Count Rumford (Essay XVI.) details a number of experiments on the comparative economy of different sized flames, produced from various burners applied to an argand oil lamp, and from which it appears that there is a certain varying proportion betwixt the diameter of each burner, the intensity of the light, and the quantity of oil consumed ; the consumption of oil being greatest in proportion to the light ob. tained when the flame was small, and gradually les sened as the light was enlarged, until it began to smoke, when the proportional consumption of oil again increased. With an argand burner of about the inch diameter at the middle of the flame, he tried the proportional consumption of oil when the light was regulated, so as to be equal to that of a successive number of candles, of from 1 to 10 ; the result, in grains of oil, which gave an equal and cor responding quantity of light for one hour, are given in the fiillowing table, and the quantity necessary to supply the light of one candle in each case is also added.
From this table it appears, that the lamp here used burned to the greatest advantage when made to give a light equal to eight or nine of the wax can dles ; and from other experiments of the like nature, the proportions above hinted at may be taken as fol lows : An oil argand burner one-third inch diameter, is most advantageously employed when giving a light equal to about . 3 wax candles.
One 7.10th (being that above parti cularized), about . 9 Do.
It was to be expected that something similar to the above would occur with regard to the combustion of gas, and this we have seen fully established by expe riments made in 1807 and later years, and without any knowledge of what Count Rumford had done.
An argand gas burner, about three-fourths of an inch diameter, when regulated so as to give light equal to one mould candle of six to the pound, con sumed Cubic feet of Gas, Cubic feet per Candle, per hour 1.43 being 1.43 when equal to 1.96 0.49 4 candles, S 6 do. 2.40 0.40 8 do. 2.95 0.37 10 do. 3.10 0.31In other experiments the consumption did not ex. ceed one-fourth of a cubic foot per hour for each candle, when the flames were as large as the burners would admit of without producing smoke.
As the general plan of re-lighting establishments will be affected by local circumstances which cannot here be considered, the disposition and arrangement the different parts is a subject on which much cannot be said. Plate LXXXIV. exhibits a plan and elevation which, where the situation will admit, will be fought as convenient as ant The retorts, b 6, are placed round a conical chimney, a a, into which all their flues enter, and in its lower part the ashes and cin ders are collected from the different fires, and taken out from time to time by an entrance from the cir cular tunnel, f f. This tunnel is furnished with apertures above, corresponding to the number of the retorts, and through which the coke is suffered to descend when discharged. The retorts are covered
with a conical roof, under which are a number of small 'openings to take off any dust, smoke, &c. which arises ; exterior to this roof (which covers the part c, where the retorts are charged and discharged) is another over e, less elevated, leaving a circular space d, to admit light ; under this may be deposited coals, &c.
The gas and other products of distillation are con veyed, by means of the pipe g, into the tar reservoir p p ; over which the condenser h, and purifier i are placed, and from this last the gas passes by the pipe k, to the gasometers m m, through the tubes 11 (in the manner represented in Plate LXXXIII. fig. 6), and from these again at n, for distribution and combustion when required.
The apparatus here delineated (on a scale of 20 feet to the inch) contains twenty-one retorts of the mensions shown in Plate LXXXII. fig. 8 ; two gaso. meters are represented together capable of contain. ing 50,000 cubic feet of gas; and in case of repairs, &c. being wanted, it will, in general, be found more convenient to have, instead of one gasometer of the full capacity required, two or more of a lesser size connected with the apparatus; and it may frequently happen that these can be placed at a distance from each other, particularly in lighting a town of any considerable magnitude, where such distribution will be attended with the advantages of more perfectly equalizing the pressure'on the gas, and of lessening materially the size of the mains necessary for ance, as_these distant gasometers can be filled during the period when light is not required, and the ing mains answer the purpose of conveying the gas • partly back again for combustion.
Preparation of Oil-Gas.
In Nicholson's Journal for 1805, Dr Henry has given an account of his experiments on the gases ob tained by the destructive distillation of oil and other substances; but no apparatus for conveniently effect ing the decomposition of oil on a large scale appears to have been constructed for many years afterwards. In 1815 Mr John Taylor obtained a patent for a mode of producing gas from bones and other animal mat ters; and the principle of action of the oil-gas ma chines, now manufactured by him, will be under stood from fig- 9, Plate LXXXIII.: a a a is a metallic or other tube placed in a furnace capable keeping it red hot ; 6, the fire place; c, a small close box or cistern containing oil ; e, a pipe leading from this box to one end of the tube a, and having a re gulating cock thereon ; f, another pipe joining the opposite end of the fire-tube, and also joining the oil-cistern at its upper side ; g, a pipe to convey a way the gas when formed, and d, another pipe with funnel attached for admitting a supply of oil from time to time.