BARKLEY: SELECTIVITY OF TOWED-NET SAMPLERS 



a 6-hr tow doubtless underestimates their actual 

 (as opposed to mean) population density by in- 

 tegrating over a 33-km distance. 



Finally, two of the four species discussed here 

 were sampled in a manner which permits use of 

 equation (8) to estimate values of K (the pro- 

 duct XoUe). Anchovy were sampled with two 

 different nets at one speed of tow, and B. stilbius 

 were sampled with one net at two ranges of 

 speed. Table 3 shows values of K calculated for 

 these two species. For any one choice of speeds, 

 values of K for B. stilbius are nearly constant 

 within the class intervals where fish were re- 

 tained by the mesh and caught in numbers large 

 enough to yield useful estimates of C1/C2, the 

 ratio of catches obtained at the two speeds. Val- 

 ues of K for anchovy, on the other hand, were 

 not constant. The latter result is not surprising, 

 since widely different values of a^o were obtained 

 for the 1-m net and the IKMT whereas Ue would 

 not be expected to differ for the two nets. Ac- 

 cordingly, the product of XoUe should not be the 

 same for both nets, which violates the basic as- 

 sumption used in calculating K with equation 



(8). 



When values of K from Table 3 are used to 

 calculate Pc from equation (7) , the results agree 

 with those obtained graphically for B. stilbius 

 at the modal speeds, but not for anchovy. Ap- 

 parently the calculated values of K for anchovy 

 in Table 3 are too low, since they yield values 

 of Pc which are too high, corresponding to val- 

 ues expected for reaction distances of about 2 

 to 3 m for both nets. Until avoidance behavior 



is better understood, equation (8) should be Used 

 with caution. 



SUMMARY AND CONCLUSIONS 



The ideal sampler for plankton and nekton is 

 one whose selective characteristics are known to 

 be appropriate, qualitatively and quantitatively, 

 for the problem under study. Major factors de- 

 termining the selectivity of towed-net samplers 

 are avoidance and mesh selection, both of which 

 are strongly dependent on species, size or stage 

 of development, and physical condition. Thus 

 the theory of towed-net selectivity must be based 

 on a combination of general principles and de- 

 tailed knowledge of the physical characteristics 

 and behavior of each species of interest, and of 

 each net design as well. 



This preliminary study deals with one aspect 

 of selectivity: the basic principles of avoidance. 

 A theoretical equation is derived for Pc, the min- 

 imum probability of capture, for animals which 

 respond individually to an oncoming sampler by 

 attempting to dodge. The theory is based on the 

 amount of time animals allow themselves for 

 avoiding the net, the animal's speed relative to 

 the net, and the geometry of the encounter. 

 Animals are characterized by their reaction 

 distance, Xo, and their escape speed, Ue. The net 

 is characterized by the radius of its mouth open- 

 ing, R, and its speed through the water, U. 

 Equations (5) and (6) show the theoretical re- 

 lationships between these variables and Pc] Fig- 

 ure 3 illustrates these equations graphically. 



Table 3. — Values of the product XqU^ = K for anchovy and Bathylagus stilbius, from equation (8). Anchovy 

 data obtained with two nets towed at one speed, day and night. B. stilbius data obtained with one net towed at two 

 speed ranges; K calculated using minimum, modal, and maximum speeds for each range. 



817 



