I seem to have overlooked this last year, but was excited to see some stable isotope results, (published last year in J. Arch Science by D'Andrea et al), from northern Ethiopian pre-Axumite and Proto-Axumite sites (i.e. 800-100 BC). This indicate a strong tendency in human bones towards predominantly (but not necessarily) exclusively C-3 plant diet, i.e. such crops as wheat and barley, but with limited C-4 input. This is of interest since Sorghum, tef and finger millet, all of which might expect to have been around are C-4 plants. So the mystery deepens as to the antiquity of these taxa in agriculture in this region-- or at least they were not very important. If memory serves there is some evidence for tef from around this period (but mainly later)-- treated in some detail by D'Andrea (and blogged previously), where as the earliest finds of finger millet and Sorghum are later and Axumite Also there is an important contrast with the domestic fauna, which have a distinctly different diet form the people. Cattle and donkey have a strongly C4 signature suggesting grazing or foddering mainly from the savannah grasslands. Sheep, goat and gazelle fall in between suggesting C4 grazing mixed with browsing of C3 shrubs etc.