THERMOMETER. 



lower ball, ;!nd confeqiiently its furface will defcend ; on 

 the contrary, as the ambient air becomes colder, that in the 

 ball is condenfcd, and the liquor prefTcd by the weight of 

 the atmofphcre will afcend : fo that the liquor in the tube 

 will afceiid or defcend more or lefs, accordinnj to the (late 

 of the air contiguous to the inftrument. To the tube is 

 affixed a fcale of the fame length, divided upwards and 

 downwards from the middle, C, into one hundred equal parts, 

 by means of which the afcent and defcent of the liquor in 

 the tube, and confequently the variations in the cold or heat 

 of the atmofphere, may be obfervcd. 



It muft be acknowledged, that the expanfion of elaftic 

 fluids affords, in fome cafes, a teft of heat, which is very 

 convenient from its great delicacy, and becaufe a very fmali 

 quantity of heat is fufficient to raife their temperature very 

 confiderably. 



A fimilar thermometer may be conftrufted by putting a 

 fmall quantity of mercury, not exceeding the bulk of a pea, 

 into the tube B C {fg. 2.) thus bent in wreaths, that taking 

 up the left height, it may be the more manageable and lefs 

 liable to harm ; divide this tube into any number of equal parts 

 to ferve for a fcale. 



Here the approaches of the mercury tovrards the ball. A, 

 will ftiew the increafe of the degree of heat. The reafon is 

 the fame as in the former. 



The defeft of both thefe inftruments confifts in this, that 

 they are liable to be afted on by a double caufe : for, not 

 only a decreafe of heat, but alfo an increafe of weight of the 

 atmofphere, will make the liquor rife in the one, and the 

 mercury in the other ; and, on the contrary, either an in- 

 creafe of heat, or decreafe of weight of the atmofphere, will 

 make it defcend. 



In winter, for example, the liquor would rife and fink too 

 much ; for a frofl condenfing the internal air, the liquor 

 would afcend, but as the air is heavier in frofty weather, its 

 prefTure on the liquor in the veffel D E {^g. 1.) being in- 

 creafed, would raife the liquor ftiU higher in the tube, and 

 thus indicate a degree of cold greater than it really is. On 

 the other hand, if the weather grows warm, as it does in 

 rainy weather in winter, the air in the ball vnll expand, and 

 the liquor defcend in the tube ; but as the weight of the at- 

 molphere is lefs in foul weather, the liquor in D E will be lefs 

 preffed than it was, and fuffer the liquor to defcend more than 

 it fhould do, and (hew a greater degree of warmth than that 

 of the ambient air. The reverfe of this will happen in fum- 

 mer : for warm weather being fair weather, and the atmo- 

 fphere being then heavier than ufual, the liquor will be made 

 to (land higher in the tube than it (hould do, and (hew 

 the degree of heat to be lefs than it really is. And as in 

 fummer, the weather becomes cold with rain ; but the 

 weight of the atmofphere being diminifhed, the liquor will 

 rot afcend fo far as it ought to afcend by the condenfation 

 of the internal air, and therefore indicate the cold to be lefs 

 than it really is : and when the two caufes, thus contributing 

 to the rife and fall of the liquor, aft equally in oppofite di- 

 reftions, the liquor would appear neither to afcend nor de- 

 fcend, whatever might be the changes in the temperature of 

 the atmofphere, on account of equal correfponding varia- 

 tions in its gravity. Befides, the air in the ball, &c. is liable 

 to be afFefted more or lefs in its elaftic quality by the vapours 

 that detach themfelves from the included liquor according to 

 the degree in vrhich it is heated or cooled. For thefe and 

 other meafures, thermometers of this kind have been long 

 difufed. 



