MONOGRAPH OF THE EXISTING CRINOIDS. 323 



it covers in the radial prolongation of the axis, so that the central space left by the 

 incomplete meeting of the valves of the basal pentagon is extended on its external 

 aspect into five broad rays, though on its internal or ventral aspect, where it is 

 bounded by the last-formed portion of the endogenous reticulation, it shows no 

 corresponding increase. This removal of the older and outer part of each basal 

 plate by resorption, and the consolidation of the newer and inner by additional 

 calcareous deposit, go on at a rapid rate, so that in specimens whose size and general 

 development show but little advance upon the earliest Antedon type we find the 

 basals already modeled into such a form that their coalescence will produce a 

 somewhat unshapely rosette. In figure 584, plate 12, is shown the dorsal aspect of 

 one of the basal plates in which the removal of the external layer has been carried 

 so much further that what is now left of it constitutes only a kind of thickened 

 margin along those sides of the plate which are received between the radials, and 

 by an extension of the same process along the median line of each plate until the 

 external layer has been completely removed from its salient angle the two lateral 

 portions of that layer separated from each other (at their distal ends) and remain 

 only as a pair of curved processes extending themselves from the inner layer in 

 such a manner as to give to the plate when viewed from its ventral side somewhat 

 of the aspect of a saddle. When the five basals thus altered are in their normal 

 apposition the curved processes on either side of each plate come into contact with 

 the corresponding processes of its next neighbor, and the junction of the two forms 

 a sort of ray curving toward the dorsal aspect (this being the rudiment of one of 

 the five radial or spoutlike processes). As each plate thus contributes the half 

 of two of these curved rays, five such rays are formed between the five salient 

 processes which are put forth by the internal or ventral layer on the median lines 

 of the five plates and are received into the retreating angles formed by the junction 

 of the radials. Very soon an actual continuity is established in the calcareous 

 reticulation along the lines of junction and the rosette is completed, although the 

 peculiarity of its shape becomes much more strongly pronounced with the subse- 

 quent increase of its size. Thus the rosette is essentially formed at the expense of 

 the secondary or ventral layer of the original basals, the ends of the curved rays (or 

 spoutlike processes) being the sole residue of their primary or dorsal layer, and 

 since, by the removal of the median portion of that layer in each plate the primary 

 basal cords are left bare on their dorsal aspect, they now pass from the central axis 

 (the chambered organ) into the canals of the radials on the outside (dorsal side) of 

 the calcareous skeleton which occupies the central part of the base of the calyx 

 instead of reaching these by passing (as they did in the first instance) along its 

 internal (ventral) face or (as at a later period) through the middle of its substance." 

 In regard to the relationship between the rosette and the axial nerve cords, 

 P. H. Carpenter says: "Each of the primary basal cords, which are interradial in 

 position, divides into two branches toward the periphery of the rosette, on the 

 dorsal (outer) surface of which it rests. These branches lie in the shallow channels 

 which mark the union of the base of each interradial triangular process with the 

 two curved lateral processes, each of which unites with a corresponding process 

 from the adjacent basal to form one of the five spoutlike processes of the rosette. 



