The efficacy of intrathecal (i.t.) chemoprophylaxis was compared with cranial radiotherapy plus i.t. methotrexate (MTX) in a Southwest Oncology Group (SWOG) study accessing 408 patients from September 10, 1974, to October 29, 1976. Randomization was stratified by prognostic groups (PGs) based on age and white blood cell count at diagnosis. All received induction therapy with vincristine and prednisone (Pred); maintenance therapy consisted of daily 6-mercaptopurine and weekly MTX. Consolidation for arm 1 employed cyclophosphamide and L-asparaginase followed by biweekly 5-day courses of parenteral MTX. The first dose of each course of MTX was given i.t. in triple chemoprophylaxis (MTX, hydrocortisone, and cytosine arabinoside). During maintenance, i.t. chemoprophylaxis was bimonthly and 28-day Pred “pulses” were given every 3 mo. Arm 2 i.t. chemoprophylaxis was initiated on achievement of remission, and arm 3 i.t. on treatment day 1; both continued 1 yr. Arm 4 induction included two doses of L-asparaginase. On achievement of remission, CNS prophylaxis (radiotherapy, 2400 rad plus i.t. MTX) was given. For all, therapy was discontinued after 3 yr of continuous complete remission. Survival and the incidence of extramedullary relapse were similar for the treatments employing either i.t. chemoprophylaxis or radiotherapy plus i.t. MTX upon achievement of remission. Among poor prognosis patients, the duration of complete remission was significantly better with the regimen using i.t. chemoprophylaxis as a component of consolidation therapy than with the regimen employing i.t. chemoprophylaxis early in induction or with the treatment using radiotherapy plus i.t. MTX for CNS prophylaxis. In poor prognosis patients, the initiation of i.t. chemoprophylaxis during consolidation was also associated with hematologic remissions that were significantly better than those achieved with the treatment employing early CNS chemoprophylaxis or with the regimen using radiotherapy plus i.t. MTX. Among average prognosis patients, therapy with CNS chemoprophylaxis during consolidation, as well as the regimen employing radiotherapy and i.t. MTX for CNS prophylaxis, produced hematologic remissions that were significantly longer than those obtained with the regimen using early CNS chemoprophylaxis. Hematologic remissions of good prognosis patients who received treatment with the regimen employing i.t. chemoprophylaxis during consolidation were statistically superior when compared to the regimen employing CNS radiotherapy plus i.t. MTX. This study indicates that i.t. chemoprophylaxis may be substituted for cranial radiotherapy when utilizing effective systemic regimens. Additionally, chemoprophylaxis may be reduced from 3 to 1 yr in patients with good prognostic factors.