TRANSACTIONS OF SECTION D. 917 



The Pelmatozoa represent a grade of structure passed through by the ancestors 

 of all echinoderms in their passage from the bilaterally symmetrical Dipleurula 

 stage to the radiate stage. The essential feature was fixation by the primitive oral 

 (anterior) end, followed by the upward passage of the mouth. The foregut in 

 that passage involved other organs of the primitive left side, especially the left 

 anterior coelom and its offshoot the left hydrocoel. The primitive pelmatozoon 

 had not acquired radiate symmetry. This was induced by the extension from the 

 central upwardly directed mouth of ciliated food-grooves (subvective system of 

 Hjeckel). The presence of anus and hydropore on the oral surface forbade absolute 

 -Symmetry of extension ; hence the primitive number of grooves was three — one 

 anterior away from the anus, one right, and one left. Bifurcation of the right and 

 left grooves produced the number five, as now we see it. It was after the pelma- 

 tozoan type of structure had been attained that some forms again relinquished the 

 attached mode of life, and assumed a position with the mouth anterior, as Holo- 

 thurians, or with it downwards, as Stelleroidea and Echinoidea. In the torsion of 

 their internal organs, in the pentamerous symmetry, and in other details, all these 

 forms bear the mark of their pelmatozoan ancestry. This need not mean that they 

 were descended from any pelmatozoan genera with which we are acquainted, 

 although the Edrioasteroidea certainly do present features in the structure of the 

 subvective grooves which enable them to be compared with the primitive Echinoidea 

 and Asteroidea. These features cause the separation of the Edrioasteroidea as a 

 distinct Class, a step already taken by Billings, Huxley, Chapman, Worchen, 

 Steinmann, Jsekel, and others. The Holothurians, Stelleroids, and Echinoids may 

 be grouped together as Eleutherozoa, without implying any genetic connection 

 between them further than that due to their independent descent, at different 

 periods, from pelmatozoan ancestors. Between the latter, however, the connection 

 is so evident that it should be recognised by the retention of them in a sub-phylum 

 Pelmatozoa. 



The mutual relations of the Classes are thus conceived, the older being placed 

 at the top : — 



/Cystidea Edrioasteroidea 



Pelmatozoa - 



Blastoidea 



I 

 ^ Crinoidea 



— Holothurioidea-y 



(Eleutherozoa) 



Echinoidea 



— Stelleroidea 



The starting-point of the Cystidea is a simple, many-plated, sac-like form, the 

 skeleton of which presents no trace of radial subvective or ambulacral systems ; 

 the porous structure of the stereom is indefinite, and no stem is differentiated. 

 Haeckel's Order Amphoridea is adapted to include this and such modifications ol it 

 as did not attain radiate symmetry. Further modifications of this type depend on 

 the mode of extension of the subvective system. In one group these extensions 

 pass over the thecal plates (epithecal), while still further extensions arise from the 

 grooves on 'exothecal' processes (brachioles). In the other group exothecal 

 brachioles spring at once close to the mouth. These two types are in the main 

 correlated with two gradual differentiations of stereom structure. In the former 

 the simple or irregular haplopores become connected in pairs (diplopores). In the 

 latter the pore-canals come to lie parallel to the surface of the test and at right 

 angles to the sutures between the plates (pore-rhombs). These canals really 

 represent stroma-strands, and sometimes, perhaps, haemal lacunae ; there were no 

 true pores. Enlargement of the thecal plates emphasised this rhomb-structure, 

 since it added strength to the plates. Diminution in size of the thecal cavity and 

 pressure of the coiled gut against its walls concentrated these structures in definite 

 areas, while the need for compensating this restriction of the assumed' respiratory 

 area led to the specialisation of the remaining pore-rhombs as ' pectini-rhombs.' 

 These facts justify the Orders Diploporita and Rhombifera, and leave only a few 

 Cystidea in the provisional Order Aporita. In none of these forms has the radial 



