This study uses the concept of stage-environment fit (Eccles et al., 1993) in conjunction with self-determination theory (Deci & Ryan, 1985) to guide an investigation of at-risk eighth graders' motivational and academic resilience. A developmentally-calibrated method was used to divide students into motivational and academic resilience groups based on their resilient, average, or stress-affected levels of academic engagement and GPA. Data from 167 eighth graders and 155 sixth graders were used to examine the extent to which students' ratings of autonomy, teacher support, peer support, and engagement in garden-based education were related to resilience group membership, and whether these four resources seemed more important to eighth graders than to sixth graders. Results provided support for the overall hypothesis that the four developmental resources (autonomy, teacher support, peer support, and garden engagement) seemed to serve as motivational and/or academic resources, but did not, generally, provide support for their being more important for eighth graders than for sixth graders. Analyses of variance and regressions showed that autonomy, teacher support, and garden engagement were all resources for eighth graders' motivation and achievement, with autonomy and teacher support as the strongest resources. Peer support was only a resource for motivation, and only for eighth-grade boys. In terms of unique effects, only autonomy uniquely predicted eighth graders' dual (motivational and academic) resilience group membership when controlling for the other developmental resources. In contrast, autonomy, teacher support, and garden engagement all uniquely predicted sixth graders' dual resilience group membership. Autonomy's status as the only unique predictor of eighth-grade resilience contrasted with an overall trend in which correlations among all variables were weaker for eighth graders than sixth graders. This contrast suggested that one part of eighth-grade risk might be an increased immunity to motivational resources. Autonomy seemed to fully mediate the relationship between teacher support and dually-resilient group membership for eighth graders, highlighting autonomy's role as a key component in motivational processes for at-risk eighth-graders. The study shed light on the educational risk posed by "eighth-gradeness" in conjunction with having an at-risk demographic status and a being traditional middle school setting, and suggested that autonomy might be a necessary, if not sufficient, route to eighth-grade motivational and academic resilience in light of this risk.