Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 20, 1887
This text is also available in PDF
(892 KB) Opens in new window
– 20 –

Art. II.—On a Specimen of Regalecus recently Stranded in Otago Harbour.

[Read before the Otago Institute, 12th July, 1887.]

Plate V.

About four years ago I communicated to the Institute (7)* a description of a fine specimen of the Great Ribbon-Fish which had been cast ashore at Moeraki, and purchased for the Otago University Museum. The skeleton was prepared, and a detailed account of it published in a subsequent paper (8). After being for some time in the Museum, it was sent to the Colonial and Indian Exhibition of 1886 as part of a collection illustrating New Zealand zoology. This specimen, which is interesting as being apparently the first complete skeleton of Regalecus on record, is now in the British Museum (Natural History), South Kensington.

The fish which forms the subject of the present communication was cast ashore in Otago Harbour, about 1 ½ miles north of the village of Portobello, and 10 miles north of Dunedin, on the 3rd of June last. It was found by a settler, Mr. Harwood, who very generously presented it to the Museum, and even took the trouble to drive into Dunedin on purpose to inform me of the capture. But for his prompt action the fish would certainly have been considerably damaged before it could have been brought to the Museum, and might have been hopelessly ruined.

The specimen, which was 11 feet long, was specially interesting from the fact that the characteristic crest or nuchal fin was practically perfect, instead of being, as in the vast majority of examples which have come under the notice of naturalists, so damaged as to make its precise characters very doubtful.

As I was anxious to secure both a stuffed specimen and a skeleton, I had the skin removed, with the exception of that of the head, which was too thin to allow of its being separated from the underlying bone. A cast of the head was taken in plaster of Paris, and was attached to a wooden model of the body over which the skin was stretched, the whole being afterwards silvered and painted from tracings taken of the fresh fish. The fins were “made up,” as the rays were required for the skeleton. In this way a specimen has been obtained which

[Footnote] * The figures in thick type refer to the bibliographical list at the end of the paper.

– 21 –

gives an excellent notion of the form, colour, markings, etc., of the fish, and in addition an almost perfect skeleton.

Since the account of the Moeraki specimen was published, I have been favoured by the authors, Dr. Chr. Lütken, of Copenhagen, and Dr. Robert Collett, of Christiania, with copies of three important papers on northern species of Regalecus. Lütken's second paper (6), which is a résumé of his first (5), is accompanied by a French translation, and is also translated into English, in the “Annals and Magazine of Natural History “for 1883. Collett's paper (1) is in Norwegian; and I should have been able to make out little beyond the description of the plates but for the kindness of Mr. C. Theilmann, who was good enough to translate both it and the necessary portions of Lütken's first paper for me.

It is perhaps not unworthy of mention that the majority of specimens found in the Northern Hemisphere have been obtained in winter or early spring. Out of 25 examples recorded by Collett (1) as having been obtained either in Scandinavia or in Great Britain during the past century, 8 were found in March, 4 in February, 3 in January, and 3 in April. Of 6 New Zealand specimens, of which the date of capture is recorded, 3 were found in winter (2 in June and 1 in July), 1 in late autumn (May), and 1 in early spring (October).* These facts are not without interest as bearing upon the case of another deep-sea Teleost, the Frost-fish (Lepidopus caudatus), which is hardly ever obtained except by being found stranded on sea-beaches during'the winter. It is also remarkable that all the Regaleci the sex of which has been ascertained have been females.

It will be convenient to discuss the present specimen under the following heads:—


Size, proportions, and number of fin-rays.


Evidence of mutilation of tail.


Colour and markings.


Characters of the crest or nuchal fin.


The skeleton.

(a.) Size, proportions, and number of fin-rays.—The chief facts under this head are best given in the form of a table, which will serve to show at a glance the main differences between the three New Zealand specimens of Regalecus which have been

[Footnote] * In mentioning the recorded occurrences of Regalecus in New Zealand in my former paper (7), I omitted two: A specimen 14 feet long is mentioned by Mr. F. E.Clarke (“Trans. N.Z. Inst.,” vol. xiii., p. 196) as having been found near Jackson's Bay by Mr. James Teer, in February, 1874; and Sir James Hector records the capture of a “species of Banks' Oar-fish, Regalecus gladius “[sic] at Cape Farewell Sandspit (” Trans. N.Z. Inst.,” vol. x., p. 533). The date of capture of this last example is not given: it seems to have been shortly before December, 1877.

