It is intended to select for discussion certain aspects of the mucosal relief pattern in malignant disease of the stomach. From the textbooks it can be seen how difficult it is to define relief changes with concise terms such as filling defect. Their visualisation, differing from the older classical method, is fully described in Barclay's monograph of the Digestive Tract, to which little can be added. We have found the use of the distinctor as devised by Holzknecht for fixation and graded compression of selected regions very effective. Well aware of the wise words written by Barclay 10 years ago, enjoining us not to be too rigid in our rules for interpreting the barium pattern, I take the view that there are some pattern changes sufficiently characteristic of malignant gastric neoplasm to be decisive. Others are only the expression of wall infiltration. But coexistence of two or more of these will usually enable one to arrive at a sound conclusion. An attempt seems, therefore, justified to summarise some of those changes in the mucosal relief encountered in gastric cancer, and to raise points of resemblance to and differentiation from non-malignant conditions.The rugal pattern is blotted out by stretching of the mucosa over an elevated infiltration. Or, with the mucosa destroyed, by protrusion of a tumour with a smooth surface. This often coexists with nodule formation. (See Figs. 4 and 9.)Protrusions devoid of barium are separated by coarse barium containing crevices of different calibre and irregular course.