230 BULLETIN : MUSEUTI OF COMPARATIVE ZOOLOGY. 



olfactory pit and brain wall as these separate from each other. I regret 

 that I am not yet in position to describe the later differentiation of 

 these cells, but it appears to me not improbable that they enter into 

 the Anlage of the olfactorius. The evidence given by many histologists, 

 from Schulze to Retzius, establishes the ganglionic character of the 

 olfactorius ; while the evidence presented by His ('89*) appears confirma- 

 tory of the view of Beard, that the olfactory plate is to be regarded as 

 the anterior of the sense organs of the lateral line, since from it are 

 derived, at least in part, the ganglionic cells of the olfactorius. There- 

 fore, if neural-crest cells also entered into the Anlage of this nerve, it 

 must be regarded as homodynamous with the sensor component of a 

 dorsal segmental nerve. 



2. Abducens, 



Another of the eye-muscle nerves, viz. the abducens, is now (65 so- 

 mites, 10 mm.) differentiated. (Compare Figs. 20 and 21 with Fig. 19.) 

 The latter nerve has arisen as an outgrowth from neuroblast cells in the 

 ventral horn of encephalomere VII, and its roots retain connection with 

 this encephalomere until the latter disappears. Zimmermann ('91) stated 

 incorrectly that its connection in Squalus is with the neural segment 

 which corresponds with my encephalomere YI. Dohrn ('90") describes 

 the nerve as having its origin from the neural tube opposite the otic 

 capsule, and between nerves YII and IX. Its position in different 

 Vertebrates seems inconstant. Some investigators (Orr, '87, Waters, '92, 

 and Hen-ick, '92) have stated that in the forms studied by them it arises 

 from the hindbrain neuromere corresponding with encephalomere IV of 

 mv figures. In the chick and swine I have found that its roots are 

 in connection with encephalomere VI, whereas in Necturus its fibres 

 may be traced from the muse, rectus posterior to a point behind the ear, 

 and thus have, as I believe, their origin from a segment of the hindbrain 

 corresponding with encephalomere VII. At least, in this form, as in 

 Squalus, it appears as a post-otic nerve. Dohrn ('91) gave a careful 

 account of the early stages of its development in embryos of various 

 Selachii. He states that the nerve first appears at a stage corresponding 

 with Balfour's stage L, arising by two roots which unite at a short dis- 

 tance from their point of exit fi'om the ventral wall of the neural tube. 

 In Mustelus the roots are more numerous than in the other forms exam- 

 ined, thei'e being as many as six on each side of the brain. The roots 

 are directed backwai-d, as in the case of spinal nerves, but later form a 

 network from which arises the stem of the nerve; this runs forward, 



