Prof. Faraday on Lighthouse Illumination. 321 
of both were shown.] In modern times the principle of total 
reflexion has also been employed, which involves the use of glass, 
and depends both upon refraction and reflexion. In all these 
appliances much light is lost: if metal be used for reflexion, a cer- 
tain proportion is absorbed by the face of the metal; if glass be used 
for refraction, light is lost at all the surfaces where the ray passes 
between the air and the glass; and also in some degree by absorp- 
tion in the body of the glass itself. There is, of course, no power 
of actually increasing the whole amount of light, by any optical 
arrangement associated with it. 
The light which issues forth into space must have a certain 
amount of divergence. The divergence in the vertical direction 
must be enough to cover the sea from the horizon, to within a cer- 
tain moderate distance from the shore, so that all ships within that 
distance may have a view of theirluminous guide. If it have less, 
it may escape observation where it ought to be seen; if it have 
more, light is thrown away which ought to be directed within the 
useful degree of divergence: or if the horizontal divergence be con- 
sidered, it may be necessary so to construct the optical apparatus, 
that the light within an angle of 60° or 45° shall be compressed 
into a beam diverging only 15°, that it may give in the distance a 
bright flash having a certain duration instead of a continuous light, — 
or into one diverging only 5° or 6°, which, though of far shorter 
duration, has greatly increased intensity and penetrating power in 
hazy weather. The amount of divergence depends ina large degree 
upon the bulk of the source of light, and cannot be made less than 
a certain amount, with a flame of a givensize. If the flame of an 
Argand lamp, {ths of an inch wide and 14 inch high, be placed in 
the focus of an ordinary Trinity House parabolic reflector, it will 
supply a beam having about 15° divergence: if we wish to increase 
the effect of brightness, we cannot properly do it by enlarging the 
lamp flame ; for though lamps are made for the dioptric arrange- 
ment of Fresnel, which have as many as four wicks, flames 32 inches 
wide, and burn like intense furnaces, yet if one be put into the 
lamp place of the reflector referred to, its effect would chiefly be to 
give a beam of wider divergence ; andif to correct this, the reflector 
were made with a greater focal distance, then it must be altogether 
of a much larger size. ‘The same general result occurs with the dioptric 
apparatus ; and here, where the four-wicked lamps are used, they 
are placed at times nearly 40 inches distant from the lens, occasioning 
the necessity of a very large, though very fine, glass apparatus. 
On the other hand, if the light could be compressed, the necessity’ 
for such large apparatus would cease, and it might be reduced from 
the size of a room to the size of a hat: and here it is that we seek 
in the electric spark, and such like concentrated sources of light, for 
aid in illumination. It is very true, that by adding lamp to lamp, 
each with its reflector, upon one face or direction, power can be 
gained; and in some of the revolving lights, ten lamps and reflectors 
unite to give the required flash. But then not more than three of 
these faces can be placed in the whole circle; and if a fixed light 
be required in all directions round the lighthouse, nothing better 
