on British Palceozoic Onnoids. 287 



lines. The radial plates are better developed relatively to 

 the orals than in the younger forms. Hence when the 

 calices are viewed from above they are seen to be only very 

 incompletely covered by the dome and project considerably 

 beyond its circumference, while distinct facets for a true 

 articulation with the first arm-joints begin to make their ap- 

 pearance (PL XVJ. figs. 9, 10). These are least marked in 

 the specimen shown in fig. 8 (PI. XVI.). In one or two 

 cases the facet is something more than a mere thickened rim 

 to the arm-groove, and shows traces both of a central canal 

 and of a dorsal fossa for the elastic ligament. The chief cha- 

 racter indicating the advanced condition of this specimen is 

 the inequality in the size of its radial plates, one of which is 

 axillary and has two arm-facets, which are not yet completely 

 developed. In the individual shown in fig. 10 (PI. XVI. _), 

 on the other hand, all the radials have well-developed arti- 

 cular facets, which are pierced by the openings of the central 

 canals ; but though the radials are unequal in size, none of 

 them is an axillary. The great reduction of the relative size 

 of the orals in this specimen is especially noteworthy. In two 

 other specimens, which also have much-reduced orals, there is 

 a remarkable inequality in the development of the arm-facets. 

 Thus, in the original of fig. 9 (PL XVI.) three of the radials 

 have well-marked articular surfaces for the first arm-joints, 

 while that of the fourth is very imperfect, and that of the 

 fifth altogether undeveloped, no trace even of an arm-opening 

 being visible. Fig. 7 (PL XV.) shows a similar irregularity. 

 The orals are relatively very small, and two of the radials 

 have distinct articular facets, whilst two others have small 

 cuboidal brachials still in contact with them. The fifth 

 radial, however, bears a very small and rudimentary brachial, 

 which is shown on the right of fig. 7, a. 



The originals of fig. 6, PL XV., and figs. 1 and 2, PL XVI.,^ 

 differ very considerably in size. xVpart from that, the chief 

 point of difference is the presence of oral plates in the latter 

 and not in the former. It does not seem to us a very im- 

 probable supposition that during the growth of the smaller 

 specimen to the size of the larger its orals would come to be 

 of such small relative size (as they djo in the development of 

 Comatula) as to be altogether lost in the fossil state, even if 

 they persisted during adult life. 



The large and small forms agree in so many points, espe- 

 cially the inequality in size of the radials, and the fact that 

 some of them may be axillary, that we do not see our way 

 to separating them specifically. It must be remembered that 

 these little fossils occur at various localities. Even the larger 



