The absorbability of four levels (20, 35, 75, and 150 ppm) of iron from dietary heme and nonheme sources was studied in normal and iron deficient rats. Beef liver was the heme iron source and a plant mixture (soy flour, wheat bran, potato flour, spinach and parsley, all at 20% wt/wt, dry matter basis) was the nonheme source. 59Fe was used to compare the availability of dietary nonheme iron to reference diets supplemented with ferrous sulfate. Hemoglobin regeneration efficiency (HRE) and apparent iron absorption were consistently higher in anemic rats and were inversely related to iron intake. No more than two-thirds of iron from all the iron sources was absorbed. 59Fe was high in the blood of anemic rats and continued to increase over the 12h study period for all iron sources and at all iron levels. Absorption of nonheme iron was similar to that of HRE. There was a high correlation (r=0.98) between percentage 59Fe absorption and apparent iron absorption. There was a very low correlation (r=0.03) in anemic rats between 59Fe activity in blood (2h after dosing) vs. percentage retention of 59Fe. Heme iron absorption was higher in anemic rats than in normal rats. We conclude that with an experimental methodology comparable to human studies, iron absorption of rats will be similar to that in humans.