Editor,—Ratjen et al suggest that differential cell counts in bronchoalveolar ravage (BAL) fluid can be of value in the differential diagnosis of pulmonary infiltrates in immuno-compromised children.1 Firstly I was confused by figure 1 whose legend states that the open bars represented children with bacterial or fungal infections. But this group, according to the figure, had the lowest, not highest, proportion of neutrophils. One or other part of the legend or the figure must be wrong. Secondly, inspection of the data in table 4 shows that the group with bacterial and fungal infection had a mean peripheral blood leucocyte count of 6.3, and neutrophil count of 4.1 × 109/1. It therefore remains possible that …