294 Anah/ses of Books. [April, 



Every body knows that cfial, when decomposed slowly, affords a larger quantity 

 of tar and amn-.oniac;il liqnor, but a less quantity of gas tlian when decomposed 

 rapidly. 



In ilie former case, the formation of the proximate products which coal is capa- 

 ble of furnishing is eftectrd properly ; the bituminous part of tlie cual is developed 

 under tlie most favourable ci; cnmstance?. 



But when coal, after beino; previously deprived of moisture, i= very suddenly 

 heated to a hiijh temperature, in thin sUala. and small portions at a lime, so that 

 the vnpoarous producls, instrad of becomiiij condensed, are made to come into 

 contact with a substance (w liichin this case is the roof of the retort i ke|)t constantly 

 at a temperature rather liijcher than tliat at which };old, silver, and copper melts 

 (SS*' AVedjewood, or 5237° Fahrenheit), a very different arrangement of princi- 

 ples takes pl.!ce. 



The greatest portion of tar which the coal is capable of furnishing, instead of 

 being produced in a liquid form, becomes then decompose I into carbureded hydro- 

 gen and oleCant gas. That portion of tar which escapes decomposition is 

 condensed in the perpendicular pipe, H, fig. '2, and falls back again into the 

 retort, where it is also deroiaj)Osed when the coal upon which it falls comes under 

 the process of decoinposiliot). 



Hence the quantity of tar obtained by means of horizontal rotary retorts is very 

 small ; it seldom exceeds the jiroportion before -!nenlioi;ed, when the retort is 

 •worked to the greatest advantage. This quantity is considerably diminished, when 

 ?fev\ castle coal, broken into pieces of the size of split pease, is decomposed in 

 strata, not exceeding two inches in thickness. The quantity of tar aftbrded by a 

 chaldron of coal then amounts to 30 lbs. ; while, at liie same time, the quality of 

 the gas is improved ; bec.iuse coal tar furnishes olcfiant gas, which the coal alone, 

 ■when distilled by means of cylindrical or other sh:iped cast-iron retorts of the usual 

 form, cannot produce, or at lea«t but in a small quantity. One gallon of coal tar 

 yields 15 cubic feet of olefiant gas, which greatly increases the illuminating power 

 «f the carburetted hydrogen. 



From what has been so far stated, it \vill be understood why one chaldron of 

 ZVcwcaslle coal, when decomposed by the new process, may readily be marie to 

 produce from 15,t)00 to 18,000 cubic feet of gas and upwards, whereas the same 

 qnantity of coal, if decomposed by the old melhod, yields only upon an average 

 10,000 cubic feet of gas. 



In the former case, the greater part of the essential nil and tar which the coal 

 'would have aflordcd is decomposed, as stated already, by virtiie of the high temper- 

 ature to wliicii the' vapourous tar is suddenly exposed in the horizontal rotary 

 retort, which is not the case when coal is decolnpo^ccl in the retorts of th^' olil con- 

 struction. 



Gninin the Quar.tilil of Cofcc. — \Viih the cylindrical or cast-iron retorts of the old 

 shapes, the qnantily of coke obtained from a given quantity of co.'il is upcm .an ave- 

 rage V.i ])er cent, increase by ineasure from Ilie best kind of Newcastle and Sun- 

 derland coa'l; but taking into account the waste incurred in breaking out and 

 removing the red-hot coke from the retort, which requires Iheapplication of rakers 

 and crow bars, a considerable portion of it becomes reduced to dust or breeze, and 

 hence no more than bulk for bulk of the coal decomposed can seldom be dejiended 

 upon as the nllimale snlrable quantity of coke. 



In the new mode of carbonizing coal by means of the horizontal rotary retorts, 

 the increase ofeokeis I.iO percent, by measure; so that one chaldrmi of Newcastle 

 coal produces two and a half chaldron of coke — this is the quantity i>rodiiccd upon 

 an avei;ise. Hut when the report is worked at a temperature to produce at the 

 rate of Iv.orin cubic feel of gas fn in the chaldron of coal, the increase of coke by 

 measure is 175 percent.; in rliat case, the lavermf coal in the coal-boxes must not 

 exceed t\\o inches in thickness; so that the volume of coke is in the ratio of the 

 <liiantity of ^as produced, and the rapidity and elevation of temperature at which 

 the decomposition of the coal is effected. 



The coke beinj wiilidraun from the place wliere it is formed by merely turning 

 the boxes containing it, up ide down, all waste is avoided. 



With respect again to the quality of the cok?, it h ill be observed that when the 

 coal is rapioly carbonized in thin l;iyers, and iias full liberty to expand freely, as 

 in the case ef the horizontal rotary retort, it affords a light ami porous coke, 

 ■whereas in the eylii;drical, parallelopipedr.l, semi-cylindrical, or ellipsoidal 

 retorts, the coke being compressed, the intense heat to which it is so lougand super- 

 iSuously exposed, renderi it extremely dense, ar.d of a stony hardness. 



