106 ME. F. B. BEDDAED ON THE [Feb. 6, 



clefts are slightly oblique. The epithelium which lines it differs 

 on the anterior and posterior faces of the cleft. Anteriorly the 

 epithelium, like that of the buccal cavity, is formed of low cells ; 

 posteriorly it is formed of tall columnar cells. These cells are 

 continuous with the ventral epithelium of the pharynx, which has 

 this character ; the dorsal epithelium being low. This phai'yngeal 

 tract of columnar epithelium extends back over the whole of the 

 branchial region, but suddenly stops short a little way in front of 

 the origin of the lungs. This fact is perhaps incidentally of some 

 little importance in view of the homology between gill-sUts and 

 lungs which was once urged. Had this modified tract of pharyngeal 

 epithehum extended to the lung and into it, as into the hyoid and 

 branchial clefts, the question might have been considered anew. 

 It will be noted that the hyoid cleft differs much from the branchial 

 clefts which follow in that the modified pharyngeal epithelium 

 only lines its posterior surface. This cleft does not open on to 

 the exterior. 



In tadpoles of June 2nd (cut longitudinally and horizontally), 

 in which the branchial basket was well developed with its vascular 

 tufts, the hyoid cleft showed no traces of being a respiratory cleft 

 and did not open on to the exterior either independently or by 

 way of the other branchial cleft. 



In a tadpole of June 5th, the opening of the hyoid cleft was 

 effected. It has the form of a comparatively narrow tube, which, 

 curving round shortly after its origin from the pharynx, opens into 

 the first branchial cleft a long way from the opening of the latter 

 on to the exterior. 



Internal Gills. — The branchial arches, as in other Amphibia, fuse 

 to form a basket-work, from the bars of which run cartilaginous 

 processes which become tufted and form the so-called filtering 

 apparatus. I observed the first traces of this filtering apparatus 

 in tadpoles of May 31. These structures become later very 

 vascular, and they must be respii'atory in function, since no other 

 internal gills are developed. In the Common Frog the tadpoles 

 possess not only these " filters" but tufted internal gills. Messrs. 

 Marshall and Bles \ while admitting the vascularity of the filters, 

 consider that, " as the blood is returned from them to the somatic 

 veins, it is probable that they are not actively respiratory." They 

 clearly must be in Xenopus, as there are no other gills. 



Exteriud Gills. — As has been already mentioned, the tadpole of 

 Xenopus is said by Mr. Leslie to possess no external gills. This 

 statement is not quite accurate, though undoubtedly complex 

 arborescent gills like those of Rana are not to be discovered. 

 Messrs. Marshall and Bles have emphasized the fact, which has 

 been rather slurred over, that the external and internal gills form 

 a continuous series of structures. In 4-5 millim. long tadpoles 

 of Rana " two pairs of external gills are present as backwardly- 



' " The Development of the Blood-vessels in the Frog," Stud. Biol. Lab. 

 Owens Coll. ii, 1S90. 



