91 



The dendrogram which resulted from the classification of the 56 

 stations using species abundances as attributes (normal classification) 

 is presented in Fig. 35. For the time being we have truncated this 

 dendrogram at the nine group level. Examination of Fig. 35 shows that 

 all of these groups are fairly discrete but some, for example groups 2 

 and 3, are candidates for further fusion. We have a great deal of faith 

 in this classification because it shows good spatial discrimination 

 (Fig. 36). 



The 10 members of site-group 1 are principally deep-water offshore 

 stations. Site-groups 2, 3 and 7 are limited to the Portland region. 

 Stations in site-groups 2 and 3 are intermingled in outer Portland 

 Harbor and are adjacent to one another on the dendrogram. This suggests 

 a close faunal affinity between them. Site-group 7 stations are found 

 on the edge of the patch of site-group 2 and 3 stations and are far 

 removed from them in the dendrogram. This reflects a real difference in 

 faunal composition undoubtedly controlled by physical factors. 

 Site-group 4 is widely scattered throughout Casco Bay with all but one 

 of the member stations occurring near shore. Site-group 5 dominates the 

 central portion of the Bay while site-group 6 members ring the Bay at 

 shallow stations. The three member site-group 8 exhibits no spatial 

 pattern and site-group 9 is a single station outlier consisting of the 

 mussel reef comminity at station 37. 



The dendrogram resulting from the inverse classification is 

 presented in Fig. 37. Only those noncolonial species occurring at over 

 10% of the stations were used in this procedure. We have tentatively 

 truncated this dendrogram at the 14 group level. The most significant 

 feature of this analysis is the distinct separation of species-group N 



