In this work package, techniques rooted in computational information geometry will be brought to bear on the inferentially demanding problem of mixture estimation, while (Obj. 2) exploring their promise, and that of further challenging extensions of CIG, in MCMC methodology and in mixture model uncertainty. The operational tool to be developed extends Lindsay’s convex geometry approach [5], which offers great insight. Exploiting the full IG of the simplex, and involving a direct search within it, the new methodology will both improve understanding of the variability of the NPMLE of the mixing distribution, and provide improved computation of it. In particular, the troublesome label-switching problem will be avoided.