the Crinoidca, Cystidea, and Blastoidea. 263 



cimens before mc instead of fip^urcs only, I should most proba- 

 bly institute a new genus for their reception. 



Our specimens of C. canadensis are well preserved, and 

 show the characters of the arms perfectly. After many careful 

 examinations under tlie microscope, I can state positively that 

 in this species the so-called " pseudambulacral fields " have 

 no ]iores. Tlie markings that have hitherto been mistaken for 

 ambulacral pores in Cadaster are not pores, but the small pits 

 or sockets which received the bases of the pinnuUe. The rays 

 therefore in this genus are not " pseudambulacral fields," in 

 the sense in which that term is used in descriptions of species 

 of Pentremi'tes, but simply recumbent arms, identical in struc- 

 ture with those of the Cystidean genera Gl?/ptoc}/stifes, Callo- 

 cystites, Ajn'oci/stites, and others. Tliey lie upon the surface 

 of the plates Avhich constitute the shell of the animals — not 

 imbedded in them, as in Pentremites. The large lateral 

 aperture is both mouth and vent, and the central opening 

 heretofore called the mouth is the ambulacral or, more pro- 

 perly, the ovarian orifice. As therefore Cadaster has the arms 

 of ApiocystiteSy the hydrospires of Pleurocystites, and the con- 

 fluent mouth and vent common to all Cystideans, I propose 

 to remove it from the Blastoidea and place it in the order 

 Cystidea. 



4. On the Genus Pentremites. 



In Pentremites the hydrospire is an elongated internal sac, 

 one side of whicli is attached to the inside of the shell, while 

 the side opposite, or toward the central axis of the visceral 

 cavity, is more or less deeply folded longitudinally. There 

 are two of these to each ambulacrum, attached along the two 

 lines of pores. There appears to be a fissure extending nearly 

 the whole length in the direction of the dotted line y (fig. 12). 

 One edge of this fissure is attached to the lancet plate, along 

 one side of the line of pores, the other to the shell, on the other 

 side of the row. The pores all enter the hydrospire through 

 this fissure. There are ten hydrospires, connected together in 

 pairs, each pair communicating with the exterior through a 

 single spiracle. The arrangement of the folds varies according 

 to the species. In P. Oodoni there are five folds, the outer 

 sides of which are close up to the inner side of the lancet 

 plate (fig. 13). In a specimen of P. ohesus, Lyon, nearly two 

 inches in diameter at the mid-height, the hydrospires extend 

 inward about three lines, the main body being about one line 

 from the lancet plate. There are five folds, each two lines 

 deep ; and thus, if the thin shelly membrane which constitutes 

 the wall of the hydrospire were spread out, it would have a 



