(1) By incubation in 0.1 M NaOH for 10 min at room temperature, it is possible to "saponify" some of the methyl carboxylate linkages in bulk yeast tRNA. By incubation with S-adenosyl(Me-14-C)methionine and either homologous (yeast) or heterologous (wheat-embryo) enzymes, it is then possible to "re-esterify" the "saponified" tRNA and thereby effect selective labelling at 5-carboxymethyluridine (Me-14-C)methyl ester residues. (2) There is also selective labelling at 2-thio-5-carboxymethyluridine (Me-14-C)methyl ester residues when "saponified" yeast tRNA is incubated with S-adenosyl(Me-14-C)methionine and homologous (but not heterologous) enzymes. (3) When selectively labelled yeast tRNA is hydrolyzed by RNase T-1, both 5-carboxymethyluridine (Me-14-C)methyl ester and its 2-thio-analogue are released as part of large oligonucleotides, each of which contains roughly 10 nucleotide residues. (4) There are at least three, and possibly four (Me-14-C)methyl ester-containing oligonucleotides released by RNase T-1 digestion of selectively labelled "saponified" yeast tRNA. A comparison of the chromatographic properties of the different (Me-14-C)oligonucleotides suggests that the same 5-carboxymethyluridine residues are probably targets for both homologous and heterologous enzymes. (5) The properties of the selectively labelled oligonucleotides are consistent with the view that some of them probably are derived from yeast tRNA-3-Glu, tRNA-2-Lys, and tRNA-3-Arg, all of which are known to contain 5-carboxymethyl methyl esters as part of their anticodon sequences.