The problems which occur in defining the thrust of a jet engine and the internal drag of a ducted body are considered, and formal definitions and names are given for the concepts considered to be of importance.

This paper is concerned with the problem of the skin–and–stringer–covered conical fuselage of circular cross section, having flexible frames and being subjected to an arbitrary distribution of loads applied in the planes of these frames or on the end cross sections. The skin thickness and stringer cross section may vary from bay to bay but are assumed to be constant in a bay, i.e. between any two consecutive frames. The stringers are assumed to be uniformly spaced round the circumference. The frames are assumed to be of uniform cross section (circumferentially) but their stiffnesses are arbitrary, as is also their spacing along the fuselage.

This paper is a sequel to one previously published in the Journal, but the method of analysis is now by minimum strain energy instead of by deflections. Owing to the generality of the investigation it is impossible to present the solution—in terms of the known loads and stiffnesses—in any explicit form, and attention has therefore been concentrated on the technique of numerical computation. A scheme of tabulation has been evolved, and its use is illustrated by a fully–worked numerical example of a moderately complex type.

The theory of stress concentration is such a fascinating study that the whole of this lecture could be most happily devoted to discussion of the stress distribution round all sorts of special boundaries. At the same time the most detailed knowledge of elastic stress distribution is of little value unless we can be sure how the stresses computed will be reflected in the actual strength of the part. For that reason this paper is divided into three sections: the first deals with certain moderately simple cases of stress and load concentration and shows how the stress analysis is confirmed by static and fatigue tests: the second describes some investigations of special two–dimensional boundaries and indicates, some fairly general theoretical conclusions which can be drawn: in the third I shall try to review the possible reasons why, in practice, the best found conclusions are not always borne out.

A considerable amount of attention has been devoted to the study of the position of the centre of shear for straight beams but there appears to have been Uttle or no consideration of this position for beams of which the centre line is either shghtly or heavily curved. The reasons are probably twofold. Firstly, no demand for such an investigation has been made by designers, and secondly, and probably not least, any very general inquiry into this problem will lead to complex expressions which would not only be difhult and tedious to evaluate but will also be of uncertain accuracy.

This problem of taking beam curvature into account starts from the point where it normally ends for most mechanical engineers, who usually require to estimate the bending stresses in members of large curvature such as hooks and chain-links. Now the classical theory of elasticity applied to these latter members has yet to be seen to conform with experimental results to such a high degree as makes the added labour attending such methods worth while.

The problem of the buckling under bi-axial compression is considered for flat rectangular isotropic plates with simple edge conditions and no lateral restraint. The buckling stress coefficient is plotted against the side ratio for various conditions of edge restraint and a known compression or tension on the sides of the plate. It is found that there is some reduction in the buckling stress if the sides of the plate are subjected to compressive stresses and, conversely, there is an increase for tensile stresses. Furthermore, for plates with large side ratios, there is a rapid decrease of the spanwise buckling stress as the chordwise compressive stress approaches its appropriate Euler instability value.

For horizontal, or nearly horizontal, steady straight flight there exist some charts from which can be determined the speed at any altitude, the inclination of the flight path, and other quantities. This can be readily done, for instance, by a graphical comparison of the forces of drag and thrust, or by a corresponding representation of the powers, required and available. These charts, however, always involve the polar of the particular aeroplane, and are thus valid only for that aeroplane; they must therefore be replotted for each example. To avoid this and obtain a generally valid chart, the effect of the aerodynamic characteristics of the particular aeroplane must be eliminated; it is thus possible to prepare a chart even for the steady circling flight at any inclination of the flight path.

The three equations for the description of the flight path may be derived on the simple assumption that the thrust acts almost along the direction of flight, i.e., that the angles of attack and of sideslip are negligibly small.