But setting aside for the time the work done by Santos-Dumont with machines heavier than air, let us consider his triumphs with balloons at the opening of his career. He had come to France about forty years after Henry Giffard had demonstrated the practicability of navigating a balloon 144 feet long and 34 feet in diameter with a three-horse-power steam-engine. But no material success attended this demonstration, important as it was, and the inventor turned his attention to captive balloons, operating one at the Paris Exposition of 1878 that took up forty passengers at a time. There followed Captain Renard to whose achievement we have already referred. He had laid down as the fundamentals of a dirigible balloon these specifications:
A cigar, or fishlike shape.
An internal sack or ballonet into which air might be pumped to replace any lost gas, and maintain the shape of the balloon.
A keel, or other longitudinal brace, to maintain the longitudinal stability of the balloon and from which the car containing the motor might be hung.
A propeller driven by a motor, the size and power of both to be as great as permitted by the lifting power of the balloon.
A rudder capable of controlling the course of the ship.
Santos-Dumont adopted all of these specifications, but added to them certain improvements which gave his airships--he built five of them before taking his first prize--notable superiority over that of Renard. To begin with he had the inestimable advantage of having the gasoline motor. He further lightened his craft by having the envelope made of Japanese silk, in flat defiance of all the builders of balloons who assured him that the substance was too light and its use would be suicidal. "All right," said the innovator to his favourite constructor, who refused to build him a balloon of that material, "I'll build it myself." In the face of this threat the builder capitulated. The balloon was built, and the silk proved to be the best fabric available at that time for the purpose. A keel made of strips of pine banded together with aluminum wire formed the backbone of the Santos-Dumont craft, and from it depended the car about one quarter of the length of the balloon and hung squarely amidships. The idea of this keel occurred to the inventor while pleasuring at Nice. Later it saved his life.
One novel and exceedingly simple device bore witness to the ingenuity of the inventor. He had noticed in his days of free ballooning that to rise the aeronaut had to throw out sand-ballast; to descend he had to open the valves and let out gas. As his supply of both gas and sand was limited it was clear that the time of his flight was necessarily curtailed every time he ascended or descended. Santos-Dumont thought to husband his supplies of lifting force and of ballast, and make the motor raise and lower the ship. It was obvious that the craft would go whichever way the bow might be pointed, whether up or down. But how to shift the bow? The solution seems so simple that one wonders it ever perplexed aviators. From the peak of the bow and stern of his craft Santos-Dumont hung long ropes caught in the centre by lighter ropes by which they could be dragged into the car. In the car was carried a heavy bag of sand, which so long as it was there held the ship in a horizontal plane. Was it needful to depress the bow? Then the bow rope was hauled in, the bag attached, and swung out to a position where it would pull the forward tip of the delicately adjusted gas bag toward the earth. If only a gentle inclination was desired the bag was not allowed to hang directly under the bow, but was held at a point somewhere between the car and the bow so that the pull would be diagonal and the great cylinder would be diverted but little from the horizontal. If it were desired to ascend, a like manipulation of the ballast on the stern rope would depress the stern and point the bow upwards. For slight changes in direction it was not necessary even to attach the sand bag. Merely drawing the rope into the car and thus changing the line of its "pull" was sufficient.
