142 COLLECTION'S FROM MELANESIA. 



become less conspicuous and much reduced; the white lines along 

 the arm are broken into by blue patches, much less extensively de- 

 veloped than in the dark form ; on the lower surface of the arm 

 the blue lines may be present as continuous tracts, or they may here 

 and there be interrupted by white. 



Port Curtis ; Thursday Island ; Port Darwin. 



16. Ophiothrix striolata. 

 Lyman, p. 36. 



Thursday Island. 



17. Ophiothrix galateae. 

 Lyman, p. 36. 



Port Darwin. 



18. Ophiothrix ciliaris. 

 Lyman, p. 35. 



Port Jackson, 0-5 fins. ; Port Molle. 



19. Ophiothrix rotata. 

 Martens, Arch.f. Nat. 1870, p. 258. 



A single specimen, without doubt referred to this species, differs 

 in one or two points from that described by Dr. von Martens. In 

 the Berlin-Museum specimen the diameter of the disk is 7 millim., 

 and the length of the arms 35 millim. In our specimen the arms 

 must have been nearly 150 millim. long, while the diameter of the 

 disk is 12 millim. The upper spines are not more than twice the 

 width of the arm, instead of four times. The original describcr 

 makes two statements with regard to the colour of the oral shields : 

 — " Unterseite der Scheibe mit den Mundschildern und die Arm- 

 stacheln blass " ; and " Das der Madreporenplatte zugehorige Hund- 

 schild ist merklich grosser, an den tSeiten nicht eingebuchtet und 

 weiss, nicht wie die andern violett." In the specimen- now under 

 examination there is some violet marking on each one of the mouth- 

 shields. 



Thursday Island, 3-4 fms. 



So far as the present collection allows me to form any ideas with 

 regard to the range of variation within the limits of a " species,'' and 

 the value of the colour-markings on which previous investigators have 

 laid, and, as it seemed, justifiably, very considerable stress, I am in- 

 clined to the view that the variation is very much greater than was 

 supposed, and that, after all, colour-marking, though an important 

 aid in the discrimination of the species, can hardly be said to have 

 the value which has been attached to it. The doubts first raised 

 by a study of 0. martensi (vide supra) are not a little strengthened 

 by the three specimens now lying before me, which, I have little 



