64 PROCEEDINGS OF THE NATIONAL MUSEUM. yol. 41. 



with the cosmopoHtan nature of many forms. The facts relative to 

 tlie distribution of the Crinoidea, on the other hand, offer many con- 

 vincing arguments in favor of the widespread freedom of the adult 

 crinoids. Inasmuch as these facts apply with equal force to all the 

 Pelmatozoa discussed in the paper, it has seemed best to postpone a 

 treatment of this phase of the subject until the different groups have 

 been discussed in detail. 



In such Crinoidea as those above noted in which a detached exist- 

 ence has been shown, there can be no question but that for a large 

 part of their lives the animals were capable of more or less free move- 

 ment. In the case of any given species where such freedom of move- 

 ment has been established, it is not held that every individual neces- 

 sarily enjoys the same freedom. In any species of Isocrinus, say, it 

 may be found that there are individuals the greater part of whose 

 lives is passed cemented or rooted to one spot, while others undoubt- 

 edly pass long periods of time during which there is no such fixation. 

 It should be borne in mind, however, that the status of such a free 

 type is not that of a sport, freedom being accidentally acquired and 

 inadvertently maintained. Rather it should be held that such a 

 type of life is, as one might say, optional with the crinoid, freedom in 

 such instances perhaps being accidentally acquired, but in most cases 

 being voluntarily maintained. In all such types freedom is the 

 expression of a general tendency toward the assumption of an 

 eleutherozoic habit and in all probability would not be perpetuated 

 were it not for this fact. 



Group II. 



As previously defined, this group contains those Pelmatozoa which 

 have lost all or the greater portion of their columns. In certain 

 fossil forms the stem is completely eliminated, not even the proximal 

 columnal being retained. In one very large, and perhaps the most 

 important division of the group, however, it is next to impossible 

 to draw a line separating the constituent members from the forms 

 characterizing Group I. In this division, the "Comatulse," the 

 proximal portion of the column is retained, which is, as a rule, pro- 

 fusely cirriferous. It seems to be a question as to whether these 

 genera should be placed in Group I or whether such a type as Milleri- 

 crinus prattii should be listed under Group II. Division has here 

 been made on the basis of the degree of specialization of the forms 

 involved. Thus, among the Comatulse a stem is only had in the em- 

 bryonic stages and is consistently lost as a column among the adult 

 crinoids. Among the ancestors of these forms, however, we come 

 eventually to a type which is indifferently stalked or detached and 

 hence in a way to be held as intermediate between Group I and 

 Group II. 



