14 



SIR GEORGE M'KENZIE's FLUE. 



them in the ground like drains ; this mode was aftervyards changed to 

 that of building them in the solid walls, either at the back or front of the 

 house. By this means much of the heat was lost in the solid building, 

 little of it finding its way into the house. But the most rational im- 

 provement which followed this, was the practice of building them quite 

 detached from all other description of walls. 



J. R. Gowen, Esq., in the " Horticultural Transactions," vol. III., pro- 

 posed flues constructed of bricks, of the usual thickness at the sides, and 

 hollowed out in the middle, vdth the ^iew, no doubt, of attaining equa 

 strength to the flue in general use, and admitting, at the same time, of 

 the more ready escape of the heat into the house through the thinner 

 parts. The covers of these flues were hoUowed^out in a similar manner to 

 the bricks mth which the sides were built. 



Sir George M'Kenzie, in the work last quoted, recommended what he 

 called an embrasure flue, of which the annexed diagram wiU give some 

 idea. The principle of tliis flue is that of exposing a greater heated 

 smface in proportion to its length. This flue was found to fall far 

 short of the expectations of its inventor, in practice, and was only in a 

 very few cases adopted. The same intelligent gentleman proposed for trial 

 a riangular cast iron flue ; but, hke the flues proposed by other indi^iduals, 

 of the same material, was found to possess no advantage whatever, but 

 were hable to many objections, of which that of being rapidly heated and 

 as soon losing then' heat, were not the least. 



The German stove, with iron pipes for conveying smoke and heat, has 

 been long used on the continent, but is hable to the objections stated as 

 belonging to the last. The improved flue recommended by Mr. Loudon, 

 in " A Treatise on Several Improvements in Hot houses,^^ Sfc, p. 33, con- 

 sisted in dividing the flue into chambers, or compartments, mth a view to 

 arrest the progress of the heat, and to fiU. each chamber with smoke and 

 heated air before the one next in fi'ont of it could become occupied with 

 it, and so on till these chambers at last became aU completely charged 

 with heat, before any could escape at the chimney top. This, as well as 

 the hot ah' flue built on the top of the ordinarj^ flue, i^commended by the 



