286 Messrs. P. H. Carpenter and K. Etlieridge, Jun., 



more than a thickened semicircular rim, which is thus con- 

 verted into the opening of a tunnel that leads inwards beneath 

 the dome. 



These notches on the upper surface of the radial plates are 

 the central ends of the grooves which are excavated in the 

 upper surface of the arm-joints, and were called the arm- 

 grooves by Miiller. They are often, although incorrectly, 

 spoken of as the ambulacral grooves. Above and partly con- 

 tained in these grooves were the vascular and generative 

 tubes of the arms, and above all these the true ambulacral or 

 food-grooves, which may or may not have been protected by 

 special plates, as in Cyatliocrinus and many other fossil and 

 recent Crinoids. All these structures entered the calyx on 

 their way to join their respective circumoral centres through 

 the openings at the edge of the vault between the radials and 

 the orals. These openings are of course the representatives 

 of the ambulacral openings round the edge of the vault of 

 Actinocrinus ; but there is no trace in AlJagecrinus of any 

 such separation of the soft parts by plates as we find in the 

 former genus. In both cases the coeliac canal, which was 

 lodged in the lowest portion of the arm-groove, is continued 

 directly downwards into the visceral cavity. It was sepa- 

 rated in Actinocrinus from the water-vessel and food-groove 

 by the subambulacral plates, which form the floor of the 

 ambulacral tunnels beneath the vault; but there is no trace 

 of these in Allagecrinus. 



In none of these small specimens is there any trace of an 

 anal opening, either directly piercing an oral plate, or at the 

 margin of the dome, between the orals and the radials. The 

 central end of one or more of the former may be marked by 

 faint tubercles (figs. 5 and 7, PI. XVI.) ; but we cannot sug- 

 gest any explanation of these. In the specimen shown in 

 fig. 6 the central portion of the dome has been removed, and 

 only the bases of the triangular oral plates are visible, 

 k Except in the characters of the stem, and in the general 

 aspect of the basals and radials, these small specimens differ 

 so much from the larger ones previously described that it 

 would seem only natural to place them in entirely different 

 families, characterized respectively by the presence or absence 

 of an oral pyramid. Fortunately, however, we have met 

 Avith a few specimens that show us such a gradual transition 

 between the two groups that Ave have been compelled to place 

 them together under one specific name. 



These intermediate forms, which are represented in PI. XV. 

 fig. 7, and PI. XVI. figs. 8-10, while agreeing in certain 

 general characters, seem to have developed along different 