– 22 –

carefully examined—viz., von Haast's New Brighton specimen, and the two which have come under my notice:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

New Brighton, 1876. Moeraki, 1883. Otago Harbour, 1887.
Ft. in. Centim Ft. in. Centm Ft. in. Centim
Total length 12 5 378.75 12 6 381.25 11 0 335.5
Greatest height of body 1 1.75 35.0 1 3.25 38.7 1 0 30.5
Length of head (jaws retracted) 0 7.75 19.5 0 9 22.8 0 9.5 24.0
Distance between snou and anus 4 11 150.0 5 6 167.75 4 9.5 146.0
Proportion of height to length 1:11 1:10 1:11
Proportion of length of head to total length 1:19 1:17 1:14
Proportion of length of head to total pre-nal region (=head + trunk) to total lentgth 1:25 1:2.27 1:2.29
Total number of dorsal fin-rays 232 205 189
(9+223) (14+191) (14+175)

It will be seen that the only differences of importance between the three specimens are those connected with (a) the size of the head, which is shortest in the New Brighton, longest in the Otago Harbour specimen; and (b) the number of fin-rays, which are most numerous in the New Brighton, fewest in the Otago Harbour specimen.

Similar differences exist between the examples found in the Northern Hemisphere. In eight of these, tabulated by Lütken (5, p. 26), the proportion between the greatest height of the body and the total length varies from 1: 9 to 1:15; the proportion between the length of the head and the total length, from 1: 14 to 1: 21; the proportion between the pre-anal region and the total length, from 1: 1.7 to 1: 3.2; and the number of dorsal fin-rays from 174 to 406.

(b.) Evidence of mutilation of tail. The tail (Pl. V., fig. 4) had evidently been broken off obliquely, probably at no very distant period, since the broken surface of the last vertebra was visible externally.

The fracture had taken place across the middle of the 89th vertebra, the remaining or anterior half of which agreed in every respect with the corresponding portion of the preceding vertebræ, and was quite different from the peculiar demi-vertebra which terminated the vertebral column of the Moeraki specimen (8, Pl. VI., figs. 25 and 26). It is, of course possible that in the process of healing the broken bone might assume the form

– 23 –

of the demi-vertebra, but the differences between the two make me disposed to doubt it. As to the form of the tail itself, it is truncated obliquely, not bluntly-pointed with a ventral emargi-nation, as in my first specimen and in von Haast's.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

All the specimens examined by Collett had the tail broken in a very similar way, the last vertebra showing a fresh-fractured surface. Collett considers that the missing piece of the tail is usually of small size, his conclusion being founded upon the fairly constant position of the anus, which is, as a rule, a little cephalad of the middle of the body. He states that the usual proportion is for the pre-anal region, or head and trunk, to be 4/10ths, the post-anal portion, or tail, 6/10ths of the total length; or, in other words, that the proportion between the length of the pre-anal region and the total length is 1: 2.5. It will be seen that, in this respect, the resemblance between the Northern and Southern Regaleci is very close.

Lütken considers that the end of the tail, with the tail-fin, is lost at an early age, and that regeneration then takes place, producing an additional piece of varying length. In this way he accounts for the great length (18 feet, or 564 centim.) and unusual number (406) of fin-rays of Lindroth's Hitteren specimen (R.grillii). He also remarks upon the fact that the examples recorded from the Mediterranean have all had uninjured tails, with small tail-fins, as shown in Cuvier's figure of R. gladius (2)

(c.)Colour and markings.—The general colour was, as in former specimens, that of pure frosted silver. The irregular sub-vertical black stripes and spots in the anterior half of the body were also of the usual character, but more distinct than in the Moeraki specimen when quite fresh (see 7, p. 293). But in addition to these the whole body was covered with oval or circular dull grey spots, formed of aggregations of chromatophores, covered and thus toned down by a thin silvery coating. These spots were from 1.5 to 0.5 inch (4—1.5 cm.) in diameter: those along a line equidistant from dorsal and ventral edges were longitudinally oval, the rest circular; each was slightly lighter in tint in the centre than at the circumference. Von Haast mentions them as “dark rings “(3, p. 248), but there is no indication of them in his figure. Their appearance was precisely that shown in Cuvier's figure of R. gladius (2, Pl. 69), in which, however, the anterior black bars are absent. A thoroughly good notion of the appearance of the present specimen would be obtained by painting on the anterior half of Cuvier's figure the black bands shown in Hancock and Embleton's drawing of R. banksii (4), or in von Haast's of R. pacificus (3). The distinction in the fresh fish between the very obscure spots, hardly visible in certain lights, and the intensely black bands on the front half of the body was very striking.

