PART 5 A MONOGRAPH OF THE EXISTING CRINOIDS 97 



Thermal range. — One record, 15.72° C. 



Remarks [by A.M.C.]. — As explained in the discussion of the subfamih- (p. 49), 

 the diiferentiation of Argyrometra and Euantedon by a small difference m the number 

 of segments in Pi seems to me very imsatisfactory, particularly in view of the wide 

 range shown by other genera or even species of the Antedoninae (notablj' Antedon 

 serraia with 12 to 28 segments). If Argyrometra is to be retained at all there seems to 

 be only the veiy large nimnber of cirri, LX-LXXX in crispa and mortenseni, to dis- 

 tinguish it from Euantedon wdth less than L cii-ri. 



Another doubtful point is whether even crispa and mortenseni are congeneric. 

 Elsewhere in this subfamily Mr. Clark has used the different proportions of the first 

 three pinnules as diagnostic of genera, although in Dorometra he includes species with 

 P2 both larger and smaller than Pi, but in these two species the sizes of all three pinnules 

 are in inverse order- — P2 and P3 progressively smaller than Pi in crispa but larger in 

 mortenseni. Perhaps mortenseni should be segi-egated in a distmct genus on this ac- 

 count, but without seeing examples of it I cannot take the step of creating a new genus. 



KEY TO THE SPECIES OF ARGYROMETRA 



a'. Pi longer than P2, which is longer than P3; 10-12 cirrus segments, the longest not much elongated 

 and with little or no expansion of the distal ends (Hawaiian Islands; 270-298 meters). 



crispa (p. 97) 



a». Pi shorter than P2, which is shorter than P3; about 17 elongate cirrus segments (northern Xew 

 Zealand; 110-175 meters) mortenseni (p. 98) 



ARGYKOMETEA CRISPA (A. H. Clark) 



Figure 7 



Iridomctra crispa A. H. Claek, Proc. Biol. Soe. Washington, vol. 21, 1908, p. 131 (listed; nomen 



nudu7ii); Proc. U.S. Nat. Mus., vol. 34, 1908, p. 213 (in key), p. 218 (description; Albatross Sta. 



3938). 

 Compsomelra crispa A. H. Clark, Crinoids of the Indian Ocean, 1912, p. 229 (listed); Unstalked 



crinoids of the Siboga-Exj>ed., 1918, p. 210 (not a Compsomelra; closely related to the species of 



Andrometra) . 

 Argyrometra crispa A. H. Clark, Journ. Washington Acad. Sci., vol. 7, 1917, No. 5, p. 128 (listed), 



p. 129 (comparison with A. mortenseni). 

 Argyrometra {Compsomelra) crispa Gisl^n, Nova Acta Reg. Soc. Sci. Upsaliensis, ser. 4, vol. 5, No. 



6, 1922, p. 130. 



Diagnostic features.- — Pi is longer than P2, which is longer than P3; the pinnule 

 segments have rather prominently everted and spinous distal ends; there are 10 to 12 

 cuTus segments of which the longest are not much elongated, with little or no expansion 

 of the distal ends. 



The arms are about 30 mm. in the unique holotj^e and the cirri are 4 mm. in 

 length. 



Description. — The centrodorsal is hemispherical, bearing about 70 cirrus sockets 

 which have no definite arrangement. 



The cirri are LXX, 10-12, 4 mm. long. The first segment is short, and the 

 remainder are longer than broad, the thu-d, fourth and fifth being the longest. The 

 opposing spine is represented by a small tubercle. 



The radials are almost entirely concealed. The IBri are short, narrowmg rapidly 

 distalh', and very deeply incised in the median line, well rounded dorsallj- and widely 



