the cod end, and were retained in the cover, weighed less than 

 fish of equal length retained in the cod end. 



In experiments conducted by Pope and Hall (1966), they 

 could find no relationship between selection factor and depth 

 or daylight vs. darkness. It is also the general opinion of 

 researchers that cod end selectivity is not appreciably affected, 

 at least directly, by the size of the vessel or gear (McCracken 

 1963; Pope and Hall 1966; Bohl 1967). 



Table 1 is a summary, from the best information available as 

 reported by Holden (1971), of the selection factors of the 

 species with which this report is concerned for New England 

 waters. 



NEW ENGLAND MESH STUDY 



This study consisted of four series of experiments in which 

 two commercial fishing vessels performed both covered and 

 uncovered cod end tows. In general, the procedures used were 

 adopted from Pope et al. (1975). All tows were 1-h duration, 

 conducted during daylight hours. The captains followed nor- 

 mal commercial practice of changing course to follow con- 

 tours, going around hard bottom (rock piles), and pursuing 

 fish traces on the echo sounder. Vessel and gear specifications 

 can be found in Appendices A and B. 



The sampling techniques were basically the same in all four 

 experiments. At the conclusion of each tow, the cod end and 

 cover catches (if a cover was used) were kept segregated. The 

 gear was meticulously checked and net damage and other oc- 

 currences that may have affected the validity of the tow were 

 recorded. Cod end and cover knots were tied tight and a piece 

 of old webbing was placed in the end to prevent leakage of 

 catch. 



After each tow, 30 cod end meshes were measured along the 

 top of the cod end in one row starting aft and running for- 

 ward. They were measured using an ICES longitudinal-type 

 mesh gauge set at 4 kg pressure. The segregated catches (cod 

 end and cover, when used) were worked up separately. Any 

 fish found forward of the cod end were excluded because they 

 may not have undergone the cod end selection process. The 

 catch was sorted by species into 1- and 2-bu baskets, weighed, 

 and length-frequency data recorded for each species. In many 

 cases, to save time, the catch was not weighed but all lengths 

 were taken and length-weight equations used to determine 

 catch weight. Randomly selected 2-bu subsamples were taken 

 if the catch was too large to handle by this means. Girth data 

 were also recorded at intervals throughout the experiments us- 

 ing tape measures. 



In 1975, mesh sizes used in the USA Subarea 5 (Gulf of 

 Maine and Georges Bank) cod and haddock fisheries ranged 

 from 110 to 129 mm (4.3 to 5.1 in), with the majority of cod 

 ends examined (>85%) having mesh sizes from 1 15 to 124 mm 

 (4.5 to 4.9 in) (ICNAF 1976). Trawl cod end mesh sizes used in 

 the 1975 yellowtail flounder fishery ranged from 110 to 139 

 mm (4.3 to 5.5 in), with most cod end meshes between 1 15 and 

 129 mm (4.5 and 5.1 in). 



The small mesh size chosen for these experiments was the 

 most commonly used "large" mesh cod end available in New 

 England. It was constructed of #102 braided nylon twine (run- 

 nage 73.76 m/kg) and sold as 4.5-in webbing. The actual 

 average dry-mesh measurement of these cod ends new was 108 

 mm (4.25 in), due to steam treatment during manufacture. The 

 larger mesh size was chosen on the basis of increasing the 



minimum size of cod to 52 cm (20.5 in) or an age-at-first- 

 capture of 3 yr. Using a selection factor of 3.6, this indicated a 

 mesh size of 144 mm (5.7 in). As no webbing of this size was 

 available, handmade cod ends of 154 mm (6.06 in) were con- 

 structed to allow for shrinkage. 



It was noted that measurements for the small "4.5-inch" 

 commercial cod ends used tended to be smaller than the 

 recorded average for the fishing fleet — 4.2 in vs. the fleet's 

 4.75 in. It was assumed that this was due to differences in 

 methodology and a mesh-measuring comparison test was con- 

 ducted. A National Marine Fisheries Service (NMFS) enforce- 

 ment agent, using a wedge-type gauge, measured 10 meshes on 

 one of the large experimental cod ends. The same meshes were 

 then measured using the wedge gauge with a 5 kg weight and 

 the ICES gauge set at 4 kg tension (Fig. 1). The average 

 readings were 144.8 mm (5.7 in), 143.0 mm (5.63 in), and 135.0 

 mm (5.3 in), respectively. Random measurements were then 

 taken on our commercial-sized cod end. The ICES gauge in- 

 dicated a little over 4 in. The wedge gauge readings were about 

 4.5 in; however, the gauge could be wedged in further to read 

 4.75 in or greater (the NMFS enforcement agent said that this 

 is the routine procedure in the field). 



Experiment One 



This experiment was conducted from the fishing vessels 

 Frances Elizabeth and Christopher Andrew on 12, 13, and 15 

 December 1977 in inshore waters off Scituate, Mass. (Fig. 2). 

 On each of the 3 d four tows were made; small- and large- 

 mesh cod ends fished covered and uncovered. The order of the 

 tows was chosen at random and followed by both vessels 

 together, usually within a kilometer of each other. Vessel 

 speed was maintained at 2.0-2.5 kn. 



All cod ends were measured dry before starting the experi- 

 ment. The small cod ends of machine-made webbing initially 



Gloucester 



j _ i 



:. 1 



50 !00 

 (" 1 









— 42°30' y> = 



/C 2 











- 

















\% / ? 











— 



\ \tp / 







- 





^^ 



1> \ ^ 







~~ 



Satuate^t 



^^?- 



_ 





too 















-42° oo r -^ 





v\ 



^ Georges 



\- 









\ ) 



Bank 



_, 







Cape Cod Bay A 



\ 



i 





New «.p 





- ^ i4) 









Bedfordf 4 



§ CAf 



^\^ 



*-— x 







? 



r~- i 









&s 



{mods 

 IHoky 





X 3 



sa 



St 



\ \ 





^^^ l' 



~" z=d 



# 



•v—^i-S: 



% A 





l=^'v 





o 





^m£) 





^=^= 



"\ 







Nantucket 



\ 



50 



i 





i 



Shools 



i 



v.-~ 



Figure 2. — Location chart of mesh experiments. 



