ON "CANNIBALISM" IN THE RAINBOW TROUT FINGERLING, SALMO GAIRDNERII IRIDEUS

抄録

It was proved that the phenomenon of "cannibalism" participated in the cause of the loss in number of rainbow trout in their early rearing stages. This phenomenon was studied in datail, and the results obtained are as follows : 1. The fingerlings were reared under the ungraded conditions for four months after the initial feeding. Approximately 20 per cent of them were recognized as dead fish but other 30 per cent disappeared during this period. The main cause of the loss may be "cannibalism". 2. In the experiment, the frequency distribution of body length in the fourth month was obtained as a curve inclined to the smaller fish, its skewness being +1.90. 3. It was observed that the smaller the predators were, the smaller the ratio of their body length to the upper limit length of the prey's was, and the smaller the range of the prey's size was. The body length ratio to prevent "cannibalism" at the first stage of the initial feeding is thought to be 1.3. 4. "Cannibalism" in rainbow trout fingerlings in pond culture is observed since approximately ten days after the initial feeding. 5. The predators' preference to the prey is thought to be considerably strong in view of their satiation rate. But the food efficiency in cannibalistic feeding could not be thought to be especially excellent compared with other natural diets. 6. "Cannibalism" does not always happen. In the population in which it does not take place, the dead fish recognized are relatively few and the health of the fish and rearing circumstances are found to be favorable.