oxygen observed in the three seasons of investigation. One sample in 1928 

 shewed only 62 percent of saturation, and in 1930 a few sarr^^les showed less 

 than 80 per cent, but in most cases the bottom water was nearly as well sup- 

 plied with oxygen as the surface water. 



Since free carbon dioxide and pH vary clossly with the oxygen, it will 

 not be necessary to discuss them in detail. Free carbon dioxide in the sur- 

 face water ranged from -5.9 to +3.1, and in the bottom water from -3.0 to 

 +7.3. The last value (+7.3) vjas associated with the very low oxygen at Sta- 

 tion $9k, and was unusually high. Values in excess of +3.0 were rare. The 

 pH of the surface water ranged from 7.7 to 8.5, and of the bottom water from 

 7.3 to 6.3. 



The data give every indication of being normal, that is, they are essen- 

 tially the same as those from some shallow inland lakes which are not polluted. 

 Lake Wingra, Wisconsin is a lake of this type which has been studied in detail 

 (Tressler and Domogalla, 1931). In Hay of both years reported, the water was 

 supersaturated with oxygen, but during the remainirig months of the period 

 April-October it was always less than saturated, with a minimum of 75 per cent. 

 Lake Wingra showed a consistent deficiency of free carbon dioxide during the 

 period in question, and in some cases reached -10 parts per million. Western 

 Lake Erie never became so deficient in carbon dioxide, and commonly contained 

 a small excess, particularly in 1930. The pH of Lake Wingra was somewhat 

 higher than that of Lake Erie in general, but the maximum (pH 8.7) was only 

 slightly higher. On the whole the two lakes agree very closely with repsect 

 to oxygen, free carbon dioxide, and pH - as closely as one would expect in 

 bodies of water so far apart. 



It is commonly believed that the surface water of an unpolluted lake 

 must be saturated or verj*" nearly saturated with oxygen. This idea is unten- 

 able. Juday and Birge (1932) reported that half of a large number of sur- 

 face samples from lakes in northeastern Wisconsin fell between 83 and 93 per 

 cent of saturation, and more samples fell in the 86-67 per cent group than 

 in any oth?r group of like interval. Many surface samples from lakes free 

 from a noticeable amount of humic substances were well below the saturation 

 point, even when the influence of elevation of the lakes on the saturation 

 point is taken into consideration. It is not uncommon to find marked reduc- 

 tions in oxygen content of surface waters as a result of the death and decom- 

 position of plankton organisms (see Whipple, 1927, page 209-210). The point 

 to be made in this connection is that the fairly consistent lack of complete 

 saturation in the Island Section was not necessarily the result of pollution. 



Thus far no mention has been made of methyl orange alkalinity. In the 

 Island Section this ranged from 85 to 103 parts per mi Hi en of calcium car- 

 bonate. According to the classification of Birge and Juday (1911> page 76), 

 Lake Erie may be regarded as a lake with medium hard water, approaching the 

 lower limit of the hard water lakes. In general the methyl orange alkalinity 

 , was higher in summer than in- spring or autumn, but there were some notable 

 exceptions to that rule. As would be expected, the vertical distribution was 

 uniform or nearly so in most cases, and when there were differences they 

 seemed to be fortuitous. 



79 



