(from θερμός hot, and μέτρον a measure). The various thermometric instruments admit of being arranged in classes according to the various phenomena they are intended to illustrate, an account of which will be found under HEAT, together with the precautions necessary to the estimation of temperature, as distinguished from the absolute quantities of heat contained in a body. Our business in the present article is to describe the thermometer, properly so called, under its most useful and important forms, referring to PYROMETER for those instruments which measure temperatures above the boiling-point of mercury. There is, however, a class of instruments named Calorimeters, the function of which is to compare the quantities of sensible heat evolved by different actions. There is another set of instruments by which the effects of different radiations or heating rays are compared; and as the indications of such instruments depend on the difference of temperature between their different parts, they have been named differential thermometers. Small quantities of radiant heat can also be measured by means of very feeble thermo-electric currents, as in the instrument called the thermo-multiplier.

The first thermometer dates from the beginning of the seventeenth century. It consisted of a tube open at one end, with a glass bulb at the other; and the open end being inserted into a cup of coloured liquid, and heat applied to the bulb, a portion of the air was expelled, and on cooling, a quantity of liquid rose by atmospheric pressure into the stem, equal to the quantity of air that had been expelled by the heat. Now, as air is sensitive to changes in temperature, any variation in that of the surrounding air would communicate its influence to the air imprisoned in the bulb, causing it to expand, and so depress the liquid in the stem when it became warmer, and to contract and cause the liquid to rise under the influence of cold. An instrument of this kind was formerly called a weather-glass, from its indicating warm and cold weather, but more commonly an air thermometer. It was improved by Boyle, who got rid of the bulb, and employed a tube open at both ends, and cemented into the neck of a bottle containing a coloured liquid, so as to confine a portion of air. By dipping the bottle into different liquids, their temperatures could be compared by the expanding or contracting air of the bottle, causing the ascent or depression of the liquid in the stem. The liquid, however, being open to the atmosphere, was liable to evaporation and contamination. A notable improvement in the instrument was made by the Florentine academicians, who removed the registering fluid from the action of the atmosphere; and instead of the ordinary scale of equal parts, constructed arbitrarily, employed fixed points such as should be applicable to all thermometers, and thus make one instrument comparable with another. The instrument consisted of a bulb and a stem of moderately fine bore, containing spirits of wine, which, being boiled to expel the air, the tube was hermetically sealed. The fixed points were, the cold of ice and snow for the lower limit, and the greatest summer heat of Florence for the upper one. But as these points are subject to variation, the instrument was still wanting in one of its essential characters. When the instrument was brought to England, Boyle substituted coloured for colourless spirit, which was an improvement. There was considerable discussion about the standard minimum of temperature, based on the erroneous notion that the freezing point of water varied in different climates. It was proposed to fix one point of the scale by noting the height of the liquid in the stem, when the bulb was placed in thawing oil or aniseed. Hooke is said to have preferred the temperature of freezing water; Halley proposed the boiling point of spirits as another fixed point; but Newton seems to have been the first to take advantage of the fact, if he did not discover it, that whenever a thermometer is placed in melting snow or ice, the liquid always stands at the same point in the tube, while the boiling point of water is almost equally constant. It was easy, then, to take these two points, and divide the interval between them into a certain number of equal parts or degrees, and continue the division above and below, and thus convert the instrument from a mere toy into an artificial organ of sense, whose observations at different times and in different places would not only be comparable with each other, but also with those of other instruments constructed under the same circumstances. As, however, the boiling point of water is higher than that of spirit, it was necessary to select as the registering liquid one that would not pass off into vapour on being immersed in boiling water. Newton suggested oil, but this was found to be sluggish in its motions, to vary in fluidity with the temperature, and to adhere strongly to the tube. Halley suggested the use of mercury, but rejected it from some mistaken notion of its small amount of expansibility; but Römer adopted mercury, and is said to have constructed the scale attributed to Fahrenheit, an instrument-maker, a native of Dantzig, resident in Amsterdam. His thermometers were so highly appreciated, that they were known all over Europe early in the eighteenth century.

