Accordingly I immediately set to work on a "No. 3," with a shorter and very much thicker balloon, 20 metres (66 feet) long and 7·50 metres (25 feet) at its greatest diameter (Fig. 6). Its much greater gas capacity—500 cubic metres (17,650 cubic feet)—would give it, with hydrogen, three times the lifting power of my first, and twice that of my second air-ship. This permitted me to use common illuminating gas, whose lifting power is about half that of hydrogen. The hydrogen plant of the Jardin d'Acclimatation had always served me badly. With illuminating gas I should be free to start from the establishment of my balloon constructor or elsewhere as I desired.
Fig. 6
It will be seen that I was getting far away from the cylindrical shapes of my first two balloons. In the future I told myself that I would at least avoid doubling up. The rounder form of this balloon also made it possible to dispense with the interior air balloon and its feeding air pump that had twice refused to work adequately at the critical moment. Should this shorter and thicker balloon need aid to keep its form rigid I relied on the stiffening effect of a 10-metre (33-foot) bamboo pole (Fig. 6) fixed lengthwise to the suspension cords above my head and directly beneath the balloon.
While not yet a true keel, this pole keel supported basket and guide rope and brought my shifting weights into much more effectual play.
On November 13th, 1899, I started in the "Santos-Dumont No. 3," from the establishment of Vaugirard, on the most successful flight that I had yet made.
ACCIDENT TO "No. 2," MAY 11, 1899
(THIRD PHASE)
From Vaugirard I went directly to the Champ de Mars, which I had chosen for its clear, open space. There I was able to practise aerial navigation to my heart's content—circling, driving ahead in straight courses, forcing the air-ship diagonally onward and upward, and shooting diagonally downward, by propeller force, and thus acquiring mastery of my shifting weights. These, because of the greater distance they were now set apart at the extremities of the pole keel ([Fig. 6]), worked with an effectiveness that astonished even myself. This proved my greatest triumph, for it was already clear to me that the central truth of dirigible ballooning must be ever: "To descend without sacrificing gas and to mount without sacrificing ballast."
During these first evolutions over the Champ de Mars I had no particular thought of the Eiffel Tower. At most it seemed a monument worth going round, and so I circled round it at a prudent distance again and again. Then—still without any dream of what the future had in store for me—I made a straight course for the Parc des Princes, over almost the exact line that, two years later, was to mark the Deutsch prize route.