Morato et al.: Diets of Rqa davata and Galeorhinus galeus 



591 



39° N 



3rw 



CORVO 



P 



FLORES 



c 



39° N 



38° N 



N 



t 



28° W 



27° W 



GRACIOSA 



r~ -t& 



• ^r^ T ERCEIRA 



S. JORGE ^"^"^^^^A * 



, FAIAL ^^^V^ ^^^^^^^^ 



^^ ^^k^ ^^^PICO Banco D Jo^o de Ci 



9H^ 4^^^C ^ Baixo de S Mateus  



Banco Pri^sa Alice ,6° W 25"W • 



^^^^^ 38° N—^ •— 



than 60n m ddoth ^ 



► Less than 600 m depth 



30° W 



28" W 



37° N 



39° N 



38° N 



Mar da Pra^B f 



'4 



24° W 



Figure 1 



Locations of the longline sets made in the Azores during the spring of 1996 (•) and 1997 (D). 



study on demersal fisheries in Azorean waters (Fig. 1). 

 Fishes were caught by longhne onboard the RV Arquipe- 

 lago. Line setting began before sunrise (approx. 05:00 hi 

 and hauhng started about two hours after setting. From 

 the fish sampled, total length (TL, to the nearest cm) was 

 measured, and sex and maturity were determined by mac- 

 roscopic examination of gonads and claspers with maturity 

 scales, as proposed by Stehmann (1987). Stomachs were 

 removed and classified as either everted, regurgitated, 

 with bait, empty, or with contents. Individuals falling in 

 any of the first three categories, as well as those that had 

 obviously eaten fish hooked on the longline, were excluded 

 from further analysis. Stomachs with contents were placed 

 in plastic bags and frozen (within about 2 h of capture) 

 for subsequent analysis. Stomach contents, which partly 

 consisted of a turbid suspension, were washed with water 

 in a nylon net of approximately 0.5-mm mesh size to allow 

 easier examination. The items were carefully separated, 

 weighed (after removing the surface water by blotting 



them in tissue paper), and identified to the lowest possible 

 taxonomic level. Individuals of each identified taxon were 

 counted. Whenever fragments were found, the number of 

 individuals was taken as the smallest possible number of 

 individuals from which fragments could have originated. 



Precision estimates in diet studies have been advocated 

 and used by several authors (Ferry and Cailliet, 1996; 

 Morato et al., 1999). We used the cumulative trophic diver- 

 sity, measured with the Shannon-Wiener index [as H'=— Z 

 P,(log^P|), where P, is the proportion of individuals in the 

 ;th species] to measure sample size sufficiency (Hurtubia, 

 1973). Cumulative numbers of randomly pooled stomachs 

 were plotted against the cumulative trophic diversity. The 

 asymptote of the curve indicates the minimum number of 

 stomachs required. Frequency of occurrence (%0), percent- 

 age number (%N), and weight (%W) for each prey type were 

 used to describe the diet of both species (for a review see 

 Hyslop, 1980; Cortes, 1997). Wet weight was used to de- 

 termine the latter value. The index of relative importance 



