210 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. 



The general coloration is light pink; the dorsal surface of the 

 disk is light; the arms are a little darker and show slightly darker 

 annotations. On the ventral surface the disk is light gray and the 

 arms are pinkish gray, with traces of annulations and a darker spot 

 toward the middle of each of the under arm plates; the spines are 

 colorless. Some dark purple dots are found on the radial shields and 

 on the upper arm plates at the base of the arms; there is not the 

 least trace of a dorsal median line. 



In 1904 I had the opportunity of redescribing the original speci- 

 men of O. aspidota, in which the diameter of the disk is 10.5 mm.; 

 at the same time I described the characters of a specimen from Trin- 

 comalee in my own collection, which I referred to O. aspidota. It 

 seems to me worth while to reproduce here (pi. 32, figs. 1, 2) two 

 photographs of this last for comparison with the individual col- 

 lected by the Albatross represented in figures 3, 4, 5. 



Miiller and Troschel's type is completely decolorized, and it is 

 impossible to learn the color which it had in life; that from Trin- 

 comalee is violet gray ; the radial shields are lighter than the rest of 

 the dorsal surface of the disk, and they show toward their borders a 

 rather irregular and noncontinuous line of dark blue ; the upper arm 

 plates show toward their distal outer angle a lighter line, and here 

 and there is an indication, very vague, however, of a lighter median 

 dorsal line; the spines are gray. The Albatross specimen differs very 

 sharply in its coloration from that from Trincomalee, but in all its 

 morphological characters I find it so conformable with the type of 

 O. aspidota that it does not seem to me possible to separate it from 

 that species ; the question of the color can not enter into consideration 

 here because the type of O. aspidota is entirely faded out. 



The radial shields of the specimen collected by the Albatross are 

 large, absolutely naked, and their surface is very finely granulose. 

 The granulation of the upper arm plates is scarcely marked. The 

 form of these plates is absolutely identical with that in the type; 

 they are trapezoidal with the distal border almost straight, and the 

 lateral angles are very sharp (pi. 33, fig. 7). I remarked in 1904 the 

 peculiar form of the adoral plates in 0. aspidota; these plates are 

 widely removed from the median interradial line and they show an 

 irregularly rounded principal portion continuing by a slender process 

 which separates the mouth shield from the first side arm plate; I 

 find this form again in the Albatross specimen, as I had also observed 

 it in that from Trincomalee. 



The only character by whch the Albatross specimen differs from 

 the two others is the form of the under arm plates, which are square, 

 as long as broad, with the distal border very gently rounded. In 

 Miiller and Troschel's type and in the specimen from Trincomalee 

 these plates are distinctly broader than long. I believe that this 



