72 THE POPULAR SCIENCE MONTHLY. 



15. The outlet should be effectually protected against any possibil- 

 ity of back-draught — indeed, it should have a considerable amount of 

 suction ; because any liability to back-draught is quite incompatible 

 with an efficient outlet. The outlet, therefore, should not communi- 

 cate directly with the out-of-doors air, but, by means of a tube or flue, 

 should pass through some permanently heated contrivance. If the 

 outlet go directly to the out-of-doors air — as, for instance, a tube from 

 over the gas to a grating in the outer wall — there will certainly be 

 back-draught ; and so also will there be if the tube lead to an opening 

 into the chimney-flue ; at any rate, when the fire is not burning, and 

 particularly if the room-door be also open, and most certainly if there 

 be also a strong draught up the chimney of another room opening out 

 of the same lobby, as, for instance, a dining-room or a kitchen. To 

 prevent any possibility of back-draught the outlet should be provided 

 with some means of constant suction, and, the more thoroughly to re- 

 move the foul air, the more suction the better, provided there is also 

 an ample inlet for fresh air : if not ample, the suction would produce 

 a smoking chimney and draughts from around the windows and doors, 

 and perhaps draw in air from foundation and drains. The necessity 

 for this suction is generally acknowledged, and it is sometimes at- 

 tempted to be gained by carrying the tube before mentioned up a 

 little way in the smoke-flue, and even by bending it down and round 

 the fireplace. But a fatal objection to this plan is, that it is quite in- 

 operative for the greater part of the year, and is of no use whatever 

 unless the fire is burning ; when the fire is not burning it may, indeed, 

 become an inlet, and then an additional objection is, that a back- 

 draught down the smoke-flue carries the soot into the room, to the 

 spoiling of the ceiling, paper, and furniture. And, to be really effect- 

 ual, the suction referred to must be constant and permanent, and oper- 

 ative both winter and summer, and day and night ; and whether the 

 apartment is occupied or not, and whether the fire is burning or not. 

 The outlet must, therefore, pass through some contrivance for keeping 

 it constantly and permanently heated. 



16. The inlet should admit only warmed air ; because the admis- 

 sion of cold air would produce dangerous draughts, and these special- 

 ly directed toward the part of the room occupied by the inmates in 

 cold weather, viz., the neighborhood of the fireplace. The inlet should, 

 therefore, open out of a warm lobby or corridor. 



17. The outlet should be sufficiently large to carry off all the foul 

 air at the time when the apartment is being put to its maximum of 

 use ; because it is just at that time the outlet is most needed, its ca- 

 pacity for other times could be regulated by a valve. The outlet for 

 a dining-room, for instance, should be calculated for a dinner or supper 

 party, and that of a drawing-room for a ball, conversazione, or soiree, 

 and should be sufficiently capacious to carry off, at the very least, 

 fifteen cubic feet per minute for each occupant. The outlet should, 



