142 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



close spiny coverin<j above it in addition to the sensitive network of flexible 

 tactile pinnules just beneath. 



There are no oral pinnules in the genera Aielecrinus and Hypalometra (fig. 

 297, p. 221). In most of the species of Pentametrocrinus P, is absent. 



In the species of Bathymetra the oral pinnules are greath' reduced. 



Whereas in the Macrophreata the tactile function of the oral pinnules is pre- 

 dominant and there is usually only a single pair present, in the Oligophreata the 

 protective function is predominant, commonly combined with the tactile in various 

 ways, and there are usually two or three pairs present, and may be as many as eight. 



The oral pinnules of the Comasteridse are very curiously specialized. They 

 are long and slender, very flexible, at least in the outer portion, and are composed 

 of from 20 to 60 (usually of between 30 and 40) segments, most or all of which are 

 about as long as broad. 



The terminal 6 to 30 (most commonlj^ from 15 to 20) segments have their distal 

 outer margin produced into a thin calcareous plate, cur\ed slightly inward and 

 basally constricted, being shaped like an arrowhead or spade, which is usually about 

 as high as the diameter of the segment which bears it, but sometimes considerably 

 higher. Occasionalh' the inner margin of the segments is produced in the same 

 way, as in the genus Nemasfer. These plates collectively form what is known as 

 the terminal comb (figs. 591-657, pp. 309-327, 1072-1082, pi. 17. and part 1, figs. 

 56-58. p. 83. and 59, 60, p. 85). 



The terminal comb in its perfected form is confined to the species of the 

 family Comasteridse, but a somewhat similar though much less developed structure 

 is found in Flarometra (figs. 1053, 1054, pi. 13), Solanometm, Anthovi/^tra and 

 PromcwhGcr'mufi, and is occasionally indicated in HeUometra (figs. 545, 546, p. 289). 



The terminal comb is primarily restricted to the oral j)innules, but sometimes 

 extends for a greater or lesser distance onto the genital pinnules, and in the species 

 of Comaster (fig. 263, p. 207) is even found at intervals on the distal pinnules to 

 the arm tip. In these cases, however, it is only upon the proximal pinnules that it 

 is found in the highest state of development; farther out on the arm fewer and 

 fewer segments are involved in its composition, and the teeth become a.s a rule 

 gradually less and less perfect. . 



The function of the terminal combs is not known. Long ago Peron observed 

 that the arms are frequently bent inward toward the disk, and this observation 

 led Lamarck to suppose that the food of the comatulids was captured by the distal 

 pinnules and conveyed by them to the opening in the middle of the large conical 

 elevation on the disk, which he supposed was the mouth. Some years ago I sug- 

 gested that as myzostomes and other parasites are extremely rare upon the arms 

 of the comasterids, this action may be for the purpose of bringing the outer portions 

 of the arms and the outer pinnules within reacli of the terminal combs on the 

 proximal pinnules and by means of the latter removing foreign objects from 

 them. Dr. H. L. Clark has also noticed that the arms of the comasterids more or 

 less constantly, but irregularly, bend inward toward the mouth, the tips just 

 brushing the disk and arm bases, but he supposes this is done "to prevent the 

 ciliated grooves from becoming clogged up." 



