450 



CONTRIBUTION TO THE 



4. OS THE CROSSOrTEItYGIA. 



My object in alluding to the present subclass is to point out the ordinal characters of Polypterus as compared with 

 others of the same primary division, and of the Aetinopteri. The elucidation of this point depends first on the 

 explanation of the structure of the dorsal fin. This is one of the most extraordinary peculiarities of the genus in 

 question, and has never been brought into harmony with that exhibited by other divisions of fishes so far as I am 

 aware. The dorsal fin of Polypterus consists of a number of independent pinnules. Each of these consists of a strong 

 spinous ray supported by a single interneural bone. Prom the posterior face of the ray projects a number of soft or 

 cartilaginous rays, obliquely backwards, forming a vortical series. Thus results a series of vertical dorsal fins standing 

 on end instead of horizontally as in other fishes. The significance of this structure appears to be as follows : 



In the Dipnoi, we observe the archetypal fin almost realized. Each neural spine bears an interneural bone, and 

 oach interneural a basal radial bone. Beyond these are the numerous distal or cartilaginous radii, which greatly ex- 

 ceed in number the interneural bones. In the Dipnoan Ph,aneroplciiron, according to Prof. Huxley's figure 

 and description, each interneural supports a basal radial, and this in its turn a number of true radii. In the caudal 

 fin of Polypterus as we pass towards the dorsals we find the transition from the rays with jointed, forked extremity 

 continuous with the base, to the dorsal pinnules, to bo complete.* The basal portion of the ray loses its jointed 

 character, and more in front than posteriorly; the anterior surface becomes sculptured like the spines that support 

 the pinnules, and tho sides begin to project in an angle which, in the next spine, becomes a free margin. The apex, 

 separating at the first joint from its terminal, jointed portion, tho latter appears to issue from a fissure in its apex. 

 In the next spine, this fissure is prolonged downwards on the posterior face, and the jointed divisions are more widely 

 separated. Their number also is increased in each by the continuation of a fissuro which divided the upper ray 

 of the last pinnule above described, for about half its length to tho base. In the anterior pinnule the inferior, now 

 horizontal, ray is considerably nearer the base of the spine than in those nearer the tail. 



The preceding succession displayed by tho pinnules from behind forwards is no doubt a developmental one, 

 representing truly tho origin of the more specialized, since tho rays of tho tail are unquestionably in tho primitive 

 condition already stated to characterize tho dorsal fin of Phaneropleuron. It has also a complete analogy to the suc- 

 cession which we trace in tho types of the pectoral fin, commencing with tho simple axis of tho Dipnoi, and passing 

 to tho axis with lateral appendages or radii, seen in SelacJte (Gegenbaur 1. c.) and that traced below in the caudal. 



Carrying tho process on, in the line already commenced, if wo bring the radii to tho plane of tho basis of the 

 basal ray or spine, we have the number of rays in tho basal radial band greatly increased. Should they ossify, as 

 we have seen them to do in passing from the caudal to the dorsal fins of Polypterus, they would be undistinguish- 

 ablefrom the basilar rays, just as the primary rays of the pectoral fin in the Urossopterygia are undistinguishable 

 from that which was originally tho axial. I am not at present acquainted with any genus of this sub-class which 

 presents this character without modification; but if wo omit a large number of tho interneural spines and their 

 increased basilar rays, all but four in fact, wo have exactly tho structure presented by the members of the family of 

 the Ccelacanths, as figured by Huxley and Egerton, 1. c, orthogonera Trisiicliopterus, Gailacanihus, Undinct and Macro- 

 poma. In several of these tho interneural spines are united in pairs, thus supporting two dorsal fins. No doubt 

 others of tho order possess this structure, and somo intermediate between it and somo of the forms already described. 

 We havo thus a succcssional specialization in the structure of tho dorsal fin represented by three types, tho 

 first being nearest to tlio Dipnoi, the last (Polypterus and Trisiicliopterus) being specialized so as to resemble tho 

 paired fins, much more so in fact than anything observed in tho Aetinopteri. The three orders may by thus de- 

 scribed. 



Basal radii equal in number to tho interneural spines, with tho radii divergent, from their extremities. Inter- 

 neural bones numerous. Hapltstia. 



Basal radii equal in number to tho interneural spines ; tho distal radii divergent from the posterior margin of tho 

 former ; interneural bones less numerous. Cladistia fam. Polypteridm. 



" This Is rooro obvious In somo specimens than others ; one from Gondokoro on the upper Nile Is most Instructive In this respect, 



