsummer rains coming after a long period of heat and of dry 

 weather. A general flooding and scouring of the surface of the 

 country, washing off of the streets of towns, and flushing out of 

 sewers, caused by heavy rains, may be supposed to bring into the 

 river water containing larger ratios of organic matter than the 

 stream itself ; and to these we must probably add the stirring up 

 and carrying off of the largely organic bottom sediments of the main 

 river. 



The carbon dioxide ratios in July and August, 191 1, varied from 

 none to 2.1 for the Kankakee River, and rose as high as 12.8 at the 

 mouth of the Des Plaines and ii.i at ^Morris and Marseilles. The 

 Illinois River ratios were lowest at Chillicothe, where, however, the 

 mean for the three sets of determinations, August 3, 17, and 29, was 

 6.3 parts per million. This is to be compared with a mean of .8 for 

 the Kankakee River for the same period. 



September, ipii. — During the second week in September, 191 t, 

 when the oxygen in the midstream at Alorris averaged .59 parts per 

 million and that of the Illinois just above the outlet of Depue Lake, 

 near Hennepin, stood at 2.65, waters from the middle of Depue Lake 

 itself contained 11.78 parts per million, equivalent to 42 per cent, 

 above the saturation point at the prevailing temperature of the lake 

 water. Even the Illinois and Michigan Canal at Morris, considerably 

 contaminated wnth Chicago sewage as it is, contained at this time 

 6.8 parts per million of oxygen when the w^aters of the Illinois River 

 gave 2.03 for the south or Kankakee side, and .88 for the north 

 bank, or Des Plaines water. 



November, ipii. — The effect of a higher temperature and higher 

 water levels w'as clearly shown in November, 191 1, by the much 

 higher oxygen ratio at all points on the upper Illinois. The river, 

 having lately fallen about three feet, was still, at this time, some 

 two feet higher than in the preceding July and August, and the 

 temperature of the water had declined from 72.5° F., the mean of 

 our midsummer readings, to 46°, the mean for November i to 8 

 (8:30 — 9 a.m.). The following is a table of parts per million of 

 oxygen at the points where comparable tests were made. 



