the great size of two feet in length, with a carapace more than twenty-one inches in circumference. Already in 1883, Mr. T. Jeffrey Parker, F.R.S., had proposed a new subgenus Jasus for those species of Palinurus which have the rostral character assigned to Palinosytus and which have no stridulating organ. He therefore claims that the name Jasus should supersede Palinosytus.

Linupāris, White, 1847, has the rostrum dilated, bipartite, with the processes flat, and the anterior margin spinulose. To this genus belongs Linuparis trigōnus (de Haan).

Panulirus, White, 1847, contains the numerous Eastern and one or two Western species, in which there is no central rostriform tooth, which have the ocular segment exposed and membranous, the flagella of the first antennæ long and slender, and their segment produced considerably in advance of the frontal margin, that being generally armed with strong teeth. Panulirus penicillatus (Olivier) has already been mentioned as having exhibited the singular monstrosity of an eye-stalk developing a flagellum or lashlike termination. In this species Spence Bate enumerates twenty-six pairs of branchiæ, this number including six pairs of 'mastigobranchiæ,' which are in fact epipods, whether accompanied or not by podobranchiæ, which also arise from the first joint. With the help of Mr. R. I. Pocock, I have come to the conclusion that Linuparis and Panulirus were not named as generally supposed by Dr. J. E. Gray, but by Mr. Adam White, in 1847, the characters of the new genera being left to be inferred from those of the known species which were transferred to them, a slovenly method of definition which is much to be deprecated.

Palinurellus, von Martens, 1878, is distinguished from Palinurus, by the feeble antennæ, the nearly smooth carapace, and its rostriform front covering the base of the antennæ and eye-stalks. The type Palinurellus gundlachi is from Cuba. The genus Synaxes, which Spence Bate established in 1881, and retained in 1888, is described as having the rostrum produced beyond the segment of the first antennæ and united with that of the second antennæ

so as to make a perfect orbit and to cover the ocular segment. It is said to have the antennæ of Palinurus, the trunk-limbs like those of Scyllarus, the carapace like that of Astacus, and the pleopods like those of Scyllarus, so as to form a truly inosculant genus. The type species is Synaxes hybridica, Spence Bate, from the West Indies. Spence Bate himself observes that Palinurellus, von Martens, according to that author differs from Synaxes in having the posterior pair of pereiopoda chelate in the female,' but does not explain how that can be any distinction, if, in Synaxes, 'the pereiopoda are like those of Scyllarus,' as he declares them to be, for in Scyllarus also the last pair are chelate in the female. The student must be prepared sometimes to find it as difficult to reconcile authors with themselves as with one another. Under the circumstances one may accept the decision of Dr. Boas, quoted with evident approval by von Martens, in 1882, that Synaxes is a synonym of Palinurellus.

The strange form known as Phyllosoma was at one time regarded as belonging to a distinct genus, but is now known to be larval, by such marks as the median eye, and the rudimentary character or unjointed condition of the various parts. A considerable number of specimens of Phyllosoma were obtained by the Challenger, of sizes varying from the seventeenth of an inch up to an inch and two-fifths, the latter being larger than some specimens of Palinurus that have attained the permanent form.

In a

general way the Phyllosoma forms may be assigned to different stages in the development of the Scyllarida and Palinuridæ, but to assign the successive stages to particular species does not seem always possible at present, and in especial there appears to be an awkward gap between the most advanced Phyllosoma and the earliest post-larval form. No such perplexity, however, affects the first larval form, or brephalos, when actually extracted from the ovum. A specimen of this kind is shown on Plate IX., in Spence Bate's figure of a juvenile Palinurus vulgaris.

A FOSSIL'S KITH AND KIN

199

CHAPTER XIV

TRIBE IV.-ASTACIDEA

THE first antennæ carry two multiarticulate flagella; the second are furnished with a scale. The trunk-legs have seven distinct joints. The first three pairs of trunk-legs, and sometimes the other two pairs also, are chelate. The first pair are the largest. The branchiæ are well developed. The first segment of the pleon has appendages, except in the Parastacidæ.

In this tribe the young are said to be hatched in the Megalopa form. It contains four families-the Eryontidæ, Nephropsidæ, Potamobiidæ, and Parastacidæ.

Family 1.-Eryontida.

