2003A&A...397..943B

We emphasize the limits of the classical paradigm explaining the periodic appearance of emission lines by shock waves induced by radial periodic pulsations. In particular, we argue that, if pulsation occurs, it would not be purely radial and the light curve resulting from the contribution of zones at different phases would have a shape different from what we actually observe. The discovery of low mass companions to numerous G stars, some of them very close to the star, led us to look at the fate and the influence of this companion when the star becomes a giant. After Rudnitskij (2002IAUCo.176..384R), we suggest that the behavior of the Mira variables is mainly due to the presence of a companion in the tenuous atmosphere of the giant instead of the effect of a pulsating atmosphere explaining the presence of emission lines and the periodicity of these variables. We argue that the presence of a companion may account for the shape of the light curve and the observed phase lag between the maximum visual and the maximum maser light.