1843.] On the Ventilation of Lighthouse Lamps. 365 



beginnings at each lamp, and being fixed to the frame which 

 supports the lamps, is made to converge together and to the 

 axis of the frame by curved lines. The object is to bring the 

 tubes together behind the reflectors, as soon as convenient, 

 joining two or more into one, like a system of veins, so that one 

 ventilating flue may at last carry off the whole of the lamp pro- 

 ducts. It is found that a pipe |ths of an inch in diameter is 

 large enough for one lamp; and where, by junction, two or 

 more pipes have become one, if the one pipe has a sectional 

 area, proportionate to the number of lamps which it governs, 

 the desired effect is obtained. 



Each of the pipes, Jths of an inch in diameter, passes down- 

 wards through the aperture in the reflector over the lamp, and 

 dips an inch into the lamp-glasses ; it is able 

 to gather and carry off all the products of 

 combustion, though, perhaps, still 2 inches 

 from the top of the flame, and therefore not 

 interfering in any respect with it, nor coming 

 as a shade between it and any part of the 

 reflector : the flame and reflector are as free 

 in their relation to each other as they were 

 before. Neither does this tube hide from 

 the observer or mariner, a part of the re- 

 flector larger than about 1^ square inch of 

 surface, and it allows of a compensation to 

 two or three times the amount ; for, when 

 in its place, all the rest of the aperture over the lamp which is 

 left open and inefficient in the ordinary service, may be made 

 effectual reflecting surface, simply by filling it up with a loose, 

 fitly formed, reflecting plate. 



At this termination of the ventilating flue an important ad- 

 justment is effected. If the tube dip about an inch into the 

 lamp-glass, the draught up it is such that not only do all the 

 products of combustion enter the tube, but air passes down 

 between the top edge of the lamp-glass and the tube, going, 

 finally, up the latter with the smoke. In this case, however, 

 an evil is produced, for the wick is charred too rapidly ; but if 

 the ventilating flue descends until only level with the top of 

 the lamp-glass, the whole of the burnt air does not usually go 

 up it, but some passes out into the chamber, and at such times 



