2l6 MORPHOLOGY OF THE ORGANS OF VERTEBRATES. 



serve to carry away the sexual cells is also suggestive. These 

 ducts, however, afford some difficulties, as it is not easy to see 

 how the continuous pronephric duct of the vertebrates could 

 have arisen from the separate ducts of the annelid. Again, the 

 ventral nervous chain of the annelid can be closely compared 

 with the spinal cord of the vertebrates, the comparison includ- 

 ing the dorsal roots of the nerves with the spinal ganglia. The 

 same is true of the similarities existing between the transverse 

 blood-vessels of the annelids and the aortic arches and inter- 

 costal vessels of the vertebrates. 



The most plausible hypothesis by which to homologize the 

 anterior portions of the nervous system is that which regards 

 the infundibulum and the nervous portion of the hypophysis as 

 representing the invertebrate mouth, while the vertebrate mouth 

 may have arisen by the coalescence of a pair of gill slits. In 

 this case the ' brain ' of the annelid would be represented by 

 the vertebrate fore brain. In certain annelids there exists a 

 subintestinal tube of entodermal origin which has been doubt- 

 fully compared with the notochord, but as yet no structures are 

 known in the annelids which can be homologized with the gill 

 slits. 



Other but less widely accepted views of the ancestry of the 

 vertebrates are those which would derive the group from some 

 arthropod not far from the limuloids, or from the nemertean 

 worms. It must however be kept in mind that the great- 

 est resemblances between vertebrates arid annelids are directly 

 or indirectly the result of metamerism ; and that it is possible 

 that this vegetative repetition of parts may have arisen inde- 

 pendently in the chordate phylum, and that the similarities 

 noted above may be expressions of convergent evolution, and 

 that the chordates may have descended from non-segmented 

 ancestors. This view receives some support from the fact that 

 metamerism also occurs in the echinoderms, where it could not 

 have been inherited from either annelids or vertebrates. 



In the following pages are numerous references to the lines 

 of descent of the various groups of vertebrates. The adjacent 

 diagram illustrates some of these. Concerning some points 

 there are differences of opinion. Thus the dipnoans are fre- 



