Two experiments were conducted with lambs fed concentrate-based diets to study the effects of dietary P and soil ingestion on performance, P utilization and mineral composition of serum and alimentary tract contents. In Exp. 1, 20 wether lambs were fed diets for 21 d in a 2 X 2 factorial arrangement of treatments with .2 or .3% total dietary P and O or 9% added Florida Ultisol (highly weathered soil with high P-fixation capacity) containing 4,600 and 6,400 mg/kg total Al and Fe, respectively. True P absorption increased (P less than .05) from 30 to 42% and from 37 to 54% when soil was added to the basal and high P diets, respectively. In Exp. 2, 24 wether lambs were assigned randomly to a 2 X 2 X 2 factorial arrangement of treatments. Lambs were fed during a 70-d period using diets described previously on either a restricted or an ad libitum basis. Soil increased Fe concentration in ruminal fluid (P less than .05) and ash percentage (P less than .001) in ruminal solids and feces. Aluminum concentration in ruminal solids and feces and Fe in ruminal solids were increased (P less than .001) by soil addition, whereas concentrations of Ca, Mg, P, Zn, Cu and Mn in ruminal solids and feces were decreased (P less than .01) by soil addition. There were time X P interactions on serum inorganic P (P less than .01) and Mg (P less than .05) concentrations. Although high in P-fixing capacity, soil fed to lambs in these experiments improved P utilization. Lambs appeared to be able to adapt to changes in intake of dietary minerals in soil, possibly through adaptation of ruminal microflora or a redistribution of body mineral pools.