8o MARKETABLE BRITISH MARINE FISHES 



to present the same stage of development, so that it is evident 

 that after spawning has once commenced there will be no long- 

 continued production of ripe eggs in succession, but that a large 

 number will be shed in a short space of time, while in the sole, and 

 also in the gurnard, eggs are found in all degrees of develop- 

 ment from the most primitive stage onwards. That the eggs 

 of the sole are shed in very small numbers at a time has been 

 strikingly confirmed by the observation of the soles in the act 

 of spawning at the Plymouth Laboratory. ^ Only a small number 

 of fertilised eggs were obtained each day from a tank contain- 

 ing a large number of ripe fish, and the eggs appeared to the 

 observer to be shed one at a time at intervals. 



It is evident that if only a few ripe eggs are shed into the 

 water at one time the quantity of ripe milt which must be 

 liberated to fertilise them will be very small, while if .several 

 hundreds or thousands of eggs are expelled within a few seconds 

 or minutes within the same time a hundred or thousand times as 

 much milt will be required as in the other case. If the forma- 

 tion of the milt occupied as long a time as that of the eggs, there 

 would be no apparent reason why the proportion in size between 

 the male and female organs should differ greatly in fishes that 

 spawn rapidly and those that spawn slowly. In both roe and 

 milt there would be a bulky accumulation of eggs and sperms in 

 process of development. But in the egg there is a gradual 

 formation of yolk which extends over weeks or months, while 

 the sperms are minute, contain a small quantity of material, and 

 can be produced in a few hours. In this fact lies the reason why 

 a slow spawning roe is so much larger than the milt of the same 

 species. If we suppose that 5,000 ripe eggs are shed by a sole 

 per day in the spawning period, and that the growth of an egg 

 takes three months, then a sole producing 500,000 eggs would 

 take TOO days to get rid of all its eggs, and on the day it com- 

 menced to spawn very nearly all the eggs would be present in 

 the roe in all stages of development or formation. On the other 

 hand it is extremely probable that the ripe milt necessary to 

 fertilise 5,OCO eggs can be produced in one day or a few days, 

 and therefore at the commencement of spawning the male organ 

 would contain not all the milt to be shed in the spawning 



1 G. W. Butler, B. A. : "On the Spawning of the Common Sole in the Aquarium,"' 

 M. B. A. /ournal, vol. iv., No. i, p. 3. 



