CLASSIFICATION I IO, 



are often pulled off and remain attached to the stem, which have given rise to 

 the idea of a persistent fusion of the infrabasals with the top columnal; this 

 has proved to be not universal in the Flexibilia, and where it exists the sounder 

 view seems to be that it represents merely one retrogressive stage pointing 

 toward the atrophy or disappearance of the infrabasals. Only in Clidochirus 

 are the characters offered by the infrabasals thoroughly unreliable. 



Within the Ichthyocrinidae the character of reduced infrabasals brings 

 together a number of genera which have always seemed to be related by a 

 certain marked similarity of habitus, yet which differed so much in details such 

 as the presence or absence of interbrachials or of a radianal, and in the mode 

 of arm branching, that I was for a long time unable to frame a diagnosis that 

 would satisfactorily hold them together. The common character producing 

 this similarity of habitus lies in the shape of the rays, which continue more or 

 less to expand from the radials up, the successive brachial series tending to 

 widen distally, so that each ray, taken from the lower angle of the radials to 

 the level of greatest expansion of the crown, is regularly fan-shaped. Thus 

 the increase in diameter of the calyx is accomplished by the brachials them- 

 selves, often without the interposition of interbrachial plates. In marked con- 

 trast to such forms as Lecanocrinus, Pycnosaccus, etc., the radials are not the 

 largest plates in the ray. It was only after I had succeeded in ascertaining the 

 actual structure of the infrabasals in Dactylocrinus, W achsmuthicriniis , and 

 Synerocrinus, that the importance of this associated character became apparent. 

 The infrabasal structure is also correlated with a more or less concave base, 

 which is found in most of the genera, the infrabasal elements being completely 

 buried under the column, so that they no longer come in contact with the sur- 

 rounding water, or contribute essentially to the protection or enclosure of the 

 soft parts; therefore they tend to become atrophied from disuse (see an ex- 

 treme example of this on PI. XLI, fig. $d). Aside from the changes due to 

 atrophy, there is but little variation in the shape and position of the infrabasals 

 among eight of the genera composing the family. As a rule they occupy a nar- 

 row space surrounding the axial canal, and lie perfectly flat, with their distal 

 faces practically vertical. In W achsmuthicriniis there seems to be a slight ten- 

 dency to form a thin flange extending outward from the lower part, but rarely 

 reaching the outside (PI. XLIII, fig. 10) ; and Clidochirus, as already stated, 

 is irregular. 



This family represents a homogeneous group, which began in the Silurian 

 and continued through the Devonian into the Lower Carboniferous, running 

 a course substantially parallel to that of the other families, with consistent de- 

 velopment of radianal and anal and modification in arm branching, just as we 

 find in those families as well as in the dominant families of the Camerata. 



