NOBERT'S LINES. 55 



the number of divisions corresponding to that quantity 

 "be multiplied by two hundred or five hundred, as the 

 case may be. 



The micrometer should not be fitted into too deep 

 an eye-piece, for it is essential to preserve clear defini- 

 tion. A middle power or Kellner eye-piece is the 

 best, provided the object-glass be of the first quality ; 

 otherwise, use the eye-piece of lowest power. The 

 lens above the micrometer should not be of shorter 

 focus than three-quarters of an inch, even with the 

 best object-glasses ; and the slit cut in the tube can 

 be closed at any time by a small sliding bar, placed at 

 Z, m, fig. 34. 



The wonderful tracings on glass executed by the late 

 M. Nobert, of Earth, deserve attention. The plan 

 adopted by him was to trace on glass ten or more 

 separate bands at equal distances from each other, 

 each band being composed of parallel lines of a frac- 

 tional part of a Prussian inch apart ; in some they are 

 l-1000th, and in others only 1 -4000th of a Prussian 

 inch separated. The distance of these parallel lines 

 forms part of a geometric series : 



O'OOIOOO lines. 0'000463 lines. 



0-000857 C'000397 



0-000735 0-000340 



C'000630 0/000292 



0-000540 0-000225 



To see these lines at all, it is requisite to use a micro- 

 scope with a magnifying power of 100 diameters ; the 

 bands containing the fewest number of lines will then 

 be visible. To distinguish the finer lines, it will be 

 necessary to use a magnifying power of 300, and then 

 the lines which are only 1 -4700th of an inch (Prussian) 

 apart will be seen perfectly traced. Of all the tests 

 yet found for object-glasses of high power, these 

 would seem the most valuable. Robert's tracings have 

 tended to confirm the undulating theory of light, the 

 different colours of the spectrum being exhibited in the 

 ruled spaces varying with the separation of the lines; 

 in those cases where the distances between the lines are 

 smaller- than the length of the violet-coloured waves, 

 *io colour is perceived ; while, on the other hand, if 



