1885.] NATURAL SCIENCES OF PHILADELPHIA. 303 



had contemplated, but we encountered great difficulties, as the 

 two forms run very closely from one into the other. Even the 

 ventral sac, the best distinguishing character, undergoes all 

 possible modifications. It dwindles down to almost nothing in 

 some of them, and its porous nature is sometimes very indistinctly 

 developed or even unrepresented. We regard these modifications, 

 as they, occur in palseontological times, as representing various 

 stages of development in the historj' of this group, and as good 

 generic characters, but do not attach to them the importance we 

 did before. We have, however, placed the genera in which a 

 ventral sac is developed as a group by themselves, to separate 

 them from those in which it is absent. The latter group, which 

 represents the larval form, w*ill be designated by us as " Larvi- 

 f omnia " the former as " Fidulata." 



Instead of the name Splncroidocriniclae, which is objectionable 

 for several reasons, we propose to use Camarata, under which we 

 have placed several additional groups. To the Articulala (nobis, 

 not Miiller or Miller), which we restrict to the articulated Palseo- 

 crinoidea, we refer the Ichthyocrinidre and Crotalocrinidae. If 

 there is any objection to re-establishing Midler's name, which 

 has been generally discarded, we might change Articulata into 

 "Articulosa." We think, however, we are fully entitled to adopt 

 the former, as the Crinoids which we refer to them are true 

 Articulata in Midler's sense. We place the Camarata, which we 

 regard as the typical form of the Palaeocrinoidas, at the head of 

 the list, the Articulata next, and the Inadunata, which in some 

 respects approach the Neocrinoidea, at the opposite end. 



The Camarata embrace all Palseocrinoidea in which the plates 

 of the test are solidly united by suture, and in which the lower 

 arm plates are incorporated by means of interradial plates so as 

 to form a part of the calyx. The underbasals are frequently 

 undeveloped. The basals of monocyclic genera are variable in 

 number, five being the exception. The primary radials consist 

 generally of three plates to each ray, rarely of two or four. 

 There is always at least one secondary radial, which ma}* give 

 off the free arms or support others, and frequentl}* radials of 

 higher orders. Interradials numerous, or not less than two; the 

 first one resting upon the sloping upper sides of the first radials, 

 or alternating with them. The interradials, together with the 

 interaxillaries and anal plates, separate the rays and their sub- 



