not-yet-known not-yet-known not-yet-known unknown Discussion Developmental environments have strong effects on offspring phenotypes in many organisms, and the strength of those effects can vary among populations and individuals (Johansson and Boix 2013). This variation provides an opportunity for natural selection to shape developmental plasticity in adaptive ways when environmentally induced phenotypes have fitness consequences. Our objectives were to quantify the effects of multivariate, natural nest environments in two distinct habitats (open vs shade). Additionally, we examined variation in family level reaction norms for phenotypes within and among populations. Although incubation conditions influenced phenotypes, the effects were not always consistent with results from previous studies which tested nest conditions in isolation (see below). For example, while open nest environments produced smaller hatchlings as expected, incubation treatments had no effect on egg survival or offspring performance. Phenotypic responses to incubation conditions varied among family groups but support for family-level plasticity was relatively weak. Moreover, phenotypes and reaction norms varied little between the two island populations. Incubation conditions influenced several, but not all, hatchling phenotypes. Open nest conditions caused eggs to hatch sooner than shaded nest environments, as increased temperature accelerates development rate in ectotherms (Zuo et al. 2012). However, incubation treatment had no effect on egg survival, although high temperatures reduce hatching success (Hall & Warner 2021). Nevertheless, as expected, the open incubation treatment produced smaller offspring than the shaded treatment, which is a pattern commonly attributed to reduced yolk catabolism under dry conditions (Miller and Packard 1991). Indeed, relatively dry incubation conditions produce smaller hatchlings with more residual yolk in A. sagrei (Warner et al. 2012) and other reptiles (Tucker et al. 1998, Ji and Du 2001). This pattern was pronounced for offspring from relatively small eggs (i.e., egg mass x incubation treatment interaction); due to a large surface area volume ratio, small eggs in the open treatment were probably less effective at maintaining water balance while converting yolk to body tissue than larger eggs (reviewed in Gatto & Reina 2022). Despite these effects on body size, incubation treatment did not influence desiccation rate or running speed even though warmer temperatures usually enhance locomotor performance (Noble et al. 2018; Pearson & Warner 2018; Hall & Warner 2020). The combination of expected and unexpected results may be explained by complex interactions among the multivariate nest conditions. A unique aspect of our experiment was to simulate natural nest environments in a controlled laboratory setting, whereas previous studies examine the effects of each incubation factor in isolation, which could hinder our understanding of these environmental effects in nature (Tezak et al. 2020). Indeed, the individual variables we used could affect phenotypes in opposing ways, masking single variable effects on phenotype or plasticity. Given that our treatments simulated real nest sites, it is possible that females choose microhabitats that generally maintain a desirable range of phenotypes rather than selecting based on individual abiotic factors. We urge researchers to exercise caution when interpreting the ecological relevance of results based on studies that isolate single environmental factors (e.g., Pruett and Warner 2021, Hall and Warner 2019, Hall and Warner 2018, discussed in While et al 2018), particularly when considering the evolution of developmental reaction norms. Maternal and offspring phenotypes did not differ between the open and shaded islands per se , providing no evidence of trait divergence between the populations. Moreover, reaction norms generally did not differ between populations, indicating little, if any, divergence in plasticity. Despite this general pattern, we show that hatchling body mass of individuals from the open island is more sensitive to incubation environments (i.e., steeper reaction norm slope) than those from the shaded island (Figure. 4a); this may be a spurious observation, but it raises speculation about different benefits of plasticity in open vs shaded habitats. For example, increased plasticity in body size in the open island population combined with the negative effect of the open nest environment on body size indicates that offspring from the open island will be relatively small. However, although the landscape on the open island is mostly devoid of canopy cover, it still contains a small area that provides shade. This shaded microhabitat, which is preferred by females for nesting (Pruett et al. 2020), will generate large hatchlings that typically have increased survival (Warner & Lovern 2014; Chejanovski and Kolbe 2019). In contrast, eggs/embryos that are laid in the open microhabitat will have accelerated development and hatch early, which in turn, also increases survival compared to those that hatch later in the season (Pearson & Warner 2018). Overall, both shaded and open microhabitats may have different benefits and costs for offspring. This level of environmental heterogeneity is not present on the shaded island as all eggs will experience shaded conditions during incubation, and thus providing little opportunity for selection to act on plasticity. Although the increased plasticity in body size in the more heterogeneous environment (i.e., the open island) is consistent with predictions from adaptive models for the evolution of plasticity (Lande 2009, Arnold et al. 2019), we recognize that our study contains no replication at the population level, and therefore are cautious about drawing adaptive explanations for the population difference in body size plasticity that we observed. Evidence for family-level differences in reaction norms was weak but mixed model comparison, G x E analyses, and individual reaction norms collectively demonstrate that the developmental environment most strongly affects hatchling morphology. Family-level variation can have important biological implications because it provides opportunity for natural selection to act on plasticity (Nussey et al. 2005) and indicates that strong stabilizing selection has not shaped reaction norms to polyphenism (Levis and Pfenning 2016). Importantly, for natural selection to drive the evolution of developmental plasticity, the nest environments must generate variation in phenotypes that have fitness consequences. Indeed, body size of juvenile A. sagrei has been shown to positively correlate with survival in the lab (Warner & Lovern 2014) and field (Delaney & Warner 2016), whereas locomotor performance is less important (Pearson & Warner 2018). In addition, stabilizing selection on these traits and/or reaction norms must be consistent over generations for evolution to reduce variance in plastic responses (Suzuki and Nijhout 2006, van der Burg et al. 2020). Lastly, reaction norms must have a genetic component for an evolutionary response to arise (Li et al 2017). While we show that reaction norms exhibit some among-family variation (a proxy for genotype), studies that examine the genetic underpinnings of this plasticity will provide insight into the evolutionary potential of reaction norms in the wild. The within-population variation in reaction norms that we document may provide an opportunity to adapt to environmental heterogeneity. This is critical in the context of invasion biology and global change (Engel et al. 2011), which are two important contemporary issues that pose challenges to natural populations. The plastic responses to natural incubation environments and the among-family variation in reaction norms we observed are features that may have contributed to the invasion success of A. sagrei in various parts of the planet. This variation in plasticity will also facilitate future adaptation to novel environments induced by aspects of global change. Although we show little evidence for population differences in plasticity, the steeper reaction norm slope for body size on the open island might generate fitness benefits in ways that lead to adaptation. While rapid morphological adaptation in response to island habitat structure has been demonstrated for A. sagrei (Losos et al. 1997), adaptive plasticity may rapidly arise due to selection associated with habitat heterogeneity. More work is needed to determine if this pattern is consistent in other populations that inhabit open and forested environments, which would provide a more robust understanding of plasticity as an adaptation to environmental heterogeneity. Overall, quantifying the complex associations among habitat-specific developmental environments, variation in the capacity of organisms to respond to environmental cues (plasticity), and their phenotypic/fitness consequences will provide insight into how populations will continue to cope with temporal and spatial environmental heterogeneity.