To determine the usefulness of a consulting panel for histopathologic review in lymphoma cases, a comparative study of the histopathologic diagnosis of the local pathologist was done and compared with the final diagnosis by an expert pathologist in 1088 cases of lymphoma studied by the Southeastern Cancer Study Group (SECSG) during the implementation of five protocols during the last ten years. The following conclusions were reached: (1) In 44 cases (4%), the material sent for review was judged inadequate to make a diagnosis; (2) In 82 cases (8%), the local pathologist's diagnosis was judged incorrect; (3) In Hodgkin's Disease (HD), the diagnosis was confirmed in 545 of 595 cases (92%). However, using the Lukes and Butler classification by subtype of HD, the expert pathologist judged only 289 of 595 cases as correct (49%). The least agreement was found in the lymphocytic predominance (LP) subtype, in which only six of 34 (18%) cases were correct. The best correlation was found in the nodular sclerosis (NS) type of HD where agreement occurred in 161 of 186 (87%) cases; and (4) In the non-Hodgkin's lymphomas (NHL) the diagnosis was confirmed in 428 of the 493 cases studied (87%). Subclassification using the Rappaport system revealed agreement in 274 of 493 (56%) cases. Subtypes associated with good prognosis were judged correct by the expert pathologist in 105 of 139 instances (76%). Likewise, agreement in subtypes of poor prognosis occurred in 254 of 351 cases (72%). These data again confirm the need for an uniform consulting panel system especially for cases involved in cooperative group trials. This has even become more important now with the increasing complexity of the different classifications currently in use.