the difference of temperature between the inflowing and outflowing air. If the inflowing air be made to vary with the temperature to be measured, and the outflowing air be kept at a certain constant temperature, then the tension in the space or chamber between the two apertures will be an exact measure of the temperature of the inflowing air, and hence of the temperature to be measured.

In operation it is necessary that the air be sucked into it through the first minute aperture at the temperature to be measured, through the second aperture at a lower but constant temperature, and that the suction be of a constant tension. The first aperture is therefore located in the end of a platinum tube in the bulb of a porcelain tube over which the hot blast sweeps, or inserted into the pipe or chamber containing the gas whose temperature is to be ascertained.

The second aperture is located in a coupling, surrounded by boiling water, and the suction is obtained by an aspirator and regulated by a column of water of constant height.

The tension in the chamber between the apertures is indicated by a manometer.

The Air-thermometer. (Prof. R. C. Carpenter, Eng'g News, Jan. 5, 1893.)-Air is a perfect thermometric substance, and if a given mass of air be considered, the product of its pressure and volume divided by its absolute temperature is in every case constant. If the volume of air remain constant, the temperature will vary with the pressure; if the pressure remain constant the temperature will vary with the volume. As the former condition is more easily attained air-thermometers are usually constructed of constant volume, in which case the absolute temperature will vary with the pressure.

If we denote pressure by p and p', the corresponding absolute temper atures by T and T', we should have

The absolute temperature Tis to be considered in every case 460 higher than the thermometer-reading expressed in Fahrenheit degrees. From the form of the above equation, if the pressure p corresponding to a known absolute temperature T be known, T' can be found. The quotient T/p is a constant which may be used in all determinations with the instrument. The pressure on the instrument can be expressed in inches of mercury, and is evidently the atmospheric pressure bas shown by a barometer, plus or minus an additional amount h shown by a manometer attached to the air thermometer. That is, in general, p = b ±h.

The temperature of 32° F. is fixed as the point of melting ice, in which case T 460 +32=492° F. This temperature can be produced by surrounding the bulb in melting ice and leaving several minutes, so that the temperature of the confined air shall acquire that of the surrounding ice. When the air is at that temperature, note the reading of the attached manometer h, and that of a barometer; the sum will be the value of p corresponding to the absolute temperature of 492° F. The constant of the instrument, K = 492÷p, once obtained, can be used in all future determinations.

High Temperatures judged by Color.-The temperature of a body can be approximately judged by the experienced eye unaided, and M. Pouillet has constructed a table, which has been generally accepted, giving the colors and their corresponding temperature as below:

The results obtained, however, are unsatisfactory, as much depends on the susceptibility of the retina of the observer to light as well as the degree of illumination under which the observation is made.

A bright bar of iron, slowly heated in contact with air, assumes the following tints at annexed temperatures (Claudel):

148° F.

Carbonic acid..

Ice

Hyponitric acid.

Nitro-glycerine..

Tallow.

16

The boiling points of liquids increase as the pressure increases. The boiling point of water at any given pressure is the same as the temperature of saturated steam of the same pressure. (See Steam.)

MELTING-POINTS OF VARIOUS SUBSTANCES.

The following figures are given by Clark (on the authority of Pouillet, Claudel, and Wilson), except those marked *, which are given by Prof. Roberts-Austen in his description of the Le Chatelier pyrometer. These latter are probably the most reliable figures. Sulphurous acid

Alloy, 1 tin, 1 lead.. 370 to
Tin

466° F.

442 to

446

Phosphorus

Acetic acid.

For melting-point of fusible alloys, see Alloys. Cobalt, nickel, and manganese, fusible in highest heat of a forge. Tungsten and chromium, not fusible in forge, but soften and agglomerate. Platinum and iridium, fusible only before the oxyhydrogen blowpipe.

QUANTITATIVE MEASUREMENT OF HEAT.

Unit of Heat.-The British unit of heat, or British thermal unit (B. T. U.), is that quantity of heat which is required to raise the temperature of 1 lb. of pure water 1° Fahr., at or near 39o.1 F., the temperature of maximum density of water.

The French thermal unit, or calorie, is that quantity of heat which is required to raise the temperature of 1 kilogramme of pure water 1° Cent., at or about 4o C., which is equivalent to 39°.1 F.

1 French calorie = 3.968 British thermal units; 1 B. T. U. = .252 calorie. The "pound calorie" is sometimes used by English writers; it is the quan

tity of heat required to raise the temperature of 1 lb. of water 1° C. 1 lb. calorie 9/5 B.T.U.0.4536 calorie. = The heat of combustion of carbon, to CO2, 18 said to be 8080 calories. This figure is used either for French calories or for pound calories, as it is the number of pounds of water that can be raised 1° C. by the complete combustion of 1 lb. of carbon, or the number of kilogrammes of water that can be raised 1° C. by the combustion of 1 kilo. of carbon; assumiug in each case that all the heat generated is transferred to the water.

