There can be no question that manoeuvrability is one of the principal requirements for a combat aircraft, and achieving the necessary amount of usable lift is never an easy design task. The lift generated by the wings ceases to be usable when it is accompanied by unacceptable levels of buffet or degraded handling or loss of control. Practically all combat aircraft have to penetrate some way into this region where things begin to go wrong, and the problem we are concerned with in the Royal Aircraft Establishment is to see how certain we can be, before the first flight of a new design, that the so-called buffet boundary will allow the specified manoeuvrability to be achieved. One line of attack is to take an existing aircraft, record and analyse its behaviour in these limiting conditions, and see whether design and test methods are good enough to predict what we find in flight.

The possibilities for technical progress in every aspect of aviation during the first half of this century have indeed appeared as a limitless horizon. Increasingly since that period, as we have moved outward toward the technical horizons, the course has become rougher, and we find ourselves in the foothills with a high and formidable range of mountains looming into view before us.

These figurative mountains are representative of the increasing interface deterrent which we are encountering at the edge of the aviation systems’ advancing fronts. These deterrents include socio-environmental and political impacts, economic factors, and the need, in many cases, for redirection and/or equivalent progress in non-aviation technical systems which interface with aviation development. Specific areas are increasing costs for R & D, increased fuel costs, environmental effects of new aircraft and air transport ground systems, changing requirements and demands of the public, increasing liability exposure, increasing regulatory constraints and conflicts, etc.

As the eighth decade of heavier-than-air flight comes to a close, it has become increasingly evident that the nature of progress in civil aviation is undergoing a significant change. The transition from an explosive growth to what has been referred to as maturity has been both swift and unexpected, and coincidentally enhanced by factors external to aeronautics. While the anticipation of timing of such a change usually calls for a predictive ability which borders on clairvoyance, the inevitabilty of occurrence of transition in the development of any technology should be clearly recognised. It should be thus neither surprising that the forecasts of the 1960s have badly missed the mark, nor that the period of transition has set in.

As civil aviation moves into a new phase of development, it is of interest to evaluate the status it has attained globally as a mature industry and service, and to consider the factors which could significantly influence its future. We shall preface the global assessment with a brief review of the major causes of spectacular growth of aviation since 1945—an exercise necessary to appreciate the basis of maturity and the constraints now evident.

Reference 1 describes a numerical procedure for calculating the roll-up of the wake behind a wing of finite span, when the wake vorticity is assumed to occur within a layer of finite thickness with thin elliptic cross-section normal to the stream. The vorticity is assumed to be a function of spanwise position only (corresponding to a linear transverse velocity profile when the wake is thin), the process is treated as a two-dimensional unsteady one, as in the work of Westwater, Moore and others, and viscous and turbulence effects are ignored.

In Ref. 1 the technique was applied to the classical elliptic loading, to calculate the very early stages of the roll-up. In the present paper these calculations are extended to larger values of the dimensionless time t* and the results are seen to tally fairly well with those of Moore, which were obtained by using a point-vortex model. Similar calculations have also been carried out for a particular non-elliptic loading which was dealt with by Clements and Maull by a two-dimensional, point-vortex method. They gave this loading as an example of a distribution which produces two roll-up cores on each half span, with an induced drag only 10% greater than that of the elliptic loading having the same Ch. The results of the new calculations are presented here, and are seen to be similar to those of Ref. 4 in a general way, both qualitatively and quantitatively.