Inftruments of this kind, when they are fubjeft to the 

 variations of the prefTure of the atmofphere, as well as to 

 thofe of its temperature, are properly called manometers, 



and require, for enabliiij; us to employ thcin as ihfrmomc- 

 ters, a comparifon with the barometer ; while, on the other 

 hand, thi-y may be ufed as barometers if the temperature be 

 otherwife afccrtaincd. They are, however, very ufeful 

 without this comparifon, in delicate experiments of (liort 

 duration ; befides, the changes of the barometer are feldom 

 very rapid, and they may alfo be wholly freed from the 

 cffefts of the prefTure of the atmofphcre in various ways. 



Bernouilli's method confifts in clofing the tube of a com- 

 mon barometer foas to leave the column of mercury in equi- 

 librium with the air contained in the bulb at its aftual tem- 

 perature, and capable of indicating, by the changes of it's 

 height and of its prefTure, any fubfcquent changes in the 

 temperature of the air, which muft affeft both its bulk and 

 its elafticity. (See_y5io'. 3.) Mr. Leflie's photometer, or 

 Z)!^ri'n/;fl/ Thermometer (which fee), has fome advantages 

 which render it better than this inftrument ; but it can only 

 be employed when the changes of the temperature can be 

 confined only to a part of the inftrument. The elafticity of 

 the air contained in the bulb is here counterafted, not by the 

 prefTure of a column of mercury, but by the elafticity of 

 another portion of air in a fecond bulb, which is not to be 

 expofed to the heat or cold that is to bfe examined ; and the 

 difference of the temperatures of the two bulbs is indicated 

 by the place of a drop of a hquid, moving freely in the 

 tube which joins them. 



M. Amontons, in 1 702, with a view of perfefting the 

 aerial thermometer, contrived his univerfal thermometer. 

 Finding that the changes produced by heat and cold in the 

 bulk of the air were fubjeft to invincible irregularities, he 

 fubftituted for thefe the variations produced by heat in the 

 elaftic force of this fluid. This thermometer confifted of a 

 long tube of glafs ( ke^g. 4. ) open at one end, and recurved 

 at the other end, which terminated in a ball. A certain 

 quantity of air was comprefTed into this ball by the weight 

 of a column of mercury, and alfo by the weight of the at- 

 mofphere. The effeft of heat on this included air was to 

 make it fuftain a greater or lefs weight ; and this effeft was 

 meafured by the variation of the column of merciu-y in the 

 tube, corrcftcd by that of the barometer, with refpeft to 

 the changes of the weight of the external air. Tiiis inftru- 

 ment, though much more pcrfeft than thofe in the room of 

 which it was fubftituted by its inventor, is neverthelefs fub- 

 jeft to very confiderable dcfefts and inconveniences. Its 

 length of four feet renders it unfit for a variety ot experi- 

 ments, and its conftruftion is difficult and complex : it is 

 extremely inconvenient for carriage, as a very fmall inclina- 

 tion of the tube would fuffer the included air to efcape : and 

 the friftion of the mercury in the tube, and the comprcffi- 

 bihty of the air, contribute to render the indications of this 

 inftrument extremely uncertain. Befides, the dilatation of the 

 air is not fo regularly proportional to its heat, nor is its dila- 

 tation by a given heat nearly fo uniform as he fuppufed. This 

 depends, as the abbfe Nollet has fuggefted, much on its 

 moifture ; for dry air does not expand near fo much by a 

 given heat, as air ftored with watery particles ; which by 

 being converted into fteam, very much nicreafe the feeming 

 volume of the air. For thefe and other reafons enumerated 

 by M. de Luc, (Recherches fur les Mod. de I'Atm. tom. i. 

 p. 278, &c.) this inftrument was imitated by very few, and 

 never came to be of general ufe. 



ConJ{ru8ion of the Florentine Thermometer. — The acade- 

 mifts del Cimento, about the middle of the feventeenth cen- 

 tury, confidering the inconveniences of the air-thermome- 

 ters above defcribed, attempted another, that fhould mca- 

 fure heat and cold by the rarefaftion and condenfation of 

 fpirit of wine ; though much lefs than thofe of air, and con- 

 fequently 



