566 



of the ganglion in question, and may even according to Froriep pre- 

 sent a ganglion (in the young embryo of Torpedo). But the ganglionic 

 portion may from the earliest days of the Craniates have been ab- 

 sorbed in the ophthalmic and so have failed to reach the brain, and 

 the motor fibres alone remain to represent it. It appears, probable 

 therefore that the ophthalmic has taken up the ganglia of the three 

 nerves,, the thalamic, mesencephalic, and trochlear. 



Thus both anteriorly and posteriorly the cranial ganglia have 

 undergone suppression, the incidence of the auditory vesicle has brought 

 about the Craniate position of the VII — VIII, and the mesencephalic 

 flexure that of the V. 



The ventral roots occur interneuromerically and are restricted to 

 the hinder rhombomeres — usually the last three persisting — to the 

 interrhombomere 5—6 (abducent), and to the intermesomere 1—2 

 (oculomotor). Most neurologists would include the trochlearis in the 

 segment behind the latter, but for reasons already given this nerve is 

 here considered the motor survivor of a mixed root. It may, as was 

 suggested in the paper on the gull, be compared with the spinal ac- 

 cessory. But we have no reason for doubting that originally all the 

 intersegments except the first and second possessed ventral roots. 

 Thus one has been suppressed in front of the oculomotor. The next 

 five have also disappeared. This brings us to the abducent. Behind 

 the latter a slightly variable number come into existence, the last three 

 of which usually persist as the cranial part of the hypoglossal. And 

 these are continued in the uninterrupted series of the spinal cord. 



We are thus led to see that some of the ventral roots have only 

 a temporary ontogenetic existence. 



Spinal dorsal roots, which are also situated interneuromerically, 

 are altogether absent from the cranial region. They are represented 

 frequently in the embryo in one or more of the post-vagal segments, 

 and they have been found further forward in the region of the hind 

 brain in the lowest Craniata (Dohrn, v. Kupffer). They do not occur 

 even in the embryo in front of the hind brain. 



B. Mesodermal Somites. 

 The presence of the ventral roots and of the fleeting dorsal roots 

 is associated with mesodermal somites. In previous papers on the sub- 

 ject reasons have been advanced for presuming that the whole region 

 above described as being provided with ventral roots presented orig- 

 inally an equal number of somites. It is only in the lowest Craniates 

 that a glimpse is obtained during development of the majority of these. 



