neal: nervous system in squalus acanthias. 211 



and it is certainly not true for Squalus, and not true, so far as I am 

 able to determine, for Amblystoma. In Petromyzon, however, as has 

 been previously stated by Scott ('87) and Shipley ('87), the spinal gan- 

 glia lie opposite the constrictions between the somites (in later stages 

 opposite the myosepta).^ Thus, inasmuch as the dorsal nerves of Ammo- 

 coetes are intersomitic and never uuite with the ventral nerves which 

 are somitic in position, and inasmuch as the dorsal ganglia show close con- 

 nection with the ectoderm in early stages of development and lose this 

 connection during development, the spinal nerves of this animal form 

 a natural transition from the nerves of Amphioxus to those of Squalus 

 and higher Vertebrates. For in Amphioxus ventral nerves are somitic 

 in position, dorsal nerves intersomitic, and the connection of the ganglia 

 of the latter with the skin is retained throughout life.^ Two chief causes 

 seem to have brought about the change in the relations of the dorsal 

 spinal nerves in the Vertebrate series. The first cause appears to have 

 been the great dorsal and anterior extension of the trunk myotomes, and 

 the second cause the posterior extension of the ramus cutaneus dorsalis 

 vagi (ramus lateralis vagi), which takes the place of the rami cutanei of 

 the spinal nerves. The physiological reason for the extension of the 

 vagus is to be found in the advantage obtained from the centralization 

 of sensory impulses in the brain. With van Wijhe ('92), Hatschek ('93), 

 and M. Flirbringer ('97), I accept the theory of Prochaska, Sommering, 

 and Gegenbaur that cranial and spinal nerves are hemodynamic, and the 

 view of Hatschek ('92) that dorsal and ventral nerv'es primitively alter- 

 nated with each other.^ Of these, the former were mixed in function and 

 the latter motor, as in Amphioxus. 



c. I^ERVE Relations ix the Cephalic Region of S. acanthias. 



In the head, where the nerve relations are much more complicated, it 

 will be necessary to trace the development of the nerves in different 

 stages. The series represented in Figures 7 to 21 (Plates 3 and 4) is 

 intended to show the changes which the neural crest (colored in blue) 

 undergoes, and likewise to show the development of the brain vesicles 



1 Because of this relation to the myomeric constrictions in Ammoccetes and tlie 

 relation of the ganglia to the expansions of the spinal cord (mj-elonieres) deducible 

 from it, it is obvious that not very great morphological value can be given to the 

 fact that in Squalus the ganglia lie opposite the constrictions of the spinal cords. 



2 I hold with Hatschek ('92) and M. Flirbringer ('97) that in Amphioxus the 

 homologues of the dorsal ganglia of Craniota are found in the cell groups at the 

 place where the dorsal nerves meet the skin. 



3 See also Ransom and Thompson ('86). 



