[65] THE EVOLUTION OF THE FINS OF FISHES. 1045 



any data derived from a study of tbe development of the ])aired fins of 

 Teleosts. In other words, since we now know that the natatory fold, 

 Irom which the unpaired median fins are developed, is sometimes dis- 

 continuous, 1 see no reason why we should not expect to find the lateral 

 Unfolds discontinuous, as there are more reasons why they should be 

 so in the Teleost than in the Elasmobranch embryo. In fact, it would 

 appear that the iireater the lonjiitudiual extent of the uni)aired fins, in 

 ])ro])oition to the lenj> th of the body of the adult, the more likelihood 

 there is of finding a cojitinuous dorsal and ventral natatory fold devel- 

 o))ed in tlie larva, and rice rcrsa. The longitudinal extent of the paired 

 fins of Teleost fishes is less, vastly less, in respect to the number of sup- 

 ])orting rays than those of (he Elasmobranchs, and in consequence of 

 this difierence alone we should not be surprised to find lateral fin-folds 

 of considerable extent in the former. Viewed in this way, we may prove 

 too much for the doctrine of the origin of the paired fins from lateral 

 folds.'"* 



The principle stated by Balfour in the following words, " the limhs 

 are the remnants of covitinuou.s lateral /n?,sy"t may jiossibly need qualifi- 

 cation, if the preceding view is justified. The case of the embryo Tor- 

 pcflo, in which the continuous lateral folds are especially well devel- 

 oj)ed, does not seem to me to be conclusive for the reasons urged above, 

 because that form and its allies Narcine, Hypnos^ &c,, have developed 

 in the adult condition what has been denominated a cephalic fin. It 

 thus turns out that we here have the iufiuence of the principle verified 

 which is palpably operative in the case of the embryos of Gadus and the 

 Flounders. Both of these having the median fins of the adult approxi- 

 mated, the tendency would uaturally be toward the development of a 

 continuous median fin-fold in their larvaj, such as we find to be actually 

 the case. It thus beconjes evident that heredity may directly afiectthe 

 mode of development of the young. The opposite condition of discon- 

 tinuity in the Lophohranchil is similarl}- explicable. 



Whether the cephalic fins of Torpedo were primitively a j)art of the 

 pectoral or not, as Gegenbaur would urge, is immaterial, for the great 

 extent of the lateral fin-system in the adult must obviously afiect the 

 development of the early rudiments of the limbs in the embryo, and 

 tend to cause their first appearance as uninterrupted folds. 



In the adult Torpedo the cephalic fin does not possess all the traits of 

 a paired fin with straight, parallel cartilaginous supports (see Gegen- 

 baur, Das Kopfskelet der Selachier). In Hypnos mbniger, on the other 

 hand, the supporting rays of the cephalic fin, according to Haswell,:j: 

 are present as nearly parallel bars, resting in part on a pair of anterior 



* Development of the Spanish mackerel (CyUum muculaium). Bull. U. S. Fish 

 Commission, I, 1881, pp. IGO, KU. 



t Monog. on the Develop, of Elaiuosbranch Fishes, p. 102. 



t Studies on the Elasmobranch skeleton. Proc, Linn. Soc, of Ne\v SputJi Wales, 

 Vol. IX, 1884, pp. 105, 107, and Uti, 



