Page images






THERMOMETER; from the Greek words Esgueos, the thermometer has had several claimants, the resheat, and mergov, a measure; an instrument adapted pective merits of whom it seems nearly impossible for the comparative estimation of the temperature at the present day definitely to determine. Though of bodies.

by some Galileo has been considered as the true That the idea of any instrumental measurement of inventor, and by others, Father Paul Sarpi, the Veheat is wholly of modern invention seems beyond a netian, this honour chiefly rests between Sanctorius, doubt, since in the writings of the ancients we find an Italian physician, and Drebbel, a Dutchman, no trace of any such comparative scale except the both men of ingenuity and original minds. Many rude indications of our natural feelings. Since we moderns support each of these claims, but Sanctoare absolutely ignorant of the state of substances rius (or Santorio has, on the whole, most in his deprived of, or saturated with heat; since our senses, favour, since, as Martine justly observes,* no one though adapted to the perception of it, are inca- but he demanded the honour of an inventor during pable of forming any accurate measure of its effect, his lifetime. His Commentaries on Avicenna, pubor of computing its real quantity upon any inde- lished in 1626 are very interesting, since they conpendent standard as we can do the pressure of the tain descriptions and curious wood cuts of the va. atmosphere by knowing the weight of an equipon- rious forms of the thermometer which he proposed derant column of mercury, we are obliged to have for medical purposes; and the extent to which he recourse to a relative, not an absolute scale, and carried it in other matters, is shown by his attempt establish an arbitrary measure of heat by the ex- described and figured at folio 22 of that work, to tent of one of its acknowledged effects, that of ex- compare the heat of the sun's and moon's rays, an pansion. (See the Article Heat.)

attempt which afterwards became one of the most Hence, it is obvious, that in order to conceive delicate in natural philosophy. the thermometer to be a true measure of heat, we Perhaps Sanctorius and Drebbel invented much assume that by equal increments of that agency, about the same time this instrument, a circumthe expanding substance employed is uniformly en- stance by no means improbable, considering the larged in bulk, a postulate not of easy proof, but then advancing state of physical science. Be this which it has been ascertained in most cases of im- as it may, the instrument contrived by each was portance is nearly true. But this we shall explain precisely similar, in which air was employed as more fully afterwards; in the meantime we propose the expanding substance, which in several respects to give some account of the invention and history is remarkably well fitted for this application. of the simple thermometer, with the various im- The air thermometer, as shown at Plate DXXIV. provements which it has gradually undergone, and Fig. 1, consisted of a ball of glass A with a stem B after some directions for its practical construction, nearly filled with air, but having the lower part of proceed to the various ingenious modifications the tube occupied by a coloured liquor, which, which it has received for different scientific pur. when the air in A was expanded, was forced into poses, and we propose to conclude with some con- the recipient C, and hence a scale being applied to siderations on the application of the thermometer the stem, the dilatation of the enclosed air by heat to use, and the defects to which it is liable, without was marked by the descent of the coloured fluid. the most scrupulous attention.

In this form the instrument was obviously unfit for Like most other important discoveries, that of trying with any ease the temperature of Auids.

* Essays on Thermometers, p. 2, where the authorities in this controversy are quoted. VOL. XVIII. Part I.


Boyle therefore subsequently proposed to include by evaporation. This instrument was constructed both the air and the coloured fluid in one bulb as much in the same way as at present, the spirit beshown at Fig. 2, where the tube AB which is open ing dilated till it filled the whole tube, when it was at top, reaches below the surface of the fluid nearly quickly sealed, and on cooling, the fluid retired, to the bottom of the receptacle C, into the neck of leaving

nearly a vacuum above it. The great defect which the tube is hermetically sealed. The same of the Florentine weather glass, as it was commonly philosopher, however, demonstrated the grand de called, was the want of any fixed scale of graduafect of the air thermometer, that by a change of tion, on which account no instrument except those pressure in the atmosphere, as shown by the graduated by the original one of the academy, barometer, the elasticity of the inclosed air is could be comparable with any other, the only direcaltered independently of temperature, so as to

