AT THE PLYMOUTH LABOEATOEY. 305 



days. Before it began, one of the smelts jumped from the fresh-water 

 tank, over the barrier, into the sea-water tank adjoining. It was left 

 in sea water afterwards, and suffered no harm. The same thing 

 happened to a second individual on the first night after the change 

 began (March 5th). 



On March 10th this second lot of smolts was put into the large 

 tank with the others. This tank, which now contained forty-nine 

 smolts, was used throughout to accommodate the survivors as long 

 as the salmon remained at the Laboratory. Its inside dimensions 

 are 15 "7 feet in length, 9 feet in width, and 4-4 feet in depth of water, 

 giving a capacity of 621 cubic feet. It was fed by eight jets, giving 

 a total normal inflow of about 385 cc. per second, and its position, 

 partly screened from the direct light by a dark-coloured blind, is such 

 that its lighting may be described as moderately low and constant. 

 The back, the ends, and the floor of the tank are formed of slate, 

 and all uprights or other portions of the framework are similarly 

 dark in colour. Air was supplied entirely by the force of the water 

 from the several jets striking the surface, which was sufficient to 

 carry the fine air-bubbles nearly or quite to the bottom of the tank. 



Feeding. — The smolts were fed twice a day, and often three times. 

 On the first day or two the food given them was broken biscuit and 

 prepared fish and flesh foods previously soaked, of the same kind as 

 that used at Endsleigh. This was then varied with raw bullock's 

 liver cut into small pieces, and the preference for this latter soon 

 became so strong that the other was very shortly discarded altogether. 

 This was the case with both lots of smolt. When the transference 

 to sea water was about half completed, it was found that the common 

 inter-tidal marine worms of the genus Nereis (N. diver sicolor) were 

 taken very readily. About the time of the completion of the change 

 s. distinct loss of appetite was shown by several of the smolts for 

 a few days. But, on the whole, feeding was fairly good during the 

 interval, and this was particularly the case with the second lot, where 

 the transference was brought about more rapidly. After the change 

 to sea water, liver was taken with the same readiness as previously. 

 Nereis was at times taken when liver was refused, but beyond this 

 little preference was shown between the two. 



Experiments in feeding with marine animals other than Nereis gave 

 entirely negative results. Among others, trials were made on several 

 occasions with the following species : — * 



* It is to be regretted that herring was never tried. But the keeping of young 

 herring alive for any length of time, even when they are obtainable, is a matter of very 

 great difficulty. 



