JAPANESE ASTEROIDEA. 287 



abactinal surface is subject to considerable variation. In some 

 specimens the black portions preponderate so much that the whole 

 abactinal side appears a uniform black at a short distance, while 

 in others they form only small scattered patches in the general 

 orange red surface. The relative distribution of the two patches 

 are entirely irregular, and in some specimens the contrast be- 

 tween the two patches is not so sharp as in others. At the 

 centre of the disk there is generally a black patch, which may 

 be very small or cover nearly the entire disk, and may be of 

 any form. In one specimen in which the patches of the two 

 colours were nearly equally distributed, there was an outline of 

 a star- shaped polygon in black, with as many reentrant angles as 

 there were arms, each reentrant angle being interradial in posi- 

 tion ; a black strand extending inwards from the apex of each 

 projecting angle and all meeting at the centre of the disk, the 

 spaces between the black strands being orange red. In another 

 specimen, in wliich the orange red portions showed a slight 

 preponderance, there was a black patch of a very irregular shape 

 at the centre of the disk, surrounded at a short distance by a 

 ring of small discontinuous black patches running close to the 

 margin of the disk. 



The actinal side is perfectly flat, but the convexity of the 

 abactinal side is somewhat different in different specimens, depend- 

 ing among other things upon the condition in which the specimen 

 was fixed, especially in dry specimens. GreneraUy speaking, al- 

 coholic specimens show a more convex abactinal side than many 

 dried ones, and this appears to me to indicate that the flattened 

 condition of many dried specimens is probably due to pressure 

 before or durino: dessication. 



