J. D. Tot/i ill — The Ancestry of Insects. 



381 



('96) supposes that the head of Triarthrus is made up of at 

 least six segments and possibly of seven ; Jaeckel ('01) finds 

 at least six head segments in a series of trilobites and supposes 

 a total of eight. These findings cannot be taken too seriously 

 for the reason that the developing trilobite brain has not been 

 examined — and it is in the ontogeny of the head that the key 

 to the same problem in other groups (Arachnida, Hexapoda, 

 Vertebrata) has been found. They are, however, highly sug- 

 gestive and it is probable that trilobites had at least as many 

 head segments as insects, perhaps the same number. 



.Returning to Triarthrus, the long antennae are instructive 

 as are also the compound eyes. 



Fig. 6. 



Fig. 6. a, b, Olenellus thompsoni ; c-e, O. gilberti ; f, Mesonacis vermon- 

 tana. Showing specialization of first three post-cephalic segments. (From 

 Grahau and Shimer, after Waleott.) 



Lindstrom (1901) among others has paid special attention to 

 the eyes of trilobites and finds at least three kinds — isolated 

 eyes or ocelli, aggregate eyes of biconvex lenses, and compound 

 eyes. The three kinds of insect eyes are roughly comparable 

 and it is conceivable that they may have been derived from 

 those of trilobites. 



In Olenellus (fig. 6) there is a regional specialization of the 

 first three post-cepbalic segments. In many trilobites, such as 

 Aloertella, Ceraurus, there is a tendency toward specialization 

 of the caudal extremity by the formation of caudal spines ; in 

 Neolenus serratus, whose appendages have been recently dis- 

 covered by Waleott on specimens from the Burgess shale 

 (Middle Cambrian), there are jointed caudal rami strongly 

 suggestive of insect cerci. 



