In this study I investigated the analytical and clinical performance of the measurement of the free beta-subunit of choriogonadotropin (hCG) in normal pregnancies and in pregnancies affected by Down's syndrome. Free beta-hCG in maternal serum has been shown to be increased in Down's syndrome-affected pregnancies and is proportionally increased in more cases than is total hCG. This study confirms previous findings of low concentrations of unconjugated estriol and alpha-fetoprotein in maternal serum in Down's syndrome-affected pregnancies. Using a multivariate risk analysis of maternal age and concentrations of alpha-fetoprotein, unconjugated estriol, and hCG in maternal serum, I determined that, at a risk cutoff value of 1 in 300, 52% of Down's cases could be detected with total hCG in the calculation, compared with 66% with the free beta-hCG concentration. The false-positive rate was 5.9% in both cases. Therefore, free beta-hCG can be used effectively in a screening program for Down's syndrome; however, further studies are required to ascertain whether the measurement of free beta-hCG has any advantages over the use of total hCG for detecting Down's syndrome.