NEAL: XEETOUS SYSTEM IN SQUALUS ACANTHIAS. 257 



(5) lu Torpedo it innervates musculature (muse, rectus posterior) de- 

 rived from two somites, viz. van Wijhe's third andfourth (Sewertzoff, '98). 

 I am not able, however, to offer direct evidence that the nerve has part of 

 its nucleus in eucephalomere IV. I am therefore not able to exclude 

 the possibility that the ventral root of a post-otic somite has been substi- 

 tuted for the pre-otic veutral nerve which once innervated somite 3. 

 That such a substitution of the fibres of a ventral nerve of one segment 

 for those of another may take place ontogenetically, I have the following 

 evidence. I find that in a Squalus embryo of 50 mm. the ventral nerve 

 of van Wij lie's 7th somite has become very rudimentary, while fibres 

 fi'om the ventral nerve of the 8th somite extend to the musculature de- 

 rived from the 7th somite, which in this stage forms the most anterior 

 segment of the lateral musculature. Now, if the ventral root of the 7th 

 somite atrophies before the adult stage is reached, and if the muscula- 

 ture derived from this somite remains the first segment of tlie lateral 

 trunk musculature of the adult, as has been stated by van Wijhe ('82) 

 and Hoffmann ('94), the conclusion seems unavoidable that we have to 

 do here with a substitution of a posterior nerve for one farther anterior. 

 Moreover, in Petromyzon we have evidence that the first five post- 

 otic myotomes of the lateral trunk musculature are innervated by the 

 ventral nerves of the last two of the corresponding somites, i. e. the 4th 

 and 5th post-otic, which in my opinion are homologous with the 4th and 

 5th post-otic somites of Squalus (van Wijhe's Sth and 9th). Here also 

 the conclusion seems to me to lie warranted that there has been a phy- 

 k'genetic, if not an ontogenetic, substitution of the nerves of posterior 

 segments for those of more anterior segments.^ "We may therefore 

 infer, with a considerable degi'ee of probability, that a similar substi- 

 tution of a post-otic nerve for a pre-otic one may have occurred phy- 

 logenetically in the case of the abducens. Such evidence, however, 

 seems to render unwarrantable the assumption of a primary and in- 

 separable connection of motor nerve and muscle. Furthermore, the 

 evidence that the motor nerves develop as axis-cylinder processes of me- 

 dullary cells given by His ('89) for spinal nerves, and by myself in this 



^ See Neal ('97, Figure 2, p. 446) for evidence tliat the fibres of a post-otic ven- 

 tral nerve (hypoglossus auctorum) extend into the pre-otic region with the muscle 

 they innervate. It would seem a very easy matter for such fibres to come into 

 nervous connection phylogenetically with the eye muscles, and especially the 

 posterior of these, with which in Petromyzon they are very closely connected. 

 Hatschek ('92) stated that the muse, rectus posterior becomes connected with the 

 anterior of the post-otic myotomes. See evidence given by M. Fiirbringer ('97) 

 and Neal ('97) upon this question. 



VOL. XXXI. — NO. 7. 8 



