In this talk I revisit the problem of gas accretion onto minihalos after reionization. I show that primordial minihalos with v(cir) < 20 km s(-1) stop accreting gas after reionization, as is usually assumed, but in virtue of their increasing concentration and the decreasing temperature of the intergalactic medium as redshift decreases, they have a late phase (at, redshift z < 2) of gas accretion and possibly star formation. As a result we expect that pre-reionization fossils have a, more complex star formation history than previously envisioned. A signature of this model is a bimodal star formation history. The dwarf spheroidal galaxy, Leo T. that inspired the present work, fits with this scenario. Another prediction of the model is the existence of a, population of gas rich minihalos that level, formed stars. A subset, of compact, high-velocity clouds may be identified as such objects but the bulk of them may still be undiscovered.