Semi-Ce7itennial Volume. 147 



placed in the containers with thoi-e for the water samples, and were back 

 for analysis within a week after shipment. Most of the analyses were 

 made from preserved material; however, one complete analysis was 

 made from each town on the living material, which was shipped into 

 the laboratory on ice, and allowed to grow in aquaria. 



In most cases the algae were present in sufficient quantities to ex- 

 amine direct from the samples. In those cases in which small masses 

 of the algae, "moss," could not be found in the basin or reservoir, a liter 

 of water from the basin was taken and the microorganisms present con- 

 centrated by the Sedgwick-Rafter filter method (Whipple, 26, page 28). 

 This method of determining the actual number of microorganisms per 

 cubic centimeter of water is especially valuable in recording the reduc- 

 tion of organisms per cubic centimeter during the copper sulphate treat- 

 ment. By this method it was often difficult to identify accurately the 

 organism under the 16 mm. objective. In order to overcome this difficulty 

 a few drops of the water containing the concentration of organisms was 

 mounted in the usual method under the 4 mm. or higher power. Most 

 of the species found by this method were diatoms and other unicellular 

 algae. 



At times samples of living algae came into the laboratory in a vegeta- 

 tive condition and the species could not be accurately determined with- 

 out the fruiting state. This was especially true of the Conjugatae. For 

 such algae Kelbs' method was used to induce the fruiting stage (Cham- 

 berlain, page 163). According to this method such forms as Spirogyra 

 in a vegetative state were placed in an aquarium of about .2 per cent 

 Knop's solution and left to grow for a few days. When a vigorous 

 growth was shown the algae were transplanted to a jar of distilled 

 water in the sunlight. In most cases these conditions caused conjugation 

 necessary for classification to species. 



In dosing the reservoirs with copper sulphate I have in most cases 

 used the Moore and Kellerman (14) method. 



