CHAPTER IV.

GREEK LAMPS, TOYS, ETC. PERPETUAL LAMPS. The ancients utilized, in their prestiges, combustible gases, which, in many places, were disengaged naturally from the earth. The Arab Schiangia, in a passage quoted by Father Kircher, expresses himself in this wise: “In Egypt there was a field whose ditches were full of pitch and liquid bitumen. Philosophers, who understood the forces of nature, constructed canals which connected places like these with lamps hidden at the bottom of subterranean crypts. These lamps had wicks made of threads that could not burn. By this means the lamp, once lighted, burned eternally, because of the continuous influx of bitumen and the incombustibility of the wick.” It is possible that it was to an artifice of this same nature that were due some of the numerous perpetual lamps that history has preserved a reminiscence of, such as that which Plutarch saw in the temple of Jupiter Ammon, in Egypt, and that in the temple of Venus, which Saint Augustine could only explain as due to the intervention of demons. But the majority of them owed their peculiarity only to the precautions taken by the priests to feed them without being seen. It was only necessary, in fact, that the wick, which was made of asbestos threads or gold wire, should be kept intact, and that the body of the lamp should communicate with a reservoir placed in a neighboring apartment in such a way that the level of the oil should remain constant. Heron and Philo have left us descriptions of a certain number of arrangements that permitted of accomplishing such an object. The same authors likewise point out different processes for manufacturing portable lamps in which the oil rises automatically. The most ingenious one is that which is at the present day known under the name of “Heron’s Fountain.”[12] [12] In 1801, Carcel and Carreau applied Heron’s system to lamps without, perhaps, knowing that they were thus returning to the primitive apparatus. The following is the Alexandrine engineer’s text: “Construction of a candelabrum such that upon placing a lamp thereon, there comes up through the handle, when the oil is consumed, any quantity that may be wished, and that, too, without there being any need of placing above it any vessel serving as a reservoir for the oil. “A hollow candelabra must be made, with a base in the shape of a pyramid. Let Α Β Γ Δ be such pyramidal base, and in this let there be a partition, Ε Ζ. Again, let Η Θ be the stem of the candelabrum, which should also be hollow. Above, let there be placed a vessel, Κ Λ, capable of containing a large quantity of oil. From the partition, Ε Ζ, there starts a tube, Μ Ν, which traverses it and reaches almost to the cover of the vessel, Κ Λ, upon which latter is placed the lamp in such a way as to allow only a passage for the air. Another tube, Ξ Ο, passes through the cover and runs down, on the one hand, to the bottom of the vessel, Κ Λ, in such a way that the liquid may be capable of flowing, and on the other, forms a slight projection on the cover. To this projection there is carefully adjusted another tube, Π, which is provided with a stopper at its upper part, and, traversing the bottom of the lamp and united with it, is wholly inclosed within the interior of the lamp. To the tube, Π, there is soldered another and very fine one which communicates with it and reaches the extremity of the lamp handle. This tube debouches in the latter in such a way that its contents may empty into the lamp, the orifice of which is of the usual size. Under the partition, Ε Ζ, there is soldered a cock that enters the compartment, Γ Δ Ε Ζ, in such a way that when it is open the water from the chamber, Α Β Ε Ζ, may pass into the compartment, Γ Δ Ε Ζ. Through the upper plate, Α Β, there is pierced a small hole, through which the compartment, Α Β Ε Ζ, may be filled with water, the air within escaping through the same aperture. [Illustration: PLATO’S LAMP] “Let us now remove the lamp and fill the vessel with oil by the aid of the tube, Ξ Ο. The air will escape through the tube, Μ Ν, and afterward through a cock which is open near the bottom, Γ Δ, when the water has flowed out from the compartment, Γ Δ Ε Ζ. Let us place the lamp upon its base, connecting it at the same time with the tube, Π. When it becomes necessary to pour oil into it, we will open the cock near the partition, Ε Ζ. The water that is in the compartment, Γ Δ Ε Ζ, as well as the air therein, being forced through the tube, Μ Ν, into the vessel, will cause the oil to rise and pass into the lamp through the tube, Ξ Ο, and the one that forms a continuation of it. When it is desired to cause the oil to stop coming over, the cock is closed, when the flow will cease. This may be repeated as often as may be necessary.” Such was, perhaps, Plato’s lamp, of which Athenæus speaks in the “Banquet of the Sophists,” and by means of which the illustrious philosopher was enabled to have a light for himself during the longest nights in the year. AN ANCIENT AUTOMATON. In his “_Spiritalia_” (written about 150 B.C.) Heron describes several automata of which figures of birds form a part; but perhaps the most