in P. frontalis it is covered, and in P. verreauxii it is enclosed; and MilneEdwards observes that many other Crustacea offer examples of these three organic forms. For instance Pagurus coenobites and Calianassa have the ophthalmic somite exposed as in Palinurus vulgaris; Homarus, Crangon, Palamon, Galathea, Lithodes, Ranina, &c. have this somite covered as in Palinurus frontalis; and Homola has the ophthalmic somite enclosed.

In Astacus the ophthalmic somite is reduced to a minimum extent, and it is only partially protected by the anterior projection of the rostrum of the

carapace.

Milne-Edwards says that, independently of the somite, the ocular appendages are formed of three "articles" or joints, a coxophthalmite, a basophthalmite, and a podophthalmite, but that ordinarily the coxophthalmite is rudimentary or obsolete.

In the genus Alpheus (fig. 10) and other fossorial marine forms the ocular appendage is reduced to an extent that allows the carapace to cover it entirely; but in the larval form the organ (fig. 11) is seen to be as well developed and as prominent as that of any aquatic species. It is in this way we may assume that the sessile condition of the organ in the Edriophthalmia (fig. 12) has been attained, first by the contraction or reduction in extent of the ocular appendage, so that the anterior wall of the carapace shall cover it, and then by the more intimate connexion of the organs with the structure of the parts that protect them, and ultimately with entire absorption of the ocular appendage; the eye receives its support from the walls of the carapace alone.

Even here the organs are themselves still liable to depreciation; thus those that exist where light is absent (which inhabit deep wells, subterranean cavos, and excavations in the depths of the ocean) first lose the dark colour of the reflecting pigments, which is soon followed by a degeneration of the character and appearance of the lens. In Ampelisca, an Amphipod that lives in muddy bottoms, all the lenses but two have disappeared, and the pigment has become red; in the well-shrimp (Niphargus) the only trace of an eye exists in some yellow-looking pigment; while in the Podophthalmia we find that Polycheles (Heller), a prawn from the Adriatic closely allied to (if not identical with) Didamia from the deep-sea dredging of the 'Challenger' expedition, and another from the Mammoth Caves of America, as well as Nephrops Stewarti (Wood-Mason) from Formosa, have the eyes wanting as organs of vision, while they retain them as obsolete appendages. The second pair of appendages is the first pair of antennæ. These M. Milne-Edwards has named (for the sake of convenience in distinguishing them from the second pair) the antennules. But as this term is one, in itself, that is suggestive of diminutiveness and inferiority, I think that it had better be employed as little as possible. Generally speaking, this pair is smaller in proportion than the second; but usually it is of a more highly organized structure, and diminishes in dimensions as it becomes important in its functional properties.

The appendage consists, in its normal condition, of three joints, homotypical of the coxa, the basos, and ischium of the true legs in Crustacea. These three joints support an extremity that is very liable to vary in form, number of branches, and general appearance; but one of them must be regarded as the primary branch, inasmuch as it is invariably furnished with a set of organs peculiar to it, and found on no other part of the animal. These are slender, delicate, membranous, thread-like processes, that are liable to vary somewhat in form and size, but are all but universally present

in aquatic Crustacea, and which, from their supposed connexion with the sense of hearing, I have elsewhere denominated aural cilia. The secondary branch is less important, and frequently divides into two or more rami. Sometimes these flagelliform branches are reduced in size to a minimum amount, and this generally corresponds with the highest character of the organ; for it appears to be in inverse ratio-the longer and more extensive the character of the terminal flagella, the less developed is the structural condition of the organ of sense contained within the peduncle; and, on the other hand, the more developed the sensational organ, the feebler and less numerous is the organism and less antenna-like is the general character of the distal portions of the appendage. To this very constant condition in the aquatic forms of Crustacea we have a variation in the terrestrial species. In the genus Oniscus and allied forms of Isopoda, as well as in the littoral varieties of Amphipoda, such as Talitrus, Orchestia, &c., the first pair of antennæ are reduced to a minimum proportion consistent with their presence, without any increased importance in the structural condition of the peduncular joints, as far as I have been able to ascertain.

