THERMOMETER. 



r«(UM«lj the »UeratijJi4 ia ttie degree of heat likely to be 

 muili loU Ifiifiblo. 



The fpirit of wine was enclofed in glafs tubes, liermctically 

 fe»le<l ; fo that llufc thcrnwiiietcrs could be fubjed to no 

 inconveuience by the evaporation of the liquor, or the va- 

 riable gravity of the incumbent atmotpherc. Inllrumciils 

 of thii kind were firll introduced into England by Mr. 

 Boyle, and they were fooii uiiiverfally ufed among philofo- 

 pliers in other countries. The Florentine thermometer con- 

 litts of a fmall narrow tube BCD {fg. 5.) conneaed with 

 a glafs ball A. The tube (hould be procured as cylmdric 

 as poflible : and it may be tried, by putting into one end of 

 it as much mercury as will till the length of one incii, and 

 letting this quantity of mercury pafs from one part of the 

 tube to another, through, its whole length ; meafure with 

 compafTes the length it occupies in every part of the tube ; 

 aiid if it every where takes up an inch, the tube is cylindric, 

 and a fcale of equal divifions will agree with it : otherwife it 

 vrill be longer where the tube is fmaller, and (horter than an 

 inch where the tube is larger ; and in this cafe, the divifions 

 mull be fuited to the contents of the bore. The glafs ball 

 may then be joined to the tube, and a fmall cavity be made 

 at the other end. Fill the ball and tube with reftified fpirit 

 of wine to a convenient height, as to C, when the weather 

 is of a mean temperature, which may be done, by inverting 

 the tube into a vefTel of llagnant coloured fpirit, under a re- 

 ceiver of the air-pump, or by many other ways. The 

 fpirit may be colonred by pouring a quantity of it on fmall 

 pieces of turmeric, which will hereby receive a red tinfture ; 

 and the fpirit may be repeatedly filtrated through browTi 

 paper, in order to feparate from it the eoarfer particles of the 

 root. Some perfons, in filling the ball and tube, for pre- 

 venting the fpirit from wholly defcending into the ball in 

 winter, recommend putting the ball into a lump of fnow, 

 mixed with fait ; or if the inftrument be made in fummer, 

 into fpring-water impregnated with falt-petre, that the con- 

 denfed fpirit may fhew liow far it will retire in extreme cold. 

 Jf it riles to too great a height from the ball, part of it 

 is to be taken out ; and that the tube may not be made 

 longer than necefiary, it is convenient to iramerge the ball, 

 filled with its fpirit in boiling water, and to mark the fartheil 

 point to which it then rifes. When the thermometer is pro- 

 perly tilled, with a lamp heat the little bubble left at the 

 end of D red-hot, and feal it hermetically, leaving, as Dr. 

 Defaguhers recommends, in the thermometer only the third 

 part of the air that was in it, which will give room to the 

 dilatation of the fpirit ; and this rarefied air will prevent the 

 air left in the fpirit, even after the air-pump has been applied, 

 from dividing the fpirit by its expanfion. To the tube apply 

 a fcale, divided into one hundred equal parts, from C 

 towards D, and alfo from C towards B. 



Now, fpirit of wine rarefying and condenfing very confi- 

 derably ; as the heat of the ambient air increafes, the fpirit 

 will dilate, and ceiifequently will afcend in the tube ; and as 

 the heat decreafes, the fpirit will defcend : and the degree 

 or quantity of afcent and dcfcent will be feen in the fcale. 

 Yet as the ratio of yefterday's heat to to-day's is not hereby 

 difcovered, this inftrument is not ftriftly a thermometer, any 

 more than the former. 



It is to be here obferved, i. That as the natural gravity 

 of the liquor makes it tend downwards, fo it refills its afcent 

 ont of the ball into the tube ; and that the more, as it rifes 

 lugher ; for which reafon, fome have advifed to have the 

 tube horizontal. 



