1G2 THE VOYAGE OF H.M.S. CHALLENGER. 



examples all the joints except those just at the base are much wider thau long 

 (PI. XXVIII. figs. 1, 2), the later cirrus-joints of the Scotch variety are relatively shorter 

 and thicker, so that in the extreme forms they are actually wider than long (PI. XXVII. 

 figs. 23, 24). 



The longest cirrus which I have found in specimens from the Tunis coast 

 measures 52 mm., and contains forty-seven joints, while in one from the Minch there 

 are fifty-one joints, though the length is only 47 mm., and in the most extreme forms 

 from this locality there are forty-eight joints in a length of but 35 mm. I have pointed 

 out elsewhere l that these two types of cirri, apparently so different, are linked together 

 by a complete series of intermediate gradations, in all of which there is a great amount 

 of variation in the characters of the terminal claw and of its opposing spine. Evidently, 

 therefore, the only character of the cirri of Antedon phalangium on which we can at 

 all rely as having a sufficient degree of stability for specific distinction is the great 

 number of their component joints. This is common to Antedon 2^'olixa and to 

 Antedon hystrix, and it serves as a convenient means of separating these three 

 species from the large group which embraces Antedon rosacea, Antedon tenella, and 

 similar forms with shorter and fewer jointed cirri. 



Marion 2 has described the second radial of Antedon phalangium as " profondement 

 enchancree pour recevoir l'axillaire, qui est tres-grande." The four figures which he 

 gives of the calyx certainly bear out his statements. But I have seen individuals 

 from Marseilles that I owe to his kindness, and others from the Tunis coast, which have 

 much less quadrate axillaries, and therefore also less deeply incised second radials. 

 Other examples from the Tunis coast correspond to Marion's figures ; but in most of 

 these and in all the Scotch specimens the second radials are oblong in their general 

 outline, and but little incised, while the axillaries are subtriangular, subquadrate, or 

 more usually pentagonal, with their bases curving slightly outwards. Not unfrequently 

 there are forward projecting lateral processes on the second radials which are much more 

 marked in some individuals than in others. The axillaries may have slight processes of 

 the same kind, and they are continued on to the first brachials as a sort of flattening of 

 their outer sides, thus affording an approach to the condition of the BasicuTva-groxvp. 



The two first brachials, just like the two outer radials, vary considerably in their 

 shape and mutual relations. Thus, for example, in the four individuals figured by 

 Muiion the two joints borne by the axillary are well separated from one another above 

 its distal angle, and the second brachials have an irregularly quadrate shape. But in 

 the specimens dredged by the " Dacia " on the Seine Bank the second brachials are 

 almost triangular in outline, and the two joints below them are closely united above 



1 On the Variations of the Form of the Cirri in certain Comatuloe, Trans. Linn. Soc. Land. (Zool.), 1886, ser. 2, 

 vol. ii. pp. 475-480, pi. lvii. 



2 Ann. d. Sci. Nat. (Zool.), 1879, sir. 6, t. viii. p. 43, pi. xviii. fig. 11. 



