The Boy's Playbook of Science Part 44

If a red-hot ball is placed in the focus of a concave metallic speculum, it gives out certain emanations that are quite invisible, but which are reflected from the surface of the mirror in the same manner as visible rays of light, and may be collected in the focus of another and second concave speculum, when they can be concentrated on to a bit of phosphorus, and will cause the combustion of that substance. If the air from a pair of bellows is blown forcibly across the rays of heat as they are being concentrated upon the phosphorus, the rays are not moved from their course, they are no more blown away than a sunbeam darting through an aperture in a cloud on a stormy, windy day. The heat has, therefore nothing to do with the air, and is wholly independent of that medium in its pa.s.sage from one mirror to the other. Such an experiment as that described would at once suggest the idea that heat is a matter _sui generis_, a component part of all bodies, and given off from incandescent matter, the sun, &c., and that it may be propagated through s.p.a.ce much in the same manner as light. (Fig. 355.) The mechanism may be very much like the corpuscular movement of light as defined by Sir Isaac Newton, and already explained in another portion of this book. Hence it has been supposed that heat is propagated through the air, water, and solid substances by a direct emission of material particles from the heat-giving agent, and that these molecules of heat force their way into, or along, or through them, according to circ.u.mstances.

[Ill.u.s.tration: Fig. 355. Heat reflected by mirror, but not blown away by air from bellows.]

Certain bodies are almost transparent to heat rays, such as air, whilst others take an intermedial position, and only stop a certain quant.i.ty of the heat molecules, such as rock crystals, mirror gla.s.s, and alum. A third cla.s.s of bodies absorbs the heat plentifully, such as charcoal, black cloth, &c.; and a fourth, when polished and placed at the proper angle, reflects or throws off the heat, as in the case of polished mirrors. The transparency or opacity of substances (so far as light is concerned) [Page 370] does not affect the transmission of heat. Light of every colour and from all sources is equally transmitted by all transparent bodies in the liquid or solid form, but this is not the case with heat.

The rays of heat emitted by the sun and other luminous bodies have properties quite different to the rays of light with which they are accompanied. From these statements it will be evident that the _material theory_ of heat is surrounded with difficulties and anomalies that cannot be reconciled the one with the other, or neatly adapted, fitted in, and dovetailed with all the puzzling phenomena that arise. Our knowledge of the theory of heat has been greatly a.s.sisted by the researches of Melloni, who has demonstrated that different _species_ of rays of heat are given off by the same body at different temperatures, which may be distinctly sifted and separated from each other. Long before the experiments of Melloni philosophers had endeavoured to weigh heat; trains of the most delicate levers were exposed, without effect, to the action of heat rays; and all attempts, experimental as well as theoretical, to define heat by the _material_ theory, are imperfect, crude, and unsatisfactory. We are perforce obliged to adopt another theory, and the one that obtains the greatest favour, as offering the best definition of heat, is the _dynamical_ theory, which is more or less a.n.a.logous to the undulatory theory of light. At pages 262, 328, 335, this theory has been partly explained, and in speaking of it again, great care must be taken not to confuse the undulations of heat with those of light. The sun and the stars swim in a molecular medium, and 39,180 vibrations or waves must occur in one inch to produce the sensation of red light, and 57,490 undulations in the s.p.a.ce of one inch to produce a violet light. As vibrations of the ethereal molecules affect the eye, so there may be other nerves in our bodies which are peculiarly sensitive to the waves of heat. It requires eight vibrations of the air to occur in a second to produce an audible sound; whilst if the vibrations of the air amount to 25,000 per second they cannot be appreciated by the human ear, although it is possible to conceive that the ears of certain animals may be so susceptible of rapid vibrations that they may be able, for certain wise purposes of the Creator, to appreciate sounds which are inaudible to human ears.

Melloni exhibited a spectrum to a number of persons, and found that there was more light apparent to some eyes than to others. Lubeck put a scarlet cloth on a donkey, and found that the two were frequently confounded together by the eyes of many spectators. These facts indicate that there may be vibrations of molecules that produce the sensation of heat, but which do not affect the nerves that are sensitive to the action of light waves, and vice versa; and it is also probable that all these different undulations, some affording heat and some light, may be generated and propagated through s.p.a.ce, as from the sun; or through shorter distances, as from burning lamps and fires, without in any way interfering with or impeding each other's progress.

