250 bulletin: museum of comparative zoology. 



form like S. acanthias, where the ear capsule shifts backward, the exit of 

 the root of the glossopharyngeus lies behind eucephalomere VII, whereas 

 iu such forms as the chick and swine, where the ear capsule does not 

 similarly shift backward, the exit of its root is from the expansion of 

 eucephalomere VII. In all Vertebrates, the roots of the glossopharyn- 

 geus and the Urvagus lie close to each other, but in S. acanthias, where 

 there is a greater amount of posterior displacement than in any other 

 Vertebrate that I have studied, these roots are more crowded together 

 than in other forms. These facts seem to me to warrant the conclusion 

 that the roots of the glossopharyngeus and the Urvagus primitively made 

 their exit from those encephalomeres which give rise to their ganglionic 

 Anlagen. And we may likewise assume that the local thickenings of 

 these" encephalomeres have their significance in this primitive relation, 

 i. e. they contained the " Kerne " of these roots. I am able to find no 

 facts which render this assumption untenable. 



On the other hand, eucephalomere IV never has nervous connection 

 with a visceral arch. From it few neural-crest cells are proliferated, 

 and in consequence it never forms the ganglionic Anlage of a nerve, 

 nor does it ever in ontogeny have a motor nei-ve in connection with it. 

 Since the other four encephalomeres are related to visceral arches, I 

 incline to think that this eucephalomere was once related to a visceral 

 arch of its own. Otherwise, so far as I can see, its existence is in- 

 explicable. In this condition, then, I find additional evidence of a 

 lost visceral arch, which van Wijhe ('82), Miss Piatt ('91^), and Hoff- 

 mann ('94) believe once existed in the region of this neuromere. These 

 investigators have found a want of exact correspondence between the 

 somites and the visceral arches in the region of the spiracular cleft. 

 Van Wijhe was led to believe that, the hyoid (2d visceral) arch is 

 double, -i. e. represents two arches, the fusion of which has resulted in 

 the obliteration of the visceral cleft between them, - while Miss Piatt 

 and Hoffmann have held that the mandibular arch is double, and that 

 an anterior gill cleft has disappeared. The disappearance of a viscei-al 

 cleft is rendered plausible, if we assume that such a loss would greatly 

 strengthen the mandibular arch when it came to function as a lower 

 jaw. The evidence from a study of mesomerism and neuromerism there- 

 fore seems mutually confirmatory. . . . n .u . 

 If eucephalomere IV was related to a lost visceral arch, it follows that 

 the lost arch must have been situated posterior to the mandibular (1st 

 visceral) arch, for the musculature of this arch is innervated from eu- 

 cephalomere III. It also follows, because of the relation of the nerve 



