THE TANK MODE OF HEATING. 



193 



To Mr Rendle, the very respectable 

 nurseryman at Plymouth, we are, how- 

 ever, mainly indebted for bringing this 

 excellent mode of obtaining bottom heat 

 into general notice. Mr R. recommends 

 wooden tanks made of the best Memel 

 pine, and covered at top with slates 

 closely jointed with cement or Aberthaw 

 lime, to prevent a superfluity of steam 

 from escaping into the house. The tanks 

 are to be supplied with water from a 

 cistern placed without the house. "In 

 some cases," he says, "I would recom- 

 mend that piers be constructed on which 

 the tank could be made to rest. There 

 would then, underneath the tank, be 

 formed an open chamber, which would 

 be of great utility, and well adapted for 



forcing sea- 

 Fig. 245. kale or rhu _ 



barb." The 

 annexed cut, 

 fig. 245, will 

 show a section 

 of the tank — 

 a the tank ; 

 b the flue; c 

 the board for 

 enclosing the 

 plunging ma- 

 terial ; e the slates for covering the tank. 

 The flue, as here shown, meets the objec- 

 tions to a certain extent thrown out 

 against this mode of heating, as being of 

 itself insufficient to heat the atmosphere 

 of the house, as well as the bed in which 

 the plants are set, because it will give 

 out a considerable amount of heat by 

 radiation from the sides ; and, indeed, 

 although the heat from the top does not 

 directly enter the house from the cover, 

 yet the tanks give off heat by conduction to 

 the brickwork above, and ultimately this 

 heat is diffused through the house. But, if 

 the tanks are supported on piers, as stated 

 in the treatise, and shown in the figure, we 

 cannot see how the flues can be carried 

 around the house — for how are they to 

 pass the spaces between the piers'? If the 

 tank is supported on two walls extending 

 to their whole length and breadth, then 

 this difficulty disappears. Were the sides 

 and ends of the tank, as here shown, con- 

 structed of slate, pavement, brick, or 

 cast-iron, then a considerable amount of 

 heat would pass through them, perhaps 

 enough to maintain the required tem- 



VOL. I. 



perature of a small house. This, how- 

 ever, cannot be expected from tanks 

 whose sides and ends are of wood. Mr 

 Rendle seems aware of this deficiency of 

 heat, and very properly says, " In cases 

 where the heat from the flue is not suffi- 

 cient to warm the atmosphere of the 

 house, it is better, although not absolutely 

 indispensable, to have carried round the 

 tank, close by its sides, a 2-inch cast- 

 iron pipe, which could be joined to the 

 tank," &c. 



Mr Rendle very properly objects to 

 wrought-iron tanks, on account of their 

 liability to corrode ; but is favourable to 

 the use of cast-iron ones — stating, how- 

 ever, which is true, their greater expense 

 than those of wood. Wooden tanks coated 

 with lead, say 5 lb. to the foot, he remarks, 

 will answer every purpose, and would last 

 for years. This, however, we may observe, 

 depends greatly on the water to be used 

 in them. 



Similar tanks coated with zinc or cop- 

 per he would not recommend ; and very 

 properly, for the first is of too temporary 

 a nature, and the last far too expensive. 



Brick and cement tanks, he is of opinion, 

 might be used advantageously ; and tanks 

 of slate of good quality he c 1 would always 

 give a preference to;" and recommends, 

 in their construction, " that the joints be 

 true and evenly filled, and cemented to- 

 gether with red lead, as well as strength- 

 ened with cramps when requisite." 



Fig. 246, from Mr Rendle's treatise, 

 shows a connection between two tanks 



Fig. 246. 



where a passage intervenes. By placing 

 a stopcock on the siphon, the circulation 

 may be carried on in one or both, at the 

 will of the owner. 



Mr Rendle deserves w T ell of his country 

 for the reform he has so eminently been 

 the means of bringing about; and al- 

 though it does not clearly appear that he 



2 B 



