Serum beta 2 microglobulin (P2m) and C-Reactive protein (C-RP) were measured in patients with Hodgkin's disease (HD) and non-Hodgkin's lymphoma (NHL). In a vertical study of 215 patients with HD and 171 with NHL, the serum P2m was raised in 8% of stage I and I1 and 47% of stage I11 and IV active untreated HD; the corresponding frequencies for patients in "complete remission" (CR) off treatment were 4% and 18%. In active NHL, p2m was elevated in 23% of stage I and I1 disease and 60% of stage I11 and I V disease; the frequencies for CR were 10% and 21%. The serum C-RP levels afforded interesting comparison: values >20 mg/l were found in 31% of stage I and I1 and 53% of stage I11 and I V active untreated HD. The corresponding frequencies for the CR groups were 5% in both. In active NHL, C-RP was elevated in 23% of stage I and I1 and 40% of stage 111 and 1V: the frequencies for the CR groups were 7% and 2%. When histopathological grade rather than clinical stage was considered in NHL, a raised P2m was found in association with "unfavorable" histologies in 78% of all untreated patients and 100% of stage IV disease. The detailed investigation of smaller numbers of patients including extended longitudinal studies provided additional information: the serum e2m was found to be independent of changes in serum C-RP, acid glycoprotein (a,AGP), fibrinogen, and the erythrocyte sedimentation rate (ESR). It was elevated in advanced untreated disease and persistent elevation probably reflected resistant or partially responsive disease. Chemotherapy produced a reduction in serum P2m levels with each individual course of treatment; this change was either part of a continuing overall reduction to normal or was transient, producing a characteristic "zig-zag" pattern. Biochemical tests in relation to fixed time points have limitations. Longitudinal studies can probably provide crucial information in only a small number of patients who are already being closely monitored. Nevertheless, serum p2m and C-RP levels can furnish useful information of potential clinical value and prognostic significance.Cancer 45318-326, 1980.