The dynamical theory seems to offer the best idea of the transmission [Page 371] of heat which is carried, conducted, or propagated through solids with variable rapidity, either by the vibration of the const.i.tuent molecules of the body itself, or by the undulation of a rare subtle fluid which pervades them. If a copper and iron wire of the same length and diameter are bound together and heated at the point of union, the waves of heat travel faster through the copper than the iron, and the former is said to be the best conductor of heat; and the fact itself is demonstrated by placing a bit of phosphorus at the end of each metallic wire, and it will be found by experiment that the combustible substance melts first and takes fire on the copper, and that a considerable interval of time elapses before the phosphorus ignites on the iron.

[Ill.u.s.tration: Fig. 356. C. Copper wire bound at A to I, an iron wire.

After the heat of the lamp has been applied for about five minutes the heat travels to C first, and ignites the bit of phosphorus placed there.

After some time has elapsed the phosphorus at I also ignites.]

The same fact is exhibited in a most striking manner by inserting a series of rods of equal lengths and thicknesses in the side of a rectangular box, allowing them to pa.s.s across the interior to the opposite side. The rods are composed of wood, porcelain, gla.s.s, lead, iron, zinc, copper, and silver, and have attached to each of their extremities, by wax or tallow, a clay marble. When the water placed in the box is made to boil, the heat pa.s.ses along the different rods, and melting the wax or tallow, allows the marble to drop off. Consequently the first marble would drop from the silver rod, the next from the copper, the third from the iron, the fourth from the zinc, the fifth from the lead, whilst the porcelain, gla.s.s, and wooden rods would hardly conduct (in several hours) sufficient heat to melt the wax or tallow, and discharge the marbles.

_Conduction of Metals._

Gold 1000 Silver 973 Copper 898.2 Iron 374.3 Zinc 363 Lead 179.6

[Page 372]

The experiment is made more striking if the marbles are allowed to fall on a lever connected with the detent of a clock alarum, which rings every time a marble falls from one of the rods. (Fig. 357.)

During a cold frosty day, if the hand is placed in contact with various substances, some appear to be colder than others, although all may be precisely the same temperature; this circ.u.mstance is due to their conducting power: and a piece of slate seems colder than a bit of chalk, because the former is a much better conductor than the latter, and carries away the heat from the body with greater rapidity, and diffuses it through its own substance.

[Ill.u.s.tration: Fig. 357. A B. Trough containing boiling water, heated by gas jets below. C. The eight rods and marbles attached, one of which has fallen. D. The tray to receive the marbles.]

The gradual pa.s.sage of heat along a bar of iron as compared with one of copper, is well ill.u.s.trated by supporting the ends of the two bars on the top of the chimney of an argand lamp, whilst the other extremities are held in a horizontal position by little blocks of wood. If marbles are attached by wax to the under side, they fall off as the heat travels along the metallic bars, and more rapidly from the copper than the iron, because the former is a better conductor of heat than the latter. (Fig.

358.)

[Ill.u.s.tration: Fig. 358. A. Section of an argand gas lamp, with a copper chimney supporting the ends of the bars of copper and iron marked C and I. The b.a.l.l.s have fallen from C, the copper bar.]

From the experiments of Mayer, of Erlangen ("Ann. de Ch.," x.x.x.), it would appear that the conducting powers of different woods are to a certain extent to be regarded as in the inverse proportion to their specific gravities--_i.e._, the greater the density of the wood the less conducting power, and the contrary.

If a cylindrical bar or thick tube of bra.s.s, six inches long, and about two inches in diameter, is attached to a wooden cylinder of the same size, the conducting powers of the two substances are well displayed by first straining a sheet of white paper over the bra.s.s, and then holding it in the flame of a spirit lamp. The heat being conducted rapidly away by the metal will not scorch the paper, until the whole arrives at a uniform high temperature; whereas the paper is rapidly burnt when [Page 373] strained over the wooden cylinder, because the heat of the flame of the lamp is concentrated upon one point, and is not diffused through the ma.s.s of the wood. (Fig. 359.)

