STUDIES ON THE POPULATION DYNAMICS OF GNATHOPOGON CAERULESCENS SAUVAGE(PISCES), IN LAKE BIWA, JAPAN : IV. FURTHER ANALYSIS ON THE CRITICAL LIFE CYCLE STAGES RELATED TO THE ANNUAL FLUCTUATION OF THE POPULATION

The author reported in his previous paper(MAKI, 1966a)that the under-yearling population of Gnathopogon caerulescens tends to fluctuate yearly towards the period between winter and spring during which they carry on their wintering and spawning activities, respectively. To confirm the findings stated above, some efforts were made in analyzing the various fishery records (1960-1966) of the fishery cooperations around the lake. The records of daily catch obtained with various gears in different periods, i.e., the spawning trap in spring, the trawl net in winter and the gill net in spring and late summer, were compared with each other. A statistical analysis of the data revealed that the amount of the total catch depends on the number of fishing efforts in the case of the spawning trap and gill net fisheries, since the correlation coefficients between the total fishing days and total catch by these two gears were all significant at one per cent level, i.e. 0.861 for the spawning trap, and in the case of gill net, 0.929 in spring and 0.877 in late summer, respectively. Thus, the catch per day by each fishing gear was used in this study as an index of the biomass of the population. The correlation coefficient between the index of the biomass of the spawners and that of the under-yearling population in early winter was 0.846(significant at one per cent level). While, no correlation was observed between the index of the biomass of the underyearling population in early winter and that in early spring. Rather parabolic relation which may indicate the presence of density dependent mortality was observed between the early spring index of the under-yearling population and the late summer one of the same population. The correlation coefficient between the index of biomass of the yearling population in late summer and that in the next spring was 0.909 with the significance at one per cent level. The "wintering" of the yearlings is not important for the annual fluctuation of the population. The indices of the under-yearlings in early winter obtained from three different areas of the lake fluctuate annually with a significant correlation from 1960 to 1966. Much the same tendency was observed between the late summer indices of the yearlings from 1960 to 1963 from two areas. These results suggest that the annual fluctuations of catch per unit effort occurred in a similar pattern at various fishing grounds around the alke. However, a different situation was observed among the early spring indices, which suggested that the mortality rate during the wintering period differed from one area to another.