As shown in Table 2, extrapolated faunal densities ranged from 873 

 individuals per square meter (control station 1 in March) to 12,831 

 individuals per square meter (borrow station 4 in June). The average 

 densities for each sampling date in individuals per square meter showed 

 a steady decline through the sampling period — 5,204 in June; 2,960 in 

 September; 2,856 in December; and 1,551 in March. 



In June, control stations 1 and 2 showed no significant differ- 

 ences in the numbers of species or individuals. These stations were 

 also very similar in their species compositions, as indicated in 

 Figure 7, which shows groupings of stations based on degrees of faunal 

 similarity. Tlie relationship between borrow stations was quite 

 different. Borrow station 3 had a significantly greater number of 

 species than borrow station 4, but the latter contained over twice as 

 many individuals. These differences, caused in part by the high 

 conceatration of E, nitens at borrow station 4, were also largely 

 responsible for the borrow stations having a relatively low degree of 

 faunal similarity at this time (Fig. 7). The combined borrow stations 

 had significantly more species and individuals than the combined 

 control stations (p < 0.001). 



In September, control station 1 contained significantly greater 

 numbers of both species and individuals than control station 2 (p < 

 0.001). As expected, these stations also showed little faunal 

 similarity (Fig. 7). Borrow station 3 yielded siginif icantly more 

 species and individuals than borrow station 4 (p < 0.001). These 

 stations also showed relatively little faunal similarity (Fig. 7). The 

 two borrow stations combined contained significantly greater numbers of 

 species and individuals than the two control stations combined (p < 

 0.001). 



In December, control stations 1 and 2 showed no significant 

 differences with respect to numbers of species or individuals, and also 

 showed a high degree of faunal similarity (Fig. 7). Both stations (1 

 and 2) contained large numbers of E. nitens (223 and 194, respectively) 

 Borrow station 3 contained more than twice as many individuals and 

 almost twice as many species as borrow station 4. Although their level 

 of faunal similarity was not particularly high, these stations did 

 occur together in one of the four major groupings in the similarity 

 dendogram (Fig. 7). The low number of individuals collected at borrow 

 station 4 resulted in no .significant differences between the two 

 control stations combined and the two borrow stations combined in terms 

 of faunal densities. However, there were significantly more species at 

 the borrow stations combined than at the control stations combined (p < 

 0.001). 



In March, the control stations showed no significant differences 

 in numbers of species or individuals, and also showed a close asso- 

 ciation in the similarity dendogram (Fig. 7). This was also true for 

 the borrow stations on this sampling date. 



18 



