ON PROJECTION g 



was all that could be desired. The fact must be, therefore, 

 that the more luminous rays from the edge of the slide, while 

 they fell upon and got through the lens, did not get through 

 the pinhole until we drew back the light. 



Let us investigate this further. With the slide in the 

 lantern, leave the lens in its place ; but this time remove the 

 condensers instead. The slide itself is now very nearly as 

 brightly illuminated as before ; but only the centre of the 

 picture is at all bright upon the screen, and we have again 

 the dark margin, all the more striking by contrast with the 

 bright centre. Considering this, we begin to understand the 

 state of the case. Our slide is very greatly transparent. 

 (It is not wholly so, or there could be no picture of it, nor 



FIG. 5 



could we see it.) Now, any imperfectly transparent object 

 scatters light falling upon it, and in this way sends out some 

 rays in all directions open in space. But the greater part of 

 the luminous rays which strike on it do go straight on from 

 the light in the lantern, or continue diverging ; and as regards 

 far the greater part of the luminous rays, therefore, the case 

 stands as in fig. 5. Those passing from the radiant, B, through 

 the outer part of the slide, s, pass outside the lens, L, and 

 give no light to the final image ; the lens only picks up from 

 the margin of the slide the few comparatively dim ' scattered ' 

 rays. On the other hand, a central cone of bright direct rays 

 does get to the lens, L, and forms a good image of the central 

 portion of the slide. 



The properties of a lens already indicated, suggested long 



