and shad are liable to numerous destructive agencies; that only a small
percentage of the eggs laid under natural surroundings ever hatch, and
that the young are subject to heavy mortality up to the time when they
leave the river and enter the salt water. Probably 5 per cent would
be much too large an estimate of the number of salmon eggs which in a
state of nature produce fish that reach the ocean. Fish-culture, on
the other hand, hatches 95 per cent of the eggs and raises 75 per cent
of the fry to the age of yearlings. Of 206,350 Atlantic salmon eggs
obtained in 1895 at the government station at Craig Brook, 206,109
were hatched and 151,761 yearling fish were liberated in the fall.
The percentage of eggs hatched was thus 99.88 and the percentage of
yearlings raised was 78.39. This is sufficient ground for interfering
with the salmon even after they have reached their spawning-beds, and
justifies the establishment of hatcheries in the headwaters of the
Penobscot, provided the supply of fish in any section is large enough
to insure a reasonable take of eggs.

No examinations of the upper tributaries of the Penobscot thus far made
have disclosed the existence of any stream on which the construction of
a branch salmon hatchery is warranted, owing to the few salmon
obtainable. The matter deserves further investigation, however, and
will receive due consideration at an early date. It is thought that a
satisfactory supply of fish may be secured by constructing a dam or
rack which will intercept fish in the main stream and lead practically
the entire run into one tributary, where they may be retained.

The operation of a branch salmon hatchery in the river above Bangor
would of course depend on the successful working of the fishways and
the enforcement by the State of the anti-poaching laws.