Mr. F. A. Bather on British Fossil Crinoids. 381 



actually represent the intermediate stages in the atrophy of 

 the radianal ; this, however, is disproved by the arms, for while 

 those of Cyathocrinus &c. are simply dichotomous, the more 

 specialized armlets are already developed in Sicyocrinus and 

 Botryocrinus. 



Clearly then the Dicyclica of Niagara and Wenlock age 

 must be divided into three groups : one with brachianal, large 

 radianal, and tendency of tube-plates to sink into dorsal cup ; 

 a second with large brachianal only ; and a third with large 

 brachianal and with small radianal. The first group has 

 rather slender, long, dichotomous arms ; the second has short, 

 rather stout, dichotomous arms ; the third has stouter arms, 

 as a rule with armlets. 



Following the fortunes of the first group, Homocrinus leads 

 us through the Lower Devonian to the Eifel-Kalk of Europe 

 and the Hamilton group of America. Here occur Pariso- 

 crinus, Poteriocrinus, and Scaphiocrinus, all closely resembling 

 one another in the anal area, and in this point also not far 

 removed from Homocrinus. The main difference is observ- 

 able in the arms ; those of Homocrinus and Parisocrinus are 

 simple and dichotomous, while those of Poteriocrinus and 

 Scaphiocrinus are already pinnulate. There is here one of 

 those gaps which a better knowledge of American Devonian 

 Crinoids would probably fill. From Scaphiocrinus through 

 Woodocrinus to Cadiocrinus and Hydreionocrinus the stages 

 are gradual and easily traced. 



It is convenient next to take the third group mentioned 

 above, viz. the Botryocrinus group. Here again is a great 

 gap : at the same time there can be little doubt that both 

 Vasocrinus and Barycrinus are direct descendants of Botryo- 

 crinus ; with them Atelestocrinus is closely connected. Deca- 

 docrinus, which comes in with Vasocrinus^ appears to me to 

 be also descended from Botryocrinus^ although it is rather 

 further removed from it than is Vasocrinus. With Decado- 

 crinus are closely connected the rather later forms Graph io- 

 crinus and Scytalecrinus. Grajyhiocrinus, however, presents 

 that modification in the anal area (PI. XIV. fig. 36) which 

 through Bursacrinus, Phialocrinics, and Synyphocrinus leads 

 on to Erisocriuus } Ceriocrinus, and Stemmatocrinus, and 

 culminates in the Triassic Dadocrinus and Encrinus. A 

 different modification of the Decadocrinus type is seen in 

 luipachycrinuSj with which Tribrachiocrinus is probably to 

 be connected. Cromyocrinus is a direct offshoot from 

 Eupachycrinus and to it in turn Agassizocrinus is closely 

 allied. 



These relationships are so obvious, and in fact so generally 



