Partial mortality in broods of altricial passerines results from various reasons, including deficiency of resources, inadequate parental care, diseases and other external factors. In this study we test two hypotheses concerning the phenomenon of partial mortality in nestling altricial passerines: (1) non-adaptive starvation hypothesis based on the assumption that partial brood losses are caused by starvation and undernourishment and has no adaptive value, and (2) adaptive starvation hypothesis assuming that losses from starvation might be elements of an adaptive strategy of reducing surplus nestlings in broods that are over-large. To test the above hypotheses we divided broods of two tit species Great Tit Parus major and Blue Tit Cyanistes caeruleus into two categories: broods with incomplete, less than 100% fledging success and broods with complete 100% fledging success. Subsequently, we compared physiological performance of nestlings between brood categories using blood hemoglobin concentration. We predicted, that if non-adaptive starvation hypothesis was correct, in broods with complete fledging success nestlings should be in better condition. Alternatively, if the adaptive starvation hypothesis was correct, no such difference in nestling condition should occur. In accordance with our prediction, we found that for both species, broods of nestlings that attained 100% fledging success had a higher mean hemoglobin concentration than broods in which fledging success was less than 100%. This result strongly supports non-adaptive starvation hypothesis.