computing the average number from a small number of observations . 



The variation in the number of eggs in samples of an order of size 1 gram is completely 

 regular in relation to the stage of maturity of the ovaries . In the various stages the number of 

 eggs per gm. varies in the following way: stage III - 3, 255 eggs, stage IV - 2, 148 eggs, stage V - 

 1,443 eggs. The relation between these figures is approximately 3:2:1. 



With a great number of observations, the coefficient of error will be of a minimum size, 

 and from the number of eggs per gm. it is therefore possible to judge the stage of maturity. 



The histological investigations indicate that the Murmansk herring has a proper total spawn- 

 ing, without any transition between the ripening ovocytes filled with yolk, and the ovocytes in the 

 follicle with one single layer of cells, and younger ovocytes (the complex of ovocytes in ovaries of 

 stage II). This is confirmed by a number of photomicrographs shown in our paper. The insigni- 

 ficant fecundity in the Murmansk herring is possibly explained by the total spawning (all the eggs 

 extruded in one batch) . 



In the Caspian herring, the fecundity is, as is very well known, considerably higher than in 

 the Murmansk herring. During the process of evolution, the faculty of adaptation increased in the 

 Caspian herring, as expressed in the partial spawning, assuring the conservation of the species. 

 During the year, or more exactly, one spawning period, the Caspian herring spawns several times. 

 As one portion of the ripe eggs is extruded, another portion will in time be mature, and so on, up 

 to 3-4 times . 



One may assume that, because of the partial spawning, the latest phases in the ripening of 

 the ovocytes (the 2nd, 3rd and 4th portion) pass on more quickly than in the Murmansk herring. 

 Accordingly, more ovocytes can be contained in an ovary of the same volume, ripen and be dis- 

 charged in spring during several intervals of time. In the single individuals of the Caspian herring, 

 the spawning period therefore lasts longer than in the Murmansk herring. In the Murmansk her- 

 ring, the total portion of eggs which is going to be spawned at the same time in spring is separated 

 simultaneously at the end of the summer of the preceding year . The spawning proceeds rapidly, 

 but the total number of eggs discharged by one single female is smaller than in the Caspian herring. 



The simultaneity in the development of the portion of eggs which was separated at the end of 

 the summer, is confirmed not only by the histological analysis, but also from the direct counts of 

 the eggs. As already mentioned, the number of eggs in ovaries of stage III and IV is nearly iden- 

 tical , 



The Structure of the Ovaries , Phases in the Development of the Ovocytes , and the 



Stages of Maturity . 



Material and Methods for the Macro- and Microscopical Investigation. 



For the macro- and microscopical investigation of the ovaries we used in main samples 

 collected at different times of the year. Our attention was also drawn to samples taken by other 

 investigators for the study of morphological questions, food, condition, and so on. Such samples 

 were usually very badly fitted for a histological investigation, as the herrings most often were pre- 

 served whole in formalin, alcohol, or merely salted. Sometimes we obtained satisfactory histolog- 

 ical preparations, but usually these samples were useful only for a macroscopical description of 

 the gonads . 



From the material collected in 1938 we analyzed 46 samples containing 5,500 individuals; 



210 



