682 ; REPORT—1904. 
5. Distribution of Magnetic Induction in Multipolar Armatures.} 
By W. M. Tuornton. 
6. On Large Bulb Incandescent Electric Lamps as Secondary Standards of 
Light.| By Professor J. A. Fiemine, 1.A., D.Sc., F.RS. 
The importance of possessing a secondary standard of light which shall be at 
once portable, convenient, and constant is generally acknowledged, and the choice 
lies between some form of flame standard and some form of incandescent standard, 
It is known that the candle power of a flame standard is affected by the 
variation of moisture in the air, atmospheric pressure, and carbonic acid, and that 
even in well-ventilated rooms changes in atmospheric moisture and pressure 
may cause variations to the extent of 4 per cent. in the candle power of a flame 
standard. 
For the last eight years the author has employed as a secondary standard of light 
a form of carbon filament incandescent lamp having a specially large bulb and a 
filament prepared in a certain manner. The size of the bulb prevents any sensible 
deposit of carbon uponit, and the particular preparation of the filament by ageing it 
previously to mounting in the bulb prevents variations in candle power provided 
the lamp is used in a particular manner and only for a short time on each occa- 
sion. These large bulb lamps are not intended for continuous use, but only to be 
employed in setting or adjusting the distance of another lamp from the photo- 
meter disc in a photometer, so as to produce on the dise a predetermined illumina- 
tion. The lamp to be measured is then substituted for the standard lamp, and by 
this process of double weighing all errors due to want of symmetry in the photo- 
meter are eliminated. If only used in the above manner the standard lamps may 
be used for hundreds of times without being in operation altogether for more than 
a few hours, and by comparing a number of these standards with one another it is 
possible to preserve a standard of light with great constancy. The illuminating 
power of these lamps is not affected by changes in moisture and atmospheric 
pressure, and the experiments described in the paper show that they are not 
sensibly affected by change in atmospheric temperature. The light of the lamp 
is therefore determined only by the current passing through it, and this can be 
measured easily with an accuracy of one part in a thousand by means of a potentio- 
meter. Hence, when the filament is traversed by the same current, the lamp 
gives the same light. The author has therefore devised an arrangement consisting 
of a large bulb lamp united with a current measuring instrument and a variable 
resistance. This instrument, however, is not graduated directly to read current, 
but is graduated to read candle power; hence all that has to be done is to place 
the instrument on a circuit supplying a steady electromotive force and vary the 
current through the lamp by means of a rheostat until the needle of the current- 
measuring instrument indicates a certain candle power. The lamp then has a known 
candle power in a certain direction. Such an arrangement, although not sensitive 
enough for laboratory purposes, is quite sufficiently accurate and very convenient 
for the workshop comparison of ordinary glow lamps. For more accurate observa- 
tions, however, a potentiometer must be employed, since the current or the yoltage 
on the terminals of the lamp must be determined to at least one part ina thousand, 
if the candle power is to be correct within half per cent. If incandescent lamps 
are compared directly with flame standards, the latter not being corrected for 
atmospheric moisture and pressure, then differences to the extent of even 4 or 
5 per cent. may be found in measuring the candle power of the incandescent lamp 
on different occasions and in different places. As this difference amounts to 
about one candle in twenty-five, it is far greater than possible errors in observa- 
tions made with due care. 
For the purpose of bringing into agreement photometric measurements made 
! Published in the Hlectrician, August 26, 1904. 
