Chanute’s kindness and Katharine’s persistence put Wilbur and Orville back on track. To prepare his speech, Wilbur had to carefully think through the results of his and Orville’s 1900 and 1901 experiments. Because their machines did not produce the expected lift, the brothers had begun to suspect that the data produced by Lilienthal -- which every other scientist in the field took for granted -- were incorrect. The more they discussed it, the more convinced they became. Wilbur decided to stick his neck out and say so in his speech.

To his relief, the speech -- which Wilbur simply called "Some Aeronautical Experiments" -- was well received and a copy was published in the prestigious Journal of the Western Society of Engineers. But now that the brothers had declared the late, great Lilienthal wrong, what data should be used to replace his tables of lift and drag? The brothers decided they would have to produce their own tables. While at Kitty Hawk, George Spratt had suggested an apparatus that balanced lift against drag to measure both these quantities more accurately, and now they set about designing it. First they built a balance from a bicycle wheel mounted so it revolved horizontally. To its rim, they attached a flat plate and a curved plate 90 degrees apart. In a steady wind, the curved plate –- a miniature wing - could be set at an angle to the air stream so that the lift generated by air rushing over its surface would be balanced again the drag on the flat plate set perpendicular to the wind. If the lift and the drag were equal, the wheel would not revolve. Wilbur and Orville set the curved plate at the angle predicted by the Lilienthal tables, mounted the wheel to one of their bicycles, and pedaled like mad. As the air rushed over the plates, the wheel turned. Something was indeed amiss with either the data or the method that Lilienthal used to predict lift and drag.

Although ingenious, the bicycle apparatus was too crude (and too exhausting) to make accurate measurements. So the brothers built a wind tunnel from an old grinder and scraps of wood, and made delicate balances from hacksaw blades and bits of wire. They tested a variety of wing shapes in the tunnel and by mid-December 1901, they had made a surprising discovery. Lilienthal’s tables were, by and large, correct! But another number in the equation he used to calculate lift, the coefficient of air pressure, was way off the mark. This number had been developed in the mid-eighteenth century by an English engineer, John Smeaton, to design more efficient windmills. It had been in general use for nearly a century and a half, long enough to be accepted as gospel.

Wilbur and Orville also found that the shape of Lilienthal’s wings –- the camber - was inefficient. The curves of the German’s wing were sections of a circle. The highest point was in the center of the chord, midway between the leading and trailing edges. Wilbur and Orville found that a wing produced much more lift if its curve followed the shape of a parabola, with the high point about one quarter of the chord behind the leading edge.