152 COrvRESPONDENCE. 



Mr. Mayall iu describiug the "demonstration" overlooks, and makes 

 no mention of certain facts that were shown, of vital imjjortancc to the 

 truth of my statements regarding the Tolles' -|^th. When Mr. Mayall 

 unexpectedly brought this objective, I could not at the time call to 

 mind all the jjarticulars of trials made two years ago, and needed re- 

 ference to my notes. However, I told him that the slit in the semi- 

 cylinder was formed by placing two strips of tinfoil across the centre, 

 with their edges approximating, having been set in position under a 

 low power by the aid of Canada balsam. I presented Mr. Mayall with 

 a semi-cylinder better polished and finished ; he brought the original 

 one, having jjut a slit iu place, and asked about the width. I said 

 that it was far too wide, and so it proved, as it gave about the same 

 immersion apertures stated by Mr. Tolles. I then made a slit nar- 

 rower and with a water contact and projjer adjustment ; the result was 

 an immersion ajierture — less than the 68° I had formerly given. 

 Mr. Mayall then protested that the thickness of the foil perhaps cut 

 otf oblique rays. Thinking that there might be some reason in this 

 (though I did not ascertain if it was so by light coming through at 

 over 100°), the remedy at once occurred to mc. I covered the plane 

 of the semi-cylinder with opaque black varnish, through which with 

 a steel point I made a fine clean cut exactly midway across the semi- 

 cylinder. The test now repeated with water again gave an angle of 

 less than 68^ ! Mr. Mayall then strongly contested that the slit was 

 " too narrow." I replied that it would bear to be made wider and still 

 bring the angle within 82°. I wished to do this, but it was not tried, 

 nor can I tell the actual width of the slit as the varnish was imme- 

 diately wiped off, and thus the " demonstration " ended, Mr. Mayall 

 only allowing such a width of slit as would support his statements, 

 and bring the aperture up to near what Mr. Tolles had asserted, and 

 I adopting a slit that would cut off lateral pencils and show the aper- 

 ture I had formerly stated ! As we could not agree upon this point, 

 I concluded that it was useless to call Mr. Mayall's notice to any 

 other measurements to prove my position. 



From recent experiments I maintain that the narrower 4,he slit, 

 the more accurate the results ; I mean, of course, a slit with thin 

 edges, that will not cut off rays within the ajierture to be tested. 

 It is obvious that the slit may be opened so wide as to be practically 

 without effect. I leave it to even the most inexperienced to judge 

 which direction is most conducive to accuracy, for the object of the 

 slit is to obtain a mere line, or film of light in the focus of the 

 object-glass. 



The -^th object-glass being out of my possession, I was unable 

 to make any further verifications free from impediment. I there- 

 fore requested Mr, Crisp to again favour me with the loan of it, 

 and with his usual courtesy and impartiality he has done so. I now 

 repeat the measiirements with a semi-cylinder having in its centre a 

 clear line cut througli black varnish, and a thin glass cover cemented 

 over the slit witli Canada balsam. On looking through the slit it 

 admitted rays through beyond an angle of 130, the object-glass was 

 focussed to slit and carefully adjusted for best definition, and though 



