[Illustration: FIRST WRIGHT GLIDER.] [Illustration: SECOND WRIGHT GLIDER.] The 1902 machine, shown in Plate XIX, had two main surfaces, measuring each 32 by 15 feet, and a front rudder measuring 15 square feet. The whole weight was 116 pounds. It will be noted that a vertical rudder was now employed. This was a reversion to the design of Chanute and Herring, but after some experience, the rudder was made adjustable, as in Henson’s aëroplane of 1842. Its surface was 12 square feet, but later reduced to six. With this machine they obtained between 700 and 1,000 glides during the season. It showed greater efficiency than its predecessors, its normal angle of descent being estimated at seven degrees or less. This was some improvement over the efficiency of the Chanute-Herring glider, partly due, of course, to placing the rider flat, instead of allowing him the more comfortable erect posture adopted later. Whatever improvements of efficiency and strength had been made, these were of secondary importance compared with the provisions for projectile stability and manual control. Here at last, after ten years’ groping, was an actual glider with sufficiently high centroid to minimize the pendulum effect, and with three rudders to give impactual torque about the three axes. These simple provisions had been previously pointed out in aëronautic writing, and, in the latter nineties, had been embodied in Mattullath’s aëroplane, but not tested in the large machine, owing to his death. The wonder is that, of all the practical inventors of aëroplanes, Mr. Mattullath was the only one of that period fully to grasp and adopt these main ideas before starting to build a man-carrying machine. However, it must be added that he had previously made small flying models, which may have suggested the advantage of kinetic stability and the three-torque system of control. If Lilienthal and his disciples, who laid so much stress on gliding experience, had started like Mattullath with three torque-surfaces, they would have missed indeed those acrobatic and picturesque kickings at the sky, but they would have reached the desired goal with less danger, time and expense. They displayed more skill in riding a fractious glider than in designing a tractable one, by providing for impactual torque about each of three axes. Had they started with a good theory of dynamic control, they could have dispensed with coasting entirely, and commenced aviating with short runs over a smooth course followed by cautious leaps in the air, after the style of certain ingenious French aviators. However, the knack of balancing was finally acquired, and thus the glider was ready to receive the propelling mechanism.