14. U. S. BUREAU OF FISHERIES 
made of brass-wire cloth, 60 meshes to the inch, tacked to a heavy 
frame. The cloth is pulled very tight before being fastened, so 
that it will present a smooth surface to the air or water jets. The 
screens should never be painted. ‘The screen frames are held in 
place by bolts that pass through the frames and into the projecting 
cleats in the trough. 
It is exceedingly important that these screens be kept clear of the - 
accumulation of egg shells and impurities in the water. This is 
accomplished preferably by means of air jets, although at some sta- 
tions water jets are used. The air jet can easily be arranged for by 
installing an air pump and carrying the connecting pipe along the 
side of each tank on the inside of the screen, thence at right angles 
parallel to the screen and about 1 inch distant from it. Half-inch 
galvanized iron pipe is used in the construction of air jets. The 
cross pipe should be perforated on one side with holes one-thirty- 
secondth inch in diameter and 3 inches apart, the perforations open- 
ing toward the screen and upward at an angle of about 45°. When 
the air is turned on an apparently solid mass of bubbles will rise 
along the whole surface of the screen. 
With such an arrangement the screen will need no attention for 
hours or even days at a time, whereas without air jet one or more 
men must be employed to keep the screen clear. Moreover, many of 
the fry are unavoidably killed by being forced against the screens 
and by the men in keeping the screens free. ‘The thorough aeration 
of the water indirectly accomplished by use of the air jet is very 
beneficial when large numbers of fry are passing. over, and twice as 
many can be handled in troughs thus equipped. 
The construction of a water jet is very similar to that described 
above except that a somewhat larger pipe is used and the holes in 
the cross pipe are larger. The water-jet system for keeping screens 
clear is not considered as efficient as is the air jet, one of the princi- 
pal objections being that it adds more water in the trough without 
materially increasing the aeration. 
The absorption of the food sac is governed by the period of incu- 
bation and by the water temperature. If 28 days have been required 
the sac will be absorbed in from five to six days, while if a shorter 
period, say 14 days, has been required approximately 10 days will 
elapse before the sac has disappeared entirely. Within a day or two 
after its complete absorption cannibalism will begin. 
PREVENTING COHESION 
Many experiments have been made from time to time to determine 
the best means for overcoming the tendency of pike perch eggs to 
eohere. This may be accomplished by stirring the eggs constantly 
from the time water is added until it fills the egg, when cohesion 
ceases. Time is lost, however, and a large percentage of the yolks 
inevitably are ruptured. Another method is to allow the eggs to 
agelutinate and stand thus until fully water-hardened, afterwards 
separating them by gently rubbing between the hands; but this also 
sacrifices time. 
The date of the first use of foreign inert substances to prevent 
cohesion is uncertain. Fine clay dust and clay in solution have been 