The advantages of mercury as the thermometric fluid are, that it enlarges in bulk more equally for equal increments of heat than most other liquids, although in common with all other substances, an equal amount of expansion is not produced by equal increments of heat at different temperatures. According to Regnault, the total expansion of mercury for each of three progressive intervals of 180° is as follows:—between 32° and 212°, mercury expands one part in 55:08; between 212° and 392°, the expansion is one in 54:61; and between 392° and 572°, it is one in 54:01. A second advantage of mercury is, that it is more easily purged of air than alcohol or oil; in the third place, it has a range of more than 700°, while oil becomes viscid at low temperatures, and water begins to expand below 394°, and both it and alcohol boil before they attain a high temperature. Another advantage of mercury is the rapidity with which it accommodates itself to the temperature of surrounding bodies. The thermometers used by Mr Glaisher in his experiments on nocturnal radiation (Phil. Trans. 1847), were so sensitive that when taken from a room at the temperature of 60° into the open air at 37°, the latter temperature was indicated in about two minutes.

In order that mercurial thermometers shall be comparable with each other, or have any scientific value, numerous precautions are necessary in constructing and using them. In order to allow a considerable range, the bore of the tube should be capillary, and of equal diameter throughout. But from the mode of manufacturing such tubes, they are usually frusta of very elongated hollow cones. By drawing a drop of mercury into the tube, so as to form a silver line of a definite length, and moving it through different portions of the tube, and measuring it in each position, the accuracy of the bore can be tested, and those tubes selected which are most equable in bore. Tubes are sometimes made with an elliptical bore, so as to spread out the mercury and make it more apparent. The mercury can be moved along the bore by tying a bag of India-rubber containing dry air to one end of the tube, and compressing it. The same method must be adopted in blowing a bulb at the end of the tube, after having made the glass soft in the flame of a table-blowpipe. Air from the mouth would introduce moisture, which would be injurious. The precautions adopted at the Meteorological Observatory at Kew will be noticed hereafter, our present object being to show the ordinary method of constructing the instrument. The size of the bulb must depend on circumstances; if it be large in proportion to the stem, the instrument will be sensitive, but if too large, there is the danger of the great mass of mercury affecting the temperature of a small amount of matter into which the bulb may have to be placed. A spherical bulb is to be preferred to a cylindrical or a pyriform one, on account of the more equal pressure of the atmosphere on a spherical body, a point of some importance, as will presently be explained. In filling in the mercury, which of course should be quite pure, the tube should be placed nearly horizontal with its open end below the surface of the metal, and the bulb be heated by means of a spirit-lamp. A good deal of the air will thus be expelled, and as the bulb cools, a portion of the metal will enter the tube and bulb, but the bulb must be again heated, so as to boil the mercury, when vapour of the metal will expel all Thermometer.

The bulb will become filled with the fluid metal. Some makers fill the instrument by tying a little paper cone round the end of the tube, putting mercury into it, and holding the bulb over a spirit-lamp. Another plan is to blow a second bulb on the tube near the open end, and on heating this bulb while the open end is immersed in mercury, a portion of the metal will pass into that bulb as the glass cools. The tube is then held upright, the lower bulb is heated, and the air escapes through the mercury in the upper bulb. The cooling and heating are repeated until the lower bulb is quite filled, and a little mercury remains in the upper bulb. The tube is next held over a charcoal fire, so as to vapourise a portion of the mercury, and expel all the air and moisture; before the tube is removed from the fire, the open end is stopped with wax for the purpose of excluding the air; and when the tube is cool, it can be inclined so as to allow a portion of mercury from the upper bulb to fall into the stem to any required height. When the quantity of mercury has been properly adjusted, a blowpipe flame is applied a little below the upper bulb, and the tube is hermetically sealed, and at the same time detached from the upper bulb. Another method, when there is only one bulb, is to heat the mercury until it rises to the point where the tube is finally sealed. The excess of mercury is thus driven out of the tube, and when the contents have cooled down to the temperature of the air, a sufficient range is still left in the tube for all ordinary purposes. The space thus left between the top of the mercury and the sealed end of the tube should be a perfect vacuum, so that when the tube is inverted, the mercury should entirely fill it, and fall with a metallic click to the end of the tube.

