72 PROCEEDINGS OF THE NATIONAL MUSEUM. tol. 41. 



the tendency toward this life is quite marked, as shown by the modifi- 

 cations of M. prattii. 



Whether there be a truly persistent proximal columnal or not it 

 certainly is true that in many genera at a certain stage in the growth 

 of the organism the proximal columnal becomes differentiated in a 

 more or less marked manner from the remaining ossicles of the column. 

 The tendency is constantly toward a higher degree of specialization 

 in this regard, and no doubt the structure becomes considerably 

 affected by acceleration. A proximal columnal may only be safely 

 called a proxunale, I tliink, when fusion, more or less complete, has 

 taken place between it and the infrabasals. Such a fusion alone 

 may definitely preclude the possibility of an intercalation of colum- 

 nals immediately beneath the theca. The infrabasals in any case 

 must be small, if not practically obsolete. In consequence of this 

 fact the proximal columnal comes to lie within the basal circlet. 

 It is evident, however, that its contact udth the circumjacent basals 

 is not of the same nature as that of a centrale with its juxtaposed 

 plates. 



The proximale as it is shown by different genera is too well kno\\Ti 

 to warrant extended description. Carpenter (1884, p. 25) describes 

 the proximale as follows in the case of Khizocrinus: 



But the uppermost joint of all is of a different character altogether. It has a pen- 

 tagonal outline, and the surface, which rises gradually from the circumference toward 

 the center by five radiating ridges into an equal number of trapezoidal fossae that 

 receive the lower ends of the elongated basals. Here, therefore, we find the top stem 

 joint presenting the same character that it does in Apiocrinus and Millericrinus, and 

 entering to some extent into the composition of the cup, while the new joints are 

 probably intercalated below it. 



Such is the essential structure of the proximale in the Bourgue- 

 ticrinidae, with which we are immediately concerned. The "proxi- 

 male" as had b}^ other families referred to the Flexibilia will not be 

 discussed, as having but slight bearing on the origin of the centro- 

 dorsal. It is to be noted in the description above that the proximale 

 of Rhizocrinus does not present ' ' the same character that it does in 

 Apiocrinus and Millericrinus." In the case of Rhizocrinus fusion of 

 the proximal columnal with the infrabasals has apparently become 

 completely effective, whereas in the other two genera cited such 

 fusion is by no means universal. 



The evolution of tJie centro-dorsal. — Having briefly outlined the 

 ontogenetic development of the centro-dorsal in the case of Antedon 

 and discussed in a general way the structure of the proximale and 

 centro-dorsal, we may indicate the stages by which these structures 

 have been acquired phylogenetically. As above mentioned the gen- 

 erally accepted type of centro-dorsal will be dealt with first. 



We must of necessity begin with a crinoid in which columnals 

 were normally produced immediately beneath the theca. In all 



