138 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



majority of the c_vstidcans, and in such holotliurians as the species of the family 

 Psolidse, these primitive plates and spicules, at first ser\-ing merely to stiffen and to 

 protect the bod}- wall, in the course of phylogenetic development gradually became 

 gathered together into groups more or less definite in position, the grouping origi- 

 nally being contingent upon mechanical considerations resulting from the localizing 

 effect of the movements of the body wall, especially of the anterior portion. 



Such a gi"oui)iug and fusmg of spicules to form a definite skeleton is not with- 

 out a parallel in other invertebrate classes. In the Tubii)ora, or organ pipe corals, 

 the tubular skeleton, with its transverse platforms, is the result of a fusion of spic- 

 ides, and the remarkably solid axial skeleton of the red corals has the same 

 origin. It is only among the echinoderms, however, that a spicular skeleton 

 develops into a solid external armament or into a series of articulated braces. 



Skeletons of the sjjicule formmg type are found only among permanently fixed 

 or more or less strictly sedentary animals, though sedentary animals do not all 

 possess them; their existence appears to be entirely incompatible with muscular 

 activity. We thus have an excellent clue to the habits of the earliest ecliinoderms, 

 and especially of the earliest crinoids, as it is in this class that the densest skeleton 

 is found. 



The sponge or alcyonarian-like skeleton of the echinoderms is undoubtedly of 

 independent origin within the group, without further phylogenetical significance; 

 also it is probably a feature of the adult organism only, without a coimterpart in 

 the larva. It does not appear before the assumption of the radial sjmimetn,', and 

 was probably phylogenetically, as it is ontogenetically, coincident mth it. 



In the cystideans and in the plated holothurians, such as the species included 

 in the family Psolidse, the body skeleton is formed directly by a simple process of 

 segregation and development of the spicules in the body wall, governed purely by 

 mechanical considerations; but this is not the case in the echinoids or in the crinoids. 

 In these classes the ultimate origin of the plates is exactly the same, but the place 

 of origin of all the plates is always about the anterior end of the digestive tube, 

 from which position they have traveled posteriorly, so that they now surround the 

 opposite apex of the body, their paths along the body wall being marked by a trail 

 of redupHcations of themselves left in the line of passage. 



In the holothurians the fortuitousness of the primitive spicule ft)rming t}-pe 

 of skeleton is seen in an extreme development; for in the species of this class no 

 calcareous matter at all may be deposited, as in Pelagothuria, there may be scattered 

 spicules of the most prunitive type, there may be highly specialized spicules, or 

 there may be very definite plates. 



In the holothurians the dermal skeleton is merely a mass of diffuse spicules, 

 not segregated into plates; in other words, of the ancestral tvjie for the echinoderms. 

 In the echuioids definite plates are present, ahnost entirely enclosing the body; but 

 these plates are extremely primitive in character; they are differentiated in each 

 radial division into a central series, composed of a varjnng number of similar col- 

 umns (interambulacrals), and bordering series of which there is usually a single 

 row on either side of the central series. 



