VARYING THE CONJUGATE AREA 129 



distance from the discontinuity which causes the diffraction. It is not 

 surprising, therefore, that phase contrast is produced only close to the 

 boundary of a particle and that the more or less complementary halo 

 is visible only in the neighborhood of the particle. 



In order to study some of the effects resulting from varying the 

 dimensions and transmission of the conjugate area of a diffraction plate, 

 a specimen was prepared which had little internal detail so that the 

 extent of the halo and the area over which contrast is produced could be 

 more easily seen. The specimen was made by shielding some areas of a 

 microscope slide with very thin iron tilings while magnesium fluoride 

 was evaporated on the slide. After the evaporation, the filings were 

 wiped off the slide, and an oil having a refractive index of 1.47 was 

 flowed underneath the cover slide. The specimen then consisted of a 

 number of oil pools of different area surrounded by magnesium fluoride. 

 The photomicrographs reproduced in Fig. III. 7 show some of the oil 

 pools as they appeared under the various conditions of the experiment. 

 As a rule, a sharp shadow was not formed at the edge of the iron filings 

 during the e^'aporation because the filings had not adhered closely 

 to the slide over their entire area. The thickness of the magnesium 

 fluoride film was such that the optical path difference between the oil 

 pools and their surround was equal to or less than 0.065 X (X = 54G1 A) 

 or 23.-1:°. The oil pools will be referred to as the particles in the discus- 

 sion of the observations made on them. The particles, then, differ in 

 area, have an optical path greater than that of the surround, and may 

 vary in thickness close to the edge. If no impurities are present, 

 the edges cannot have an optical path exceeding that at the center of a 

 particle. 



These observations of changes in the image w'hich resulted from 

 varying the dimensions and transmission of the conjugate area in- 

 cluded only diffraction plates that produced dark contrast. One 

 photomicrograph of a bright-contrast image is shown for purposes 

 of comparison. All the diffraction plates in this experiment were 

 made with a coating of magnesium fluoride and Inconel such that the 

 optical path difference between the conjugate area and the comple- 

 mentary area was — X/4 (X = 5461 A). The diffraction plates were 

 separable from the objective and were mounted behind all the lenses 

 so that the same objecti\"e remained in the optical system as different 

 diffraction plates were substituted. A selected Spencer achromatic 

 objective of 4-mm focal length, N.A. 0.66, was chosen. In the manu- 

 facture of such an objective, a separation between two lens components 

 is adjusted to balance the objective as well as possible for spherical 

 aberration. For Fig. III. 7, the separation was chosen to give as good 