– 24 –

According to Collett, the body in the Northern forms is also marked with 4–6 brownish black longitudinal bands. I am disposed to think he must refer to the elevated longitudinal ridges present in most of the accurately described specimens. There was no special development of pigment on them in either of my examples, but they would assume the appearance of dark bands in a dried specimen.

As in the Moeraki specimen, the raised tubercles with which the body is beset are composed of thick fibrous tissue. This was made very obvious when the skin was allowed to dry; instead of standing out even more prominently than in life, as they would have done if made of bone, they almost disappeared, and are barely visible in a thoroughly dried piece of skin.

(d.) Characters of the crest.—The precise characters of the crest, nuchal fin, or first dorsal fin of Regalecus, seem always to have been doubtful. In my former paper (7) I gave a résumé of all previous descriptions which had come under my notice, as well as outline sketches of the more important published figures, of which it will be seen no two are alike. Lütken gives a figure of a specimen found at the Faröe Islands (5, p. 20), in which the crest is shown to consist of two distinct nuchal fins, the anterior rather less, the posterior a little more than thrice the height of the head, and the rays of both terminating in simple points. Collett (1, Pl. II.) figures the crests of two specimens, one from Nordfjord, the other from Stavanger; in both, the rays are broken off short, and the membrane between them is lost. These are the only additional figures I have met with since the publication of my former paper. In the present specimen, as already stated, the crest was nearly perfect, the only broken rays being the seventh and ninth. The membrane of the fin was very little damaged, and by floating the whole crest out in a dish of water, its characters could be perfectly well ascertained.

I find that in all essential respects the crest of the Otago Harbour specimen (fig. 1) agrees with that of Cuvier's figure of R. gladius* in the illustrated edition of the “Règne Animal.” It is distinctly divisible into two portions or “nuchal fins,” an anterior consisting of five, and a posterior of nine rays; so that the total number of rays in the crest is fourteen. In my former paper I gave the number conjecturally as fifteen, stating that what I took to be the last six rays were broken; judging from the present specimen, it must have been the last five rays of the crest and the first of the second dorsal which were damaged. Cuvier's figure shows five rays in the anterior, seven in the posterior division.

[Footnote] * Reproduced in 7, Plate xxiv.

Picture icon

Regalecus Argenteus

– 25 –

The length of the rays is as follows:—

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Crest, or “First Dorsal.”
In. Cm. Condition.
First Nuchal Ray I. 22 56 Perfect.
" II. 20.5 52 "
" III. 18 46 "
" IV. 14.5 37 "
" V. 13 33 "
Second Nuchal " VI. 29.5 75 "
" VII. 21 53 Distal end wanting.
" VIII. 20 51 Perfect.
" IX. 1 2.5 Broken off shoort.
" X. 12.5 32 Perfect.
" XI. 9 23 "
" XII. 7.1 18 "
" XIII. 3.2 8 "
" XIV. 1.4 3.5 "
"Second Dorsal."
Ray XV. 1.4 3.5 "

As in most other carefully-described specimens, including the two figured by Collett, the first ray is stout at its proximal end (4 mm. in diameter), the next four—i.e., the remaining rays of the first nuchal—extremely slender (about 1.75 mm.), and those of the second nuchal stout, their thickness diminishing, however, pari passu with their length, so that the last is of the same thickness as the rays of the second dorsal. The reason of this variation in thickness is apparent, when it is seen, as described in the following paragraphs, that the rays of the first nuchal are united to one another for about their proximal half by membrane, while those of the second nuchal are free, except at the base.

The first ray is united by membrane to the second for at least its proximal 34 cm., probably a little more—its distal portion being fringed by a delicate wavy membrane, which terminates in a simple point, and is continued into a very narrow band, edging the anterior face of the distal end of the ray (fig. 2). The four following rays are similarly joined, the vertical height of the uniting membrane diminishing progressively in successive interspaces, from 34 cm., between the 1st and 2nd rays, to 21 cm. between the 4th and 5th.

