they exhibit in a marked manner characters which distinguish them from their homologues belonging to the hinder feet."

"This nomenclature, which at first appears a happy one, has many inconveniences, and is frequently inapplicable to particular species. In the first place, the appendages of the cephalic lobe are sometimes quite different from each other in formation, as well as in structure, on which account most authors have given them different names. What a difference, for example, there is between the palpi (lateral antennæ, And. and Ed. Qtfg.) and the frontal antennæ of the Lycoridians. The first are fleshy, many-jointed, partially retractile, filled by the prolongation of the largest nerve in the body; the second are filiform, simple, non-retractile, poor in nerves. An equal distance separates the palpi (Kimberg, and all recent authors) and the true antennæ of the Aphroditians. . . . A second inconvenience of M. Quatrefages's nomenclature is its inapplicability in all the cases in which the anterior segments are very condensed, and in which it is not possible to determine to which segment a pair of appendages belong. The learned Academician, enamoured of his theory, suppresses by a stroke of his pen the buccal segment of the major part of the Sigalionides-at least, he ascribes to them only "an indistinct buccal ring, deprived of appendages." Nothing is, however, more distinct than the buccal segment of these worms, only it carries a pair of feet with their bristles, and cannot be a buccal segment according to the theory of M. Quatrefages. Unfortunately, the author does not know that all the Polynoes have bristles on the segment which he regards as buccal, and that it would be necessary to imagine for them a distinct buccal segment without appendages."

"M. Quatrefages, however, gives a rule difficult of application, but still a rule for the determination of segments and their appendages. The cephalic lobe and the antennæ, he says, receive their nerves from the cerebral ganglion; the buccal segment and its tentacles, from the oesophagal connections; the tentacular cirri, from the ventral ganglionic chain. This proposition cannot be maintained in the face of modern embryology. Schaum had already asserted that, throughout the articulata, each segment is characterized by the possession of a ganglion, and from this principle he denies that the head of the Anthropoda is formed of many united segments. This doctrine was immediately refuted. In fact, the nervous system differentiates itself relatively very late among the Articulata. On the contrary, the appearance of segments-protozonites, as they are called-is, in many cases, the result of one of the first modifications of the blastoderm. These primitive segments unite in groups, and frequently solder themselves together

long before the differentiation of the nervous system; and when this system is developed, the number of ganglions is not necessarily identical with that of the primitive segments. Amongst Annelids, in particular, the formation of a nervous system sometimes follows pretty closely that of the segmentsamong the embryos of the Capitelles, for example-but it is usually later. I do not dispute that among many Annelids the birth and distribution of nerves conform to the rule of M. Quatrefages. But we see that in certain cases it is not the buccal segment only, but also some of the following segments, which receive their nerves from the oesophagal connexions. It is so among certain Aphroditians, Hesionians, etc. . . . . I employ the name Antenna for all the appendages of the cephalic lobe; but when two of these appendages spring from the inferior portion of the lobe, and present special anatomical and physiological characters. I, in common with the majority of authors, call them palpi. The modified cirri of the buccal and following segments are called in this memoir tentacular cirri."

"Without wishing to dilate on the conformation of the feet of the Annelids, I would indicate the relation of the bristles to the tissues environing them. Some authors consider them as enclosed in a pocket which is merely an invagination of the teguments, and others think they are engendered in an internal follicle, and only arrive in a secondary way at the surface. This second opinion is correct. Among the Hesions and others, for instance, the whole bunch of bristles come in a compact form out of a single pedal opening, but in other cases each bristle has its own special aperture. This is especially the case with the flabelliform tufts. The issuing pore of each bristle is not preformed. The bristle makes its own perforation, which is easy when the tissues are soft, but it is not so when the worm is protected by a resisting cuticle, and when the bristle, armed with hooks in different directions, might entangle themselves in the tissues, and produce serious rents. In these cases the extremity of each bristle is surmounted with a small provisional apparatus, terminating in a very keen blade, destined to cleave a free way for the bristle, and to avoid tearing. The form of this cutter varies much with that of the bristle, and especially of the hooks, whose passage must take place without tension of the adjacent parts."

Teguments and Muscular Apparatus." The teguments are composed of two layers-one internal and cellular (Corium, Rathke, derme, Quatrefages), corresponding to the sub-cuticular or chitinogenous layer of other Articulata; the other extracellular, the cuticle (epidérme, Rathke, Qtrfg.) sometimes very delicate, sometimes composed of a thick layer of chitin. Up

to the present the teguments have only been carefully studied by Kölliker, to whom we owe many excellent papers on the histology of the Annelids, which are unfortunately quite unknown to the author of the "Histoire Naturelle des Annelés."

"The superficial layer deserves the name which Kölliker has given it. In a histogenetic point of view it belongs entirely to the category of cuticular formations. The subcuticular layer (hypoderm, Weissman) which secretes it-often called with Kölliker epithelium-in most cases does not enable us to recognize the boundaries of its constituent cells. The nuclei appear sown with great regularity in a continuous granular layer, as M. Bauer has seen in certain Arthropoda. Whenever the cuticle attains a certain thickness it is seen to be ornamented with two systems of striæ at right angles to each other, or more frequently at about 70°, as noticed by M. Kölliker. The tubular pores (porenkanale of the Germans), wherever they exist, are disposed in lines conforming with the striæ. M. Kölliker has been struck with the difference of these pores from each other. Often, he says, they correspond only with a subjacent cell, and he asks if these openings are really homologous with the tubular pores of the Arthopoda, and do not rather resemble the openings of cutaneous glands discovered by Leydig in the Piscicoles, or the hairs of insects and crustaceans. To these questions I can reply in a positive manner that both sorts of pores exist amongst the Annelids. Those which serve for the outpouring of certain secretions appear to exist in all species. In large species they are sometimes of considerable diameter, but usually very limited. Sometimes they are found united in groups. The canalicular pores are much smaller, much nearer together, and have no resemblance to glands. They are only found in species which have a thick skin, and not in all of these. The subcuticular layer-the derme of M. Quatrefages-appears always to contain glandular follicles, and that in all parts, even in the cirri and antennæ. These follicles

empty themselves externally by means of the glandular pores. Some only secrete a thick liquid, others engender bundles of little rods (batonnets), and these I have named bacilliparous follicles; others only secrete granules."

