OPHIURANS OF THE PHILIPPINE AND ADJACENT WATERS. 123 



All have large radial shields. The arm spines are rather numerous, 

 from seven to nine, rather long in the three first species, without, 

 however, being very long, and decidedly short in O. codonomorpha. 

 The dorsal plates of the disk are small, and sometimes bear rather 

 large granules (O. leucorhabdota) , sometimes small club spines (0. 

 eurypoma), or even a mixture of granules and true spines (O. 

 hylacantha) , or they are almost completely bare (O. codonomorpha}. 

 The upper arm plates, of various forms, are not truly in contact at 

 the base of the arms except in 0. leucorhabdota and 0. hylacantha. 



It is thus seen that tkese four species show very different characters, 

 and they certainly do not represent a natural assemblage. Ophia- 

 cantha eurypoma and O. hylacantha approach most closely the type 

 of the genus Ophiophthalmus. Ophiacantha hylacantha, in which 

 the disk carries both spines and granules, these latter passing onto 

 the distal edges of the upper arm plates, may be left in this genus, 

 as well as O. eurypoma and O. codonomorpha. But O. leucorhab- 

 dota, with its peculiar upper arm plates, with its numerous mouth 

 papillae, and with, or so it seems to me, a few tooth papillae, differs 

 markedly from all the other species. Would it not better find its 

 place in the genus Ophiomitra in the restricted sense ? 



There remains two species forming a third group; these are 

 Ophiomitrella placida and Ophiomitra miwophttax. I should be 

 disposed to place them in a genus different from that including the 

 preceding, on account of the form of the adoral plates. 



In my memoir on the ophiurans of the Siboga expedition, in dis- 

 cussing the genera separated by Verrill from the old Ophiomitra, I 

 expressed myself on the subject of the genus Ophiomitrella (Kcehler, 

 '04, p. 123) as follows: 



Verrill says in the characterization of this genus that the mouth shields 

 touch the first side arm plate; but the Siboga has found two new species, 

 which I shall describe further on under the names Ophiomitrella placida and 0. 

 moniUformis, in which the adoral plates are broadened outwardly into a lobe 

 which separates the mouth shield from the first side arm plate. It would 

 therefore be necessary either to maintain the genus Ophiomitrella strictly 

 within the narrow limits assigned to it by Verrill, or, broadening these limits, 

 to admit the species with the adoral plates extending between the mouth shield 

 and the first side arm plate. But it is indubitable that when Verrill established 

 his genus Ophiomitrella, no species were known which could be assigned to it 

 in which the adoral plates show this arrangement; he therefore did not 

 have to take it into account. I have thought that it would not be unseemly 

 to place in the genus Ophiomitrella the two species discovered by the Siboga, 

 modifying the diagnosis of this genus by stating that the adoral plates may 

 or may not separate the mouth shield from the first side arm plate. The 

 remarks which I have given above on the form of these plates in the genus 

 Ophiacantha show that this form is not perhaps so constant as Verrill thought, 

 and the genus Ophiomitrella is quite sufficiently characterized in spite of the 

 slight modification which I propose to introduce into the original diagnosis. 



