This article discusses the race, migration, growth pattern, mortality and dynamicsof the herring (Clupea harengus pellasi Valenciennes) inhabiting in Huang Hai, andattempts to make prediction and management on fisheries. The results are summarizedas follows: 1. IIerring inhabiting in areas north to the latitude of 34°in Huang Hai all the year round it is believed to be a race of the Pacific herring. Hence they are traditionallycalled the Huang Hai herring. 2. The growth pattern of Huang Hai herring can be described by von Bortalanffyequation, and as of year 1972 the value of parameters for the equation is IV_∞ = 314g.,L_∞ = 308mm, and K = 0. 58. The maximum growth rate appears at t = 1.3 years. FromFebruary 1972 to March 1973, the total Instantaneous mortality estimated by data forthe catch per effort was 1.05 which consists of the fishing mortality 0. 939 and thenatural mortality 0. 111. 3. The numbers of the 1970 year-class in early January were estimated with twoprocesses, Delury's method and the method of estimation of fishing mortality. The dif-ference in valuo of the calculation between the two methods is about 5%. the averagevalue is 2082×10⁶ individuals, it coincides with actual catches as shown in the table 6. 4. In the 1970s, the fluctuation in the numbers of the population happened to be maximum in the year of 1972 and minimum in 1977. The former were ten times greaterthan the latter. These drastic dynamics were considered to be a natural phenomenon rather than the result of over-fishing. The statistical analysis shows the interdependencebetween actual catches and the abundance index obtained from fish--searching in October. The relative coefficient was 0. 9710. 811 (P0. 05), which accounts for thefact that catches can be predicted by abundance.index. 5. After making a careful study of the optimal fishing mortality and mesh--sizefor the fisheries, and considering the economical value of the herring eggs, we workedout a strategy on the management of the fisheries for Huang Hai herring. Since thevalue(including the yield, average body length and weight as well as the age of the 1970year-clase) obtained by Boverton-Holt model and S--E model are similar to the resultssampled from fishcries, it is suggested that the conclusion here submitted in the papermay be valid.