Neotanais) lævispinosus and Strongylura arctophylax, all of Norman and Stebbing, taken off the coast of Ireland, may be regarded as British. The last, however, is not very markedly distinct from the Norwegian species, Strongylura cylindrata, Sars. The Australian species, Paratanais linearis, Haswell, should perhaps be transferred to Anarthrura, hitherto represented only by Anarthrura simplex, Sars, from Norway.

There are difficulties connected with the study of the Tanaidæ, owing to the differences that often exist between the two sexes, to the likenesses that sometimes exist between the females of different species, and to the prevailing minuteness of size, which descends even to one-twentieth of an inch. That many of the genera are blind is readily to be explained in connection with their habit of living ensconced in the sand. When their little white or pellucid bodies are discerned amidst the fragments of crystal, shells, polyzoa, foraminifera, and spines of urchins, it is seen that these components of the sand often exceed the Tanaids in size. It is not unlikely, therefore, that many of the species have been frequently overlooked, and that in the future those already known will be discovered in many fresh localities, and that many fresh species will eventually be brought to light.

There is a still pending dispute whether the Tanaidæ should maintain their position among the Isopoda or be transferred to the Amphipoda, or be separated from both and raised to the dignity of an independent sub-order. Of these plans the least advisable seems to be that which would mix them up with the Amphipoda, for that enormous group is at present separated from all other Crustacea by characters of the branchial organs and the pleon, in which the Tanaidæ have no share, while the Tanaida, on their side, have characters of the mouth-organs foreign to all the Amphipoda. The form and position of the heart, extending in the Tanaida from the first to the last segment of the peræon, very nearly as in the Amphipod Corophium, and the other resemblances in the circulatory apparatus of the two groups, pointed out by M. Yves Delage, afford the

AN UNSETTLED QUESTION

329

most important argument for the proposed alliance between them. But the position of the heart, when considered in connection with the branchial arrangements, points rather in the direction of the Cumacea than the Amphipoda, and although in the Isopoda the heart is usually in the pleon, there are various genera in which it extends more or less into the peræon, thus establishing a gradation to the Tanaids in which it is wholly there. From the point of view of embryology, Bate and Westwood say of Tanais and Apseudes that the development is after the manner of the Amphipoda rather than that of the Isopoda,' but in point of fact, though the curvature of the embryo in the egg is inward as in the Amphipoda, the young leave the mother's pouch with the last peræopods undeveloped, as is customary with the Isopoda.

CHAPTER XXII

TRIBE II -FLABELLIFERA

THE name refers to the circumstance that in all the members of this tribe the terminal segment in conjunction with the uropods forms a caudal fan as in the Macrura, although here the telson is often not articulated, but fused with the preceding segment. The heart is situated chiefly in the pleon, only penetrating one or two of the terminal segments of the peræon. Respiration is effected by the help of pleopods acting as branchiæ.

The families included are the Anthuridæ, Gnathiidæ, Cymothoida, Serolidæ, Sphæromidæ, and Limnoriidæ. The term Cymothoidæ is used comprehensively for a closely connected group of families, as will be explained hereafter.

Family 1.-Anthuridae.

The body is long and narrow, subcylindrical or depressed. The head is usually shorter than the following segment. The peræon has its seven segments distinct. The first five segments of the pleon may be all distinct or completely or partially coalesced; the sixth is sometimes fused with the linguiform telson. The mouth-organs are suctorial. The first gnathopods are usually the larger, subchelate; the second gnathopods and first peræopods resembling them in general form. The other peræopods are ambulatory. There appear to be no separate marsupial plates. The first pleopods are large, commonly expanded to cover the rest. The uropods have a two-jointed inner branch, and the one-jointed outer branch usually so articulated as to arch more or less over the back of the telson.

Some eight genera are included in this family. The first four to be mentioned agree in having two rounded lobes to the lower lip, the mandibles with a cutting edge of two or three blunt teeth and a sort of semicircular saw in place of molar and spine-row, the 'palp' three-jointed; the first maxillæ are simple, with apical teeth, without 'palp; the second are slightly cleft at the apex; the maxillipeds have from three to six broad flattened joints including the basal piece.

Anthura, Leach, 1814, has the flagella of both pairs of antennæ few-jointed in the female, of the first multiarticulate in the male; the maxillipeds three-jointed; the first five pleon-segments fused in the female, partially distinct in the male. The telson is not coalesced with the preceding segment.

Cyathura, Norman and Stebbing, 1886, has the flagella of both antennæ rudimentary, of the first perhaps not greatly developed in the male; the maxillipeds fourjointed; the first five pleon-segments fused, at least dorsally, in the female.

Anthelura, Norman and Stebbing, 1886, has the flagella of both antennæ multiarticulate, of the first pair brushlike in the male; the maxillipeds six-jointed; the pleonsegments distinct, and the telson in the type-species separate from the sixth segment.

Hyssura, Norman and Stebbing, 1886, has the flagella of both antennæ multiarticulate; the maxillipeds sixjointed; the seventh segment of the peræon seemingly devoid of limbs; the first five pleon-segments distinct, fully half as long as broad; the pleopods alike, the first pair not covering any of the following; the uropods with the outer branch long and narrow, not arching over the narrow lanceolate telson. The animal is vermiform.

Eisothistos, Haswell, 1884, has both antennæ very short in both sexes with few-(8-6) jointed flagella; the first gnathopods not subchelate, smaller than the following limbs; the first five pleon-segments short but distinct; the first pleopods covering the rest; the uropods with outer branch narrow, not arching over the broad quadrate

telson. The animal, especially the male, is vermiform. The mouth-organs have not been described. Mr. Haswell speaks of the outer branch of the uropods as an appendage which is directly articulated with the posterior border of the segment,' a very strange peculiarity.

Haliophasma, Haswell, 1880, would seem to be a synonym of Anthura, since the species Haliophasma maculata, Haswell, has been described by Chilton under the name Anthura affinis, and stated by him to be a true Anthura.

Ptilanthura, Harger, 1878, is probably also a synonym of Anthura, the one-jointed mandibular 'palp' being due, it may be supposed, to a defect of the specimen observed.

The next two genera agree in having the lower lip bipartite but acuminate; the mandibles without teeth, lancet-like, lobes at the base forming a channel for the liquid drawn by the thrust of the lancets; the first maxillæ spear-like, distally channelled and serrate; the maxillipeds elongate, consisting of four or five joints, the second of which is elongate.

Paranthura, Bate and Westwood, 1866. The flagellum of the first antennæ in the male forms a large multiarticulate brush, in the second it is rudimentary. The mandibular 'palp' is three-jointed. The pleon has six segments and the telson distinct. The outer branch of the uropods is short, variable in width.

Calathura, Norman and Stebbing, 1886. The antennæ of both pairs in both sexes have many-jointed flagella, that of the first pair perhaps not greatly developed in the male. The pleon has six segments and the telson distinct. The outer branch of the uropods is short and wide, overarching.

Cruregens, Chilton, 1882. Both pairs of antennæ are short, few-jointed. The second pair have a small exopod. There is no mandibular 'palp,' but otherwise the mouthorgans bear a general resemblance to those of the two preceding genera. The seventh segment of the peræon is very short and devoid of limbs. The pleon has six segments and the telson distinct. The outer branch of the uropods is narrow, not overarching the squamiform telson.