1885.] NATURAL SCIENCES OP PHILADELPHIA. 239 



of interradials contains not one alone, but invariably three or 

 more plates, placed side by side, all resting upon the upper faces 

 of the first radials. Only the middle plate, the one placed upon 

 the outer ends of two radials, corresponds with the first interra- 

 dial of other groups; the plates at the sides are accessory pieces, 

 and rank as interradials of the second and third row, respectively. 

 Species with a discoid base have sometimes five plates in the same 

 row, of which only the outer ones meet the second radials. In 

 the simplest form of Platycrinus, the middle plate connects 

 directly with the proximals, and at the azygous side with anal 

 plates. In most species, however, the first row is succeeded by 

 other interradials, which either connect laterally with their fellows 

 of adjoining sides, forming with them a continuous belt around 

 the peristome (PL 5, fig. 9, and PL ?, fig. 6), or are separated by 

 radial structures. In either case there is no dividing line 

 between the plates of the outer and inner rows, and the upper 

 rows always rest against the proximals. The case is the same in 

 Marsujriocrinus ( PL 8, fig. 7), Hexacrinus, Dichocrinus and Tala- 

 rocrinus, and similar in Coccocrinus and Culicocrinus, which we 

 shall discuss farther on. 



In the organization of the Actinocrinidse, Melocrinidse, Euca- 

 lyptocrinidse, Rhodocrinidse, Glyptasteridse and Reteocrinidag, 

 the interradials form even more important parts than in the two 

 groups above mentioned- That here the plates of the ventral 

 side form a continuation of the interradials at the dorsal side, is 

 clearly indicated in genera whose arms are given olf in clusters, 

 or in which the rays are formed into lateral extensions. In 

 such species, the interradial series are not disturbed by so many 

 radials, nor by interaxillary plates, and the interradials decrease 

 in size gradually all the way from the first interradial up to the 

 proximals. In species, however, in which the arms are arranged 

 in a continuous ring, the interradials decrease in size more or less 

 from the poles toward the periphery. This decrease in the size 

 of the plates toward the equatorial regions is easily explained by 

 the extravagant increase of arms in those species, and by the 

 nature of the interradial plates, which, as stated, are accessory 

 pieces, filling up spaces. An occasional interruption of the 

 series, therefore, is no proof that the two sections represent 

 different elements. 



In some genera the interradials of the ventral side are exceed- 