The remaining rays—those of the second nuchal—are united only at the base: how far is uncertain, the membrane being torn, but probably between 2 and 4 cm. These rays are all fringed posteriorly by a wavy membrane, which terminates distally in a thickened lanceolate lobe (fig. 3), as described in my former papers. These lobes were present on all the rays of the second nuchal, except the 7th and 9th, the former of which was damaged at the tip only, the latter broken off short. Curiously enough, the 7th and 9th were the only two perfect rays in the Moeraki specimen. The presence of a small but

– 26 –

distinct lobe (about 8 mm. long) on the 14th ray, and the absence of anything of the kind on the 15th, allows of a clear distinction being drawn between the crest or “first dorsal” fin and the “second dorsal,” although the adjacent rays being united by membrane, there is, strictly, only one continuous fin.

A comparison of Pl. V., fig. 1, with Cuvier's figure (7, Plate xxiv., fig. 6) shows that the only difference of importance between the two is that the second nuchal in the latter has seven rays instead of nine, and that the terminal lobes are represented as much larger than in my specimen.

In the first nuchal the pink membrane uniting the rays was marked with small circular spots of a deep crimson colour, (about 2–3 mm. in diameter,) and arranged in a single row in each membranous interspace. Examined microscopically, these were found to be produced by aggregations of well-marked sub-circular chromatophores, having interspersed among them branched black pigment-cells.

In the second nuchal each ray with its membrane was marked with nearly equidistant crimson blotches, about 8 mm. in diameter, generally occurring in the wider parts of the wavy membrane. These also contained both red and black chromatophores; the former faded considerably after two or three days, the spots then appearing of a dull grey colour.

These large patches of red are shown in Cuvier's figure; but the small spots of the first nuchal are not indicated either in that or in any other figure with which I am acquainted.

All the descriptions and figures of the crest of Regalecus with which I am acquainted can be accounted for on the theory that it had, when uninjured, the characters described above, with the single exception of von Haast's. He states (3, p. 248) that, in R. pacificus, the 2nd and 7th rays were perfect, and were respectively 7 and 7.75 inches long; he also says that the 3rd, 4th, and 5th rays were nearly as thick as the first, and that all were beset with minute upwardly-directed hooks on the anterior and posterior edges.

(e.) The skeleton.—This agrees in all essential respects with that of the Moeraki specimen, the only important difference being in the number and position of the ribs. In the Moeraki specimen there were ribs on the 8th–25th vertebræ inclusive, and in Imtken's Faroe Islands specimen (5 and 6) on the 8th–24th. In the present skeleton the 6th vertebra has a rib on the right side only; form the 7th to the 20th there are well-developed ribs, and rudiments on the 21st–28rd.

Both pelvic (= ventral) rays were broken off short, but attached to one of them by a shred of membrane there was, as I am informed by my assistant, Mr. Jennings, a bony rod about 6 inches long, of the same thickness as a pelvic ray, curved at its proximal end, and fringed with membrane. The fact of the

– 27 –

bone not being straight is peculiar, and its thicker or proximal end did not correspond with the broken attached end of the pelvic ray. Jennings assures me, however, that he removed it from its attachment himself; moreover, it is obviously a Regalecus bone, and agrees in appearance, size, etc., V with nothing but the pelvic rays, so that its curvature is probably teratological.

The Otago Harbour specimen of Regalecus is obviously of the same species as that from Moeraki, the only difference of importance being the faint spots of the former. Whether these are a matter of age it is impossible to say, but apparently not, as both individuals were adult or sub-adult females. I therefore assign the present specimen to the species R. argenteus, Parker, the diagnosis of which (see 7, p. 295) must be amended in the following particulars:—


The number of dorsal rays and the proportion of height of body and length of head to total length are variable, the caudal extremity being subject to mutilation.


The first fourteen rays of the dorsal fin form a crest about three times as high as the head, and divided into an anterior portion of five, and a posterior of nine rays: the former have their lower halves connected, their upper halves fringed posteriorly by membrane, and terminate in simple points; the latter are connected only at the base, fringed posteriorly, and terminate in small lanceolate cutaneous lobes.*


The body is silvery, marked on its anterior half with irregular sub-vertical stripes and spots, and having sometimes, in addition, faint grey spots over the whole surface.

