Migraine is one of the few pathologies which gave rise to a tremendous number of physiopathological hypotheses. The variability of its clinical features and of the crisis initiating triggers, together with the numerous functional and/or biological abnormalities reported in migrainous patients, led to multiple 'theories' about migraine. For instance, migraine attacks may be associated with modifications of cerebral blood flow, and/or alterations at the cellular (neuronal and peripheral: platelets, mast cells, etc.) and subcellular (mainly mitochondrial) levels leading to variations of parameters such as serotonin, vasoactive neuropeptides, histamine, nitric oxide, neuroactive amino acids, etc. However, these modifications are mainly related to migraine attacks but not to migrainous patients. These emphasize how important is the distinction between the crisis mechanism(s) and the determinism of migraine illness. Despite the absence of any true animal model of migraine attack, the obtention, through the activation of the trigemino-vascular complex, of an experimental meningeal neurogenic inflammation was a clear breakthrough for the understanding of the migraine attack. Concerning the determinism of migraine, its familial characteristic has been known for a long time, but genetic studies started only recently. Despite some important contributions, the respective roles of genetic and environmental factors, as well as the transmission mode of migraine, remain largely to be determined. Practically, these genetic data, which really concern only a very peculiar form of migraine--the familial hemiplegic one--do not have presently any diagnostic or therapeutic application.