If the selected curves of the same series are observed together, 

 it appears that they move to the right by enlarging the meshes | that 

 is to say, more undersized fish escape by enlarging the mesh widtho 



Howeverj there are other factors responsible for the direction 

 of the selection curves o Figure 5d shows selection curves of cod 

 enS of 12 1/2 centimeter double twine and 12 centimeter single twine 

 torwed by steam trawlers (Herrington, 1935). The somewhat narrower, 

 single^twine cod end clearly gives more undersized fish an opportunity 

 for escapee This is understandable although single and double knitted 

 nets of the same mesh widths are knitted on the same shuttle, the 

 double knitted ones have smaller openings especially when the knots 

 swell after submerging (Davis, 1929), 



Complaints came from many small trawl fishermen about the loss 

 of marketable sole when a minimum mesh width of 70 millimeters for 

 trawl nets was ordered in 1937o The complaints were investigated 

 (de Veenj 1937) and indeed it appeared that a fairly large amount 

 of marketable sole was lost with the 70 millimeter meshes, while by 

 the use of meshes of 55 millimeter practically no very smallo 

 marketable plaice were being caught. This pertains chiefly to very 

 small ships, most of which were sailing craft with low towing speeds. 

 The faster towing draggers and cutters caught considerable amounts 

 of small sole, even with the 70 millimeter meshes. The only con= 

 elusion that can be made is that the chance of escape diminishes 

 with the increase of the towing speed. Figure 5a for haddock shows, 

 in addition to the series of 3 cod ends with meshes of 5 l/2, 6 l/2 and 

 8 3/4 centimeters, a selection curve for a cod end with a mesh size 

 of 10 centimeters (Davis, 1934 o) . The position of the curve 

 (judged with the length group whereby 50 percent remained in the net) 

 is equal to that of the curve of 8 3/4 centimeters. It is very 

 probable that the 10 centimeter curve moved to the left, compared 

 to the other three, since the first curve was determined on a 

 trawler with greater towing speed than the one used in the other 

 tests. For the tests in which plaice and sole were caught by 

 sailing trawlers, the single-meshed nets and the slow towing speed 

 caused a moving to the right of the selection curves, in relation 

 to the curves prepared for plaice with the aid of double-twine nets 

 towed by steam trawlers. Unfortunately we do not possess curves 

 for sole prepared from tests made with steam trawlers and double- 

 twine nets. 



It is clear from the shape of the selection curves that the net 

 does not make a clear selection but always catches a few fish below 

 a certain size and allows a few above that size to escape. Therefore, 

 it is never possible to avoid, the catching of all fish below the 

 minimum sizeo There is a certain non-selective action and as this 

 action lessens the curve becomes steeper. There are two factors 

 which probably encoiirage the selectivity and therefore make the curves 

 steeper© These are a smaller towing speed and a single -knitted mesh* 



42 



