332 



NATURE 



\yuly 25, 1878 



of the azygos and paired fins themselves. It remains to 

 speak of the supporting structures of the paired fins, the 

 pelvic cartilages or bones, and the shoulder-girdle. At 

 first it appears that a formidable objection against the 

 similar nature of the paired and azygos fins may be 

 drawn from the existence in the former of these sup- 

 porting structures (which serve in the pectoral region to 

 fix the pectoral fins to the axial skeleton), while no 

 such connection ordinarily exists with regard to the 

 azygos fins. 



We have seen, however, that in Pristis and Pristio- 

 phorus the dorsal fin becomes directly continuous with 

 the axial skeleton by a mass of cartilage large enough to 

 warrant comparison with the shoulder-girdle itself, while 

 it is more or less firmly united with the axial skeleton in 

 Rhyncobates, Sqtiaiina, Acanthias, Spinax, ChimcBra, 

 and Callorhynchus. It must be admitted, however, that 

 the attachments of the dorsal fin to the axial skeleton is 

 horizontal, direct, and continuous, while the structure 

 supporting the pectoral fin (the shoulder-girdle) extends 

 vertically, is arched in shape, and only abuts at one 

 end against the axial skeleton, while ventrally it joins its 

 fellow of the opposite side. These characters seem at 

 first to tell against the similarity of nature of the dorsal 

 and pectoral fins. But three things should be borne in 

 mind — (i) the pectoral fin-support could not continuously 

 adhere to the axial skeleton antero-posteriorily without 

 impeding the lateral flexure of the body in swiinming ; 

 (2) the pectoral fins join the body at too low, a level for 

 their support to extend in horizontally to the skeletal 

 axis ; (3) and did it so extend inwards in a straight line, 

 even obliquely, it would intrude upon the visceral cavity. 

 For these reasons the pectoral (and ventral fins also) 

 must (if they are to rest on a solid support to facilitate 

 their flapping motion) have a narrow connection with a 

 sustaining structure, which structure must not be directly 

 continuous, in a straight line, with the skeletal axis. 

 Moreover, to obtain a firm basis, this limb-support, if it 

 is attached obliquely upwards to the skeletal axis, must 

 have some point to abut against ventrally also. Thus 

 such support must assume the form of a limb-girdle. 



I think, then, that there is sufficient evidence to war- 

 rant a belief that the skeletal structures of the paired 

 fins of fishes (and therefore the limbs of higher verte- 

 brates also) are the result of the centripetal growth and 

 coalescence of a primitively distinct, parallel series of 

 cartilaginous rays, developed in a pair of lateral fins 

 similar to those* developed, and more or less coalescing 

 and centripetally extending in the median fins above and 

 below. 



But what about the limb-girdles themselves ? Mr. James 

 K. Thacher,^ of New Haven, Connecticut, has thrown out 

 the suggestion that the pelvic bones and cartilages of 

 fishes (and therefore limb-girdles generally) are also due 

 to the further extension inwards of such centripetal 

 growth. I regard this as a most happy suggestion, and 

 adopt it myself. The mystery of the limb-girdles is thus 

 satisfactorily explicable ; they are neither modified bran- 

 chial arches, extra-branchials, nor ribs, but parts siii 

 generis, due to the ingrowth of originally superficial 

 structures — exoskeletal hardenings which have grown 

 inwards and become endoskeletal. 



It remains to consider the question of the development 

 of the original digit-bearing limbs, cheiropteiygiufii, from 

 the primitive fin, or archipterygmm. 



Gegenbaur at first regarded the elasmobranch fin as 

 derived from a limb formed like that of Lepidosi?-en, but 

 he subsequently adopted that of Ceratodiis as the archi- 

 pteiygium, in which view Huxley coincides. The former 

 naturalist, however, considers the shark's fin and the 

 cheiropterygium as formed by the ail-but complete abor- 

 tion of the rays on one side of the ceratodus limb-axis, 

 with the simultaneous shortening and thickening of that 



See Trans. Connecticut Academy, vol. i i. 



axis into a metapterygium, while the rays of the other 

 side of the axis coalesce to form the meso and pro- 

 pterygium. The latter anatomist (Huxley), on the con- 

 trary, regards the ceratodus-limb axis as forming by its 

 progressive shortening (or drawing-in) the mesoptery- 

 gium of the shark's pectoral, and the limb-axis of the 

 cheiropterygium, the latter being perfected by the atrophy 

 of the proximal lateral rays and the hypertrophy of the 

 distal ones, the distal end of the axis becoming the 

 middle digit of the hand. Of these two views the latter 

 seems to me much to be preferred, but it demands the 

 unity of the centrale carpal ossicle, which now seems most 

 probably to have been primitively double, as it is so not 

 only in cryptobrancliiis^ but also in both limbs of three 

 species of Siberian Urodeles.^ 



I believe, however, that the limb of Ceratodus is 

 far from showing us a primitive form, but is, on the 

 contrary, a very special and peculiar structure, which is 

 carried to a still more abnormal development in Lepi- 

 dosiren. This view seems warranted by the theory of evo- 

 lution, according to which air-breathing vertebrates must 

 have been amongst later developments, and therefore 

 have post-dated creatures with limbs more or less like 

 those of Elasmobranchs and Teleosteans. The secondary 



Fig. 17. — Pectoral fin o£ Acanth'as (from Gegenbaur). i, propterygium ; 

 ms, mesopterygium ; wt, metapterygium. The line drawn through vtt 

 indicates the fundamental line of the archipterygium or Ceratodus limb- 



fringing rays of the central limb axis of these Dipnoi 

 may (as Peters pointed out as long ago as 1845) have 

 arisen like the secondary fringing rays of the dorsal of 

 the primary rays of the dorsal fin of Polyptenis. 



As to the formation of the cheiropterygium, I think 

 that there are some reasons which favour the acceptance 

 of the propterygium as the part in Elasmobranchs which 

 has most relation to its primitive axis. Such are (i) the 

 preaxial position, in the limb, of the line of the Propte- 

 rygium — which is the line of support needed for the fore- 

 Umb of a quadruped which necessarily extends preaxiad, 

 distally ; (2) the apparently complete atrophy of the 

 mesopterygium in Chiloscylliuin and its partial atrophy in 

 Polyptenis, and other forms ; (3) the large size of the 

 propterygium in Chiincera, Callorhynchus, Cestracion, 

 Scyllitim, and Pristiurus. 



On the whole, then, I feel much persuaded that verte- 

 brate limbs have been formed as follows : — 



I. Two continuous lateral longitudinal folds were 

 developed, similar to dorsal and ventral median longi- 

 tudinal folds. 



' See " Morphol. Jahrbuch " vol. ii. 3rd Heft, p- 421, H. 29. 



