280 THE VOYAGE OF H.M.S. CHALLENGER. 



mentioned thirty (which is clearly a misprint for twenty) in Lamarck's type at Paris. 1 

 But the number of cirrus-joints in the original of Asterias pectinata, Retzius, was stated 

 by Miiller as only thirteen, and that of the cirri as sixteen. 2 I have lately examined a 

 large number of specimens from eighteen different localities, all of which agree in having 

 no more than sixteen cirri with a small number of joints, and in a generally less robust 

 appearance than that of the type specimens of Actinometra Solaris. In the majority of 

 cases the number of cirrus-joints is eleven or twelve (PL LIII. fig. 15), but it may fall as 

 low as nine, though it occasionally rises to fifteen. This reduction in number is not due 

 to immaturity, as in the case of the small specimen represented in PI. LIII. fig. 1, the 

 very youthful condition of which is indicated by the shape of its arm-joints, as compared 

 with those of the larger individual shown on the same plate (fig. 2). But it is a character 

 of much constancy, accompanied by others which will be noticed immediately ; and as 

 such it will serve, I think, for the separation of the two species Actinometra pectinata and 

 Actinometra Solaris. 



To those who know the extent of the variation in the number of cirrus-joints within 

 the limits of individual species of Antedon, the difference between an average of twelve 

 and another of eighteen or twenty may seem altogether insufficient to serve as a basis of 

 specific distinction. But it must be remembered that as the number of cirrus-joints in 

 Actinometra rarely exceeds twenty-five, it has but a very slight range of possible varia- 

 tions. The number does vary in each of these two types — from nine to fifteen in the 

 one and from seventeen to twenty-three in the other, — while there are other characters 

 which also help to separate them. There may be as many as twenty-five cirri in 

 Actinometra Solaris, which is a much more robust species than Actinometra pectinata, 

 i.e., it has stouter joints, both in the cirri and in the arms and pinnules. Thus, for 

 example, taking an immature individual of Actinometra Solaris with about the same 

 " spread " as an Actinometra pectinata from the same locality, I found the corre- 

 sponding pinnules with nearly the same number of joints to be one-third longer in the 

 former than in the latter, while the arm-joints were also wider. 



The two species present a parallel series of colour-variations. The) 7 are sometimes a 

 deep purple, and sometimes brown in various shades, greyish, reddish, and blackish, or 

 occasionally more or less mottled purple and brown. Miiller described the medio-dorsal 

 line of the arms in Actinometra pectinata as marked by a white band with a dark one on 

 either side. I have seen this also in examples of Actinometra Solaris from Hong Kong 

 and Billiton. In the dry state the arms of Actinometra pectinata have a slightly raised 

 ridge in the medio-dorsal line which is hidden by the white band just mentioned, when 

 this is present. But the ridge always exists and is very frequently visible in spirit 

 specimens, though sometimes, as in that dredged by the Challenger, it is scarcely visible 

 till the arm is dried. It is never so marked in Actinometra Solaris, however, as it may be 



1 Archivf. Naturgesck, 1843, Jahrg. ix. Bd. i. p. 135. 2 Ibid., p. 133. 



