20 R. THRELFALL. 
so that not only does ninety-five per cent. of the energy coming 
from such a lamp refuse to affect our eyes at all, but the greater 
part of what does so affect them does it by means of rays to which — 
they are only slightly sensitive. That a much higher efficiency is — 
attainable we know by Langley’s study of the glow worm, the 
most efficient lamp in existence, for the heating power of the | 
energy emitted is too small to be measured by ordinary means, or 
a glow worm “burns cold.” Ebert claims that his lamp will give 
a ‘serviceable’ light for an expenditure of a millionth of a watt. 
Suppose ‘a serviceable light’ means half a candle power, then for — 
a fifteen candle power lamp we should require only three 
hundred thousandths of a watt, instead of the fifty watts say we 
require at present ; of course further experience may show that 
this estimate of Ebert’s is too smal], but it is possible, and in any 
case there is no doubt that an enormous efficiency is gained. This 
is not quite new by the way, we always knew that phosphorescence 
was an efficient form of illumination, but the figures given, are 
I believe, the first to have been anything like fairly estimated. 
In connection with this I might remark that incandescent burners — 
of the Welsbach type, using gas, probably work in virtue of the — 
substance of the mantle being rendered phosphorescent by rise of — 
temperature. Whether anything will come out of Ebert’s lamp! — 
cannot say, for the practical difficulties appear insuperable at 
present. 
Before I leave this subject, I will mention a result I got some ; 
years ago in a discussion on sensitive galvanometers. I found — 
that I could make an instrument so sensitive that a definite 
motion. of the mirror corresponded to an energy supply of 10 
ergs per second. This I have now increased so that an effect = 
produced by a supply of 10 ergs per second, or, say, 10° ; 
horse power. Langley’s numbers for the ee efficient light 
required to produce vision are about 10° ergs per second. 
Consequently, the human eye must look to its laurels, or its claim — 
as the most sensitive mechanism will have to be rejected. 
