12 U. S. BUREAU OF FISHERIES 
in the haddock populations over the entire period, which is marked 
by three periods of scarcity and two of abundance. Periods of 
abundance apparently result from a series of years with a high sur- 
vival of young, while periods of scarcity result from a series with a 
poor survival. The years since 19380 are an exception, for since then 
the level of abundance has remained relatively low in spite of the 
consistently good production and survival of young. This condition 
appears to be the result of the increased intensity of the fishery in 
recent years. Cycles of good and poor survival of young appear to 
be related to the abundance of adults on the nursery grounds, the 
most favorable level being approximately that which prevailed in 
1922 to 1924, 1929, and 1936. 
These developments indicate that we are now approaching the 
stage where it will be possible to explain the causes for the great 
cycles of abundance which have been observed in the haddock fishery. 
It appears that the cycles of abundant and poor year classes are 
related to the varying abundance of large fish on the banks. Poor 
production or survival of young may result either from an over- 
abundance of adults on the nursery grounds or from too great a 
reduction in the spawning stock. If the completed analysis of the 
material substantiates these indications, it will provide information 
of tremendous value toward development of a consistently good yield. 
Population composition—A considerable amount of information 
concerning the abundance and distribution of haddock was obtained 
from the trawled material taken on the spring A¢lantis trip. One- 
and two-year fish were abundant, but older fish were relatively scarce. 
The 1-year fish were distributed about as in previous years, but 2-year 
fish, which previously were found in great numbers on Central 
Georges, in 1938 were almost entirely absent from this region. They 
were found in large numbers only on southeastern Georges and in 
South Channel. The latter location was particularly unusual, as but 
few fish of this age had been found there in previous years. 
Tagging.—FEarly in this investigation a series of tagging experi- 
ments was carried out, using haddock held in tanks and live cars to 
determine the most satisfactory tag for this species. A variety of 
special tags were tested in addition to the disk type (plaice label) 
and strap tag used by other investigations. The experiments indi- 
cated that a button type was most successful with the disk type al- 
most as good. This work was followed by field experiments in 1932- 
83, Involving the tagging from otter trawlers of 440 haddock with 
button tags and 52 with belly tags, and from line trawlers of 296 
with button and 117 with belly tags. No returns were received from 
any of these experiments except for one button tag from the line- 
trawl tagging. In view of the successful retention of tags in the 
live-car experiments, it appeared that the failure of the offshore 
work was the result of high mortality due to the gear and depth. 
Experiments next were tried from small line trawlers in shoal water 
along the Maine coast. In this attempt, 296 button and 117 belly 
tags were attached. ‘There were 6.4 percent of the button and 6 per- 
cent of the belly tagged fish recaptured and reported. 
Owing to limited personnel and funds, no further work could be 
done on this subject in spite of its importance until Dr. G. A. Rounse- 
fell took charge of the work in 1938. In June of that year, 265 
haddock were tagged with button tags, 266 with disk tags, and 100 
