THE 732 THE

end ed, is fixed a short piece of a tube of black glass, of sucli adiameter as to pass Ireely up and down within the small lube of thethermometer. The lower end, floating on the surface of the mer-cury, is carried up with it when it rises ; while the piece at theupper end, being of the same diameter, keeps the body or the in-dex parallel to the sides of the thermometiical tube. From thetipper end of the body of the index at 0, is drawn a spring of glassto the fineness of a hair, about -j of an inch in length, which beingset a little oblique, presses lightly against the inner surface of thetube, and prevents the index from following the mercury when itdescends, or being moved by the spirit passing up or down, orby any suddeiv motion given to the instrument by the hand orotherwise ; but at the same time the pressure is so adjusted, as topermit this index to be readily carried up by the surface of therising mercury, and downwards whenever the instrument is to berectified for observation. To prevent the spirit from evaporating,the tube at the end i is closely sealed. The common contrivancefor a self-registering thermometer, now sold in most of the Lon-don shops, consists simply of two thermometers, one mercurialand the other of alcohol, (fig. 6,) having their stems horizontal;the former has for its index a small bit of magnetical steel-wire,and the latter a minute thread of glass, having its two ends formedinto small knobs by fusion in the flame of a candle. The niag-lietical bit of wire lies in the vacant space of the mercurial ther-mometer, and is pushed forward by the mercury whenever thetemperature rises, and pushes that fluid against it; but when thetemperature falls, and the fluid retires, this index is left behind,and consequently shews the maximum. The oilier index, or bitof glass, lies in the tube of the spirit-thermometer immersed m thealcohol; and when the spirit relires by depression of temperature,the index is carried along with it in apparent contact with its in-terior surface ; but on increase of temperature tbe spirit gees torward, and leaves the index, which therefore shews the minimum oftemperature since it was set. As these indexes merely lie in thetubes, their resistance to motion is altogether inconsiderable. Thesteel-index is brought to the mercury by applying a magnet onthe outside of the tube, and the other is duly placed at tbe end otthe column of alcohol by inclining the whole instrument. In 1750,Lord Charles Cavendish presented to the lloyal Society of Lon-don , a thermometer in two different forms; the one contrivedto mark the greatest degree of heat, and the other the greatest de-gree of cold. The first consists of a glass-lube AB, (fig. 9,) witha cylindrical bulb B at tbe lower end, anil capillary at tbe top,over which there is fixed a glass-ball C. Tbe bulb and part ofthe tube are filled with mercury, the lop of which shews the de-grees of heat as usual. The upper part of the tube above tbemercury is filled with spirit of wine ; the ball C is also filled withthe same liquor almost to tbe top of tbe capillary tube. Whenthe mercury rises, the spirit of wine isaFo raised, and falls into theball C, which is so made that the liquor cannot return into thetube when the mercury sinks; consequently the height of thespirit of wine in the ball, added to that in tbe tube, will give thegreatest degree of heal to which the thermometer has pointedsince last observation. When a new observation is to be made,the instrument must be inclined till the liquor in the ball cover theend of the capil'ary tube. In this thermometer it is evident thatthe mercury must be affected by tbe weight and elasticity of thespirit ot wine, and therefore it will not correspond lo any ofthe common nvrcurial thermometers. The thermometer lorshewing the greatest degree of cold is represented in fig. 10, bytbe crooked tube A BCD. This instrument is filled with spirit ofwine, wnli the addition of as much mercury as is sufficient to fillboth legs oi the siphon, and about a fourth or filth part of thehollow ball C. We are not told what the proportion of mercurywas to that of spirit of wine. The degrees of heat are shew n bythe rise or fall of the mercury in the leg AB. The thermometermaiks the greatest fall by m-ans of the hollow ball C. When themercury m the longer leg sinks by cold, that in the shorter willrise and run over the ball C, from which it cannot return when I liemercury subsides in the shorter and rises in the longer leg. Theupper part of the she ter leg will therefore be filled with a columnof spirits ot a length pioportional lo the increase of heat; the bot-tom or lower surface oI wlnili, by means of a proper scale, willshew how much the mercury has been lower than it is: which beingsubtracted from the present height, will give the lowest point to

