than a few percent cover on any plates although several of these 

 species also appear to have limited depth maxima (Table 2). 



Observations of the two-year settling plates indicate that 

 the low percent cover of encrusting species was maintained at 

 the two deepest stations, but that the two shallower stations 

 achieved 100 percent cover of bryozoans, with the introduction of 

 other phyla including ascidians ( Aplidium sp. ) and crustose 

 coralline algae. The erect bryozoans appear fouled by algae on 

 the 30 m plates and other bryozoans and ascidians were 

 overgrowing both erect and encrusting bryozoan colonies. Similar 

 results were seen on granite blocks. The first year photographs 

 show bryozoan-dominated communities similar to those on settling 

 plates, with Spirorbis , Filograna and a few Anomia . The depth 

 pattern was also similar, with high percent cover at the two 

 shallow stations and low percent cover at the deepest station. 

 The horizontal blocks at 30 m showed greater algal colonization 

 than did vertical settling plates, but the second year's vertical 

 granite slabs showed similar bryozoan dominated assemblages 

 although at lower percent cover. When the second year's granite 

 slab arrays were brought to the surface all mobile species were 

 counted and collected (Table 3) except for those on the 30 m 

 slabs which were left in place and photographed by divers. Of the 

 species found, only Margarites spp. , Leptasterias , Henricia and 

 possibly the other gastropods are likely to be bryozoan 

 predators. These species were most abundant at 50 m, 

 corresponding with the highest bryozoan cover. 



DISCUSSION 



The early successional assemblage in deep rocky habitats of 

 the Gulf of Maine was dominated by approximately twenty species 

 of erect and encrusting bryozoans, two polychaetes, and one 

 bivalve species. The bryozoans occupied well over 95 percent of 

 the available space after one year of colonization, with the 

 other groups rarely accounting for more than one to five percent 

 cover. This preponderance of bryozoans in the assemblage is 

 unique in comparison with other similar studies from shallower 

 temperate communities, although bryozoans form a large fraction 

 of the species in cryptic reef habitats (Buss and Jackson, 1979; 

 Buss, 1980; Jackson, 1977a, b, 1979) and on certain temperate zone 

 subtidal rock walls (Rubin, 1980) and settling plates 

 (Sutherland, 1977; Karlson and Sutherland, 1982). The appearance 

 of ascidians and species from other phyla in the second year of 

 our study indicates that this assemblage will not persist and 

 that community development is likely to continue toward the 

 sponge, ascidian, and sea anemone dominated assemblage typical of 

 unmanipulated rock walls at these depths. 



Idmidronea atlantica appears to play a major role in 

 structuring the fouling community at the three shallowest 

 stations, but especially at the 50 m station. This erect species 

 overgrows the only other species with an erect growth form, 

 Tubulipora lilliacea , and many of the other species as well. It 



59 



