SYSTEMATIC PART. 157 



Fistulata. The former embrace the families Haplocrinidae, Pisocrinidse, 

 Symbathocrinidse, and Cupressocrinidae ; the latter the Hybocrinidae, Hetero- 

 crinida3, Anomalocrinidge, Belemnocrinida3, Gasterocomidas, Catillocrinida3, 

 and Calceocrinidae of monocyclic forms, and the Dendrocrinidse, Cyathocri- 

 nidoB; Poteriocrinidse, Astylocrinidas, Encrinidas, and Pentacrinida3 among 

 dicychc forms. The arrangement is substantially the same as that proposed 

 by us in 1885, in Part III. of the Eevision, except that we withdraw the 

 Gasterocomidae from the Larviformia, and place them among the Fistulata. 

 This change was announced by us in 1890,* when we restricted the Larvi- 

 formia to those Inadunata in which the orals rest against the radials, and the 

 ventral surface is covered exclusively by the orals, i. e., Crinoids which 

 remain persistently in the larval state. 



A different division of the Inadunata has been lately proposed by Mr. 

 Bather,t who subdivided the Inadunata into " MonocycHca " and " Dicyc- 

 lica ; " but whether they should be ranked as suborders, he leaves as yet in 

 doubt. In alluding to the Larviformia and Fistulata, Bather says these divi- 

 sions " cannot well be maintained. Many genera hitherto included in the 

 Larviformia have quite as good a ventral sac as some acknowledged Fistu- 

 lata.'' We do not know of any group to which this remark can be applied, 

 unless Mr. Bather undertakes to homologize the narrow anal tube of Syvibatho- 

 crinus and Pisocriniis with the ventral sac of the Fistulata. Symbathocrinus 

 has no ventral sac, but simply an anal tube, nor has it an anal plate, or 

 perisomic pieces as we once supposed ; its asymmetrical oral pyramid rests 

 directly upon the radials, and its anal tube is supported by the radials and 

 orals together. Bather further says: "they (W. and Sp.) excluded Hetero- 

 crinus and Calceocrinus, in which it has at all events never been proved 

 that other plates beside orals occur in the tegmen." We supposed it was 

 now admitted that the ventral sac represents morphologically the highly 

 developed posterior inter-palmar area of the disk, as was proved by the 

 position of the anus, which is situated either at the anterior side of the sac, 

 or not within the sac at all, but in front of it (toward the oral centre) in the 

 main part of the tegmen. Admitting this, the presence of the sac proves 

 that the Heterocrinidae had a complicated disk. 



The case is very similar in the Calceocrinidae and Catillocrinidse. Both 

 agree with the Larviformia in having no anal plate, at least no anal :v; but 



* Proceed. Acad. Nat. Sci. Phila., 1891, p. 355. 



t The Crinoidea of Gotland, Part L, with ten plates (Stockholm, 1893), p. 20. 



