by the locus-specific primer is directly subjected to TaqI digestion and is analyzed on an agarose gel without radioisotopic hybridization. It also appears to be more accurate because all three distinct chimeric CYP21P/ CYP21 genes had 3.2-kb fragments (Fig. 1C, lanes 3 and 4). Simply assuming that the 3.2-kb fragment from conventional Southern analysis (6 ) is a deletion of the CYP21 gene has been shown to be incorrect. The authors of one report have stated that the 3.2-kb fragment analyzed by Southern blotting identified a fused CYP21 gene, but in this study, Southern blotting was performed with radioisotope hybridization and a different endonuclease (8 ).In conclusion, the coexistence of 9.3-and 3.2-kb fragments is consistent with a chimeric CYP21P/CYP21 gene. The method of analysis we have described is a novel tool for identification of such a molecule. These chimeric genes were found to occur frequently in CAH caused by steroid 21-hydroxylase deficiency in a Taiwanese (ethnic Chinese) population. The development of benign prostatic hyperplasia (BPH) is dependent on androgens, primarily dihydrotestosterone (DHT) (1 ). To estimate the activity of 5␣-reductase, which catalyzes the conversion of testosterone to DHT, testosterone and DHT have been quantified in biological samples (2-4 ). Because of their lipophilic properties, they are usually found in the conjugated form, i.e., linked to a hydrophilic sulfuric moiety or -glucuronic acid, which are excreted mainly (Ͼ95%) in human urine. Despite reports of a close association between BPH and testosterone and DHT (1-4 ), to our knowledge, there is no published simultaneous quantitative data for these metabolites as their glucuronides in the urine of patients with BPH.The major problem associated with quantification of total steroids is incomplete hydrolysis, attributable mainly to the matrix effects of urine on the efficiency of enzymatic hydrolysis, which reduces the reproducibility of the assay (5,6 ). Given that it is often necessary to analyze many samples in a short time with good sensitivity and reproducibility, system throughput is a critical issue for many clinical mass spectrometry (MS) groups. In 322 Technical Briefs