458 CHARACTERS OF PALKOZOIC FISHES COPE. 



ternal border of the flu ou both sides short aud undivided parallel rays 

 diverge towards the body. 



It is evident that this fin does not resemble that of Ceratodus, nor 

 that of any of the unibasal fins of the distichous type. The axial ele- 

 ment is not segmented, unless it be near to the extremity, nor is it 

 branched. It supports simple rays alone, and these at the internal edge 

 only. 



The section of the ventral fin shows, like the pectoral fin, rays on the 

 internal border, and also at the extremity. There are none on the ex- 

 ternal border, where the axial bones are close to the integument. The 

 base of the pubis is exposed. The extremity is concavo-truncate, is 

 coarsely cellular within, and is bounded by a thin external bony layer. 

 Like the pectoral fin, the greater part of the ventral axis is occupied 

 by a single element, which is rod-like, slightly constricted medially, and 

 truncate at the extremities. The proximal extremity equals less than 

 half of that of the i)ubis, but it constitutes the axis of the limb, as may 

 be readily seen by reference to the external form of one of them. This 

 fin is bent at this articulation, and is bent again at the extremity of the 

 axial rod, beyond which the apex tapers rather rapidly. The section 

 shows no second axial segment in the distal part of the main axis, but 

 two i)airs of nodules and distad of these two transverse rows of segments 

 of three radii, more distal segments being lost. Those of the proximal 

 row are longer than those of the distal one, and the external are the 

 most robust. The structure resembles that of a Batrachian tarsus 

 or carpus rather curiously ; but this may be due to the position of frac- 

 tures of the radii distad to the axial rod. Another cartilaginous, bony- 

 sheathed segment appears in this fin, which is half the dimensions of 

 the principal one and projects a little beyond it on its inner side, lying 

 parallel to aud-<^lose to it. It is followed, after an interspace, by seven 

 radii which lie closely parallel, and soon terminate, probably owing to 

 injury. Each is divided into two or three segments, but whether nor- 

 mally or abnormally can not be stated. There is no segment connect- 

 ing this one with the pubis, but there is a rounded extremity of possibly 

 a short stout segment opi)osite the extremity of the latter, within the 

 proximal extremity of the principal axial segment. Whether this fin 

 is unibasal or pluribasal remains therefore uucertain; but if there be 

 more than one elemental axis, there are not more than two. 



It remains therefore demonstrated that the fin structure in Megalich- 

 thys is very simple, and does not in the least resemble that of Polyp- 

 terus on the one hand nor that of Ceratodus on the other. It seems to 

 be intermediate in character between that of the latter genus and that 

 of Pterichthys, or perhai)s that of the imperfectly known Arthrodira. 



