PRIMARY PRODUCTIVITY 385 



Statistical Measurements 



The data from each sampHng date were treated with analysis 

 of vari£ince (ANOVA) to determine if there was a statistically signifi- 

 cant effect of temperature or treatment (Snedecor £ind Cochran, 

 1967). The data for the year were graphed as a regression of primary 

 productivity on temperature. The graphs of data from both stations 

 exhibited a curvilinear relationship. Two hnear components could be 

 analyzed; however, by separating the data at 25°C and performing a 

 least-squares hnear regression on each component. In the equation 

 for the regression hne (Y = o: + j3X), a is height, where X = X, and j3 

 is slope (Cohen and Cohen, 1975; Snedecor and Cochran, 1967). 



RESULTS 



Physicochemical Measurements 



The temperature range at Station 1 was 17 to 39°C, and the 

 average temperature for the year was 27°C. At Station 2 the range 

 was 10 to 34°C, and the average was 21.5°C. 



The difference in temperature (AT) between Stations 1 and 2 

 was greater in the winter months. The largest AT (8°C) occuiTed in 

 January, and the smallest (2°C) occurred three times, in October and 

 November 1976 and June 1977. 



The pH ranged from 6.5 to 7.0 at Station 1 and 6.4 to 7.0 at 

 Station 2 and was frequently the same at both stations. The seasonal 

 variation was shght. 



The transparency of the water decreased during the fall and 

 winter months and increased during the spring and summer, with the 

 maximum occurring in August and the minimum in January. The 

 range of the Secchi disk depth was 1.13 to 2.20 m. Although we 

 attempted to sample on clear, cloudless days, daily variation in solar 

 radiation was considerable, ranging from 530 to 120 langleys/day. 



Alkalinity, as CaCOs, varied from 89 to 120 mg/liter and was 

 generally the same at both stations. Alkalinity was caused by 

 bicarbonate since phenolphthalein alkahnity was never detected. 

 Seasonal variation was slight. 



Nitrate nitrogen ranged from undetectable to 0.45 mg/liter. 

 Orthophosphate was never detected by the method used. Dissolved 

 oxygen was always over 90% saturation and often over 100% 

 saturation at both stations. 



