CLASS n ('YSTOIDEA 181 



st ri.le one another like the draw-tubes of a spy-glass. In other cases the upper 

 part, and occasionally, indeed, the entire stem, is composed of vertical rows of 

 alternating plates. These plates, as a rule (Dendrocytiites), enclose a large 

 ..titial space, which may be regarded as a prolongation of the body-cavity. . 



The Cystideans constitute the oldest and least specialised group of the 

 Pelmatozoa. Appearing first in the Cambrian, they develop a great variety of 

 forms in the Ordovician and Silurian, but become extinct before the close of 

 the Permian. While their own ancestry is obscure, it is highly probable that 

 from them have descended both the Crinoids and Blastoids. If, on the one 

 hand, the Aristocystidae, Spliaerodontidae, Camarocystidae, and Echinosphaeritidae 

 differ radically from the Crinoids in respect to their numerously and irregularly 

 plated calyx, or as regards the feeble development or even total absence of 

 their arms ; nevertheless, the Cryptocrinidae and the unfortunately ill-preserved 

 Cambrian genus, LicJierwides, evince a striking similarity, especially as concerns 

 the more or less regular arrangement of the calyx plates, and a certain approach 

 to radiation. In fact, so far as the construction of the calyx alone is concerned, 

 forms like Porocrin-us and Hypocrinus may be assigned with equal propriety to 

 . it her the Crinoids or Cystids. The presence of pectinated rhombs and 

 calycine pores, however, and the porosity of the calyx plates, are characters 

 which positively identify them as belonging to the latter class. 



If we can account for the derivation of Crinoids from the Cystideans on the 

 supposition that the calyx plates of the latter gradually took on a more 

 definite arrangement, while the loss of pores and pore-rhombs was counter- 

 balanced by a stronger development of the arms and the stem ; so, too, it is 

 possible to explain the origin of the Blastoids from the same source. Although 

 hydrospires are clearly wanting in the Cystids, nevertheless, other characters, 

 such as the recumbent attitude of the arms upon the sides of the calyx, or 

 their insertion in grooves on the ventral surface, as well as an approach to 

 ambulacra! areas in certain forms (Asteroblastus), predicate an intimate relation- 

 ship with the Blastoids. Probably the most notable similarities are presented 

 by the peculiarly modified families, Callocystidae and Agelacrinidae. Various 

 attempts have been made to affirm a connection between Agelacrinus and the 

 Astero&ka, and between Mesites and certain of the primordial Echinoidea (Cysto- 

 '/"/v'.s); but such hypotheses are scarcely warranted, since they proceed from 

 an over-valuation of mere external resemblances, which are in nowise indicative 

 of kinship. 



The classification of the Cystoidea remains in a very unsatisfactory condition. 

 This is owing in part to' the comparative scarcity of material, as well as its 

 frequently imperfect state of preservation ; and in part to our insufficient 

 understanding of many of the structural modifications. The classification of 

 Johannes Miijler was based primarily upon the structure of the calyx plates, 

 and three main groups were recognised : Aporitidae, Diploporitidae, and Rhombi- 

 /'//. These divisions, however, embrace a number of very heterogeneous 

 elements, and in nowise fulfil the requirements of a natural system. The 

 classifications proposed by Barrande and Steinmann are still more inadequate. 

 It is advisable, therefore, to follow the example of Neumayr and Bernard, and 

 merely arrange the different genera into families ; the establishment of larger 

 divisions being deferred until the group shall have been monographed. 