remarkable for its ingenious simplicity is No. 44, the illustration of which we reproduce. The description of this, as given by Heron, is somewhat meager and unsatisfactory, but the drawing is so very plain that, taken in connection with other mechanism in his work, operated in a similar way, it is easy to understand how the desired result was accomplished. An air-tight box of metal was provided, which was divided into four compartments, 1, 2, 3, 4, by horizontal diaphragm plates. On the top of this box was a basin, O, for receiving the water of a fountain. Around this basin were four birds, A, B, C, D, perched upon branches or shrubs, which apparently grew out of the top of the box. Each of these branches was hollow, and communicated with one of the compartments already named, by one of the pipes, 9, 10, 12, and 13, which passed but a very short distance through the tops of the several compartments. The bodies of the birds were also hollow, and were connected with the hollow branches by tubes in their legs. In the hollow body of each bird were two musical reeds or whistles of different note. One of these would sound when air was forced outward through the beak of the bird, and the other would only respond to air drawn inward. This alternate action of the air, and consequent variation of note, was produced by the peculiar way in which the water supplied by the fountain was made to pass through the several compartments. The water from the basin, O, entered compartment 1 near its bottom by the pipe 11, and as it rose in the compartment, it compressed the air above it, which escaped through the beak of the bird, A, and caused its first note to sound; but when the water reached the top of the bend of the siphon 5, it at once began to discharge by that siphon into compartment 2; but as the siphon 5 was so proportioned that it discharged the water much faster than it was supplied by pipe 11, the level of the water in compartment 1 gradually fell, and the air in passing into this compartment through the beak of the bird, A, caused its second note to sound. As the water rose in compartment 2, it compressed the air above it, which passed by the pipe 10, to the bird, B, which then sounded its first note, while the bird, A, was sounding its second, and this state of affairs continued until all of the water was discharged from the compartment 1, and compartment 2 was filled to the top of the bend of siphon 6, which then began to discharge into compartment 3; and as siphon 5 had ceased to operate, the water gradually fell in compartment 2, and the air entering by the beak of the bird, B, sounded its second note. While this was taking place, compartment 1 was again filling, and the first note of bird, A, sounding; and compartment 3 was also filling, and the air above the water therein was being forced by the pipe 12 into the bird, C, and causing its first note to sound. [Illustration: AN ANCIENT AUTOMATON.] By following out the operations described, and tracing the action of the flux and reflux of the water in the compartments 3 and 4, it will readily be seen that the bird, C, will sound its second note when the compartment 3 is being discharged by siphon 7 into compartment 4, and at the same time the bird, D, will sound its first note, and that eventually the water will escape from the automaton by the siphon 8, causing the second note of the bird, D, to be heard. It is evident that by simple and well-known means any or all of the bird notes can be made to trill, and that it is only necessary to properly proportion the discharging capacity of the siphons to insure the repetition and admixture of the notes in a bird-like manner; and it is further evident that the employment of the ideas involved is not of necessity confined to but four birds, as several birds, each having different notes, might be operated from the same compartment, and of course as many compartments as may be wished can be used. Furthermore, the wings of the birds could be made to move, and their beaks to open and shut, by the movement of the same air which acted upon the musical reeds or whistles. Each of the siphons in the automaton was intermittent in its action, ceasing to flow when its compartment was emptied, and beginning again spontaneously when the water reached the level of the top of its bend. The antiquity of intermittent siphons is of special interest from the fact of their comparatively recent application in sanitary plumbing. Chaucer was not much in error as regards his own time (1328-1400), and his words are only somewhat less true to-day: “For out of the old fieldes, as men saithe, Cometh al this new corne fro yere to yere; And out of old bookes, in good faithe, Cometh all this new science that men lere.” A GREEK TOY. Upon a pedestal there is fixed a small tree around which is coiled a dragon. A figure of Hercules stands near by, shooting with a bow, and there is an apple lying upon the pedestal. If this apple be lifted from the latter, Hercules will shoot his arrow at the dragon, and the latter will hiss. [Illustration: A GREEK TOY.] _Mechanism of the Toy._--Let Α Β be the water-tight pedestal under