RNA–DNA ratio of herring and sand lance larvae from Port Moller, Alaska: Comparison with prey concentration and temperature

ABSTRACT A key assumption of hypotheses that link the production of prey for larval fish with year‐class strength of fish is that larval growth and condition is food‐limited. We tested this assumption by comparing whole‐body RNA‐DNA ratios of individual Pacific herring, Clupea pallasi , larvae and Pacific sand lance, Ammodytes hex‐apterus , larvae from Port Moller, a subarctic Alaskan estuary, with prey concentration and temperature. RNA‐DNA ratios were correlated with larval length, but not with prey concentration or temperature. Ratios were not significantly different between a warm, well‐mixed station with low prey concentrations and a colder, stratified station with higher prey concentrations. Using RNA‐DNA ratios, we classified as starving 11 to 23% of first‐feeding (< 13 mm long) herring larvae and 45% of first‐feeding (< 7 mm long) sand lance larvae. However, starvation could not have been caused by low concentrations of prey because micro‐zooplankton prey concentrations were high enough (16 to 84 prey L _l ) to support relatively high rates of growth. Therefore, starving larvae were either abnormal or they were stIII learning to forage. We conclude that the magnitude of starvation among first‐feeding herring and sand lance larvae, and, therefore, the total contribution of starvation to year‐class strength, is dependent not only on prey concentration, but on the proportion of a population of larvae that can feed effectively.