BARBADOS-ANTIGUA REPORTS 21 



The same thing is true in the fixed and sedentary marine ani- 

 mals, and is well illustrated by the Caribbean erinoids. 



The thermal and actinic range of the stalked species is small, 

 as they are for the most part inhabitants of deep water where 

 conditions are practically uniform; their range of variation is 

 therefore slight and there is little difficulty in delimiting their 

 species. 



But many, if not most, of the unstalked species in the Carib- 

 bean Sea have an unusually great bathymetrical range, their 

 representatives in the littoral and sublittoral zones living under 

 a great variety of conditions varying in temperature and illumi- 

 nation, and therefore also in the quality and quantity of the 

 food. From this circumstance it comes about that each of the 

 more abundant types occurs in a number of more or less distinct 

 varieties which, intergrading in every conceivable way, are quite 

 plant-like in their interrelationships, and which appear to be 

 proportionate in number to the thermal and actinic range of the 

 species. 



Thus the interrelationships of the forms in the genus Com ac- 

 tinia — Hartlaub recognizes fifteen varieties of Comactinia ecMn- 

 optera — and in the genus Crinometra, with fifteen nominal spe- 

 cies, recall the interrelationships of the more difficult sections of 

 such plant genera as Crat^gus, Rubus or Rosa, and those of the 

 forms in the genus Nemaster the interrelationships of our local 

 species of Circ^a. 



The comatulids differ from all the other fixed and sedentary 

 animals in being reduced to practically nothing but a food col- 

 lecting apparatus, the organs not concerned, as are the arms and 

 pinnules, in the collection of food being reduced to an absolute 

 minimum. In the plants the species of Rafflesiaceae are for the 

 most part reduced to a flower only, without leaves, stem or true 

 root, and it is interesting to note that the range in size of the 

 species of Rafflesiaceae and of the comatulids is the same, from 

 less than an inch in diameter to about three feet {E-afflesia 

 arnoldi and Heliometra maxima). 



In the erinoids the excretion of waste products is for the most 

 part effected by the formation of small globules (sacculi) chiefly 

 along the ambulacral grooves which are superficially quite similar 

 to the "glands" dotting the leaves in the Hypericaceae and 



