Trade-offs occur when two traits have opposing fitness effects such that positive selection on one trait is constrained by the negative fitness consequences of the other trait. To understand why trade-off may arise we need to study the genetic and non-genetic factors that influence associated traits because these may respond differently to selective pressure. Research into trade-offs has largely focused on the genetic basis of associated traits, yet both maternal effects and epigenetic effects have recently been shown to affect life history traits that play a role in trade-offs. In this study, we analyze genetic, epigenetic and life-history predictors of one of the most important trade-offs, that between offspring number and offspring mortality. Using a large-scale 3-generational intercross between two divergent mouse lines C57BL/6J and DBA/2J, we show that litter size differences between these lines, although significant, are surprisingly not the most important predictors of mortality. Offspring genotype, maternal effects and their interactions are the most influential factors determining mortality. We found significant paternal effects suggesting an important influence of paternal care or potentially the role of imprinted genes. Perhaps contrary to expectations our results further show that the trade-off between offspring number and mortality is not just a simple function of the two factors yielding, on average, an 'optimal' litter size at weaning. Indeed if one focused on litter size and mortality alone, the slope of relationship is the same for the two lines, yet they differ in the number of young at weaning. Our study reveals that a perceived trade-off between two traits is governed by a more complex set of interactions between genetic and non-genetic effects. © 2011 Springer Science+Business Media, LLC.