Migraine is a frequent neurological syndrome with a heterogenous clinical presentation. Its pathophysiology remains poorly understood but a genetic aetiology has been suspected for a long time, as well as a significant influence of the environment. Familial hemiplegic migraine belongs to monogenic migraines, which are very rare entities with an autosomal dominant transmission. In that type of migraine, environment has a minor role, even if some attack triggering factors have been identified. The common forms of migraines, with and without aura, episodic or chronic, are polygenic and the results of genetic studies concerning these migraines are rather disappointing. The recent use of genome-wide approaches (linkage analyses and genome wide association studies) opened new perspectives and more convincing results are eagerly awaited in the next years. The environmental contribution (endogenous or exogenous) is much more pronounced in the common forms of migraine, with the identification of numerous predisposing or triggering factors, among which only some can be avoided. Finally, a recent behavioural hypothesis coming from adaptative darwinian theories has proposed a genetic-environment integrative model for common migraine. The latter would result of hereditary physiological adaptative defence mechanisms which would be progressively impaired by repeated brain homeostasis imbalances due to the environment.