The Mechanical Equivalent of Heat is the number of footpounds of mechanical energy equivalent to one British thermal unit, heat and mechanical energy being mutually convertible. Joule's experiments, 1843-50, gave the figure 772, which is known as Joule's equivalent. More recent experiments by Prof. Rowland (Proc. Am. Acad. Arts and Sciences, 1880; see also Wood's Thermodynamics) give higher figures, and the most probable average is now considered to be 778.

1 heat-unit is equivalent to 778 ft.-lbs. of energy. 1 ft. lb. = 1/778.0012852 heat-units. 1 horse-power = 33,000 ft.-lbs. per minute = 2545 heat-uuits per hour 42.416+ per minute = .70694 per second. 1 lb. carbon burned to CO, 14,544 heat-units. 1 lb. C. per H.P. per hour =254514544 = 17% efficiency (.174986).

Heat of Combustion of Various Substances in Oxygen.

In burning 1 pound of hydrogen with 8 pounds of oxygen to form 9 pounds of water, the units of heat evolved are 62,032 (Favre and S.); but if the resulting product is not cooled to the initial temperature of the gases, part of the heat is rendered latent in the steam. The total heat of 1 lb. of steam at 212° F. is 1146.1 heat-units above that of water at 32°, and 9 X 1146 110,315 heat-units, which deducted from 62,032 gives 51,717 as the heat evolved by the combustion of 1 lb. of hydrogen and 8 lbs. of oxygen at 32° F. to form steam at 212° F.

By the decomposition of a chemical compound as much heat is absorbed or rendered latent as was evolved when the compound was formed. If 1 lb. of carbon is burned to CO2, generating 14,544 B.T.U., and the CO, thus formed is immediately reduced to CO in the presence of glowing carbon, by the reaction CO, + C = 2CO, the result is the same as if the 2 lbs. C had been burned directly to 2CO, generating 2 X 44518902 heat-units; consequently 14,54489025642 heat-units have disappeared or become latent, and the

"unburning" of CO, to CO is thus a cooling operation. (For heats of combustion of various fuels, see Fuel.)

SPECIFIC HEAT.

Thermal Capacity. The thermal capacity of a body is the quantity of heat required to raise its temperature one degree. The ratio of the heat required to raise the temperature of a certain weight of a given substance one degree to that required to raise the temperature of the same weight of water one degree from the temperature of maximum density 39.1 is commonly called the specific heat of the substance. Some writers object to the term as being an inaccurate use of the words "specific " and "heat." A more correct name would be "coefficient of thermal capacity

Determination of Specific Heat.-Method by Mixture.-The body whose specific heat is to be determined is raised to a known temperature, and is then immersed in a mass of liquid of which the weight, specific heat, and temperature are known. When both the body and the liquid have attained the same temperature, this is carefully ascertained.

Now the quantity of heat lost by the body is the same as the quantity of heat absorbed by the liquid.

Let c, w, and t be the specific heat, weight, and temperature of the hot body, and c', w', and t' of the liquid. Let T be the temperature the mix

ture assumes.

Then, by the definition of specific heat, c × w × (t − T) = heat-units lost by the hot body, and c' × w' × (T- t') = heat-units gained by the cold liquid. If there is no heat lost by radiation or conduction, these must be equal, and

c'w' (T-t')

w(t-T)

Specific Heats of Various Substances.

The specific heats of substances, as given by different authorities, show considerable lack of agreement, especially in the case of gases.

The following tables give the mean specific heats of the substances named according to Regnault. (From Rontgen's Thermodynamics, p. 134.) These specific heats are average values, taken at temperatures which usually come under observation in technical application. The actual specific heats of all substances, in the solid or liquid state, increase slowly as the body expands or as the temperature rises. It is probable that the specific heat of a body when liquid is greater than when solid. For many bodies this has been verified by experiment.

In addition to the above, the [following are given by other authorities. (Selected from various sources.)

METALS.

Light carburetted hydrogen, marsh gas (CH4). .5929

Sulphurous acid..

Blast-furnace gases....

.2277

Specific Heat of Salt Solution. (Schuller.)

Volume.

.1246

.4683

.8909 .8606 .8490 .8073

Specific Heat of Air.-Regnault gives for the mean value

Between -30° C. and + 10° C...

0.23771

Hanssen uses 0.1686 for the specific heat of air at constant volume. The value of this constant has never been found to any degree of accuracy by direct experiment. Prof. Wood gives 0.2375 ÷ 1.406 0.1689. The ratio of