The gills in the present example are, fortunately, preserved, but as I cannot refer to Lütken's paper^* in which Himantolophus reinhardtii was described and figured, I am not in a position to decide the question as to generic identity. It is, however, possible that with the aid of the following description others more fortunately situated may be able to do so.

The figure published in illustration of Clarke's paper^† is somewhat crude, and, gauged by the characters of our example, incorrect as regards the cephalic tentacle and the number and disposition of the dermal scutes. I have therefore thought fit to refigure the species from the assembled remains of the specimen intrusted to me. I have also essayed a description of the specimen, but owing to the imperfect condition it will be understood that the proportional measurements are merely approximate, or, it may be, even conjectural. These remarks apply, however, only to the relative width and depth of the body and head, the bones being so flexible that the character of the head may be made to assume either depressed or compressed condition, while, as before stated, the absence of the soft portions of the body renders its original shape largely conjectural.

D. I, 5; A. 4; V. O; P. 17; C. 9; Vert. ? 17.

Head enormous and grotesque, its length half that of the total, computed from the tip of the snout to the base of the caudal fin; its depth is one-fifth greater than its length, and its width is a little more than half its length. The cheeks are subvertical, and the eye is placed in a large shallow depression rather high in the head. The eye is very small, about 12.3 in the head; it lies midway between the tip of the snout and the supraorbital spine; the latter marks the termination of the supraoccipital ridge; this is widely separated from its fellow where it originates behind the premaxilla; these ridges diverge behind, but are somewhat contracted in the middle. The interorbital space is deeply concave, and from its centre the remarkable tentacle takes its origin.

The gape is very wide, and the mandibular articulation is in advance of the eye, and even in front of the tip of the snout. When closed the mouth is almost vertical.

Teeth.—The teeth are in about three irregular rows, the innermost containing the largest; they are spine-like, slightly recurved, and depressible; they are slightly longer in the lower than in the upper jaw; the longest are one-fifth more than the diameter of the eye. There are no teeth on the vomer or palatines. Upper pharangeal teeth only are present; they form two clusters, which appear to act in apposition, the teeth of each group being directed towards each other to form a grasping apparatus. The teeth are similar to those in the jaws, but shorter and stouter, their combined number being 14. There are no teeth on the lower pharangeals.

The chin forms the anterior contour of the head, projecting far beyond the mouth when it is closed. There is an extensive frenum behind the teeth in both jaws.

The branchiostegals are 6 in number on each side; they do not bear teeth, as stated by Clarke, who possibly wished to express the character of the branchial arches. The gill-opening is small, and placed below the base of the pectoral fin. The gills may perhaps be denoted by the formula applied to Himantolophus—namely, ½2½ pairs—but a more detailed account of their character will be advisable,

The outer branchial is free only in its posterior half, the anterior portion being adnate to the ceratohyal. This attached portion only bears gills: they are much smaller than those of the other arches, on which they are of considerable length. There is no trace of paired arrangement in the gills of this outer arch. A paired disposition is apparent in the gills of the two middle arches, for, though the rays are set in continuous series, they are of heteracanth nature. The inner arch is wholly adnate to the membranes at the lower part of the tongue, and is fully furnished with gill-rays. The gill-rakers are spiny tubercles; there are 12 on the first arch, one of which is on the upper limb, just above the angle; the rakers on the median arches are in two rows, arranged alternately, there being 19 on the second arch.

Fins.—Some idea of the character of the dorsal tentacle will be derived from Clarke's figures, but as it was evidently imperfect. and is even more complicated than drawn and described, the following description will not be out of place:—

The tentacle lies in a deep groove between the supraorbital ridges, its bulbous base being rather nearer to the mouth than is the eye; the shaft is very stout, and it terminates above in a large semispherical bulb, its total length from base to summit being 1.7 in the length of the head. From a cup in the summit of the bulb arises a freely movable stout tentacle, which divides at a short distance above its insertion, each branch throwing off 2 smaller twigs at about half its height. Inserted in the bulb and behind the cup are 2 thick branches, which, however, arise from a common base: they become flattened distally, and each, after throwing off a twig from its inner side, divides into 3 arms; these are again subdivided, but the divisions are not the same in the two branches The illustration accurately depicts the condition. Also, on the hinder part of the bulb, but nearer its base and sides, are two other small twigs. The word “frond” would perhaps be more appropriate, for the whole tentacle may be likened to a plant of Fucus, the so-named twigs being quite like the fronds of a seaweed, while the main and secondary stalks answer to the stem and branches of the plant. There are, in all, 20 terminal fronds, and the distal portion of each is nacreous white, and is no doubt luminous in life. When the tentacle is bent forwards these luminous tips dangle just in front of the mouth, and are no doubt very effective lures. It will be apparent that the tentacle was incomplete in the type specimen, the stalk arising from the middle of the cup being absent, and doubtless leading its author to conclude that the substance within the cup was luminous, though he does not actually say so.

