FOSSIL CRINOIDS. 139 



I have prepared the table on the following page showing the occurrence of the 

 genera having these two respective types of articulation, — the inquiry being 

 limited, as before stated, to the dicyclic Inadunata: 



Leaving out of consideration the rare Lecythiocrinus, the latest Palaeozoic 

 occurrence of the horseshoe facet is in Barycrinus, in the Warsaw (possibly 

 St. Louis), where it sometimes shows the remnants of a transverse ridge (PI. IV, 

 figs. 14, 15, 16). Neither this genus, nor Cyathocrinus, nor any other with round 

 facets, has been found in the Kaskaskia, where Eupachycrinus and its congeners, 

 with the perfect tran.sverse articulation, flourished in profusion, and from there 

 up into the Upper Coal Measures and Permian. 



Such being the general line of succession and order of development, it is not 

 to be expected that here, any more than in the case of other characters relied 

 upon as the basis of large divisions, we shall find any hard and fast boundary 

 line separating the groups represented by plans 1 and 3, neither morphological 

 nor stratigraphic. The two overlapped geologically, and we may expect to find 

 intermediate stages pending the disappearance of the one and the establishment 

 of the other. Thus in Cyathocrinus, which began in the Silurian, there are 

 occasional traces of fossae, and of an imperfect, discontinuous ridge; also in 

 Barycrinus. But these are irregular, occurring in only a few species; and the 

 facets show no tendency to become straight, but retain the deeply concave, 

 rounded form, for which these genera are so well known. Perhaps the best 

 example of a transition is found in Poteriocrinus, in which there seems to have 

 been a struggle to get rid of its specialization; it begins in the Devonian with 

 round facets and a slight trace of a ridge, but in the Carboniferous it has devel- 

 oped a very distinct, straight ridge, within a facet that is still relatively narrow, 

 much less than the width of the radial. When the facet comes to fill the entire 

 distal face of the radial, the form is called Pachylocrinus (ohm Scaphiocrinus) — 

 there being no other material difference between them, unless in the ventral sac. 

 But among the genera in which the complete articulation has become a fixed 

 character there are no exceptions, or tendencies to lose it. When No. 1 was 

 once established, the plan held absolutely, within its own genera and in general, 

 through a long range of geological time, until the present. 



In rearranging the genera of the dicyclic Inadunata under a phylogenetic 

 classification, in which morphological considerations were sent to the rear 

 (Lankester ZooL, III, 171), Mr. Bather thought that our great divisions into 

 Gyathocrinidae and Poteriocrinidae could not meet the needs of the phylogenist. 

 He therefore established two suborders, Dendrocrinoidea and Cyathocrinoidea, 