The carapace is dorsally depressed, with little or no rostrum. The eyes are wanting or abnormal. The second antennæ have a long multiarticulate flagellum. The third maxillipeds are pediform. The pleopods, except the first pair, have a process attached to the inner branch (the stylamblys of Spence Bate's terminology). The uropods have no transverse suture. The telson is tapering.

To this family are assigned seven genera, but one of these, Eryon, Desmarest, 1820, which gives its name to the family, is a fossil genus from the Lias of England and the lithographic limestone of Bavaria. It is only in recent years that the depths of the ocean have yielded forms which appear to be properly classified in close proximity to the ancient fossil species.

Polychēles, Heller, 1863, has the anterior angles of the

carapace projecting; the eye-stalks obscure, 'immovably lodged in an orbit excavated in the dorso-frontal margin of the carapace, more or less covered by the antero-lateral margin of the carapace;' the second antennæ terminating in a long and slender flagellum; the first four pairs of trunk-legs chelate; the fifth pair simple in the male, sometimes chelate in the female; the pleon not longer than the carapace. The type-species, Polycheles typhlops, Heller, was first taken in the Mediterranean. Since then various species have been recorded from both the Atlantic and the Pacific, and from depths varying from 220 to 1,070 fathoms. Since in the female all the legs are usually chelate, the generic name, meaning 'with many chelæ,' is not inappropriate.

Pentacheles, Spence Bate, 1878, meaning the creature with five chelæ,' seems to differ from Polycheles only in the particular alluded to in the generic name, the male in this instance, as well as the female, having the fifth pair of legs chelate. The genus has a wide range in both the great oceans, and the species descend to great depths.

Spence Bate observes that Pentacheles euthrix (v. Willemoes Suhm) has a close general resemblance to his own Polycheles baccata.

Stereomastis, Spence Bate, 1888, is said to differ in nothing externally from Pentacheles, but to be established 'to receive those species in which the mastigobranchial lash does not exist.' It was probably foreseen that some apology would be expected for such a definition, and the remark is accordingly appended, that difference of internal structure as a specific character is of more value than any external distinction, which, though more convenient for classification, is of little importance if it does not represent structural variation.' Yet the example of the present genus gives but feeble support to this sententious aphorism, especially as in the two preceding genera the mastigobranchial lashes are for the most part of great tenuity, and in Stereomastis Suhmi, Spence Bate, the third maxillipeds have a rudimentary mastigobranchial plate,' though the trunk-limbs are without any. The meaning of the generic name would naturally imply the presence

WHERE THE FOOD THERE THE FEEDERS 201

of a solid lash, but it is explained to mean the absence of a lash.'

Willemoesia, Grote, 1873, was at first named Deodamia by Dr. v. Willemoes Suhm, but that name was preoccupied. Here the eye-stalks are rudimentary, not lodged in a notch in the dorsal surface of the carapace, but in the frontal space. The first antennæ have the first joint produced to a scale-like process, which is forced up into a crest-like ridge; the two flagella are very unequal. The trunk-limbs are all chelate in both sexes. The telson tapers to a joint. The type species, Willemoesia leptodactyla (v. Willemoes Suhm), occurs in the Mediterranean, Atlantic, and Pacific between the depths of 1,300 and 2,225 fathoms. This and the species of the kindred genera are almost always taken on an oceanic floor of globigerina ooze, and Mr. Spence Bate infers from this that the character of the food may have been one of the most permanent influences in their geographical distribution. The remark is capable of a very extended application.

Family 2.-Nephropsida.

The carapace is sub-cylindrical, with a pronounced rostrum. The second antennæ have a long multiarticulate flagellum. The segments of the pleon are dorsally imbricated. The outer branch of the uropods has a transverse suture. The 'mastigobranchia' or epipodal plates are large, having a well-developed podobranchial plume attached to all the trunk-legs except the last pair.

Six genera are assigned to this family. Spence Bate calls it the Homarida, from Homarus, the name which Milne-Edwards gave to the genus containing the common lobster, but since that genus was already named Astacus by Leach, Homarus must be discarded as a synonym. Since a freshwater genus in a different family has been misnamed Astacus, by which the application of Astacidæ as a family name has been confused, it seems better to give a new family name to the lobsters, and for this purpose Nephropsida readily suggests itself.