To the Editor.Ìý —The case report by Pauzner et al¹ describing sustained hyperventilation in a woman with primary cerebral lymphoma holds great interest in correctly identifying how often this type of tumor has been associated with the neurogenic overbreathing syndrome. One or two points, however, deserve comment.The authors conclude that their patient did not have tumor invading the brain stem because computed tomography revealed no abnormality. Perhaps. But primary lymphoma often spreads to paraventricular structures and meninges. Furthermore, computed tomographic scanning overlooks small brain-stem lesions, and not all primary lymphomas take up contrast marker. Admission neurological examinations on the patient describe drowsiness, left-gaze paralysis, diffuse phasic spasticity (increased deep tendon reflexes in four extremities), and bilateral extensor plantar responses. This combination in the absence of lateral ventricular dilatation or computed tomographic evidence of downwards herniation or shift could equally well, or better, be taken as signifying primary brain-stem