UINTACRINUS: ITS STRUCTURE AND RELATIONS. 51 



null free ossicles, are the same in both. And they are entirely different 

 from the parallel structures in mnny species of Antedon, and the recent 

 PentacrinidtB, in which the jjinnules are short and straight. There are 

 some species of Antedon, however, which seem to have pinnules much 

 like these. 



I have before ine both alcoholic and dried specimens of several of 

 the best known species of Actinometm, including Act. jMucicirra, Act. solans, 

 Act. carinata, Act. meridionalis, and notably the protean Act. fulchclla from 

 the West Indies, the latter being kindly sent me from the Museum of 

 Comparative Zoology at Harvard College, from the collections made by 

 Mr. Agassiz on the " Blake." In all of them the above mentioned 

 characters are strikingly shown. In Act. ipidcliella the anal tube is large, 

 and upon it, as well as scattered at different places on the disk, are 

 strong, irregular plates, all embedded in the membrane, resembling pustules, 

 and none of them suturally connected. ^ 



The unstable character of the base of Uintacrinus has been shown by the 

 facts hereinbefore given. It is most interesting and significant to note that 

 a somewhat similar instability exists in the base of Actinomefra. The centro- 

 dorsal, which in the adult Comatulae generally is the cirrus-bearing relic of 

 the stem, sometimes in Actinomctra loses its characteristic form, and becomes 

 a mei-e flat plate, without any trace of cirrus-sockets, lying flush with the 

 radials, Avithiii their circlet, and abutting by its outer sides against their 

 inner foces (PI. IV., Fig. 7). In Act. paucicirra the centrodorsal is reduced 

 to the condition of a flat, pentagonal plate, within the ring of radials.* In 

 these cases it might very appropriately be called a " centrale." In this 

 species the form of the centrodorsal undergoes a remarkable series of varia- 

 tions, among others from pentagonal to stellate t (PI- IV., Figs. 7, 8, 9). 



In Act. tyipica and sotne other species \ the sides of the centrodorsal are 

 resorbed, leaving clefts between it and the radials, and the remnant of the 

 plate has such a markedly stellate appearance that a distinct genus — 

 Phnnogenia — was proposed by Loven for the reception of these forms (PI. 

 IV., Fig. 10). In some cases the points of the basal rays are just visible 

 outside the angles of the centrodorsal (PI. IV., Fig. 8). These variations 

 in the base occur only in Actinometra,neyer in Antedon ; but the remarkable 



* Clmll. Rep. Comatuls, pp. 13-14. 



t Ibid., PI. LIV., Fi,?s. 3-9. 



+ Ibid., PI. LVIL, Fig. 1 ; and PI. LXV., Figs. 2-6. 



