This paper refers to the work of Moeckel (1952) on the interaction of an oblique shock wave with a shear layer in steady supersonic flow and the work of Chester (1955) and Chisnell (1957) on the propagation of a shock wave down a non-uniform tube. It is shown that their basic results can be obtained by the application of the following simple rule. The relevant equations of motion are first written in characteristic form. Then the rule is to apply the differential relation which must be satisfied by the flow quantities along a characteristic to the flow quantities just behind the shock wave. Together with the shock relations this rule determines the motion of the shock wave. The accuracy of the results for a wide range of problems and for all shock strengths is truly surprising.

The results are exactly the same as were found by the authors cited above. The derivation given here is simpler to perfom (although the original methods were by no means involved) and of somewhat wider application, but the main reason for presenting this discussion is to try to throw further light on these remarkable results.

In discussing the underlying reasons for this rule, it is convenient to use the propagation in a non-uniform tube as a typical example, but applications to a number of problems are given later. A list of some of these appears at the beginning of the introductory section.

In most work on the theory of stability of laminar flow, infinitesimal disturbances only have been considered, so that only the initial growth of the disturbance has been determined. It is the object of the present paper to extend the theory to larger amplitudes and to study the mechanics of disturbance growth with the inherent non-linearity of the hydrodynamical system taken into account.

The Reynolds stress (where averages are taken with respect to some suitable space coordinate) is the fundamental consequence of the non-linearity, and its effects can be anticipated as follows. Initially a disturbance grows exponentially with time according to the linear theory, but eventually it reaches such a size that the transport of momentum by the finite fluctuations is appreciable and the associated mean stress (the Reynolds stress) then has an appreciable effect on the mean flow. This distortion of the mean flow modifies the rate of transfer of energy from the mean flow to the disturbance and, since this energy transfer is the cause of the growth of the disturbance, there is a modification of the rate of growth of the latter.

It is suggested that, in many cases, an equilibrium state may be possible in which the rate of transfer of energy from the (distorted) mean flow to the disturbance balances precisely the rate of viscous dissipation of the energy of disturbance. A theory based on certain assumptions about the energy flow is given to describe both the growth of the disturbance and the final equilibrium state, and application is made to the cases of Poiseuille flow between parallel planes and flow between rotating cylinders. The distorted mean flow in the equilibrium state can be calculated and from this, in the latter case, the torque required to maintain the cylinders in motion. Good agreement is obtained with G. I. Taylor's measurements of the torque for the case when the inner cylinder rotates and the outer cylinder is at rest.

When a layer of fluid is heated uniformly from below and cooled from above, a cellular regime of steady convection is set up at values of the Rayleigh number exceeding a critical value. A method is presented here to determine the form and amplitude of this convection. The non-linear equations describing the fields of motion and temperature are expanded in a sequence of inhomogeneous linear equations dependent upon the solutions of the linear stability problem. We find that there are an infinite number of steady-state finite amplitude solutions (having different horizontal plan-forms) which formally satisfy these equations. A criterion for ‘relative stability’ is deduced which selects as the realized solution that one which has the maximum mean-square temperature gradient. Particular conclusions are that for a large Prandtl number the amplitude of the convection is determined primarily by the distortion of the distribution of mean temperature and only secondarily by the self-distortion of the disturbance, and that when the Prandtl number is less than unity self-distortion plays the dominant role in amplitude determination. The initial heat transport due to convection depends linearly on the Rayleigh number; the heat transport at higher Rayleigh numbers departs only slightly from this linear dependence. Square horizontal plan-forms are preferred to hexagonal plan-forms in ordinary fluids with symmetric boundary conditions. The proposed finite amplitude method is applicable to any model of shear flow or convection with a soluble stability problem.

At sufficiently high temperature the oxygen and nitrogen molecules in air dissociate into atoms. Energy is required for this decomposition and, furthermore, if the temperature is not uniform, concentration gradients are formed and energy is transferred by the consequent interdiffusion of atoms and molecules. In this paper, a simple model of a gas is formulated to illustrate the effect of these processes on heat transfer in three situations: (1) in a layer of gas at rest between two walls at different temperature, (2) in Couette flow, and (3) in a laminar boundary layer on a flat plate.