In the course of the highly philosophical experiments of Sir H. Davy, which led him gradually to the discovery of the construction of the safety lamp, he connected together, by a copper tube of a small bore, two vessels, each containing an explosive mixture composed of fire damp and air. When the mixture was fired in one vessel he found that the flame did not appear to be able to travel, as it were, across the bridge--viz., the copper tube--and communicate with the other magazine, because it was deprived of its heat whilst pa.s.sing through the tube, and was no longer flame, but simply gaseous matter at too low a temperature to effect the inflammation of the mixture in the second box.

[Ill.u.s.tration: Fig. 359. Cylinder, half bra.s.s and half wood. The paper strained over the wood is taking fire. The other extremity, shaded, is the bra.s.s portion.]

A ma.s.s of cold metal may be suddenly applied to a small flame, such as that of a night light, and depriving it rapidly of heat (like the case of the unfortunate Russian described at page 354), it is almost immediately extinguished (fig. 360), not by the mere exclusion of the oxygen of the air, but on account of the withdrawal of the heat necessary for the maintenance of the combustion.

[Ill.u.s.tration: Fig. 360. A. Small flame from night light. B C. Large ma.s.s of cold copper wire open at both ends to place over flame, and by conduction of the heat to extinguish it.]

Sir H. Davy first thought of making his safety lamp with small tubes, which would supply fresh air, and carry off the burnt or foul air, at the [Page 374] same time they were to be so narrow that no flame could pa.s.s out of his lamp to communicate with an outer explosive atmosphere; and in speaking of his lamp with tubes he says:--"I soon discovered that a _few apertures_, even of very small diameter, were not safe unless their _sides_ were very _deep_; that a single tube of one-twenty-eighth of an inch in diameter, and two inches long, suffered the explosion to pa.s.s through it; and that a _great number_ of small tubes, or of apertures, stopped explosion, even when the depths of their sides was only equal to their diameters. And at last I arrived at the conclusion that a _metallic tissue_, however thin and fine, of which the apertures filled more s.p.a.ce than the cooling surface, so as to be permeable to air and light, offered a _perfect barrier_ to _explosion_, from the force being divided _between_, and the heat communicated to an _immense number of surfaces_. I made several attempts to construct safety lamps which should give light in all explosive mixtures of fire damp, and after complicated combinations, I at length arrived at one evidently the most simple, that of _surrounding the light entirely by wire gauze, and making the same tissue feed the flame with air and emit light_."

If a number of square metallic tubes of a fine bore are placed upright side by side, and a section cut off horizontally, it would represent the wire gauze which possesses such marvellous powers of sifting away the heat from a flame, so that it is destroyed in its attempted pa.s.sage through the metallic meshes; and of this fact a number of proofs may be adduced.

A gas jet delivering coal gas may be placed under a sheet of wire gauze, the gas permeates the gauze, and may be set on fire at the upper side, but the flame is cut off from the mouth of the jet by the cooling action of the wire gauze. The same experiment reversed, by holding the gauze over the gas burning from the jet, shows still more decidedly that flame will not pa.s.s through the metallic tissue. (Fig. 361.)

[Ill.u.s.tration: Fig. 361. A A. A number of square tubes placed upright.

The arrow shows the direction of the section to obtain a figure like wire gauze.]

Sir H. Davy again says: "Though all the specimens of fire damp which I had examined consisted of carburetted hydrogen mixed with different small proportions of carbonic acid and common air, yet some phenomena I observed in the combustion of a _blower_ induced me to believe that small quant.i.ties of olefiant gas may be sometimes evolved in coal mines with the carburetted hydrogen. I therefore resolved to make all lamps safe to the test of the _gas produced by the distillation of coal_, which, when it has not been exposed to water, always contains olefiant gas. I placed my lighted lamps in a large gla.s.s receiver through which there was a current of atmospherical air, and by means of a [Page 375]

gasometer filled with coal gas, I made the current of air which pa.s.sed into the lamp more or less explosive, and caused it to change rapidly or slowly at pleasure, so as to produce all possible varieties of inflammable and explosive mixtures, and I found that iron gauze wire composed of wires from one-fortieth to one-sixtieth of an inch in diameter, and containing twenty-eight wires or seven hundred and eighty-four apertures to the inch, _was safe under all circ.u.mstances in atmospheres of this kind_; and I consequently adopted this material in guarding lamps for the coal mines, when in January, 1816, they were immediately adopted, and have long been in general use."