as to tion being that the cold of ice and snow should render the indications of the same instrument make it stand at 20 deg. and the greatest summer not comparable at different times. The air heats at Florence, at 80 degrees. thermometer was subsequently modified by the The spirit thermometer was faulty in several ingenious Hooke, in order to act as a barometer, other respects, yet it cannot but be thought fortuwhich it obviously does, if the effects of tempera- nate that this fluid, which is esteemed the second ture be corrected and those of pressure alone best for filling thermometers, should have been so shown, just as on the other hand, if the result of early thought of. The Florentine academicians variable pressure were neutralized, that of tempera- sometimes bent the tube of their instrument into a ture would be truly expressed. Hooke attached a toruous form, but this does not appear to be the mercurial thermometer to the original instrument, first attempt of the kind, for we observe at fol. 220 the temperature of which thus indicated gave the of Sanctorius's Commentaries on Avicenna, a curidata for separating the influence of dilatation ous figure of a similar contrivance. caused by heat, which was performed by means of From the invention of the spirit thermometer a sliding scale. By this elegant modification Hooke may be dated the epoch of the application of truly converted the air thermometer into a marine ba- scientific effects to the art of thermometry. Boyle rometer, which, however, was soon abandoned having received one of the Florentine instruments from the absorption that was found to take place of in England, set himself to effect some radical imthe excluded air by the coloured fluid. This defect provements upon a contrivance which it was obhas more lately been in a great measure remedied vious was capable of shedding new light and preciby the substitution of hydrogen gas instead of the sion over every branch of natural inquiry. He common included air, by Mr. Adie, who has reviv- therefore proposed, as a fixed point for graduation, ed this instrument under a very elegant and porta- the thawing of oil of aniseeds, which he preferred ble form, and under the name of the sympiesometer. to the freezing of water, (another point he also See METEOROLOGY.

mentioned,) as being more easily obtained in all The only other air thermometer we intend to no- seasons of the year, and because he had great tice is that of Amontons, who made the indicial doubts as to the constancy of the point of freezing fluid a column of mercury twenty-eight French of different kinds of water. Indeed this doubt inches long, so that the included air was subjected served for long after to delay the adoption of the to the pressure of two atmospheres; by this me- fixed points at present employed. Derham, Halthod he was enabled to measure high temperatures, ley, Musschenbroek, and other philosophers, till such as that of boiling water, without a scale of far on in the last century, believed that water froze such great length as the dilatation of air under the at different temperatures in different latitudes, common pressure would have required. It was, which Martine,+ only ninety years since, cites some however, subject to the same great defect as the experiments of his own to disprove. Besides this, Sanctorian thermometer, and was besides very un- Boyle falls into another mistake in not taking two wieldy and liable to accident. In the seventeeth fixed points, but endeavouring to compute the abcentury a modification of the air thermometer was solute expansion of the spirits, and graduating the proposed by Van Helmont and by Sturmius, which scale to ten-thousandths, or some other fixed numhas recently been revived under the names of Ther- ber of parts of expansion, a proposal which, though moscope and Differential thermometer; as this, extremely philosophical in the abstract, is too however, does not belong to the simple form of the difficult of execution to be successfully adoptinstrument, we shall give some account of it in ed in practice. A similar plan was proposed by another part of this article.

Hooke. The defects of the air thermometer having been The acute Halley afterwards turned his attention duly appreciated by the Florentine Academia del to this subject of growing importance. He first Cimento, that enterprising body published, in the proposed the temperature of deep pits for a standfirst volume of their transactions,* a description of ard, naturally enough suggested by the constancy a new thermometer, in which spirit of wine was of the thermometer observed by De La Rive and used as the expanding substance, which, as it others, in the caverns of the Observatory at Paris. might be hermetically sealed up in a glass tube or But the difficulty of finding such situations, indebulb, was free from any defect arising from press. pendently of the variation of their temperature in ure, as well as the possibility of any loss of fluid different latitudes, was enough to set aside this pro


Saggi di Naturali Esperienze.

† Essays, p. 15.

| Micrographia, p. 39.



posal. Halley afterwards devised the excellent of Fahrenheit, nor does it freeze with a great degree of standard derived from boiling liquids, though he cold; its expansibility too is great, being not much did not appreciate its real merit. He tried the less than that of alcohol; but with all these recompoint of vaporization of water, mercury, and spirit mendations, linseed oil has been found so viscid as of wine, and unfortunately preferred the latter; a to render any accuracy of observation impossible, point very uncertain, as depending greatly upon the and from the quantity which always adheres to the strength

of the fluid. The boiling point of water, tube, to make the actual height very uncertain. though Halley very much overlooked it, he found We now come to mention the greatest improveto be very steady in its indication. * Nor is it easi- ment made upon the thermometer since the period ly affected by minute accidental admixtures. The of its invention,--the practical introduction of principal source of error is that the vaporization of mercury as the expanding fluid. Halley had proAuids is affected by the state of atmosphere at the posed this substance, but was prevented from trytime, as to pressure, the temperature being highest ing it practically, by the consideration of its low when the pressure is greatest, and the reverse. By expansible power, which indeed he underrated, subsequent examination, this source of error has making it 7 between freezing and boiling water, become merely the object of an equation table instead of 1 as has been most lately determined. which will be given in the course of this article. This defect was certainly of consequence in the

