fluctuations. Initially, it was believed that estimation of the number of vertebrae In 100-200 speci- 

 mens of herring was sufficient to establish the race of the herring thus examined. Later on, 

 however, it was found that the mean index of the number of vertebrae in herring from different 

 regions varies from one year to another. Thus, observations conducted by Prof. M. P. Somov on 

 herring from the Barents Sea over the course of almost twenty years showed that the mean number 

 of vertebrae changes considerably from one generation to another. The Norwegian scientist 

 Rasmussen (62) arrived at analogous conclusions. Thus, it became obvious that when considering 

 the mmiber of vertebrae as a criterion for determining the race, it must be taken into account 

 that this characteristic varies from one generation to another. 



Morphological investigations yielded certain results only in the initial stages of the study 

 of distribution and migrations of herring. For present scientific demands and practical require- 

 ments this method is entirely unsuitable . 



In the first decade of our century, an attempt was made to follow the movement of schools 

 of Norwegian herring hatched in 1904, the scales of which bore a mark due to poor feeding con- 

 ditions in 1906. Meek (59) drew a map of the migrations of young Norwegian herring on the basis 

 of the distribution of the herring bearing these marks . This map is presented in his work, as 

 well as in the work by P. Yu. Shmidt (32) entitled "Migrations of Fishes" and in a number of text- 

 books on ichthyology . 



More recent investigations on the rate of growth of young Norwegian herring carried out by 

 Ottestad (60), led him to conclude that a sharp drop in the increment in the third year of growth is 

 a relatively frequent phenomenon and can be attributed to herring lingering in the littoral zone . 

 Thus, Ottestad questioned the observations and map of Meek. At present it is difficult to give an 

 opinion on this subject without a more thorough analysis of all factual material. It is almost im- 

 possible to assume that the unfavourable feeding conditions of 1906 only affected the extensive 

 Norwegian herring region. If this is the case, it is only logical to assume that these "marked" 

 specimens might also have been found among other races of herring, among herring from the 

 North Sea in particular. Personally, we consider the map of migrations suggested by Meek as 

 rather open to question and it will be discussed further. 



As a result of the work of Norwegian scientists a method of "reading" the scales of herring 

 has been developed, and this permits determining the origin of herring (Northern and Southern 

 type), the season during which they move into the open sea, as well as the age at which sexual 

 maturity occurs. The Barents Sea herring investigations of P. A. Murashkintseva (18) showed that 

 the criterion on which the Southern and Northern types of rings on the herring scales are dis- 

 tinguished is subjective. In a great number of specimens, we observed rings of transitory type. 

 Moreover, the Northern and Southern types of rings may form not only due to the physical and 

 geographic conditions characteristic of the latitude inhabited by the given herring, but due to local 

 conditions as well (such as littoral and open sea zones, where the feeding conditions vary from one 

 year to another, etc.). 



It is extremely difficult to solve the problem of migrations by examining the structure of 

 scale rings . This index is Important only in the study of dynamics of the populations, in fore- 

 casting the sexual maturation and spawning of adult fish schools . 



The differences in the rate of growth are also of little importance in the solution of the 

 problem of migrations. The reason for the latter is that there are considerable individual fluctu- 

 ations in the annual rate of growth, and in the peculiar circumstances of the formation of schools 

 even within one and the same region. Therefore, the presence In a region of herring bearing some 

 definite indices of growth may complicate the problem instead of solving it. 



