254 ON THE THEORY OF THE MICEOSCOPE. 



gathered from examination of the coloured fringes which appear 

 on the outlines of the object, in the centre and outside circle of 

 the field of vision. But it must be borne in mind that aberra- 

 tions of this kind, as they mostly occur in outline images, have a 

 practical significance only when, daring ordinary microscope 

 practice and with direct light, they appear prominently with a 

 central position of the mirror. 



In the test objects used for the above trials, two precautions 

 are necessary. Firstly, they mast be thin and even, in order that 

 outline and stractural detail may be seen lying in the same level ; 

 secondly, the diffraction pencils must possess a great intensity of 

 light, in order that the effect produced by them may be fairly 

 appreciated in comparison with the effect of the image produced 

 by direct light. For this last reason objects prepared dry, with. 

 vigorous and strong markings, are most suited for tests, as they 

 always give — as maybe seen in the apeitare image — the brightest 

 diffractive phenomena, because the interference of intensely bright 

 rays is required to bring out strong contrasts of clear and dark 

 parts in the microscope picture. 



For low and middle power objective an abundance of suitable 

 objects maybe selected from insect scales and the coarser diatoms. 

 For the higher powers, on the contrary, the selection is limited, 

 because of the necessity of thin and even preparations. The 

 " jS^obert-plate " is unsuited for this kind of trial, since it gives 

 no "absorption image," but a pure "diffraction image," and 

 therefore the most important element for judging of the action of 

 a lens, fails in this case. Fleurosigma angulatum corresponds, 

 perhaps, best, in regard to fineness of detail, to the angular 

 aperture of immersion lenses, and may, in fact, be employed for 

 the highest powers when fragments of delicate specimens with 

 sharply fractured edges are selected, and attention paid to the 

 qaality of the image close to such a fractured edge. But neither 

 the natural edges nor the middle rib offer any certainty of uniform 

 level. To test the high power dry lenses, fragments of coarser 

 specimens of the same object can be used, though the markings 



