A small drop placed on a horizontal surface will spread under the action of capillary forces until it reaches an equilibrium position. The rate at which it spreads provides a means for testing certain hypotheses about moving contact lines; namely that there must be slip between the fluid and the solid boundary near the rim of the drop to avoid a force singularity there, and that the contact angle measured at the rim itself does not show the dynamic behaviour observed by measurements that ignore rapid changes in slope in the immediate vicinity of the rim but remains equal to its static value.

By the use of matched asymptotic expansions, an equation for the rate of spread of a drop as a function of the radius of the contact circle is obtained. Experiments on the spreading of small drops of molten glass allow a comparison to be made between the spreading of a drop determined experimentally and that predicted theoretically, which supports the use of the proposed hypotheses as appropriate for the study of fluid motions containing moving contact lines.

The object of this paper is to derive the added mass and damping coefficients associated with the periodic motions of a floating hemisphere. Two physically distinct cases are considered; namely those of heave and surge, where these nautical terms refer respectively to a vertical or horizontal oscillation of the body. Computations have been done and the values found for the various force coefficients are presented in tabulated form. A brief derivation of the long- and short-wave asymptotics of these coefficients has also been included.

Cut-off frequencies are well known in acoustic ducts to be the thresholds of propagation and evanescence. If at one end of a duct the piston oscillates at very near the cut-off frequency, cross-duct resonance occurs and the linearized theory breaks down. This paper studies the nonlinear response, near a cut-off frequency of a guided wave, as an initial-boundary-value problem. The asymptotic state is shown to be governed by a modified cubic Schrodinger equation. Numerical solutions are then obtained for inputs of finite and long duration. In addition to the characteristics of the input envelope, two quantities control the transient phenomenon: frequency detuning and nonlinearity. Under certain circumstances, energy can be trapped near the piston long after a short-lived input has expired, while for a sustained input there is no sign of a steady state. Dissipation is not considered.

Flow visualization of artificially triggered transition in plane Poiseuille flow in a water channel by means of 10–20 μm diameter tihnium-dioxide-coated mica particles revealed some striking features of turbulent spots. Strong oblique waves were observed both at the front of the arrowhead-shaped spot as well as trailing from the rear tips. Both natural and artificially triggered transition were observed to occur for Reynolds numbers slightly greater than 1000, above which the flow became fully turbulent. The front of the spot moves with a convection speed of about two-thirds of the centreline velocity, while the rear portion moves at about $\frac{1}{3}U_{\rm cl}$. The spot expands into the flow with a spreading half-angle of about 8°. After growing to a size of some 36 times h (the channel depth) at a downstream distance x/h of about 130, the spot began to split into two spots, accompanied by strong wave activity. The spot(s) was followed visually downstream of its origin a distance x/h of about 300. These results indicate that wave propagation and breakdown play a crucial role in transition to turbulence in Poiseuille flow.

An unsteady boundary layer on a rotating disk in a counter-rotating fluid was shown in Bodonyi & Stewartson (1977) to develop a singularity at the axis of rotation after a finite time. The structure proposed to describe the singularity is, however, incomplete, even after an additional term was added by Banks & Zaturska (1981). A description is given here in which the boundary layer is divided into three regions; this description seems to be free of the weaknesses of the earlier studies, and is in good agreement with data from the numerical solution of the governing equations.

Camber distributions are chosen for airfoils of zero thickness that maximize the angular velocity induced by a sudden decrease in free-stream velocity. Those optimal shapes that are in equilibrium in steady forward flow are also neutrally stable in that position.