28 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



to the greatest depths, where only very small species are found. Below 100 fathoms 

 plant life disappears, but this disappearance is offset by the gradual increase in the 

 rain of carcasses from above, so that an equilibrium is obtained down to about 600 

 fathoms; below 600 fathoms gradual decomposition of the carcasses lessens the food 

 value and we find the crinoids decreasing in size. It is pointed out that the pen- 

 tacrinites, which are remarkably uniform in size when compared with the comatulids, 

 are almost exclusively confined to the 100-600 fathom belt; the two species which 

 descend to below 1,000 fathoms are both small. 



The unusually large size of Heliometra glacialis in certain localities is explained as 

 follows: On the west coast of Greenland, by the fresh-water ice which by melting 

 kills millions of small organisms, which fall to the bottom and serve as crinoid food : 

 about Spitzbergen by the fresh water from the snow and ice on the land which acts 

 in the same way; on the Grand Banks by the mingling of the warm Gulf Stream and 

 the cold northern water, by which organisms in both are killed; in the Sea of Japan 

 by the mingling of the warm tropical and cold northern water; here areas of warm 

 bottom water are dovetailed in with areas of cold and the size of the crinoids in both 

 is similarly increased. 



The occurrence of pentacrinites in shallow water off Cuba, Guadeloupe, and south- 

 ern Japan is explained by the existence of fresh water streams the water from which 

 kills the plankton but does not penetrate deeply enough to injure the crinoids. 



In the East Indies very large species of comatulids occur abundantly in very 

 shallow water, often just below low tide mark, and these decrease in size with depth. 

 This is explained by the killing of the plankton in the littoral water by the action 

 of the sun's heat and by rain, the dead organisms being deposited just beyond the low 

 tide mark. It is pointed out that the large littoral species occur only where the rain- 

 fall is abundant. 



The comatulids with large eggs, termed "Antedonoida," have a relatively 

 restricted distribution ; the large eggs are assumed to develop rapidly, implying a 

 relatively short duration of the active free-swimming stage and consequent inability 

 to travel widely. The species with small eggs, included in the Thalassometroida (in 

 which group Thalassometra, Charitometra, and Tropiometra are mentioned), have a 

 relatively large range ; the assumed longer duration of the free-swimming period results 

 in greater facilities for dispersal and the assumed slower development of the larva 

 possibly a greater power of adaptation. 



The coloration of the recent crinoids in life is considered in detail. All colors 

 are found in the crinoids except blue, though true black is confined to the disks of the 

 species of Pentametrocrinidae and to lines and spots on two species of Coccometra. 

 Yellow is the commonest color in the group, and is the color of all the more primitive 

 forms and of the young of almost all the others; it may, therefore, be taken as the basic 

 crinoid color. The pentacrinoids of Antedon bifida are sometimes pink, though 

 usually, like the pentacrinoids of the other forms in which they are known, yellow, 

 and certain other types are dull pinkish in all stages. Two basic colors are therefore 

 assumed, yellow and red, the latter an intensification of the former and found gener- 

 ally in the more specialized forms. 



The derivatives from these two basic colors as they occur in the crinoids are 

 grouped as follows: 



