198 BULLETIN 82, UNITED STATES NATIONAL MUSEUM 



the cirri are up to XX, 15-25; the cirrus sockets, in mature individuals, are arranged 

 in about 15 crowded columns; the IBri is very short; the IIBr series are 2 or 4 (3+4); 

 the IIIBr series, when present, are 2, developed internally; and the first two brachials 

 are united by syzygy, at least after a IIBr 4(3+4) series, or a IIIBr series; PI and P 2 

 are the longest pinnules on the arm, with more than 20 segments; the pinnules then 

 rapidly decrease in length, the middle pinnules being about three-fourths or half the 

 length of the proximal ones; the genital pinnules are moderately expanded over a 

 number of segments. He said that within these limits Pachylometra inaequalis must 

 be included in the genus, and perhaps P. levigata also. He remarked that Pachylometra 

 distincta, which in 1909 I made the type of Pachylometra, is in reality a Crossometra, 

 as pointed out by him in 1928; Crossometra therefore becomes a synonym of 

 Pachylometra. 



In reviewing the 13 genera at present accepted in the family Charitometridae it 

 was found that much confusion existed. Several of the genera were based upon 

 characters of minor importance, or so variable as to be unreliable and undiagnostic. 

 These characters were chiefly those found in the arms, especially the type of arm 

 division and ornamentation; the type of genital pinnules; the length of the oral pinnules; 

 the distribution of the cirrus sockets ; and the number of cirrus segments. The characters 

 presented by the details of the cirri were disregarded; they were overlooked because 

 of the singularly uniform general appearance of the cirri throughout the family. 



The number of arms is of little significance. Several genera include only species 

 with 10 arms, in others the included species may have 10 or more than 10 arms, in 

 still others the number of arms appears always to be more than 10. In a few genera 

 the IIBr series are always 2, but in most they are chiefly 4(3+4), more rarely 2, though 

 hi some individuals all the IIBr series present may be 2. Although in a few species, 

 as in Glyptometra tuberosa, ornamentation is always present and characteristic, in most 

 of the species in which it occurs it is highly variable. The extreme is reached in Crino- 

 metra brevipinna in which some individuals have the most highly developed and com- 

 plex ornamentation to be found in the family while others are entirely plain. Some- 

 times the adults will have a characteristic ornamentation though the young are plain. 

 But strong lateral compression of the arms combined with carination involving the 

 whole dorsal surface is a reliable feature. The characters presented by the genital 

 pinnules seem to be reliable. In many species, however, these are variable, the genital 

 pinnules of some individuals being almost or quite unmodified, of others much swollen. 

 In individuals of some forms of Crinometra brevipinna the genital pinnules have the 

 third and fourth segments abruptly expanded as in Strotometra, though the slender 

 portion beyond these is always much longer than the swollen part. There appears 

 to be in general a line of demarcation between the species with the cirri in 10 columns 

 on the centrodorsal and those in which the cirri are in 15 or 5 columns; but in some, 

 as Glyptometra inaequalis, they may be in either 10 or 15 columns, in some they may 

 be in 10, 11, or 12 columns, and in others they are in 10 columns in young individuals 

 and in 15 in the fully grown. 



Of the genera heretofore recognized five seem to be without valid significance. 

 Diodontometra is based upon the new species D. bocki which is a synonym of Antedon 

 garrettiana, the type species of the genus Chlorometra. Pachylometra as defined is 

 characterized in part by having the cirri in 15 columns on the centrodorsal, though 



