In the summer montiis only 2 -year and older haddock are tabulated. The 

 reason for this is that the commercial boats which used gear having approximately 

 a 3 -inch mesh cod end did not sample zero and one-year-old fish adequately. Fish 

 of this size that were retained in th.e net were discarded at sea and hence not included 

 in the catch statistics. During the spring months still older fish are discarded in- 

 cluding most of the 2 -year -olds and a portion of the 3-^ear-olds . This increase in 

 the culling age during the spawning season is due in part to the greater availability 

 of older haddock in the area fished by the commercial boats and in part to the fact 

 that a fish of a given age is smaller in the spring when the "haddock year" begins 

 than in the summer. This increase m the catch of older haddock during the spring is 

 evidenced by the landings of the commercial fleet over a number of years and is due 

 to the migration of older fish into shoaler water as shown above. Thus, in the age 

 group data shown in table 18 and figure 16, two groupings were made of the spring 

 data in order to demonstrate the differences and similarities in the age composition 

 as determined from commercial and census figures. 



The age composition of the commercial and census data are similar during July 

 and August of 1949 and 1950 for all ages and during April of 1950 when only 4-year- 

 old and older fish are considered. The above comparison greatly attests to the reli- 

 ability of the census data and its value in population studies when account is taken of 

 the fact that these census data are based on a comparatively few completely random 

 sam.ples compared with that of the effort expended by the commercial fleet, which 

 effort is necessarily biased in that it is concentrated on certain age groups and re- 

 stricted areas . 



Mortalities based on such a limited sampling of a population are necessarily at 

 best rough approximations, but it is interesting to note that the mortalities determined 

 from the weighted census data and commercial data for July and August of 1949 and 

 1950 are quite similar. These mortalities are given in table 19 and it is seen that 

 the mortality for all ages appear exceptionally high which lends credence to similar 

 figures determined from census data based on all subareas as shown in table 2 . 



Mortalities determined from census data for the period of July and August of 

 1949 to April of 1950 were consistently lower for all ages than those determined for 

 the longer period of July and August of 1949 to July and August of 1950. However, 

 similar figures based on commercial data give negative values for 5 year and older 

 haddock for the 9 month period. These negative results are due largely to the 

 migration of older fish during the spawning season from deep-water areas little fished 

 by the commercial fleet into shoaler water where the greatest effort of the commercial 

 fleet is expended. 



The average mortality rates determined from the yearly abundance data of the 

 commercial fleet obtained over a 20 -year period show wide fluctuations throughout the 



43 



