Evolution and Symmetry in the Order o f the Sea-pens. 133 



the genera Cavernularia (some species), Actinoptilum, Echinophilum, and 

 the aberrant Eenilla, that we find that the axis is absent, and in Cavernularia 

 (some species), Veretillum, and Lituaria that we find it incomplete. 



(4) In the study of the Veretillidse, Kiikenthal discovered that the system 

 of endodermal canals in the rachis has a closer resemblance to that of the 

 other Aleyonacea than that of any of the higher sea-pens. 



(5) In this connection, a small piece of evidence, but not an unimportant 

 one, is found in the genus Lituaria, a genus closely allied to Veretillum, but 

 occasionally showing incipient bilateral symmetry in the presence of a 

 short groove on one side of the rachis free from autozooids, similar to the 

 groove seen in Echinoptilum, and more rarely in Actinoptilum. In this 

 genus only, we hnd, interspersed with other spicules of the types found in 

 other Veretillidse, a number of thick spicules of the shape of a dumb-bell, 

 with sharp tubercular spines standing from each side of the swollen 

 extremities. This type of .spicule, known as the twin star (" doppelsternige," 

 Kolliker) or '"' capstan " spicule, is not found in any other genus of the 

 Pennatiilacea, but is a common type in the Aleyonacea. This fact is 

 significant, because the Pennatulid spicules, except in Lituaria, are quite 

 different to those found in the Aleyonacea. The thin flat flakes of Yeretillum, 

 the smooth flat rods of Cavernularia. the smooth cylindrical rods of 

 L'mbellula and Pteroeides, and more particularly the remarkably specialised 

 three-flanged spicule of Pennatula and several other genera, are quite 

 jjeculiar to the Pennatulacea, and could not be confounded with any type 

 of spicule found in the other Alcyonaria. The conclusion seems to be 

 inevitable that the types of radially symmetrical Pennatulacea are not 

 degenerate, but do represent the nearest approach to the ancestral type of 

 which we have any knowledge. 



The evidence in favour of the view- that the radially symmetrical Penna- 

 tulacea are the most primitive appears to me so strong that it is worth while 

 to consider how they arose from the Alcyonacean stock. Such consideration 

 must necessarily be purely speculative, because at present no species has been 

 discovered that can be regarded as intermediate in structure between the 

 Pennatulacea and the other orders of the Alcyonaria and because the ontogeny 

 of the Pennatulacea so far as it is known does not really shed any light on 

 the matter. There seems to be little doubt that the ancestral form of the 

 sea-pens was colonial in habit. At any rate there is no evidence derived from 

 embryology or morphology to show that the origin of the stock dates back to 

 the original solitary Alcyonarian polyp. If we are justified in making this 

 assumption, as all previous writers have done, there are two hypotheses as to 

 the further evolution. Either the Pennatulacea are derived directly from a 



