1879.] NATURAL SCIENCES OF PHILADELPHIA. 231 



inner side of the forked plate. In most genera, they are con- 

 structed upon the same general plan. There are ten sacs (com- 

 pare PI. 17, Fig. 5) which do not connect with each other, disposed 

 in pairs, one pair to each ambulacrum, and each pair separated by 

 the lancet piece. Toward the visceral cavity, they are folded into 

 a number of longitudinal plications, which show neither pores nor 

 passages. The inner and outer folds alternate with each other, 

 and are distended at their closed ends. On approaching the apex 

 of the body, they coalesce to form two separate sets of tubes. 

 The tubes from the inner folds are formed by the adhesion of the 

 walls of the outer, and, vice versa, so that the folds which open 

 toward the visceral cavity give rise to the outer set of tubes, while 

 those opening in the opposite direction become the inner tubes. 

 The former terminate within coridors leading to the so-called 

 ovarian orifices, those of each two adjoining hydrospires of two 

 different ambulacra terminating in one orifice; while, the latter 

 communicate with an annular organ located against the inner wall 

 of the test and surrounding the oral centre. The number of folds 

 varies from three to nine or more. The walls of the sacs, which 

 were evidently composed of fine membranous substance, must 

 have been strengthened by the secretion of calcareous particles, 

 or they would not be found so well preserved; and they were 

 flexible since we find the folds in various degrees of expansion. 

 In Codaster, one of the earliest and probably one of the lowest 

 type of the Blastoids, and in Codonites, its subcarboniferous repre- 

 sentative, there is in place of the folded sacs a large number of 

 tubes placed side by side, and arranged parallel with the external 

 fissures or grooves. This structure of the hydrospires so closely 

 resembles that of some of the Cystideans that Billings proposed 

 to remove Codaster from the Blastoids and place it among the 

 Cystideans. This we cannot endorse, but we do agree with him, 

 that whatever may have been the functions of the calycine pores, 

 pectinated rhombs, and internal tubes in the Cystideans, those of 

 the parallel tubes or folded sacs in the Blastoids must have been 

 very similar if not identical. 1 



1 We have given above the description of the hydrospires in Pentremites. 

 Those of Granatocrinu8 and Nucleocrinus vary in some of the details. 

 One of us, who devoted much time to the study of the Blastoids, made a 

 large number of sections of the hydrospires, in difi'erent genera, and finds 