A number of tubes having been thus prepared, the first step, preparatory to graduation, is to determine the fixed point—namely, that of freezing for thermometers of atmospheric range, and also the boiling point for chemical thermometers. To determine the first point, the tubes are placed in a vessel containing melting snow or ice, when the mercury will sink in each stem down to a certain point, and remain steady while the liquefaction is going on. The amount of depression in each case will depend on the temperature of melting ice or snow, which is constant; also on the relative sizes of the bulb and stem, and on the quantity of mercury inclosed in them. A scratch is made with a file on each tube, at the point where the column of mercury terminates; and this is the freezing-point of the instrument, or the temperature of melting ice, which is the same in every part of the world, and at all seasons of the year. The boiling-point, however, varies with the height of the barometer; but supposing this to represent mean pressure, or 30 inches, the boiling-point of pure water would always be 212°. The bulb should be placed just over boiling water contained in a metallic vessel, so that both the bulb and stem may be enveloped in the steam. A portion of the stem may pass out through a perforated cork in the top of the vessel, and a file-mark be made on the glass opposite the point where the mercury remains stationary.

About the middle of the last century, the Royal Society of London appointed a committee to inquire into the best method of constructing the thermometer. They recommended that the boiling-point should be fixed at a pressure of 29°80 inches; that the bulb be not immersed in the water, because the steam, under such circumstances, being slightly under pressure, indicated a somewhat higher temperature; that the boiler be of tinned iron, with an easily-fitting cover, made steam-tight by a ring of woollen cloth; that the cover have two apertures, one to serve as a chimney, not less than half a square inch in area, and about 3 inches high, for conveying away the steam, and the other hole for a cork through which the thermometer-stem was Thermometer-ter. to pass, so that the bulb and stem should be completely surrounded by steam, and no more of the stem be exposed above the cork than was sufficient to show the height of the mercury when the water was boiling briskly. When the barometer was not standing at 29°80, a table of corrections was furnished, in which the interval between the freezing and the boiling points was divided into 1000 parts, a certain number of which was to be added to or taken from the interval, according as the observed height of the barometer was above or below the assigned mean.

After the bulb has been blown at the end of the tube, the mercury filled in, and the vacuum produced, some molecular change takes place in the glass, either in consequence of the unbalanced atmospheric pressure on the outside, or the disturbance in the particles occasioned by the blowing and the filling; but whatever the cause, the effect is, that the fixed points become changed, if the graduation be made immediately after the filling of the instrument. It was noticed by Professor Danielli, in the case of two thermometers constructed with much care by the Hon. Henry Cavendish, where the degrees were very large, and distinctly divided into 10ths; that on immersing the bulbs into pounded ice, and leaving them there for half an hour, the freezing-point on the scale of one instrument was 0°4 too low, and in that of the other 0°35°. Hence it is recommended to allow 10 or 12 months to elapse between the sealing and the graduation of a thermometer, since it is found that the contraction of the tube takes place for the most part soon after it has been sealed.

The portion of the tube between the two fixed points corresponds to the expansion of the mercury between the freezing and the boiling-points of water, and this amounts to \(\frac{1}{35}\)th part of the volume which it occupies at the temperature of melting ice; so that the capacity of the tube between the two fixed points must always be equal to \(\frac{1}{35}\)th of the capacity of the bulb, and that portion of the stem which is below the freezing-point. In Fahrenheit's scale, the interval between the two fixed points is divided into 180 equal parts or degrees, either on the tube itself or on a stick of hard, dry wood, such as box, or a frame of ivory, glass, or metal. The tube being properly fixed to the wood, &c., a line is engraved opposite to the fixed points, the interval divided into 180 equal parts, which are continued above and below the two fixed points, beginning with 32° below the freezing-point as the zero of the scale; so that the boiling-point of water is 212°. Fahrenheit fixed his zero by immersing the bulb in a mixture of salt and snow; and erroneously supposing this to be the greatest amount of cold that could be produced, he made it the commencement of his scale. When, however, temperatures below zero are noted, the scale is continued below zero with the negative sign affixed. Fahrenheit's scale was not used in France; but up to the time of the Revolution Réaumur's was in use, in which the zero of the scale commenced from the freezing-point, and the interval between it and the boiling-point was divided into 80 parts; so that a degree on this scale was larger than one on Fahrenheit's in the proportion of \(2\frac{1}{2}\) to 1. Thus, to convert an observation on Réaumur's scale into a corresponding one on Fahrenheit's, we must multiply Réaumur's degrees by \(2\frac{1}{2}\), and add to the product 32, to allow for the distance of the points at which the scale commences. In 1742 Celsius, the Swedish astronomer, contrived a thermometer, the scale of which also commenced at the freezing-point of water, while the interval between it and the boiling-point was divided into 100°. This instrument was adopted by the French at the time of the Revolution under the name of Thermomètre Centigrade, because it harmonised with their decimal system of weights and measures. Each of thermometer's degrees is equal to 1/5th of a degree of Fahrenheit; so that to convert a centigrade temperature into a corresponding one of Fahrenheit, the number of degrees must be increased in the proportion of 5:9, and 32 added to the result. The following formulae represent all the cases that can occur:

