A significant part of the variation in the early life history traits of fish can be ascribed to the parental origin of the individual larvae. The primary source of this parental contribution has been attributed to maternal effects and evidence for paternal effects is equivocal. Maternal effects are a non-genetic contribution of a female to its offspring but most reported maternal effects are products of both genetic and non-genetic contributions, i.e. female effects. In this study, parental effects on traits of larvae of Atlantic herring (Clupea harengus L.) at hatch were investigated at one temperature using a 5 × 3 factorial mating design (North Carolina Design II). This allowed estimation of the true maternal effect and the additive genetic variation (heritability). Furthermore, relationships between individual traits were examined and for the first time nucleic acid content and otolith size at hatch were examined together. A significant correlation between the two was found and it is argued to support the notion that otolith growth is more related to metabolic rate than to somatic growth. Maternal effects were detected in larval weight and yolk-sac volume, while paternal and, hence, genetic effects appeared in larval length, yolk-sac volume, RNA : DNA ratio, and lapillar area. The findings suggest that an increased emphasis should be placed upon the importance of male influence on success of early larval fishes.