O’Shea—The giant octopus Haliphron atlanticus 7

identified cephalopod species within our waters, thespecies was removed from the New Zealand faunalinventory (O’Shea 1997). The present specimen pro-vides the valid evidence necessary to confirm H.atlanticus as an undoubted member of the New Zea-land fauna.

New data on the recognised bathymetric and geo-graphic distribution, anatomy, and diet of H.atlanticus is presented here. The incomplete natureof this specimen, and the lack of comparable descrip-tions for any specimen of similar size and maturity,precludes differentiating it from others attributed tothe cosmopolitan (fide Thore 1949) H. atlanticus.Unfortunately, type material of H. atlanticus is lim-ited to an arm (Kristensen & Knudsen 1983), pre-cluding meaningful comparison at present.

Fragments of the echinoids Gracilechinusmultidentatus, Spatangus sp., Dermechinushorridus, and the gorgonian Thouarella sp. wereattached to the skin of the Haliphron carcass, indi-cating that the trawl contacted the bottom, consist-ent with accounts referring to adult Haliphron asbathypelagic, residing on or close to the seafloor(Alvarino & Hunter 1981; Young 1995).

There is a suite of pelagic octopus species hav-ing a global tropical to subtropical distribution.Haliphron atlanticus is likely to be the last of themto be reported from our waters. It brings the totalnumber of octopus species reported from New Zea-land to 40, and adds a further family of octopuses,the Alloposidae, to our region. These 40 recordedspecies are distributed throughout each of the 12currently recognised families of octopus—effec-tively giving New Zealand the highest octopod fam-ily diversity known for any region.

SYSTEMATICS

Alloposidae Verrill, 1881 (= Haliphronidae sensuHochberg et al. 1992).DIAGNOSIS: Suckers uniserial proximally, biserialdistal to edge of web; web deep between all arms;body short, gelatinous, densely pigmented; mantle

STEVE O’SHEADepartment of Earth and Oceanic SciencesAuckland University of TechnologyPrivate Bag 92 006Auckland, New Zealand

Abstract Haliphron atlanticus Steenstrup, 1861 (=Alloposus mollis Verrill, 1880) is reported from NewZealand waters on the basis of a single specimen ofgiant proportions caught recently by fisheries trawloff the eastern Chatham Rise. This specimen, pos-sibly the largest known of this species and of alloctopods, proves to be the first validated record ofHaliphron from the South Pacific. Although exten-sively damaged, details of its morphology andanatomy are described.

Keywords Mollusca; Octopoda; Haliphronatlanticus; Alloposidae; New Zealand

INTRODUCTION

Since the recent revision of the New Zealand octo-pus fauna (O’Shea 1999), several additional deep-water species of octopus have been caught in NewZealand waters. One of these, Haliphron atlanticusSteenstrup, 1861, is here confirmed from both NewZealand waters and the South Pacific.

Haliphron atlanticus (under the synonym nameAlloposus mollis Verrill, 1880) was admitted into theNew Zealand fauna by Spencer & Willan (1995)based on reports by Clarke & MacLeod (1982) andImber (1992) of beaks identified from stomachcontents of long-distance foraging marine predatorsrecovered in or near New Zealand waters.Because neither record truly demonstrated theoccurrence of H. atlanticus, nor any other similarly

Z02025; Online publication date 4 March 2004Received 22 August 2002; accepted 2 May 2003

New Zealand Journal of Zoology, 2004, Vol. 31: 7–130301–4223/04/3101–0007 © The Royal Society of New Zealand 2004

The giant octopus Haliphron atlanticus (Mollusca: Octopoda)in New Zealand waters

8 New Zealand Journal of Zoology, 2004, Vol. 31

Fig

. 1,

2F

resh

(pr

efix

, po

st t

haw

) w

hole

Hal

iphr

on a

nim

al:

1, p

arti

ally

rec

onst

ruct

ed m

antl

e an

d ar

ms,

obl

ique

per

spec

tive

; 2,

Hal

iphr

on h

ead.

Scal

e ba

r 0.

5m

. Fig

. 3, 4

Arm

suc

kers

: 3,

pro

xim

al a

nd 4

, dis

tal

suck

ers,

sho

win

g un

iser

ial

and

bise

rial

arr

ange

men

t, re

spec

tivel

y. S

cale

bar

s 15

and

20

mm

,re

spec

tivel

y.

O’Shea—The giant octopus Haliphron atlanticus 9

opening wide; funnel embedded in gelatinous tissue;radula heterodont; eyes large, diameter about 40%mantle length (ML), hemispherical; entire right armIII hectocotylised, detachable, developed in pouchin front of eye; shell vestige absent; mantle-lockingapparatus distinct, well developed; funnel organ W-shaped.

