The relation of fecundity to age was investigated for two age groups, five- and six -year-old 

 herring, as shown below: 



6--9--12--15--18--21--24--27--30--33 thousand eggs 

 5-year old 2 12 



6-year old -- -. 2 



M 5 years = 16,380 

 M 6 years = 21,620 



The average fecundity of the 5-year-old herring is 16, 380 eggs, of the 6-year-old herring 

 21,620 eggs. 



Approximately the same figures were obtained by Esipov for 6 Murmansk herrings (see 

 Table 1). Unfortunately, the age of the herrings is not given in this table, and we have therefore 

 no opportunity of comparing our results completely with those of Esipov. 



The great variation in the data on the fecundity of the North Atlantic herrings is, according 

 to Esipov, explained by the fact that up to the present no systematic investigation has been carried 

 out on the racial composition of the herring shoals . 



We do not agree with this conclusion, and find it necessary to add two more reasons. 



First, various authors state the fecundity without taking into account the age of the fish. In 

 the Murmansk herring there are considerable variations in relation to the age of the fish. 



Second, some authors, for example, Mitchell (100), are of the opinion that all the ovocytes 

 present in the ovary will mature during one single season. Our investigations have given the oppo- 

 site result. Of the total crop of ovocytes only a known part, from which the size of the fecundity 

 is determined, will be spent during the next spawning period. 



If one would agree with Mitchell's arguments we had to expect that only empty follicles were 

 seen in the histological sections of the spent ovary. All our preparations of ovaries in stage VI 

 show a quite different picture. Among the empty follicles many young ovocytes without yolk are 

 present. TTiese ovocytes belong to the crop of the next season, and will not ripen and be extruded 

 until the following spawning period. Such ovocytes were not taken into consideration in the deter- 

 mination of the fecundity of the Murmansk herring, 



Mitchell counted all the ovocytes, and, accordingly, over-estimated the fecundity of the 

 Atlantic herring, with a maximum number of 280, 000 eggs. 



In fish with total spawning, in this case the Murmansk herring, the number of ripening eggs 

 must be the same in the various stages of development. At the end of the summer a certain number 

 of ovocytes begin to ripen, the ovaries gradually piass into stage III. The same number of ovocytes 

 evidently reaches the IV and V stages and will be extruded in the approaching spawning period. 



From our computations there are on an average 17, 113 eggs in ovaries of stage III, 16, 660 

 eggs in ovaries of stage IV, and 20,670 eggs in stage V. The absolute number of eggs in herring 

 in stage III and IV does not differ essentially from each other. The difference is only 0.98%. 



However, in stage V, the absolute number of eggs in the ovary is 8.3% above that of stage 

 in. This discrepancy is probably caused by an error in the computation of the eggs, and also by 



209 



