DIFFRACTION THEORY UNIVERSALLY APPLICABLE 75 



is in this case confined to the proportion between the bright and the 

 dark interspaces of the striation and to the appearnnce of the con- 

 tours of the stria\ 



' If not more than the said tiro rays of the total diffraction fan are 

 admitted, the dark and the light intervals are always shown of 

 approximately equal breadth, even if the real proportion of both 

 intervals differs much from 1:1; and the dark and bright stria? show 

 always gradually increasing and decreasing brightness; in other 

 words, want of distinct contours. 



'This phenomenon shows the typical picture of every regular 

 striation for the depiction of which not more than two diffraction 

 rays can be utilised. For example, Amphipleura pellitdda, or any 

 other striation which is near to the limit of resolution for the optical 

 system in use, and, therefore, even with oblique light, brings only 

 one diffracted beam into the objective. 



ii. ' Whenever a structure gives rise to a diffraction fan of con- 

 siderable angular extension, the observation with a centra] incident 

 beam or axial light may lose a greater or smaller portion of the 

 whole diffracted light if the angular expansion of the fan extends to 

 the aperture of the objective in use. But oblique illumination must 

 always involve a loss, and this loss is not confined to the external 

 (peripheral) rays of the diffraction pencil (as is the case in central 

 light), but the portion lost will more and more extend to one full 

 half of the whole when the obliquity is gradually increased to the 

 utmost limit, so that the direct beam touches the edge of the aper- 

 ture. It follows that the images which are obtained with oblique 

 light will always be incomplete and not similar to a geometrical 

 projection of the object ; and generally (though not always) more dis- 

 similar than those by central light in regard to the minuter details. 



Strictly similar images cannot be expected, except with a central 

 illumination with a narrow incident pencil, because this is the 

 necessary condition for the possible admission of the whole of the 

 diffracted light.' 



Let it be noted that these principles of the diffraction theory of 

 microscopical vision relate to structures of all kinds, whatever may 

 be their physical and geometrical composition. Irregular structures, 

 isolated elements of any shape, equally produce diffraction effects, 

 observed either by transmitted or reflected light, and being eitl i r 

 transparent, semitransparent, or opaque. 



The value of a = n sin indicates the number of rays which 

 an objective can admit ; the aperture equivalent measures the very 

 essence of microscopical performance. It measures the degree in 

 which a given objective is competent to exhibit a true, complete 

 delineation of structures of given minuteness, and conversely the 

 proportion of a in different objectives is the exact measure of the 

 different degree of minuteness of structural details which they can 

 reach, either with perfect similarity of the image or with an equal 

 degree of incompleteness of the image, provided that the purely 

 dioptrical conditions are the same. 



'Resolving' power is thus a special function of aperture. The 

 limit of visible separation in delicate structure and striation is 



