CLASSIFICATION IC>7 



This evident inadequacy of the definition has led to various proposed 

 remedies, such as placing the Pentacrinidae under the Fistulate Inadunata, and 

 the comatulids together with the Apiocrinidae as a subdivision (Pinnata) un- 

 der the Flexibilia. None of these has proved satisfactory, least of all the last, 

 for the lack of any sufficiently definable connection between the so-called Pin- 

 nata and the Paleozoic Flexibilia. It is true the pliant calyx of the latter recurs 

 in the pelagic comatulids, Marsupites and Uintacrinus, and the close lateral 

 union or partial incorporation of lower brachials is found to some extent among 

 the Apiocrinidae, Pentacrinidae, and some comatulids. But it has become 

 increasingly evident that the Flexibilia were a specialized group, derived from 

 the Inadunata, and ending like the Camerata with the Paleozoic. 1 



While there is no valid ground for any such divisions as Palaeocrinoidea 

 and Neocrinoidea, as proposed by Wachsmuth and Springer and by Carpenter, 

 and afterward abandoned, yet it cannot be denied that with the exception of 

 the Triassic Encrinus and the Plicatocrinidae the known crinoids of Mesozoic 

 to Recent time have an assemblage of features by which they are broadly dis- 

 tinguished from their Paleozoic ancestors. It is believed that this may be 

 expressed under the group Articulata as enlarged by Johannes Midler, dis- 

 tinguished not by any single character peculiar to itself, but by the fact that a 

 large number of characters belonging to different groups of Paleozoic crinoids, 

 and by which they were differentiated, have become associated, fixed and gen- 

 erally constant in this. It is by the combination of the following characters, 

 therefore, that the definition of this group, as herein given, becomes logically 

 effective : 



1. The predominance of the stem as a source of differential characters. In the older 

 crinoids the modifications marking the larger divisions occurred wholly in the calyx, and 

 with the culmination of these calyx modifications the types themselves became extinct. In 

 the survivors the general type of calyx became more fixed by decrease in size and the conse- 

 quent elimination of disturbing elements, while the larger changes in the structure of the 

 class took place in the stem and its attachments, which furnish the chief characters for the 

 later major groups. 



2. The mode of union of the radials with the plates they bear by complete muscular 

 articulation. This was the character chiefly relied upon by Miiller, and while subject to 

 modification or exception in a few genera, and to loss in senile stages of living forms, it 

 holds good throughout the group. And this is also the strongest character of the latest sur- 

 viving family of the Paleozoic crinoids, which passes on to the Triassic, — the Poteriocrinidae. 



3. Open mouth and food-grooves. 



4. Axial or dorsal canals extending into the arms within the calcareous substance of the 

 plates. 



5. The uniserial, and complete absence of biserial, character of the arms. 



1 The nearest approach to a relation between this group and the later crinoids has been worked out by 

 Mr. Clark, in connection with the second part of his Monograph now in preparation ; he finds in the larval 

 stage of the Recent comatulids before the appearance of the first pinnules and of the cirri, and in that 

 stage only, remnants of the ancestral structure of the Palaeozoic Flexibilia. 

 8 