In the model, the law for the rate of reaction (dissociation and recombination) is of an especially simple form with the speed of reaction characterized by a single parameter. When the parameter becomes large, the gas approaches chemical equilibrium; at the other extreme, no reaction occurs within the gas.

Two chemical conditions of the walls are considered, one being that the walls are catalytic (the surface reaction rate is assumed sufficiently high to hold the gas at the wall in chemical equilibrium), and the other that the walls have no effect on the reaction, i.e. they are non-catalytic.

The most important of the simplifications made are: (a) The reaction rate law is put in a form in which the equilibrium concentration of atoms varies linearly with temperature. Thus, there is only one temperature at which the gas is undissociated instead of the actual range of temperature. (b) In problems (1) and (2), μ, κ, and ρD are taken to be constant. (c) In problem (3) the Lewis number is assumed to be unity. (d) Only binary mixtures of atoms and molecules are considered (so that ionization and more complex dissociation processes are not covered). (e) Coupling of irreversible flows, such as that causing thermal diffusion, is neglected. (f) Only problems in which the pressure is constant are considered.

The transfer of energy through a dissociated diatomic gas in Couette flow is considered, taking oxygen as a numerical example. The two extremes of chemical equilibrium flow and chemically frozen flow are dealt with in detail, and it is shown that the surface reaction rate is of prime importance in the latter case. The chemical rate equations in the gas phase are used to estimate the probable chemical state of the gas mixture, this being deduced from the ratio of a characteristic chemical reaction time to a characteristic time for atom diffusion across the layer. The influence of the surface reaction appears to spread outwards through the flow from the wall as gas-phase chemical reaction times decrease. For practical values of the surface reaction rate on a metallic wall, the energy transfer rate may be significantly lower in chemically frozen flow than in chemical equilibrium flow under otherwise similar circumstances.

Similar phenomena to those discussed will arise in the more complicated case of boundary layer flows, so that a treatment of the simpler type of shear layer represented by Couette flow may be of some value in assessing the relative importance of the various parameters.

An analysis has been made of the two-dimensional flow of an incompressible constant-conductivity fluid through an elliptically shaped solenoid containing a constant magnetic field directed normal to the flow plane. The effect of both Hall current and ion slip has been included in the generalized Ohm's law used for the fluid. The analysis is based on a perturbation procedure in two parameters, one being the magnetic Reynolds number Rm and the other the ratio S of magnetic force per unit area to dynamic pressure. Calculations have been carried to the first order in each parameter, and closed-form analytic expressions have been obtained for the force and moment on the solenoid, the current density, stream function, magnetic field and other pertinent physical quantities.

It was found that, to the zeroth order, there is a force but no moment on the solenoid. To the first order in S, where the flow field is modified but the magnetic field is not, there is a moment and a force, the latter being anti-parallel to the zeroth order force. To the first order in Rm, where the magnetic field is modified but the flow field is not, there is a moment but no force. Thus, to the first order the lift to drag ratio is the same as in the zeroth order. Graphs which illustrate some of the effects of angle of attack, fineness ratio of the ellipse, Hall current and ion slip, on the forces and moments are presented.

The paper deals with the calculation of steady heat transfer from a surface of arbitrary temperature distribution to a laminar semi-infinite stream of arbitrary velocity distribution. Lighthill's method is improved by a correction which accounts for the departures from linearity of the velocity profile within the thermal boundary layer, and which comprehends the influences of Prandtl number, pressure gradient, body forces, and non-coincident start of velocity and thermal layers. Methods are given for evaluating the total heat flux directly, and for integrating the differential equations for the growth of the boundary layer thickness by means of quadratures.

Shock waves generated in a shock tube by use of hydrogen or helium as a driver gas and air, nitrogen, oxygen or argon as a driven gas have higher velocities than predicted by simple theory when sufficiently large diaphragm pressure ratios are used. Expected shock-tube performance curves have been constructed using the equilibrium Hugoniot for the driven gas, both for the usual model of shock tube flow, which assumes instantaneous diaphragm removal, and for a suggested model based on a finite rupture time for the diaphragm. Agreement between experiment and the latter model is in general good, and the differences are qualitatively accounted for by the pressure waves expected to result from mixing between driver and driven gases at the contact surface. These waves may be either compression or expansion waves, depending on the relative heat capacities of the two gases. The maximum shock strength observed as a shock goes down the tube was found to occur at a distance from the diaphragm which increases with the shock strength, and the strongest shocks were found to be still accelerating at the end of a 42 ft. long shock tube of 3 1/2 in. square cross-section. Diaphragm breaking time has been measured and found to be consistent with the observations on the shock formation distance.

