56 BARKER : 



In this connection it may be stated that images ma}' be 

 obtained through detached crystalline cones of some insects 

 in any direction with more or less distortion (Plate i, Fig. 5), 

 and an image practically free from distortion is obtained 

 when two of these cones are crossed at right angles, and the 

 image forming rays pass through both in a direction perpen- 

 dicular to the axes. 



Grenacher repeated the experiments of Gottsche, using 

 moths' eyes in which the crystalline cones are firmly attached 

 to the cornea (Plate i, Fig. 4), so that it was possible to 

 remove the soft parts without disturbing the cones. Under 

 these circumstances he found that images were formed zvithin 

 the cones. Exner obtained these images by cutting off the 

 inner ends of the cones and focussing on the surfaces thus 

 exposed, and the writer has a slide in which they may be seen 

 in optical section without amputation.* Exner had previ- 

 ously experimented with the eyes of Hydrophihis, the great 

 black water beetle, but the pyramids always came away from 

 the cornea. He calculated the corneal refractions, however, 

 and concluded that an image on the retina was impossible. 



In 1885 Felix Plateau began a series of experiments with 

 arthropods and other animals, the results of which were com- 

 municated to the Royal Acadeni}' of Belgium in 1887 and 

 1888. The keynote of Plateau's work is found in the quota- 

 tion at the head of this article, which may be translated as 

 follows: "However rational a theory may seem, it has but 

 little value so far as it has not been confirmed by repeated 

 and varied experiments." He finds that nearly all of his 

 predecessors and contemporaries have fallen into error by con- 

 fining themselves to anatomical investigation, neglecting the 

 important testimony of behavior under control conditions, 

 and though we may not accept all of his conclusions we must 

 admire his skill and patience in handling a difficult question, 



*In this case the field of view of each of the unitri embraced an 

 angle of about 20°. 



