
Plate IX.
The genus Amphiura, to which the little brittle-star which forms the subject of this paper belongs, contains more species than any other genus of the Ophiuroidea, and is found in all-parts of the world, and at all depths. It is not therefore surprising to find another species of this cosmopolitan genus in-habiting our waters. The only species hitherto known from. New Zealand are Amphiura parva, Hutton (Trans. N.Z. Inst., vol. xi., p. 305, 1879), from Dunedin Harbour, and Amphiura lanceolata, Lyman (“‘Challenger’ Report,” vol. v., Ophiuroidea, p. 133, 1882), dredged by the naturalists of the “Challenger” expedition, off the east coast of the North Island, in 700 fathoms water. I now add a third, discovered in my first attempt at dredging in Wellington Harbour, 20th January, 1892, and since then every haul of the dredge in deep water has revealed many specimens. They exist in great numbers in all the deeper parts of our harbour—from 10 to 15 fathoms—-where the bottom is composed of soft, grey mud, from which they derive their nutriment by extracting the organic matter. I find them always associated with a small heart-shaped sea-urchin—Echinocardium australe (E. zelandicum, Gray)—which evidently delights in the same localities and conditions.
This species appears to have the peculiar habit of throwing off its disc and renewing it, which has also been observed by Professor Verrill in Amphiura abdita* (Annals Nat. Hist., ser. 5, vol. ix., p. 476). All my specimens dredged from January to April had perfect discs, with fully developed scaling; but in a dredging taken on the 14th October, when a large number of specimens were obtained, three of them had no discs at all, the teeth and bases of the arms above being quite bare, showing where the disc had been. The other specimens had discs in different stages of development, some consisting merely of granular skin with impressions where the radial
[Footnote] * It is well known that several other Echinoderms undergo evisceration and regeneration of the disc and visceral mass—as, for instance, Antedon rosacea, (Ann. and Mag. Nat. Hist., of Sept., 1893, p. 197)–but it has not yet been proved whether this takes place spontaneously or is the result of accident. My observations on Amphiura rosea seem to indicate, that this species does eviscerate spontaneously. Further observation will, however, be necessary to decide this point satisfactorily.

shields were forming, in others the scaling was more advanced and the discs full size. They were so delicate that it was very difficult to handle them without breaking them to pieces, and when taken out of spirit the discs collapsed completely, as if they were only covered with skin, the scaling not being sufficiently formed to give any support. One or two of the arms of several specimens had also been broken off and were partially restored.
Amphiura rosea, sp. nov. Plate IX.
Disc small, slightly tumid, roundly subpentagonal in form, with constrictions in the interradial spaces, covered above and below with rather coarse, circular, overlapping, very irregular scales, among which the primaries can usually be made out; scaling on the interradial spaces below more regular and finer than above. Radial shields long and narrow, bluntly pointed within, meeting without (in some specimens they do not quite meet), separated within by a wedge of two or three long narrow scales; they do not reach the edge of the disc, but are separated from it by a few irregularly-shaped scales. Two short, stout, bluntly-pointed, rounded mouth-papillæ at the apex of the mouth-angle, and one short, stout, and bluntly-pointed on either side at the base of the mouth-angle. Five roundish, flattened teeth, the ends truncated or bevelled, the uppermost longest and divergent, the lowermost smallest and sometimes pointed, resembling a mouth-papilla. Mouth-shields rather large, circular in form, usually with a slight peak within; madreporic shield distinct, and a little larger than the others. Side mouth-shields trigonal, with re-entering curves, the longest angle within where they do not meet. Arms long, slender, and regularly tapering. Upper arm - plates transversely oval. First under arm-plate irregularly hexagonal, small, often indefinite in shape, those beyond pentagonal, a truncated angle within, and a slightly re-entering curve and rounded angles without. Side arm-plates with prominent spine-crest, meeting neither above nor below. Arm-spines rather slender, cylindrical, tapering, divergent, four sometimes five near the disc and three on the outer part of the arms, the two lower ones rather longer than the upper. Two very minute tentacle scales, one on the under arm-plate, the other on the side arm-plate, sometimes only one on the under arm-plate.
Colour in life often a generally-rosy red; sometimes the disc is grey or purplish-grey, and brownish towards the centre above. The arms are often grey or reddish-brown near the disc—this colour extends to a greater distance from the disc on some of the arms than on others even in the same specimen—and light-red or white beyond. In a number of speci-

mens red is by far the prevailing colour. The red colour soon fades in dried specimens, and even more quickly in alcohol.
Diameter of the disc 6·3mm. Length of arm from centre of disc about 72mm. Width of arm close to the disc, without its spines, 0·9mm. Length of longest arm-spines about 0·6mm. Length of radial shield 1·52mm.; breadth 0·45mm. Length of upper arm-plate near the disc 0·47mm.; breadth 0·72mm. Length of lower arm-plate 0·5mm.; breadth 0·4mm.
This species, I should say, would find its place in the “Table of Species of Amphiura” given by Mr. Lyman in his Monograph (“‘Challenger’ Report,” vol. v., Ophiuroidea, p. 123) near Amphiura bellis, to which it appears to be closely allied. It can, however, be readily distinguished from that species by the coarser and irregular scaling of the disc and the shape of the under arm-plates and mouth parts.

