A MONOGRAPH OF THE EXISTING CRINOIDS 547 



Comaster, he continued, further differs from all the comatulids with which we 

 are acquainted, and also from Solanocrinus, in the fact that the nervous cords are 

 not lodged in canals which perforate the pieces of the calyx, but lie freely on the 

 superior surface of the segments, the opposed terminal faces of which lie flatly against 

 one another. The muscles and ligaments lie along their concave inner sides and cover 

 in the freely exposed nerve cords. From the IIIBr axillaries onward, however, all 

 the segments have articular surfaces of the usual character, and are perforated by 

 central canals in which the nerve cords lie. This condition of the segments of the 

 calyx of Comaster is of great interest, for, besides being the normal permanent condi- 

 tion in the tessellate crinoids, it is the embryonic condition, so far as the position of 

 the nerve cords is concerned, in the comatulids. 



Carpenter said that these facts will suffice to show the very great differences 

 that exist in the skeleton alone between Comaster and the other comatulids, including 

 Solanocrinus, with which genus it has been united on account of the appearance of 

 the basals on the exterior of the calyx. In Solanocrinus, however, as in the other 

 comatulids, the radials are perforated by central canals for the nerve cords, and the 

 absence of this character hi Comaster woidd alone justify our referring these 2 forms 

 to separate genera, even if this were the only difference between them, which is by 

 no means the case. 



In 1879 Carpenter discussed Comaster at some length. He seems to have begun 

 to suspect that there was something wrong with Goldfuss's description and figure of 

 the calyx and associated structures, for he said that as Comaster has not been seen 

 by any naturalist since the time of Goldfuss its position must still remain in doubt. 

 He included Actinometra be7inetti in the list of recent comatulids to be referred to 

 the genus Actinometra. 



In 1881 Carpenter redescribed the type specimens of bennetti at Leyden. He 

 mentioned a specimen from Uea in the natural history museum at Stuttgart, and 

 another from the Pelew Islands in the museum at Copenhagen. At the same time 

 he described a new species, Actinometra peronii, based on a specimen collected by 

 Hoedt on the south coast of Ceram. This he believed to be identical with one in 

 the Bonn Museum which had been referred to by Miiller in 1849, and another in the 

 Paris Museum collected by Pe"ron and Lesueur in 1803, both of which he had per- 

 sonally examined. 



In 1882 Carpenter described 2 specimens in the Hamburg Museum, one from 

 Singapore and the other without locality. In the same year Prof. F. Jeffrey Bell 

 published specific formulas for both bennetti and peronii. 



In the Challenger report on the stalked crinoids published in 1884 Carpenter said 

 that he had been unable to find sacculi in bennetti. He divided the genus Actino- 

 metra, here used as including all the comasterids, into 8 groups (A-H) on the basis of 

 the characters exhibited by the arm division. The last group, in which the IIIBr 

 series are 4 (3 + 4), he called the group of Actinometra bennetti. He now had become 

 more than ever uncertain about the actual existence of a form with the peculiarities 

 described for Comaster, and he referred to it as "the doubtful Comaster." 



In 1887 Prof. Ludwig von Graff described the myzostomes from a specimen 

 from the Moluccas in the Amsterdam Museum. 



