Would the vibration of the explosive engine shake me back and forth, strain at the ropes until it had unequalised their tension, and then break them one by one? Would it jar the interior air balloon's pump and derange the big balloon's valves? Would it continually jerk and pull at the silk hems and the thin rods which were to hold my basket to the balloon? Free from the steadying influence of the solid ground, would the jumping motor jar itself until it broke? And, breaking, might it not explode?

All this and more had been predicted by the professional aeronauts, and I had as yet no proof outside of reasoning that they might not be right on this or that topic.

I started the motor. I felt no particular vibration, and I was certainly not being shaken. I increased the speed, and felt less vibration! There could be no doubt about it—there was less vibration in this light-weight tricycle hanging in the air than I had regularly felt while travelling on the ground. It was my first triumph in the air!

I will say frankly that as I rose in the air on my first trip I had no fear of fire. What I feared was that the balloon might burst by reason of its interior pressure. I still fear it.

Before going up I had minutely tried the valves. I still try them minutely before each of my trips. The danger, of course, was that the valves might not work adequately, in which case the expanding of the gas as the balloon rose would cause the dreaded explosion. Here is the great difference between spherical and dirigible balloons. The spherical balloon is always open. When it is taut with gas it is shaped like an apple; when it has lost part of its gas it takes the shape of a pear; but in each case there is a great hole in the bottom of the spherical balloon where the stem of the apple or the pear would be, and it is through this hole that the gas has opportunity to ease itself in the constant alternations of condensation and dilatation. Having such a free vent, the spherical balloon runs no risk of bursting in the air; but the price paid for this immunity is great loss of gas and, consequently, a fatal shortening of the spherical balloon's stay in the air. Some day a spherical balloonist will close up that hole; indeed, they already talk of doing it.

I was obliged to do it in my air-ship balloon, whose cylindrical form must be preserved at all cost. For me there must be no transformations as from apple to pear. Interior pressure only could guarantee me this. The valves to which I refer have since my first experiments been of all kinds—some very ingeniously interacting, others of extreme simplicity. But their object in each case has always been the same: to hold the gas tight in the balloon up to a certain pressure and then let only enough out to relieve dangerous interior pressure. It is easy to realise, therefore, that should these valves refuse to act adequately the danger of bursting would be there.

This possible danger I acknowledged to myself, but it had nothing to do with fire from the explosive motor. Yet during all my preparations, and up to the moment of calling: "Let go all!" the professional aeronauts, completely overlooking this weak point of the air-ship, continued to warn me against fire, of which I had no fear at all!

"Do we dare strike matches in the basket of a spherical balloon?" they asked.

"Do we even permit ourselves the solace of a cigarette on trips that last for many hours?"

To me the cases did not seem the same. In the first place, why should one not light a match in the basket of a spherical balloon? If it be only because the mind vaguely connects the ideas of gas and flame the danger remains as ideal. If it be because of a real possibility of igniting gas that has escaped from the free hole in the stem of the spherical balloon it would not apply to me. My balloon, hermetically closed, except when excessive pressure should let either air or a very little gas escape through one of the automatic valves, might for a moment leave a little trail of gas behind it as it moved on horizontally or diagonally, but there would be none in front where the motor was. (See Fig. 4.)