FISHERY BULLETIN; VOL. 87, NO. 4, 1989 



the purse seine is more efficient for the burrow- 

 ing species (Callinectes spp., Citharichthys 

 stampjlii, Penaeus notialis) and for T. guine- 

 ensis. 



Size selectivity appears for two species (T. 

 guineensis and G. nigri) when the ratio of 

 catches is calculated by size gi'oup (Table 9). As 

 it has been shown above for the marked species, 

 the size composition of fishes in the enclosure 

 during the experiment does not differ from that 

 of the final catches of the beach seine. Two fac- 

 tors remain explaining this size selectivity: 1) a 

 size selective accessibility within the enclosure 

 and 2) a size selective catchability by the purse 

 seine. The first factor is impossible to assess. 

 The second one may happen with active gear, 

 hke the seines. Larger individuals may be better 

 able to avoid capture because of their higher 

 maximum swimming speed (Bainbridge 1958; 

 Blaxter 1967). This type of size selectivity has 

 been shown in sampling plankton larvae with an 

 experimental active gear (Murphy and Clutter 

 1972), and may here explain the decrease in the 

 ratio of the catch rates with size in T. guine- 

 ensis. 



In G. nigri, the selectivity is reversed, small 

 sizes being underrepresented in the purse seine 

 catches. This point seems difficult to interpret, 

 and probably complex mechanisms are involved: 

 enclosure effect (size-dependent accessibility) 

 and size dependent catchability owing to com- 

 plex behavior. Some descriptions of complex be- 

 havior of fish during a fishing operation are 

 given in the literature. For trawlers and Danish 

 seines, the flight is triggered by a stimulus, 

 mainly visual, from the moving gear at a certain 

 distance (MacMullen 1981). Different species 

 react differently; some demersal species jump 

 perpendicularly, while others jump in random 

 directions (Hemmings 1967). Anchovies sur- 

 rounded by a purse seine tend to move into 

 deeper waters (Inoue and Ayodhyoa 1967). 



Efficiency 



For the results of the efficiency measurement 

 to be valid, all of the assumptions must be met. 

 Consequently, efficiency estimates may not be 

 completely reliable. 



Efficiency of the purse seine for fishes larger 

 than L,oo is very low according to our results 

 iChrijsichihys spp., 12% and T. guineensis, 

 18%). Actually, purse seining is an efficient tech- 

 nique when based on spotting and surrounding 

 pelagic fish schools. However, it probably be- 



comes very inefficient for "blind" fishing of 

 demersal species, as was done in this experiment 

 and as is often practiced in Ivory Coast lagoon 

 fisheries. The main cause of this inefficiency is 

 most likely the avoidance during the surround- 

 ing phase of the operation. 



The efficiency of a large, nonmotorized beach 

 seine, reaching the ground in shallow waters, 

 depends mostly on a low escapement rate after 

 closing the net. Estimation of the escapement 

 rate by mark-recapture can thus provide a sim- 

 ple and rehable upper estimate of efficiency. 

 However, it is important to stress that a general 

 application of such values to the entire fishery is 

 not possible unless the variability of gears and 

 fishing gi'ounds is considered. 



ACKNOWLEDGMENTS 



Our thanks go particularly to Jean Raffray, 

 Yann Moreau, Aka Ledjou, and Bile Adoueni 

 for their valuable help during the field experi- 

 ment; to the fishermen from Anga; to Nassere 

 Kaba for the initial translation into English; and 

 to A. E. Dizon for his thorough review and his 

 help. 



LITERATURE CITED 



Bainbridge, R. 



1958. The speed of swimming of fisli as related to size 

 and to frequency and amplitude of tail beat. J. Exp. 

 Biol. 3.5:109-133. 

 Barkley, R. A. 



1964. The theoretical effectiveness of towed-gear sam- 

 plers as related to sample-size and to swimming speed 

 of organisms. J. Cons. int. Explor. Mer 29:146-157. 

 1972. Selectivity of towed-net samplers. Fish. Bull., 

 U.S. 70:799-820. 

 Beverton, R. J. H., and S. J. Holt. 



1956. A review of methods for estimating mortality 

 rates in exploited fish populations, with special refer- 

 ence to sources of bias in catch sampling. Cons. Int. 

 Explor. Mer Rapp. Proc-Verb. 140:67-83. 

 Blaxter, J. H. S. 



1967. Swimming speeds of fish. FAO Fish. Rep. 

 62:60-100. 

 Charles-Dominique, E. 



1983. Evaluation des ressources et reglementation 

 des peches d'une lagune cotiere de Cote d'lvoire: 

 la lagune Aby. FAO Stud. Rev. GFCM 61(1);233- 

 251. 

 Dagnelie, P. 



1975. Theorie et methodes statistiques. Applica- 

 tions agronomitjues II. Presses agronomiques Gem- 

 bloux, 463 p. 

 Durand, J. R., J. B. Amon Kothias, J. M. Ecoutin, F. 

 Gerlotto, J. P. Hie Dare, and R. Lae. 



1978. Statistiques de peche en lagune Ebrie (Cote 



920 