The approximate Tricomi equation relevant to sonic speed of two-dimensional small-distrubance flow is solved by separation of variables, where these are certain stipulated functions of the Cartesian velocity perturbations. The symmetric flow patterns obtained from this solution are shown to correspond to those about half-bodies, whose ordinates vary as xn where 0·4 &\les n,<1, and x is the distance along the plane of symmetry. The surface pressures on such bodies are deduced.

In particular, the body whose ordinates vary as $x^{2|5}$ has a sonic surface velocity, except at the nose, where an edge-force which causes a drag force is shown to exist. On the assumption that a free-stream breakaway (at sonic velocity) occurs at the shoulder of a body, this solution thereby yields the flow about an aerofoil of the same shape having a flat base. This bluff-nosed section has only a little more than half the drag of a wedge on the same chord and base.

This paper deals with the stability of a two-dimensional laminar jet against the infinitesimal antisymmetric disturbance. The curve of the neutral stability in the (α, R)-plane (α, the wave-number; R, Reynolds number) is calculated using two different methods for the different parts of the curve; the solution is developed in powers of (αR)−1 for obtaining the upper branch of the curve and in powers of αR for the lower branch.

The asymptotic behaviour of these branches is that for branch I,$\alpha \rightarrow 2, \;\; c \rightarrow \frac{2}{3}$ for $R \rightarrow \infty$; and for branch II, $R \sim 1\cdot12\alpha^{-1|2},\; c \sim 1\cdot 20 \alpha^2$ for α → 0. Some discussion is given on the validity of the basic assumption of the stability theory in relation to the numerical result obtained here.

This paper examines the flow characteristics of a body of small slope planing at high Froude number over a water surface. An equation is obtained relating the slope of the planing surface to an integral containing the pressure distribution on the planing surface. The equation is expanded for large Froude number and a solution is obtained by an iteration process. At each stage of the iteration process the integral equation of ordinary thin aerofoil theory is solved. The pressure distribution on the planing surface is derived as a series in inverse powers of the Froude number F, as far as the F−4 term. Computations are performed for the planing of a flat plate, a parabolic surface, and a suitable linear combination of these shapes which results in a flow without a splash at the leading edge.

Assuming local thermodynamic equilibrium in the fluid, an expression is derived for the rate of destruction of the mean square of the temperature fluctuations by radiative transfer of heat. This takes a particularly simple form (a) if the fluid is effectively transparent over distances equal to the scale of the turbulent motion, when the effect appears as a decay time for temperature fluctuations from the mean, and (b) if the fluid is effectively opaque, when the effect is of an increased conductivity due to radiation. A theory of the interaction of the temperature and velocity fields developed in a previous paper shows that, if the radiative effects are relatively weak, a sudden collapse of the turbulent motion occurs while the flux Richardson number is still less than one. If the radiative effects are strong, the turbulent intensity approaches zero as the flux Richardson number approaches one. The effects of radiation are always to increase the critical value of the ordinary Richardson number. Criteria for fully turbulent motion of an unrestricted flow are given in terms of the gradients of mean velocity and mean temperature and of the rate of radiative cooling. The relevance of these calculations to motions of the atmosphere is briefly discussed.

The equations of motion for a chemically reacting gas in the absence of viscosity and heat conduction are set up. It is shown that the characteristic speed defined by this set of equations is the high-frequency limit of the phase velocity of sound waves as long as the reaction rate is finite. At infinite reaction rate (chemical equilibrium) the characteristics suddenly change to the lowfrequency sound speed. The nature of this transition is discussed in connection with a recent paper of Resler (1957).