\[ \text{Fahrenheit to Centigrade} = \frac{5}{9} (F^\circ - 32) = C^\circ \]

\[ \text{Centigrade to Fahrenheit} = \frac{9}{5} C^\circ + 32 = F^\circ \]

\[ \text{Réaumur to Fahrenheit} = \frac{9}{4} R^\circ + 32 = F^\circ \]

\[ \text{Fahrenheit to Réaumur} = \frac{4}{9} (F^\circ - 32) = R^\circ \]

As mercury freezes at -39° Fahr., thermometers intended to register temperatures lower than that point must have spirits of wine or alcohol substituted for mercury as the registering fluid. The method of graduating spirit thermometers will be referred to presently.

Mercurial thermometers exposed to high temperatures and great changes of temperature, especially when made of different kinds of glass, are not strictly comparable; and even the same thermometer is not always uniform in its indications, apart from the shifting of the freezing-point, as already noticed, and which will be again referred to. Professor Dixon, in his Treatise on Heat, says, "Different glasses have different co-efficients of expansion, and also vary in the law of their dilatation at high temperatures; and as the amount of absolute dilatation of mercury is small, this variation in the expansion of the glass envelope produces irregularities of considerable magnitude in the apparent dilatation of mercury. As the real expansibility of air is much greater, its apparent expansion in glass is not affected to the same extent by these variations in the rate of expansion of the latter; and accordingly, in an air thermometer, the rate of expansion of the glass may be considered as sensibly uniform. When corrected, therefore, for the expansion of its envelope, such an instrument forms the most perfect thermometer with which we are acquainted in the present state of science." An air thermometer, corrected for the expansion of its envelope, compared with a mercurial thermometer constructed with the peculiar description of glass employed by M. Regnault in his experiments, the agreement between the two instruments was perfect up to 200°C; whereas in a mercurial thermometer made of ordinary tube, compared with an instrument in which the tube was of crystal glass, although they agreed from 0° to 100°C, yet at high temperature the discrepancies were as follows:

| 1st Thermometer | 2d Thermometer | Difference | |-----------------|---------------|-----------| | 126.51 | 191.66 | 115 | | 246.68 | 249.36 | 268 | | 251.87 | 254.57 | 270 | | 279.08 | 282.50 | 342 | | 310.69 | 315.28 | 459 | | 333.72 | 340.07 | 635 |