In the highest types of Crustacea the coxal joint is considerably enlarged (vide pl. i. fig. 8 b, Report for 1875), and contains within it a complicated chamber and highly developed organ of sense; while in the Macrurous forms a less complicated chamber exists, with an external opening into which small grains of sand find their way in others, as first shown by Professor Huxley in a species of Stomapod, well-developed forms resembling otolithes. are present; this Dana has observed, and I have been enabled to confirm in a species of Anchistia from Australia (Pl. II. figs. 13 & 14).

In some genera, as Mysis among the Stomapoda, they vary in form according to sexual distinction. The male animal has the two terminal flagella feeble and slender, while a fasciculus of strong hook-formed hairs are planted on the inner and lower angle of the most distal extremity of the second joint of the peduncle, while a similar but less powerful group of spinelike hairs are planted on a strongly projecting process on the inferior distal extremity of the first joint (Pl. II. fig. 15). There are other hairs implanted on the lower margin of this joint of a very delicate ciliated character. The peduncle of this antenna is very powerful, and there can be little doubt but that it is useful as an organ of prehension, most probably employed in securing the mate. These several facts are demonstrative evidence that the first pair of antennæ are connected with the acoustic properties.

Of this I purpose treating, as well as discussing the observations made by Dr. Hesen in his researches (published in 1864) on the auditory organs of the Decapod Crustacea, when I report on the internal structure of the animal.

Contrary to a possible condition of all other appendages, the coxal joint of the first pair of antennæ is never absorbed into or fused with the sternal portion or ventral arc of the somite to which it belongs.

The third pair of appendages consists of the second pair of antennæ. These are often very large and powerful organs, frequently adapted as weapons of offence and prehension. They consist of two divisions similar to the first pair, that is, a peduncular and flagelliform part. Of these the peduncular consists of five joints, the flagelliform extremity of a strong, solitary, multiarticulate rod in its most normal condition; but it very frequently varies in form, but never increases in the number of its branches. In the Macrura generally the flagellum is produced, on an average, to about the length of the animal, and is mostly multiarticulate in its character, the small articuli varying in number and length. Sometimes, as in Scyllarus

(fig. 16), it consists of a single disk-like plate. But the greatest tendency to variation in form exists in the Amphipod and Isopod Crustacea. In some of these it reaches to a very considerable length and is multiarticulate, but in others it is reduced sometimes in length, sometimes in form. In Talitrus it is reduced without alteration of character to a very small size; so it is in Hyperia; but while in the former it stands on a long and powerful peduncle, in the latter the peduncle is short and feeble. In Chelura the flagellum is broad, flat, and uniarticulate, and fringed with a dense mass of soft hairs. In Podocerus and a few closely allied genera the flagellum is formed of one or two large articuli or joints, and the hairs are reduced in number but increased in strength, and become hook-like spines. In Corophium the whole antenna bears a near resemblance to a true walking-appendage, and is no doubt used to assist in progression, as is mostly the case with Crustacea that inhabit tubes and hollows of their own excavation or building. The peduncle of this antenna is invariably formed of five joints. These

are:

The first, for which Professor Milne-Edwards has suggested, in the memoir quoted, the name of coxocerite. This contains within it an organ of sense which Milne-Edwards believes to be connected with that of hearing; but I think there will be little difficulty, when reporting on the internal anatomy, in showing that it is connected with the olfactory sense. In the Amphipoda and Isopoda, with but few exceptions, such as Talitrus, Orchestia, &c., the first joint is free; but so it is in many of the Macrurous forms, such as Astacus, Homarus, &c. But in Palinurus it is strongly built into and fused with the ventral arc of the fourth or next approximating somite. These parts are still more closely associated in the Brachyurous form, so that it is difficult to determine where the antennæ end and the region named by Latreille the epistome commences.