M. Claparède explains that the bibliography of these bacillary corpuscles is very rich, and he wonders that it has altogether escaped the notice of M. Quatrefages. "Certain families," he says, "have their teguments literally covered with bacciliparous follicles, even in the cirri and the antennæ. This is especially the case with all the Aricians and Spiodians, and a great part of the Chatopterians. Their abundance is likewise remarkable in a crowd of Phyllodocians, and in certain Hesionians. Among the latter especially, their grouping and

their relation to the external pores is very remarkable. The formation of these organs is still entirely problematical. I have sometimes compared them to the cells full of needles of the Turbellarians, and to the stinging organs of the Acalephs, etc., but this is mere hypothesis."

"Tubular glands full of spherical granulations have been pointed out by us in divers Annelids. Sometimes they reach considerable dimensions, particularly among the Lycoridians, and in this case the glomration formed by the junction of glandular tubes was known to old authors and considered to be a sac. M. Quatrefages knew of one of the passages in which I had spoken of these organs, and cited an analogous observation of M. Keferstein, but by a singular blunder, he makes us describe circumvolutions of blood vessels when we spoke very positively of glandular branches (boyaux).

Muscular system.-The muscles of Annelids exhibit extraordinary varieties in their histological character. Sometimes they are composed of parallel fibres quite destitute of a nucleus, and at others of fibre-cells furnished with large nuclei. The existence of these fibre-cells of a muscular character in Annelids has been, it is true, denied by M. Schneider; but "although this savant is right in the majority of cases, we shall see that the rule suffers certain exceptions in the pharynx of some Nereids, tentacles of many Terebellians, etc. Sometimes the muscular fibre separates into two distinct layers, one axial, the other cortical, as Leydig first observed."

"The Historie Naturelle des Annelés' describes between each segment a sort of tendinous raphis, in which the muscular fascia insert themselves. These raphids have no existence. It is easy in longitudinal sections of Annelids to see that the longitudinal fascia are continued without interruption the whole length of the worm. This is what Blainville, Della Chiaje, Rathke, Meckel, etc., observed. Muscular plates more or less complete, sometimes divide the previsceral cavity into several chambers. Thus the Ophelians, Polycirridians, and many Terebellians, the Aphroditians, and the majority of the wandering Annelids have three divisions in the perivisceral cavity, the Glycercians two, etc."

Digestive Organs.-M. Quatrefages has endeavoured to establish a new nomenclature for different parts of the alimentary tube, the necessity for which may be disputed. Why, for example, in the Syllidians give the name dental region of the proboscis to the organ with glandular walls, which does not form part of the proboscis, and contains no teeth.* The

* In a note, M. Claparède says that M. Quatrefages has enumerated certain Syllidians as having teeth in this region, but has mistaken the species, which belong to quite different families.

names employed by other authors such as fleshy portion of the pharynx (M. Edwards), gizzard (Williams), proventriculum (Oersted), appear very preferable. Are there sufficient reasons for rejecting the names ventricle and glands of the ventricle, employed first by Rathke in reference to the Nereids, and repeated by his successors? Is it necessary to replace them by oesophagus, and salivary glands? . . . Salivary glands are usually more or less connected with the buccal cavity, while the glands in question are often twenty or thirty segments behind it. In certain Annelids the posterior region of the intestine following the biliary region, presents a special appearance. Its walls appear filled with cells containing curious concretions, without doubt destined to be eliminated with the fœces. I call this portion of the intestine the urinary region, though it does not appear to contain uric acid."

The Perivisceral and Circulatory System.-We owe to Quatrefages and Williams, especially to the former, a profound study of the perivisceral cavity and of the lymph which it contains. The perivisceral cavity is lined with a delicate membrane which is only demonstrable in large species, and which M. Quatrefages attributes the discovery of to himself. Had he searched the works of Della Chiaje and Rathke, he would have found the membrane and its name. The structure of this peritoneum (tunica sierosa, tunica peritonicale of Della Chiaje) is subject to great variation. In some species the perivisceral cavity is supplied with vibratile cilia borne by this membrane. Sharpey was, if I am not mistaken, the first to describe it among the Aphroditians, Williams afterwards observed it on the branchiæ of the Glycerians, and I described it throughout their perivisceral cavity. It was also noticed in the Tomopteridians. M. Quatrefages, who only cites en passant the observations of Williams, adds that this ciliary movement was known to him long ago in a number of Annelids, and that it may be seen in all species if any trouble is taken. This opinion is not correct. The immense majority of Annelids do not exhibit a ciliary movement in the perivisceral cavity, except at the entrance of the segmentary organs. I only know a perivisceral ciliation in the following groups :-all the Aphroditians, Glycerians, Polycirridians, Tomopteridians, and, lastly, in an abnormal Terebella-T. vestita. It is striking that all these Annelids, excepting the little Terebella and the Aphrodita aculeata, are totally destitute of vessels, and of these two exceptions, one, the Aphrodita, is an animal of rudimentary vascular system belonging to a vesselless family; the other, the Terebella, belongs to a family in general vascular, but of which, one tribe, the Polycirridians, is vesselless. In presence of these facts, I must regard the perivisceral ciliary