218 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
first grade is at a height corresponding to the level attained in 
the jar by a measured half pint of water; the succeeding grades 
are determined by the introduction of additional half pints of 
water. When constructing a scale the feed-tube should be 
stopped at the lower end, so it may displace an amount of water 
equal to the amount of eggs it will displace in practice. It is de- 
sirable to have a separate scale for each variety of eggs hatched. 
All measurements are made with the feed-tube in place, the 
water shut off, and the eggs allowed to thoroughly settle. All 
semi-buoyant eggs should beinthe jar several hours before 
measuring, so that they may swell to their normal size. The 
short leg of the square is placed over the top of the jar, the 
long leg hanging down, and the scale read from the point 
where the top layer of eggs shows in the jar. Eeference has 
been made to measuring the fish. It is possible to obtain by 
measurement an almost correct estimate of the number of fish 
produced by ajar of serai-buoyant eggs. Just before the time 
of hatching all dead eggs should be carefully drawn off, and 
when the first fish is seen swimming in the jar the eggs should 
be again measured. It is an ascertained fact that scarcely any 
semi-buoyant eggs die, under proper conditions, after hatching 
out has commenced, so that the last measure furnishes us a 
close approximation to the number of fish. 
No. 4. 
