324 MARSHALL AND MELLINGER 



cadmium, but S' was by far the more sensitive index, especially in 

 the range of to 5 fig/liter. The mean value of S' at 5 jUg/liter is 

 reduced to 73% of that for the controls, and the difference is highly 

 significant (P < 0.001); whereas the mean value of S at 5 jUg/liter is 

 not significantly different. 



The effects of enclosure, in contrast to the effects of cadmium, 

 were mostly insignificant. This is shown by the results of t-tests of 

 the significance of differences between mean numbers in the 

 incubated controls and those in unincubated samples taken either at 

 the end of experiments 1, 3, and 5 (48 tests) or at the beginning of 

 each experiment (80 tests). These results are summarized in Table 3. 

 Both sets of tests indicated that most of the differences (58 to 62%) 

 were not significant (P > 0.05). The two sets of tests also indicated 

 that 35 to 37% of the significant differences were caused by larger 

 populations in the incubated controls as compared with the 

 unincubated samples. Most of the significant differences, further- 

 more, are accounted for by relatively few species. Holopedium 

 gibberum, Eubosmina coregoni, and immature copepods account for 

 54 to 68% of the significant decreases, and Chydorus sphaericiis and 

 Cyclops bicuspidatus thomasi account for 57% of the significant 

 increases (Table 3). 



DISCUSSION 



The results of this study showed that added cadmium concentra- 

 tions as low as 5 jUg/liter caused pronounced effects on both 

 functional and structural attributes of a Lake Michigan zooplankton 

 community in 9 days, but the effects of enclosure were relatively 

 minor up to 15 days. The results indicate further that added 

 cadmium concentrations less than 5 ^tg/liter probably would cause 

 detectable effects within 9 days and that incubations up to 15 days 

 are feasible. 



The responses of species' r to environmental changes are, 

 collectively, a fundamental functional attribute of a plankton 

 community because these responses affect mineral-cycling and 

 energy-flow pathways by determining the absolute and relative 

 abundances of the different species. In this study the observed 

 effects of cadmium on values of r for the different species were 

 probably caused mostly by direct toxicity because the effect in all 

 cases was a reduction of r. Furthermore, the duration of the 

 experiments was probably too short for secondary effects to 

 manifest themselves. Hurlbert (1975) defined secondary effects as 

 those which take place in an ecosystem followdng and as a result of 



