258 IOWA ACADEMY OF SCIENCES. 



third millions to the c.c. If this were the principal cause 

 of fluctuation the effluent of November 14th should con- 

 tain about 41,000 bacteria to the c.c. other things being 

 equal. 



Along with the above the amount of organic matter 

 present would have a considerable influence, as it would 

 serve as food for the bacteria. Hence fission would be 

 more rapid and the number of bacteria to the c.c. 

 increased, but no data bearing on this point are at hand. 



The third factor would be the time of taking the 

 samples, whether at the beginning or toward the end of 

 the discharge. It is presumed that during the period that 

 the bed is resting the bacterial life increases; accumulat- 

 ing in the interstices between the material of which the 

 filter is composed. When the discharge comes on the 

 beds the pressure and hence the force being greater at 

 that time than at any other, also the number of the bac- 

 teria in the interstices being greatest then, might not the 

 force of the sewage wash these bacteria free and hence 

 through the bed into the effluent? If such be the case the 

 number of bacteria to a c.c. would be greatest at the 

 beginning of the discharge and least at the end. While I 

 have not been able to make experiments to fully elucidate 

 this point I feel quite confident from numerous observa- 

 tions in taking samples that such may be the case. 



Of course all of these factors and probably others acting 

 in unison complicate the problem to such an extent that 

 until more data is at hand it will be impossible to accu- 

 rately determine the exact amount of variation caused by 

 each factor. 



By referring to the tables containing the average num- 

 ber of germs per c.c. for each month, of manhole, tank, and 

 effluent, it will be observed that there is considerable fluct- 

 uation. It will also be noticed that the results for the 

 manhole, tank, and effluent decrease on the whole together. 

 The month containing the lowest average for the effluent 

 is August in 1900 as well as 1899. The largest average for 

 the effluent in 1900 is March after which there is a gradual 

 decrease until September. Several things must be taken 



