208 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



There are four possible ways of escape from such a calamity: (1) The column 

 may be discarded; (2) the indivadual columnals may become greatly shortened, 

 the motion lost through the great diminution in tlie original bevchng at the articu- 

 lations being conipousated by the great!}' increased nund)cr of articulations in a 

 given section of stem; (3) the articulations may alter the direction of their fulcral 

 ridges so that, instead of each being at right angles to the preceding and succeeding, 

 they may each he at only a slight angle to the preceding (all diverging toward the 

 same side), thus mutually bracing each other and attaining a collective rigidity, 

 hke a pile of narrow boards built up spirally; or (4) the original fulcral ridge may 

 (hsintegrate, each half breaking up longitudinally and spreading out fan hke, the 

 two fan-like figures eventually uniting to form an articular surface composed of 

 numerous luiiform radiating hnes, each Une representing a narrow ridge, and the 

 joint face becoming circidar in outhne instead of narrowly elhptical. 



The comatuUds fulfill the conditions of the first possibiUty; before the animal 

 is large enough to cause any danger of "buckling" the column is discarded at the 

 articulation between the topmost colunmal which remains unmodified, and the 

 centrodorsal. Phrynocrinus (fig. 2, p. 61) is the best recent example of the second 

 case, though all the larger species of the various genera of the Bourgueticrinidae, 

 as for instance of Democrinus (fig. 138, p. 205) exhibit the same feature in varying 

 degrees of perfection. The curious fossil Platycrinvs (fig. 516, pi. 1) typifies the 

 third. Among the recent forms Proisocrinus (fig. 128, p. 199) (probably also Car- 

 pentcrocrimis) , and possibly Ilyocrinvs, Tlialassocrinvs (fig. 145, p. 209) , Gephyro- 

 crinus,Ptilocrinus (fig. 144, p. 207), Calamocrinus, and the pentacrinites (see beyond), 

 (figs. 126, p. 195, and 127, p. 197) are instances of the fourth. 



In the genera Ilyocrinus, Ptilocrinus, Calamocrinus, Gephyrocrinus and TTialas- 

 socrinus the column is attached by a sohd terminal stem plate, and the indi\-idual 

 columnals are cyhndrical with their circular articular faces marked with radiating 

 lines; the proportionate length of the columnals varies with the size of the animal, 

 the columnals being longest in the smallest species. 



There is no evidence whatever that these columnals were derived through 

 colunuials of the bourgueticrinoid type, or that young intUviduals possess co- 

 lumnals in any way different from those of the adults. 



There is no trace whatever of a proximale; in Calamocrinus, where the topmost 

 colunmal has been investigated with great care, it has been found to be a very tliin 

 quinquelobate structure, the quinquelobate form undoubtedly resulting from the 

 mechanical hmitations imposed upon it by its place of origin, just below the five 

 basals. 



Wlule we know that this type of column may be derived through the bour- 

 gueticrinoid typo, as it is in the pentacrinites for instance, we are not justified in 

 assuming that in these genera it has undergone an}' such development. It is quite 

 possible, even almost probable, that we have here a case of the survival of the 

 typical palaeozoic column in a recent group. 



The change from the type of cohnnn characteristic of the young of Antedon 

 to that characteristic of Phrynocrinus may be traced step by step in the family 

 Bourgueticrinidse, beginning with the httle 7?. lofotensis and ending \vith the 



