55 



in Necturus allows one to follow the further fate of cells not 

 contributing to the Anlage of the peripheral nervous system. That 

 many such cells exist in 'other embryos has been known for 

 some time. They are, however, usually similar in appearance to 

 cells migrating at the same time from the walls of the head-cavities 

 (or from homologous "mesodermic" tissue) which has made it often 

 difficult, if not impossible, to determine the number and ultimate fate 

 of these migrating mesectoderm cells. Yet I believe that we are not 

 for this reason warranted in assuming that no actual difference may 

 exist between the individual cells composing an apparently homo- 

 geneous mesenchyme. 



Since there is no such union of mesectoderm with mesendoderm 

 as Goronowitsch has described in the chick, it is evident that the 

 homologue of the "periaxial Strang" does not exist in Necturus. 



The position and behavior of cells here taking part in the for- 

 mation of the peripheral nervous system recall strikingly observations 

 of similar phenomena in the Elasmobranchs. The sensory nerves arise, 

 as is well known, in the skin, and become apparently connected with 

 the brain through the mediation of ganglion cells. In the formation 

 of the lateral-line nerve, the migration of cells from the ectoderm into 

 the nerve throughout its entire length is so great as to assimilate 

 this nerve to a prolonged ganglion both in structure and in manner of 

 formation. 



I have not observed the migration of cells from the brain into 

 the ganglia ("nervenführendes Gewebe" of Goronowitsch) except 

 through the neural crest, and if such migrations occur at an early 

 stage of development, they are not common. I have moreover seen 

 that some of the original mesectoderm cells which take part in the 

 formation of the Gasserian ganglion "spin" fibres to muscle cells lying 

 in the mandibular mesendoderm. These are undoubtedly motor fibres 

 "spun" from cells that have not migrated from the brain, unless through 

 the neural crest. Connecting this fact with observations made by 

 Froriep 1 ) and myself on the development of the trochlearis in 

 Selachians, I am inclined to think that before the great motor and 

 sensory tracts of the central nervous system are established , indivi- 

 dual peripheral cells may complete in themselves a reflex arc, either 

 as an ancestral characteristic repeated in embryonic development, or 

 as an advantageous larval acquisition. 



1) A. Feobiep, Zur Entwickelungsgeschichte der Kopfnerven. Ver- 

 handlungen der Anat. Gesellschaft, 1891. 



