229 



further factor of uncertainty not present, to a like degree at 

 least, in the reservoir waters of the lake. 



That natural waters in this locality are not subject to the 

 presence of chlorine (sewage) in such excess or in such fluctu- 

 ating amounts appears on contrast of the chlorine curves of 

 Spoon River (PL XLVL and XLVII.) with those of the Illinois 

 (PI. XLIII.-XLV.). In the former, barring a few instances of 

 apparent contamination by river water (October, 1897, January 

 and February, 1899), the chlorine runs uniformly low through- 

 out the year, dropping but a trifle with rising floods. 



In Quiver Lake (PI. XLVIII. and XLIX.) the chlorine 

 similarly runs low during the period of individuality of the 

 lake, that is, of low water. The increase in chlorine comes 

 only at times of invasion of flood water from the river or the 

 bottom-lands above, as, for example, in November and Decem- 

 ber, 1898 (PI. XLIX.). The periods of fertilization of this lake 

 by sewage thus depend upon floods, and occur at times of 

 greatest dilution. 



In Thompson's Lake (PI. L. ) the chlorine (sewage) content 

 exhibits the same general tendencies found in the river, from 

 which its water supply is derived. The chlorine content runs 

 high during low water and drops with the rise of the flood. 

 The abrupt and numerous fluctuations of the chlorine of the 

 river do not, however, appear in the lake, being diminished and 

 equalized by its greater permanency. Even under these favor- 

 able conditions it is difficult to find any constant or well- 

 defined correlation between the chlorine pulses and those of the 

 plankton. It may be that the fertilizing elements of the sew- 

 age which the chlorine is regarded as representing have already 

 been exhausted, so that the chlorine curve no longer represents 

 a commensurate fluctuation in the fertility. In a few cases, as, 

 for example, in December, 1897, and in January and September, 

 1898 (PI. L.), a slight correlation in the chlorine and plankton 

 curves appears, though the only relation between the two may 

 lie in the effect of changing river levels upon both, a declining 



