Turbidity 



The determinations of turbidity indicate rela- 

 tively clear water at the surface of nearly all the sta- 

 tions throughout the lake. In the vicinity of the 

 Deep Hole the bottom water was always clouded and 

 showed the highest figures. The whole range of tur- 

 bidity was from to 100 p. p. m. of SiOg at the bot- 

 tom and to 25 p. p. m. at the surface. 



In June the Buffalo station gave high turbidity 

 figures at both the surface and bottom. 



Figure 36 shows the distribution of turbidity for 

 September. Nowhere else in the lake does the tur- 

 bidity approximate the values shown for the bottom 

 waters of the Deep Hole. In July the bottom waters 

 were even more clouded than in September. The 

 surface samples were quite clear; in fact, no tur- 

 bidity was noted at a large number of the stations 

 east from Long Point. It is suggested that bottom 

 currents may have kept the dense waters of the Deep 

 Hole in a constant state of agitation with the result 

 that the fine bottom mud was dissipated into the over- 

 lying water mass. 



to -2. 53 p. p. m. (table 12). At the bottom there 

 was positive free carbon dioxide in June with a 

 change to negative free carbon dioxide in July. Con- 

 siderable increase in the negative value occurred in 

 September. The probable errors of the mean carbon 

 dioxide values are correspondingly greater than was 

 the case for oxygen saturation. In other words, there 

 was a greater dispersion &om the average of the car- 

 bon dioxide determinations throughout the lake. This 

 dispersion was relatively greater at the bottom in 

 July and August, when there were almost as many 

 negative as positive carbon dioxide quantities. 



Relatively great changes in carbon dioxide pro- 

 duce small pH alterations, and hence only small 

 fluctuations were observed in the mean surface and 

 bottom pH values. Slight increases are nevertheless 

 apparent with advance of the season. The surface 

 means are slightly, but consistently more alkaline 

 than the bottom. The largest probable deviations 

 appear at the bottom toward autumn when there oc- 

 cuned a marked separation of the bottom waters into 

 two areas, J., e. , with and without free carbon dioxide 

 which correspondingly affected the hydrogen -ion con- 

 centration. 



GENERAL DISCUSSION 



In the interpretation of a set of random sam- 

 plings, such as that from the lake with which we are 

 dealing, it is often very useful to apply certain 

 standard statistical methods. In making the compu- 

 tations, aid was obtained &om the standard treatices 

 of Merriman (1901) and Chaddock (1925). The 

 short formulae for the probable errors (r and r^) were 

 taken from the Smithsonian Physical Tables (Fowle 

 1927). 



In figure 37 are shown the mean values and 

 probable deviations of the mean for oxygen satur - 

 ation, carbon dioxide, and pH. At a glance we can 

 see the general trend of conditions. 



As concerns oxygen saturation the surface more 

 nearly approached complete saturation than did the 

 bottom in the 4-month period from June to Septem- 

 ber. Also the probable deviations are relatively 

 small. The values are closely grouped about the 

 central tendency. 



The mean negative carbon dioxide values at the 

 surface throughout the season ran from -1. 76 p.p. m. 



There was a rather constant mean alkalinity con- 

 tent all through the summer, and probable errors are 

 small (fig. 38). In general the bottom values were 

 slightly higher than those for the surface. Toward 

 autumn there was a slight reduction in the mean al- 

 kalinity along with the general decrease in free car- 

 bon dioxide. 



Through the summer a decrease in the chloride 

 content of the lake was witnessed at both the surface 

 and the bottom. Though there was very little ver- 

 tical difference of the averages, the bottom values 

 for each month showed a slight margin above tliose 

 at the surface. The probable errors are so small 

 that it was not possible to show them on the par- 

 ticular scale used in plotting. The chloride values 

 in the lake, hence, may be said to fluctuate rela- 

 tively little from the mean. 



The striking difference between the values at 

 the surface and bottom is shown in the graph for 

 turbidity. In all months the bottom water shows a 

 far greater mean cloudiness. This vertical dif- 

 ference seems to have become most pronounced in 

 July when the mean turbidity at the bottom was at 

 a maximum of 13. 2 p. p. m. Si02. At tliis time 



91 



