I9l8.] E. H. WiiiTEiioirsE : The, Caudal Fins of Fishes. 1137 



adult fish ; in tlie adult, tlie dorsal and anal fins have not completely- 

 joined round the extremity and the vertebral column has exactly the 

 appearance of having been artificially abbreviated. 



It does not, however, necessarily follow that all the symmetrical forms 

 of the tail which follow the asymmetrical have once been heterocercal 

 and that this fin has been lost entirely to be replaced by a gephyrocercal 

 form. It is possible that something of this nature occurred : both the 

 later heterocercal and diphycercal forms were descendants of a primi- 

 tively symmetrical (protocercal) type ; at a certain period, a divergence 

 took place, some forms adopting the heterocercal tendency, while others 

 continued in the line of their ancestors and retained their symmetry. 

 This may be represented by a simj)le diagram (fig. 1). 



Fig. 1. 



It should be noticed that this does not necessarily regard all diphy- 

 cercal fins as protocercal ; a reduction of the terminal axial elements 

 might proceed by which the original caudal element might be eliminated 

 and the dorsal and anal elements made to contribute to the new caudal 

 fin. Thus the question of the symmetrical fin is left open, and the fossil 

 diphycercal forms might be regarded either as protocercal or as gephyro- 

 cercal. 



This interpretation would appear to be a perfectly natural one ; 

 an order of things which might easily have occurred, for the modification 

 of such an organ is only the result of a change in habit ; it cannot be 

 expected that all fishes adopted the same form of locomotion in early 

 times, since it is not likely that all adopted the same change of habit. 



I have said that the development of heterocercy was due to a change 

 of habit, and it is necessary to enquire what change was consequent on 

 the adoption of the heterocercal caudal fin, or rather how heterocercy 

 could bring about any change. An explanation is not easy even if 

 possible. In this connection we may recall Ryder's theory of the use 

 of heterocercy ; he likens the use of the tail fin to the sculling action 

 of the boatman, who propels his boat by a side to side motion of a single 

 oar from the hinder end. The analogy is a good one and explains the 

 use of the tail as a propeller, but it still leaves it uncertain why the axis 

 turned upward and not downward ; why the ventral side was chosen 

 for enlargement and not the dorsal. It is a question mainly for the 

 physicist to answer, and innumerable enquiries I have addressed on the 

 point to physicists have all failed to obtain an answer. 



It is thought that early fishes were bottom dwellers, and that in the 

 attempt to explore the upper waters there was a mechanical stimulus, 

 the response to w^iich caused the ventral lobe to increase in order to cope 

 with the demand made upon it. The rearing of the trunk to reach 



