1888.] NATURAL SCIENCES OF PHILADELPHIA. 373 



criuidae. The mode of insertion of the higher radials upon the first 

 primary is similar to that found in Pterotocrinus (PI. XIX, fig. 6), 

 and Marsupiocrinus (PI, XIX, fig. 7, and also Angelin's PI. XXII, 

 figs. 1, and 28, PI. XXVII, fig. 4), and is upon the very same 

 principle that prevails in the Platycrinidae generally. 



A further striking resemblance to the Platycrinidae is to be 

 observed in the structure of the vault. We give for comparison fig- 

 ures of three of the most perfect vaults of Crotalocrinus ever found. 

 Fig. 4, on PI. XIX is from the Swedish specimen already described 

 as the original of Angelin's PI. XVII, fig. 3a. Fig. 3 is from a 

 specimen formerly in the Fletcher collection at Dudley, but now be- 

 longing to Cambridge University. It differs somewhat from the 

 others in the form of the four smaller orals, which are nearly equi- 

 lateral instead of clavate, and in having a few more interradials. 

 Fig. 2!!. is from a Dudley specimen in our own collection. We can 

 see enough of the vault in our specimen of G. pulcher from Gothland 

 to show that it is built upon the same plan as in the three specimens 

 of C. rugosus illustrated, but it cannot be exposed sufficiently to afford 

 a good figure without mutilating the specimen more than is justi- 

 fiable. 



Taking all these facts together, the vault of Crotalocrimis seems 

 to have been composed of well developed oral plates (four proximals 

 and a central), large interradials, several anal plates, with anus in 

 form of a subcenti-al opening or a tube, and covering plates. The 

 latter are solidly inserted in the vault between the other plates, so as 

 to form a part of the wall, contrary to the Inadunata, in Avhich the 

 covering plates, and the ambulacra generally, rest ^lpon the edges of 

 the other vault plates. 



Taking now for comparison the vault of Marsnpiocrinus tennes- 

 seejisis (PI. XIX, fig. 7), we find the same arrangement of oi'als; the 

 same solid covering pieces incorporated into, and forming part of the 

 vault, originating at the re-entering angles of the five orals, and pass- 

 ing outward to the arm bases ; we also find a system of interradial 

 and anal plates substantially like that of the Cambridge specimen of 

 Crotalocrinus (PI. XX, fig. .3). Indeed, if we had the vaults alone 

 of these two specimens under examination, it would not be a very 

 easy matter to point out why they might not belong to the same ge- 

 neric type. Certainly no one can look at the two figures, and not be 

 entirely convinced that they represent the same plan of summit 

 structure. And if we then compare the parts above the first radials 



