238 THE CEINOIDEA CAMEEATA OF NOETH AMEEICA. 



municates with the perforation in the arms ; axillary plates well developed ; 

 arms round and generally set at right angles to the body." He took the 

 upper openings, i. c, the central perforation following the median line of the 

 appendages, to be " efferent tubes," carrying off the water used for respira- 

 tion, which he thinks in other groups is performed by the anal tube. 



From numerous specimens in our collection, some of them with all of the 

 two kinds of appendages preserved to their full length, we are enabled 

 to fully confirm the opinion of Meek and Worthen that the smaller, pinnule- 

 bearing appendages are arms, and that the stouter, tubular ones are struct- 

 ures unlike those of other Crinoids. What the functions of these tubes may 

 have been can only be conjectured, but they were probably not identical 

 with those of the cirri, as Meek and Worthen supposed ; and we are inclined 

 to think, from the fact that their canals communicate with the subtegminal 

 galleries at the inner floor of the ventral disk, that the functions, as sug- 

 gested by Grenfell, were respiratory, and that the canals performed a similar 

 office to that of the respiratory pores of Batocrinus and the spiracles of the 

 Blastoids. 



Meek and Worthen, in separating the European species from the Ameri- 

 can, were probably not aware that the coalesced appendages are actually 

 pairs of distinct tubes, each one having a canal of its own, and the canals 

 of the same pair communicating with different ambulacra, in a way similar 

 to that of the paired spiracles in the Blastoid genus Pentremites. It is in this 

 respect exceedingly interesting that in Gilhertsocmius fiscelhis the posterior 

 appendages are simple, and are widely separated by a row^ of anal plates, 

 exactly as in the European species ; while those of the four other sides are 

 united at the base. The species thus represents at two of its sides the 

 European form of the genus, and at the other three the American. 



Another interesting fact in the developmental history of the genus is 

 that in the species of the Burlington limestone the consolidated tubes are 

 composed of the single cylindrical plates of two simple tubes united laterally 

 by suture. In the transition beds between the Burlington and Keokuk 

 groups appears the rare species G. ohovatus, in which the two rows of plates 

 composing the double tubes of its predecessors are roofed over by four more ; 

 and this character becomes constant in the Keokuk, where the genus is 

 found quite abundant in some localities. 



The genus Rhodocrinus^ which has been so frequently confounded with 

 Gilhertsocmius, has arm openings only, the tubular appendages being un- 

 represented ; otherwise the two genera cannot be distinguished. 