The dorsal fin has a slightly more forward insertion than the anal, and has one more ray. The first is simple, the other four being divided nearly to their bases. The third is the longest, being 3-2 in the head. The last ray is connected to the peduncle, just free of the upper caudal ray. The anal is very similar, but the first two of its 4 rays are simple. The pectoral is short and rounded, and is placed nearly midway between the end of the snout and the base of the caudal rays. The caudal is large and rounded, arising from a very compressed and short peduncle, whose depth is equal to the longest dorsal ray.

Armour.—The skin is soft and loose, warty on snout and chin, and, excepting the top of the head, cheeks, lower jaw, and all parts in front thereof, studded with round cartilaginous scutes, each of which bears in its centre a hard low thorn with roots radiating into the body of the scute.

It was the evident intention of the authors to allude to these cones in naming the species, for on the plate accompanying the description the figures are designated Oreosoma coniferum, whereas in the text the name Oreosoma atlanticum is used.

The New Zealand example exhibits characters which are not referred to in the description of the Atlantic specimen, and these will be mentioned later.' The following is a description of the fish taken at Lyall Bay:—D. VI, 30; A. III, 28; V. I, 7; P. 20; C. 13 + 4; L. lat. 90.

Length of head, 2–64; height of body, 1–3; length of caudal, 4–7 in the length; diameter of eye, 2–27; interorbital space, 2–63; and length of snout, 2–94 in the head.

Head compressed, eyes lateral, the supra- and post-orbital ridges armed with a number of denticles, of which one in the middle of the series is larger, forming a short spine. Preopercle very oblique; a ridge across the opercle. Eyes lateral. Interorbital space flat. Nostrils close together, in front of the upper anterior margin of the orbit; the anterior nostril large, directed forward. Jaws equal; mouth protractile; the cleft subvertical. Dentary produced downwards into an acute angle. The maxilla, whose length is less than the diameter of the eye, scarcely reaches the anterior margin of the orbit when the mouth is closed. Gills 4, a small orifice behind the fourth; gill-rakers moderate, bristle-like; pseudobranchiae present.

Teeth.—The teeth are extremely small and villiform in character. A narrow band exists in the lower jaw, but no teeth are to be found in the upper jaw; they are present on the vomer, but there are none on the tongue or palatines.

The upper and hinder parts of the body are compressed and normal; a pronounced median keel runs from the occipital region to the origin of the dorsal fin, lying between the swellings on which the dorsal cones are situated. The whole of the ventral portion of the body is enormously, naturally, and permanently distended, so that a section across the body is not unlike that of Lactophrys trigonus.

Fins.—The dorsal fin arises midway between the end of the snout and the base of the caudal. Its spines are short, the second and longest being little more than half the diameter of the eye. The first spine is very short, and the second and following are graduated. The longest rays occur behind the middle of the second dorsal, and are nearly as long as the eye. The anal spines are quite small, almost hidden within the folds of the posterior dilatations of the abdomen. The rays are similar to those of the dorsal, but have a somewhat more posterior hinder insertion. The ventrals are noticeably separated, and of considerable length, the slender spine being one-half longer and the first ray twice the length of the orbit. The pectoral is rounded, and its length is equal to the diameter of the eye. The feeble tail is also rounded, and the depth of the slender peduncle is less than half the eye-diameter.

Scales.—The scales are nowhere imbricate, but form a mosaic, the components varying greatly in different parts of the body. They are minute on the interorbital space, small on the cheeks and opercles, and on the upper and hinder part of the body. They are larger immediately behind the opercles and on the sides of the body, while those on the ventral surfaces are tubercular. All are 6-sided and concentrically striated. The lateral line is well marked: it originates behind the opercle and rises above the

pectoral fin to a point in advance of the first dorsal cone; it thence drops to the mid-line of the body and passes along the middle of the caudal peduncle.

Cones.—The remarkable cones which give the fish such a striking appearance are disposed as follows: The swellings on each side of the dorsal ridge above referred to support 2 pairs of small size; the hinder pair lie at the base of the dorsal spines, and are directed outwards; the pair in front of these have a more upward aspect. All the other cones exist on the ventral portion of the body; the largest form a series of 5 pairs disposed along the lateral margin, the centre one on each side being the largest, and directed straight from the body, those before and behind being divergent. A smaller cone is placed immediately in front of each ventral fin, and a similar, though larger one, on each side of the vent between these ventral and anal cones. There are 3 pairs of much smaller ones, which thus complete the vertical armament. In these latter each cone is set close to its fellow. The mosaics in the mid-ventral line form small tubercles, but quite distinct in size and character from the true cones, which, as will be seen, number 12 pairs—namely, 2 dorsal, 5 lateral, 2 subventral, and 3 ventral. The cones, which arise from an enlarged series of mosaics, are as high as, or higher than, their diameter, and are sculptured with both radiating and transverse striae, the former being straight and the latter wavy. The area between each radial is flat. These correspond in number with the basal mosaics, of which there are 16 surrounding the largest cone —namely, that in the middle of the lateral series.

