Some cases of primary hyperphenylalaninemia are not caused by the lack of phenylalanine hydroxylase, but by the lack of its cofactor tetrahydrobiopterin. These patients are not clinically responsive to a phenylalanine-restricted diet, but need specific substitution therapy. Thus, it became necessary to examine all newborns screened as positive with the Guthrie test for tetrahydrobiopterin deficiency. Methods based on urinary pterin or on specific enzyme activity measurements are limited in their availability, and the simplest method, based on the lowering of serum phenylalanine after loading with cofactor, was discouraged by the finding that some dihydropteridine reductase-deficient patients were unresponsive. The preliminary observation that this limitation could be overcome by increasing the dose of the administered cofactor prompted us to reevaluate the potential of the tetrahydrobiopterin loading test in hyperphenylalaninemia. Fifteen patients, eight with ultimate diagnosis of phenylketonuria, three with 6-pyruvoyl tetrahydropterin synthase-, and four with dihydropterine reductase-deficiency, have been examined by administering synthetic tetrahydrobiopterin both orally, at doses of 7.5 and 20 mg/kg, and i.v., at a dose of 2 mg/kg. All the tetrahydrobiopterin-deficient patients, unlike those with phenylketonuria, responded to the oral dose of 20 mg/kg cofactor by lowering their serum phenylalanine concentration markedly below baseline to an extent easily detectable by Guthrie cards. This method allows for a simple screening method when enzyme or pterin studies are not available.