242 Appendices to Fifth Annual Report 



' the greater or less salinity which made the difference, as the sudden 

 ' transition from the Salter to the fresh water. 



' This was confirmed by the fact, that a number of deep sea cod kept in 

 ' confinement in the wharf from the one year to the other, were found to 

 ' contain eggs in excellent condition for hatching. 



' The results of this observation was, that all the brood fish purchased 

 ' for this year were brought to tanks in the hatchery, and thus is 

 ' obtained a greatly increased quantity of roe, which will be easily under- 

 ' stood when I state that the number of fish in the best spawning time 

 ' was about 150, against 400 to 500 in the previous year; and that, 

 ' nevertheless, and in spite of the increased loss through the lower tempera- 

 ' ture during last winter, a larger number of fry were hatched than ever 

 ' before. The result of the low temperature of the sea water during winter 

 ' was the great loss of eggs in the hatching boxes, the loss being 52 '9 per 

 ' cent., against 39 '3 per cent, in 1885. This is not to be understood as if 

 1 the cold was directly hurtful to the eggs or the fry, but rather that the 

 ' development goes on so much more slowly, which necessitates the eggs 

 ' remaining longer in the apparatus where from knocking against the bottom 

 ' and from other causes they greatly suffer. 



' The longest previously observed hatching time for cod eggs was 26 days, 

 1 while this year it lasted 42 days, though the average difference in tem- 

 1 perature was only 2° '5 R. 



1 With a temperature of + 4° R. it occupies 22 to 24 days, and as the 

 ' fry appear strongest after such a hatching period, it is reasonable to 

 ( suppose that this is the most natural time for the cod. 



' If the water remains 6° warmer, e.g., the development is accomplished 

 4 in 15 to 18 days ; but the fry are tender and weak, which is also the 

 ' case, though in a less degree, when the temperature sinks to+ 2° or lower.' 



As to rearing the fry, Dannevig says : — ' I have found that to rear the 

 ' fry successfully, it is necessary, not only that the temperature should be 

 1 suitable, but that the density of the water should be at least 1*022, so 

 ' that the tender young may keep clear of the bottom without too great 

 ' exertion. "When the specific gravity of fry is greater than that of the 

 ' water, they have a tendency to sink, and as their swimming powers are 

 ' small in the first stage of their growth they sink, in spite of all efforts, 

 ' to the bottom, where their destruction is certain. The only way to get 

 ' the necessary salinity, and at the same time have the water completely 

 ' under control, was to build a great basin on land, and fill it with the 

 ' help of steam pumps. A basin was constructed in 1885 (measuring — 

 'greatest length, 43 metres; greatest breadth, 30 metres; and greatest 

 ' depth, 5 metres), with a capacity of 2500 cubic metres. It is situated 

 ' in a natural hollow, and has, on the south side, a wall 40 metres long, 

 ' and on the north-east a similar wall 15 metres long. 



i After the basin was cleaned and supplied with several sea plants, 

 ' about 500,000 cod fry (hatched 27th April) were introduced on the 

 ' 3rd May, and the pumping was then continued till the basin was full, 

 1 and afterwards p.t suitable intervals.' 



The results of this experiment will be best gathered from the following 

 extracts from Mr Dannevig's diary : — 



' May 3-8. Saw the fry every day, and always most in the east end of 

 ' the basin, always swimming. 



' 13. Introduced 5000 young flounders (Pleuronectes lemandu) into the 

 ' basin. 



' 18. The majority seem to keep near the supply pipe, probably for 

 : the sake of the fresh water streaming in. 



' 21. Townrds evening observed the greatest number at the west end of 



