VARIOUS METHODS OF HEATING DESCRIBED IN DETAIL. 215 



reason is, the metal is a good conductor, and quickly conveys 

 away the heat from the water by imparting it to a colder body, 

 while the wood is a bad conductor, and retains it. Again, the 

 metallic tank will have the same temperature as the water 

 within, before the wooden one is sensibly warm. In fact, the 

 wooden tank retains and accumulates the heat, while the metal- 

 lic one gives it off as soon as it receives it. None of this heat, 

 however, is lost to the atmosphere of the house ; for though the 

 wooden tank parts with its accumulated heat more slowly, it as 

 certainly parts with it, in the course of time, as the metallic one. 

 It parts with its heat gradually till it is reduced to the same 

 temperature as the atmosphere around it. A house heated with 

 a wooden tank will maintain an average temperature with less 

 expenditure of fuel than a thin metallic one, the other circum- 

 stances being equal, which is accounted for by the fact that 

 when a house is suddenly heated, the wann air is forced rapidly 

 upward, and, coming in contact with the glass, is rapidly cooled, 

 descends, and is again warmed, till the warming surface is 

 entirely deprived of its heat ; then the temperature falls. On 

 the other hand, when the heat is disseminated at a low temper- 

 ature, the atmosphere is less agitated, and the ascending air 

 less rapid in its motion. Not so much escapes through the laps 

 of the glass, or is cooled down by the external cold upon its 

 surface ; and hence, the same quantity of specific heat maintains 

 a given temperature for a longer time, when gradually given off, 

 than when suddenly given off at a high temperature. 



Although the sudden rise and fall of temperature by thin 

 metallic tanks be apparent, we do not condemn their use for all 

 purposes. As we have already said, they may be profitably 

 used in many kinds of erections, and for various purposes ; and 

 I consider them worthy of more extended trials. But I do not 

 believe that they will ever supersede cast metal pipes for the 

 general purposes of heating by hot water ; and for a retention- 

 tank, I would decidedly prefer wood. Fig. 45 represents a 

 house with a wooden tank, in which the water circulates by 

 various divisions, after it enters from the flow-pipe. This tank 

 was erected in a plant-house beneath the stage, as shown in the 

 end section, Fig. 46, which may be objectionable as regards 



