236 BULLETIN: MUSEUM OF COMPAEATIYE ZOOLOGY. 



lie above the superior oblique muscle. Thus the permanent trochlearis 

 arises from two sources, from the brain and from ganglion cells." Fi- 

 nally, KupfFer ('91) stated that he had found a nerve in Amraocoetes, 

 which he thought to be the trochlearis (for reasons not clear to me), 

 directly connected with the second epibranchial ganglion. ^Vere this 

 opinion correct, the trochlearis would be the serial homologue of a 

 branchial (dorsal), not of a spinal dorsal nerve. 



From this summary of previous embryological evidence bearing on 

 the question of the morphology of the trochlear nerve, it is clear that 

 little support is given to the view, based on the later histological struc- 

 ture and relations, that it is morphologically a ventral segmental nerve. 

 Only Kastschenko ('88) finds the nerve in early stages fibrillar in 

 structure. The following evidence, however, leads me to conclude that 

 its mode of development is the same as that of the oculomotorius and 

 abducens, and that therefore it must be regarded, like these, as a ven- 

 tral (medullary) nerve. I first find tlie trochlearis in sections of embryos 

 of 19-20 mm. as a fibrillar nerve bundle extending from the dorsal con- 

 striction between eucephalomeres II and III. Two roots are already 

 present at this stage, but neither in these, nor in the nerve bundle as 

 far as its fibres may be traced in the mesenchymatous tissue at the sides 

 of the brain, are nuclei to be found. "While proximally the nerve fibres 

 are united in a compact bundle, they distally separate so as to form a 

 loose brush of structureless fibres, which are lost in the mesenchyma at 

 a considerable distance dorsal to the muse, obliquus superior (Figure K). 

 While I am able to offer no direct evidence in favor of the view that the 

 fibres of the trochlearis, as above described, are processes from neuro- 

 blast cells in the ventral horn of encephalornere III, I hold that they 

 are such, since their well known later histological relations support this 

 conclusion. Dorso-veutral fibres in this region of the neural tube may 

 indeed be traced in embryos of this stage, but their connection with the 

 fibres of the trochlearis is not clear to me. The dorsal chiasma of fibres 

 is present.^ Of a ganglion, or of any grouping of cells which might 



is that portion of the trigeminus Anlage which I have for convenience called its 

 trochlear portion, which persists for some time in the constriction between midbrain 

 and hindbrain vesicles. Since the proof of its morphological value has not been 

 given, and since the " permanent " trochlearis is not developed from the " primary " 

 trochlearis, as Miss Piatt herself states (p. 96), the use of the latter term appears 

 to me apt to mislead. 



1 The explanation for this dorsal chiasma may be sought in some physiological 

 advantage in coordination gained, but it may also be seen that in case the dorsal 

 exit of fibres were of physiological advantage, it would be easy for the fibres to 



