1887.] PAIRED FINS OF CERATODUS. 21 



me as conclusive against Huxley's view, as is the character of the 

 embryonic fin against that of Gegenbaur." 



Examination of the pectoral fin of Ceratodus shows that the 

 elements which I hold to represent the metapterygium and its rays 

 (figs. 5, 6, & 8, mt. & r.) are related to the postaxial border of the 

 second mesomere. There is at most a bare suggestion of a relation- 

 ship to the proximal mesomere {m.p.). In the hind limb this is 

 otherwise, for those parts of the skeleton which most nearly repeat the 

 characters of the presumed metapterygium of the fore limb are 

 unmistakably connected (figs. 1 & 4) with the proximal mesomere. 

 If they really represent the metapterygium and its rays as they occur 

 among the lower fishes, it is, I think, not unlikely, p tting together 

 these facts and those recorded by Davidoff, that while the metaptery- 

 gium has for the most part disappeared, the proximal mesomere may, 

 after all, turn out to represent the proximal end of that structure as 

 defined by Huxley, early differentiated and segmented off. 



The above suggestion, should it be substantiated, would explain 

 the fact that the proximal mesomere of Ceratodus is the only con- 

 stituent of the fin-axis whose characters are constant. It would 

 simplify our conceptions of the fins of the Ganoids and Dipnoi, and 

 bring into harmony the supposed divergent modifications of the 

 fins of opposite extremities ; while it would show the pelvic member 

 to be, on the whole, less modified than is usually thought. I am 

 disposed to think, moreover, that it receives support from the 

 absence of preaxial rays in connexion with the basal mesomere 

 of Ceratodus; from the complete exclusion of the mesopterygium 

 from connexion with the shoulder-girdle in Polypterus ; and from 

 the condition of the pelvic fin of that animal, already alluded 

 to, no less than from the marked tendency towards an increased 

 development of the proximal end of the pectoral metapterygium 

 among the living Ganoids. 



Still more suggestive is the condition of the basal elements of 

 a Protopterus pectoral fin represented in fig. 8 a. Wiedersheim has 

 (as I have already mentioned, p. 5) shown that the proximal piece 

 of the pectoral fin-skeleton of this animal bears ray-like elements. 

 He describes a smaller distinct ventral (postaxial) one and a larger 

 dorsal (preaxial) one, which is confluent with the main piece (proximal 

 mesomere as compared with Ceratodus). I have examined two 

 specimens ; in one of them the latter is much smaller than in his 

 example, while in the other (fig. 8 a) there is no trace of it. I can 

 only conclude therefore that it is a lobe of the basal mesomere, 

 variable in character. Not so with the former ; that is in both 

 perfectly distinct, being separated from the basal mesomere by a 

 fibrous tract, such as subdivides any two segments from each other. 

 In that specimen which was destitute of the preaxial process (fig. 8 a) 

 its characters are still further noteworthy. It is elongated and 

 shows traces of subdivision into two pieces, the basal one of which 

 is swollen and enlarged in common with the proximal mesomere 

 {m.p.), and from that it appears most clearly to have been derived. 

 The second segment of the axis is in relation with both the proximal 



