SHA-FISHERIES LABORATORY. 163 
There is a great deal of valuable material in these 
statistics which, whether or not it leads to any definite 
conclusions, will at least help us to see what further 
observations are required, and what measure of success we 
may hope to attain in the proposed census of the sea. 
HatTcHING WoRK AT PIEUL. 
(ANDREW SCOTT.) 
I. FisaH HATCHING. 
At the conclusion of the spawning season of 1898 it was 
found that, in order to deal with large quantities of fish 
eggs we should have to increase the capacity of our hatching 
apparatus. After careful consideration it was decided that the 
“ Dannevig system,” which has given satisfactory results in 
Norway and in Scotland, should be adopted. This appar- 
atus for incubating fish eggs consists of a series of movable 
boxes, each of about a cubic foot in capacity, all floating 
in separate compartments of a tank. The bottom of each 
box is covered with fine silk gauze or hair cloth, the 
meshes of which are of sufficient size to allow the water 
to pass through freely and yet keep back the smallest size 
of egg. Hach box is placed in a separate water-tight 
compartment, to one side of which it is hinged. The 
water enters the box over a small shoot, and passes out 
into the compartment through the perforated bottom and 
then overflows into the next box. The apparatus is 
constructed in sets, each set consisting of a double row of 
five boxes in their compartments. At the ends of each 
row there are smaller compartments, one at the top for 
the distribution of the inflowing water, and one at the 
bottom for collecting the waste, which is led away to the 
nearest drain by an overflow pipe. The whole apparatus, 
when placed in working order, is set on the floor at a 
