190 



PROGRESS IN MICROSCOPY 



As shown in Fig. 7.10, the shift amounts approximately to one third 

 of the space between two fringes. Then, K' = 1/3. However, the 

 shift d' does not necessarily equate the actual path-difference ^ brought 

 in the object as, in monochromatic light, all fringes are identical. 

 Now, were the object-originated path difference to equate a wave- 

 length whole number /?, nothing would be visible in monochromatic 

 light. Were a shift d' = K'a, smaller than a fringe inter-space, to be 

 observed, then, the actual shift may equate a whole number times /., 

 viz., pX, plus the fraction A". The actual path-difference S would be: 



d =(n'-n)e =pk + d' =pX + K'X = KX 



(7.8) 



In Fig. 7.10, the path-difference d' is the actual path-difference if, 

 in the case of the image A[, p is zero and the shift is to be located 

 in relation to the right-hand fringe. How to eliminate these difficulties 

 is shown later. 



In the above-described method, the fringes are at right angles 

 to the direction of duplication. Such a direction is not too con- 



a, JO J 



Fig. 7.11. Lengthwise extended object at right angles to the direction of duplication. 



venient for carrying out measurements when the object extends over 

 some distance in a given direction or when the straight edge of 

 a phase-shifting object is being investigated. Such a lengthwise ex- 

 tended object appears as shown in Fig. 7.11: owing to the shift, it 



Fig. 7.12. Transmitted wave by a straight edge of a dephasing object. 



may occur that no dark fringe develops in the two images. The 

 straight edge of a phase-shifting object gives rise to a wave similar 

 to the one shown in Fig. 7.12 or looks like the Fig. 7.13: devoid of 



