FISHES. 



117 



is not bent upwards, but divides the caudal fin-rays into two 

 nearly equal portions, and the symmetrical tail-fin thus pro- 

 duced is said to be " diphy cereal." In the great majority of 

 the Bony fishes the tail- fin appears on inspection to be 

 divided into two equal lobes, and it is then said to be " homo- 

 cereal" (fig. 491, a). This apparent symmetry is due to 

 the fact that the spinal column seems to terminate in the 

 centre of a triangular bony mass, to the free edges of which 

 the fin-rays are symmetrically attached. In reality, how- 

 ever, the notochord is prolonged into the upper lobe of the 

 tail ; and as there is a much larger number of fin-rays below 

 the bent-up notochord than above it, the tail is truly unsym- 

 metrical in its fundamental structure. Lastly, in the Elas- 

 mohranchii, and most Ganoids, the tail is conspicuously 

 unsymmetrical (fig. 491, b), and is then said to be " hetero- 



Fig. 491.— A, Sword-fish, showing homocercal tail ; b, Sturgeon, showing the 

 heterocereal form of tail. 



cereal." In these cases, the lower lobe of the tail is con- 

 spicuously larger than the upper, owing to the disproportion- 

 ate development of the hsemal spines, and the spinal column 

 is prolonged into the upper lobe of the tail. 



In a recently published and important memoir, Professor A. 

 Agassiz has shown that in Pleuronectes and various other liv- 

 ing Bony fishes, the tail of the early embryo is rounded, and 

 is symmetrically developed at the hinder end of the straight 

 notochord (" leptocardial stage "). Soon the chorda becomes 

 arched upwards, and there appears the first trace of a sepa- 

 ration of the tail-fin into two portions, only one of which is 



