36 



to the fact that both of these condensers do not focus at a sharp point, 

 but rather build up a patch or fleck of light,' no difficulty is experienced 

 in the centeruig of the condenser, and although the thickness of the 

 microscope slide or observation chamber may be varied slightly 

 (0-3 mm.) this does not seriously interfere with the observation. 

 The illumination used is either an inverted incandescent gas mantle 

 or a Nernst lamp. For the rapid examination of a series of solutions 

 the paraboloid is not so convenient as the Zsigmondy arrangement 

 since the microscope setting must be distrubed after each reading 

 and the slide and cover glass most carefully cleaned. For use with 

 the Reichert instrument a small observation cell has been constructed 

 through which a series of solutions may be passed. 



3. Stokes' Law. — Although useless for direct quantitative work, 

 Reichert and Paraboloid ultra-microscope when arranged horizontall3^ 

 are well suited for the purpose of determining the radius of sub- 

 microns from their speed of settlement by the appUcation of Stokes' 

 Law (see above under Microns). The method cannot conveniently be 

 used for submicrons of less than 20yLt./x diameter since calculation 

 shows that gold jjarticles of this size take seven hours to fall 0-1 mm. 

 By super-imposing on gravity an electrical field, the speed of settle- 

 ment is increased and Burton^ by tliis means has applied the Law 

 to the " weighing of the particles " (c./. also Westgren'^^). 



4. Density of Distribution. — These mstruments may also be used 

 to count the number of particles n and n° of radius r density p in 

 equilibrium at heights o and h. The radius is then calculated from 

 Perrins^i equation (c./. also Oden^*) — 



, n° , , h r3 10" 



2. Double Reflecting Condensers. — The Siedentopf Cardioid^^ and 

 Jentzsch^* are practically free from spherical aberration and are 

 almost aplanatic. By means of a micrometer eye-piece the area of 

 an illuminated layer of fixed depth — shghtly moe than the depth of 

 vision — is determined and the volume calculated. This volume 

 cannot be varied as in the slit or immersion ultra-microscope (see 

 later), and the determintions are not so reliable. On the other hand 

 owing to the increased intensity of the illumination it is possible to 

 measure particles which are too small for counting in the slit ultra- 

 microscope, but which can be coimted in the new Zsigmondy*^! 

 immersion apparatus. The Jentzsch or Cardioid are particularly 

 suitable for the exammation of thread-like bacteria of microscopic 

 length but ultra-microscopic breadth. Such bodies in the sUt or 

 immersion ultra-microscope appear as points of fight. Siedentopf^^ 

 showed that this was due to filumination from one side only, and that 

 the appearance of such bodies depends on the azimuth of the incident 

 light. Owing to the fact that the iUumination in all reflectmg 

 condensers is concentric, the complete structure of such bodies is 

 revealed. Snice the rays are brought to a sharp focus it is essential 

 that double reflectmg condensers should be accurately centred and 

 that the special slide, ceU, and cover be " ultra-microscopicaUy clean." 