A set of two-dimensional subsonic flows past certain cylinders is obtained using hodograph methods, in which the true pressure-volume relationship is replaced by various straight-line approximations. It is found that the approximation obtained by a least-squares method possibly gives best results. Comparison is made with values obtained by using the von Kármán-Tsien approximation and also with results obtained by the variational approach of Lush & Cherry (1956).

An extension of the linearized theory of supersonic flow past quasi-cylindrical bodies of almost circular cross-section has been found which enables a direct calculation to be made of the overall forces on wings mounted on such bodies, subject to certain restrictions on the plan-form. The method is applied to two examples: (i) the effect of an arbitrary body distortion on static stability at supersonic speeds; and (ii) the effect of wing-body interference on rectangular wings mounted on a cylindrical body. The drag calculations in the second example are compared with the results of the supersonic area rule, which is found to be in error for moderate values of the ratio of wing chord to body radius, though the discrepancy is not serious from a practical point of view.

The subterranean mixing in permeable media of sea water and ground water is studied. The model for this mixing process which was suggested by C. K. Wentworth is adopted, but is soon discarded in favour of a more tractable formulation whose equivalence to the original model is established. The analysis is carried to the point where the determination of the salinity distribution of the ground water in a given subsoil requires only the solution of an elementary linear ordinary differential equation.

The plane steady decelerated flow of a dust-gas mixture is analysed in an approximate manner. The problem, which has a five-parameter family of solutions, is reduced to a form such that the analysis can be completed by the integration of a first-order non-linear differential equation and a quadrature. A few integral curves are given and the characterizing features of the flow field are discussed.

This paper describes an experimental investigation of the form of the large scale motions in turbulent flow. These motions have been found to be more ordered than has usually been supposed and their origin and dynamics are discussed in terms of physical models of typical eddies.

Nine components of the double velocity correlation tensor have been measured at a number of positions in the wake of a circular cylinder and in a ‘flat plate’ boundary layer. These have been supplemented by measurements of correlations with separations in directions other than the axial ones. In the wake, the correlations at large values of the separation are explained in terms of two types of large scale motion. One of these is a pair of vortices, side by side and rotating in opposite directions with axes aligned approximately normal to the plane of the wake. The other typical motion is a series of jets in which turbulent fluid is projected outward from the core of the wake. It is suggested that these are the result of an instability of the turbulent shear stress. A qualitative explanation of the apparent structural equilibrium of the wake is given in terms of this instability. The vortex pair eddies were not found in the boundary layer but there is evidence of jets much like those in the wake.

Correlations measured in grid turbulence have been found to be highly anisotropic and consistent with the presence of vortex pair eddies. When a plane strain was applied to grid turbulence, the effect on the correlations suggested the presence of a stress instability similar to that postulated for the wake.

Four alternative theoretical treatments of ‘displacement thickness’, and, generally, of the influence of boundary layers and wakes on the flow outside them, are set out, first for two-dimensional, and then for three-dimensional, laminar or turbulent, incompressible flow. They may be called the methods of ‘flow reduction’, ‘equivalent sources’, ‘velocity comparison’ and ‘mean vorticity’.

The principal expression obtained for the displacement thickness δ1 in three-dimensional flow may be written $\delta_1 = \delta_x - \frac{1}{Uh_y} \frac {\partial}{\partial y}\int^x_0 \delta_y dx,$ if, as orthogonal coordinates (x, y) specifying position on the surface, we choose x as the velocity potential of the external flow, and y as a coordinate, constant along the external-flow streamlines, such that hydy is the distance between (x, y) and (x, y + dy); and if also δx and δy are the streamwise and transverse ‘volume-flow thicknesses’ $\delta_x = \frac {1}{U}\int _0^\infty (U - u)\;dz,\;\;\;\;\;\;\delta_y = \frac {1}{U}\int ^\infty_0 v\; dz,$z is the distance from the surface, u and v are the x and y components of velocity, and u takes the value U just outside the boundary layer.

A new water tunnel, incorporating a slotted wall working section, was found to suffer from severe vibration. A theoretical explanation is given for this, together with experimental evidence gleaned from this water tunnel and a small wind tunnel. It is shown that the oscillations are hydrodynamic in origin and are associated with the slotted wall design. Consideration is given to methods of elimination or reduction of the oscillations.