REPORT UPON THE ACTION OF A TUBULAR BOILER. 263 



teracting 5° of cold, the apparatus was found in such a case unfit ; 

 for a gentle fire was so damped by being chiefly enclosed by metal 

 kept comparatively cool by the circulation of water within the 

 tubes, that it soon went out. This is a defect which can never 

 be completely obviated whilst the fuel is in contact with a rapid 

 conductor of heat. I have seen a quantity of fuel put in, and 

 well ignited, which ought to have continued burning till seven 

 o'clock next morning, but which did not keep in till four a.m., 

 the time when heat is usually most wanted. For my own part, 

 knowing from experience that there is no difficulty in making 

 up a fire to keep in till morning in a furnace lined witli fire-brick 

 or similar non-conducting materials, I should never recommend 

 metal enclosures, wherein a prolonged combustion can never be 

 insured, unless a stoker is afforded throughout the night. 



The apparatus in question is also objectionable in regard to 

 economy of fuel. It was found that the portion of fuel next the 

 arched tubes, constituting the sides of the fire-place, and kept 

 cool by the water, was only charred, instead of being consumed, 

 as would have been the case in a fire-place constructed with non- 

 conducting materials. It is well known that in the process of 

 charring or forming coke the hydrogen constituent of the coal 

 must be driven off". In this case it could not have been con- 

 sumed, otherwise the carbon, which remained in the state of coke, 

 would have been likewise consumed ; for the intense heat result- 

 ing from the inflamed hydrogen most certainly would have had 

 the effect of maintaining a red heat in the other portion, and 

 consequently its combustion would have gone on. The coke 

 thus formed was, however, found in the morning not in an ignited 

 state. It, of course, was used in again making up the fire. It 

 was made at a loss — that of the hydrogen constituent of the coal, 

 as above stated. The exact amount cannot be easily ascertained ; 

 and it also differs in the various descriptions of coals employed. 

 Tredgold supposed the loss in such a case might be one-third ; 

 but supposing it is only one-fourth of the heating effect, it is a 

 very important consideration. I prefer estimating the probable 

 loss at one-fourth, although from the subjoined calculations, 

 which I have made from the best data, Tredgold's supposition of 

 one-third is pretty well confirmed. 



Data. 



1st. 24 C-|-13H-i-0=Mineral coal. 



Authority, Bichardson and Regnavlt. 

 2nd. 20 C+9 H-l-0=Coking coal. 



Authority, Liehig. 

 3rd. 1 lb. of mineral coal will melt 62^ lbs. of ice. 



Authority, Watt, Tredgold. 



