Neurosurgery can be complicated by the clinical situation commonly referred to as “tight brain,” in which the brain presses against the inner table of the skull or protrudes through the craniotomy site. We report here a retrospective study of 32 patients who had undergone elective craniotomy for resection of supratentorial mass lesions. We determined the relationship between lumbar cerebrospinal fluid pressure (CSFP) and brain relaxation and whether brain relaxation varies with anesthetic technique. Patients had received one of four anesthetic techniques: 1 MAC isoflurane (ISO), 1 MAC desflurane (DES), 50% N2O with 0.5 MAC ISO, or 50% N2O with 0.5 MAC DES. Lumbar CSFP had been.recorded before the induction of anesthesia (baseline) and immediately prior to dural incision. Charts were retrospectively reviewed for evidence of tight brain, which was considered present if mannitol had been administered, CSF had been drained via the lumbar needle, or the surgical dictation noted the brain was tight at the time of dural incision. Tight brain occurred in 10 of 32 patients. CSFP (mean ± SD) was significantly greater in the tight than in the nontight group both at baseline (11 ± 5 vs. 8 ± 3 mm Hg, p < 0.05) and immediately prior to dural incision (13 ± 7 vs. 9 ± 4 mm Hg, p < 0.05). Tight brain did not occur in any patient with CSFP < 6 mm Hg, but it did occur in all patients with CSFP > 17 mm Hg. Within the range of 6–17 mm Hg, CSFP was not predictive of brain relaxation. Tight brain was more common in patients receiving 1 MAC ISO or DES (9 of 20 patients; 45%) than in patients receiving 0.5 MAC ISO or DES with 50% N2O (1 of 12 patients; 8%, p < 0.05). We conclude that in patients undergoing elective craniotomy for resection of a supratentorial mass lesion, brain relaxation is not predictive of CSFP. Although CSFP values at the extremes of the observed distribution (>17 mm Hg or <6 mm Hg) did correlate with brain relaxation, within the range of 6–17 mm Hg, CSFP did not predict brain relaxation. Additionally, the data from this study suggest that in patients undergoing elective craniotomy for resection of a supratentorial mass lesion, tight brain may occur with a lower frequency in patients receiving 0.5 MAC ISO or DES with 50% N2O than in patients receiving 1 MAC ISO or DES.