In many circumstances plasma free tryptophan correlated better than plasma total tryptophan with brain tryptophan concentration (immobilization, fasting, acute liver failure, some drugs). Also, using a modified Oldendorf method it was found that changes of plasma tryptophan binding considerably affected brain tryptophan uptake. Usually, changes of plasma tryptophan binding and non-esterified fatty acid concentration were associated. This led either to changes of plasma free and brain tryptophan concentrations (see above) or to "buffering" in which the proportion of plasma tryptophan in the free state changed but not its concentration. The plasma free tryptophan-brain tryptophan relationship was confirmed in rats after portocaval anastomosis or sham operation. In these experiments brain tryptophan changes did not correlate with plasma amino acids competing with tryptophan for transport to the brain. Determinations on plasma, lumbar and ventricular CSF from psychiatric patients suggest that plasma free tryptophan concentration provides an index of CSF tryptophan and 5-hydroxyindoleacetic acid concentrations and hence of 5-hydroxytryptamine turnover in the human central nervous system.