32 VERTEBRATE ZOOLOGY 



that allows water to pass out of the pharynx to the exterior. The 

 opening is made embryonically by an out-pouching of the"p!ia^TLX 

 and an in-pouching of the ectoderm, which meet and break through. 

 There is therefore always an ectodermal and an endodermal part of 

 a pharyngeal cleft. When gills are formed they are derived sometimes 

 from the ectodermal and sometimes from the endodermal part of the 

 cleft. In terrestrial vertebrates no gills, except minute transitory 

 rudiments of gill filaments, ever develop, though the pharyngeal 

 clefts may appear and either break through or remain closed. The 

 number of pharyngeal clefts ranges from upwards ol fifty in Amphi- 

 oxus to one pair in Cephalodiscus and none in Rhabdopleura. Usually 

 a framework of cartilage or bones constitutes the branchial skeleton 

 and serves to support the clefts, their accompanying branchial tissues, 

 and their blood supply. 



Medullary Plate (Neural Tube). The term "medullary plate" is 

 perhaps somewhat more widely applicable than "neural tube" be- 

 cause in some of the more primitive chordates the plate never reaches 

 the tubular condition. The ^medullary plale4s a dorsaLarea of ecto- 

 dgrrn that typically becomes first foM^d^jn_^ngitudinajlv^Jii1k) a 

 neural groove and then converted into a tube with^ a central canaj_or 

 neurocotir. !CT^uf Lliu VefteBrattes piTtpeT^rt^soine ^period haveThe 

 central nervous system external (in the medullary plate condition), 

 but some of the more primitive chordates have merely a diffuse plexus 

 of nerve cells in the skin with a tendency for them to concentrate 

 along the dorsal side. Such individuals can hardly be credited with a 

 medullary plate, much less a neural tube. Some of their immediate 

 relatives, however, have the beginnings at least of an infolding 

 that makes it probable that the structure concerned is a medullary 

 plate. 



In dealing with the various groups that lay claim to membership, 

 along with the vertebrates, in the chordate fraternity, it will be 

 necessary to test their claims by a rigid examination of their creden- 

 tials: notochord, pharyngeal clefts, and medullary plate. 



The heterogeneous collection of types that has been assembled by 

 comparative anatomists and called chordates is customarily sub- 

 divided into four sub-phyla, which, beginning with the group that 

 shows most certain affinities to the vertebrates and followed by those 

 forms that haye a less valid claim to vertebrate relationship, ^re as 

 follows : 



