In the present study, in order to simplify the procedure of helper T lymphocyte precursor (HTLp) and cytotoxic T lymphocyte precursor (CTLp) assays, reduce the sample size required and compare the sensitivity of HTLp and CTLp assays in the detection of alloreactive responses, the same set of limiting dilution mixed lymphocyte cultures was used and evaluated for both HTLp and CTLp frequencies. Ninety-four normal volunteer or patient/donor pairs were tested. High frequencies of HTLp (1:385 to 1:28,571) and CTLp (1:554 to 1:25,000) were detected for HLA unmatched unrelated individuals. The range of HTLp and CTLp frequencies of HLA identical siblings was 1:67,000 to 1:1,000,000 and 1:3700 to 1:1,000,000, respectively. In HLA matched unrelated pairs both HTLp and CTLp frequencies ranged from 1:1000 to 1:500,000, overlapping the values obtained from the former two groups. High CTLp and HTLp frequencies were found in 73% and 84% of HLA matched unrelated patient/donor pairs, respectively, suggesting that these in vitro functional assays may provide additional and essential information for the selection of unrelated bone marrow donors. The results of this study support the hypothesis that the combined single limiting dilution system is an efficient and informative method to quantitatively analyse HTLp and CTLp frequencies against major and minor histoincompatibilities with the advantage of being less time-consuming, less laborious and requiring fewer lymphocytes. The results also suggest that the HTLp assay is at least as efficient as the CTLp assay for the detection of alloreactivity other than of HLA identical siblings and may be useful for predicting graft-versus-host disease after bone marrow transplantation using HLA matched unrelated donors. The clinical value of this modified limiting dilution assay to predict graft-versus-host disease in bone marrow transplantation using HLA identical sibling donors and matched unrelated donors is being assessed.