THE RESPIRATORY SYSTEM 



335 



their surface. The fact that the peribranchial cavity of amphioxus 

 develops from the ventral side of the body, makes it seem impossible to 

 compare this with the atrium of the urochordates which arises on the 

 dorsal side. See p. 66i, Fig. 535. 



The history of gills in vertebrates is one of continuous reduction in 

 number and modification in function. The transformation of their 

 skeletal supports has been described in the history of the skeletal system. 

 Even in amphioxus, some of the gill pouches of the embryo are modified 

 or lost. The first pair become the endostyle while the second pair form 

 the larval club-shaped gland. The third left slit never has a correspond- 

 ing right sUt and disappears early in ontogenesis. The first permanent 

 gill sUt of amphioxus is therefore the fourth of the ontogenetic series. 



ENDOSTYLE 

 _3RD RWRED 

 /GILL POUOCS 



5— VENT. AORTA 



(s) ^^ 



AJONGUE-BAR 



S*->CILL POUCHES CILL SUTS' 



C. D 



Fig. 280. — Four stages in the development of the endostyle of amphioxus. The 

 evidence indicates that the endostyle is developed from a pair of modified gill-pouches. 

 The evidence is equally strong that the endostyle is the homolog of the thyroid gland 

 of vertebrates. The conclusion is drawn that the thyroid has evolved from a pair of 

 gill-pouches. There is no evidence, however, that endostyle and thyroid are functionally 

 similar. (Redrawn after van Wijhe.) 



There are cogent reasons for homologizing this slit with the spiracle of 

 elasmobranchs, but there is Uttle agreement among morphologists in 

 regard to the exact homology of serial organs in chordates. 



The popular belief among morphologists that the vertebrate mouth 

 has been formed by the coalescence of a pair of gill slits is supported by 

 the mode of development of the mouth in amphioxus. The endoderm 

 takes the initiative in the development of the mouth of amphioxus, as 

 would be the case if it were a gill slit. In this respect, the mouth of 

 amphioxus differs from that of vertebrates, in which the ectoderm initiates 

 development. Later the mouth of amphioxus assumes a left-sided posi- 

 tion and becomes enormously enlarged. Partly from the fact of its 

 left-sided position, van Wijhe (1893) homologized the mouth of amphioxus 

 not with that of vertebrates, but with the left spiracle of elasmobranchs. 

 Other morphologists have not accepted this homology, but are not wholly 

 agreed that the mouth of amphioxus is homologous with that of verte- 

 brates. At metamorphosis, the mouth of amphioxus assumes a position 



