for each species constituting more than 1 percent of the total macro- 

 fauna (Tables 31 to 42) . The total fauna, including both macrofauna 

 and larger meiofauna, varied from 230 (site 3, transect VI, October 

 1975) to 132,700 individuals per square meter (site 1, transect IV, 

 July 1976). Densities were generally highest on transect IV and 

 lowest on transect VI and meiofauna species, primarily copepods, had 

 the highest densities. 



The burrowing amphipod, L. dysticus, was abundant with to 3,300 

 individuals per square meter. Another amphipod, Monoaulodes sp., was 

 common with to 1,360 individuals per square meter. Two other 

 amphipods, Leptoaheirus plumulosus and Gammarus sp., were collected on 

 the beach, but their densities were low and their occurrence was 

 sporadic . 



The burrowing polychaete, Scoleeolepides vlvidis , was found on all 

 sites, but was most abundant in deep waters with to 435 individuals 

 per square meter. Two other polychaetes, Lysippi-des gvayi and 

 Laeoneveis culvevi, occurred sporadically and in low numbers, but their 

 frequency of occurrence was highest during July 1976. 



Chironomid larvae and the oligochaete, Pelosoolex sp . , were usually 

 abundant with to 2,579 and to 41,480 individuals per square meter, 

 respectively. They were most abundant on transect IV and on site 8 

 of all transects. 



The last major species was the brackish water clam, R. cuneata, 

 which was commonly collected on the deep sites with to 115 indivi- 

 duals per square meter. 



d. Analyses of Variance . Three-way analyses of variance were 

 performed on the total macrofauna and on all major species, using site, 

 season, and transect as the independent variables (Table 43). The two- 

 way interactions between site and season, transect and site, and season 

 and site were often highly significant. The three-way interaction 

 between site, transect, and season was also usually significant. 

 Significant interactions make interpretation of the main effects 

 difficult, because the levels of the measured variables are affected 

 nonadditively by the levels of two or more independent variables. The 

 interpretations were made after re-examining the original data and 

 determining the direction and magnitude of the interaction term. The 

 total fauna showed significantly different densities for site, transect, 

 and season. Total fauna increased on all sites during May and July 

 1976, but greatly increased on sites 1 and 2 of transect IV because of 

 marsh development. 



Scotecotepides viridis showed highly significantly different 

 densities for site and season but no significant difference in densities 

 for transect. This species was most abundant on the deep sites with a 

 significant difference in density between samplings with an increase 

 during March and May. 



57 



