The role of starvation in the population dynamics of larval Pacific herring, Clupea harengus pallasi

Starvation has long been hypothesized to be an important agent of mortality in first-feeding larval fishes. I measured the degree of starvation in four natural cohorts of larval Pacific herring, Clupea harengus pallasi, from Bamfield Inlet, British Columbia, in order to determine if starvation occurred, and, if it did, to determine its relative importance in the dynamics of the populations. A morphometric condition factor was used to diagnose the nutritional status of the larvae rather than a chemical or histological index in order that large numbers of larvae could be screened. The condition factor was derived from five body measurements and dry weight. It was calibrated with learning samples of fed and starved larvae reared from the egg in laboratory aquaria and in in situ culture chambers suspended in Bamfield Inlet. This condition factor was shown to be superior to the other major morphometric condition factor, Fulton's factor, because it was independent of the size of the fish, because its constituent variables were rendered orthogonal to (independent of) each other, and because its constituent variables could be assigned some biological meaning. The effects of the shock of net capture on the postpreservation morphometry and dry weight of Pacific herring larvae were measured from towing experiments. Only larvae less than 14 mm long were affected by net capture. Gompertz models were developed to correct the morphometry and dry weight of field-caught larvae so that they could be compared with the learning samples of fed and starred laboratory-reared larvae. Delayed feeding experiments showed that starving Pacific herring larvae have 6-8.5 days at 6-10° C after the exhaustion of the yolk in which to feed before they enter a state of irreversible starvation. This is a relatively long period of time compared to other species that have pelagic larvae. The length of this time period was shown to be due almost entirely to the rearing temperature. Other rearing experiments showed that delayed feeding and starvation affected the onset and rate of ring deposition in the sagittal otoliths as well as morphometry and dry weight. Only populations with an average linear rate of growth of 0.36 mmd⁻¹ or higher deposited rings at a daily rate. Wild larvae were sampled with towed plankton nets and with a stationary night-light in Bamfield Inlet and surrounding areas. Starvation was diagnosed in 4.9-90.0% of the 7-20 mm long. Most of these starving larvae were 9-15 mm long and they were found within 3 km of the hatch site. The rates of growth in length and weight, and the weight-length exponents, decreased with increasing population density and increasing percent of the population diagnosed as starving. The diffusion coefficients, mortality rates, and dispersal velocities, increased with increasing abundance and percent starving. These results support the hypothesis that starvation plays a role in the early life history of Pacific herring but the direct role is minor—starvation was directly responsible for only 3-23% of the total mortality rate. Starvation may have an additional indirect effect on mortality by increasing the vulnerability of the fish to predators. This hypothesis cannot be rejected. Neither can the hypothesis of density-dependent predation. A model of mortality rate in marine pelagic fish eggs and larvae was developed in order to test the hypothesis of a mortality-patchiness interaction. The model was based on the observation that the daily instantaneous mortality rates of marine pelagic eggs and larvae are higher than expected from the trend of mortality rate with dry weight in the sea. This extra mortality was positively correlated with the patchiness of the spatial distribution of the eggs and larvae, which suggests that it is caused by the aggregation of predators on patches of their prey. The product of weight-dependent mortality, 5.26x10⁻³W[sup -0.26], and 1 + Lloyd's patchiness index predicts mortality rates close to those that have been measured from the field by other workers. The results of the field study and the modelling exercise offer strong support to the idea that starvation is a relatively minor phenomenon in wild herring larvae. Mechanisms that control the spatial patchiness of eggs and larvae and, therefore, their vulnerability to predators, appear to be far more important. The mortality-patchiness hypothesis offers a new route to the study of the effects of predation on natural planktonic communities.Science, Faculty ofZoology, Department ofGraduat