Drops formed from the breakup of charged cylindrical liquid jets have been shown to be smaller than those formed in the unelectrified case. This change has been related to electrically induced instabilities on the jet. Small-perturbation analysis was used to formulate the Lagrangian and write the differential equation for the jet. Non-dimensionalized plots of the solutions exhibit the stabilizing or destabilizing influences of surface tension and electric effects, and allow these influences to be related back to liquid physical properties. Drop diameter to be expected from breakup of the electrified jet was calculated as a function of jet diameter, physical properties of the liquid, jet electrification, and the mode of instability dominating the breakup process.

The flow past a circular cylinder moving close to a free surface at high Froude number is investigated by the method of matched asymptotic expansions. In contrast with the linearized solution in which the dimensionless depth of immersion h = h′g/U′2 is kept constant, in the present analysis h → 0 as Fr → ∞.

The inner flow model is that of a non-separated non-linear gravity-free flow past a doublet, while the linear outer solution is that of a singularity a t the free surface. At deep submergence the solution coincides with the linearized solution. At moderate immersion depths the linearized solution is still valid, provided that the depth is replaced by an effective depth, larger than the actual one. For a body close to the free surface the non-linear solution differs significantly from the linearized solution.

The transport of heat and salt across a density interface containing salt fingers is investigated when turbulence produced by vertically oscillating horizontal grids is imposed on the deeper layers above and below the interface. The fluxes of heat and salt are measured as functions of the stirring frequency. The results are discussed with reference to the parameter $\lambda = w^2/(\overline{u})^2$, where w is the velocity of the fluid in the fingers in the undisturbed state and $\overline{u}$ is the r.m.s. horizontal velocity of the turbulence. It is found that the salt flux has a minimum (as a function of the stirring frequency) when Λ ∼ 0·3 and that when Lambda; [lsim ] 0·05 the transport across the interface is dominated by mechanical mixing. The ratio r of the contributions of heat and salt to the buoyancy flux increases with decreasing Λ and r > 1 when Λ [lsim ] 1·3. The latter result implies that if turbulent intensities are such that Λ [lsim ] 1·3 in a particular oceanic situation, mechanical mixing will affect the vertical transport of heat and salt in such a way that the salt finger mechanism will be unable to produce layered temperature and salinity microstructure in the manner described by Turner (1967).

Benjamin (1970) has calculated the upstream influence in open-channel flow, and has argued that a similar effect occurs when a body is moved along the axis of a tube of uniformly rotating fluid. In the present paper Benjamin's work is extended to the case of interfacial waves in a two-fluid system. It is shown that there are certain special flows for which the upstream influence vanishes.

We investigate the behaviour of infinitesimal perturbations introduced into an unstably stratified horizontal Couette flow. We assume that the fluid is Boussinesq and contained in an infinite conducting rectangular channel which is uniformly heated from below. The sidewalls are rigid and the Couette flow is generated by moving them with equal and opposite velocities along the channel. The top and bottom are assumed to be free so that we can separate variables.

Without shear, the preferred modes of convection closely resemble transverse ‘finite rolls’ (Davies-Jones 1970). Shear increases the critical wavelength so that the preferred modes become longitudinally elongated cells, or even longitudinal rolls in some cases. The critical Rayleigh number increases quite rapidly at fist with Reynolds number, but at higher Reynolds numbers it levels off to a constant value (which cannot be greater than the shear-independent Rayleigh number at which longitudinal disturbances fist become unstable).

We also find that the disturbances are tilted in the same direction as the shear, and that the marginally stable ones transfer kinetic energy from the mean flow to the perturbations. Except at low Reynolds numbers, the long wave perturbations gain more energy through the conversion of mean flow kinetic energy than through the release of potential energy, even though the instability is convective in origin.

The discussion concerns free wave motions generated from rest in a finite region of an ocean of heavy liquid lying on a horizontal plane. It is shown that the horizontal fist moment of the free-surface displacement varies linearly with time. Hence, if the total volume displaced is not zero and therefore the centroid of the displacement is definable, the centroid travels with a constant horizontal velocity as the wave motion evolves. This conclusion holds exactly for waves of any amplitude and even remains applicable subsequent to the breaking of waves.

The 23rd EUROMECH colloquium on finite-amplitude and diffusive effects in acoustics was held in Rapperswil (Switzerland), 5–7 April 1971. There were 41 participants from 7 countries, and the authors were the chairmen of the meeting organization. References quoted in this report give the titles of the talks and sources for further details of the work described at the meeting; there will be no other publication of the proceedings. The subject matter of this meeting was more strongly restricted than is indicated by its title, inasmuch as papers motivated solely by sonic-boom research problems were not included. Included in particular were problems of acoustic damping by relaxation, dust, moisture, etc.; damping in ducts; effects of turbulence; acoustic streaming; and thermo-acoustic effects.

Upon review of past experimental results and theoretical efforts it is apparent that the mechanism by which combustion noise is generated is not well understood. A theory of combustion noise is developed in this paper which follows rigorously from the principles of fluid mechanics. Lighthill's approach, used in his studies of aerodynamic noise, is closely followed in the present work. The sound radiated from open, turbulent flames is found to depend strongly upon the structure of such flames; at present their structure is not well known. However, meaningful bounds and scaling rules for the sound power output and spectral content are derived based upon the present limited knowledge. A framework is developed which explains past experimental work and the origin of combustion noise.