consideration, provided with a diaphragm, Γ Δ. To this latter there is fixed a small, hollow, truncated cone whose apex points toward the bottom of the vessel, and from which it is just sufficiently distant to permit the water to pass. To this cone there is adjusted with care another one, Θ, which is fixed to a chain that, passing through an aperture, connects it with the apple. Hercules holds a small horn bow, whose string is stretched and laced at a proper distance from the right hand. The left hand is provided with a detent. To the extremity of this latter there is fixed a small chain or a cord that traverses the top of the pedestal, passes over a pulley fixed to the diaphragm, and connects with the small chain that joins the cone with the apple. This cord passes through the hand and body into the interior of Hercules. A small tube, one of those used for whistling with, starts from the diaphragm, rises through the top of the pedestal, and passes into the interior of the tree or around it. Now, if the apple, Κ, be raised, the cone, Θ, will be raised at the same time, the cord, Χ Φ, will be tightened, the catch will be freed, and this will cause the arrow to shoot. The water in the compartment Α Γ, running into the compartment Β Γ, will drive out the air contained in the latter, through the tube, and produce a hissing. The apple being replaced, the cone, Θ, will adjust itself against the other, stop the flow, and thus cause the hissing to cease. The arrow and its accessories will then be adjusted anew. When the compartment Β Γ is full, it is emptied by means of a spout provided with a key, and Α Δ is again filled as we have indicated. THE DECAPITATED DRINKING HORSE. The optical delusion known as the talking decapitated person has already been described in Book I., Chapter I., of the present work. The ancients invented an analogous trick, but one that was founded upon a very ingenious mechanical combination. This is found described at the end of Heron’s “Pneumatics,” under the title, “To cut an animal in two and make him drink.” It is as follows: “Let us suppose a hollow pedestal, A B C D, divided in its center by a diaphragm, E F. Above the pedestal there is fixed a statuette representing a horse and traversed by a tube, M N, which terminates on the one hand in the horse’s mouth, and in the other in the upper part of the compartment, A B E F, after following one of the legs. It will be conceived, in the first place, that if the said compartment be filled with water through an aperture, T, which is afterwards stopped up, and that then a cock be opened, so as to form a communication between the upper compartment and the lower (which latter is itself provided with an open air-hole), the water will flow, and, in doing so, tend to cause a vacuum in the tube, M N, so that when a vessel of water is brought near the animal’s mouth the water will be sucked up. “If the cock be so arranged as to present its key upon the top of the pedestal, and if to the key there be adapted a statuette representing a man armed with a club, things may be so arranged that the animal shall drink when the man has his back turned, for example, and that he shall stop drinking when the man threatens him with the club. “The following is the way in which a knife may be passed through the animal’s neck without causing the head to fall or interrupting communication between the mouth and pedestal. The head and body form two distinct pieces, which are adjusted according to the plane, O P (Figs. 1, 2, and 3). The tube, M N, is interrupted to the right of this slit, and the two parts of it are connected by a smaller tube, α β, which enters by slight friction into the interior of each of them; and to this small tube, α β, there are fixed two racks, δ and ε. Above δ and under ε are placed two segments of toothed wheels, π and ρ, which are movable around axles fixed in the body of the animal. Over the whole there is a third wheel, which is likewise movable around an axle fixed in the animal’s body, and the thickness of which keeps increasing from the centre to the circumference. This wheel is cut out into three parts of circles, μ, ν, and ξ, which have for diameters three of the sides of the inscribed hexagon. It is inclosed in the neck in such a way that the circular cavity containing it embraces just four of the sides of the inscribed hexagon, the two other sides projecting outside of the plane, O P. In the piece that forms the head a circular cavity is formed capable of containing this projecting portion of the wheel, and a wedge-shaped profile is given it, so that when one tooth of the wheel, σ, is engaged therein by the edge, it can also only leave it by the edge. Let us now suppose the wheel, σ, free; let us engage one of its teeth in the cavity, χ ψ; let us cause the head and body to approach; let us fix the wheel, σ, in the body by means of the movable axle traversing it; and let us introduce a knife into the slit, O P, and see what will happen. [Illustration: HERON’S DECAPITATED DRINKING HORSE.] “The blade, on entering the space, ξ, will press against one of the teeth, and cause it to descend until it, as well as the