PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 781 



While Carpenter recognized certain similarities between the arms and cirri of his 

 Thaumatocrinus renovatus and those of the species of Pentametrocrinus {"Eudiocrinns") 

 with elongated cirri, and also noted the occurrence of orals and of evanescent plates 

 comparable to intcrradials in the peutacrinoids of Antedon, he could not divest himself 

 of the profound impression made upon him by the curious appearance of his specimen, 

 and especially by the process on the anal interradial, and so considered these similarities 

 as purclj' superficial. 



The new genus Thaumatocrinus as understood from the characters presented by 

 this young specimen was very widely noticed, and its peculiarities and curious cor- 

 respondences with various Paleozoic crinoids attracted much attention. It was even 

 made the type of a new familj', the Thaumatocrinidae, by Dr. F. A. Bather in 1899. 



In my preliminary studies on the comatulids in 1907 I found that the species 

 referable to the genus Eudiocrinus as understood by Carpenter which I had at hand 

 {varians, seiwperi and tuberculatus) were in all respects, excepting only in the number of 

 radials, closely similar to Carpenter's Promachocrinus naresi and P. abyssorum as I 

 understood them from analogy with a related form (borealis) which I had the year before 

 dredged in Japan. But Promachocrinus kerguelensis, the type of the genus Promacho- 

 crinus, was obviously, excepting for its 10 instead of 5 radials, very closely allied to 

 Heliometra. Adopting Minckert's genus Decameirocrinus to include Promachocrinus 

 naresi and P. abyssorum, as well as my new species borealis, I created the new family 

 Eudiocrinidae to cover Eudiocrinus, as understood on the basis of the three species 

 (varians, scmperi and tuberculatus) at hand, and Decameirocrinus, and placed Proma- 

 chocrinus (including P. kerguelensis only) in the Antedonidae next to Heliometra. 

 This arrangement was published on May 14, 1908, in a preliminary reclassification of the 

 comatulids from which the genus Thaumatocrinus was omitted, as, although it did not 

 appear to me to be nearly so anomalous as was commonly supposed, yet I could not 

 satisfactorily place it anywhere. 



After this paper was written, I received from Mr. Alan Owston through Mr. 

 Frank Springer a single specimen of a new species of Eudiocrinus from Japan which was 

 evidently allied to Semper's indivisus, the tj'pe of the genus. This specimen showed 

 conclusively that the Eudiocrinus of Semper, in spite of its undivided arms, is allied to 

 the genus Zygometra, and that Eudiocrinus indivisus, E. granulatus, and the new species 

 E. variegatus have nothing else in common with E. varians, E. atlanticus, E. seinperi, 

 E. japonicus, and E. tuberculatus. 



The difference of greatest importance between the Eudiocrinus indivisus type and 

 the group of species typified by E. japonicus was found to be that in the former a IBr 

 series is present, though the second ossicle is not axillary, whUe in the latter the arms 

 are morphologically comparable to the arms of other comatulids beyond the last 

 axillary or its equivalent. 



The genus Pentametrocrinus was created to include those species of Carpenter's 

 genus Eudiocrinus in which IBr series are absent (atlanticus, semperi, varians, japonicus 

 and ivberculatus), and the family Pentametrocrinidae was suggested to cover all the 

 comatulids in which the arms are similarly simple, embraced within the genera Penta- 

 metrocrinus, Decametrocrinus and Thaumatocrinus. 



Although the paper in which these conclusions were stated was published on 

 April 11, 1908, it was actually written some time after the one published on May 14, 1908. 



