THE TANK SYSTEM OF HEATING. 
241 
means of drying the atmosphere of the pit — in either case to 
the manifest increase in the consumption of fuel. This is 
entirely obviated in the proposed method. It will be seen by 
the annexed plan that the tank is formed precisely as recom¬ 
mended by Mr. Rendle, and running to within a foot of either 
end walls, onty there are added shelving-boards, from the sides of 
the tank to the s ; des of the pit: on these (and bearers) is laid 
the false bottom or covering to the tank, which being raised six 
or eight inches above the top of the tank, allows the heat arising 
therefrom to spread under the whole of the bed, and is so dis¬ 
tributed equally throughout; and this position of the side boards 
also forms a considerable vacuity, from their undersides to the 
ground which receives the heat given off on either side of the 
tank. This heated air is allowed to pass on to the surface of the 
bed through apertures formed by the tank being made shorter 
by a foot at each end than the pit; where an upright board is 
required to keep up the soil of the bed, and a cover to the 
opening may also be had, which will give complete command of 
the surface heat. 
The well-known aerostatic fact of the volatility and expan¬ 
siveness of rarefied air will at once explain the cause of a cir¬ 
culation, for this under heat will necessarily be highest at the 
warmest end of the tank; and here it will ascend through the 
opening between the end of the tank and the outer wall of the 
pit, to the glass, with greater velocity than at the opposite end, 
thus leaving a vacuum into which the cooler air immediately 
rushes, which in its turn will become heated, and again rise ; 
by these alternations giving a free circulation of warm air 
throughout the whole pit, which will materially tend to lessen 
the degree of moisture generally observable on beds over tanks. 
The idea originated with J. H. Schroder, Esq., of Brixton, a 
gentleman much attached to and a most liberal patron of hor¬ 
ticulture, to whom we are indebted for this opportunity of 
making known what we consider a desideratum in the tank 
system. 
It must be understood, though we use the term boards, when 
speaking of the inclined pieces from the edge of the tank to the 
sides of the pit, any other convenient material may be used: 
slate would perhaps be best, but for the false bottom, we greatly 
prefer wood. When the system was first brought into notice, 
