382 



stain slightly less than those found elsewhere which latter are pressed 

 together and are of a long ellipsoidal shape (Fig. 2). 



These three groups of larger nuclei form the retinal portion of the 

 three lateral eyes. They are spindle-shaped or rather wedge-shaped, 

 narrower towards the exterior surface. Later they are separated from 

 the surrounding nuclei at the narrower end hy a ring-like furrow, 

 peculiar to the lateral eyes, and the surrounding nuclei grow and 

 meet together over the retinal portion forming a layer of cells — the 

 vitreous body. 



The lateral eyes receive their nerves from a portion of the brain, 

 formed by the lateral vesicle. Thus the latter is the optic ganglion, 

 formed from an invagination independently of the semicircular cephalic 

 groove which gives rise to the brain proper. 



The common hypodermic thickening of the lateral eyes of spiders 

 is most probably homologous to the hypodermic thickening of the 

 lateral compound eye of Limulus 3 , as the position is just the same and 

 an invagination similarly produced in both cases. Then, are the 

 peculiar groups of larger nuclei of the former homologous to the 

 ommatidia of the latter? I am inclined to believe that it is the case 

 and the lateral eyes of spiders are separated, enlarged, and 

 modified ommatidia of a compound eye of their ancestor. 



In PI. IV illustrating Parker's paper «The Eyes in Scorpions« 

 we distinctly see that all the lateral eyes of the scorpion also arise from 

 a common thickening of the hypodermis. Thus we see that this 

 interesting phenomenon does occur not only in spiders, but also in an 

 allied order Scorpionidea. Parker, however, does not seem to con- 

 sider this interesting fact as of much value, if he did not indeed 

 overlook it. 



Lankester and Bourne arrived at the same conclusion as mine 

 from the study on the structure of the eyes of adult Limulus and 

 Scorpio; but they wanted the embryological proof. After these authors 

 much work has been done on the development of spiders, scorpions 

 and Limulus] but no one has confirmed the conclusion of Lankester 

 and Bourne. 



The facts that the number and the relative position of the simple 

 eyes of spiders and scorpions are very much variable and that they are 

 placed more close together in the embryonic than in later stages, speak 

 in favor of my conclusion. 



3 The detailed description of the development of the lateral eye of Limulus I 

 reserve to a paper on the embryology of our king-crab, which will be published in 

 the course of time in the Journal of the College of Science, Imperial University, 

 Tokyo, Japan. 



