LIEUT. DRUMMOND ON THE ILLUMINATION OF LIGHT-HOUSES. 387 
Encyclopedia, in 1811, showed that it might be built of separate pieces; and 
this was an important step, inasmuch as it rendered of easy execution what was 
before impracticable. To Dr. Brewster therefore the priority of suggesting this 
improvement is due. To MM. Arago and Fresnel, obviously unacquainted 
with what had been previously done or recommended, belongs the praise of 
having first got such a lens constructed, of combining it with a very powerful 
lamp, and above all of giving it a most useful and beneficial practical appli- 
cation. 
The Cordouan light-house at the mouth of the Garonne, the difficult entrance 
to Bourdeaux, has been fitted up in this manner ; and as the lens and lamp used 
in our experiments were purchased at Paris of the same makers, a pretty accu- 
rate estimate may be formed of its merits when compared with the light-houses 
of this country. 
Such are the methods at present in use in the best light-houses of Great Bri- 
tain or France. The third and last method is that which I have ventured to 
propose, and in which the light is derived from a source altogether different 
from the preceding two ; a ball or cylinder of lime, intensely ignited, being 
substituted for the Argand lamps. 
For the purposes of a survey, when portability rather than economy is the 
more important object, this intense heat was obtained by directing a stream of 
oxygen gas through a flame of spirit of wine : but for a light-house, where, from 
the long continuance of the light, economy is a primary object, and portability 
is no longer required, it was desirable if possible to substitute hydrogen gas for 
the alcohol. As the effects produced in this manner are very remarkable and 
considerably exceed those formerly obtained, I shall now give a short descrip- 
tion of the apparatus, and then proceed to state the results of our experiments. 
Plate XII. fig. 1. represents the lamp. The two gases, oxygen and hydrogen, 
proceding from separate gasometers, enter at o and h, but do not mix till they 
arrive at the small chamber c, of which fig. 2. is a section ; into this chamber 
the oxygen gas from the inner tube is projected horizontally through a series of 
very small apertures, and the hydrogen gas rises vertically through a series of 
similar apertures at d. The united gases then pass through two or three pieces 
of wire-gauze placed at e, and being thus thoroughly mixed, issue through 
the two jets against the ball b. To prevent the wasting of the ball opposite 
