September, 1911. 



KNOWLEDGE. 



355 



general test object, to illustrate different observers" 

 theories. It has also been used to measure the 

 advances, or the supposed advances, in the revealing 

 of structure. Sometimes it has served as a proof 

 that it can offer no evidence of the real structure at 

 all — Dr. Abbe to wit. In the September number 

 of Harper's Magazine, for 19(17, there is an article 

 on photo-micrograph\- with the ultra-violet ra\'s, 

 under which it is claimed that the resoh'ing power 

 of the microscope has been recently doubled. The 

 customary P. angulatiiin is ushered in to make valid 

 this claim. B\- an ingenious and. I am afraid, a 

 somewhat disingenuous arrangement, two photo- 

 micrographs of this diatom are placed side by side, 

 one taken under the ordinary 

 illumination and the other 

 under the ultra-\'iolet light, to 

 prove it. 



Now, somebod\'. somewhere, 

 must ha\-e been taken in at 

 the same time, for this sup- 

 posed new structure was seen, 

 and photographed, by the 

 present writer twent\' \ears 

 pre\'iousl\-. It had not then, 

 and has not now. any special 

 significance for him, except in 

 illustrating his theory of focal 

 images. Neither is it a severe 

 test of the resolving [)()wer of 

 the microscope : an\' oil im- 

 mersion of 1-JO will suffice. 

 The statement is from 

 Germany, which produces some 

 of the best micro-objectives in 



the world, and is also the home of some of the 

 worst micrograph\'. 



Of course, one does not wish to imply that the 

 resolving power of the microscope cannot be 

 increased, or even doubled, by the use of the ultra- 

 violet light, onlv that the e.xample given does not 

 prove it. As far as memory serx-es, the method 

 stated as being employed did not give much promise 

 of future great discoveries. The object was not 

 seen at all, sureh- fatal to accurate and refined 

 definition. To know one's subject thoroughly it is 

 necessary to work out the details under the 

 microscope while sitting down, in order to recognise 

 what to go for, before attaching it to the camera. 

 More practical, the late Dr. Dallinger advocated 

 increasing the resolving power b\' the judicious use 

 of coloured screens. "^'et. after ex[jeriments, he 

 recognised that the objectives should be corrected 

 throughout for an\- particular ray. 



Now. turning tt> the inner side of the valve, 

 enlarged as before, for better comparison, I'^igure 18 

 should give us the ke}- to the much-discussed 

 hexagonal structure. It offers, however, this charac- 

 teristic difference from the corresponding side of 

 P. formosum, that the fibrils or short bars of silex 



Figure 18. The inner 

 X 3770, showing 



instead of being placed lengthways are set obliquely 

 across the valve. It will be seen also that the zig- 

 zag arrangement of the isolated strip, if joined to the 

 broken edges bordering the clear spaces on each side, 

 would require no crossbars to complete the so-called 

 hexagons. .Another peculiarit}" is that this seems to 

 be real structure in which there are no focal images 

 to obscure it. The analogw in fact, would be more 

 with the hexagonal middle structure of the large 

 discoid and other forms. 



There still remains the question of the intercostals, 

 which Mr. E. M. Nelson maintains are the products 

 of the second order of spectra only. If these are 

 not real entities, then why are the\- produced trom 

 this one side of the valve 

 alone ? The structure on both 

 sides is of equal fineness : 

 given the same conditions of 

 aperture and lighting, should 

 there not be produced the 

 same results ? However, per- 

 haps it is not well to express 

 too pronounced an opinion on 

 such a matter, ^^"e are dealing 

 here with appearances so 

 minute that six dots are 

 arranged around one that once 

 taxed the whole powers of the 

 microscope to define. 



I'igure 19 is from a valve 

 of P. haJticiim. still showing 

 fibrils and also indications of 

 the structure underneath. The 

 appearances here have been 

 put down to moisture in 

 dr\- mount : but. as it happened, Mr. 

 E. M. Nelson exhibited the same diatom under 

 the celebrated new objective of 1-6J N..\., pres- 

 ented to the Royal Microscopical Society by 

 Zeiss. The object then was of a necessity 

 mounted in a dense medium, yet in all respects 

 it offered the same details as are figured here. 

 So much for moisture in the mount. It will be 

 seen that the fibrils differ to a certain extent 

 from the preceding examples. They appear in 

 this print as extended bars, with swellings at 

 regular intervals, seemingly, when together, gix'ing 

 the semblance to beads or squares according to the 

 aperture employed. This difference can be ascribed 

 to one of two causes. It mav arise from the true 

 fashion of the structure in itself, or from imperfect 

 resolution. Some countenance is given to this last 

 view bv referring back to F"igure 11*, taken with the 

 dry lens, where it will be seen the fibrils come out 

 almost straight, though b\- no means so when under 

 the wider aperture of an oil immersion. E\-ery 

 earnest student of Nature knows that often we arrive 

 at a point where we stand groping helplessl}' outside 

 the boundarv of knowledge. We mav hazard a guess 

 as to the right wa\' in ; but there our efforts must end. 



side of P. angulatiiiii 

 the isolated sUp. 



the 



"See "Knowledge" for August, 1911, page 293. 



