140 • BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



Among the holothurians we find many cases of large fenestrated plates provided 

 witli inwardly projecting processes, wliich are probably primarily compounded from 

 several smaller primitive plates and underlying spicules (fig. 70, j). 127). 



Tlie cl('vi'l(ii)ment of the large crinoiilal plates, wliich are of quite (hfferent 

 pliylogenetical significance, is fundamentally a continuation of just such a contli- 

 tion, the original plates as formed growing inward by means of long spicular 

 outgrowths which anastomose accorthng to a definite plan, and finally give rise to 

 more or less dense and very definite calcareous masses. 



Although in the earliest stages of the ontogeny phylogenetically far advanced 

 over the body plates of the cystideans or of the Psolidse, we apj)ear to have in the 

 crinoids, as in the other liighly calcareous echinoderms, evidence that the large 

 and definite plates, perfectly and characteristicaUv formed as they now are, arose 

 primarily through the union of several plates and a great development of spicules 

 just within them; in other words through a secondary, doubtless purely mechanical, 

 grouping of the elements of a primitive diffuse spicular skeleton. 



Had the ecliinoderms remained as inactive as the sj)onges or the alcyonarians 

 they, too, would doubtless have developed a similar dense, but diffuse and more or 

 less amorphous, spicular skeleton, and in them it would have been cliiefiy con- 

 fined to the outer body layers; but all of the ccliinoderm classes retained to a 

 greater or lesser degree their primitive bodily, if not their locomotor, activity, and 

 this activity has been sufficient to prevent, except in such inert groups as the cysti- 

 deans, and the Psolidse among the holotiiurians, any development from the original 

 spiculated skeleton other than a remarkable specialization, in certain cases, of the 

 indi^•idual spicules; indeed in the pelagic holothurians there has remained, or there 

 has been secondarily acquired, so much activity that it has resulted in the entire 

 suppression -of the skeleton. 



Autotomy. 



Autotomy of essentially the same type, frsquently more or less restricted to 

 definite specialized regions, is common to the echinoderms and crustaceans, and in 

 both it is developed to a very varying degree in different classes. It is quite possible 

 to regard the adolescent autotomy of the crinoids as comparable to a crustacean 

 moult. 



This process, strange as it is, really is not so anomalous as it would appear at 

 first sight. Except for a thin ventral band of perisome underlain by attenuated 

 extensions from the rmg systems about the mouth, the crinoid arms are composed 

 of solid calcareous plates developed by the growth inward of what is, reduced to its 

 lowest terms, a calcified cuticle. The brachials, being mostly composed of a solid 

 calcareous mass, are not able to increase in size with sufficient rapidity to meet the 

 exigencies imposed by the rapid larval growth, with the single exception of the 

 first (more rarely, in the more speciaUzcd types, of the first three), which has a much 

 less extensive skeleton than the succeeding. Development of the first brachial 

 ^\Tthout a corresponding development in those succeeding, or in the ligaments 

 between it and the second, inevitably results in an increasing tension in the hga- 

 ments the development of which is arrested, and which therefore are not able to 