The following synonymy is complete for NewZealand citations, but is elsewhere limited to sub-stantial accounts wherein specimens are both de-scribed and/or illustrated. Numerous uncriticalcitations from regional checklists, or citations basedonly on paralarvae or juveniles, are excluded.

Haliphron Steenstrup, 1861

TYPE SPECIES: Haliphron atlanticus Steenstrup,1861; by monotypy (Hochberg et al. 1992).Haliphron atlanticus Steenstrup, 1861: 332.Kristensen & Knudsen 1983: 221; Willassen 1986:35–40, fig. 1–4; O’Shea 1997: 265; O’Shea 1999:6; Norman, Hochberg & Lu 1997: 375, 376, fig. 11e.Alloposus mollis Verrill, 1880: 394. Joubin 1895(1995 English translation): 15–18, pl. 5, fig. 1, 3, 10–11, pl. 6; Robson 1932: 215–217; Thore 1949: 67–72, fig. 61–69; Voss 1956: 169–170, fig.18a; Clarke1980: 297–299, fig. 233; Alvarino & Hunter 1981:26–32, fig. 1–4; Clarke & MacLeod 1982: 35;Hochberg, Nixon & Toll 1992: 227–228, fig. 244–245.Alloposus pacificus Ijima, 1902: 87. Sasaki 1929:17–19, pl. 8, fig. 6–8, text fig. 5; Robson 1932: 218–219.?Alloposus hardyi Robson, 1930: 397–400, text fig.17, 18, pl. 4, fig. 1; Robson 1932: 217–218.?Alloposina albatrossi Robson, 1932: 220.MATERIAL EXAMINED: NZOI Stn Z10911: F, ML0.69 m, TL 2.90 m, weight 61.0 kg (pre-fix, wetmeasured defrosted; specimen incomplete),44°33.37–33.17¢S, 175°45.33–44.21¢W, 922–920 m,0704 h, 26 Oct 2001, bottom temperature 6.0°C,R.V. Tangaroa Stn TAN0117/03.RECOGNISED DISTRIBUTION: New Zealand, ChathamRise; otherwise tropical to subtropical, cosmopoli-tan, and nowhere common; surface to 3180 m. Ju-veniles are pelagic in surface waters; they descendthrough the water column with increasing maturity;adults are presumed to be bathypelagic.DESCRIPTION: Mature female large, incomplete, withestimated intact total length 4 m, weight 75 kg (Fig.1). Mantle extensively gelatinous; skin smooth (Fig.2). Eyes laterally oriented, and probably deeply

recessed into head tissues; head not clearly differ-entiated from mantle. Mantle aperture wide; funnellarge, almost entirely fused to ventral surface ofmantle, head and arm bases 4; funnel organ lost(freezing artefact); gills with 10 lamellae per innerand outer demibranch.

Arm fragments thick, tube-like, gelatinous. Webextensively damaged, thick, with sector remnantspresent to at least level of 11th proximal sucker. First8 or 9 suckers uniserial, thereafter biserial (Fig. 3,4). Suckers widely spaced proximally, more closelyspaced distally, superficially embedded in arm tis-sues; first 2 suckers marginally smaller than thosethat follow, the 4th or 5th attaining greatest diam-eter (16.2 mm), then similarly sized to the 9th sucker,thereafter gradually decreasing in diameter to thedistal-most arm fragments; approximately 100 suck-ers per most complete arm fragment, with one-thirdto one-half of that arm missing. Suckers (Fig. 5) withbase wider than aperture; suction aperture and cham-ber large; suction pad and wall ring muscular; grasp-ing ring poorly developed.

Reproductive system (Fig. 6–9) with large ovarysac distended with eggs. Proximal oviduct narrowbasally, expanding distally, with extensively ridgedglandular lumen. Proximal and distal oviducts ofcomparable length (Fig. 6, 7); distal oviduct ex-panded, stout and thick walled, appearing com-pressed or collapsed, with extensively ridged,glandular lumen; genital pore terminal. Oviducal ball(Fig. 8) two chambered, with proximal chamberappreciably smaller than distal chamber. Eggs ab-sent from proximal and distal oviducts, and ovidu-cal ball; ovarian eggs (Fig. 9) teardrop shaped, to16.0 ¥ 5.0 mm greatest dimension, excluding stalk.

