/ 



420 Prof. Airy's Remarks on Sir David Brewster's Paper 



and predicting new phaenomena? This principle has been 

 recognised by every philosopher, and is tacitly acted upon in 

 every investigation which is going on in every other branch of 

 science. Is optics to be excepted ? Or am I to understand you 

 to say that Newton's and Biot's molecular theories will apply 

 to the explanation of phaenomena of various classes equally 

 well with the undulatory theory? If this is your meaning, the 

 onus probandi is upon you. It is certain that observations have 

 been calculated upon the undulatory theory, and have been 

 found to agree with the calculations, which have not been cal- 

 culated on any other theory : it is certain that phaenomena have 

 been predicted from the undulatory theory, and have corre- 

 sponded exactly to the prediction, which have not been pre- 

 dicted from any other theory # . If you intend that the sen- 

 tence above quoted should be received literally, you are bound 

 to point out some steps at least of the calculation on other 

 theories. 



Nothing appears to me more prejudicial to the progress of 

 science than vague statements of such a kind as that to which 

 I allude. I am desirous of avoiding this error, and I will 

 therefore point out several instances in which the two theories 

 that you have mentioned fail. 



Newton's theory of alternate states of easy reflexion and 

 easy transmission will not explain the jetty blackness of the 

 central spot in Newton's rings. It will not explain the dila- 

 tation of the rings on increasing the angle of incidence, with- 

 out another principle (the lengthening of the fits), which is 

 negatived by every use made of light which has passed ob- 

 liquely through glass. It will not plausibly explain Grimaldi's 

 fringes, and fails totally for the fringes produced by narrow 

 openings. It will not in the slightest degree explain the fringes, 

 &c. in the shadows of bodies of different forms. It will not 

 explain the interference-bars produced by two mirrors f. It 

 will not explain the spectra formed in telescopes by Fraunho- 

 fer's gratings. 



Biot's theory of moveable polarization will not explain the 

 ordinary polarized rings of Iceland spar, in different positions 



* As a simple instance of calculation, I may point out the polarized rings 

 of Iceland spar, in different positions of the analysing plate ; and as a simple 

 instance of prediction, the change in the character of Newton's rings at a 

 certain angle of incidence, when the lower plate is metallic and polarized 

 light is used. Perhaps the most remarkahle prediction that has ever been 

 made, is that lately made by Professor Hamilton. [See present vol. p. 112 

 and 207.— Edit.] 



f In some place, to which I am at present unable to refer, I have seen 

 a hypothesis to account for the destructive interference of light on a theory 

 of emission. I envy the imagination of any one who can form such a con- 



