ORGANIC DEPENDENCE AND DISEASE 55 



tions, the interpretation may seem invasive. Nothing, I am 

 disposed to believe, can be more illuminative of the prog- 

 ress toward and in intelligence than the early case before 

 ns, in which a directed habit has already become fixed by 

 heredity. 



Taken as a whole this combination is very complicated 

 commensalism from a date so ancient as the Devonian, more 

 extreme than any other yet known from the Paleozoic rocks. 

 We find a somewhat parallel case in the present fauna de- 

 scribed by Bouvier as occurring in the Gulf of Aden — a 

 coral and a worm growing together, and hidden in the coral 

 substance a gastropod on which both settled down when the 

 partnership began ; furthermore there appears to be a small 

 bivalve in association with the worm. Other similar cases 

 might be cited from the existing fauna. 



One stands with wonder before such evidence as this from 

 the ancient faunas, questioning how such a habitude came 

 about, what conditions impelled, stabilized and restricted 

 it, and our wonder is none the less because here we stand 

 at the inception of such associations and contemplate it 

 from a world that is full of them today. And the inquiry 

 naturally arises : What became of this organic combina- 

 tion? It reached its acme only as the coral genus became 

 old, indeed in the last of its representatives. Riddled with 

 commensals, overloaded with boarders who fed at the same 

 table and flourished, it may be that the worm became an 

 effective parasite which helped to bring about the extinc- 

 tion of its host. 



Commensalism between the worms and sponges. This 

 combination appears in the late Devonian, but there is evi- 

 dence that it is of earlier date. We have just cited the 

 presence of a minute sponge in the Hicetes innexus, the in- 

 colant worm of Pleurodictyum, and there is an undescribed 

 spreading sponge of the Middle Devonian (Hamilton 

 group) which indicates the presence of coexistent annelids. 

 The simple ancient instance I can here illustrate is that 



