THE recent isolation of asparagine from an enzymic digest of edestin [Damodaran, 1932] provided the first direct evidence for the existence of a dicarboxylic acid amide in the protein molecule. As was pointed out in that paper it is, perhaps, from the point of view of demonstrating the validity of the amide hypothesis more important to isolate glutamine, because the majority of proteins on acid hydrolysis yield much larger amounts of glutamic acid than aspartic acid. But glutamine was known to be unstable in aqueous solution [Chibnall and Westall, 1932] so that at the end of the period required for the digestion of a protein to the amino-acid stage only a portion of the glutamine set free would still be present as such in the digest. Furthermore, the work of Schulze on the isolation of glutamine from plant extracts suggested that the separation of this amide from the digest would be difficult, and it was for these two reasons that the isolation of asparagine, which is stable in aqueous solution, and which crystallises readily in the presence of other amino-acids, was first attempted. With the experience thus gained a successful attempt has now been made to isolate glutamine from an enzymic digest of gliadin. This protein was chosen because of the high proportion of glutamic acid which it gives on acid hydrolysis; in terms of total protein-N it contains 25-7 % of amide-N, 22-9 % of glutamic acid-N, 1-17 % of hydroxyglutamic acid-N, and only 05 % of aspartic acid-N. In other words, if the validity of the peptide and amide hypotheses be assumed the gliadin molecule should yield 42-2 % of glutamine.At the outset an unforeseen difficulty arose, in that this prolamine was not as susceptible to the action of proteolytic enzymes as, for example, the edestin used in the asparagine research. While with this, and many other of the better known proteins, the successive action of pepsin, trypsin and either intestinal erepsin or yeast dipeptidase could accomplish the liberation of 90 % of the amino-N obtained on complete hydrolysis [Damodaran, 1932; Frankel, 1916], with gliadin it was found that the same enzymes split little more than 70 % of the total peptide bonds. The phenomenon, which invites further