We have already mentioned that Amontons con- 17th century, when the art of making thermometers structed an air thermometer which could measure with bores of great tenuity was neither understood temperatures as high as boiling water, which he not appreciated; and for the advantageous introheld as a fixed point in graduation, but hitherto the duction of mercury as a fluid, a proportionate imimportance of the points of freezing and boiling provement in manipulation was requisite. water was imperfectly estimated, and had never This practical discovery appears, as far as the been united in graduating the same instrument. It testimony of contemporary writers can guide us, was reserved for the powerful mind of Newton to to be due to Olaus Römer, the ingenious discoverer put this branch of science upon the same secure of the progressive motion of light, who, according footing with every other which he had studied. to Boerhaave, t likewise proposed the scale now Had not his mind been led to more magnificent known under the name of Fahrenheit's, and so earfields of investigation, he would probably have ef- ly as the year 1709 observed the mercury to sink fected the admirable refinements in thermometric by a natural cold to the zero of that scale; but it is science to which he only led the way, rejecting the more commonly conceived that Fahrenheit himself partial views of his predecessors, and laying the made that observation, and founded his scale upon foundation, as that of every philosophical super- it, which was contrived in the year 1720, and de- . structure should be laid, in principles as rigidly scribed to the Royal Society of London in 1724. I accurate in theory as they are simple and consis- From this period the thermometer became of scientent in their application.

tific utility, and its indications being founded upon Newton rejected spirit of wine as a thermometric fixed principles, gave the assurance that all instrufluid, as having too low a range, and selected lin- ments made with equal care would be strictly comseed oil as being capable of bearing unaltered in- parable. There is even now, after the lapse of a tense heat and cold. Like Boyle he computed the century, no prospect of materially improving the absolute expansion of the fluid in ten thousandths plan of the simple thermometer, or of finding any of its bulk, but he showed his judgment in not substance fitter than mercury for forming a scale making this the basis of practical graduation, of heat. In fact it unites the most important qualiwhich was always to be performed by means of two fications. fixed points, and dividing the interval into an arbi. Ist, It is found to measure with more accuracy trary number of degrees. He found that supposing than any fluid hitherto employed by equal spaces, the mass of oil at the temperature of melting ice equal increments and decrements of heat, which to be = 10000; at the heat of the human body it being the fundamental postulate in all thermometric was 10256, boiling water = 10725, and of melt- fluids, is of the most primary consequence. 2d, ing tin = 11516. As an empirical measure he di- The facility with which it is divested of air. wided the space between melting snow and the Though this might at first sight appear a trilling common temperature of the body into 12 parts, recommendation, it is found in practice one of the and proceeding from the data above given, he num- highest value; for the difficulty with which spirit

725 x 12 bered the boiling point = 34. The prin

of wine is freed from particles of air is one great 256

obstacle in forming good instruments with that ciple of his thermometer, and a scale of various fluid, the air deranging much its operations, and temperatures examined by it are to be found in a giving an independent degree of expansibility. Oil paper by Newton, printed anonymously in the 22d is still more faulty in this respect, but mercury, Vol. of the Philosophical Transactions in 1701, and from its immense cohesion of parts, the high temare also partly given in the Principia.

perature to which it may be brought, and the power The oil thermometer of Newton, though a happy of distilling it readily into a state of almost chemiapproach to a universal standard, had striking de- cal purity, has a decided superiority over every fects. The fluid, indeed, does not boil below 600° other substance yet applied to the purpose. 3d,

[blocks in formation]

Phil. Trans. vol. xxxiii.

[ocr errors]

