CAL

CAL

( iso )

-of metal were, made, the upper half of Iron ,and the lower of copper, and heat applied frombeneath, no expanfion of the lower portion couldcaufe a mixture of the two metals, or a paflageof one through the other. But quite the reverfehappens when a liquid or gafeous body is par-tially heated ; for being compofed of fluid par-ticles of great mobility, thole that receive thefirit imprellion of heat and become expandedby it, acquire thereby a comparative levity, theirfpeciiic gravity being diminifhed, and, if at thelower part of the mafs, they immediately rifethrough the cooler portion to fhe furface, givingoff in their paflage more or lefs of flieir acquiredheat to the particles of the cooler fluid withwhich they fucceffively come in contact.

This eflential difference in the mode of com-munication of heat renders it neceffary to con-fider feparately its paflage through folids ; andit would be even convenient in philofophicallanguage to diitinguifh thefe very different mo-tions of caloric by appropriate terms.

Of the Propagation of Heat in Solids.

When a folid body (a bar of metal for exam-ple) is expofed to the influence of any fource ofheat, this laid pervades the metal, not with theimmeafurable velocity of radiant heat, but witha flow and gradual progrefs, from the neareftto the mod remote part; and if the heat bebelow fufion, producing no other apparent effectthan expanfion, and a fomewhat different degreeof cohefion of the partieles of the metal. Asradiant heat may be confidered as caloric in itsdate of the greated freedom, and the moduncombined, and as then it is found to movewith a velocity utterly incalculable by any ex-periment, it mud follow that the extreme re-tardation of its motion in palling through a folidis owing to a fucceflive combination with allthe particles or drata of particles which oppofeits progrefs. Hence the particles of folid mattermult have an affinity for caloric to be able todetain it for (comparatively) fo great a lengthof time ; and as this time is in no degree regu-lated by the fpeciiic gravity or denlity of thefolid, that is by the fuppofed number of integrantparticles of which it is compofed, it feems evi-dent that each fubftance has its peculiar andrelative force of affinity with caloric. This forc.eof affinity is obvioufly in direct oppolition to its.conducing power, for the greater force it exertsin detaining caloric, the lefs eafily will it con-duct, or allow the paflage of caloric through itsfubdance. Hence too the terms good or badconductors ; the former term being applied to

thofe bodies that have the weaked affinity forheat; the latter, to the dronged.

The paflage of conducted heat through abody has its. limits. Owing, to the affinity be-tween the two, a certain portion of the caloricis detained within the fubdance heated, and theexpanfion produced by the combination feems.to affid in this detention. We in fa£t find thatif one end of a very long bar be heated mo-derately and for a ffiort time only, the heat isfenfibly conducted only to a fmall didance, andin a condantly diminilhing intenfity. There isno good reafon however to fuppofe that any ofthe acquired caloric is permanently detained inthe bar, after the heated end has returned to thecommon temperature and the whole has con-tracted to its original dimenfions. The intenfityof conduced heat, and the didance to which itmay be propagated, are alfo determined by thetemperature at which the fubdance begins tomelt, confume, or to undergo any other chemi-cal change ; for when this occurs, all the furplusheat above this point is employed in the chemi-cal changes that are going on, and does notmaterially add to the quantity that paffes off bythe conducting power. Therefore (other thingsbeing the fame) an indefinite intenfity of heatwill pafs to a greater didance through a lefsthan a more fufible metal, and further throughan incombuftible than a combuftible fubftance.

Though caloric is capable of palling throughfolids in every direction, it appears that it morereadily proceeds upwards than downwards. Theexperiments of M. Pi£tet confirm this opinion,which has long been entertained. The appa-ratus employed by this very able experimenterwas a glafs tube about 4 feet long, containing acylindrical bar of copper, hollowed a little attop and bottom, to receive the bulbs of two finethermometers. The bar and tube were fuf-pended vertically, and the bar was heated ex-actly in the middle by the fun’s rays thrownupon it, and the exact fpot was fecured byenclofing the whole apparatus in a woodenframe, perforated at the place on which the heatwas to fall. Moreover the glafs was exhaufted ofits air, to a trifling refidue, by a good air-pump,otherwife the mere afcent of the heated air incontact with the copper bar would certainlyheat the upper part fooner than the lower, anddefeat the object of the experiment. On heat-ing the middle of the bar and obferving thetimes required to raife the touching thermome-ters to the fame heights, it was found that the'upper one was confiderably fooner heated than