In the first place, the airship, flying-machine, aerostat, or whatever we please to call it, must not merely move, but also lift itself. Neither a ship nor a locomotive is called upon to do this. Its ability to lift itself must depend upon either the employment of large balloons or upon sheer power. In the first case the balloon will, by reason of its size, be unmanageable in a high wind; in the second case, a breakdown in the machinery would probably prove fatal. Even supposing that our aerostat can lift itself successfully, we encounter the difficulties connected with steering in a medium traversed by ever-shifting currents of air, which demands of the helmsman a caution and capacity seldom required on land or water. Add to these the difficulties of leaving the ground and alighting safely upon it; and, what is more serious than all, the fact that though success can be attained only by experiment, experiment is in this case extremely expensive and risky, any failure often resulting in total ruin of the machine, and sometimes in loss of life. The list of those who have perished in the search for the power of flight is a very long one. Yet in spite of these obstacles determined attempts have been and are being made to conquer the air. Men in a position to judge are confident that the day of conquest is not very far distant, and that the next generation may be as familiar with aerostats as we with motor-cars. Speculation as to the future is, however, here less profitable than a consideration of what has been already done in the direction of collecting forces for the final victory. To begin at the beginning, we see that experimenters must be divided into two great classes: those who pin their faith to airships lighter than air, _e.g._ Santos Dumont, Zeppelin, Roze; and those who have small respect for balloons, and see the ideal air-craft in a _machine_ lifted entirely by means of power and surfaces pressing the air after the manner of a kite. Sir Hiram Maxim and Professor S. P. Langley, Mr. Lawrence Hargrave, and Mr. Sydney Hollands are eminent members of the latter cult. As soon as we get on the topic of steerable balloons the name of Mr. Santos Dumont looms large. But before dealing with his exploits we may notice the airship of Count Zeppelin, an ingenious and costly structure that was tested over Lake Constance in 1900. The balloon was built in a large wooden shed, 450 by 78 by 66 feet, that floated on the lake on ninety pontoons. The shed alone cost over £10,000. The balloon itself was nearly 400 feet long, with a cylindrical diameter of 39 feet, except at its ends, which were conical, to offer as little resistance as possible to the air. Externally it afforded the appearance of a single-compartment bag, but in reality it was divided into seventeen parts, each gas-tight, so that an accident to one part of the fabric should not imperil the whole. A framework of aluminium rods and rings gave the bag a partial rigidity.