490- 



NA TURE 



[Fkbruary 25, 190Q 



method differs from all others. It would happen with 

 a powerful source of light, unless the apparatus were 

 thus arranged, that so inany small particles would be 

 illuminated that the diffraction discs would overlap 

 and simply produce a general diffusion of light, and 

 but very few individual particles, and of these only the 

 more isolated ones, would therefore be visible. This 

 is exactly what happens with any ordinary method of 

 dark ground illumination where the particles are 

 numerous or where they are distributed throughout 

 a considerable area. In the Siedentopf method, where 

 the light is simply concentrated on one spot in the 

 field alone, little or no interference between the dif- 

 fraction discs results, internal reflections between the 

 components of the optical system are reduced to a 

 minimum, and consequently an image is obtained 

 which under other conditions would be impossible. 



Considerable developments have taken place recently 

 in methods of dark ground illumination. It is now 

 possible to view such objects as bacteria with ease, 

 without resorting to the excessive staining that is 

 unfortunately so prevalent, enabling them also to be 

 observed while in the living condition. The most 

 satisfactory of these is the parabolic illuminator 

 recently introduced by Messrs. Zeiss (Fig. i). It may 



easily be said, and 

 , ♦ , it is perfectly 



true, that the 

 parabolic illum- 

 inator, either dry 

 or oil immersion, 

 is no novelty, but 

 the one made bv 

 Messrs. Zeiss has 

 been developed 

 along scientific 

 lines, and is the 

 result of careful 

 computation, 

 whereas those 

 ^^'^- I- m a d e i n t h e 



earlier days of 

 microscopy were largely the result of chance, or at 

 least trial and error, and it was a fortunate cir- 

 cumstance if they gave a result which was entirely 

 satisfactory. 



In the case of the Zeiss parabolic illuminator, the 

 light is so reflected from the internal surface of the 

 paraboloid that the annular cone of rays is projected, 

 and has its focal point exactly where the objective is 

 focussed. It would not, of course, give the same 

 results as Siedentopfs method with such objects as 

 ruby glass or colloidal solutions, but for observing 

 minute living bacteria or similar transparent objects 

 it leaves little to be desired. The illuminating ravs, 

 too, are exactly confined within the limits of a numer- 

 ical aperture of I'l to i'4, so that, if using, say, a 

 4-millimetre apochromatic objective with an aperture 

 of o'95, no rays would enter directly, and it would 

 only be those reflected or refracted by the object that 

 would pass into the objective at all, the objects, in 

 fact, behaving as self-luminous bodies. This is a 

 definite improvement on the method of introducing a 

 stop into the substage condenser, so that the central 

 rays are blocked out, and only the peripheral rays are 

 allowed to pass, as there is much less spherical and 

 chromatic aberration, the image being to a large 

 extent dependent on reflected light. This appliance 

 will render particles visible that might be termed ultra- 

 microscopic, and in any solution or preparation of 

 bacteria in water a great number of diffraction discs 

 will be visible that by ordinary direct light could not 

 be seen. 



NO. 2052, VOL. 79] 



.Another method that fulfils its purpose is the reflect- 

 ing condenser made by Messrs. Leitz, of Wetzlar. 

 In this, two reflecting surfaces, the one internal and 

 the other external, as shown in Fig. 2, are so shaped 

 as almost completelv to unite the rays at a point P. 

 The light enters from below, and takes the direction 

 as shown bv the dotted lines, ultimately converging 

 on the point P, which is the position of the object, 

 and is the focal point of the objective. It is obvious, 

 therefore, that there is no chromatic or spherical 

 aberration. The adjustments for centring are exactly 

 the same as for an ordinary substage condenser, and 

 the optical portion is contained in the mount that 

 slides into the ordinary substage carrier. This ap- 

 paratus, as well as the Zeiss, requires that the object- 

 slide and the cover-glass shall be of a certain 

 thickness, and cedarwood oil is used between the 

 top surface of the condenser and the slide. It may 

 be used as shown with any dry lens, but the best 

 results are obtained with an apochromat, especially 

 with the 4-millimetre 0*95 N..^. The cone of the 

 illuminating rays is confined within the same limits 

 as the Zeiss apparatus. 



It is necessary to remember, however, that because 

 a particle that is invisible by axial illumination 

 becomes 

 visible by ob- 

 lique light, it 

 does not, 

 therefore, fol- 

 low that it is 

 ultra - micro- 

 s c opi c. Its 

 t r a n sparency 

 may be too 

 great, or iK 

 refractive in- 

 dex may toi. 

 nearly coincid( 

 with the me- 

 dium in which 

 it lies for it to 

 be visible by 

 direct light. Fic. 2. 



By oblique 



illumination a much smaller difference in refractive 

 index between the object and its medium will be 

 sufficient to form an image. 



It is very difiicult within the limits of a short 

 article such as this to go into the matter sufficiently 

 thoroughly to deal with all the points at issue. It 

 may easily be urged that particles that are ultra- 

 microscopic can be made visible by methods other 

 than those described. It is well known that even 

 passing a very powerful beam of light through a 

 darkened room will render visible a large number of 

 particles that cannot be seen by ordinary methods of 

 illumination, and it is more than probable that many 

 of the particles so observed are, in fact, ultra-micro- 

 scopic. Faraday was able, by projecting a powerful 

 beam of sunlight through a piece of ruby glass, to 

 view the diffraction discs arising from the gold 

 particles in the glass without any microscope at all. 

 The whole question resolves itself really into the 

 necessity of having primarily a sufliciently strong 

 source of light, and so to arrange the incidence of the 

 light that only those particles in the field of view are 

 illuminated. In many arrangements that have been 

 described for microscopic illumination, these two con- 

 ditions have not been combined, and it is only recently 

 that it has been thoroughly realised that either one 

 without the other will not give the looked-for result. 



J. E. B.^RNARD. 



