February i6, 1922] 



NATURE 



205 



Letters to the Editor. 



Xh& Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.'\ 



Test-piates for Microscopes and Microscopic Definition, 



In Nature of September i last (vol. io8, p. lo) I 



iggested that dry films of some of the aniline colours 



light prove suitable surfaces on which to rule the 



»ry fine lines required for test-plates. Since that 



le I have made some trials with films, using 



irious dyes and various materials for the ruling- 



>int. The dyes were eosin, saffranin, cyanine, 



jthyl-green, methyl-blue, and methyl-violet. The 



St of these gave the most uniform films, leaving, 



the evaporation of the solvent, a bright surface 



from structure. Cyanine gave good films, though 



quite so opaque as the methyl-violet for the same 



ickness. The others, either from a tendency to 



rstallise or from drying with a dull surface, were 



)t so satisfactory. 



[The points for ruling were of steel, garnet, car- 

 rundum, and diamond. In all cases, except that of 

 eel, the natural points found among broken frag- 

 ;nts were chosen, and I doubt whether, without a 

 lewhat elaborate grinding tool, an artificial point 

 ild be made so sharp as that given by the natural 

 jakage. 



The points have to be selected by trial, as the micro- 

 >pe is of little assistance in determining their real 

 Terminal shape. The only apparatus which was at 

 hand for ruling purposes was a Cambridge rocking 

 microtome, and the minimum interval between the 

 lines (which corresponds, of course, with the thinnest 

 section it could cut) was somewhat greater than 

 1/60,000 of an inch. No great accuracy in the spacing 

 of the lines could be expected, but it is to the credit of 

 the design of this microtome that there was no dif!i- 

 culty in getting well-separated lines at 30,000 per inch, 

 and occasionally the 60,000 lines were quite apparent. 

 The uncertainty with these close lines was due, I 

 think, chiefly to the rather rough workmanship of 

 the bearings of the rocking arms, but in part, perhaps, 

 to those of the point-holder. 



The only real difficulty in close ruling is that of 

 finding a fine enough point and applying it to the 

 surface with a small enough force. 



As mentioned in my previous communication, this 

 force must not exceed a small fraction of a grain, 

 and should remove the film but not scratch the glass. 

 In the trials the point-holder was made as shown in 

 Fig. I, the materials being reed and split-cane put 

 together with silk splicing and shellac cement. 



This light and rigid frame could rock on the needle- 

 points, one of which entered a conical pit, and the 

 other a V-shaped groove in a fixed fitting which re- 

 placed the knife of the microtome. The whole forms 

 a pendulum the effective weight of which (a few 

 grains) is, say, w, of length L, which if displaced by 

 n amount a exerts a force wa/L in the direction 

 i'posite to the displacement. Thus by placing the 

 plate to be ruled at a suitable distance from the 

 undisturbed position of the ruling-point the force can 

 be adjusted as required. 



While the film is being advanced for a fresh line 

 the ruling-point is withdrawn by means of a silk fibre 

 ^' ading from the swinging frame to a bell-crank 

 ounted on the fixed support. 



A few photographs of bands ruled with this ap- 

 paratus are given in Fig. 2. Ail these were made 

 with diamond points. With a newly ground hardened 

 •NO. 2729, VOL. 109] ' 



steel point the 30,000 band was well shown, but the 

 tool soon became too blunt for any spacing less than 

 15,000 per in. Garnet and carborundum points lasted 

 fairly well for 30,000 per inch lines, but, as might be 

 expected, were inferior to diamond so far as wear 

 was concerned. 



In the earlier trials from ten to sixty lines were 

 ruled for each band, but later it was found that four 

 or five lines were quite sufficient to show all the effects 

 which various kinds of stage illumination have on the 

 definition. 



It is only with opaque and very thin objects that 



Fig. I. — Holder for ruling point. A, Reed; BBB, split cane; CC, needle 

 points ; D, fixed support ; E, diamond point ; F, aniline film ; G, silk 

 fibre ; H, bell-crank. 



these can be exainined to any advantage, thinness in 

 this case beirrg in comparison with the wave-length 

 of light. 



The methyl-violet films from which the photographs 

 were taken were less than one-tenth of a wave-length 

 in thickness, and, though not quite opaque, trans- 

 mitted only a deep blue with some little red. The 

 measurements of thickness were made by noting the 

 displacements of the Newton's rings formed between 

 a lens and the film at a place where part of the latter 

 had been removed. 



Test objects, such as diatoms, are much thicker 

 than this, and, with them, what is seen in the field 



Fig. 2.— Photograph of bands ruled on films of methyl-violet. 

 a, 10 lines about 15,000 per in. X350 | e, 4 lines about 37,000 perin. x6oo 

 ^> I' 11 20,000 ,, ,, /, 5 ,, ,, 3S,ooo ,, ,, 



c, ,, ,, 17,500 „ ,, g, Intersection 



d, ,, ,, 2t,ooo „ ,, j of bands 17,500 ,, ,, 

 The photographic lines are not nearly so well defined as they appear when 



examined by the eye. 



is merely a phenomenon in which the thickness and 

 wave-length are both concerned. 



I think that the late Lord Rayleigh was the first 

 to emphasise the fact that optical definition in general 

 depends on the difference of the optical length of the 

 paths of the rays the convergence of which forms the 

 images of contiguous objects. Let A^, Aj (Fig. 3) be 

 two objects in a line making an angle P with the 

 focal plane of the lens, B,, Bj their images, and 

 /j, /^ the conjugate focal lengths. 



