NEAL: NERVOUS SYSTEM IN SQUALUS ACANTHIAS. 239 



later stages, as does its ventral root, the hypoglossus of the adult having, 

 according to Gegenbaur ('72) and M. Fiirbringer (97), only two ventral 

 roots without dorsal ganglia. Since the reduction of dorsal and ventral 

 roots takes place from the anterior towards the posterior, these two ven- 

 tral roots of the adult hypoglossus are in all probability the posterior of 

 the five roots of the embryo.^ A similar process of fusion of dorsal ganglia 

 with the dorsal ganglion of the vagus takes place in Petromyzon ; but in 

 that animal the fusion of the ganglion — viz. that of the " spinalartige 

 Vagusanhang," which for reasons already stated by me ('97, pp. 454, 

 455) I regard as the exact homologue of the dorsal ganglion of van 

 Wijhe's eighth somite in Squalus ^ — appears by a comparison of the 

 results of Wiedersheim ('80), Schneider ('80), Ahlborn ('84^), Hatschek 

 ('92), KupfFer ('96), and M. Fiirbringer ('97) to be a variable one. 

 This union of dorsal spinal ganglia with the ganglion of the vagus, taken 

 in connection with the fact previously stated by me ('97, p. 453), that 

 the dorsal ganglia of the glossopharyngeus and vagus lie primitively * 

 median to the dorsal portion of post-otic somites, is a further link in 

 the chain of evidence which shows that no fundamental distinction be- 

 tween spinal and cranial nerves exists. From the foregoing account it 

 will be seen that, as in the case of spinal nerves, we are able, using as 

 criteria the central and peripheral relationships of the motor fibres, to 

 divide cranial motor nerves (roots) into two classes, viz. : (1) dorsal 

 (splanchnic) roots, having their nucleus in the lateral horn of the neural 

 tube and their peripheral distribution in the musculature (ventral) of 

 the visceral arches; and (2) ventral (somatic) roots, which have their 

 nucleus in the ventral horn of the neural tube and their peripheral 

 distribution in the musculature (dorsal) of the somites (somatic muscu- 



1 Hexanchus and Heptanchus both have five hypoglossus roots in the adult 

 (M. Fiirbringer, '97). 



■^ Homologized, however, by Ahlborn ('84") with van Wijhe's tenth somite, and 

 by Hatschek ('92) with van Wijhe's seventh somite. 



^ Goronowitsch ('92) first observed in the chick that the topographic relation of 

 the vagus to the head somites is the same as that of the spinal nerves to tlie trunk 

 somites. Sewertzoff ('95, p. 92) also states that "Die Beziehung der Kopfmyotome 

 zu den Kopfnerven, z. B. zuni N. vagus ist dieselbe, wie diejenige zwischen den 

 Rumpfmyotomen und Riickenmarknerven, d. h. sie liegen nach aussen von Xerv 

 (Cyclostomata, Ganoidei chondrostei, Urodela, Reptilia, Aves)." This is stated, 

 however, by Kupffer ('94, '96) not to be a primitive relation of the post-otic myo- 

 tomes in Petromyzon. My own observations and conclusions, however, differ from 

 those of Kupffer (see Neal, '97. p. 453). Miss Piatt's ('97) observations on Nec- 

 turus, and her conclusions likewise, confirm the conclusions of Goronowitsch and 

 Sewertzoff. 



