132 PROFESSOR MARSHALL. 



while the anterior root is devoid of ganglion cells. The roots of origin 

 of the anterior root are also very generally multiple, while those of 

 the posterior roots, whether primary or secondary, are apparently 

 invariably single. 



Balfour's observations were soon extended to birds and mammals, 

 and the description given above is now recognised as that of the 

 general and typical mode of development of the vertebrate spinal 

 nerves. It was further found that the neural crest is not confined to 

 the spinal cord, but extends forwards along the top of the brain, 

 and that certain of the cranial nerves are developed from it in the 

 same way as the posterior roots of the spinal nerves. By this 

 discovery a new and veiy reliable clue to the segmental value of the 

 cranial nerves is obtained, for it is clear that if certain of the cranial 

 nerves do, and others do not, conform to the mode of development of 

 the typically segmental spinal nerves, there is strong reason for 

 regarding the former as being of segmental value, and the latter 

 as not. 



Embryology has furnished us with one further test of the segmental 

 value of cranial nerves, for which again we are indebted to Mr. Balfour, 

 who has shown that in Elasmobranchs (and the observation has since 

 been extended to other groups) the two halves of the ccelom or body- 

 cavity at an early period extend forward on either side of the neck 

 into the head, and that on the appearance of the visceral clefts each 

 of the halves becomes cut up in a series of isolated compartments, 

 one in each visceral arch. 1 If the visceral clefts and arches are 

 segmental, it is clear that these " head-cavities," as they are called, 

 must be also, and that they will therefore afford an additional clue 

 to determining the segmental value of the nerves associated with them. 



Summary of Evidence of Segmental Value of Cranial Nerves.— 

 From what has been said above it will be evident that we have now 

 several independent tests of the metameric or segmental value of the 

 cranial nerves, — tests with all of which a nerve ought to comply to 

 entitle it to rank as segmental. For convenience of reference, these 

 tests, the majority of which have already been discussed, may be 

 enumerated here : 2 — 



1 Balfour, Elasmobranch Fishes, pp. 206-209; also Marshall, " Head-Cavities and Asso- 

 ciated Nerves of Elasmobranchs," Quart. Journ. of Micros. Science, January 1881. 



2 Cf. Marshall, " The Morphology of the Vertebrate Olfactory Organ," Quart. Journ. of 

 Micros. Science, July 1879, p. 317. 



