ARTHROPODAL VISION. 51 



tracing paper, and support it beneath the stage. It will 

 be advantageous to arrange it' to permit rotation, but that 

 is not indispensable. Now illuminate the design with the 

 strongest beam of light obtainable, and focus upon the 

 mounted fragment of cornea (this having been placed upon 

 the stage, of course, in the usual position), using a moderate 

 or low power. The hexagonal facets of the posterior or inner 

 surface of the cornea will now be seen (Plate i. Fig. i), and 

 on focussing downward the anterior surface will come into 

 view. But now rack up slowly until the facets disappear, 

 and when a certain plane a small fraction of an inch above 

 the object is in focus, the images of the design beneath the 

 stage will appear, one complete image above each facet in the 

 field of the optical combination in use (Plate i. Fig. 2). 



If the prepared design is such as to show change in orient- 

 ation it will be observed that the images are not inverted as in 

 Leeuwenhoek's experiment, and this is merely because 'the 

 compound microscope erects the inverted images formed by 

 the corneules at their foci, which the simple microscope did 

 not do. 



Certain measurements made by the writer may be tabulated 

 as follows, the subject being the dragon-fly's eye : — 



Distance of design below the stage, 2.37 inches 



Diameter of design 187 



Diameter of a cornenle 003 " 



Focns of a cornenle 012 " 



Diameter of image 0015 " 



These figures will indicate the order of the dimensions 

 with which we have to deal, though there is much variation 

 even in the same eye. 



But these experiments or demonstrations do not throw 

 much light upon the nature of arthropodal vision. Many of 

 the creatures of this group have only simple eyes, called 

 ocelli, and resembling, superficially, the eyes of vertebrates. 

 Some have both simple and compound eyes, some compound 

 eyes only, and some no eyes whatever — yet they see. 



