184 Management of Light in Illumination. 



These experiments must be performed at night, or, 

 if made by day, a room must be chosen from which day- 

 light can be effectually excluded. 



Three tables will be necessary in making these ex- 

 periments : on one of them the photometer is to be 

 placed, and on each of the others one of the lights 

 that are to be compared. The heights of these tables 

 should be such that the two flames of the lamps or 

 candles that are to be compared and the centre of the 

 field of the photometer may be at the same horizontal 

 level, or nearly 'so ; and, in order that the- photometer 

 may be at a proper height for observing with conven- 

 ience the shadows projected on its field by its cylin- 

 drical columns, it may be placed on a small stand set 

 down on the table, or on the flat bottom of a square 

 wooden box of a proper height, turned upside down. 

 The height of the photometer should be such that 

 when the observer is seated in a chair before it his eye 

 may be on a level with the upper extremities of the two 

 columns by which the shadows are projected. 



Suppose now that it were required to determine the 

 relative intensities of the light emitted by two candles, 

 the one made of wax, the other of tallow. The three 

 tables are first to be placed at the distance of about 

 eight feet from each other in the middle of the room, 

 or as far as possible from its walls ; the photometer, 

 elevated to a proper height, being placed on one of 

 these tables, and one of the candles on each of the two 

 others. 



The observer is now to seat himself before the table 

 on which the photometer is placed, and with his back 

 turned to the two other tables. 



He will find two shadows in the field of the photom- 



