The absence of stainable iron in a bone marrow aspirate is widely considered to be diagnostic of iron deficiency anemia (IDA). We re-evaluated this concept by studying a cohort of 108 consecutive bone marrow specimens from an unselected series of patients who were seen at a hematology clinic and in whom iron stores were reported as being absent. A review of the pathologic reports revealed 19 inadequate specimens and 15 with decreased, but not absent, iron stores. Thus, only 74 of the 108 cases had been accurately reported. In 37 of these cases, adequate clinical and laboratory data were available and allowed further analysis. In 18 patients, careful review of patient history, physical examination, results of endoscopic procedures, and follow-up information failed to suggest the presence of IDA (group A). The review process in the remaining 19 patients suggested the possibility of IDA (group B). Significant differences between groups A and B were observed in serum ferritin (P = 0.001) and red blood cell mean corpuscular volume (P = 0.004). In contrast, the two groups did not differ significantly in hemoglobin concentration, serum iron, total iron-binding capacity, transferrin saturation, or erythrocyte sedimentation rate. These observations suggest that a pathology report of absent bone marrow hemosiderin may be inaccurate in more than 30% of cases and, even when accurate, may not necessarily signify the presence of IDA. Measurement of the serum ferritin level is needed to confirm a clinical diagnosis.