compressed into the form of a long thin package, of the length of the straightened bows and of the thickness of the folded envelope. Such a vessel, when full and tight, would exhibit an appearance precisely similar in kind to that of a stretched umbrella. Of course, the same mode of construction is equally applicable to vessels for associate rowing, or for mechanical propulsion. Indeed, I would recommend that the first experimental vessels, which it would be desirable to build as cheaply and as small as possible, should be built in this manner. By this method sufficient stiffness will be given to vessels of no great length, and, as will appear hereafter, some of our other requisites will be equally satisfied by the same arrangement. Some additional appliances will be necessary to enable the vessel constructed on the umbrella-bow plan, to support a burden suspended to it. But this will be more appropriately considered in another place.

The way is now prepared for the consideration of our next condition.

CHAPTER VI.

CONDITION 4.—THE GAS-VESSEL MUST BE FIRM, NOT YIELDING TO THE PRESSURE OF THE AIR.

THE tediousness of the last chapter will be somewhat compensated by the brevity of the present one, of which, indeed, the chief work has been done in the pages just concluded.

In the case of the envelope stretched upon a skeleton framing, nothing more need be said; the head of the vessel is, of course, as soon as the bows are bent, as firm as the material will admit of being made, and ready to face without yielding any resistance which the material is strong enough to withstand.

If the gas-vessel is made of solid materials, such as metal plate, gutta-percha, or papier-maché, of course the same property of firmness is equally secured for it, without any further arrangement. For the head of the vessel will, whatever be its form, be of the nature of an arch or vault, calculated to meet any external pressure in the line of the axis of the vessel, with the most effectual resistance.

Should the bows ('bous,' I mean, not 'bws') of the vessels built on either of these fashions require any further strengthening, it

may be given to them thus. CAD (fig. 37) being the curved outline of the prow, a the extreme end of the long axis, let A B be a sprit fixed at A, and let stays B C, B E, B F, BD, radiate from its end B, to points symmetrically arranged on a circle girdling

the head of the vessel about the part where it requires support. To these points the stays must be firmly attached. They must be furnished with one of the hundred possible contrivances for tightening them, each and all, at will. If they were attached at B to a piece of tube as are the struts of an umbrella frame, the tube being moveable on the rod at pleasure, and admitting of being fixed in its place by a screw, the tension of the whole might be regulated at pleasure to a nicety. In the case of the flexible bows with the envelope stretched on them, the tube E F (fig. 36), being sufficiently lengthened, would answer the purpose of the rod A B (fig. 37), and the stays, being each attached to one of the bows, would aid the bow-string in keeping the figure in shape. Further, it is probable that if the cords BC, BE, &c., were covered with a cone of canvas resting on them, the resistance of the air to the vessel (unless the curvature of the prow be very sharp, in which case the beak-stays would not be likely to be required at all) would be diminished. The form indeed thus presented-a cone capping a spheroidal figure-is, under certain proportions, Newton's theoretical solid of least resistance.1 But this point would be determined by the experiments on the resistance of the air, which would be made according to the suggestions given in a former chapter.

In the remaining case, which will probably be the commonest, of gas-vessels in which the envelope is provided with an outer skeleton shell, constructed according to the method described in the last chapter, the means of furnishing it with the requisite outward firmness are simple and obvious. All that is necessary is to cover the framework with a continuous web stretched over it, and strong enough to resist the pressure which it will meet with from the air. The network of bamboo will thus be converted into an outer envelope, as hard and unyielding as can be desired. The opposing pressure of the air acts, of course, directly only on the front part of the gas-vessel; it will, therefore, be only at the fore-quarters and bows that the outer covering will be necessary as a shield. But it will be requisite to invest the whole of the skeleton-shell with a hide or enclosing

1 See 'Principia,' Lib. II. Prop. iv. Scholium.

4

eases less and less rapidly, as the actual size and

So that the larger the vessel, the less really, ss, is its monstrosity. Besides, reader, there is etween the earth and the moon, for all the airr launch in this world, however huge their icularly anxious, too, to show ground for ing-air craft may be made of reasonably small did not think I could first accomplish this, I tured to sketch a plan for the building of or ulterior use.

is also necessary. Indeed, I regard this as one of the most important advantages to be derived from the mode of building air-craft which I propose. The aerial architect and navigator will have, as one of their chief duties, to provide for the careful nursing of the delicate envelope. So long as it is proposed to leave this life-maintaining membrane open to all the destructive influences which are continually threatening its safety when it is free in the air, it is useless to bestow upon its first construction all the care which may so easily be given to it, which is so requisite in any serious attempt to use hydrogen in aerial navigation, and which may all be wasted by the rough handling of a single gale. But when it may be guarded on every side from the inclement touches of wind and weather, as well as from the severer violence of accidental shocks, it becomes imperatively requisite to provide it with the most effectual protection. The material, therefore, of the outer covering must not only be strong enough to resist, or at least to break, the force of all ordinary blows to which the vessel may be liable, but to shelter its contents completely from atmospheric influences. It should, therefore, be waterproof. It should likewise transmit changes of temperature at its surface as slowly as possible to the gas within. It should be made of material that conducts and radiates heat as slowly as possible. This checking of the liability of the gas to sudden contraction and expansion by cold and heat from without, will be a very important function of the outer shell.

Again, I have already alluded to the serious inconvenience which the deposition of large quantities of rain, dew, or snow, upon the surface of the gas-vessel may entail upon the aerial voyager, by loading his vessel, perhaps suddenly, with a great additional weight. Almost the whole of this precipitation of moisture, in the case of our air-craft, will fall upon the outer shell of the gas-vessel, chiefly upon its upper part. The common balloons suffer greatly in this respect by reason of the cords of the net, which not only soak up an immense quantity of water, but, by the meshes resting on the upper part of the envelope, form a vast number of little shallow pools in which the water

1 See p. 79, above.