266 Proceedings of the British Association. 



" In the Second Part," says Prof. Airy, " of the Transactions for 

 1840, the Royal Society has published a memoir by me, explaining 

 on the Undulatory Theory of Light the apparent new polarity disco- 

 vered by Sir D. Brewster, which explanation is based on the assump- 

 tion that the spectrum is viewed out of focus ; an assumption which 

 corresponded to the circumstances of my own observations and 

 to those of some other persons. Since the publication of that me- 

 moir, I have been assured by Sir D. Brewster that the phenomenon 

 was most certainly observed with great distinctness, when the spec- 

 trum was viewed so accurately in focus that many of Fraunhofer's 

 finer lines could be seen. This observation appeared to be contradic- 

 tory to those of Mr. Talbot, cited by me in p. 226 of my memoir, as 

 well as to my own. With the view of removing the obscurity that 

 still appeared to embarrass this subject, I have continued the theore- 

 tical investigations for that case which was omitted in the former 

 memoir, — namely, when the spectrum is viewed in focus, or when 

 a=o (p. 229 ;) and I have arrived at a result which appears com- 

 pletely to reconcile the seemingly conflicting statements." — Phih 

 Trans., 1841, p. 1. 



Now, in the investigations which this paper contains, and which 

 Prof. Airy considers satisfactory, there are two points which require 

 special attention. The first of these is the assumption, necessary for 

 the explanation, that even when any single point of the spectrum is 

 seen accurately in focus, it forms a different image on the retina, the 

 extent of the difference being exceedingly less than the interval 

 between the bands. The supposition appears to me quite untenable, 

 and one which cannot for a moment be admitted. The second point 



relates to the expression of - , - which Prof. Airy obtains for the 

 interval between the bands ; from which it follows, that this interval 

 is inversely as the radius of the pupil, or the area of the object-glass. 

 But the intervals have no such rotation, and Prof. Airy does not say 

 that such a rotation was ever noticed in any of his experiments. I 

 have made the experiment repeatedly and carefully, and can state 

 with confidence, that the fringes do not vary with the diameter of 

 the pupil or the operations of the object-glass. Their interval re- 

 mains the same, whether we look through a pin-hole or with the 

 pupil in its fullest expansion ; and it is equally invariable when the 



