246 NATURAL SCIENCE. Oct., 



simple Auricularia larva, possessing a terminal blastopore, a con- 

 tinuous circumoral (longitudinal) ciliated ridge, with a subjacent 

 central nervous system, and an adoral ciliated band, which extends 

 round the anterior and lateral margins of the mouth, and is drawn 

 inwards and backwards in the mid-ventral line to form a ventral 

 oesophageal ciliated loop. 



The Tornana-\a.rva. of Balanoglossns is equally derivable from the 

 Auricularia-type', and now Mr. Garstang suggests that the true 

 Chordata also trace their origin from the same primitive pelagic 

 form, thus following out in a definite manner the interesting views of 

 Brooks in this direction. (See Natural Science, vol. iii., p. 222.) 



The basis of his theory is the homology which he- proposes 

 between the medullary folds of vertebrate embryos and the paired 

 circumoral longitudinal ridge of Auricularia, the homology being 

 founded on the identity of the relations of these two sets of structures 

 to the mouth, the apical pole, the blastopore, and the central nervous 

 system. In the Chordata th^, ciliated medullary ridges eventually 

 fuse to form the medullary or neural canal, thus enclosing the sense- 

 organs of the apical pole (optic pits) and the blastopore. 



In the Echinoderma, as was shown by the beautiful researches of 

 Semon, the central nervous system becomes segregated at an early 

 period from the ciliated ridges, and migrates ventrally to form the 

 circumcesophageal nerve-ring, while the ridges eventually atrophy 

 and the blastopore persists as the anus. 



An especial importance naturally attaches to the relations of the 

 nervous system and ciliated bands of the developing Balanoglossns. The 

 diagrams exhibited at the reading of the paper, and the new point of 

 view from which the author regarded the course of the nerve-tracts in 

 Balanoglossns, undoubtedly revealed some most significant, if not 

 decisive, evidence in favour of his theory. Unlike his predecessors, 

 Mr. Garstang holds that the dorsal nerve-cord in the trunk of Balano- 

 glossns is not the morphological continuation of the collar- cord, but is 

 — along with the ventral cord — a nerve-tract that is entirely confined 

 to the group Enteropneusta, developed pari passu with the posterior 

 elongation of the body. He holds, on the other hand, that the collar- 

 cord, though ontogenetically median and unpaired, has been phylo- 

 genetically derived by the fusion of two lateral cords, and he finds 

 support for this view in its divarication posteriorly to form the nerve- 

 ring which runs along the posterior rim of the collar. The collar-ring 

 is upon this view homologous with the periblastoporal portion of the 

 nerve-plate in vertebrate embryos, and with the posterior transverse 

 loop of the central nervous system oi Auricularia. 



As to the medullary folds of the collar which give rise to the 

 neural canal, it was pointed out that, according to the researches of 

 Morgan on Tornaria, the medullary folds do not cease at the posterior 

 limit of the collar-cord, but bend round the sides of the body and 

 persist through life as the posterior rim of the collar, the area of separation 



