square meter. The haustorid amphipod, P. longimerus , occurred with up 

 to 4,000 individuals per square meter (particularly on site 8). 



Site 9 was farthest from shore and was located in the middle of the 

 buildup zone where megaripples occur in the rough facies. Parahaustorius 

 longimerus was dominant in this area with up to 8,000 individuals per 

 square meter; however, both S. squoanata and Donax sp. occurred in large 

 numbers. A second haustorid amphipod, Bathyporeia quoddyensis , also 

 occurred in the area with up to 750 individuals per square meter. No 

 other significant species were collected. Eight crustaceans and five 

 polychaetes occurred over the sites but did not dominate any area. 



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

 performed using site number, transect number, and season as the 

 independent variables, and total macrofauna and major macrofaunal 

 species as the dependent variables (Table 17). The two-way inter- 

 actions between site and transect, site and season, and transect and 

 season were often highly significant. The three-way interaction be- 

 tween site, transect, and season was usually significant. When 

 interaction terms are significant, interpretation of the main effects 

 is difficult because the level of the measured variable (number of 

 individuals collected) is affected nonadditively by the levels of the 

 two or more independent variables. An interpretation was made after 

 re-examining the data and determining the direction and magnitude of 

 the interaction effects. 



All species showed a significant difference in density due to 

 season (probability, p <0.05), and most showed a highly significant 

 difference (p <0.001). Densities generally increased from October to 

 June and decreased in July. An exception was the seasonal distribution 

 of Donax sp., which showed no significant change in total numbers 

 between June and July. Clam spats (probably Donax) were numerous in 

 June, but were reduced considerably in July. This indicated that a 

 reduction in the Donax populations between the two samplings was 

 counteracted by maturation of some juveniles. About 50 percent of the 

 major species showed a significant difference in density between 

 transects. Three of the four most abundant species were significantly 

 different between transects (p <0.01), but total macrofauna was not 

 significantly different. Although macrofaunal species composition 

 varied, the total number of organisms at a site did not vary 

 significantly among transects (excluding temporal variation) . 



All the major species except Microphthotmus sczelkowii, Gtyoeva sp., 

 and Petatosars-la declivis showed a significant difference in density 

 between sites. A poster-iovi tests (Student-Newman Keuls' procedure for 

 differences between means at the 5-percent level) were performed on the 

 species showing a significant difference between sites. Emerita 

 taZpoida was most abundant on sites 1 and 2; P. longimerus was most 

 abundant on sites 8 and 9; and S. squamata was most abundant on site 7, 

 but its densities on sites 5 and 8 were also significantly different from 

 other sites. Juvenile Donax sp. were significantly more abundant on 



38 



