VII. Morphology and Bionomics. 1915 214 



Cambrian rocks, was certainly not fixed, and can have had at most 

 a loose and variable attachment. Between these extremes lie all the 

 forms of attachment found in the Edrioasteroidea, differing from 

 genus to genus, and even from species to species, according to the 

 needs of the environment. Thus, evidence has been adduced to 

 show that, in spite of Jaekel's contrary opinion, JEdrioaster and 

 Dinocystis were not actually fixed (Study II, 1900, p. 200 ; Study IV, 

 1914, p. 169). Direct evidence is wanting in the case of Dinocystis, 

 but the indirect proof is the same as for Edrioaster, namely, that the 

 fossils are never found resting on any hard surface, but have the 

 apical face covered with shale or sand, which cannot have afforded 

 a firm basis of attachment. It is by parity of reasoning that 

 Pyrgocysth grayae and its "Wenlockian successors are supposed to have 

 been not fixed but inserted in a sea-floor of similar loose consistency, 

 although P. sardesoni, in its firmer surroundings, seems to have 

 been fixed. 



The general situation of such Agelacriuidae as c Agelacrtnus* 

 sensu lato, Hemicystis, and Streptaster, upon the smooth hard surfaces 

 of shells and similar objects, suggests a permanent fixation, and I can 

 recall no very clear evidence to the contrary. Dr. Eoerste, however, 

 says (1914, op. cit., p. 407): "In all of the Ordovician species 

 referred to Agelacrinus or Zepidodiscus, the animal evidently was 

 capable of attaching itself to various objects for support, although 

 this attachment was not permanent, and occasional specimens are 

 found unattached." The nature of this temporary attachment is 

 imagined by Dr. Eoerste to have been essentially the same as that 

 suggested by me for Edrioaster and Dinocystis. As explained in 

 Study II (1900, p. 202), it is supposed that this was effected by an 

 action comparable to that of a limpet or a sea-anemone or of any 

 mechanical sucker. In all the species of Edrioasteridae the necessary 

 elements of the theca were present, namely an apical concavity, 

 a rigid frame, and a central area of flexible integument. We know, 

 it is true, nothing of the muscles within the thecal cavity that may 

 have raised up this central integument, but the radial muscles of the 

 flexible-tested sea-urchin Asthenosoma indicate how readily the 

 necessary muscles may have been developed. Another mechanism, 

 however, may be conceived. If, as here maintained, the pores 

 between the floor-plates of the subvective grooves led from podia 

 fringing the grooves to ampullae within the thecal cavity, then the 

 distension of the ampullae by influx of water through the hydropore, 

 or by retraction of the podia, would exert hydraulic pressure on the 

 walls of the theca and their flexible portions would be inflated. If 

 now, the ampullae were contracted and all their contained fluid forced 

 into the podia or even out through the hydropore, then the flexible 

 parts of the thecal wall would necessarily be drawn inwards. Thus, 

 if the thecal margin were resting on a sandy bottom, a vacuum would 

 be created, with consequent sucker action. 



In so far as this latter hypothesis helps to account for the presence 

 of pores in the free Edrioasteridae, just so far does it fail to harmonize 

 with their presence in the fixed ISteganohlastus with its more rigid 

 theca. Moreover, if the Agelacriuidae were attached by sucker 



