Both within the US, and on the North Atlantic, there has been a dramatic reduction in governmental control over airline pricing routeing and service policies. In the short run, the most visible and important development has been the substantial decrease in average fares, spearheaded by greatly reduced fares available in certain restricted circumstances.

Because decisions about price, routeing and service interact, I find it most useful to talk about the loosening of governmental strictures on them, collectively, as deregulation, and I want to concentrate my remarks on the impact of deregulation from my perspective as an observer of the industry, and as an airport operator. I hope also to develop the theme that loosening one of these strictures without attention to the others can adversely affect air service.

Studies made of the flow about the wing-tip of a wind tunnel model showed that the vortex-type flow generally associated with the rolled-up vortex sheet well behind the tip existed just behind the trailing edge and seemed to be forming over the top of the wing-tip itself. As a result small cambered aerofoils were attached to the wing-tip to redirect the flow locally in such a way that each experienced a force with a significant thrust component. The effect is similar to that generated by the sail of a close-hauled yacht; consequently these small tip extensions were called sails. Reference 1 describes the original wind tunnel and the preliminary flight tests of sails mounted from the tip tanks of the Paris aircraft of the Cranfield Institute of Technology.

The local buckling behaviour of structural sections composed of isotropic flat plate elements is well documented. The results are most conveniently presented in the form of curves for the minimum buckling coefficient, Kmin, against the flange/web ratio λ2. A series of such curves for a range of structural sections composed of orthotropic plates subjected to uniform end loads and simply supported end constraints has been presented by the author. For isotropic sections of practical proportions, it is generally assumed that the buckling coefficients for columns with simply supported and built-in end conditions approach the common value Kmin, and that the half wavelength of the buckle mode in the axial direction is of the same order as the cross-sectional dimensions of the column. For structures composed of highly orthotropic materials such as carbon fibre reinforced plastic (CFRP), the half wavelength is much greater than that of the isotropic counterpart. In consequence, it is only for much longer sections that the common value of Kmin, can be assumed for the buckling coefficient of built-in orthotropic sections without excessive error.