292 BULLETIN 82, UNITED STATES NATIONAL MUSEUM. 



metridse, i[ariametridffl (fig. 344, p. 287), and Stophanometridtc (fig. 340, p. 287), the 

 structure of the cirri appears to be quite comparable to that of the Charitomotridje, 

 and transition segments appear never to have occurred. On the other hand, in 

 a single genus of the Macrophreata, Zenometra, a more or less marked transition 

 segment is found, comparable in every way to that of certain of the Oligophreata. 



There is a curious correlation between the cirri and the proximal pinnules; 

 species in which the latter are large, as Craspedometra acutidrra, as a rule have long 

 cirri with numerous segments, while species in which they are not especially de- 

 veloped, as Heterometra quinduplicava, commonly have cirri with fewer segments. 

 Again in certain species, as in Oligometra serripinna and in the species of Prometra, 

 there may be more or less variation in the number of segments in the enlarged 

 proximal pinnules; this is found upon examination commonly to agree directly 

 with a similar variation in the number of cirrus segments. This correlation is most 

 marked and most obvious in the Comasteridse. In this family species with large and 

 very long proximal pimiules which are stout basally, like Oomanthus hennetti or C. 

 pinguis, have very large and stout cirri with a large number of segments, wliile 

 species with a few small and weak cirri, or none at all, as Oomanthus annulata or C. 

 parvicirra, or many of the species of the genus ComaMer, have the proximal pinnules 

 small. 



This interrelationship between the cirri and the proximal pinnules appears to 

 be confined to the Ohgophreata, and in this group it is of more or less uncertain 

 occurrence, being by no means general. 



There is a closer and more widespread agreement between these two sets of 

 structures in regard to the modification of the distal ends of the component segments, 

 an agreement which is further correlated with a similar modification of the ossicles 

 of the calyx, the di%nsion series and the arm bases. In cases where, as in Thalas- 

 sometra vilhsa, Stylometra spinifera, or in the species of the genus Colohometra, the 

 distal ends of the cirrus segments are produced and spinous, the calyx and arm bases, 

 as well as the distal edges of the segments of the proximal pinnules, will also be found 

 to be spinous, though this spinosity is less, and may be entirely suppressed on the 

 brachials, from the fourth onward, and on the genital and distal pinnules. This 

 type of correlation is not found outside of the Ohgophreata, except in the genus 

 Zenometra. 



The striking correlation, both in structure ajid in function, between the cirri 

 and the ungrooved pinnules in ComatuleUa hrachwlata has already been discussed 

 in detail. 



Mention must also be made of the curious case illustrated by the families 

 Thalassometridse, CharitometridiE and Tropiometridse. In the Charitometridse 

 and TropionietridEe smooth and very stout cirri accompany very slender many 

 jointed proximal pimiules; the long and spiny cirri of the Thalassometridse occur 

 together with greatly enlarged, swollen, and elongated proximal pinnules, the 

 accentuation of these characters in the latter being to a considerable degree cor- 

 related with the proportionate length of the cirri. 



Though in Asteromctra, Pterometra and Ptilometra (which together form the 

 subfamily Ptilometrinae) the cirri are excessively long, and are in structure just hke 



