THE STELLEROIDEA 249 



lysactinic Echinodenus in which there is an ambulacral groove. The 

 arms generally pass gradually into the disc, but in some cases are 

 sharply marked off from it. The digestive system generally has an anus, 

 and shares in the stellate disposition of the body. Pentameric repetition 

 is more often exceeded in this than in any other class, and asexual repro- 

 duction is not uncommon. Respiration diffuse. The madreporic aperture 

 is generally abactinal. 



This diagnosis at once sharply separates the Asteroids from all 

 Echinoderms except Ophiuroicls, between which, as we have seen (p. 238), 

 it is not possible at present to draw any precise line of separation. The 

 Asteroidea, however, always have an open actinal groove, whereas this 

 is exceptional among the Ophiuroidea ; the arms usually pass gradually 

 into the disc, and generally contain throughout prolongations of the genital 

 and alimentary systems. 



ORDER 1. Phanerozonia, Sladen. 



Asteroidea with large marginal plates, and with the dermal branchiae 

 or lymph gills limited to the abactinal surface. 



This order includes a group of starfish which began in the Cambrian 

 age and has lived on till the present time. Its members can be readily 

 distinguished by the large size of the marginal plates. The limitation of 

 the dermal branchiae to the abactinal surface is a more primitive con- 

 dition than that met with in the Cryptozonia. Embryological evidence, 

 and the greater importance of the order in the Palaeozoic and Mesozoic 

 eras also suggest that this is the simpler of the two orders of star- 

 fish. 



The Palaeozoic genera appear to be normal member? of this order, 

 and some of them may be included in existing families. They are, how- 

 ever, often all grouped together as an order, the u Pala j asteroidea," and 

 separated from all the later, or " Euasteroidea." The character on which 

 this separation is based is the alternation of the ambulacral ossicles in the 

 former, whereas they are said to be always opposite in post-Palaeozoic 

 Asteroids. This character is of great importance among Ophiuroids, for 

 when the ossicles are alternately arranged, they cannot be united into verte- 

 bral ossicles. But when the ossicles are narrow, thin plates, closely packed 

 into two series, one on either side of a ray, and when the separate ossicles 

 meet those of the other side only by their narrow ends, then alternation 

 is very likely to arise from growth pressure. In fact, one part of an 

 arm may have the ambulacral ossicles alternate, while in another part 

 they may be opposite. The character, moreover, is one on which little 

 reliance can be placed when cipplied to fossils, for a slight movement ia 

 sufficient to alter the relative positions of the two series. It is difficult 

 to explain the relative positions of the ambulacral ossicles in different 

 arms of the same Asteroid, except on the assumption that the two series 

 moved past one another during the lateral bending of the arm. Alterna- 

 tion of the ossicles was probably an original character ; but as the arms 

 became flexible with the reduction of the external skeleton, and as the 



