274 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 
nection with them; and (4) of the development of at least a part of the 
musculature innervated by them from splanchnic mesoderm. Thus 
there is to-day a distinct conflict as to the morphology of the eye-muscle 
nerves, one party to the conflict being supported by histological evi- 
dence, the other by embryological. The assumption by His (88), that 
the eye-muscle nerves develop as processes of medullary cells (neuro- 
blasts), — which is involved in his contention that they are the serial 
homologues of ventral spinal nerves, —has never hitherto received the 
requisite embryological confirmation. In fact, the latest embryological evi- 
dence concerning the development of the oculomotorius and trochlearis 
seems quite irreconcilable with the view of Schneider (’79), van Wijhe, 
and His. In regard to the latter nerve, Hoffmann (89, p. 338) says, 
if one disregards the fact that no ectodermal fusion takes place, “so 
gleicht die Anlage des Trochlearis in sehr jungen Entwicklungsstadien 
[of Lacerta] vollkommen der eines segmentalen Kopfnerven, besonders 
der des Trigeminus.” Froriep also finds that the trochlearis possesses 
in early stages a ganglion, and is differentiated from neural-crest cells in 
situ. Miss Platt ('91*, p. 259) likewise states that “in Acanthias the 
development of the trochlearis in all essential respects so completely 
corresponds to that of the trigeminus and facialis, that like them it must 
be considered to combine primarily those dorsal and ventral elements 
which have separate roots in the nerves of the trunk. It can, therefore, 
not be regarded as the ventral root of another segmental nerve.” More- 
over, Kupffer (’95, 96) finds the trochlearis to possess in Ammocoetes 
both dorsal and ventral roots. 
With regard to the oculomotorius, the conclusions of embryologists 
are even more conflicting. While Dohrn (’91) finds that this nerve is 
formed by the migration of cells from the ventral wall of the midbrain, 
and considers it a motor nerve, Miss Platt (’91*) states that she has 
shown the oculomotorius to be “undoubtedly originally sensory.” Her 
observation that the nerve develops from the ganglion toward the brain 
has been confirmed by both Mitrophanow (’93) and Sedgwick (95). 
Nevertheless the evidence which has been stated by me in division VI. 
shows conclusively, as I believe, that all the eye-muscle nerves, oculo- 
motorius, trochlearis, and abducens, develop, like ventral spinal nerves, 
as processes from neuroblasts lying in the ventral horn of the medul- 
lary tube. Therefore, from their development, as well as their adult 
histological structure and relationships, the eye-muscle nerves must be 
regarded as the serial homologues of ventral spinal nerves. Finally, 
with the accumulating evidence given by many investigators, — among 
