neal: nervous system in squalls acanthias. 251 



of eucephalomere V (facialis) to the present spiracular cleft, that this 

 was once the second visceral cleft instead of the first (disregarding for 

 the present the possibility that the mouth i-epresents a pair of gill clefts), 

 as it now is. It seems entirely possible that the outpocketing of the 

 present first viscei'al (hyo-mandibular) cleft was originally a double one,^ 

 and that the fusion of these two outpocketings resulted in the loss of the 

 A'isceral arch which once separated them, and therefore in the loss of the 

 nerve primitively related to that arch. Moreover, between the second 

 head somite of van Wijhe, which extends into the mandibular arch, and 

 the fourth somite, which is widely connected with the mesoderm of the 

 hyoid arch, there lies the third head somite, in correlation with which 

 there is no intermediate visceral arch. This somite (the 3d) lies opposite 

 the posterior constriction of neuromere IV, and speaks plainly for the 

 previous existence of a lost head segment, for which neuromere IV may 

 once have furnished the nerve centre. Did such an arch exist, each of 

 van Wijhe's somites from the second to the sixth, and each of the en- 

 cephalomeres from III to VII would correspond with a visceral arch. 



I give a brief summary of the line of reasoning which leads me to 

 believe that the significance of the hindbrain neuromeres lies in their 

 primitive relationship to the visceral arches. In the young embryos 

 of S. acanthias two facts, both so far as I know new, present them- 

 selves. In the first place, the hindbrain neuromeres, five in number, 

 are found to be successive similar thickenings of the lateral zones of 

 the medulla. Secondly, from four of them, viz. Ill, V, VI, and VII, 

 are proliferated the ganglionic cells of the four cranial nerves which in- 

 nervate the first four visceral arches, viz. the trigeminus, the facialis, 

 the glossopharyngeus, and the Urvagus. A clue to the significance of 

 the local thickenings of the neural wall in the tract of the encephalo- 

 meres is given in the fact that from those two encephalomeres which 

 (in other Vertebrates as well as in S. acanthias) most closely retain 

 these primitive nerve relationships, viz. Ill and V, emerge the fibi-es 

 which innervate the visceral arches (primitively) related to them. The 

 thickenings are the first expression of the "Kerne" (nuclei) of the 

 nervous centres related to the visceral arches, and possibly also, priuii- 

 tively, of those related to the somites. 



1 Kupffer ('93) finds in Acipenser embryos an entodernial outpocketing or pouch, 

 which soon disappears, just anterior to tlie liyomandibular pouch. The position of 

 tliis pouch would identify it with the cleft whose former existence seems probable 

 on the evidence given above. Houssay ('91) also recognizes in Amblystoma a vis- 

 ceral cleft between the oral and the liyomandibular. 



