614 ZOOLOGY rkct. 



growth of the animal, as in Ampliioxus, and the division of each 

 by a " tongue " is very similar in the two cases. Further homologies 

 have been suggested by comparing the snout of Aniphioxus with 

 the proboscis or pre-oral lobe of Hemichorda and its pre-oral pit 

 with the proboscis- pore. 



On the whole, although it is by no means certain that the 

 " chordatc " peculiarities of the Hemichorda may not have been 

 independently evolved, it is convenient to retain them in the 

 present phylum, pending further knowledge of their true 

 affinities. 



Various zoologists have supported the view that the nearest 

 relatives of the Chordata among the Invertebrata were forms 

 characterised, like all but the lowest of the former phylum, by 

 the occurrence of metamerism, i.e., Annulata or Arthropoda ; 

 while others have looked upon the metamerism of the Chordata as 

 independently evolved. Among the non-metameric groups 

 which have been supposed to have been ancestral to the Chordata 

 are the Nemertines — the proboscis-sheath being compared with 

 the notochord and the proboscis itself with the pituitary invagina- 

 tion. But this view has received little support. On the other 

 hand, the supporters of the Annulate ancestry have been more 

 numerous — some seeing in the Cha^topoda, others in the 

 Hirudinea, the nearest Chordate relatives. The metamerism of 

 the Annulata is regarded as supporting this theory, as well as the 

 structure and arrangement of their circulatory, nervous, and 

 excretory systems. The notochord might have been supposed 

 to have originated in an Annulate from a typhlosole or from an 

 intestinal diverticulum or siphon, But a very serious difficulty is 

 met with when we proceed to try to derive the Chordate nervous 

 system from that of the Annulata. Not only must the ancestral 

 Annulate, in order to become transformed into the primitive 

 Chordate, have reversed its habitual position, so that its neural 

 surface became dorsal instead of ventral ; but its mouth, which 

 pierced the central part of the nervous system, must have been 

 abolished, while a new one was developed on the hromal side of 

 the head. 



Elaborate comparisons have been instituted between the brain 

 of Cyclostomes and Fishes and those of Crustacea and Xiphosura, 

 and it has been sought to explain the neurocoele as the discarded 

 Arthropod enteric canal. But if Amphioxus and the Urochorda, 

 to say nothing of the Hemichorda, are branches from some low 

 part of the chordate stem — a fact it seems hardly possible to 

 doubt — it is obvious that there can be no direct connection with 

 the highly specialised classes referred to. If, for instance, the 

 lower Craniata sprang either from a Choetopod-like or from a 

 Limulus-like ancestor, Amphioxus and the Tuuicates must either 



