Abstract: Hierarchical radiosity and Monte Carlo radiosity are branches of radiosity research that focus on complementary problems. The synthesis of both families of radiosity algorithms has however received little attention until now. In this paper, a procedure is presented that bridges the gap. It allows any proposed hierarchical refinement strategy to be investigated in the context of an arbitrary discrete Monte Carlo radiosity algorithm. The synthesis of Monte Carlo radiosity and hierarchical radiosity yields more reliable and easy-to-use radiosity algorithms. Our experiments show that storage requirements for rendering complex models can be reduced to about 20% compared to hierarchical radiosity. At the same time, computation times for images of very reasonable quality can be reduced by one order of magnitude.