ORGANIC DEPENDENCE AND DISEASE 



79 



ties of their observation occurs on no other crinoicl species 

 except Scytalocrinus rohustus, which also possesses a long 

 anal tube but with the aperture far down toward the base 

 and within easy reach of the starfish. We here illus- 

 trate a Barycrinus {B. Hoveyi Hall), a species without 

 anal proboscis, engaged with this starfish. No evidence 

 exists that this habitude continued beyond the Lower Car- 

 boniferous. One familiar with the feeding habits of the 

 starfish from ancient Devonian times to the present would 

 be disposed to suggest that the association under present 

 consideration can be only for feeding purposes. That this 

 may have been ''a favorite resting place" (Wachsmuth and 



Fig. 66. The crinoid Banjerinus Hoveyi with the starfish OnycJiaster flexUis inter- 

 twined with the arms. Mississippian. 



Springer) for the starfish will hardly explain it. The small 

 boy in an apple tree is not there to rest. This affiliation, 

 its constant recurrence always between the same species of 

 starfish and a very few species of contemporary crinoids; 

 limited apparently to the very brief period which repre- 