The second joint, named by Milne-Edwards the basocerite, is generally short and supports at its extremity a movable squamiform appendage, to which the same carcinologist has given the name of scaphocerite. This appendage is constant in all Macrurous forms of Crustacea. It appears to be wanting in the genus Palinurus only; but even here it is represented, as I had the opportunity of showing, in the Report on "The Marine Fauna of Devon and Cornwall," by a figure of it incorporated in the integument of the succeeding joint, as if it were absorbed by pressure against it.

This appendage (scaphocerite) does not exist in any of the forms higher or lower than the Macrura, except Pontia (Pl. II. fig. 18) in the Entomostracous forms, and that peculiarly interesting little Isopod Apseudes, in which genus we find a small squamiform plate resembling and probably homologous with it. The third joint the above author has named the ischiocerite, and the two following the mesocerite and the carpocerite, while the multiarticulate flagellum, which corresponds "to the penultimate joint of the thoracic member," he calls the procerite. It is rather a curious oversight that, while Milne-Edwards has been most particular in identifying the several parts of the second antennæ by an especial name, he has omitted to give any to those of the first pair of antennæ, the three joints of the peduncle of which are homotypical of the coxocerite, the basocerite, and the ischiocerite of the second pair of antennæ; but the flagellum, instead of being homotypical of the procerite, represents the mesocerite and the successive articulations.

In the Macrura generally the joints of the peduncle are distinctly separated from one another; but in some of the higher forms, such as Astacus, Homarus, and Palinurus, they exhibit a tendency to crowd and coalesce with each other,

that is increased in the Anomura, and carried to such a degree in the Brachyura, that in some, as in Menethæus, Leptopodus, Maia, &c., the first two or three articulations are not to be distinguished from the surrounding structure except by the position of the olfactory opening.

In the Canceridae all the joints of the peduncle (Pl. II. fig. 17) are fused together and are so closely implanted in the structure of the facial portion of the two first somites that they assist more or less perfectly in forming the walls of the ocular orbit, the several variations of which are made use of by Alphonse Milne-Edwards as a means of assisting him to distinguish the several genera of the Cancerides from each other, and which, from their easily accessible position, might be found a convenient aid in assisting to determine genera among fossil forms.

Among the Amphipoda all the several articulations are distinct from one another and from the body of the animal, and the olfactory organ is carried in a long tooth-like process that is open at the extremity. This arrangement is not so distinct in the Isopoda and the terrestrial Amphipoda. It also disappears in certain abnormal forms of aberrant and parasitic Isopoda.

The next succeeding, or fourth pair of appendages is among the most constant in the subkingdom. Within certain limits the mandibles vary with every genus, and would form when detached a very certain means of generic diagnosis. In the most simple condition, where they approximate in form to that of the peduncular portion of the second pair of antennæ, they exist in Nebalia (Pl. III. fig. 19). But, as stated by Milne-Edwards ("Squelette tégumentaire des Crustacés décapodes," p. 256, Ann. des Sc. Nat. 1854), the mandibles are not appendages simply applied against the mouth, but occupy of themselves a special cavity, flanking on either side the entrance to the alimentary canal, which, when the two are brought into juxtaposition in the median line, they generally close. The mandible in Nebalia (Pl. III. fig. 19) is formed of a long osseous process that projects internally, and is secured by muscular attachments to the internal dorsal surface of the carapace; a large obtuse-pointed process is projected inwards across the mouth, and antagonizes with a corresponding process on the one opposite. This process is very liable to vary in form in different genera. Beyond this process, at the root of it, springs a cylindrical osseous continuation, at the apex of which are articulated two equally long and important joints. These two joints are homologically the same that form the small appendicular appendage attached to the mandible of all Crustacea (Pl. III. fig. 21) so persistently that their absence is a fact to be recorded in the structure of special genera, such as Talitrus and Orchestia among the Amphipoda. In a scientific point of view, this appendage must be part of the primary portion of the theoretical limb. This idea also receives confirmation in the form of the mandibles of the genus Pontia of Milne-Edwards, where may be observed a secondary ramus attached to the extremity of the first joint of the appendicular branch (Pl. III. fig. 20).

