220 bulletin: museum of comparative zoology. 



ill f miction, and to contribute to the mesenchyma of the head.^ The 

 ear capsule now lies with only its anterior portion opposite encephalo- 

 mere VI. Behind the ear capsule and opposite the posterior portion 

 of eucephaloniere VII lie the cells of the glossopharyngeus, as yet with- 

 out fibrillar connections with the neural tube. Behind the glossopha- 

 ryngeus and now separated from it lie the cells of the vagus, extending 

 ventrally as a broad sheet between the mesoderm and ectoderm into 

 the region of the pharynx, where the Anlage becomes segmented by the 

 formation of the visceral clefts. The trochlear and tlialamic portions 

 of the trigeminus soon disapj^ear without assuming fibrillar relation 

 with the neural tube. 



At a stage with 52 somites, w-heu the embryo is about 8 mm. in length, 

 the thalamic portion remains as a group of cells lying in the constriction 

 between the forebrain and midbrain vesicles (Plate 4, Fig. 18), but with- 

 out connection with the ophthalmicus profundus. It very soon disappears 

 entirely, and I think probably contributes to the loose mesenchyma of 

 this I'egion. In precisely the same way the disassociation of cells of 

 the trochlear portion takes place, scattered clumps of cells indicating 

 its previous extent. The Gasserian ganglion and the ganglion of the 

 ramus ophthalmicus profundus (mesocephalic ganglion) are both clearly 

 differentiated. Three branches of the fifth nerve may now be distin- 

 guished, viz. the two sensor branches, r. ophth. profundus and r. maxil- 

 laris (inframaxillaris ^ Dohrn), and the mixed mandibular branch. Xerve 

 relations to the neural tube remain the same as in the previous stage. 



d. Development of the 



1. OCULOMOTORIUS. 



By the time the embi-yo has reached the length of 8 mm. (52 so- 

 mites), the oculomotorius has however appeared as a fibrillar process 

 from the base of the midbrain (encephalomere II, Figures F to H), arising 

 as processes from neuroblast cells in tlie ventral horn of this encephalo- 

 mei-e. Since this nerve throws light on the morphology of the pre- 

 mandibular somite, whose musculature it innervates, its development is 

 of great interest and has been studied by many investigators ; viz. Mar- 

 shall ('81), Rabl ('89), Dohrn ('91), Piatt ('91), Mitrophanow ('93), 

 and Sedgwick ('94). Neither Marshall ('81) nor Rabl ('89) saw the 



1 Kastschenko ('88), Goronowitsch ('92), Miss Piatt ('9.3). 



- This is, I believe, the nerve which in Ceratodus van TVijIie ('82) named ramus 

 maxillaris superior, whicli in Amphibia Strong ('05) called accessory branch of 

 the fifth, and Miss Flatt ('90) r. buccalis profundus V. 



