THE PARTS OF A LANTERN 23 



however, is that the lenses are not ground thin at the edges ; 

 nearly all lantern -makers being far too careless in this respect. 

 Not only is a needlessly thick lens much more likely to crack, 

 and more absorbent of light, but it is distinctly worse in optical 

 performance. The lenses should be ground to as nearly 

 knife-edges as possible, being only just edged down for fixing 

 into their cells. 



Both lenses should be mounted so loosely in their brass 

 cells, that they can be turned round with the fingers, else they 

 may crack merely from expansion when heated, and they often 

 become hotter than the hand can bear. Holes must be 

 pierced in the margin of the cell, to allow of the escape of the 

 aqueous vapour which always forms when the lantern is first 

 lit. Only the back lens ever cracks from heat alone. If the 

 crack be irregular, the lens must be replaced ; but it often 

 happens that the crack is quite straight across a diameter, in 

 which case the cracked lens will answer perfectly well if the 

 crack is arranged perpendicularly, and the operator may feel 

 certain that his lens will never crack again. 



12. Size of Condensers. The best general size of con- 

 densers is 4 inches diameter next the slide. The standard 

 size of slides is 8J inches square, and 4 inches will cover 

 1 cushion ' slides on such square of glass. Of course a 

 diameter of 3^ inches is ample for the usual circular slides, 

 which give a disc of 3 inches diameter. If it is desired to 

 cover square slides of three inches, as the corners are never 

 absolutely square, it is better to have 4| inches diameter than 

 4| inches, because practically only a certain angle of light 

 can be taken up from the radiant, and the larger surface this 

 is spread over, the more it is diluted, so that for the usual 

 size of slides, less light is passed through them by a large 

 condenser than a small one. Theoretically any angle might 

 be taken up by a condenser, but practically it is limited ; first 

 by the thickness of the glass, which would both crack, and lose 

 as much light by absorption as was gained in angle ; and also 



