ELEVATIONS AND DEPRESSIONS, ETC. 221 



The principal planes of the groove-shaped portions of the mem- 

 brane, from which the virtual images are at equal distances, do not 

 therefore lie in the same level. For instance, under the relations 

 represented in Fig. 121 for the grooves concave upwards, they 

 coincide with the line a I, and for those alternating with them with 

 c d. The virtual images themselves lie therefore alternately higher 

 and lower, and since they act as microscopically small sources of 

 light, the different colours which are observed with higher and 

 lower focal adjustment may hence be explained. As an example, if 

 the instrument is focused to a plane that lies higher than the 

 lower, and lower than the upper virtual images, the latter will 

 appear bluish, the former reddish. 



4. ALTERNATE SOLID AND AQUEOUS LAYERS. 



Aqueous layers which alternate with solid ones, when viewed 

 vertically, act as attenuated places or crevices in a homogeneous 

 substance. Equivalent effects are obtained if the ratios of re- 

 fraction between the solid and the aqueous layers are the same as 

 between the homogeneous substance and the fluid occupying the 

 crevices. The microscopic image exhibits, therefore, just as with 

 fibrous thickenings, alternately bluish and reddish lines, corre- 

 sponding to the virtual and real images of the portions which are 

 not exactly in the plane of adjustment. The difference in bright- 

 ness, which is produced by alteration of the object-distance at 

 definite image-points, as well as the intensity of the shadows, 

 offer to the practised eye important data for estimating approxi- 

 mately the differences in density. To distinguish without further 

 investigation, when in a given case actual crevices or merely 

 aqueous layers are present, is theoretically impossible. 



5. ELEVATIONS AND DEPRESSIONS AS OPPOSED TO DENSE AND 

 LOOSE LAYERS. 



Since elevations, as already pointed out, act optically exactly 

 as corresponding thickenings of the substance, the solution of the 

 question, whether fine markings (such as occur in diatoms, striated 

 cell-membranes, &c.) are due to the differences of density or of 



