value was 97. p. p. m. , and in September 95. 2 

 p.p. m. There was a fairly consistent trend toward 

 decreasing alkalinity at both the surface and the bot- 

 tom with advance of the season from June to Septem- 

 ber. There was also some decrease through the sum- 

 mer of 1928. 



This phenomenon appears logical when it is re- 

 membered that carbon dioxide is being removed from 

 the lake to such an extent that some of the CaC03. 

 H2CO3 is being turned to the relatively insoluble cal- 

 cium carbonate. The normal carbonate probably 

 precipitates out very slowly and sinks to the bottom, 

 thus removing from the lake some considerable 

 amount of alkaline matter. Probably a large amount 

 of this is recovered when the carbon dioxide increases 

 again in the winter and the lake overturns in the 

 spring. 



Hydrogen -ion Concentration 



The water became slightly more alkaline as the 

 season progressed and carbon dioxide was gradually 

 removed from the lake. The mean surface pH in 

 June was 8. 18 and in September 8. 22. At the bot- 

 tom the pH was on the average somewhat lower than 

 at the surface. In June the mean bottom pH stood at 

 7. 98 but by September it had risen slightly to 8. 06. 



The general correlation between carbon dioxide 

 content and pH is rather good (figs. 31-34). When 

 the carbon dioxide curve dips the pH rises, and vice 

 versa. Also there is a relation between fluctuations 

 in oxygen and pH, as a corollary. The lowest pH 

 value was 7. 6 and the highest was 8.4. These ex- 

 tremes were practically always accompanied by 

 correspondingly great inverse fluctuations in the 

 amount of carbon dioxide. 



Toward September when vertical temperature 

 and carbon dioxide gradients were decreasing in the 

 shallower regions of the lake, the pH curves for the 

 surface and bottom also showed a gradual tendency 

 to approach each other. The graph for September, 

 in figure 34, shows very little difference in the pH 

 of surface and bottom determinations except in the 

 region of the Deep Hole. There, 2 p. p. m. of &ee 

 carbon dioxide at the bottom caused a pH of 7. 6 

 in contrast to the surface with negative carbon di- 

 oxide and a pH of 8. 3. 



agree with those of 1929 as to range, average, and 

 general distribution. 



Chloride 



Chlorides are washed from the rocks of the 

 drainage area into the lake waters and are hence a 

 natural constituent among other solutes. The chlo- 

 ride content of Lake Erie ran somewhat higher than 

 the figures given by Birge and Juday (1911) for the 

 Wisconsin lakes. These workers report an average 

 of about 5 p. p. m. and a maximum of 10 p. p. m. of 

 chloride, while the analyses made on Lake Erie in 

 the summer of 1929 show a range from 8 to 16 p.p. m. 

 with a surface mean of 11. 93. Very little vertical 

 difference in the chloride content was noted through- 

 out the lake (figs. 31-34). 



One of the conspicuous features of the hori - 

 zontal distribution of chloride was that at certain 

 points along the lake shore relatively high values 

 occasionally occurred (fig. 35). The diameters of 

 the circles signify the parts per million of chloride 

 deviation from the arithmetic mean of all the deter- 

 minations for the surface and for the bottom in Sep- 

 tember. At most of the stations only very small de- 

 viations occurred. Conspicuously high values were 

 found near Port Colborne, off the mouth of Eighteen 

 Mile Creek, near Erie, Fairport, and Cleveland. As 

 pointed out in the report on "Pollution, " these high 

 chloride figures are probably due to indusUial wastes 

 from manufacturing centers. 



Only upon one occasion was a high chloride 

 value found near the Buffalo water intake. In June 

 the analyses showed low oxygen and high chloride 

 at station 01 near Buffalo. It is possible that at 

 times certain amounts of waste materials may find 

 their way out into the lake opposite the Buffalo 

 Harbor. 



At the extreme western end of the lake, a puz- 

 zling low chloride area may be seen in figure 35 and 

 figures 31 to 34. The values are consistently low in 

 the line of stations running across from Lorain, Ohio, 

 to Point Pelee. Just what is the significance of this 

 condition is difficult to say without knowing the state 

 of affairs further to the west. 



The pH observations in the summer of 1928 



85 



