A commentary onOxidative stress as a cost of reproduction: beyond the simplistic trade-off model by Speakman, J. R., and Garratt, M. (2014). Bioessays 36, 93-106. doi: 10.1002/bies.201300108 Current theory proposes that life history traits cannot be simultaneously maximized because there are costs paid in the currency of fitness when a change in one trait, for example increased reproduction, results in a detrimental change in another, for example decreased survival. The underlying reasoning is that resources (e.g., nutrients, energy, time) are finite and this gives rise to trade-offs in their allocation among traits, activities, or processes that compete for the same resource. Therefore, allocation of a resource to one trait means that less can be invested to other traits. This simple concept, referred to as a trade-off, has had a prominent role in our understanding of how life history variation arises.We currently know very little about how these costs of reproduction are actually incurred, since the majority of studies have focussed on the outcomes rather than the mechanisms. In the last decade, there has been substantial interest in the role of oxidative stress as a mechanism underlying the cost of reproduction and other life-history trade-offs. Oxidative stress is a complex multifaceted biochemical condition of the organism that occurs when generation of oxidative molecular damage increases (Costantini, 2008(Costantini, , 2014Metcalfe and Alonso-Alvarez, 2010;Isaksson et al., 2011). Several recent studies have, however, found weak or no support that reproductive effort increases oxidative damage (Costantini, 2014). In a recent article, Metcalfe and Monaghan (2013) have pointed out that several of these studies testing the role of oxidative stress as a cost of reproduction did not include data on non-breeding controls, making it hard to exclude seasonal or ontogenetic causes of changes in oxidative stress. They also highlighted, however, that females may tailor their own level of reproduction to avoid oxidative damage, hence it would be also important to manipulate the reproductive effort. In another article, Speakman and Garratt (2014) reinforced the idea of testing the oxidative cost of reproduction by randomly assigning animals to two groups, one of which is forced to reproduce and another one that is prevented from doing so. This is because even if individuals can choose their level of reproduction, what they cannot avoid, is an increase in metabolic rate, relative to a non-reproductive control animal. Recent experimental meta-analytic (Blount et al., 2015) work has, however, found that oxidative damage is lower in reproducing than in non-reproducing individuals, providing again little support for the oxidative cost of reproduction hypothesis.An important question then is whether non-reproducing individuals are truly appropriate controls. The answer does not seem to be so straightforward. Non-reproducing individuals may differ from reproducing individuals in several characteristics, which might make ...