t JIAI-. in.] NOTES ON STORMONTFIELD EXPERIMENTS. Ill 



than 80,000 ova could be obtained, in consequence of the 

 unfavourable state of the river for capturing gravid salmon. 

 Peter Marshall has proved a most able pisciculturist. The loss 

 of eggs under his management forms an almost infinitesimal 

 proportion of the total quantities hatched at Stormontfield. 



Mr. Buist has favoured me with the following notes, which 

 were compiled from his day-books at an early stage of the 

 Stormontfield experiments : 



" 1. Of the marked fish which were liberated from the 

 pond at Stormontfield, four out of every hundred were re- 

 captured, either as grilse or salmon. 



" 2. We find that more than 300,000 fish were reared in 

 the pond, and allowed to go into the Tay. Thus forty fish 

 out of every thousand were re-captured ; and as 300,000 were 

 in all liberated, it follows that 12,000 of the salmon taken 

 in the Tay were pond-bred fish. But as the fish did not all 

 go away in one year, this 12,000 must be distributed over two 

 years. 



" 3. We find the average number of salmon and grilse 

 taken in each year is 70,000. It follows, then, if there be any 

 truth in figures, that nearly one-tenth of the fish taken in the 

 Tay for the last two years were artificially bred. This is 

 equivalent to a rise of 10 per cent in the rental of the fishings ; 

 and such we find is the result. 



" It may be urged that if the salmon from which the ova 

 were taken had been left at liberty, the result would have 

 been the same ; but this we know could not have been the 

 case, for, according to a careful calculation made by Mr. 

 Thomas Tod Stoddart and others, each pair of salmon, 

 although they produce upon an average 30,000 eggs, do not 

 rear above five fish. Three female fish, if every egg they 

 deposit was to produce a salmon, would produce all the fish in 

 the Tay. When left in their natural state, 30,000 ova produce 

 four or five fish fit for the table ; whereas the same number of 



