Abundances in HII Galaxies

François Ch. Cuisinier¹

(1) GEMAC/OV/UFRJ

Abundances serve two purposes: (1) assess stellar evolution, through the study of relevant abundances ratios sensitive to the nuclear cyclings of progenitor stars; (2) study galactic evolution, using general prescriptions of chemical evolution. In the case of HII galaxies, two elements are of particular interest, because their abundances are easy to derive from emission lines arising from the gas: oxygen and nitrogen. The origin of oxygen is relatively well known: it is produced in massive stars, and expelled nearly imediately after a few millions of years in the interstellar medium. Its recycling, however, is not necessarily instantaneous, as generally assumed in chemical evolution models: its release is connected to the explosion of supernovas, producing a very hot [EQUATION]-[EQUATION]K gas phase. This gas phase takes at least a few tens of millions of years to cool down, and the mixing of oxygen should happen nearly at the same time as the bulk of intermediate mass stars products. Nitrogen, on the other hand, is believed to produced both in high and intermediate mass stars - its exact origin being still subject to discussion. "State of the art" yields predict that massive stars produce secondary nitrogen, whereas intermediate massstars produce primary. We will first present a new chemical evolution model that takes into account the delay in the mixing of oxygen (as well presented as a poster (Ferraresi et al.) at this meeting). We will then discuss the origin and recycling of oxygen and nitrogen, based on the spatial variations of abundances in HII galaxies. Oygen, passing through a hot gas phase before cooling down into star forming clouds, should present quite homogeneous abundances. Nitrogen, believed to originate principaly from intermediate mass stars in HII galaxies, should be recycled imediately after the dissipation of the planetary nebula phase, at the place where the progenitor star was born; its abundance should therefore present much higher spatial variations. We will see how recent observations can shed light on these problems