Since Aug. 1979 through Dec. 1989, 270patients (200 males, 70 females, age ranged from 10 to 79, median age 44) with, malignant lymphoma underwent bone marrow examination including bone marrow trephine biopsy for histopa-thological examination as well as bone marrow smear for cytological examination. There were 34Hodgkin's lymphomas (HD) and 236 non-Hodgkin's lymphomas (NHL) based on Rye classification and modified working formulation classification (Chengdu conference 5, 1985.) . According to Ann Arbor staging system, patients with staged I were 28, staged II 73, staged III 128, staged IV 35, unknown 6 . There were 107cases with B symptoms and 111 cases were no prior chemotherapy. The incidence of bone marrow involvement (BMI) in 270 cases was 21.1%. The major factors influencing BMI were sex, pathology and clinical stage (female Vs male, P<0.05; HD Vs NHL, P<0.01; I-II Vs III-IV, P<0.001) . BMI was observed most frequently in patients with diffuse small lymphocytic (75%) , small cleaved in diffuse cleaved (50%) and lymphoblastic (29.6%) . After bone marrow examination, 49 cases (18.1%) were upstaged. Our studies revealed that bone marrow trephine biopsy assisted by bone marrow smear (21.1%) was better than biopsy alone (14.8%) or smear (13.5%) alone for the diagnosis of BMI in malignant lymphoma. Follow-up examination showed that BMI disappeared in 90% patients after complete of partial remission . Key words: Malignant lymphoma; Hodgkin's disease; non-Hodgkin's lymphoma; bone marrow examination; bone marrow involvement