which (he mercury has fallen. That llie thermometer may befitted lor a new observation, the mercury must be made to runback from the ball into the short leg, by inclining the tube andheating the ball. It might at first sight be imagined that equal in-crements of heat would cause fluids to expand equably ; viz. thatif the heat is increased gradually by one degree, two degrees,three degrees, &c. the fluid tlm> heated would expand its bulk bya certain quantity, then by twice that quantity, three times thatquantity, and so on. But this is not the ease, and every fluidseems to follow' a particular law of expansion. Mercury seems toexpand more equably than any other fluid. Yet its incrementsof bulk are not quite proportional to the increments of heat. Withother fluids the irregularity of expansion is very considerable.On account of tbe great irregularity of the rate of expansion, mer-cury and spirit of wine are the only two fluids which can be usedfor thermometers ; observing that some compensation must bemade in the scale of the spirit-thermometer, in order to make itcorrespond with the scale of the mercurial thermometer. But themercurial thermometer cannot indicate a temperature higher than600. Hence various ingenious persons have endeavoured to con-trive instruments capable of indicating the higher degrees of heat,which would be of great use in philosophy, chemistry, and variousarts ; but the only useful contrivance of this sort was made by thelate Mr. Wedgewood. This ingenious gentleman applied to themeasuring of high degrees of heat, a singular property of argilla-ceous bodies, a property which obtains more or less in every kindof them, as far as has been examined. This property is, that anargillaceous substance, when exposed to fire, is diminished in bulkby it, imr does the bulk increase again after cooling; and this di-minution of bulk is proportionate to the degree ot heat to whichthe substance has been exposed. This property may seem to bea deviation from 1 he general rule, viz. that beat expands all sortsof bodies, and llial a diminution of heat enables them to contracttheir dimensions ; but in this case it must be considered that theclay-pieces contract, and remain contracted, because some sub-stance, viz. water and an aeriform fluid, is separated from them bythe action of the lire. Mr. Wedgewood’s thermometer, or appa-ratus for measuring tbe high degrees ot heat, consists of smallpieces of day ot a determined length, which are to be placed intlie furnace, crucible, &c. whose degree of beat is to be ascer-tained ; and of a guage to measure the eontracted dimensions ofthe clay-pieces, alter they have Deep exposed to the fire. Fig.7, represents the guage, which is either ot brass or of porcelain.Fig. 8, represents a section of (lie same; and the letters refer tothe like parts m both figures. F.FGI1 isa smooth flat plate; AC,BI), are two rulers or flat pieces, a quarter of an inch thick, andfixed fast upon the plate, so as to form a converging canal ABCD,whose width.at Cl) is Ihree-litihs of the width at AB. The wholelength of the canal from AB to CL), is divided into 240 equalparts, ami the divisions are numbered from the wider end. It isevident that if a body, so adjusted as to fit exactly the wider endof this canal, is afterwards diminished m its bulk by the action oflire (as the thermometrica) pieces which will be described in tbenext paragraph^ it may then be passed further in the canal, andmore so accord ng as tbe diminution is greater. The thermome-trical pieces are small cylinders of clay, a little flattened on oneside. They are nearly as much in diameter as they are in length.When one of these pieces is to be used, it is proper to measure itfirst by placing it in the guage al AB; for sometimes the piecesare a few degrees larger orsmallcr than the distance AB, which ex-u ss or defect being ascertained, must afterwards be allowed for.P represents one of these pieces set in the guage for measurement.The piece is then placed in the furnace, or crucible ; and if it istaken out either at tbe end of the operation, or at any period, and,when grown cold, is measured by sliding it as far as it will go intothe canal of the guage, the number of divisions against the platewhere it stops w ill shew the contracted dimensions of the piece, andof course the degree of heal to which it has b 'en exposed. It willbe found that these pieces will go very little beyond 0 in the canid*if (hey have been exposed to a visible red beat ; will go to 27° >*they have been exposed to the beat in which copper melts; toabout 00° if exposed lo the welding-beat ol iron ; about 160° if e *'posed lo the greatest heal that can be piodiued with charred pit'coal in a well-constructed common air-ftiinace, &c. The saw®thermometrical piece which has been used before, may be use«

agoi*