The Meteorological Observatory at Kew, established and supported by the British Association for the advancement of Science, has devoted considerable attention to the construction and comparison of Meteorological instruments. The late Mr Welsh, in his valuable Report on the thermometer, has adopted the plan of operations recommended by M. Regnault, which may be arranged under the following heads:—1st, The calibration of the tube; 2d, The graduation of the scale; 3d, The determination of the scale coefficients. We will give a few details under each of these heads, referring the reader who is desirous of more minute information to the Report itself. 1. In the calibration of the tube, a short column of mercury having been introduced into it, the tube is attached to the frame of Perreaux's dividing engine, and, by means of flexible tubing, is connected at both ends with India-rubber air-bags, the pressure on which is regulated by screws. The mercury is brought to that part of the tube where the graduation is to be commenced. The cutting frame of the engine carries a small microscope, with cross wires in its focus; on turning the dividing screw, the microscope wire is made to coincide with the first extremity of the mercury, and the screw is then turned forward until the wire reaches the second extremity; the length of the column is thus given in revolutions of the screw. The mercury is then made to move along the tube by pressing on one of the India-rubber bags, until the first end again coincides with the microscope wire, when the length of the column is again measured, and the mercury again moved forward; this process is repeated until the column has been measured for each length of itself through the whole extent of the proposed scale. Permanent marks are made on the glass at the points of commencement, and ending of calibration. If the progress of the numbers shows any considerable irregularity in the tube, and as a verification of the first set of measures, the calibration may be repeated, commencing at a point one-half the length of the column in advance of the original starting point. A series of measures, interpolated from the two sets, may then be adopted. 2. With respect to the graduation, the measured lengths of the column of mercury, in its successive steps along the tube, correspond to equal volumes, so that, assuming the calibre of the tube not to vary throughout the small length of the calibrating column, it is evident that, by dividing the spaces occupied successively by the mercury into an equal number of parts, the divisions will represent the same capacity, although they may be of different lengths. Before making the tube into a thermometer, the divisions of the scale may be verified by introducing a longer column of mercury, and examining whether the column occupies an equal number of divisions in equal parts of the scale. Should there be any irregularity, a table of corrections is to be formed, but this is seldom necessary. The divisions are cut with a fine needle-point upon a coating of engraver's varnish, and are afterwards etched with fluoric acid. The required dimensions of the bulb may be found approximately by weighing a measured length of the mercurial column, and from the known expansion of mercury and its specific gravity, computing the capacity of the bulb. 3. The thermometer having been filled with mercury, we have an instrument in which the divisions of the scale represent equal increments of the volume of the fluid, but have an entirely arbitrary value. The chief object in graduating the tube with an arbitrary scale, is the convenience it affords of testing the divisions before it is converted into a thermometer. The instruments are kept sufficiently long to allow of the settlement of the freezing-point, and this is determined by placing the thermometer in finely pounded ice, from which the water is drained off as it melts. The boiling-point is taken at the temperature of steam of the same elasticity as that of the atmosphere. The steam is generated from distilled water. The height of the barometer is observed at the time; and the correction to a uniform height of 30 inches (reduced to 32°) is found from Regnault's table. In determining the fixed points, the stems of the thermometers are kept vertical. But if the thermometers are to be used in any other position, the fixed points must be determined in that position. The tendency of the freezing-point to rise if the graduation take place immediately after the filling, has already been noticed; but Mr Welsh points out another peculiarity in this respect:—After a thermometer has been exposed for some weeks to the ordinary temperature of the air, if its freezing-point be ascertained, and it be suddenly exposed for a short time to the temperature of boiling water, and again immediately placed in ice, the latter determination of the freezing-point... Thermometers will be lower than the former by as much as from 0.1° to 0.2°, and the freezing-point does not recover its former position for some time, probably two or three weeks.

The apparatus used for comparing the indications of different thermometers consists of a cylindrical glass vessel 15 inches deep and 8½ inches diameter, together with a stand for supporting the thermometers, and an agitator or flat ring of tinned iron, fitting easily within the vase for agitating the water, so as to preserve an equable temperature throughout. The stand is a vertical rod supported by a small tripod, resting on the bottom of the vessel. Hooks sliding on this rod serve for suspending the thermometers, which are arranged with their bulbs at the same height in a circle of 3 inches diameter round the rod, and are kept fixed by being strapped with elastic bands against a projecting six-rayed frame, attached to the supporting rod. In this way, six thermometers can be compared at once. The whole apparatus is placed on a wooden revolving stand, and in taking the observations, the observer first agitates the water briskly for some time, turns the revolving stand until each thermometer is brought in turn opposite to his eye, and reads off the scales as quickly as possible to an assistant who takes down the numbers. The six thermometers can be read off and recorded in twenty seconds. It is desirable to make more than one set of readings for each temperature, and also to reverse the order of observing the instruments, in order to avoid as much as possible the changes which may occur during the reading off.

In graduating mercurial thermometers, the practice is to consider the increments of volume as proportional to increments of temperature. This cannot be assumed to be the case in spirit thermometers. In testing some spirit thermometers graduated for low temperatures intended for the Arctic Expedition under Sir Edward Belcher, Mr Welsh proceeded to determine the rate of expansion of alcohol in glass as compared with that of mercury. The alcohol had been carefully prepared by Dr Miller, and its specific gravity at 60° was 0.796. A tube was calibrated and divided with an arbitrary scale, and its divisions were found, upon verification, to be of exactly equal capacity throughout. The tube was then furnished with a bulb, of the same dimensions as those intended to be supplied to the Admiralty, and was filled with alcohol. Comparisons were then made between the readings of this instrument and those of a standard mercurial thermometer, through as wide a range as was found practicable. The comparisons above the freezing point were taken in water; those below 32° were taken in freezing mixtures of ice and salt, or chloride of calcium. From these comparisons the law of expansion was deduced, for the details of which we must refer to Mr Welsh's Report.

