74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 7/ 



cular thickening of the cuticiila, with the sense cehs differentiated 

 at the inner end of the cup and a crystalHne body formed as a secre- 

 tion in the distal part beneath the lens. These eyes occur in a group 

 on each side of the larval head, frequently at the place where the 

 compound eye is to be developed in the pupa, but it has been shown 

 that the larval eyes degenerate without taking any part in the forma- 

 tion of the compound eye, and there appears to be no ground for 

 the idea that the larval ocelli are ommatidia of the adult eye. So 

 far as known to the writer, the roots of the ocellar nerves have not 

 been traced in the larval brain, and until this is done the status of 

 the ocelli probably cannot be decided. In a caterpillar the ocellar 

 nerve trunks arise from the lower parts of the lateral brain lobes, 

 immediately lateral to the antennal nerves, and follow a long semi- 

 circular course forward and outward to the ocelli. The optic lobes 

 of the adult are developed from tissues within the larval brain. 



The typical frontal or dorsal ocelli of adult insects are without 

 doubt the primitive ocelli of insects since they occur in all the prin- 

 cipal orders except the Collembola and Ephemerida. In these ocelli 

 the hypodermal elements become arranged in two horizontal layers, 

 usually by a process of delamination of the cells where the eye is 

 formed. The cells of the outer layer are the corneagenous cells. 

 Generally they secrete a thick biconvex cuticular lens and then be- 

 come reduced to a thin transparent sheet over the inner layer. The 

 latter consists of the retina cells which become grouped into rhab- 

 dom-forming retinulag. Typically, there are three frontal ocelli, one 

 median and two placed more laterally, but there is evidence in the 

 strvicture and development of the median eye that it is the product 

 of the fusion of an original pair of eyes. Sometimes the median ocel- 

 lus is lacking where the others persist. The lateral simple eyes of 

 adult fleas show by their structure that they are ocelli of the 

 frontal type ; probably they are the paired frontal ocelli that 

 have moved to the sides of the head in the absence of compound eyes. 



The frontal ocelli of adult Ephemeridae are in certain respects 

 similar to the frontal ocelli of other insects, but they dift'er from all 

 other insect eyes in having a multicellular hypodermal lens formed 

 apparently by an invagination of the outer surface of the eye. 



The large, single, lateral ocellus of the larvae of sawflies (Ten- 

 thredinoidea), finally, is somewhat of an anomaly among insect 

 .eyes. In structure it resembles the frontal ocelli of adult insects, but 

 these are eyes that no other larva possesses. On the other hand, it 

 has certain characters that have suggested the idea that it is a pro- 

 totype of the compound eye, a claim disproved, however, by other in- 



