BOWERING: DISTRIBUTION OF WITCH FLOUNDER 



Figure 2.— Nominal catches of witch flounder in NAFO 

 Div. 2J, 3K, and 3L from 1963 to 1984. 



that overall catch might be affected were not in- 

 cluded in the analyses. All fishing tows were of 

 30-min duration. At the end of each set, bottom 

 temperature was measured using an expendable 

 bathythermograph (XBT). 



The geographic distribution and relative abun- 

 dance is shown by indicating, in 1/2° latitude and 

 1° longitude rectangles, the average numbers of 

 fish caught per 30-min set. A preliminary exami- 

 nation of distribution of witch flounder by year 

 (all divisions) and season (Div. 3L only where 

 enough seasonal data were available) showed no 

 differences in geogi'aphic distribution; therefore, 

 all trips were combined for the period 1977-83. 

 These years were chosen for the distribution 

 study since surveys during this period were based 

 upon random distribution of sets throughout the 

 fishing area whereas prior to this time some areas 

 were surveyed using fixed station line transects 

 not covering the whole area. The number of sets 

 used in the presentation of the distribution is 

 shown in Figure 3. 



Ages were determined from otoliths (Powles 

 and Kennedy 1967). Age composition is presented 

 for males and females separately by division. 

 Comparisons of age composition were made for 

 Div. 2J for the periods 1973-78 and 1979-83, 

 Div. 3K for the periods 1970-78 and 1979-83, and 

 Div. 3L for the periods 1968-78 and 1979-83. For 

 Div. 2J and 3K most data were collected during 

 the last quarter, whereas for Div. 3L most data 

 were collected over the last half of the year. How- 

 ever, considering the extent of the data and the 



very slow growth rate of witch flounder (Dower- 

 ing 1976), slight differences in the timing of data 

 collection are not sufficient to invalidate compari- 

 sons among divisions. 



Growth (cm) was expressed in terms of log-log 

 regressions (Log^, Length = a + b log^. Age 

 (years)). Growth curves were computed for each of 

 the age compositions stated above using data for 

 each fish and not mean length at age. Differences 

 in weight at age were then calculated between 

 the earlier period and later period for each divi- 

 sion by applying the length-weight relationship 

 of Bowering and Stansbury (1984) to the observed 

 mean length at age. 



The maturity rates were calculated as the 

 length (cm) and age (years) at which 507f offish 

 were mature (M50) as determined by probit analy- 

 sis according to the method of Bliss (1952) as ap- 

 plied to witch flounder in the Gulf of St. Lawrence 

 by Bowering and Brodie (1984). The results are 

 presented with 957^ fiducial limits. Males for Div. 

 2J were not included in the calculations due to too 

 low numbers of immature fish in the samples to 

 allow for a significant probit analysis. 



RESULTS 



Geographic Distribution and 

 Relative Abundance 



Witch flounder was caught throughout the 

 stock management zone of Div. 2J, 3K, and 3L 

 from the northern tip of Hamilton Bank to the 

 northern half of the Newfoundland Grand Bank 

 (Fig. 4). Catches were insignificant in the Hamil- 

 ton Bank area of Div. 2J with mean numbers per 

 30-min set generally <1.0. The general area of 

 highest abundance in this division was to the 

 south in the Hawke Channel. Here mean num- 

 bers per set ranged from about 2.0 in rectangle 

 Q16 to 11.0 in P17, the highest in Div. 2J (Fig. 4). 

 In Div. 3K, mean numbers per set per rectangle 

 were considerably higher than those in Div. 2J. 

 For many rectangles the mean numbers per set 

 were in excess of 10.0. The general areas of 

 highest abundance occurred along the deep 

 waters around Funk Island Bank where mean 

 catches ranged from about 13 to 53 fish/set. The 

 highest density occurred in rectangles P21, P22, 

 Q22, Q23, and R24 where mean catches ranged 

 from 43 to 53 fish/set. In Div. 3L the only signifi- 

 cant catches occurred along the northern slope 

 and northeast cape of the Newfoundland Grand 

 Bank. In this area, catches ranged from about 2 to 



613 