Registering Thermometers, or those which show the maximum and minimum of temperature between the taking of one observation and the setting of the instrument previous to another, are numerous. Many of them are very complicated and liable to get out of order; hence they have gradually gone out of use in the presence of simpler and more modern contrivances: the practice of photographic registration adopted in observatories, is still more fatal to their existence. For a long time the best known register thermometer was that by Mr Six of Canterbury. It consists of a long cylindrical bulb and a tube, in the form of a siphon, with parallel legs, the shorter leg proceeding from the bulb, and the longer leg terminating in a small cavity. A portion of the two legs of the siphon is filled with mercury, and the bulb and the rest of the tube with spirits of wine. This double column of mercury gives motion to two indices, each consisting of a piece of iron wire capped with enamel at both ends, and pressing with a little friction against the tube by means of a thread of glass or a bristle, so as to remain stationary at the place where it is left by the retreating mercury. To prepare the instrument for an observation, the indices are brought down by means of a magnet, so as to touch the mercurial column. After this, should the temperature rise, the spirit expands, depresses the mercury in one leg, and causes it to rise in the other leg, and in doing so, drives the index before it until the maximum temperature has been attained. Then, as the temperature declines, this second index does not follow the retreating mercury, but remains fixed in the tube, to indicate the maximum temperature. When, on the contrary, the spirit contracts by cold, the mercury follows it, and pushes up the other index until the contraction ceases, when, the temperature rising, the spirit begins to expand, and forces back the mercury, but leaves the index fixed in the tube as a record of the minimum temperature which has been obtained. If the instrument be read every twenty-four hours it will give the greatest and least temperatures during the day. This instrument cannot be trusted for accurate observations, as it is apt to get out of order; and the use of two liquids of different degrees of expansibility is also an objection. The glass spring of the index is also liable to break, or the bristle to lose its elasticity, so as to render the indices useless; and if this were not so, the vertical position of the tubes renders any little agitation from wind, &c., liable to start the indices. Rutherford's maximum and minimum thermometers are free from some of these defects. Two thermometers, one a spirit and the other a mercurial, each furnished with a distinct scale, are mounted on the same frame with the stems horizontal. The mercurial thermometer contains a small piece of steel wire, which is pushed forward as the mercury expands by heat, and remains to mark the highest temperature, when the mercury begins again to contract. The minimum thermometer contains a bead of glass with a small enamelled knob at each end; it rests in the tube, and the expanding spirit passes it freely; but as the spirit contracts by cold, there is sufficient adhesion between the last film of spirit and the glass bead to enable the spirit to drag it to its utmost limit of contraction, where it will remain to note the minimum temperature. The instruments are set for a new observation by gently inclining the frame, but this must be carefully done or the indices will get entangled with the liquids. This has been to a certain extent remedied by placing a piece of iron wire within the axis of the bead; so that this, as well as the other index, may be set by means of a small magnet. Messrs Negretti and Zambra have an ingenious maximum thermometer, in which the tube placed horizontally is contracted near the bulb, so that, although the expanding mercury passes freely through the contraction, yet when the temperature declines, the mercury cannot return through the contracted part; so that the maximum temperature is indicated by the extremity of the detached mercurial column. In setting the instrument for a new observation, all that is necessary is to place the tube in a vertical position, when the detached column, assisted by a gentle swing, will unite itself with the rest of the mercury. Professor Phillips has constructed a maximum thermometer, in which a portion of the mercurial column is separated from the rest by means of a small bubble of air. When the mercury begins to contract, the portion above the air-bubble remains fixed, while that below it retreats towards the bulb. Walferdin's maximum thermometer has at the top of the tube a small reservoir to receive the mercury which overflows by expansion. In preparing the instrument for an observation it is inverted, so that the end of the tube is immersed in the mercury of the reservoir. In this position the instrument is exposed to a temperature lower than any that it will meet with in the observations that are about to be made. The instrument is then brought back to its proper position when the tube is full to the top. As the temperature rises, the mercury expands and is forced out of the tube into