1887.] NATURAL SCIENCES OF PHILADELPHIA. 107 



the dorso central be represented at the actinal side when there is no 

 actinal stem, in this or any other group of the Echinoderras ? The 

 dorsoeentral in the Echinozoa represents in a wider sense the whole 

 column in its simplest form, although in a narrower sense it is the 

 homologue of the first part of the stem that makes its appearance in 

 the embryo. If there was such a thing as an orocentral in fossil Cri- 

 noids, Blastoids and Cystids, it seems to us, it certainly w^ould be rep- 

 resented in the early larva of the living types Ijefore the parting of 

 the orals, and in the closed oral pyramid of the Cystids and Steph. 

 anocrinus; but unfortunately for Carpenter's theory w^e meet with 

 no trace of it in either one of those forms. The plate which he re- 

 gards as orocentral, occupies the place of the five orals in other groups, 

 and in a similar manner as these, covers the peristome and the 

 origin of the ambulacra. This is conclusively shown by comparing 

 the ca.se of Caryocrinus in which the ambulacra start from beneath 

 the central plate and branch twice underneath the surrounding 

 plates,withthecaseof /SpAaero/^ites (fig. 1) and Stephanoeriims (fig. 3), 

 in which the ambulacra start from beneath a penta-partite oral pyramid. 

 Does this indicate that the five plates constituting the latter, are the rep- 

 resentatives of the proximals ? We doubt it, for the structural resem- 

 blance is with the central piece. We think the distribution and arrange- 

 ment of the surrounding plates in Carijocrinus proves conclusively 

 that these cannot be orals, for the most ingenious speculator would be 

 unable to reconstruct five primitive plates from such an assemblage 

 of pieces as we find in Caryocrinus and in Von Koenen's new genus 

 Juglandocrinus^. What those plates may be, whether actinal 

 or abactinal structures, we will not pretend to decide, but we do un- 

 dertake to say that they are not orals, otherwise the rule that there 

 are always five primitive orals meets with a very serious exception. 

 Somewhat more favorable perhaps to Car^^enter's views is the ar- 

 rangement of the proximals in the Palaeocrinoidea and Blastoidea, 

 in which the plates surrounding the central piece are unquestionably 

 actinal structures, and there is a possibility of reconstructing from 

 the six, seven, or more pieces, five primitive plates. We also admit 

 that in all cases where those plates come in direct contact with the 

 anal structures, their arrangement might possibly have been disturbed 

 thereby, but this explanation is not applicable to forms like Megisto- 

 erinus, Dorycriims and many others, in which the anus is lateral or 

 moved away from the center to the arm regions or even beneath them. 

 But there are several other equally serious objections. 



*Neues Tahrbuch fur Mineralogie 1886, Bd. II, Taf. IX, Fig. 3. 



