68 THE TUEBOT AND TUEBOT CULTUEE 



16 inches. At the end of a few weeks the wild fish began to feed. 

 They were then given very short rations — i. e. a bit of plaice 

 about as big as the hand once a week to each fish. Anthony 

 noted that overfeeding of the fish in previous experiments had 

 probably been one reason why turbot had so ojten refused to 

 spawn in captivity. To keep the basins clean a conger eel and 

 a dogfish were kept in the basins as scavengers. 



The first eggs were found on the 18th July ; and the fish 

 spawned again on the 18th, 21st, 28th, 29tli July, and on 

 3rd August. There were ' thousands and thousands ' of eggs, 

 all normal and normally fertilized. A certain number of eggs 

 were transplanted to lO-gallon glass barrels (Fig. 5) in which 

 the water was kept in motion by a revolving disk driven by 

 a small one-fortieth horse-power hot-air motor. This was 

 a modification introduced by Bietrix and Fabre Domergue into 

 an apparatus invented by Browne of Plymouth, who agitated his 

 water by means of the rise and fall of a plunging disk. Part of 

 the water was renewed for ten minutes several times a day — 

 and dead eggs and other matter were cleared out from the 

 bottom. The best temperature was found to be 59° F. to 68° F., 

 the latter being the highest hmit of safety. The eggs hatched 

 out between the sixth and eighth day after spawning — the same 

 period which was observed in this country by Holt, and by 

 Fulton and Harold Dannevig at Dunbar. 



Tivo or three days after hatching the larvae were fed with 

 live ' plankton ' collected at sea, and carefully sifted in the 

 laboratory. Plankton was placed in the barrels once a day and 

 in great abundance ; the revolving disk of course maintained 

 a constant current in the barrel, and kept the plankton alive, 

 so the young fry always had their natural food at hand directly 

 they began to feed, as they did long before the yolk-sac nm.s 

 absorbed. 



The yolk-sacs- finally disappeared on the fourteenth or fif- 

 teenth day. It is noticeable that at Dunbar (where the larvae 

 were not fed) this stage was reached on the seventh day. This 

 suggests to the writer's mind that the Dunbar fish died of 

 starvation. Probably the yolk-sac does not contain in itself 

 enough food to carry what trout-breeders call ' alevins ' through 

 the larval stage. Lebour notes that all the ' large-mouthed ' 

 flat fishes, soles, turbot, brill, dabs, and plaice (?), start feeding 

 on copepods almost as soon as they are hatched. Probably the 

 contents of their sacs must be supplemented by other food — 

 in the case of turbot within a day or two after hatching — or 

 the fry will be weakly by the time the yolk has gone. In other 

 words, the babies must be weaned gradually ; in Scotland they 



