6 Dr. P. IT. Carpenter on the 



to one and the same set of plates, a method whicli, as it seems 

 to nie, is still more likely to confuse tlie student. The 

 German palasontologists have naturally followed Zittel, and 

 continue to speak of the dicyclic base as composed of para- 

 basals and infrabasals, a course which will not be made easier 

 by some recent discoveries. Thus, for example, de Loriol 

 has found infrabasals in two species of Miller icn'nus *, and 

 the plates above them, hitherto called basals, must now be 

 known as parabasals in these two species, though retaining 

 the simpler name in all the remaining species of the genus. 

 This v\ill be an endless source of confusion, and anotlier is 

 afforded by Zittel 's own description of the calyx of Penta- 

 cri'mis. He states that it contains five basals, but adds that 

 five infrabasals are sometimes present. According to his 

 terminology, however, the species possessing them f should 

 have no basals, but parabasals ; but he gives no hint of this. 

 Then, again. Bury has recently demonstrated the presence of 

 infrabasals in Antedon 7'osacea ; so that in Zittel's termin- 

 ology the plates hitherto called basals in this type must now 

 be known as parabasals, though their honiologues in the 

 apparently monocyclic fossil Comatulce will retain their old 

 name. In these three genera therefore — Millericrinus^ Penta- 

 crinus (in the widest sense), and Antedon — some species are 

 known to be dicyclic, while others are not, though the latter 

 are in all probability only pseudomonocyclic, to use the con- 

 venient term proposed by Bather \. But in Zittel's termin- 

 ology the generic diagnosis will have to run somewhat as 

 follows : — " Calyx composed of radials and basals, or of radials, 

 parabasals, and infrabasals." Would it not be infinitely 

 simpler and less confusing to say " Calyx composed of radials 

 and basals, sometimes with the addition of infrabasals " ? 

 If this be admitted, it is clear that the same principle may be 

 extended to definitions of families and larger groups, and the 

 misleading term parabasals will then have to be finally 

 abandoned. 



The term " subradials " was proposed in 1854 by de 

 Koninck and Le Hon instead of parabasals, and was generally 

 adopted by the leading American palaeontologists, e. g. Hall, 

 Billings, Meek and Worthen, and Whitfield. As long as 

 the homology of the plates so named with the basals of 

 monocyclic Criuoids remained unrecognized, this name w^as in 



• ' Paf^ontologie Fraugaise,' Terrain Juriissique, tome xi. pt. i. pp. 553, 

 566. 



t These species are now refen-ed to Extracrinus. 



t "British Fossil Crinoids," Ann. & Mag-. Nat. Hist. 1S90, ser. 6, 

 vol. V. p. 316. 