It appears very probable that further researches on this interesting genus will necessitate the union of more or fewer of the species, but the information is at present insufficient to decide the question. As Lütken remarks, “il est clair que le genre est assez cosmopolite, mais on ne peut rien dire encore quant au nombre de ses especes.”


1. R. Collett.—“Om de i vort Aarhundrede ved de norske Kyster strandede exemplarer af Slægten Regalecus.” “Christiania Videnskabsselskabs Forhandlinger,” 1883, No. 16.

2. Cuvier.—“Règne Animal, Poissons,” p. 148, and pl. 69.

3. v. Haast.—“Notes on Regalecus pacificus, a new Species of Ribbonfish from the New Zealand Seas.” “Trans. N.Z. Inst.,” vol. x. (1876), p. 246.

[Footnote] * In the specific description referred to (7, p. 295) the words occur: “three or four of them [the rays of the crest] terminate in lanceolate cutaneous lobes:” the words in italics were substituted by a printer's error for “more or fewer.”

– 28 –

4. Hancock and Embleton.—“Account of a Ribbon-fish (Gymnetrus) taken off the Coast of Northumberland.” “Ann. and Mag. Nat. Hist.,” 2nd series, vol. iv., 1849, p. 1.

5. Lütken.—“Trachypterus arcticus og Gymnetrus banksii (grillii).” “Vidensk, Meddel. fra den naturh. Foren. i Kjobenhavn,” 1881, p. 190.

6. Lütken.—“Nogle Bemærkninger om Vaagmæren (Trachypterus arcticus) og Sildetusten (Gymnetrus banksii):” with French translation. Oversigt over d. k. D. Vidensk. Selsk. Forhandl., 1882, p. 206. Translation by W. S. Dallas in “Ann. and Mag. Nat. Hist.,” 5th series, vol. ii., 1883, p. 176.

7. T. J. Parker.—“On a Specimen of the Great Ribbon-fish (Regalecus argenteus, n. sp.) lately obtained at Moeraki, Otago.” “Trans. N.Z. Inst.,” vol. xvi. (1883), p. 284.*

8. T. J. Parker.—“On the Skeleton of Regalecus argenteus.” “Trans. Zool. Soc.,” vol. xii., part 1 (1886), p. 6.

Postscript.—Since writing the above I have seen Professor McCoy's description of a Ribbon-fish caught in the waters between the Victorian and Tasmanian coasts, in May, 1878. The description occurs in the 15th decade of the “Prodromus of the Zoology of Victoria,” and is illustrated by a coloured plate. The specimen is especially interesting from the condition of the tail, which tapered gradually to a vertical height of 1 cm., when it was broken off: thus, even if it tapered, when perfect, to a veritable point, it cannot well have lost more than 3 or 4 cm. In correspondence with this, we have the important fact that the number of dorsal fin-rays is 17 + 406, that is exactly the same as in Lindroth's Hitteren specimen (R. grillii), in which the number of rays is given as 406, the nuchal rays being, according to Collett, counted separately.

Unfortunately, McCoy does not give the position of the anus: the remaining chief measurements are as follows:—

ft. in. cm.
Total length 13 7 424
Length of head 0 7.5 19
Greatest height of body 0 7.25 18.5
Proportion of length of head to total length 1:22
" " height of body " " 1:23

From the analogy of this specimen one would conclude that the Otago Harbour Regalecus described above must have lost at least 6 feet (180 cm.) of its length, or, in other words, that in the uninjured condition it must have been fully 17 feet (528 cm.) long.

The markings of McCoy's specimen, as shown in the plate, are peculiar. The black bands on the anterior part of the body are more nearly vertical, and more irregular in form and size, than usual; the oval or circular spots do not extend over the

[Footnote] * Further references to the literature of the subject are given in this paper.

– 29 –

anterior part of the body, and are quite black; and five longitudinal black stripes are shown, which, however, seem to be intended for the grooves between the elevated ridges.

No detailed description of the crest is given—the figure shows a single nuchal fin, the rays terminating in simple points and regularly diminishing in length from before backwards.—T.J.P.

Dunedin, January 6th, 1888.