Abstract: This dissertation research is concerned with diffusion flames generated by a porous spherical burner. It consists of two parts: the structure and extinction of weakly buoyant, nearly spherical, stationary flames, and the structure and dynamics of these flames in response to rotation of the burner in micro-gravity. In the first part of the investigation, normal-gravity experiments were conducted with nearly spherical, inverse diffusion flames of small density difference with the ambient to study the chemiluminescent flame structure and oscillatory extinction. The flames were imaged by a UV camera, with narrow-band-limited filters corresponding to electronically excited OH and CH. The experimental results were then compared with computations allowing for detailed chemistry and transport. While the comparison was very satisfactory for the hydrogen flames, OH* chemiluminescence exhibited two peaks for the hydrogen/methane flames, demonstrating the importance of the H + O + M OH* + M reaction. By decreasing the reactant concentrations in the ambient, the transient extinction behavior of these flames was also studied. In particular, pulsating instabilities were experimentally observed and measured for a spherical diffusion flame. This was further validated by comparing the measured frequency of oscillations to that obtained from computations, showing good agreement. In the second part of the investigation, the coupled effects of the rotational motion and non-unity Lewis number diffusion for both fuel and oxidizer were first studied theoretically through perturbation analysis. The analysis showed that the rotational motion induces a secondary flow that distorts the otherwise spherical flame into a pancake shape. The flame temperature was also affected, such that the flame became more susceptible to extinction either at the poles or the equator depending on the system Lewis numbers. Microgravity experiments were subsequently conducted at the NASA Glenn Research Center not only to verify the theoretical predictions, but also to extend and supplement the theories to study the flame response at high angular velocities where the perturbation theory no longer applies. Results show that, in the limit of low angular velocity, no visible changes were observed as the perturbations predicted by the theory were also small. However, in the opposite limit, the effects of rotation were clearly visible and the response agreed surprisingly well with the perturbation theory for most conditions.