358 



Mr. G. Dodd on 



room, the principal object being to shew the relative con- 

 struction of a, h, c. 



Fig. 2. 



Now, the question is, sir, are these discrepancies ever 

 observed in practice 1 Do the mere facts of changing the 

 point of incidence, changing the direction of passage, or 

 varying the distance at which the spectrum is observed, 

 make such great changes in the chromatic distribution of 

 the spectrum ? I think not. It is generally thought 

 desirable that the central ray shall pass parallel to the 

 base ; but I am noi aware that in any of Newton's, Fraun- 

 hofer's, or Brewster's experiments the mere position of the 

 point of incidence, with respect to the apex and the base, 

 was any matter of moment : as I am aware that parallelism 

 to the base is generally sought for in practice, I have con- 



