231 
cient aeration. The eggs develop in these boxes and the 
larve after being retained for some time are taken out to 
sea and “planted” in a suitable locality. It will be 
obvious from what we have stated above with regard to 
migrations that the larve must be set free in such a place 
that the prevailing winds and tidal drift will carry them 
to a nursery (which must be present in the area dealt with) 
in which the conditions for further development are suit- 
able and in such a time that the assumption of the 
demersal habit will coincide with the arrival of the larve 
on the nursery grounds. The ultimate aim of the hatch- 
ing operations is to rear the larve through the period of 
their metamorphosis under artificial conditions. It has, 
however, been found extremely difficult to rear even a 
small proportion through the period referred to since great 
numbers die during the period immediately following the 
absorption of the yolk sac. In practice, therefore, the 
larvee are set free before this mortality has seriously com- 
menced. ‘The Scottish Fishery Board, in addition to their 
utilitarian object of adding to the number of young fish 
in the sea, have since 1897 been devoting attention to the 
interesting experiment of placing some millions of fry 
yearly in one limited area where they have reason to think 
they may be able to test the result by periodic observations 
on the young fish fauna. In the period between 1894 and 
1899 inclusive the Board set free 136 millions of arti- 
ficially hatched larvae, and since 1897 these have been 
planted in Loch Fyne, the area chosen for experiment. 
The argument for the utility of sea fish hatching 
rests to some extent on the hypothesis that the period 
during which the fish are being dealt with in the hatchery 
is that during which the mortality in nature is greatest. 
Since, in the hatchery, the eggs and larve are safe from 
enemies or prejudicial changes in their physical surround- 
