The purpose of this research is to quantify a mechanism for lateral melt migration beneath mid.ocean ridges; namely freezing induced melt channeling. Preliminary results suggest that rapid freezing near the sloping base of the lithosphere may lead to the fromation of high permeability m lt channels that allow melt to migrate to the ridge axis. The questions remaining are: (1) What are the fundamental parameters that control the process? (2) Are channels expected in the earth? (3) What is the efficiency of this process? (4) What are the observable consequences for volume timing and chemistry of mid.ocean ridge volcanism? By combining high resolution 2.D magma migration code with large scale thermal and mantle flow models, a systematic series of problems will be investigated to address these questions.