km/h (3.2 kn); Pat San Marie, 5.9 km/h (3.2 kn); and Oregon, 

 3.7 km/h (2.0 kn). 



The standard duration of tows (on bottom) at all stations was 

 30 min, with the exception of some 60-min tows taken during the 

 comparative fishing trials between Oregon and Pat San Marie. 

 The timing of each tow was started when the trawl winch brake 

 was set, after having allowed the trawl to settle to the bottom. The 

 end of tow was marked when the trawl winch was started to 

 retrieve the net. Loran-C readings were taken at the start and end 

 of each tow, and the computed distance between these two points 

 was used as the best estimate of actual distance sampled. These 

 distances varied significantly between vessels because of dif- 

 ferences in towing speeds. Within the series of trawls taken by 

 each vessel, between-tow variance in distance sampled was caused 

 by varying ocean currents, winds, and bottom friction. 



Environmental data collected with each trawl sample were posi- 

 tion, mean bottom depth, sea surface temperature, bottom water 

 temperature, and observations of cloud cover. Seawater tempera- 

 tures were measured by bucket thermometer readings and ex- 

 pendable bathythermograph casts on the Miller Freeman, Anna 

 Marie, and Oregon. 



Sampling 



Initial handling of the catch. — The method of processing the 

 catch depended on its size. If the catch was less than approx- 

 imately 1,150 kg, it was dumped directly onto a sorting table and 

 completely processed. Larger catches were subsampled using the 

 procedures described by Hughes (1976). 



When subsampling was required, a deck bin was lined with a 

 retainer net, upon which a subsampling net was placed over ap- 

 proximately one-half of the bin. The total catch or a split of the 

 catch was then brought aboard, weighed with a dynamometer, 

 and released into the deck bin. When the total catch had been 

 weighed and loaded into the bin, the subsampling net was lifted 

 and emptied onto the sorting table. The unused portion of the 

 catch remaining in the deck bin was then quickly returned to the 

 sea by lifting the retainer net. 



Sorting and weighing the catch. — After the catch had been 

 transferred to the sorting table, it was sorted by species into wire 

 bushel baskets and tubs. For catches numerically dominated by a 

 single species, sets of two to three baskets were filled at the same 

 time with that species, weighed, and placed on the deck in rows. 

 When a large number of baskets (15-25) were required for a single 

 species, only 1 basket was kept from each set of 2-3 baskets, and 

 the others emptied and reused for sorting the remainder of the 

 catch. While the dominant species was being sorted, other species 

 were sorted into single baskets or other containers. The procedure 

 of filling single or sets of baskets was repeated until the entire 

 catch for subsample) was sorted and weighed. Baskets were placed 

 on deck in order of sorting, so that they could be identified as if 

 they were from the top, middle, or bottom of the trawl sample, 

 lost organisms occurring in the trawl samples were identified 

 to species, although those that were difficult to reliably identify 

 vere grouped by genus or combined within a higher taxonomic 

 level. Catch weights for all taxa were determined by weighing 

 baskets to the nearest 0.5 kg on a 140 kg capacity platform scale. 

 Numbers of individuals were determined by direct count or by ex- 

 panding the number determined from a weighed subsample. 



Subsampling for biological data. — After weighing and counting, 

 the catches of 1 1 species of principal interest were further processed 



for length-frequency and individual specimen data. These species 

 were (in order of priority for data collection) 



1 . Walleye pollock, Theragra chaicogramma 



2. Yellowfin sole, Limanda aspera 



3. Rock sole, Lepidopsetta bilineata 



4. Flathead sole, Hippoglossoides elassodon 



5. Pacific halibut, Hippoglossus stenolepis 



6. Pacific cod, Gadus macrocephalus 



7. Sablefish, Anoplopoma fimbria 



8. Pacific ocean perch, Sebastes alutus 



9. Greenland turbot, Reinhardtius hippoglossoides 



10. Arrowtooth flounder, Atheresthes stomias 



1 1 . Alaska plaice, Pleuronectes quadrituberculatus. 



For the collection of length-frequency data, subsamples of 

 50-200 individuals were randomly selected from the baskets of as 

 many of the 1 1 fish species occurring within the trawl catch as 

 time permitted. Species having highest priority were selected and 

 measured first. Individuals within the subsamples were sorted by 

 sex (male, female, undetermined), then the lengths within each sex 

 group were recorded. Fish lengths were measured to the nearest 

 centimeter either as fork lengths (FL), from the tip of the snout 

 (or longest jaw) to the center of the caudal ray fork, or as total 

 length (TL), from the tip of the snout to the longest extension of 

 the caudal fin. The sex of small juvenile fish could not always be 

 determined. 



Specimen data (descriptive observations from individual fish) 

 were collected by two sampling approaches. Length-age and 

 length-weight observations were collected with a selection of in- 

 dividuals (samples) stratified by sex and centimeter length-size 

 class. Observations of the reproductive condition of walleye 

 pollock (length-maturity data) were recorded from the same ran- 

 domly selected individuals used for length-frequency measure- 

 ments. Each vessel was assigned a list of species and types of 

 specimen data to collect from the different survey areas. 



For each species, length-age data were collected by taking 6-10 

 age (skeletal) structures for each sex-centimeter length-size group, 

 within each assigned otolith area. Saccular otoliths were used for 

 determining the ages of all species except Pacific cod, for which 

 scales were used because they showed clearer annual growth rings. 

 Otoliths were stored in glass vials in 50% isopropanol. Pacific cod 

 scales were stored dry in small paper envelopes. 



Length-weight determinations were made for five individuals 

 within each sex-centimeter length-size group. Whole, freshly 

 caught individual fish were weighed at sea to the nearest gram on 

 a triple-beam balance. 



Age Determinations 



Age determinations made as a part of the present study have 

 assumed that rings in otoliths and scales (zonal discontinuities) in- 

 dicated annual marks. Previous studies have supported the 

 hypothesis that an annual time scale can be fitted to the patterns 

 of skeletal ring formation of walleye pollock (LaLanne 1975 J ) and 

 yellowfin sole (Hatanaka 1968). Although similar analyses have 

 not yet been conducted for other Bering Sea species, zonal discon- 



'LaLanne, J. J. 1975. Age determination of walleye pollock {Theragra 

 chaicogramma) from otoliths. Unpubl. manuscr., 19 p. Northwest and .Alaska 

 Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake 

 Boulevard East, Seattle, WA 98112. 



