Research on some Standards of Light. 



491 



gen is burning at the rate of 0*8 cubic foot per minute, we then regu- 

 late the supply of gas so as to keep the ratio gauge in balance. At the 

 end of fifteen minutes we press the key which stops the gas and swings 

 the diaphragm into position. No photometric readings are taken at 

 the end of this first fusion for two reasons : 1 . The walls of the fur- 

 nace have not yet had time to heat up, and the platinum, though 

 fused on the surface, is probably still solid underneath. 2. Some par- 

 ticles of lime dust are pretty sure to be floating on the surface if the 

 ingot of metal has only just been put into a newly made crucible. We 

 therefore merely use this fusion to adjust the position of the furnace 

 and cover, so that the centre of the platinum surface and the centre of 

 the aperture in the. cover should be on a vertical line passing through 

 the centre of the diaphragm. 



This done, we re-start the blow-pipe, and fifteen minutes later, as the 

 metronome rings, we press the electric key, keeping an eye on the 

 photometer. The photometer is kept in balance, and every ten seconds, 

 as the metronome rings, the position of the photometer head is marked 

 off. By the time some ten or fifteen readings have been recorded the 

 platinum has cooled below its melting point. The distances are then 

 read off at leisure, the blow-pipe is swung back into position, the 

 chopper re-set, the mirrors are interchanged, and everything is ready 

 for a fresh start. Fifteen minutes later another series of readings can 

 be taken. 



A Lummer and Brodhun photometer* and photometer bench were 

 used diiring these experiments, the distance between the lights being 

 331*4 cm. The position of both the sources of light was fixed, the 

 photometer head alone being movable. To ensure the maximum sen- 

 sitiveness it is well to keep one eye exclusively for the photometric 

 observations, covering it when the readings, are not being taken with 

 a black screen. In all the experiments incandescent lamps were used 

 as standards of reference. When the necessary precautions are taken 

 these form very reliable standards, remaining absolutely constant for 

 many hours.! The pressure on the terminals of these lamps must be 

 adjusted with the greatest care, as the light varies with the sixth 

 power of the electromotive force. In the present case fifty-volt lamps 

 were used, a specially constructed divided resistance being placed 

 across the terminals of the lamp. By means of a potentiometer the 

 electromotive force on l/20th of this resistance was compared with 

 the electromotive force of a Clark's cell. To avoid any rapid changes 



* The photometer head and part of the photometer bench are shown in fig. 4. 

 To avoid any stray light, the photometer was hang with black velvet curtains. 

 These have been drawn aside in fig. 4. Tor a full description of this instrument 

 see ' Zeitschrift fur Instrumentenkunde,' p. 41, 1892, and ' J. fur G-asbeleuchtung 

 und Wasserversorgung/ vol. 37, p. 61, 1894. 



f ' Zeitschrift fiir Instrumentenkunde,' vol. 10, p. 119, 1890. 



