Abstract Both Big-Bang and stellar nucleosynthesis have outcomes related to the density of baryonic matter, but whereas in the first case there is a standard model that makes very precise predictions of light element abundances as a function of the mean density of baryons in the Universe, in the second case various uncertainties permit only very limited conclusions to be drawn. As far as Big-Bang synthesis and the light elements are concerned, existing results on D, 3He and 7Li indicate a value of ΩNh20 greater than 0.01 and less than 0.025, where ΩN is the ratio of baryonic density to the closure density and h0 is the Hubble constant in units of 100 km s-1 Mpc-1; probably 0.5 < h0 < 1. New results on the primordial helium abundance give a still tighter upper limit to ΩN,ΩNh20 < 0.013, which when compared with redshift surveys giving Ω > 0.05 implies that the observed matter can all be baryonic only if the various uncertainties are stretched to their limits.