[6] THE EVOLUTION OF THE FINS OF FISHES. 985 



The development of the rays seems to begin peripherally in the fin- 

 IV»lds in membrane and approaches the basilar interneural elements from 

 without, these latter being the most peripheral of the median series, 

 which are developed in true cartilage from mesoblast. The rays are 

 therefore sub-epiblastic, though their proximal ends may embrace 

 nodules of cartilage of epaxial and hypaxial mesoblastic origin. The 

 words peripheral, distal, and proximal will, therefore, be occasionally 

 met with as used above. 



The words concrescence, coalescence, are usually synonymous here as 

 applied to describe the blending of series of bars of cartilage. The 

 terms homonym and homonomous, segmental, and metameric are nearly 

 synonymous in some cases. Homonomous, homonomy, homodynamous, and 

 homodynamy are used in the sense defined in Gegenbaur's Elements of 

 Comparative Anatomy. The term serial homology (Owen) is equiva- 

 lent to the last of the above-mentioned words. Metamerism refers to 

 the segmented condition of the vertebrate embryonic, and even, in fishes, 

 to the adult axis, in which we find similar successive segments or somites- 

 The word urochord is used a few times to designate the membranous 

 part of the axis of the embryo when it is exserted beyond the hypural 

 cartilages, to distinguish it from the ossified urostyle of the adult. 



The term protopierygian, to designate the stage when the embryonic 

 fin-rays first appear, was chosen to designate that condition when it may 

 be said that true fins first appear. Pterygoblasts refers to the proto- 

 plasmic bodies from which the embryonic fin-rays are developed. The 

 word orthaxial is used to designate the archaic straight type of verte- 

 bral axis, which is not bent upwards at its posterior extremity. 



The word actinost as used by Gill applies only to the distal cartilagi- 

 nous or bony elements of the limbs of fishes, which support the rays, or 

 those more especially of Teleosts and Ganoids resembling them, and, 

 inasmuch as these elements are clearly those from which their isomeral 

 equivalents are evolved in the higher forms bj^ concrescence, their 

 homology throughout the Lyrifera, becomes apparent, though very fre- 

 quently fusion has occurred to form elements called the pro- meso- and 

 metapterygium. Eegarding the whole of the mobile axial skeleton of 

 the limbs of the Lyrifera as essentially homologous, I will call the dis- 

 tal parts which directly supi)ort the true rays actinophores. While the 

 Teleosts have had these elements much shortened as compared with 

 the same parts as found in the Kays amongst Elasmobranchs, such a 

 shortening is merely the result of extreme specialization growing out 

 of adaptation. The basilar interneural and interhsemal nodules in this 

 terminology become median epaxial or hypaxial actinophores. 



II. — The theory of the development of the median fins. 



The median fins of fishes normally present five well-marked condi- 

 tions of structure, which con-espond to as many stages of develo[)ment, 

 which in typical fishes succeed each other in the order of time. A sixth 



