254 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 
parts of the original segments. Fortunately, however, with the knowl- 
edge that neuromeres and mesomeres correspond numerically, we are 
able to see that the majority of changes which have occurred are cor- 
related ones, and therefore capable of explanation. We furthermore see 
that the greatest changes have taken place in the more anterior meta- 
meres, chiefly and primarily by the loss of the ventral parts of these 
metameres. Since the more posterior of the cephalic segments have 
indubitable metameric value, I shall discuss in detail only those anterior 
ones (viz. I to VIL) concerning which there is most disagreement among 
morphologists, beginning with the consideration of the seventh, whose 
relations are least modified. 
VIII. Primitive Relations of Cephalic Segments. 
a. RELATIONS or ENCEPHALOMERE VII. 
Opposite the posterior constriction of this encephalomero in very early 
stages lies van Wijhe's 6th somite, which develops embryonic muscle 
fibres and is universally considered a true somite. I therefore regard 
this as the mesomere corresponding with encephalomere VII, whose 
neural-crest cells first meet the mesoderm opposite the anterior constric- 
tion of this somite (Plate 3, Fig. 13). These cells form the Anlage of 
the anterior branch of the vagus (Urvagus), and I assume that the 
primitive relations of this nerve were with the myoseptum between the 
5th and 6th somites. The intermediate position of the Urvagus with 
respect to the myotomes and its ontogenetic union with spinal ganglia 
in some Vertebrates serves to show that there is no fundamental 
difference in this respect between cranial and spinal nerves. For rea- 
sons which will be stated in connection with the study of the relations 
of encephalomere IV, I regard the abducens (Plate 4, Fig. 21), whose 
fibres have their exit from the ventral horn of encephalomere VII, 
as representing in part the ventral nerve of this segment. Furthermore, 
I assume that the mesoderm of the 6th somite was primitively connected 
with the mesoderm of the 4th visceral arch (Plate 3, Fig. 16); because 
that somite in Ammocotes which I regard as its exact homologue, viz. 
the 2d post-otic somite, is certainly in early stages thus connected. 
Consequently the present 3d visceral cleft bounds ventrally the visceral 
(splanchnic) portion of this segment, 