The range of the mercurial scale is so great, and to see, that as the quantity of heat required to between such limits as to be extremely convenient raise mercury to any temperature is less than is refor all ordinary experiments; it does not boil under quisite for any other fluid, the acquisition of tema temperature of 660° of Fahrenheit according to perature will, cæteris paribus, be more quickly acDalton, or 656° according to Petit and Dulong, and complished. Add to this that mercury combines it does not freeze until the extreme cold of -- 39 of the rapid conducting power of a metal with conduc, the same scale. Deluc, from whom the principal tion by locomotion, which is peculiar to the fluid facts of the thermometric properties of mercury state, and which, communicating the temperature have been taken,* is much mistaken in asserting through all its parts, facilitates the change of conthat mercury bears a more intense cold than alco- dition. Dr. Trail found, (Nicholson's Journal, xii. hol. The experiments made about the middle of p. 137.) by a series of experiments, in which the the last century were very erroneous in regard to real conducting power of substances was alone inextreme degrees of cold, since as mercury contracts dicated, that mercury conducted a definite portion enormously at the moment of solidification, it was of heat in 15'' for which alcohol required 10' 45'' supposed that a temperature of — 261 of Deluc's making the conducting power as 43 to l. From scale, or 489 Fahrenheit, had been measured these circumstances a mercurial thermometer acwithout the freezing of mercury. On the other quires much sooner the temperature of any me. hand it was asserted,t that the French Academi. dium than one of alcohol; a fact which, though cians had found the spirits in Reaumur's thermo- long since ascertained, was a subject of great surmeter to freeze at 37 at Tornea in Lapland. prise and perplexity to philosophers. The deterThis proves nothing but the weakness of the spirits mination of specific heats being one of great nicety, employed, which indeed were always adulterated experimenters vary very much in their results, with water, as will be shown when we describe that but probably we are not far wrong if we consider thermometer. Pure alcohol has never yet been alcohol to possess double that of mercury. As a frozen by any cold, natural or artificial, though it satisfactory proof of the superiority of mercury in has been reduced as low as — 91 Fah. by Mr. this respect, we shall quote a specimen of Martine's Walker, of Oxford, and it has even been alleged experiments on the subject, which were among the that a temperature of - 150° has been produced. I first to throw light on the subject, taken from his Foi intense cold, therefore, such as was frequently little volume of Essays on Heat,ll a work which at experienced in Captain Parry's late voyages, spirit any period would have done honour to its author, thermometers are requisite, but within all ordinary and for the time at which it was written, certainly bounds, as we have already stated, mercury enjoys one of uncommon merit. The following experian ample and convenient range.

ment was made by placing two thermometers, 4th, Mercury accommodates itself more speedily one containing 31 ounces of water, the other near. to the temperature of the medium in which it is ly 48 of mercury (or equal volumes) before a large placed, than any other fluid, gases excepted. The fire. . theory of the heating and cooling of bodies is a

Water Mercury

Water Mercury very curious one, and till lately was very imperfecto

Times. heating. heating.

heating. heating. ly understood. From its great importance, a short O'


32' 83 963 outline of it may here find a place. Equal quanti

97 ties of heat thrown into different bodies do not ne

69 cessarily produce the same temperatures in each.


99 An equal bulk of mercury having a temperature of 16 68 841

91 100 100° and of water at 40° will not produce a mixture



100 having a temperature at a mean of these, or 70°, as

24 77 92

56 933 100 equal volumes of water would have done; the re

80 95

60 94 995 sulting temperature will only be 90°, showing that the 40° of heat lost by the mercury when transfer- Here we observe that the mercury reached the red to the water had only the power of raising it point of 925° in 24', for which the water required 20° of temperature. Speaking generally, therefore, 52', whence to find the relative capacity we say as mercury is capable of receiving only half as much 52 :1:: 24:0.46, or exactly the most modern deheat as water without being raised to a higher tem- duction. Mercury being thus compared to water perature, or to use the language of modern chemis- as a standard, the following experiment by the try, the specific caloric of water being 1, that of same author gives the relative value of spirit of mercury by equal volumes is 0.5.8 Now it is easy wine:









87 89






Modifications de l'Atmosphere, i. 285.

† Maupertuis, figure de la Terre, p. 58. We know not upon what authority this is stated in a late publication.

Henry's Chemistry, 10th edition, vol. i. p. 146. The difference between specific caloric and capacity for caloric, is sometimes ill explained, or not explained at all, in chemical works, which use both terms. The former expression is properly applied to substances taken by equal weights and the latter by equal volumes; yet they are sometimes interchanged. See Henry, p. 151; and we think that the latter term might very advantageously be done away with. See also Murray's Chemistry, i. 392, &c. To prevent confusion, it is proper to add, that the specific heat of French writers is identical with the absolute caloric of Crawfurd and others. Such abuses of nomenclature ought to be discarded as fast as possible from science.

| Page 77. The specific heat of mercury by volume is usually stated at .445, water being 1. That of alcohol varies exceedingly under the hands of different observers, some making it almost equal to water, which seems too great, and on the other hand, Mr. Leslie (Essay on Heat, p. 548) has certainly rated it too low in making it only .57.

« ՆախորդըՇարունակել »