NEAL : NERVOUS SYSTEM IN SQUALUS ACANTHIAS. 153 



SO far as tlier have been studied. In both tables it has been impossible 

 to exclude much that is theoretical, and in view of this fact general 

 conclusions are obviously dangerous. One important result, which 

 should be borne in mind during the discussion of the evidence presented 

 in tiiis paper, is established, viz. the constancy, in all classes of Verte- 

 brates, of five "hindbrain neuromeres " ("Falten" or "vrais replis"), 

 and of their nerve relations. When six have been counted, usually the 

 Anlage of the cerebellum has been included with them, and when seven, 

 (see HoflFmann, '90), another fold behind the true fifth neuromere has 

 been counted. There is consensus of opinion that from the third 

 "■ hindbrain neuromere " (designated in Table II. as 5, and as Y in my 

 figures) the acustico-facialis nerve takes its origin. In counting hind- 

 brain neuromeres, then, this may safely be used as a check. In regard 

 to the presence of true neuromeres, comparable with those of the hind- 

 brain, in the region of the encephalon anterior to the hindbi-ain, much 

 is theoretical, and, as I believe, uncritical. Morphologists have natu- 

 rally been more or less prejudiced in favor of the view that a serially 

 homologous segmentation extends throughout head and trunk. This 

 preconception has led to the search for resemblances at the risk of dis- 

 regai'ding differences which obviously exist, and as a result structures in 

 the encephalon which are morphologically incomparable with the rayelo- 

 meres have been homologized with them. ]\Ioreover, this has been done 

 in utter disregard of their relations to the segments of the mesoderm. 



The study of neural segments and their relations to nerves and somites 

 in embryos of Squalus acanthias has given me some facts bearing 

 on the problem of cephalic segmentation, which are, so far as I know, 

 new. The conclusion ^hich I have reached is as follows. In S. acan- 

 thias there exists in early stages a contimious primitive segmentation of" the 

 nervous system serially homologous throughout head and trunTc, — the 

 '■'• neuromeric'^ iegmentation. In later stages there appears in the en- 

 cephalon a secondary (in time) segmentation resulting in the so called 

 vesicles, which are not serially homologous with the segments of the 

 myelon, but give rise to an anterior cephalic tract, which is a region sui 

 generis. 



In the following discussion I propose (1) to trace the development of 

 neuromeres; (2) to compare the structure of the segments of the enceph- 

 alon with those of the myelon ; and (3) to note the relation of the 

 neuromeres to the sensor and motor nerves, to the mesodermal somites, 

 and to the visceral arches. I shall begin with the description of the 

 first appearance of neural segmentation in the embrvo. 



