Dr. F. A. Bather — Studies in Edrioasteroidea. 321 



(1914, pi. i, fig. 9), and Jaekel's diagram of Eophiura, Palaeura, and 

 Bohemura (1903), and it will be seen that the older the form, the 

 more obvious is the composition of the mouth-frame from a series of 

 ambulacrals diverging and becoming overlapped by the corresponding 

 adarnbulacrals. 



Perhaps the most instructive figure in this respect is that of the 

 mouth-angle of Siluraster perfectus from the Upper Ordovician of 

 Bohemia (Jaekel, 1903, fig. 3). Here is seen a series of adarnbulacrals 

 each articulated with its corresponding ambulacral. The proximal 

 adambulacral is enlarged to form a " Mundeckstiick ", but beneath it 

 is still the ambulacral, though correspondingly reduced in size. 



It seems a legitimate inference that all these plans of mouth-frame 

 have been derived from one just a little simpler than the simplest of 

 them, namely, a plan in which the ambulacrals were continuous right 

 round the angle from ray to ray, each accompanied by its adambulacral. 

 At first, no doubt, they were undifferentiated, so that the particular 

 numerals attached to them, either here or in later forms, have no 

 profound significance. Such a simple plan is essentially that of 

 JEdrioaster (Study IV, 1914, p. 164, PI. xiv, fig. 2), but here already 

 there is a condensation of the interradially placed floor-plates into 

 a mouth-angle plate, a divergence of the proximal floor-plates of 

 the groove, and an overlap producing in oral aspect the deceptive 

 appearance of distinct perradial elements. The chief point of 

 difference from any primitive Asterozoon lies in the fact that the 

 adarnbulacrals of Edrioaster are still serving their primitive function 

 as cover-plates. 



The structure of Edrioaster suggests that even the interradially 

 placed odontopbore may be not of true interradial or interambulacral 

 origin, but derived from the outer fused portion of the interradially 

 placed floor-plates. On the other hand it is not quite certain that 

 in Edrioaster itself this portion may not be in part an interambulacral 

 with which the floor-plates have fused. 



The curiously close resemblance between the ambulacral and 

 peristomial structures of an Edrioasterid and those of a primitive 

 Asterozoon would be strange indeed if their modes of feeding had 

 been as different as those usually connoted by the names Pelmatozoon 

 and Asteroid. We have, however, seen reason to believe that 

 Edrioaster and Dinocystis were not permanently fixed ; whence it 

 may be inferred that they possessed some slight power of independent 

 locomotion, towards which movements of the peripheral podia may 

 have contributed. The existence of well-developed podia (a fact that 

 cannot reasonably be doubted) further renders it probable that those 

 organs helped in the transport of the larger food-particles to the 

 mouth. Some Asteroids, on the other hand, and even some Ophiuroids, 

 can use their podia for the same purpose ; and some of the less active 

 and less rapacious Asteroids can, as Dr. Gemmill has lately proved, 

 subsist in part, if not entirely, by ciliary nutrition (March, 1915, 

 Proc. Zool. Soc, pp. 1-19). Though a predatory mode of life was 

 assumed at least as early as Devonian times by some starfishes 

 with well-developed mouth-frame, e.g. Xenaster eucharis (see J. M. 

 Clarke, 1912, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 15, 



DECADE VI. — VOL. II.— NO. VII. 21 



