230 rROCEEDIXGS OF THE ACADEMY OF [Marcll, 



to it. Thus in the figures the region immediately dorsal to the fold 

 is the mandibular arch, whereas that ventral to it is the hyoid 

 arch. In fig. 7 the efferent portion of the mandibular aortic arch 

 can be seen as a transversely placed vessel (m.a'.) just above the 

 roof of the pharynx and extending outward above the hyoman- 

 dibular diverticulum. Internally the vessel unites with the carotid 

 (car.). The course of the mandibular aortic arch can be followed 

 by comparing the figures. At first it is very small, as seen in fig. 

 5 (?n.a'.). Tracing it forward, however, it is soon found to be 

 continuous with a much larger vessel with a Avell-marked lumen. 

 This vessel is the afferent portion of the mandibular aortic arch 

 {m.a"). Immediately beneath the antcro-inferior extremity of the 

 hyomandibular fold the mandibular aoitic arch is joined by the 

 hyoidean aortic arch, and the common trunk thus formed communi- 

 cates with the large inferior jugular sinuses beneath the mouth. 



The other structure to which I desire to call attention is the 

 hyomandibular ramus of the facial nerve. The facial ganglion at 

 present lies just back of and above the dorsal margin of the 

 hyomandibular fold. The anterior edge of the ganglion is in 

 actual contact with the outer margin of the fold (fig. 7, vii). 

 From the ventral surface of the ganglion the hyomandibular ramus 

 (fig. 6, vil h.) is given off as a large nerve supplying the muscles 

 of the hyoid arch. It is hence posterior to the hyomandibular 

 fold. 



Stage ///(Pis. VII, VIII, figs. 10-14, 16-18).— Young tadpole. 

 External gills prominent and considerably branched, not covered as 

 yet to any marked extent by the opercular fold. Third visceral- 

 cleft opening to the exterior. Mouth communicating with pharynx. 

 The tail has attained its full development. 



A considerable departure from the conditions observed in the 

 preceding stage is shown in the present. The different organs are 

 (|uite clearly differentiated, while the Anlngen of the more impor- 

 tant cartilages can be made out as dense aggregations of the mesen- 

 chyme. The first visceral-cleft especially has undergone marked 

 modifications. We can follow out its course by comparing figs. 

 10-14. Consulting fig. 10, we notice that the pharyngeal wall is 

 separated from the exterior by a considerable interval occupied by 

 scattered mesenchyme cells, which in the region immediately sur- 

 rounding the pharynx are segregating to form the AnJagen of the 