The remarkable conducting power of wire gauze is further shown by placing some lumps of camphor on a piece of this material, and when the heat of a spirit-lamp is applied on the under side of the gauze, the camphor volatilizes, and as the vapour is remarkably heavy, it falls through the meshes of the gauze, and takes fire; but the most curious and further ill.u.s.tration of the conducting power of the wire meshes is shown in the fact that the fire does not communicate through the thin film of gauze to the lumps of camphor placed upon it.

The camphor may be ignited by applying flame to the upper side of the gauze, showing that, although this substance is so exceedingly combustible, it will not take fire even if placed at no greater distance from flame than the thickness of the wire gauze, provided the latter material is interposed between it and the flame.

A square box made of wire gauze, with a hole at the bottom to admit a candle or spirit-lamp, may have a considerable jet of coal gas forced upon it from the outside, or a large jug of ether vapour poured upon it; and although the box may be full of flame, arising from the combustion of the gas or ether, the fire does not come out of the wire box or communicate with the jet or the ether vapour as it is poured from the jug. (Fig. 362.)

[Ill.u.s.tration: Fig. 362. A box made of wire gauze, with a hole in the bottom to admit a spirit lamp lighted. A hot jug full of the vapour of ether may be poured on to the flame, but it only burns inside the box, and does not communicate with that in the jug.]

Sir Humphrey Davy's safety lamp consists of a common oil-lamp, _f_, with a wire through the cistern for the purpose of raising or depressing the cotton wick without uns.c.r.e.w.i.n.g the wire gauze; _b_ is the male screw fitting the screw attached to the cylinder of wire gauze, which is made double at the top. The entire lamp is shown at A, whilst the platinum coil which Sir H. Davy recommends should be wound round the wick is shown at _h_. The small [Page 376] cage of platinum consists of wire of one-seventieth to one-eightieth of an inch in thickness, fastened to the wire for raising or depressing the cotton wick, and should the lamp be extinguished in an explosive mixture, the little coil of platinum begins to glow, and will afford sufficient light to guide the miner to a safe part of the mine. With respect to this platinum coil, Sir H. Davy gives a careful charge, and says:--"The greatest care must be taken that no filament or wire of platinum protrudes on the exterior of the lamp, _for this would fire externally an explosive mixture_."

[Ill.u.s.tration: Fig. 363. Sir Humphrey Davy's safety lamp.]

Since the invention of the Davy lamp, a great number of modifications have been brought forward, some of which for a short time have occupied the public attention, but whether from increased cost or a sort of inertia that arrests improvement, it is certain that the lamp originally devised by Sir Humphrey Davy is still the favourite. It was perhaps unfortunate that the lamp was called the _safety_ lamp, because it is not so under every circ.u.mstance that may arise, unless it happens to be in the hands of persons who have taken the trouble to study it and understand how to correct the faults. The lamp might have escaped the incessant attacks that have been made upon its just merits, if the name had simply been that of its ill.u.s.trious inventor--"a Davy lamp." No one could carp at that, whilst "safety" was held to mean perfect immunity from every possible and probable danger that might arise in the coal-pits. The lamps are now usually placed under the charge of one man, who trims them and ascertains that the wire gauze is in perfect order; this latter is usually locked upon the lamp, and as it is a penal offence, and punishable by a heavy fine and imprisonment, to remove the wire gauze from safety lamps in dangerous parts of the mine, of course the miners are being gradually brought to a sense of the obligations they owe themselves and their brother-miners, and the rash, ignorant, and foolhardy offences of breaking open safety lamps for more illumination, or to light pipes, are becoming much less frequent than formerly. One of the most ingenious "detector lamps" is that of Mr.

Symons, of Birmingham. (Fig. 364.) It consisted of the old-fas.h.i.+oned Davy, but [Page 317] inside the rim of the wire gauze is placed a small extinguisher and spring, which does not move so long as the gauze is screwed _on_ to the lamp, but directly the gauze is unscrewed, the reversed movement releases the detent, and the extinguisher falls upon the light. In spite of the manifest ingenuity of this lamp, it is not adopted, because it costs a trifle more than the ordinary "Davy." To show the remarkable perfection of the wire gauze principle, some turpentine may be poured upon a lighted safety lamp, when a great smoke is produced by the evaporation of the spirit, but no flame pa.s.ses through to the outside, although the turpentine burns inside the lamp.

