All of our patients with dialysis siderosis received intravenous iron dextran. We agree with Bregman and colleagues that iron dextran appears to be the principal cause of the dissociation between the hepatic and the bone marrow iron in these patients. However, this may not be the sole cause of this phenomenon. Recently, we observed this paradox of hepatosplenic siderosis and marrowiron depletion in nondialysis patients with liver cirrhosis.Other published observations of Bregman et al1 concerning HLA-linked iron overload, myopathy, marrow iron, and serum ferritin are in basic alignment with ours. They furnish no data concerning hepatic iron stores. Their argument seems to center about the validity of the serum ferritin as a marker of bone marrow iron. Since our aforementioned study, we have reported the association of high serum ferritin levels and depletion of marrow iron stores in ten of 36 dialysis patients with hepatosplenic siderosis.