Year-class strength is believed to be determined in the early life stages of marine fish, and the variations in survival rates underlying the variable year-class strength have often been attributed to the influence of a fluctuating environment on feeding and growth in the larval and early juvenile stage.

Using field data in combination with laboratory experiments, I was able to show that cod larvae from the eustarine Baltic Sea have to migrate from hatching depth below the halocline to the surface in order to optimize feeding conditions, as defined by food abundance and light dependent capture success. On all cruises I found a fraction of the older larvae below the halocline, which indicate that they had not managed the migration towards more optimal feeding conditions. The older larvae caught below the halocline were all in poor nutritional condition, whereas in the surface layers a large proportion of similar aged larvae exhibited moderate to high protein growth rates. In order to phenotypically compare these two larval groups, I analysed their otoliths with regard to hatch-check sizes, which is suggested to correlate with either larval size at hatch or larval metabolic rate during the embryonic stage. My analysis showed that larvae with small hatch-checks died around day 12 after hatch. The hatch-checks of similar aged larvae were significantly larger in the surface compared to similar aged larvae in the deep water indicating that larvae with larger hatch-checks are more likely to complete the migration through the halocline. Having obtained RNA/DNA-ratios and otoliths from individual larvae I compared these two phenotypic traits. Using otolith fluctuating asymmetry (= small random deviations from bilateral symmetry) as a measure of feeding related stress, I was able to show that the level of fluctuating asymmetry in larvae with low RNA/DNA-ratios was significantly higher than in groups of larvae with either moderate or high RNA/DNA-ratios.

The results from these studies have enhanced our knowledge about the recruitment processes in marine fish, and by combining well-tested and new methodology I have identified environmental and phenotypic ‘bottlenecks’ separating viable and moribund Baltic cod larvae.