258 PROCEEDINGS OF THE SOCIETY. 



to Helmlioltz's work on the diffusion disk whilst he himself was unaware 

 of it, and founded his autipoiut theory on the earlier work of Airy. 

 But I did not discuss at all the question as to the source of the two 

 authors' knowledge : I merely pointed out that the conclusion they came 

 to was exactly the same, viz. that the spurious disk theory might be 

 extended to the Microscope by taking into account the modification 

 which the spurious disk undergoes owing to the peculiar modes of 

 lighting microscopical objects. This extension, if proved sound, would 

 be a distinct addition to the astronomical theory of the spurious disk, 

 and would thus justify a claim of originality on the part of both 

 Dr. Altmann and Mr. Gordon, and I do not doubt that hithei'to every- 

 body has considered his suggestions for an antipoint theory of the 

 Microscope as Mr. Gordon's principal scientific achievement. 



I cannot see anything in my paper which could be interpreted as 

 aimins: at the extension of my somewhat curt reference to the " smiles 

 and sneers " of Dr. Altmann for the Abbe theory to Mr. Gordon's treat- 

 ment of the same doctrine. Still, personally, I would rather put up 

 with smiles and even with sneers than be accused of such extraordinary 

 elementary blunders as those which Mr. Gordon attributed to Professor 

 Abbe. 



Mr. Gordon deals very extensively with the aerial image of a grating, 

 and states, on the whole correctly, the views on the subject which follow 

 from the undulatory theory as usually interpreted, and which were 

 implicitly beheved until about ten years ago. This belief was then 

 rudely shattered by a crucial experiment which Dr. Johnstone Stoney, 

 F.R.S., showed at the Royal Society, and which I am showing here 

 to-night. This experiment proves conclusively that under the conditions 

 prevailing in the Microscope the flame image behaves exactly as if it 

 consisted of diffused light ; the diffraction-spectra refuse to vanish, no 

 matter how carefully the condenser may be focused and adjusted.* 

 I have, indeed, gone to the length of using an apochromatic objective as 

 a condenser — an experiment which I showed when reading my first paper 

 on microscopical theories : still the diffraction spectra will not vanish. 



When a crucial experiment of this kind disagrees so obviously with 

 theory, the latter has to be modified ; the last word has not yet been 

 spoken on this point, and I therefore prefer to leave the theoretical 

 aspect undiscussed. But there can be no doubt whatever that the 

 diffraction-spectra cannot be got rid of by using critical light. 



* The experiment shown by me accurately represented the Microscope in its 

 normal working condition ; the focusing of the lamp-flame was therefore im- 

 perfect for the reason which I fully explained in my first paper on Microscopical 

 Theories (see this Journal, 1904, p. 612). On applying the calculation there first 

 exemplified to the present case, I find that the diffusion of focus of the foremost 

 and rearmost portions of the flame employed would cause diffused light reaching 

 in maximo over about four adjoining slits when the edge of the flame was em- 

 ployed, but so slight as not to bridge even two adjoining slits when, as was usually 

 the case, the broad side of the flame was presented to the condenser. Even in 

 the first case the diffraction-spectra should have been considerably modified, whilst 

 in the second case they should have vanished. As all those present saw, they 

 refused to vanish in every case. In fact, in the paper already cited, on p. 631, I 

 also mentioned that no matter how delicately the experiment may be carried out, 

 the diffraction-spectra can only be got rid of by illuminating only one single slit ; 

 but no one would or could expect grating-spectra from a single slit. 



