A major goal of evolutionary biology is to understand the genetic architecture of the complex quantitative traits that may lead to adaptations in natural populations. Of particular relevance is the evaluation of the frequency and magnitude of epistasis (gene-gene and gene-environment interaction) as it plays a controversial role in models of adaptation within and among populations. Here, we explore the genetic basis of flowering time in Arabidopsis thaliana using a series of quantitative trait loci (QTL) mapping experiments with two recombinant inbred line (RIL) mapping populations [Columbia (Col) x Landsberg erecta (Ler), Ler x Cape Verde Islands (Cvi)]. We focus on the response of RILs to a series of environmental conditions including drought stress, leaf damage, and apical damage. These data were explicitly evaluated for the presence of epistasis using Bayesian based multiple-QTL genome scans. Overall, we mapped fourteen QTL affecting flowering time. We detected two significant QTL-QTL interactions and several QTL-environment interactions for flowering time in the Ler x Cvi population. QTL-environment interactions were due to environmentally induced changes in the magnitude of QTL effects and their interactions across environments--we did not detect antagonistic pleiotropy. We found no evidence for QTL interactions in the Ler x Col population. We evaluate these results in the context of several other studies of flowering time in Arabidopsis thaliana and adaptive evolution in natural populations.