Page:My Airships.djvu/198

 60 cubic metres (2118 cubic feet). The ventilator that was to feed it formed practically a part of the motor itself. Revolving continually while the motor worked, it would serve air continually to the compensating balloon whether or not the latter would be able to hold it. What air it could not hold would escape through a comparatively weak valve ("Air Valve," Fig. 10) communicating with the outer atmosphere through the bottom of the air balloon, which was also the bottom of the great outer balloon. To relieve the great balloon of its dilated hydrogen when necessary I supplied it with two of the best valves I could make ("Gas Valves," Fig. 10). These also communicated with the outer atmosphere. Imagine, now, that after a certain condensation of my hydrogen the interior compensating balloon should have filled up in part with air from the ventilator and so maintained the form of the great balloon rigid. Shortly after, by a change of temperature or altitude, the hydrogen would begin to dilate again. Something would have to give way, or the balloon would burst in a "cold explosion." What ought to give way first? Evidently the weaker air valve ("Air Valve," Fig. 10). Letting out part or all of the air in the