342 Dr. J. W. Gregory — On Pakeozoic Starfishes. 



been accepted by most palaeontologists who have referred to the 

 group. Thus they were adopted by von Zittel/ who simply 

 redefined the genera, and by Stiirtz,^ whose important papers 

 included the first attempt at a formal classification. 



The character on which Bronn's groups were based is the 

 arrangement of the ambulacral ossicles. In the Encrinasterite 

 (or Paleeasteroidea as they are sometimes called) the ambulacral 

 ossicles are said to be alternate, whereas in the Asteriaa verse or 

 Euasteroidea they are said to be opposite in position. The alternation 

 of the ambulacral ossicles is no doubt the primitive arrangement ; 

 and it is a character of great importance among opbiurids, for 

 so long as it was retained the union of the pairs of ambulacral 

 ossicles into vertebral ossicles was impossible. But among the 

 asterids the character is not essential ; for the ambulacral ossicles in 

 each pair never fuse, but remain as independent bars, meeting at 

 their narrow ends. The alternation may at any time result from 

 growth pressure. In some starfish the ambulacral ossicles in one 

 part of a ray may be alternate, while in another part they may 

 be opposite. The character is, moreover, difficult to use, as in some 

 asterids the two series of ambulacral ossicles appear to have 

 a certain amount of movement past one another during the lateral 

 bending of the arm. 



The alternation or opposition of the ambulacral ossicles is 

 therefore not a good character for the primary subdivision of the 

 Asteroidea. That the older asterids had alternate ambulacrals is 

 only to be expected, for that was naturally the primitive arrange- 

 ment. It was gradually lost as the ambulacrals became reduced 

 in size and the arms more flexible. The artificial nature of the 

 divisions based on this character is shown by the action of Stiirtz, 

 who divided several of his families into two parts, and placed one 

 part in the Encrinasterise and the other part in the Asterite (or 

 StelleridsB) verge. Thus his family PalEeechinasteridse comprised 

 the genera EchinastereUa and Palasteriscus. The former he placed 

 in the Encrinasterise and the latter in the Asteriaa verse. 



Nor is it possible to use the alternation or opposition of the 

 ambulacral ossicles as a basis for the divisions Palseasteroidea and 

 Euasteroidea, for Bronn, Yon Zittel,^ and Nicholson^ have correctly 

 included some of the Paleeozoic asterids in the Asteri^ verae. 



Bronn's classification appears, therefore, neither morphologically 

 nor historically justifiable, and a more convenient system is 

 desirable. In 1889 Sladen '" founded the two groups, the Phanero- 

 zonia for asterids having large marginal plates and the dermal 



1 K. von Zittel : Handb. Palaeont., vol. i (1879), pp. 451-4. 



^ B. Stiirtz, "Neuer Beitr. Kennt. palaeoz. Seesterne " : Palscontogr., vol. xxxyi 

 (1890), pp. 246-7. " Uber verst. uud leb. Seest." : Verb, uaturh. Ver. preuss. 

 Rheinl., ser. 5, vol. v (1893), pp. 76-7. 



=* K. von Zittel : Handb. Palaeont , vol. i (1879), p. 454. 



* H. A. iM'ichoIson : Man. Pal., vol. i (1889), p. 397. 



^ "W. P. Sladen, "Report on Asteroidea collected by 'Challenger'": Eep. 

 Chall. E.xp., Zool., vol. xxx, pp. xxviii and xxxiv. 



