1. Despite the central importance for life-history theory, egg-size effects on offspring fitness are still considered ambiguous. Most previous studies were only observational and consequently might suffer from uncontrolled correlations between egg size and parental/territory quality. Even after cross-fostering is performed, direct genetic effects and parental adjustment of post-natal care might confound our estimates of egg-size effects per se. 2. I performed a full cross-fostering experiment in the collared flycatcher (Ficedula albicollis) exchanging the whole clutches between pairs of nests. I statistically controlled for direct genetic effects and parental feeding frequencies. I followed young until recruitment to estimate the long-term effects of egg size and parental provisioning. In addition, I compared the effects obtained in the cross-fostering experiment with those obtained from a set of unmanipulated nests. 3. Egg size per se affected offspring morphology in both the short and long term, while having no effect on offspring survival and immunity. Egg-size effects were not confounded by parental post-natal care and direct genetic effects. 4. The number of care-givers was an influential predictor of nestling performance. Apart from the variation caused by this factor, feeding frequencies had no consistent effect on offspring performance. 5. Fitness benefits of large eggs may be difficult to establish due to variation of egg-size effects between years and habitats. Feeding frequency may affect offspring state but offspring state may also affect feeding frequency. Varying causality between feeding rate and offspring state may preclude the detection of a positive effect of the former on the latter.