70 



for Northsea-herriugs and also witli those of spring-herrings of 

 the Baltic — the elaborate reports over these herrings pnblished 

 by Heincke enabling us to do so — there remains uot the least 

 doubt that our Zuiderzee-herring dififers greatly from the first 

 mentioned herring, and bears great resemblanee to the second. 

 The position of the dorsal and ventral fins is highly charac- 

 teristic in this respect: in the spring-herring of the Baltic these 

 fins have a much more advanced position, and a larger index in 

 consequence, than is the case with the Northsea-herring, and 

 also with the auturan-herring of the Baltic. Therefore, whereas 

 the formulae of the last-raentioned herrings are combinations con- 

 taining 1 a, 2 a and 1 b, the spring-herring-formulae show com- 

 binations containing 2 b, 2 c, 3 b and 3 c. 



Under the 200 herrings of the Zuiderzee which I studied there are 

 only 88 (19'^/o), the formulae of which are combinations containing 

 1 a, 2 a and 1 b. Combinations containing 2 b, 2 c, 3 b and 3 c on the 

 contrary are represented under these 200 155 times (77.5^/o). I am 

 not quite decided whether we are to cousider those lO'^y which 

 represent the Northsea type as true Northsea-herrings which 

 entered the Zuiderzee along with shoals destiiied to spawn in 

 that sea, or as Zuiderzee-herrings, which in some respects still 

 display in structure their near relationship with Northsea-herrings. 

 As an illustration of the importance of this characteristic feature 

 the formulae of a few Northsea-herrings, which were taken in 

 the summer of 1887 off Peterhead by the nets of a Vlaardingeu 

 herring-logger, may be of use. There were only six of them in- 

 vestigated by me but they show the Northsea type in such a 

 remarkable way as to give me liberty to insert their formulae here. 