If some coa.r.s.e gunpowder is laid upon two thicknesses of fine wire gauze, it may be heated from below with the flame of the spirit lamp, and the sulphur will gradually volatilize without setting fire to the ma.s.s of powder. To show the security of the Davy lamp, it may be lighted and hung in a large box with gla.s.s sides, open at the top, and a jet of coal gas supplied at the bottom; as this rises and diffuses in the air, the mixture becomes explosive, and the fact is at once evident by the alteration in the appearance of the flame of the lamp, which enlarges, flickers, and frequently goes out, in consequence of the suddenness with which the explosion of the mixture takes place inside the lamp, producing a concussion that extinguishes the flame. In this case the utility of the platinum coil is very apparent, and it continues to glow with a red heat until the explosive character of the air in the box is changed.

[Ill.u.s.tration: Fig. 364. Symons' self-extinguis.h.i.+ng Davy lamp.]

If a large washhand-basin is first warmed by some boiling water, which is then poured away, and a drachm of ether thrown in, a highly-combustible atmosphere is obtained, and when a lighted Davy lamp is placed into the basin so prepared, the flame inside the lamp immediately enlarges and flickers, but is not extinguished, and does not communicate to the combustible vapour outside. The contrast between the safety lamp and an unprotected flame is very striking; if a lighted taper is thrust into the basin, the ether catches fire, and burns with a very large flame. The solid conductors of heat, which are said to enjoy this property in the highest degree, are the metals, marble, stone, slate, and [Page 378] other dense and compact solid substances; whilst the opposite quality of being non-conductors, or nearly so, is possessed by fur, wood, silk, cotton, wool, eider and swansdown, paper, sand, charcoal, and every substance which is of a light or porous nature. The practical application of this knowledge is very apparent in the affairs of every-day life. Thus we rise in the morning, and immediately after the necessary ablutions, if it is winter time, proceed to encase the body in non-conductors, such as flannel and wool. When we sit down to the breakfast table to make tea, we may notice the contrivances for preventing the handle of the top of the urn, or that of the teapot, from becoming too hot for the fingers, by the interposition of ivory or wood.

If asked to place water in the teapot from the kettle, we instinctively seek for the well-worn kettle-holder made of Berlin wool, and therefore a bad conductor. As we cut our meat or fish at the same meal, we may s.h.i.+ver with cold, but our fingers are not quite frozen by contact with the steel knives, as we hold them by ivory handles; and we are agreeably reminded that some metals are good conductors of heat, by the pleasant warmth of the silver teaspoons, as we stir our tea or coffee.

Even the polish of the well-rubbed mahogany is protected from the heat of the dishes by non-conducting mats, and plates are handed about, if "nice and hot," with a carefully-wrapped non-conducting linen napkin.

Supposing we prefer a bit of fresh-made toast, the fork is provided with a non-conducting handle; and should we peep out of window some wintry morn whilst the baker delivers his early work in the shape of hot rolls, we notice they come out of nicely-wrapped flannel or baize, which being a bad conductor is employed to retain their heat. We read, occasionally, in the military intelligence, statements respecting some newly-constructed sh.e.l.ls which are to burst and scatter melted iron (!!); and of course the idea of the interposition of a good non-conductor of heat between the bursting charge and the molten metal must be realized in their construction.

The _central heat_ of our globe is a reality that cannot be disputed, and after digging beyond a depth of twenty feet the thermometer gradually rises at the rate of one degree of Fahrenheit's scale for every fifteen yards. The bad conducting power of the crust of the earth must, therefore, be apparent, as it is easy, knowing the diameter of our globe, to calculate that the increase of heat downwards amounts to 116 for each mile, consequently at a depth of thirty and a half miles below the surface, there will be a temperature most likely equal to 3500, or a heat that might easily melt cast-iron, and would help to account for the earthquakes and eruptions of volcanoes, which still remind us by their terrible warnings, that we live only on the bad conducting upper crust of a globe, the inside of which is still, perhaps, in a liquid and molten state. Monsieur Fourier has demonstrated the non-conducting power of this sh.e.l.l by calculating that, supposing the globe was wholly composed of cast-iron, the central heat would require myriads of years to be transmitted to the surface from a depth of 150 miles; and by inverting the process of reasoning, we may come to the conclusion that the [Page 379] internal heat must be excessive, because it is confined and shut out from those influences that would carry off and weaken the intensity.

