38 H. B. POLLARD. 



One of the most remarkable features is the Rostrum or modification 

 of the internasal septum. It is most marked in Trichomycterus and 

 Auchenaspis. It becomes invaded by the so called dermethmoid bone. 

 We have only to consider the rostrum somewhat prolonged to obtain a 

 typical Sturgeon rostrum. The Sturgeons cannot be very far removed 

 from the Siluroids, more especially the Hypostomidae, as indeed is 

 suggested by Huxley's observations on the relation of the fossil forms. 



The comparative anatomy of the hyomandibular is of very great 

 interest, but, since I have already dealt with the subject in a paper on 

 the suspension of the jaws, I need not refer to it in detail here. The 

 hyomandibular articulates with the pterotic ridge by a long articula- 

 tion. The immobility of the suspensorium of Hypostomidae is well 

 known. 



In Clarias the pterotic ridge is produced far outwards, the arti- 

 culation of the hyomandibular lying some way from the cranial wall. 

 The Siluroids approach near to an autostylic condition, or, to speak 

 more correctly, are little removed from it. 



Histology op the Tentacular Skeleton. 



I venture to give the following sketch of the varieties of cartilage 

 present in the tentacular skeleton. 



Condensed embryonic tissue, known as precartilage, developes in 

 various directions. It may give rise to an intercellular matrix with a 

 tendency to become refractile. Such a tissue, for the sake of com- 

 parison, 1 term soft Myxinoid tissue (A), inasmuch as it forms the 

 axis of the tentacles of Myxine. The nuclei and protoplasm may dis- 

 appear, and the intercellular matrix become very hard, as in the hard 

 tissue of Myxine. 



On the other hand the intercellular matrix may become fibrillar and 

 and the cells and protoplasm degenerate as in the tentacles of Clarias 

 (B). Or the precartilage may develope into hyaline cartilage (C) or 

 persist to a considerable extent in its embryonic condition (D). The 

 refractile matrix is stained blue by Bleu de Lyon, while the nuclei and 

 protoplasm remain unstained. The core of the tentacle may attain 

 very little developement, the tentacle then being very flexible, as in 

 Misyurnus, or, on the other hand, as in Mote/la it may be formed by 

 structureless bone, with a layer of osteoblasts round it. I have drawn 



