PEARCY: DISTRIBUTION AND ABUNDANCE OF SMALL FLATFISHES 



for plaice under 7 cm to 15-307o for plaice over 15 

 cm. Riley and Corlett (1965) and Edwards and 

 Steele (1968) estimated catch efficiencies of 579r 

 and <459f for 0-group plaice in 2-m and 4-m beam 

 trawls, respectively. In this study, the 3-m beam 

 trawl caught about one-fourth the biomass of 

 flatfishes per square meter caught by Demory and 

 Hosie (see footnote 7) off Oregon, indicating a low 

 catch efficiency. 



The length-frequency distributions of Pacific 

 sanddab and rex sole caught in the beam trawl are 

 compared in Figure 2 with those caught in two 

 larger otter trawls with large-sized mesh by Dem- 

 ory and Robinson^ off Oregon. The small-meshed 

 beam trawl retained appreciably smaller indi- 

 viduals than the otter trawls, verifying that it was 

 most effective in capturing small flatfishes and 

 that large fishes effectively avoided this small net. 



The disparity between the estimates of abun- 

 dance by otter trawls and the beam trawl may be 

 magnified by the effects of otter doors, bridles, and 

 towing cables. The bridles from the otter doors to 

 the net (sweeplines and dandy lines) in combina- 

 tion with the wake from the doors can herd fishes 

 into the net from a wide area in front of the trawl, 

 thereby increasing the effective mouth opening 

 (Loverich^^). Thus the abundance estimates cited 



^Demory, R. L., and J. G. Robinson. 1973. Resource surveys 

 on the continental shelf off Oregon. Fish. Comm. Oreg., Annu. 

 Rep., 18 p. 



'"Loverich, G. 1975. Trawl nets evaluated by expert. 

 Fisherman's News, Sept. 1975 and pars, commun. 



cc 



CL 20 



15 

 10 

 5 



5 10 15 20 25 30 35 

 TOTAL LENGTH (cm) 



40 45 



Figure 2. — Comparison of length- frequency distributions of (A) 

 rex sole and (B) Pacific sanddab captured in the beam trawl (BT) 

 with 13-mm mesh with those captured by Demory and Robinson 

 (see footnote 9) in 3'/2-in (89-mm) mesh and 2'/2-in (64-mm) mesh 

 otter trawls on the continental shelf ofTOregon in 197 1 and 1972. 



above that are based on the horizontal spread of 

 otter trawl nets are probably too large. The towing 

 cable of the beam trawl, on the other hand, drags 

 on the bottom immediately in front of the trawl 

 and therefore may frighten fishes away from the 

 net tow path. Size-dependent responses of fishes to 

 the bottom disturbances created by trawl doors, 

 bridles, and cables may contribute to the differ- 

 ences found in size of fishes captured by the two 

 types of trawls. Large fishes, with better swim- 

 ming capabilities, may be herded more effectively 

 by otter trawl bridles than small fishes. Large 

 fishes may also rapidly swim away from the tow- 

 ing cable of the beam trawl. 



Environmental Effects 



An analysis of variance, using a linear model 

 with depth as a continuous variable and sediment, 

 season, and year as indicator variables, was com- 

 pleted on the un transformed data for numbers and 

 weight of fishes per square meter and numbers 

 and weight per tow. Sediment was classified as 

 four types based on the percent of sand (2-4, 28, 

 63-69, 84-99%). Trawls were combined into two 

 seasons, October-April (winter), and May- 

 September (summer). Levels of significance, 

 P<0.05 and P<0.01, are shown in Table 4. Be- 

 cause of the large number of tests, only effects with 

 P<0.01 are considered significant, although 

 P<0.05 are shown to provide indications of possi- 

 ble trends. Effects with P<0.05 will only be dis- 

 cussed if they reinforce an effect of P<0.01, or if 

 two effects of P<0.05 were found in the same 

 species-effect category. 



Sediment Effects 



The number and biomass of slender sole caught 

 per tow appeared to be affected by sediment type 

 (Table 4). This is curious because this species is 

 primarily a pelagic feeder (Pearcy and Hancock 

 1978). Largest catches per tow were made at sta- 

 tions with high percentages of clay and silt, and 

 lowest catches occurred on sandy sediments. 

 Based on sediment types (and stations), catches of 

 slender sole ranked as follows: 2-37f sand (6-8) 

 >63-697. sand (2-7) >28% sand (23) >84-99% 

 sand (22-15) (see also Table 2). The same trend 

 with sediments is indicated (P<0.05) for slender 

 sole weight per square meter, but because catches 

 per square meter were more variable or mean dif- 

 ferences were less, differences were not significant 

 atP<0.01. 



635 