2. Since there mull of neceffity be fome air left in the 

 void part of the tube, over the hquor, that air, by its elaf- 

 ticJty, will tend downwards, and of confcquence will rcCft 



the rife of the hquor, and be compreffed by it at it doM 

 rife : its clafticity therefore is thus increafed. 



3. Since it is found from cxpei-iencc, that a lefs degree of 

 heat is comniunicatod more eafily to the fpirit of wine in the 

 ball than a greater, tlie rarefaftions of the fpirit of wine are 

 not proportionable to their producing caufes ; efpecially fincc 

 a greater degree of heat finds more liquor in the tube than a 

 lefs does, to which, notwithilanding, the heat may be more 

 eafily communicated tiian to that ftagnating in the ball. 



4. Spirit of wine is incapable of bearing very great heat 

 or very great cold. It boils fooner than any other liquor, 

 and, therefore, the degrees of heat of boiling fluids cannot 

 be determined by this thermometer. And though it retains 

 its fluidity in pretty fevere cold, yet it feems not to condenfe 

 very regularly in them : and at Torneo, near the polar 

 circle, the winter cold was fo fevere, as Maupertuis informs 

 us, that the fpirits were frozen in all their thermometers. 

 So that the latitude of heat and cold, which fpirit of wine 

 is capable of indicating, is much too limited to be of very 

 great or univerfal ufe. On thefe accounts, the Florentine 

 thermometer, though it has been much ufed, is far from 

 being an accurate meafure of heat, &c. to which may be 

 added what Dr. Halley obferves in the PhilofopliicalTranf- 

 a<ftions, that he has learned from thofe who have kept fpirit 

 of wine long, that it always lofes part of its expanfive force 

 in courfe of time. 



This objeftion, fuggefted by Dr. Halley, and often in- 

 filled on by others, has, according to Dr. Martine, no great 

 weight. Well reftified fpirit of wine, if fealed up in a glafs, 

 is in a confiderable degree unalterable. It cannot evaporate ; 

 and by many years experience its force of expanfion has con- 

 tinued the fame ; as, befide other obfervations, we know 

 efpecially from the Annual Regifters of M. de la Hire's 

 fpirit thermometer, that have been kept in the Obfervatory 

 for many years. 



Another great defeft of thefe, and other thermometers, is, 

 that their degrees are not comparable with each other. They 

 mark, indeed, the different degrees of heat and cold ; but 

 each marks only for itfelf, and after its own manner ; be- 

 caufe they do not proceed from any point of heat, or cold, 

 that is common to them all. It is with them as with two 

 clocks, which for want of having been firfl: fet to the fame 

 hour by the fun, will, indeed, mark that one, two, or more 

 hours are paffed, but not what hour it is by the day. Nor 

 can we be aflTured, that when the liquor is rifen a degree in 

 two different thermometers, they have both fufFered the fame 

 imprcfTion of an equal additional heat : fince the fpirit of 

 wine may not be the fame in both ; and, in proportion as the 

 fpiiit is more or lefs reftified, it will rife more or lefs high by 

 the fame heat. Nor is this all ; for in graduating thermo- 

 meters, they often take equal lengths of the tube for equal 

 afcents of the fpirit : whereas, fuppofing the diameters of 

 the tube equal throughout, which very rarely happens, there 

 are fo many irregularities withinfide, that a certain length 

 of tube fometimea. requires double the quantity of liquor to 

 fill it, that the fame length in another tube of the fame dia- 

 meter requires. All which arifes from the unequal thickneffes 

 of the parietes of tubes in different places ; and from acci- 

 dental prominences and cavities, always found in the inner 

 furfaces of tubes ; and efpecially from their being always 

 bigger at one end than the other. 



Befides, the divifions of the fcale cannot accurately indi- 

 cate the quantity of rarefaftion, unlefs the proportion of the 

 cavity of the tube D B to that of the ball A were known. 

 Hence it is, that the comparifon of thermometers becomes 

 fo precarious and defeftive. Yet the moft curious and inte- 

 refting ufe of thermometers is, what ought to arife from 



fuch 



