as Canada, U.S.S.R., Poland, German Democratic 

 Republic, and Federal Republic of Germany, who 

 have prosecuted this fishery for over 2 5 yr from Davis 

 Strait to the northern Grand Bank and, more recent- 

 ly, the Gulf of St. Lawrence. 



The only published information on age and growth 

 of Greenland halibut in the northwest Atlantic was by 

 Bowering (1978a), who reported growth in terms of 

 arithmetic linear least squares regressions for select- 

 ed areas where data were available. Walsh and 

 Bowering (1981) presented information on the valid- 

 ity of field observations on sexual maturity in female 

 Greenland halibut using histological techniques; the 

 data, however, were confined to one area of northern 

 Labrador. To my knowledge, this is the only available 

 information on sexual maturity of Greenland halibut 

 in the northwest Atlantic. 



This paper presents a detailed account of the age 

 and growth of Greenland halibut in the Canadian 

 northwest Atlantic from Baffin Bank to the northern 

 Grand Bank and into the Gulf of St. Lawrence. An 

 analysis of sexual maturity data on female Greenland 

 halibut throughout the region (except the northern 

 Grand Bank) is also discussed. 



MATERIALS AND METHODS 

 Age and Growth 



For examination, the data were organized according 

 to seven geographical areas: 1) Baffin Bank; 2) Sag- 

 lek Bank; 3) Nain Bank; 4) Hamilton Bank; 5) North- 

 east Newfoundland Shelf; 6) northern Grand Bank; 

 and 7) the Gulf of St. Lawrence, since these are dis- 

 crete fishing areas where data were available (Fig. 1). 

 All data were collected during random- stratified re- 

 search vessel surveys for groundfish (Pinhorn 1972) 

 stratified by depth. The Baffin Bank data were col- 

 lected by the French research vessel Cryos in Oc- 

 tober 1977. This vessel is a stern trawler, 46 m long, 

 which uses a Lofoten bottom otter trawl with a 5 mm 

 small mesh liner in the cod end. All sets were of 30- 

 min duration during daylight hours at a towing speed 

 of 3.5-4.0 kn. Data from sets in which the gear was 

 badly damaged or other reasons considered to inter- 

 fere with normal retention of the catch were not used. 

 All other data were collected by the Canadian re- 

 search vessel Gadus Atlantica. The Gadus Atlantica 

 is a stern otter trawler, 85 m long, which uses an En- 

 gels high-rise bottom otter trawl with a 12 mm small 

 mesh liner in the cod end. All sets were of 30-min 

 duration at 3.5-kn towing speed, and fishing was car- 

 ried out on a 24-h basis. 



The fish ages were determined from the left sac- 



FISHERY BULLETIN: VOL. 81, NO. 3 



cuius otoliths which, in the case of Greenland halibut, 

 were more suitable for age determination, since the 

 annuli were spaced more evenly and were more dis- 

 tinct. Otoliths were ground on the convex surface ex- 

 posing the center more clearly and were placed in a 

 black watch glass containing ethanol in order to facili- 

 tate reading. Age frequencies were compiled for 

 males and females separately using age-length keys 

 for each area with the distributions calculated as 

 numbers per thousand at age of the total catch. 



For ease of statistical comparison, growth was ex- 

 pressed in terms of semilog curves (length = a + bin 

 age) similar to that used by Bowers (1960) for witch 

 flounder and suggested in Roff (1980). Growth curves 

 were computed separately by sex for each area. The 

 growth curves were weighted by the number of obser- 

 vations at each age, since many of the older age 

 groups were based upon as little as one observation, 

 and the age reading of very large fish is often ques- 

 tionable (Lear and Pitt 1975). 



Otoliths from 30 males and 30 females of the 1972 

 year class were selected from each area and ex- 

 amined to back-calculate length at age. Since the 

 Baffin Bank data were collected in 1977 and the Gulf 

 of St. Lawrence data were collected in 1980, only the 

 1972 year class was abundant enough in the samples 

 for comparison. Subsequently, in order to standard- 

 ize the number of ages for consideration, calcula- 

 tions could only be made up to age 5. Otoliths were 

 measured by means of a drawing tube attached to a 

 binocular microscope, using a technique similar to 

 that described by Moores and Winters (1978). An an- 

 nulus was considered to be the width described by an 

 opaque (summer) zone and a translucent (winter) 

 zone accounting for a growth increment in a par- 

 ticular year. The ratio of otolith length to total fish 

 length was used to back-calculate the average length 

 at each age, using the direct proportionality method 

 described by Lea (1919). In order to validate the 

 direct proportionality method as it applied to Green- 

 land halibut, a linear least squares regression was 

 performed on a random sample of 1 2 3 measurements 

 of total fish length to total otolith length (3 

 measurements per 1 cm group from 10 to 50 cm). 



A covariance analysis (Zar 1974) was performed on 

 the back-calculated data for males versus females in 

 order to determine if there were significant differ- 

 ences among the regression coefficients (slopes) or 

 the adjusted means (y- intercepts). Where no dif- 

 ferences between the sexes occurred the sexes were 



FIGURE 1. — Map of study area with corresponding 

 major names mentioned in text. 



600 



