116 ME. W. G. EIDEWOOD ON THE HTOBEAKCHIAL 



some physiological purpose, probably to give the larynx greater- 

 freedom of motion. 



Both Xenopus and Pipa have, it is true, no tongue, but is it 

 not just possible that there is some physiological connexion, 

 between the absence o£ tongue, the union of the eustachian 

 recesses into a median ostium pharyngeum, the excessive develop- 

 ment of the hyobranchial wings, the formation of a hyoglossal 

 foramen, the elongation of the bronchi, the absence of vocal 

 cords, the enlargement of the larynx, and its support by the 

 hinder part of the branchial skeleton ? Admitting that the same 

 conditions acting independently on the two genera may have 

 resulted in the loss of the tongue or some other of the above 

 specified peculiarities, it is quite comprehensible that the other 

 modifications have followed as concomitant variations. The 

 similarity of the hyobranchial and laryngeal skeleton, so far as I 

 am aware, has not yet been employed as one of the characters 

 pointing to a natural affinity between Xenopus and Pipa; but 

 were the attention of the systematist directed to this feature, he 

 would probably interpret it as corroborative detail, whereas if,, 

 as above suggested, it is physiologically related to the suppres- 

 sion of the tongue, he would be doubling or perhaps trebling the 

 importance of the aglossal character, and would not be adducing 

 new evidence. It may, in fact, be that the enlargement of the- 

 larynx as an accessory sexual organ is the primary cause of all 

 this modification, including even the suppression of the tongue.- 

 In the Discoglossidse and Pelobatidse, where, as already pointed 

 out, the larynx is much more enlarged than in most Anura,- 

 and where the tongue shows unmistakable signs of reduction, 

 we have some such correlation as that suggested, and, what is 

 more, a diminution of the antero-posterior diameter of the basaL 

 plate of the hyoid, coupled with a great development of the 

 lateral outgrowths of the plate. 



In Bomhinator and Peldbates the mesial ends of the hyoideau 

 cornua are connected by a dense layer of connective tissue 

 situated below the hyoglossal muscle ; and in Pelodytes the 

 cornua are actually in contact in the median line, so that only 

 a fusion of the two is required to make a perfect hyoglossal 

 foramen. This fact is very significant, since it is only in the.«e' 

 genera, outside the Aglossa, that any indication of a hyoglossal 

 foramen is to be met with. The anterior cornu of Xenopus is 

 broad and expansive, and thereby exhibits a retention of a- 



