604 



Fishery Bulletin 92(3). 1994 



Koslow and Ota, 1981; Houston and Haedrich, 1984), 

 and feeding habits (Cartes and Sarda, 1989; Cartes, 

 1991) may also be related to distribution of A. 

 antennatus. Ghidalia and Bourgois (1961) and 

 Bombace (1975) associated certain shrimp species 

 with water masses that were of a characteristic 

 septentrional type, with a low temperature of 12.8C 

 and high salinity (38.1-38.8 ppt). However, it has 

 not been possible to confirm these hydrographic hy- 

 potheses for A. antennatus, because the study by 

 Ghidalia and Bourgois ( 1961 ) reported few catch data 

 of this species. No other studies on related species 

 have established more specific hypotheses. 



The growth, abundance, parent stock biomass, and 

 recruitment of commercial species are important 

 when calculating parameters directly related to their 

 population dynamics and exploitation (Caddy and 



Figure 4 



Mean number of indi- 

 viduals of rose shrimp, 

 Aristeus antennatus, 

 showing 95% confidence 

 intervals (x 2 =8.67; P< 

 0.01): 7;<'=number of 

 samples below median; 

 'n>'=number of samples 

 above median; Q3-Q1= 

 difference between 

 maximum and mini- 

 mum value. US=upper 

 slope; MS=middle slope; 

 LS=lower slope. 



Sharp, 1988; Sarda, 1993). Size-frequency data for 

 A. antennatus caught by commercial fishing vessels 

 (Demestre, 1990; Fig. 7) do not show progressions in 

 monthly length-frequency modes. Apparent "nega- 

 tive growth rates" between different months suggest 

 that vessels were following a moving stock and that 

 catches were taken at the most commercially profit- 

 able locations. Procedures for analyzing this type of 

 stock have been considered by Jones (1984), Caddy 

 ( 1982, 1987), and Caddy and Garcia ( 1986) from fish- 

 ery catches. Procedures for treating migratory stocks 

 have been considered primarily by Sousa (1988) in 

 the fish Decapterus russelli. Bias in size frequencies 

 due to migratory effects in D. russelli is similar to 



£• 2 ° 



1 15 



g 10 



H 5 

 8 



> 



6 



Illl'i'i 



Figure 5 



Monthly catch rates (A) and coefficient of variation 

 (B) based on daily catches of rose shrimp, A. 

 antennatus, taken by a vessel during 1984, 1985, and 

 1986 (redrawn after Tobar and Sarda, 1987). Num- 

 bers are total number of samples by month for all 

 three years combined. 



Table 3 



Results of applying multifactorial nonparametric analysis of variance to the basic data matrix presented in Table 2 

 for collections of rose shrimp, Aristeus antennatus, with and without lower slope (LS) data. M = males; F = females; 

 J=juveniles, rc=total number; ( + )=significant difference (P<0.05); ( — )=nonsignificant difference. 



With LS data 



Without LS data 



M 



M 



Seasonality 



Depth 

 Interaction 



