or 



-I 2-| 



■0.4 



co 

 a. 



O 



<r 

 en 



UJ 



o 



UJ 

 Q. 



co 



CO 04- 



C 

 D 



E 



F 



G 

 H 



I 



STATION GROUPS 



I 2 3 4 5 6 



8 9 10 



-12 -04 04 CONSTANCY 



SIMILARITY . . __ 



1 I Very Low <0 I ^BHigh>_0 5 



f=l Low >0 I I . I Very High >07 



I I Moderote >0 3 



Figure 6. — Nodal constancy for station and species groups as 

 defined by cluster analysis for the 1978 late spring sponge-coral 

 habitat survey. 



CO 



Q. 



Z> 



o 

 cc 

 o 



CL 



co 



or 



-I 2- 



4 



5 



CO 04- 



FISHERY BULLETIN: VOL. 81, NO. 3 

 STATION GROUPS 



_r: 



I 2 3 4 5 6 7 8 9 10 II 12 



-12 

 SIMI 



■04 0.4 



LARITY 



FIDELITY 



I INeqotiveO I 1 Moderote >2 



L==JLow>J BBHigh23 



FIGURE 7. — Nodal fidelity for station and species groups as defined 

 by cluster analysis for the 1978 late spring sponge-coral habitat 

 survey. 



TABLE 6. — Number of species/tow, diversity, evenness and species richness values for demersal 

 fishes taken spring 1978 in 3 A Yankee trawl tows in the open shelf and sponge-coral habitat, South 

 Atlantic Bight, LCL and UCL = lower and upper 907c confidence limits. 



compared closely with values for a reef at a depth of 

 37 m off the South Carolina coast (Powles and 

 Barans 1980), they were one to two orders of 

 magnitude lower than tropical reef systems (Table 

 7), with the exception of an estimate of 40.4 kg/ha 

 based on five visual transects at Rabbit Island in 

 Hawaii (Brock 1954). This latter value, however, was 

 indicative of marginal reef habitat, since the estimate 

 was derived from counts taken largely over sandy 

 bottom with an occasional rocky ledge or coral head. 



Edwards (1968) provided a rationale for obtaining 

 biomass estimates from otter trawl catches for 

 resource surveys in the western North Atlantic. 

 Catches are adjusted on a species-by-species basis 

 using a correction factor which is derived from the 

 individual species' availability and vulnerability to 

 the survey gear. In addition, areal and seasonal 

 changes in the species' distribution pattern are 

 figured into the adjustment. Correction factors are 

 determined from long-term distribution and relative 



546 



