MORPHOLOGICAL PART. 79 



their mode of growth, and their gradual introduction in geological time. 

 In the young dicyclic Crinoid, as we had occasion to observe especially well 

 in the genus Platycrimm (Plate LXXIIL Fig. 10, and Plate LXXV. Fig. 11), 

 the arms are uniserial throughout, their outlines waving, the plates decidedly 

 wedge-shaped, the pinnules proportionally large and given off alternately as 

 in true uniserial arms. In somewhat older specimens, the plates at the tips 

 gradually interlock, and the new ones still forming at the distal end are 

 strictly biserial. With advancing maturity the interlocking gradually ex- 

 tends to the proximal ends, until finally in the adult Platycrinus the whole arm 

 becomes biserial, except perhaps as to a few plates near the calyx, which 

 permanently retain their larval condition. Similar modifications occurred 

 in geological time. In the Lower Silurian the arms of monocyclic Camerata 

 are uniserial, almost without exception. In the Niagara group and Wen- 

 lock limestone, however, they rapidly change into biserial. It is very signi- 

 ficant that among the species of that epoch we find as persistent characters 

 all the phases through which the arms of the individual Crinoid pass in early 

 life. This is well shown in the case of the Batocrinidae.* The arms of 

 Hahrocriniis and Desmidocrimis are uniserial ; but while the plates of the 

 former are always rectangular, those of the latter in some species are de- 

 cidedly cuneate. The same modifications can be observed among the arms 

 of Fatelliocrinus and Stelidiocrinus, but in some of their species the cuneate 

 plates already begin to turn into biserial by interlocking. We thus find in 

 the same genus, and almost contemporaneously, all the variations from uni- 

 serial arms to biserial; and, what is most significant, the arms of all Devo- 

 nian and later Batocrinidse are strictly biserial. Turning to the Platycrinidse, 

 we find that the Upper Silurian Gordylocrinus has uniserial arms, formed of 

 rectangular or cuneate joints, while in Marsupiocriniis, CuUcocrinus, and Platy- 

 criniis they are biserial. We may note also the case of the Hexacrinidge, in 

 which the development of the arms took place at a later period. The arms 

 of the Devonian genus Hexacrinus are uniserial, and also those of the earlier 

 species of Dichocrinus. The plates of the latter are rectangular in all Kinder- 

 hook species, and also in about half of those from the Burlington and Keokuk 

 groups ; in most of the others they are wedge-shaped, and in a few of them 

 the arms fairly enter the interlocking stage. All species, however, of the 

 Kaskaskia group have biserial arms, and likewise the contemporaneous 

 Talarocrinus and Pterotocrinus. Among the MelocrinidaB, Glyptocriniis and 



* We have separated the Batocrinidse from the Actinocrinidse, referring to them only those genera in 

 which the anal plate is followed by three interbrachial pieces, instead of two as in the latter family. 