Colours.—The ground-colour is brownish-yellow, and the markings form wide open reticulations, consisting of a black line merging into bluish-grey, which extends so as to nearly obscure the ground-colour. The latter remains fairly pronounced on the cheeks, the lower edge of the caudal peduncle, and an area at the base of the anal fin, due to the absence of markings on these parts. The membranes of the first dorsal fin and of the anterior ventral rays are black; the other fins are colourless.

Length, 80 mm.

One specimen only.

There is a temptation to give the Pacific fish a distinct specific name, not only on account of certain described differences in the two known individuals, but also in consideration of the widely separated habitats, the one being taken in the Atlantic and the other in the Pacific Ocean. The fish must have rather limited powers of progression, for its locomotory fins are feeble, and the general conformation of the body is opposed to even moderate progress. The original specimen was supposed to have been taken in the surface-net, and, as the New Zealand specimen was secured alive on the beach, it becomes fairly evident that we have either two very closely allied species, or, like Tetragonurus, a single species of pelagic habit, of which examples have been obtained from both Atlantic and Pacific Oceans without any intermediate occurrences.

The differences noted between the two specimens may be due to certain characters in the smaller one having been overlooked. It is unlikely, for example, that the 3 spines preceding the anal rays were absent, or that the ventrals had only “le nombre ordinaire de ⅕.” Other differences may be noted in the descriptions of the dental armature. The French authors apparently found teeth in both jaws, whereas my specimen exhibits them in the lower jaw only. There is agreement as to the presence of teeth on the vomer, but I find none on the palatines, their presence being

direct to Wellington. Mr. Hamilton was fortunate in being able to secure a specimen for me, which confirmed my supposition, and it is this larger specimen which forms the basis of the subjoined description.

I understand that the occurrence of the fishes at the Chatham Islands was of short duration only, and that, though they were quite plentiful at the period of their appearance, they are not now to be obtained.

During a subsequent visit to Sydney I was permitted to re-examine the specimen of Eurumetopos johnstonii, and compared with it a cast of the smaller of our two examples. I found them to be specifically identical.

The Tasmanian specimen exhibits the following characters:—B. VII; D. VIII, I, 20; A. III, 15.

The length of the head equals the depth of the body, and the pectoral is as long as the head.

The radial formula, as given by Morton^* in his original description, appears to have been slightly mutilated by the printer, producing a very misleading result, which in all probability accounts for the non-recognition of the affinities of the species for such a long period. The figures D, 9 1–9, were intended for D. 9, 19, or, as now more usually written, D. VIII, I, 19. The anal formula is III, 13.

The following is a description of the Chatham Island specimen: B. VII; D. VIII, I, 20; A. III, 15; V. I, 5; P. 20; C. 24+6. L. lat. 84; L. tr 18+34. Vert. 10+12=22.

Length of head, 3.0; height of body, 2.7; and length of caudal, 5.5 in the length; diameter of eye, 5.2; interorbital space, 2.7; and length of snout, 4.0 in the head.

Head rounded, compressed, naked and porous above, tumid over the nostrils; the latter are close together, the anterior being circular, while the posterior one is an oblique slit lying midway between the end of the snout and the eye; snout truncate; the interorbital is broad and convex; the eye is relatively low in the head and is somewhat overhung by an obtuse ridge. The cleft of the mouth is horizontal, and the maxilla, which has a supplemental bone, extends to below the second third of the orbit; its distal portion is rounded and its width nearly half the diameter of the eye. The opercular bones are thin and entire, and the angle of the preopercle is greatly, though roundly, produced. Gill-membranes united far forward, not attached to the isthmus; gill-rakers long, 21 in number on the first arch, of which 16 are on the lower limb; pseudobranchiae present, but ill-developed.

Teeth.—The teeth are confined' to the jaws, the rest of the mouth being edentulous; they are small, set close together, and form a single series along the whole margins of both jaws.

Fins.—The dorsal fin commences over the edge of the operculum; the fourth and fifth spines are the longest, three-fourths the diameter of the eye; the last spine is continuous with the rays, the anterior of which is the longest and twice the diameter of the eye. The anal commences beneath the eighth dorsal ray, and is similar in character to the dorsal, terminating more posteriorly, however. The pectoral is falcate, and its seventh ray is as long as the head. The ventral spine is long and slender, its length one-half more than the diameter of the eye; the length of the first ray is twice the orbital diameter; the fin lies below the pectoral. Caudal emarginate;