respect to nitrogen, oxygen, carbon dioxide, and pH. In 1929 the investiga- 

 tion was extended to cover all of the lake east of Western Lake Erie, but 

 determinations of nitrogen were discontinued. The results of this study- 

 have not been published. The Michigan Stream Control Commission has recently- 

 made a study of the oxygen content of Huron and Raisin Rivers, as well as of 

 Black River for some distance above its confluence with St. Clair River at 

 Port Huron, Michigan. The data have not bpm oublished, but have been made 

 available to the writer in the form of blue-printed charts. 



No attempt will be made to review here the results obtained in the chemi- 

 cal studies of Lake Erie and its tributaries', but in the body of this report 

 those which have a bearing on the problem will be discussed, and, in some 

 cases, compared with those obtained in the present investigation. 



I 



Scope and Methods 



This report is based on a large number of analyses of the water of 

 Western Lake Erie made in the years 1928, 1929, and 1930. Analyses for dis- 

 solved oxygen, free and fixed carbon dioxide, and hydrogen ion concentration 

 were made in all three years. In addition the program in 1930 was expanded 

 to include determinations of chlorine as chloride, and nitrogen as free and 

 albuminoid ammonia, nitrite, and nitrate. With the exception of a few 

 samples taken in February, 1930, the samples were taken vdthin the period 

 April to October, inclusive. Samples for chemical analyses were taken with a 

 Kemmerer-Foerst water bottle sindlar to the one described by Birge (1922). In 

 1928 the analyses were made on the boat, but in 1929 and 1930 this work was 

 done in the laboratory at Put-in-Bay. The few samples taken in February 1930 

 were analysed in a terrporary laboratory at Monroe, Michigan. 



In 1928 dissolved oxygen was determined by the Winkler method as modified 

 by Rideal and Stewart (American Public Health Association, 1925). In 1929 

 and 1930 the original Winkler method was used on water from the Island Section, 

 while the Rideal-Stewart modification was retained for sartples near the rivers. 

 Numerous parallel tests showed that the two methods gave concordant results in 

 the Island Section, but not in the polluted areas. All chemical values in 

 this paper are reported in parts per million (milligrams per liter), except 

 those for percentage of oxygen saturation and hydrogen ion concentration. 

 Percentage of saturg.tion was computed according to the table on page 62 of the 

 reference work note! above, which is generally known as "Standard Methods". 



Free carbon dioxide was determined by the Seyler method (Birge and Juday 

 1911) in 1928, 1929, and until June lU, I930. At that time the procedure 

 recommended in Standard Methods was adopted. In the tables of this paper, the 

 presence of free carbon dioxide is indicated by a plus sign, and a deficiency 

 of the gas by a minus sign. Figures following a minus sign show the amount 

 of carbon dioxide which would have to be added to change all of the calcium 

 carbonate to calcium bicarbonate and render the water neutral to phenolphtha- 

 lein. 



60 



