At the same time, however, though it was at the time difficult to pick out, there was very real progress being made, and, though a number of ‘freak’ machines fell out by the wayside, the pioneer designers of those days learnt by a process of trial and error the right principles to follow and gradually succeeded in getting their ideas crystallised. In connection with stability mention must be made of a machine which was evolved in the utmost secrecy by Mr J. W. Dunne in a remote part of Scotland under subsidy from the War Office. This type, which was constructed in both monoplane and biplane form, showed that it was in fact possible in 1910 and 1911 to design an aeroplane which could definitely be left to fly itself in the air. One of the Dunne machines was, for example, flown from Farnborough to Salisbury Plain without any control other than the rudder being touched; and on another occasion it flew a complete circle with all controls locked, automatically assuming the correct bank for the radius of turn. The peculiar form of wing used, the camber of which varied from the root to the tip, gave rise, however, to a certain loss in efficiency, and there was also a difficulty in the pilot assuming adequate control when desired. Other machines designed to be stable--such as the German Etrich and the British Weiss gliders and Handley-Page monoplanes--were based on the analogy of a wing attached to a certain seed found in Nature (the ‘Zanonia’ leaf), on the righting effect of back-sloped wings combined with upturned (or ‘negative’) tips. Generally speaking, however, the machines of the 1909–1912 period relied for what automatic stability they had on the principle of the dihedral angle, or flat V, both longitudinally and laterally. Longitudinally this was obtained by setting the tail at a slightly smaller angle than the main planes.