Developmental plasticity allows genomes to encode multiple distinct phenotypes that can be differentially manifested in response to environmental cues. Alternative plastic phenotypes can be selected through a process called genetic assimilation; a process by which acquired phenotypes can become genetically inherited. The importance of genetic assimilation for evolutionary diversification has been debated for decades, even though the mechanisms for assimilation and variation in phenotypic plasticity are still largely unknown. In this dissertation, I investigate how variation in plasticity and genetic assimilation might evolve. In Chapter 1, I investigate the mechanisms that regulate chromatin accessibility, a major driver of gene regulation and cell fate. I show that distinct sets of transcription factors are predictive of chromatin opening at different developmental stages. I show that spineless binding is a major predictor of opening chromatin. Surprisingly, binding of ecdysone receptor (EcR), a candidate accessibility factor in Drosophila, and phenotypic plasticity in butterflies, was not predictive of opening, but instead marked persistent sites. This work characterizes the chromatin dynamics of insect wing metamorphosis, identifies candidate chromatin remodeling factors in insects, and presents a genome assembly of the model butterfly Junonia coenia. To investigate how variation in plasticity might evolve, I assimilated a seasonal wing color phenotype in a naturally plastic population of butterflies and characterized three responsible genes. I found that the transition of wing coloration from an environmentally determined trait to a predominantly genetic trait occurred through selection for regulatory alleles of downstream wing patterning genes. This mode of genetic evolution is likely favored by selection because it allows tissue- and trait-specific tuning of reaction norms without affecting core cue detection or transduction mechanisms. Lastly, I review recent evidence showing that ecdysone-mediated plasticity in different pattern elements such as color or eyespot size can evolve independently. I propose that environmental modulation of ecdysone titers leads to alternate chromatin regulation, and subsequently results in different phenotypes in butterflies. I present a model on the evolution of ecdysone responsiveness that integrates evolution, development and functional genomics, and propose a set of testable hypotheses on how seasonal plasticity evolves.