Some simple similarity solutions are presented for the flow of a viscous fluid near a sharp corner between two planes on which a variety of boundary conditions may be imposed. The general flow near a corner between plane boundaries at rest is then considered, and it is shown that when either or both of the boundaries is a rigid wall and when the angle between the planes is less than a certain critical angle, any flow sufficiently near the corner must consist of a sequence of eddies of decreasing size and rapidly decreasing intensity. The ratios of dimensions and intensities of successive eddies are determined for the full range of angles for which the eddies exist. The limiting case of zero angle corresponds to the flow at some distance from a two-dimensional disturbance in a fluid between parallel boundaries. The general flow near a corner between two plane free surfaces is also determined; eddies do not appear in this case. The asymptotic flow at a large distance from a corner due to an arbitrary disturbance near the corner is mathematically similar to the above, and has comparable properties. When the fluid is electrically conducting, similarity solutions may be obtained when the only applied magnetic field is that due to a line current along the intersection of the two planes; it is shown that the effect of such a current is to widen the range of corner angles for which eddies must appear.

Ten years have elapsed since the passing of George Keith Batchelor (8 March 1920–30 March 2000), formerly Professor of Fluid Dynamics at the University of Cambridge, and Founder Editor of the Journal of Fluid Mechanics. It is fitting to remind the readers of this Journal what a great scientist he was, both in respect of his own contributions to our subject, and even more in respect of his inspirational influence on generations of research students and younger colleagues, and also more widely on the international stage, on which he was a revered, if sometimes controversial, personality.

Stokes flow of a viscous fluid in a cylindrical container driven by time-periodic forcing, either at the boundary or through oscillation of the cylinder about an axis parallel to its generators, is considered. The behaviour is governed by a dimensionless frequency parameter $\eta$ and by the geometry of the cylinder cross-section. Various cross-sections (square, rhombus, and sector of a circle) are first treated by either finite-difference or analytic techniques and typical transitions of streamline topology during flow reversals are identified. Attention is then focused on the asymptotic behaviour near any sharp corner on the boundary. For small $\eta$, a regular perturbation expansion reveals the manner in which local flow reversal proceeds during each half-cycle of the flow. The behaviour depends on the corner angle, and different regimes are identified for both types of forcing. For example, for an oscillating square domain, eddies grow symmetrically from each corner and participate in the subsequent flow reversal in the interior. For large $\eta$, the corner eddies merge into Stokes-type boundary layers which drive the interior flow-reversal process. In general, the local corner analysis provides a key to an understanding of the global flow evolution.

The analysis reported in Part 1 is extended here to the case in which the conductivity κ is large compared with the viscosity ν, the conduction ‘cut-off’ to the θ-spectrum then being at wave-number (ε/κ3)¼. It is shown, with a plausible and consistent hypothesis, that the convective supply of $\overline {\theta^2}$-stuff to Fourier components of θ with wave-numbers n in the range (ε/κ3)¼ [Lt ] n [Gt ] (ε/ν3)¼ is due primarily to motion on a length scale of order n-1 acting on a uniform gradient of θ of magnitude $[(\overline {\nabla \theta)^2}]^{\frac {1}{2}}$. The consequent form of the theta;-spectrum within this same wave-number range is $\Gamma (n) = \frac {1}{3}C \chi \epsilon ^{\frac {2}{3}} k^{-3}n^ {-\frac {17} {3}}.$

The way in which conduction influences (and restricts) the effect of convection on the distribution of θ at these wave-numbers beyond the conduction cut-off is discussed.

Frictionless flows with finite voticity are usually made determinate by the imposition of boundary conditions specifying the distribution of vorticity ‘at infinity’. No such boundary conditions are available in the case of flows with closed streamlines, and the velocity distributions in regions where viscous forces are small (the Reynolds number of the flow being assumed large) cannot be made determinate by considerations of the fluid as inviscid. It is shown that if the motion is to be exactly steady there is an integral condition, arising from the existence of viscous forces, which must be satisfied by the vorticity distribution no matter how small the viscosity may be. This condition states that the contribution from viscous forces to the rate of change of circulation round any streamline must be identically zero. (In cases in which the vortex lines are also closed, there is a similar condition concerning the circulation round vortex lines.)

The inviscid flow equations are then combined with this integral condition in cases for which typical streamlines lie entirely in the region of small viscous forces. In two-dimensional closed flows, the vorticity is found to be uniform in a connected region of small viscous forces, with a value which remains to be determined—as is done explicitly in one simple case—by the condition that the viscous boundary layer surrounding this region must also be in steady motion. Analogous results are obtained for rotationally symmetric flows without azimuthal swirl, and for a certain class of flows with swirl having no interior boundary to the streamlines in an axial plane, the latter case requiring use of the fact that the vortex lines are also closed. In all these cases, the results are such that the Bernoulli constant, or ‘total head’, varies linearly with the appropriate stream function, and the effect of viscosity on the rate of change of vorticity at any point vanishes identically.

The effect of turbulence on an applied magnetic field is considered in the case when the magnetic Reynolds number Rm is large compared with unity but small compared with the ordinary Reynolds number R of the turbulence. When the applied field is sufficiently weak, it is argued that its effect on the velocity field is negligible. The equation for the field is then linear and its spectrum may be obtained throughout the equlibrium range of wave-numbers. It appears that the spectrum increases as $k^{\frac {1}{3}}$ up to a wave-number kc marking the threshold of conduction effects, and falls off as $k^{-\frac {11}{3}}$ beyond kc. The net effect of the turbulence is expressed in terms of an eddy conductivity equal to $R_m^{- \frac {5}{2}}$ times the electrical conductivity of the fluid. The effect of magnetic forces when these are not negligible is also tentatively considered.