knife, is disengaged. The tooth above the space, ξ, will then be disengaged in its turn and connect the head with the body again. The knife-blade, which is now under the wheel, σ, rests on the inclined plane that the figure shows in the segment, π, and, on pressing thereupon, causes the wheel to turn, and with it the rack, δ, and the tube, α β, which latter leaves the tube, M, and gives passage to the blade between it and the extremity, α. Then the blade comes in contact with the lower projection of the sector, ρ, which has been carried upward by the motion of the rack, ε, that is connected with the rack, δ. On pressing against such projection the blade causes the segment, ρ, to revolve in a contrary direction, brings ε toward the left, and causes the small tube, α β, to enter anew the tube, M. Communication between M and N is thus reëstablished.” M. de Rochas has never found elsewhere than in the “Pneumatics” a description of this system of toothed wheels, although he has read the majority of books treating of this class of ideas. The description given by Heron is itself so confused and so mutilated, and the figure that accompanies it is so incomplete, that in all the Latin editions it is suppressed as incomprehensible. [Illustration: THE DECAPITATED HORSE. DETAILS OF THE MECHANISM IN THE NECK.] ODOMETERS. In the inventory of the objects sold after the death of the Emperor Commodus, drawn up by Julius Capitolinus in the life of Pertinax, we find mentioned, among other valuable things, “vehicles that mark distances and hours.” Vitruvius (X, 14) describes the mechanism of these vehicles, but the figures that must have served to throw light upon the text have been lost, so that his description is somewhat obscure. Fortunately, as a sequel to a manuscript of the Dioptra of Heron, there have been found two Greek fragments upon this same subject, dating back probably to the Alexandrine epoch and accompanied with figures. The following is a translation, says M. de Rochas: TO MEASURE DISTANCES UPON THE SURFACE OF THE EARTH BY MEANS OF AN APPARATUS CALLED AN ODOMETER. Provided with this instrument, instead of being obliged to measure land slowly and laboriously with the chain or cord, it is possible in traveling in a vehicle to know the distances made, according to the number of revolutions of the wheels. Others, it is true, have, previous to us, made known certain methods of accomplishing the same object; but every one will be able to decide between the instrument described here by us and those of our predecessors. [Illustration: FIG. 1.--HERON’S ODOMETER FOR VEHICLES.] Let us imagine an apparatus in the form of a box (Fig. 1) in which is contained the entire machine that we are to describe. Upon the bottom of the box rests a copper face wheel, Α Β, having, say, eight teeth. In the bottom there is an opening in which a rod, fixed to the hub of one of the wheels of the vehicle and engaging at every revolution, pushes forward one of the teeth, which is replaced by the following one, and so on indefinitely. Whence it results that when the wheel of the vehicle has made eight revolutions, the face wheel will have made one. Now, to the center of the latter there is fixed perpendicularly, by one of its extremities, a screw which, by its other extremity, engages with a crosspiece fixed to the sides of the box. This screw gears with the teeth of a wheel whose plane is perpendicular to the bottom of the box. This wheel is provided with an axle whose extremities pivot against the sides of the box. A portion of this axle is provided with spirals formed in its surface, so that it becomes a screw. With this screw there gears a toothed wheel parallel with the bottom of the box. To this wheel is fixed an axle, one of the extremities of which pivots upon the bottom, while the other enters the crosspiece fixed to the sides; and this axle likewise carries a screw that gears with the teeth of another wheel placed perpendicular to the bottom. This arrangement may be continued as long as may be desired, or as long as there is space in the box; for the more numerous are the wheels and screws, the longer will be the route that one will be able to measure. In fact, every screw, in making one revolution, causes the motion of one tooth of the wheel with which it gears; so that the screw carried by the face wheel, in revolving once, indicates eight revolutions of the wheel of the vehicle, while it moves only one tooth of the wheel upon which it acts. So, too, the said toothed wheel, in making one revolution, will cause the screw fixed to its plane to make one revolution, and a single one of the teeth of the succeeding wheel will be thrust forward. Consequently, if this new wheel has again thirty teeth (and this is a reasonable number), it will, in making one revolution, indicate 7,200 revolutions of the wheel of the vehicle. Let us suppose that the latter is ten cubits in circumference, and this would be 72,000 cubits, that is to say, 180 furlongs. This applies to the second toothed wheel. If there are others, and if the number of teeth likewise increases, the length of the journey that it will be possible to