The two most common materials used in the New England 

 fishery are nylon (polyamide) and polypropylene. In com- 

 parison fishing these two materials, Bohl (1966) found that for 

 haddock a polyamide cod end gave selection factors about 

 7-10% higher than a polypropylene one. He reasoned this was 

 due to the greater extensibility of the polyamide and the fact 

 the polypropylene webbing had larger knots. In further 

 studies, Bohl (1968) compared three different types of 

 polypropylene twine (splitfiber, continuous, and monofila- 

 ment); results indicated no significant difference in selectivity 

 even though physical properties were very different. Bohl 

 (1971) also found no significant differences in the selection 

 factor between a "normal" polyamide cod end and an extra- 

 strong one. He also failed to find a correlation between 

 elongation and selectivity. In general, polyamide gives the 

 highest selection factors, followed by polyester, 

 polypropylene, and manila (Pope et al. 1975). 



Very little is known about the relationship between towing 

 speed and selectivity. This is probably due to the practical dif- 

 ficulty of accurately measuring the speed of the trawl over the 

 bottom and maintaining other parameters constant. Trawl 

 mensuration studies at the Northeast Fisheries Center on "36" 

 and "41" Yankee trawls indicated that varying towing speed 

 within the range of 2.5-4.0 kn can change the headrope height 

 by several feet. If, for example, the larger fish of a particular 

 species stay further off the bottom than the smaller ones, by 

 varying speed so as to increase headrope height the trawl will 

 select the larger fish. This will ultimately show up in the selec- 

 tion factor calculated for the cod end. 



It has also been shown that towing speed affects the 

 hydrodynamics of the trawl. Beverton and Margetts (1963) 

 found the drag increases approximately exponentially with 

 towing speed. They calculated, at speeds of 3-4 kn, drag forces 

 on 53, 69, and 215 mm mesh cod ends of 800, 700, and 150 lb, 

 respectively. There is little doubt that speed affects the tension 

 in the twine of the cod end meshes and thus probably the selec- 

 tivity. The Russians, realizing this fact, have studied this ap- 

 proach in their trawl design efforts (Treschev 1963). Saetersdal 

 (1960) did find a tendency of the selection factors for cod to in- 

 crease with decreasing speed in the range of 2-3 kn as indicated 

 by the ship's speed log, but this was not evident for haddock. 



Clark (1963) found that the longer the tow the higher the 

 escapement and thus the selection factor for haddock. The 

 selection factors went from 3.0 for 20-min tows to 3.4 for 

 80-min tows. Pope and Hall (1966) did not find a marked ef- 

 fect, like Clark, for haddock but did see a tendency for higher 

 selection factors in 2-h tows compared with 1-h tows. The 

 general explanation for the above phenomenon is that the 

 longer tow time gives a fish more opportunity to make 

 repeated attempts at escape. As tow time increases so usually 

 does the catch and this may have a counterbalancing effect. 



Clark (1963) found that for haddock the selection factor 

 decreased with larger catches; the 50% retention point decreas- 

 ing by as much as 5 cm. McCracken (1963) reported no change 

 in haddock selection factors for catches up to 1,000 fish/tow; 

 however, there was a slight drop in selection factors for larger 

 tows. He could not demonstrate this effect for cod. Hodder 

 and May (1964) presented data indicating slight decreases in 

 selection factors for cod and haddock with larger catches, but 

 not of a magnitude to affect assessments. There are several 

 papers that report no apparent effects (ICES 1965; Pope and 

 Hall 1966). 



There are a number of reasons that have been advanced to 

 explain lower selection factors for larger catches. The fish 

 would have less of a chance to be selected by the larger meshes 

 at the aft end of the cod end. There may be more tension on 

 the meshes making them less flexible, or the meshes may just 

 become blocked. Schooling behavior may even come into play. 

 On the other hand, Pope et al. (1975) reported that this effect 

 has only been observed in covered cod end tows and thus may 

 be an artifact of the method. With larger catches more fish 

 may be swimming back into the cod end from the cover or may 

 be escaping forward of the cover, thus reducing the apparent 

 selectivity. 



While it is generally assumed that selection factors are 

 relatively constant through a range of mesh sizes, this has been 

 shown not to hold in certain cases. Clark (1963) demonstrated 

 that for silver hake the selection factor increases with mesh 

 size. He reasoned that this was due to a greater flexibility of 

 the larger mesh allowing more fish to force their way through. 



Another aspect of selectivity that varies with mesh size is the 

 selection range, the area between the 25% and 75% retention 

 lengths on the selection curve where most of the escapement 

 occurs. The smaller the selection range, the sharper the selec- 

 tion. Clark et al. (1958) found that for haddock the selection 

 range for a 75 mm mesh was 4 cm compared with 14 cm for a 

 150 mm mesh. 



As mentioned previously, trawl efficiency apparently in- 

 creases with cod end mesh size for most species. Davis (1934b) 

 was one of the first to observe this phenomenon for haddock. 

 A larger mesh caught more of the larger size fish. Clark (1963) 

 and Templeman (1963) reported similar results. Evidence ex- 

 ists that indicates this increased efficiency is not related to an 

 increase in speed or ground covered by the larger mesh (Bever- 

 ton and Margetts 1963; Clark 1963). Beverton and Margetts 

 also indicated that the decrease in drag of a trawl caused by 

 having a larger mesh cod end is relatively insignificant. 



The escapement ability, hence the selection factor, can vary 

 considerably from one species to another. The relationship 

 between the shape of the mesh and the shape of the fish is con- 

 sidered important. Roundfish tend to have a cross-sectional 

 shape more nearly matching that of a mesh than flatfish, and 

 thus tend to have a higher escapement rate for a particular 

 length. The behavioral response of a particular species to a net 

 is a key factor also. Clark (1963) has demonstrated for silver 

 hake that this species has a lower escape response when com- 

 pared with other species. In general, for roundfish, when girth 

 is compared with mesh circumference, the majority of the fish 

 that theoretically can fit through do in fact escape. Draganik 

 and Zukowski (1966) found that haddock which escaped from 



Table 1. — Selection factors. 







Single- 



twine 





Species 



Polyamide (nylon) 



Polypropylene 



Atlantic cod 



3.6 







3.5 



Haddock 



3.4 







3.3 



YeDowtail flounder 



2.3 







NA' 



Winter flounder 



NA 







NA 



American plaice 



2.3 







NA 



Pollock 



NA 







NA 



'NA = Not available. 



