Althou^ only Georges Bank was surveyed consistently with tl^e Albatross III, 

 occasional trips were made to adjacent areas (Subareas E, F, and Q in area XXII 

 and area XXIII) and in these locations the concentration of zero-ring haddock was 

 sometimes greater than that on Georges Bank. The abundance of zero-iing haddock 

 in all locations sampled is shown in figure 3 . 



It appears likely tliat the Georges Bank stock of haddock may be dependent, 

 at least in part, upon the immigration of immature fish from off the bank proper, and 

 that the St very young migrants may have been spawned either bn the bank or in the 

 GuLf of Maine . 



Some of the 1948 data are inconsistent with the 1949 and 1950 records. For 

 instance, the abundance figures for 2, 5,, 7, and 9-year-old fish in 1949 were greater 

 than those for 1, 4, 6, and 8-year-old fish in 1948. It is our opinion that the sampling 

 technique was not so good in 1948 as ia other years. Another iindication of this is the 

 lower number of species per tow and the greater number of tows with 0, 1, and 2 

 species in that year (Taylor 1953). The reason for this poorer sampling in 1948 is 

 not immediately apparent for, as seen in table 1, the number of tows by subarea, 

 depth, and bottom type were consistent with those of other years . One way in which 

 'the 1948 sampling technique differed from that of 1949 and 1950 was in the time ire- 

 quired to cover the designated area. In 1948, 97 days were required to sample the 

 area, while in 1949 and 1950, 23 and 53 days, respectively, were needed. Also, dur-, 

 ing the 1948 cruises towing speeds varied considerably which possibly contributed to 

 the sampling error. Because of this discrepancy oaly the If 49 and 1950 data were used 

 in estimating mortality rates and in determining mean values . 



Since the catch per unit of effort varied markedly from year to; year, the age , 

 composition was expressed as percentages of the total catch per tow in order to give 

 a clearer picture of the differences in the age composition of the catch during the three 

 years. These percentages are shown Ih figure 4. 



In 1948, more than 90 percent of the catch was made up of 0-3 -year-old fish, 

 with 2 -year -old fish predominant (37pere«it)o In 1949, over 90 percent of the catch 

 consisted of 1 throu^ 3 -year -old fish with 1 -year -old fish predominant (75 percent) . 

 The percentage of zero-ring fish in 1949 was negligible. In 1950, 89 percent of the 

 catch were in the throu^ 2 -year age group with 2 -year-old fish predominating 

 (49 percent). The relative hig^ percentage, as well as actual numbers, of zero -ring 

 fish in 1950 indicates a successful 1950 year class, and the low percentage and numbers 

 of 1-year-old fish, indicates the weakness of the 1949 year class. TTie slightly higgler 

 percentage of 5-year-old fish, which were found principally in subareas G and H in 

 1950, is not readily explainable from the age composition of the 1948 and 1949 catches. 



