36 



Thomae** — ^is provided with quartz windows at right angles and 

 held in a special clip so that the condenser and water immersion 

 objective are directly opposite the windows. On account of the 

 Brownian Movement it is more difficult with Uquids than with 

 soUds to count the particles in a known Ulurainated volume. The 

 solution should be so diluted, that at the same time, about four 

 particles only are visible and ten momentary counts are made. The 

 solution in tlie cell is changed ten times, counted each time, and the 

 mean of the 100 counts is found. A control observation should be 

 made with a different dilution. With experience the total time 

 required for a determination is about 30 minutes, and it is recom- 

 mended by I^ng-", that for rajDid work, momentary illumination by 

 means of a pendulum be used, and the momentary observations 

 recorded on a tjqjevvriter. With this apparatus hydrosol submicrons 

 of 2/x to 5/x/x can be measured, provided the refractive mdex* of the 

 particle is sufficiently different from that of the water — a proviso 

 which applies to all ultra-microscopic observations. 



Shortly • after the introduction of the " sUt " ultra-microscope 

 other systems of dark ground illumination, well known to earlier 

 Enghsh microscopists, were used for the examination of colloidal 

 solutions. These systems differ from the Zsigmondy arrangement 

 in that the Uluminating beam is coaxial with the microscope, and is 

 brought to a focus in the coUoidal solution by means of a special 

 substage condenser, provided with a central stoj), so that the solution 

 is illummated from all sides by rays comprised \\athin the apertural 

 zone of 1-0 to 1-3 NA. It is evident that the value of the apertural 

 zone need not be greater than the refractive index of the medium, 

 e.gf., for hydrosols 1-33. The earUer condensers of this type are oi 

 historical interest only, as also are the Cotton and Mouton*, Scarpa^*, 

 and modmed Abbe condensers used with centre stop objectives of 

 high numerical aperture. Reference should be made to the useful 

 summary by Burton^. 



The substage ultra-condensers suitable for colloidal solutions or 

 the observation of living and unstained bacteria may be classified 

 as (1) single reflecting, Reichert and the Zeiss Paraboloid^' ; (2) double 

 reflectmg, Leitz^^-Jentzsch^^ ; and Zeiss Cardioid^^. 



1. Single Reflecting Condensers. — The Reichert and so-called 

 paraboloid! although different in principle have both the same uses 

 and Umitations. They are suitable only for comparatively coarse 

 hydrosols, and camiot be used for dkect quantitative estimations, 

 since both show spherical aben-ation and in the paraboloid the images 

 produced by the different zones vary in size. Consequently the 

 intensity of iUumination decreases from the centre of the field. 0\ving 



* The intensity of scattered light 



Ig oc 



[^'J 



where fi is the refractive index of medium, and /n, of the submicron. 



t Siedentopf modified the shape of the true Paraboloid of Wenham (1584) 

 and so made the technical reproduction of the condenser cheaper and less 

 tediousi 



