If the spawning had lo be characterized only by the external features of spawning, all the 

 known fishes might be referred to the partial spawning type. We do not know any fish in which all 

 the ripe eggs are discharged in one batch. Interruptions in the spawning process are evidently ob- 

 served in all species of fish, but these breaks may last from a few minutes or hours to a few weeks. 



If we were guided only by the intervals between the different extrusions of ripe eggs, would 

 it then be possible to draw a definite limit between partial and non-partial or total spawning? 



In our opinion, this is only possible in those species of fish in which the time needed for the 

 development of the single portions of developing eggs is at least one and a half weeks . An exact 

 idea of the character of the spawning can only be obtained from a study of the processes in the or- 

 ganism itself. 



From a histological picture of the ovary the type of spawning may be determined without any 

 error. By studying a series of histological preparations, we came to the conclusion that the Mur- 

 mansk herring had a proper total spawning (Figure 1, a, b). In addition, a count of the eggs in the 

 ovaries indicates that their number does not vary during the ripening. During the whole spawning 

 period only as many eggs are spawned as were separated in the ovaries at the beginning of the 

 ripening process . 



Several papers mention partial spawning, i.e., in the genus Alosa , which is related to the 

 herring, in carp, and several other species (in cyprinoid fishes partial spawning has been thorou^- 

 ly described by Dryagln (17), but he did not undertake any histological study of the gonads). 



The conclusion of Hickling and Rutenberg (88) that the oceanic herring has partial spawning 

 is explained by the incertitude in the method used (the ovocytes were measured in histological sec- 

 tions) . From the measurements the authors divided all the ovocytes in the ovary into two diameter 

 size groups: J) 0.1-0.2 mm., and 2) 0.5-1.2 mm. In the graph they showed two separate and 

 clearly expressed peaks. The authors inferred that after the ripening of the ovocytes of 0.5-1.2 

 mm., the size group 0.1-0.2 mm. would mature and be extruded after having reached a size of 

 0.5-1.2 mm. Accoraingly, the Atlantic herring would discharge the eggs in two portions. It is to 

 be regretted that the authors in their paper had neither drawings nor photographs of the ovaries in- 

 vestigated. It is therefore difficult to decide to which development phase the ovocytes in the first 

 or second size -group can be referred. 



We do not think that the ovocytes of the Murmansk herring differ very much in size from 

 those of other varieties of the oceanic herring Clupea harengus . The ovocytes of the first size 

 group are probably in the phase of the follicle with one layer of cells . 



In our paper (52) ovocytes in the phase of the single -layered follicle measure 86.87-148.7 

 mu, average 103.66 mu. In this phase the ovocytes from the Murmansk herring correspond in size 

 with those of the first size group from the oceanic herring. The development of the ovocytes from 

 single -layered follicles to the ripe ovocytes (the ovulation) requires not less than 8 months. It 

 would therefore be incorrect to conclude that the ovocytes of the first size group (single -layered 

 follicle) would attain the size of the ripe ovocyte (the second size group - 0.5-1.2 mm.) during say 

 2-3 weeks. 



On this basis we arrive at the following conclusions: 1) in the ovaries of the oceanic her- 

 ring only a certain part of the ovocytes ripens, and 2) the ovocytes of the first size group (0.1-0.2 

 mm.), constituting the so-called reserve of the coming year, will partly ripen and be spawned in 

 the following spawning season. The two size groups of eggs, which might perhaps correspond with 

 two spawning periods, are explained by other causes. Different periods in the spawning of the same 



204 



