328 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



inner end of the ray, which is occupied by a considerable depression forming the 

 peripheral end of the groove contained in the spoutlike process. In the natural con- 

 dition when the basals are in place and in connection with the radial pentagon, the 

 inflected edges of this process unite with those of the axial interradial furrow to form 

 an axial interradial canal. This terminates on the dorsal surface of the radial 

 pentagon by a small opening situated at the central end of the dorsal interradial 

 furrow, in which furrow the tertiary element of the corresponding compound basal 

 is received. The depression at the central end of the ray lies over this opening, 

 and thus forms a blind end to the axial interradial canal, just as the depressions on 

 the ventral surface of the centrodorsal of Antedon receive the blind ends of the 

 axial radial canals which open on the dorsal surface of the radial pentagon by the 

 five large openings." 



The basal rays are most uniformly developed and largest in the Comasteridse 

 (figs. 416, 419-422, 424-427, p. 321, 229-234, p. 247, 236-242, p. 249, and 243-249, 

 p. 251). So far as is at present known they occur in all the species of the family, 

 though frequently they are not long enough to reach the exterior of the calyx. 

 They may form long prisms with parallel sides, or may be more or less expanded 

 at the base or distally. Frequently the terminal portion bifurcates so that the tip 

 is bilobed (fig. 229, p. 247). 



In the species of the families Himerometridse (figs. 253-255, p. 253, and 256-258, 

 p. 255), Mariametridge (figs. 259-261, p. 255), and Colobometridse they are frequently 

 lacking; I have not found them in Himerometra martensi (fig. 254, p. 253), Amphi- 

 metra philiberti (fig. 258, p. 255), A. ensifer (fig. 256, p. 255), Craspedometra acuticirra 

 (fig. 255, p. 253), or in Mariametra sulcarinata (fig. 260, p. 255). When they do occur 

 they form slender prismatic rods which often do not reach to the exterior of the 

 calyx. These rods have parallel sides, and are more slender than similar structures 

 in the Comasteridse. 



The basal rays of the species of Thalassometridse (figs. 267-273, p. 259), as a rule, 

 are small, like those of the Himerometridse or Mariametridas, or may be entirely 

 wanting. In Ptilometra (figs. 267, 271, p. 259) and Asterometra (fig. 268, p. 259) the 

 basal rays are only faintly indicated. They do not appear to be found as such, but 

 the radial areas on the ventral surface of the centrodorsal are delimited by more or 

 less numerous parallel grooves under the interradial angles of the radial pentagon. 

 In a very few forms, as in Stylometra spinifera (fig. 273, p. 259), however, they are 

 large and prominent. 



As in the Thalassometridse, the basal rays of the species of Charitometridse (figs. 

 274-279, p. 260) are, as a rule, small, or may be entirely wanting. Occasionally 

 they are large and prominent, as in Crinometra (fig. 276, p. 260). The largest basal 

 rays ever observed in any recent crinoid were in a young specimen of a species 

 of Charitometridse (fig. 415, p. 319). 



Basal rays are entirely absent from the species of the family Calometridse 

 (fig. 263, p. 257). 



In the family Tropiornetridse (figs. 264-266, p. 257) the basal rays are well 

 developed and have a regular distal taper. 



