ECHINODERMATA 633 



(Echinopluteus), owing to the fact that the former of these larvae has 

 fewer arms than the latter and that in it the arms known as the " pos- 

 terolateral arms" are the largest and are directed forwards, whereas 

 these arms, if they are present in the Echinopluteus , are there small 

 and directed outwards or backwards. The larva of the Crinoidea has 

 no longitudinal band, but five rings of strong cilia around the body. 

 In the development of the Holothuroidea the Aiiricularia is succeeded 

 by a stage known as the pupa, in which the longitudinal band breaks 

 up and rearranges itself into a series of five transverse rings some- 

 what resembling those of the crinoid larva. The Bipinnaria of the 

 Asteroidea is succeeded by a Brachiolaria which differs from it in 

 possessing in the preoral region three processes by which the larva 

 can hold fast to objects. 



The larvae become transformed into adults by a metamorphosis 

 which differs in the several classes. In all it involves an alteration of 

 the position of the mouth, which in groups other than the Crinoidea 

 is removed to the left side, and in the Crinoidea to the posterior end, 

 taking with it the coelomic cavities of the left side. The fate of the 

 several divisions of the larval coelom has been described above 

 (p. 627). In the Crinoidea and Asteroidea the larva becomes fixed by 

 the preoral lobe at the time of metamorphosis, 2. fixation disc develop- 

 ing for the purpose. In crinoids the fixation persists, at least until the 

 adult is completely formed. In starfishes it is only temporary. 



The fixation of the sea lilies, and the fact that starfishes are fixed 

 when the bilateral symmetry of the larva changes to the radial sym- 

 metry of the adult, are interesting facts in view of the fixation which 

 is general in the other great group of radially symmetrical animals, 

 the Coelenterata. Radial symmetry is essentially the symmetry' of a 

 sessile animal, which is in the same relation with its surroundings on 

 all sides, whereas bilateral symmetry is that of a travelling animal, 

 which needs differentiation not only of the upper side from that which 

 faces the ground, but also of the fore from the hind end. It is likely 

 that at one time all echinoderms were fixed, and that those which are 

 now free retain the radial symmetry of their sessile ancestors. This 

 supposition is supported by the fact that the earliest known fossil 

 members of the phylum were fixed. 



For the rest, the Dipleurula and its metamorphosis suggest that 

 the early sessile echinoderms were descended from a free, bilateral 

 ancestor ; and the close resemblance between the Auricularia and the 

 Tornaria larva of Balanoglossus, together with the history of the 

 coelom (see p. 660), and the nature of the nervous system, indicate an 

 affinity between that ancestor and the Enteropneusta. 