measure will increase proportionally. But it is well to make use of an apparatus so constructed that the distance which it will be able to indicate does not much exceed that which it is possible to make in one day with the vehicle, since one can, after measuring the day’s route, begin anew for the following route. This is not all. As one revolution of each screw does not correspond with mathematical accuracy and precision to the escapement of one tooth, we shall in an express experiment cause the first screw to revolve until the wheel that gears with it has made one revolution, and shall count the number of times that the wheel will have revolved. Let us suppose, for example, that it has revolved twenty times while the adjacent wheel has made a single revolution. This wheel has thirty teeth; therefore, twenty revolutions of the face wheel correspond to thirty teeth of the toothed wheel moved by the screw. On the other hand, the twenty revolutions allow 160 teeth of the face wheel to escape, and this makes a like number of revolutions of the wheel of the vehicle, that is to say, 1,600 cubits; consequently, a single tooth of the preceding wheel indicates 53-1/3 cubits. Thus, for example, when, in starting from the origin of the motion, the toothed wheel will have revolved by fifteen teeth, this will indicate 800 cubits, say two furlongs; upon this same wheel we shall therefore write 53-1/3 cubits. Making a similar calculation for the other toothed wheels, we shall write upon each one of them the number that corresponds to it. In this way, after we ascertain how many teeth each has moved forward, we shall know by the same the distance that we have traveled. Now, in order to be able to determine the distance traveled without having to open the box in order to see the teeth of each wheel, we are going to show how it is possible to estimate the length of the route by means of an index placed upon the external faces. Let us admit that the toothed wheels of which we have spoken are so arranged as not to touch the sides of the box, but that their axles project externally and are squared so as to receive indexes. In this way the wheel, in revolving, will cause its axle with its index to turn, and the latter will describe upon the exterior a circle that we shall divide into a number of parts equal to that of the teeth of the interior wheel. The index should have a length sufficient to describe a circumference greater than that of the wheel, so that such circumference may be divided into parts wider than the interval that separates the teeth. This circle should carry the number already marked upon the interval wheel. By this means we shall see upon the external surface of the box the length of the trip made. Were it impossible to prevent the friction of the wheels against the sides of the box, for one reason or another, it would then be necessary to file them off sufficiently to prevent the apparatus from being impeded in its operation in any way. Moreover, as some of the toothed wheels are perpendicular to and others parallel with the bottom of the box, so, too, the circles described by the indexes will be some of them upon the sides of the box and others upon the top. Consequently, it will be necessary to so manage that the side that carries no circle shall serve as a cover; or, in other words, that the box shall be closed laterally. Another engineer, probably Græco-Latin, since he expresses distances sometimes in miles and sometimes in stadia, has pointed out an arrangement of a different system for measuring the progress of a ship. We shall describe this apparatus, which we illustrate in Fig. 2. Let Α Β be a screw revolving in its supports. Let us suppose that its thread moves a wheel, Δ, of 81 teeth, to which is fixed another and parallel wheel, Ε (a pinion), of nine teeth. Let us suppose that this pinion gears with another wheel, Ζ, of 100 teeth, and that to the latter is fixed a pinion, Η, of 18 teeth. Then let us suppose that this pinion gears with a third wheel, Θ, of 72 teeth, which likewise is provided with a pinion, Κ, of 18 teeth, and again that this pinion engages with a wheel, Λ, of 100 teeth, and so on; so that finally the last wheel carries an index so arranged as to indicate the number of stadia traveled. On the other hand, let us construct a star wheel, Μ, whose perimeter is five paces. Let us suppose it perfectly circular and affixed to the side of a vessel in such a way as to have, upon the surface of the water, a velocity equal to that of the vessel. Let us suppose, besides, that, at every revolution of the wheel, Μ, there advances, if possible, one tooth of Δ. It is clear, then, that at every distance of 100 miles made by the vessel the wheel, Δ, will make one revolution; so that, if a circle concentric with the wheel, Λ, is divided into 100 parts, the index fixed to Λ will, in revolving upon this circle, mark the number of miles made by the number of the degrees. Odometers, like so many other things, have been reinvented several times, notably in 1662 by a member of the Royal Society of London, and in 1724 by Abbot Meynier. [Illustration: FIG. 2.--ODOMETER FOR VESSELS.]