Division in the Echinodcnnata. 325 



isidicola was simply the result of an accidental natural in- 

 jury. The regularity with which it is manifested proves 

 sufficiently that this is not the case, but that we have to do 

 with a true spontaneous natural division, the object of which is 

 multiplication. The specimens that we possess of the other 

 species of Ophiothela are not numerous ; but I have ascertained 

 that the four species with which I am best acquainted present 

 an analogous peculiarity ; for together with regular individuals 

 with six (exceptionally five) arms, we find others in which the 

 three (or rarely two) pairs of radial plates on one side of the 

 disk are smaller than the others, and the corresponding arms 

 only developed in the same proportion : in the Japanese spe- 

 cies the four very small specimens at my disposal were all in 

 this unsymmetrical state of regeneration ; in the other species 

 it was comparatively rarer. I believe, however, that this 

 character is sufficiently general to allow us to reckon sponta- 

 neous division among the generic characters of Ophiothela. 



As has already been mentioned by Steenstrup*, Sarsf, and 

 myself J, the same phenomenon is observed in other small six- 

 armed Ophiurids, especially those of the genus Ophiactisj 

 which, like the OpMotheke, live upon corals and sponges ; but 

 I have never found any trace of it in the species of that genus 

 whicli have normally five armS; whilst, as far as I can judge, 

 it may be observed in all the species with six arms. As a 

 supplement to the brief remarks which I have published upon 

 this subject in connexion with Ophiactis Miilleri, Krebsit, and 

 virescens, I will here communicate the observations which I 

 have made more recently. From a sponge from the Red Sea 

 I extracted 16 specimens of Oj)hiactis Savignyi. Most of 

 them (of average size, with a disk 2-3 millims. in diameter) 

 are regularly furnished with six arms, which present no 

 striking difference in length. In some of them, indeed, two or 

 three arms on one side ai-e sliorter than the three or fom- others ; 

 but the difference is so slight that it would hardly be remarked 

 in other Ophiurids, and it might be supposed that these shorter 

 arms had been broken and sprouted afresh. (One specimen 

 has seven arms, one of which is distinctly shorter than the 

 rest.) But it is very clearly seen that a division has taken 

 place in the four larger and smaller specimens. In the smallest 

 of all (with a disk a little more than 1 millim. in diameter) 

 one half of the disk and the three corresponding arms are en- 



* Forhaudl. ved de Skand. Naturforsk. syvende Mi)do i Cliristiania 

 (1857), p. 2.m 



t Bidrap til Kundskab. oni ]Middolliavets Littoralfauna, i. p. 1)7. 



X Addilanieiita ad histnriaui Ophiuridannn, ii. pp. 127, 12'J, and 14G, 

 tab. 4, fij,'. hd, iii. p. 38, 



