This study evaluates psychological and immunological functioning after bereavement and the influence of group counseling. Eighteen widows (bereaved within 3 months of enrolment) and a reference group of 10 married control subjects were asked to fill in self-report scales and to donate a blood sample (T1). After T1, half of the widows (the experimental group) were randomly assigned to grief counseling (13 sessions over 4 months), while the other subjects (the control group) received no treatment. Seven months after bereavement (T2) or, in the case of the experimental group, immediately after the intervention, a follow-up was conducted in the widowed subsample using the same measures. Blood samples were analyzed to determine the total number of white blood cells, number of lymphocyte subsets, natural killer cell activity (NKCA) and lymphocyte proliferative response to phytohemagglutinin (PHA), anti-CD3 and pokeweed mitogen (PWM). At T1, we found significant differences between widows and non-widows regarding both psychological and immunological measures. Widows felt more anxious, depressed, hostile and agoraphobic. At T1, widows had a lower number of the CD19+CD5+ B cell subpopulation. The cell function tests for T and B cells showed higher responses in widows (lymphocyte proliferation response to PHA, anti-CD3 and PWM). No significant difference in NKCA was found between widows and non-widows. At T2, there appeared to be no significant difference between widows and non-widows on the psychological measures. With respect to the immunological measures, widows and non-widows showed no significant differences for the total number of white blood cells, number of lymphocyte subsets and NKCA. Consistent with our findings at T1, the lymphocyte proliferation response to PHA, anti-CD3 and PWM at T2 appeared to be higher in widows than in non-widows. Comparing the experimental group (widows) and the control group (widows) with respect to psychological measures at T1, widows in the experimental group felt more insufficient and had more sleep disturbances. With respect to the immunological measures, no differences were found between those two groups. When the same two groups were again compared at T2, no differences were found in any of the psychological or immunological measures (lymphocyte sub-populations, proliferation tests and the NKCA).