This appendage M. Milne-Edwards, in the nomenclature that he has given, proposes to name the protognath; but the first joint, or true mandibular portion, he calls the proto-coxognathite, and the second joint the protobasognathite, and the other joints in succession after the names of the respective joints in the ideal appendage which they homologically represent. While wishing to give all honour to that distinguished carcinologist for the care and exactitude in determining the several parts of the structure of a crustacean by means of a distinct nomenclature, it is with regret that I am compelled to admit that they would be more practically useful, and consequently more generally adopted, if the terms were less

lengthy, and with a less redundancy of expression. I shall therefore in this report, as far as possible, adopt the terms of definition proposed by MilneEdwards, but omit generally the appendicular term so constantly repeated by him. Thus the terms coxa, basos, ischium, mesos, carpus, propodos, and dactylos will be sufficient for whatever appendage I may be writing about, without repeating the name of the appendage, whether gnathite, podite, cerite, or other, after that of each individual joint.

But it is only just that Professor Milne-Edwards's reasons for adopting these terms should be reported in his own words. Writing of the appendages of the mouth, he says::

"Depuis les beaux travaux de Savigny sur la bouche des animaux articulés, on s'accorde généralement à considérer tous ces organes comme étant des homologues des pattes, mais on les distingue presque toujours entre eux sous les noms particuliers de mandibules, mâchoires proprement dites et mâchoires auxiliaires ou pattes-mâchoires; ces désignations spéciales sont quelquefois utiles; mais, dans la plupart des cas, il est préférable de considérer tous ces appendices masticateurs comme des membres d'un seul et même groupe organique, de leur donner un nom commun, et de spécialiser ce nom par l'adjonction d'une racine adjective; on pourrait de la sorte les appeler protognathe, deutognathe, etc. et faire entrer le mot gnathite, comme racine constant, dans la composition des noms appliqués à chacun des articles, ou éléments sclérodermiques, dont ils sont formés. Ces gnathites seraient différenciés à l'aide d'un certain nombre de racines adjectives indiquant leur position dans le membre, et lorsque dans les descriptions zoologiques on aurait à en parler, on pourrait se borner à ajouter aux noms composés, qui appartiendraient en commun à tous les termes de chaque série des pièces homologues, un numéro d'ordre pour indiquer leur position dans cette série organique, c'est-à-dire les appendices auxquels ils appartiennent. Ainsi je proposerai d'appeler coxognathite, basignathite, mésognathite, etc. les articles qui, dans la série des appendices maxillaires correspondent au coxite, au basite, etc. dans les autres membres, et d'appeler premier coxognathite la pièce de cet ordre qui appartient au protognathite, deuxième coxognathite celle qui appartient au deutognathite, etc. Ce système de nomenclature est à la fois si bref, si commode et si éminemment significatif, que je demande aux carcinologistes la permission d'en faire usage non seulement dans les considérations morphologiques dont je m'occupe ici, mais aussi dans les travaux taxologiques que je me propose de publier prochainement."-" Squelette tegumentaire des Crustacés décapodes," Ann. Sc. Nat. 1854, p. 267.

The mandible or protognathe is sometimes very large, and at others reduced to a rudimentary condition. In Palinurus it occupies on each side one half of the breadth of the animal, and to remove the two mandibles is almost to decapitate the animal. In some of the parasitic forms it is reduced to a rudimentary condition. In the female of Anceus (Pranisa) it, with other appendages, coalesces to form a probing or lancing instrument that projects like a proboscis beyond the head; while in the male of the same genus the mandibles are situated on the anterior margin of the head, and stand projecting like a pair of rude irregular antennæ. But in this animal the mouth is closed, or at most represented by a microscopic aperture, as it, in this stage, exists without eating.

In most forms of Crustacea the space that exists between the anterior margin of the protognathe or mandible and the posterior margin of the epistome is occupied by a fold of the membranous tissue that encloses the oral cavity. This fold is frequently ossified and projected into a strong