The incorporation of tritium into the C-8 position of purine residues in yeast tRNA(Phe) is shown to be markedly dependent on the conformation of the molecule. The completely unfolded molecule incorporates tritium at a rate commensurate to that expected for free purine nucleotides, but partially or completely folded forms incorporate proportionately less. The labeling of specific purine sites is determined by digesting the nucleic acid with a specific nuclease and analyzing each of the radioactive fragments produced. This analysis reveals that the amount of labeling of a purine in the folded form is strongly dependent upon its position in the sequence, whereas purine labeling in the unfolded form is independent of sequence position. The labeling pattern of the different bases in the folded form agrees fairly well with what is expected, based on existing solution and x-ray data on the conformation, i.e., residues involved in helical sections or apparently masked by tertiary interactions label more slowly than those on the "outside" of the molecule. Finally, folding and unfolding of specific regions of the macromolecule may be followed by the isotope labeling.