Maternal effects directly and indirectly modify an offspring’s phenotype during development, preparing the offspring for an environment similar to an environment experienced by the mother. Evolutionarily, this could be adaptive if organisms did not experience a large amount of environmental variation across generations. We argue increased anthropogenic changes have led maternal effects to be maladaptive in plants. Similar problems in humans have been proposed, where metabolic diseases may be increased by a mismatch between the in utero environment and the adult environment; often referred to as the Barker Hypothesis. Plants may experience similar metabolic and functional disorders, caused by maternal effects generated from unreliable environmental cues. In the last 200 years, human activities have increased habitat fragmentation, changed patterns of nutrient deposition, and increased climatic variation within and among years. In this manuscript, we suggest that ecologists should consider the increased negative influence of maternal effects on plant growth and reproduction in spatially fragmented and temporally stochastic landscapes. Several implications for conservation and research arise from this discussion: (1) researchers should focus on evolutionarily relevant measures of spatial and temporal heterogeneity, (2) the maternal history of seeds used for population restoration should be considered, and (3) contiguous habitats should be constructed to reduce the impact of evolutionary traps.