608 INVERT BERATE MORPHOLOGY. 



be postponed for the present, and attention called to the suggestive char- 

 acter of the Tornaria. Its first describer took it for an Echinoderm larva, 

 and the majority of succeeding authors have been inclined to regard it as 

 indicating aifiuities with that group. The arrangement of the praeoral and 

 postoral ciliated bands, and the occurrence of the proboscis-pore, suggest 

 the Echinoderm larva without doubt, but it must still be regarded as a 

 decidedly open question whether or not these features indicate an aflSnity. 

 Further information is required both in regard to the ancestry of the 

 Echinoderms and as to the life-histories of the Pterobranehia, before the 

 question can be settled. 



Another line of ancestry must also be mentioned, namely, one which 

 leads back to ancestors common to the Hemichordates and the Prosopygia. 

 The similarities of the Pterobranehia to the Polyzoa are striking, there 

 being the same bending of the intestine, similar lophophorelike tentacular 

 structures, and, what is of considerable importance, a dorsally situated 

 nervous system arising as an invagination of the ectoderm. A further 

 point perhaps of some importance may also be mentioned, i.e., the occur- 

 rence of three sections in the body-cavity of the Brachiopoda. In following 

 out the line of descent suggested by these similarities, we are, however, 

 quickly brought to a halt by the uncertainty connected with the origin of 

 the Prosopygia, and we are left standing between two lines, one leading 

 back to the Prosopygia and the other to the Echinoderms. Whether or not 

 these two lines converged to common ancestors in pre-Cambrian times can- 

 not be ascertained, and the solution of the problem must be left to future 

 embryological investigations. 



II. Class Cephalochokda, 



The class Cephalochorda contains a single genus, Amphi- 

 oxus (BrancMostoma), which is exclusively marine in habitat, 

 being found buried in an upright position in the sand, the 

 anterior end of the body alone showing at the surface. 



The body in all species is elongated (Fig. 278) and some- 

 what flattened from side to side, and bears along the mid- 

 dorsal line an unpaired fin, formed as a fold of the body-wall, 

 and containing a cavity traversed by numerous skeletal rods 

 (see Pig. 279) which serve as a support for the fin. Poste- 

 riorly it becomes somewhat higher, and forms a caudal fin 

 surrounding the posterior end of the body, while on the ven- 

 tral surface two fins run forward a short distance, both these 

 and the caudal fin being supported by fin-rays. Some dis- 

 tance from the hind end of the body on the left side of the 

 caudal fin is situated the anal opening (Fig. 278, a), while in 