There are no two words, says Tyndal, with which we are more familiar than _matter_ and _force_. The system of the universe embraces two things, an object _acted upon_, and an agent _by which_ it is acted upon; the object we call matter and the agent we call force. Matter, in certain respects, may be regarded as the _vehicle_ of force; thus, the luminiferous ether is the vehicle or medium by which the pulsations of the sun are transmitted to our organs of vision. Or, to take a plainer case, if we set a number of billiard b.a.l.l.s in a row, and impart a shock to one end of the series in the direction of its length, we know what will take place; the _last ball_ will fly away, the _intervening_ b.a.l.l.s having served for the transmission of the shock from one end of the series to the other. Or we might refer to the conduction of heat. If, for example, it be required to transmit heat from the fire to a point at some distance from the fire, this may be effected by means of a conducting body--by a poker, for instance; thrusting one end of a poker into the fire, it becomes heated, the heat makes its way through the ma.s.s, and finally manifests itself at the other end. Let us endeavour to get a distinct idea of what we here call heat; let us first picture it to ourselves as an agent apart from the ma.s.s of the conductor, making its way among the particles of the latter, jumping from atom to atom, and thus converting them into a kind of _stepping stones_ to a.s.sist its progress. It is a probable conclusion, even had we not a single experiment to support it, that the mode of transmission must, in some measure, depend upon the manner in which those little molecular stepping stones are arranged. But we must not confine ourselves to the molecular theory of heat. a.s.suming the hypothesis, which is now gaining ground, that heat, instead of being an agent apart from ordinary matter, consists _in a motion of the material particles_; the conclusion is equally probable that the transmission of the motion must be influenced by the manner in which the particles are arranged. Does experimental science furnish us with any corroboration of this inference? It does.

More than twenty years ago MM. De la Rive and De Candolle proved that heat is transmitted through wood with a velocity almost twice as great along the fibre as across it. This result has been recently expanded, and it has been proved that this substance possesses three axes of calorific conduction; the first and greatest axis being parallel to the fibre; the second axis perpendicular to the fibre and to the ligneous layers; while the third axis, which marks the direction in which the greatest resistance is offered to the pa.s.sage of the heat, is perpendicular to the fibre and parallel to the layers.

If many solids are bad conductors of heat, they are at all events greatly surpa.s.sed by fluids, and especially by water. The conduction of heat by that fluid is almost imperceptible, so much so, that it has even been questioned whether liquids do really conduct heat downwards at all.

It has, however, been found that liquid mercury will conduct heat downwards, and therefore by a.n.a.logy it may be a.s.sumed that other liquids must possess a conducting power, although it may be exceedingly limited.

[Page 380]

In order to prove that water is an exceeding bad conductor of heat, a tube with a large gla.s.s bulb blown at one end is partly filled with tincture of litmus, until it will just sink below the surface of water placed in a tall cylindrical or open jar. If a copper basin, containing burning ether, is now floated on the top of the water, so as to leave about a quarter of an inch between the top of the air thermometer--viz., the bulb containing the coloured liquid--and the bottom of the copper pan, it will be noticed that whilst the water surrounding the latter almost boils, not the slightest effect arising from the conduction of heat can be perceived in a downward direction. After the ether has burnt out of the copper vessel, it may be removed, and the boiling water stirred down and around the air thermometer, when the air within it expands, drives out the colouring liquid, and the bulb becoming specifically lighter, rises to the top of the containing gla.s.s. (Fig.

365.)

[Ill.u.s.tration: Fig. 365. A A. Cylindrical gla.s.s full of water. B. The gla.s.s air thermometer containing the coloured liquid just standing upright, the mouth of the tube at C being open. D D is the copper basin containing the burning ether. E shows how the gla.s.s bulb and tube rise after the upper basin is removed, and the hot water comes in contact with and expands the air, making the thermometer light, and causing it to rise.]

Again, if the tube of an air thermometer is placed through a cork in the neck of a gas jar, inverted and standing on a ring stand, and the [Page 381] jar is then filled with water, and boiled at the top with a red-hot iron heater, the heat does not pa.s.s downwards and affect the thermometer. By introducing a syphon the water surrounding the thermometer at the bottom of the jar may be drawn off, until the hot water is within a fraction of an inch of the air thermometer, and still no heat is conducted, and the liquid in the latter remains stationary.