Approximate solutions for the critical load of an axially loaded strut, and for the deflection produced in an eccentrically loaded strut, are obtained by the Energy Method. For most practical purposes these solutions compare favourably with solutions based on successive approximation, since for the same accuracy the time required for a solution by successive approximation is much greater.

Investigations have shown that with the addition of Kirsch's supplementary stresses to the elastic expansion and rivet side-thrust, there is a liability to dangerous stress concentration at the edge of the rivet hole. The influence of these edge stresses on the adjacent hole(s) and the plate margin, also requires investigation.

However, since in spite of this disadvantage, riveting is in practice found to be superior to welding, it is evident that the edge stresses must be effectively neutralised by combined elastic and permanent deformation of the material around the hole. At the same time, the unavoidable weakening of plate and hole edges by hair cracks, etc., makes it advisable to pay special attention to the following points :—

1. 1. The initial stresses, in the plates to be joined, should be kept as low as possible.

2. 2. Rivet holes should be accurate and cleanly finished.

3. 3. Rivets should not be too closely spaced.

4. 4. The plate margin should be wide enough to avoid risk of breakdown through stress concentration due to hair-cracks or small notches.

5. 5. The edges of the plates and rivet holes should be reamed to remove hair cracks and local crumbling.

In common with other machines, each new type aircraft must be put through comprehensive and well-seasoned tests. Unlike other machines, however, the aircraft exists to fly and by so doing severs its contact with the earth. Even this would not present a peculiar test difference if the atmosphere were rigidly connected to the earth and had a constant pressure and constant temperature at all heights, for then the engine power would not change with height or place and the speed over the ground would equal the speed through the air. Likewise the rate of climb, the manœuvrability and the behaviour of the controls would be independent of height. In such an atmosphere, tests would be required at one height only and measurements such as speed, rate of climb, take-off and landing could all be obtained accurately and directly from ground apparatus.

Although this paper is primarily concerned with the structure of current German military aircraft, no apology is needed in considering for a few moments some of the outstanding developments that took place 20 or 30 years ago.

In the early days of any branch of engineering, there are always a few individuals in each country who stand head and shoulders above their fellow workers in the same field, and aviation is no exception. But, whereas in this country quite a number of people spring to mind as having made exceptional contributions to aeronautics, in Germany one name stands out pre-eminent— that of Junkers. Hugo Junkers was born in 1859 and soon showed promise as a scientist and engineer. His technical training was varied, and at the age of 36 he founded at Dessau the engineering firm which bears his name.

Our engineering habits form slowly, and once formed are slow to change. This is perhaps less true in aeronautics than in other branches of engineering. Nevertheless some aeronautical customs are deeply rooted, so much so that it is easy to confuse the superstructure of habit with the basic foundations. In this article I have reviewed some of the customary methods of specifying strength conditions and have tried to disentangle the underlying principles. Current changes in these methods emphasise the need of a proper appreciation of the principles, and a study of these principles suggests even more radical changes of methods in future.