Dr Swallow’s observations of the vertical profile of currents show how different a view of the deep ocean is now emerging. The tendency of currents to increase with depth below the permanent thermocline is so striking that it seems timely to point out that just this possibility was suggested recently by linearized theory (Rhines 1970). The work showed how short, yet quasi geostrophic, Rossby-topographic waves have a dominant mode that ceases to be independent of depth, and instead has faster currents near the bottom, due to the combined action of rotation, stratification and bottom slopes. This comment is not the first (nor probably the last) application of linear theory to such observations, for both Longuet-Higgins (1965) and Phillips (1966) have suggested their interpretation as Rossby waves of a constant-depth ocean model. Noting this vertical shear, Phillips was forced to invoke a mixture of the familiar baroclinic and barotropic modes. There are, however, objections to this, since those baroclinic modes are exceedingly slowly propagating, and cannot be expected to persist with such strong currents. Also, as Dr Swallow points out, the current directions do not vary greatly in the vertical, suggesting an explanation as a single mode rather than a mixture. The bottom slopes, however, can greatly alter both horizontal propagation and vertical stucture. In particular the presence of vertical shear no longer implies impossibly small frequencies.