FISHERY BULLETIN: VOL. 80, NO. 1 



Table 1.— Groundfish cruises of the RV Dolphin. 



tinental shelf and upper continental slope be- 

 tween Cape Fear, N.C., and Cape Canaveral, 

 Fla., except in spring 1974 when sampling ex- 

 tended to Cape Hatteras. A preassigned number 

 of stations was selected randomly (Grosslein 

 1969) with a set number in each of six depth 

 zones (9-18 m; 19-27 m; 28-55 m; 56-110 m; 111- 

 183 m; 184-366 m). Bottom water temperatures 

 were measured at each station with mechanical 

 or expendable bathythermographs. 



Thirty-minute trawls were made continuously 

 (day and night) from the RV Dolphin, at 6.5 km/h 

 with a towing wire scope of 2.5-3.0:1. The trawl 

 was a 3/4-scale version of a "Yankee No. 36" with 

 a 16.5 m footrope, 11.9 m headrope, and 1.3 cm 

 stretch mesh cod end liner (Wilk and Silverman 

 1976). 



Fork lengths (later converted to total lengths) 

 of all fish collected by trawl were recorded to the 

 nearest centimeter. Frozen fish samples were 

 taken to the laboratory for further investiga- 

 tions. 



An index of relative abundance (Musick and 

 McEachran 1972) was calculated for each depth 

 zone by the following expression: 



Index of Relative Abundance 



2 1n(x+l) 



n h 



where n h = number of trawls in the Mh depth 

 zone, and x = number of individuals for each tow 

 in a given depth zone. Because previous investi- 

 gators have shown that trawl catches are usually 

 distributed as a negative binomial (Elliott 1971; 

 Taylor 1953), a In (x + 1) transformation was 

 made on the relative abundance data to permit 

 statistical tests to determine if the differences 

 among habitats within depth zones were signifi- 

 cant. 



Estimates of biomass standing stock were cal- 

 culated with both transformed, In (x + 1), and 

 untransformed data for comparison of the result- 

 ing values. The stratified mean catch/tow (Coch- 

 ran 1977) was calculated by the expression: 



*- = N k iNhVh] 



where y st = stratified mean catch(kg)/tow, 

 N = total area, 

 Nh = area of Mh depth zone (from plani- 



meter chart measurements), 

 y h = mean catch/tow in the Mh depth 



zone, and 

 k = number of zones in the set. 



The area of live-bottom habitat in each depth 

 zone («44.5%) was estimated from the frequency 

 of occurrence of sponge and coral in catches dur- 

 ing 5 yr of bottom trawling with the stratified 

 random sampling design. The areas of sandy- 

 bottom habitats were obtained by subtraction. 

 The estimated population variance of the mean 

 catch(kg)/tow was also calculated by Clark and 

 Brown (1977): 



S 2 = 



N mi 



[N h y 2 ]- NyJ + ! Sf \(N h - 



*=i 



r 



+ 



(N h -N)(N h -n h ) 



N 



n h 



] 



where S = estimated population variance, and 

 S h 2 = variance of the Mh zone. 



The mean catch/tow (y h ) of the transformed In 

 (x + 1) data was estimated for each depth zone 

 following the methodology of Bliss (1967): 



E(y h ) = exp (y h 



+ S 2 /2) 



where E(y h ) = the estimated (retransformed) 

 mean catch(kg)/tow in the Mh depth zone, y h and 

 S 2 , both expressed in logarithmic units, are the 

 zone mean and its variance. The same methodol- 

 ogy was applied to obtain the stratified mean 

 catch/tow from transformed data for the whole 

 study area. Biomass estimates were expanded by 

 the area swept method (Rohr and Gutherz 1977), 

 using 



S&ot = X (P h ) (A„) 



h -1 



where SStot = total standing stock, 



