SKELETON AND LA.ETNS OF XENOPIIS AND PIPA. 99 



The ceratohyal lias not shared in this modification, and so a 

 vacuity now exists between its outer end and the branchial 

 skeleton. The anterior portion of the basihyal has been ab- 

 sorbed, and a cleft is beginning to apjDear between the median 

 extremities of the ceratohyals. The distance between the outer 

 ends of the ceratohyals is now slightly less than tbe maximum 

 transverse diameter of the branchial skeleton. The larynx, 

 bronchi, accessory pulmonary lobes, and the lungs are all larger 

 than before, but are not otherwise noteworthy. 



^TAQ-E III. (Plate 11. fig. 3.) Tadpoles with well-developed fore 

 limhs, sliarply tent. Absorption of tlie tail not yet commenced. 



Considerable changes bave taken place in the hyobranchial 

 skeleton since Stage II., far more than would be suspected from 

 a superficial examination of the tadpoles. The five cartilages to 

 be seen in Stages I. and II are no longer distinguishable. The 

 basibyal has either been absorbed, or bas fused up with the 

 neighbouring cartilages in such a way as to leave a deep cleft, 

 enlarging posteriorly between the median ends of theceratobyals. 

 The posterior end of this cleft marks the position of the future 

 hyoglossal foramen ; and it only requires a further enlargement 

 of the cleft, accompanied by a secondary union of th.e ceratohyal 

 cartilages in front, to bring about the adult relations. It is a 

 matter of the keenest regret that I am unable, from lack of 

 material, to add a circumstantial account of these later changes. 



The mandible and ceratohyal have changed but little ; and the 

 concavity at the back of the latter, vacated by the first branchial 

 arch, still remains. The branchial skeleton, on the other hand, 

 has been considerably reduced. The first branchial arch is now 

 no thicker than the second, and a wide space intervenes between 

 it and the ceratohyal. All trace of inflation has disappeared, and 

 the flattened branchial apparatus bears a much closer resemblance 

 to that of the common frog than it did before. The three 

 branchial clefts have not altered except in size, and, judging by 

 analogy with tadpoles of other species of Anura, they will not 

 do so. The arches will be absorbed, and the clefts will disappear 

 in consequence ; but the latter will play only a passive part in 

 the change. The distance across the branchial skeleton is now 

 considerably less than the extreme width of the hyoid arch. That 

 the absorption of the branchial skeleton has been a marginal one 

 is seen by the fact that in the first stage the external edge of the 



7# 