Buccal bulb destroyed; beaks (slightly damaged)recovered from trawl; radular ribbon, anteriormostoesophagus and salivary glands lost, limiting de-scription to visceral components of alimentary ca-nal (Fig. 10) and beaks only. Anterior oesophagusnarrow, thick walled, opening centrally into capa-cious crop; crop with corresponding substantial an-terior diverticulum, thin walled, with extensivelyridged lumen; posterior salivary glands paired, ovoidand heart-shaped. Stomach weakly divided intomuscular proximal section and thin-walled distalsection; lumen continuous. Spiral caecum thinwalled, forming single incomplete whorl. Intestinelonger than combined length of crop and residualoesophagus, proximally thick walled, centrally thinwalled and dilatated, distally constricted proximal toanus. Hepatic ducts paired; digestive gland ofmetallic mustard colour, large, widest proximal to

10 New Zealand Journal of Zoology, 2004, Vol. 31

Fig. 5–9 5, Cross section through largest dorsal arm sucker, scale bar 5 mm; 6–9, female reproductive system: 6, entiresystem, ventral view; 7, oviducts, dorsal view; 8, oviducal ball; 9, ovarian eggs, scale bars: Fig. 6, 7, 10 cm; 8, 3 cm;9, 10 mm. Fig. 10–14 Alimentary canal: 10, entire system, excluding buccal bulb and anteriormost oesophagus (lostat capture); 11, 12, upper, and 13, 14, lower beak. Scale bars: Fig. 10, 10 cm; Fig. 11–14, 3 cm.

O’Shea—The giant octopus Haliphron atlanticus 11

hepatic ducts, constricted proximal to ink duct, andwithout apparent pancreas. Exposed portion of inksac short, c. 25% digestive gland greatest length, intowhich it is imbedded; ink duct long, opening directlyinto anus.

Crop full of non-cuticular red-coloured prey, mostlikely cnidarian (cf. Atolla sp.). Also several tinystones (~2 mm greatest dimension), and 2 tinyamphipods (~4 mm dimension), probably ingestedaccidentally.

Beaks (Fig. 11–14) dark pigmented with narrowtranslucent hood and lateral wall margins. Upperbeak (Fig. 11, 12) squat; hood large, projectingposteriorly, extending high above the lateral wallcrest; hood and lateral wall margins translucent; jawirregularly micro-serrate, but with serrations possi-bly attributable to opposing beak wear; rostral tipacutely pointed. Lateral walls with rounded profileand crest, and very shallow wall fold, extending fromthe jaw to the lateral wall base. Lower beak (Fig. 13,14) squat (height 75% width and 83% base length);hood broad, long, with rounded crest and very shal-low posterior notch; rostrum well developed, dam-aged. Jaw irregularly serrate but possibly damaged;lateral wall with pronounced fold extendingposteriorly to lower edge of lateral wall. Upper beakmeasures (mm): rostrum length 5.6, hood length42.3, crest length 49.3, shoulder length 26.5, lateralwall height 38.5; and lower beak measures (thosemarked by an asterisk are damaged; reconstructedfigures in parentheses): rostrum length 5.0*(~5.5–6.0), hood length ~26.0, crest length 41.0, shoulderlength 9.0, wing length 34.0, beak height 30.5 andbeak base length 55.0.

DISCUSSION

It is often stated (Thore 1949; Nesis 1987; Hochberget al. 1992; Young 1995) that Haliphron juvenileslive in shallow water and adults at depth. If correct,it is remarkable, given the preponderance of juve-niles elsewhere, that Haliphron juveniles have notbeen captured in New Zealand waters. The numberof fine-meshed-net research trawls that have beenundertaken here, which have successfully caughtnumerous other comparatively shallow-dwellingjuveniles of other pelagic genera (e.g., Amphitretus,Ocythoe, Tremoctopus, and Argonauta), might beexpected to have produced juvenile Haliphron ifthey were common. Moreover, given that the south-ern slopes of Chatham Rise have been extensivelybottom- and near-bottom trawled to depths of

1200 m, it is remarkable that an apparently widelydistributed, slow-moving and large animal such asadult Haliphron has until now escaped capture. It haslong been held that nets are inefficient at capturingcephalopods in general, and that Haliphron is capa-ble of out-manoeuvring them (Alvarino & Hunter1981). It is possible that both juveniles and adultsare absent from our records because New Zealandwaters are at the limits of their distribution.

This female was mature, but not breeding. Theovary sac was completely distended with large eggs(to 16.0 mm length, excluding stalk), but eggs wereabsent from both the proximal and distal oviduct andoviducal gland, and no detached male hectocotyluswas recovered from the mantle cavity. Brooding isreported for Haliphron, with the egg mass, albeitdifficult to see, described around the buccal regionof a female estimated to be of 1 m total length; theeggs are further described as “considerably smallerthan the suckers and grouped into numerous grape-like clusters that formed a single mass of eggs”(Young 1995). The largest suckers and the ovarianeggs from the New Zealand specimen are of com-parable size, and had the specimen been intact itwould have measured ~4 m total length and pos-sessed suckers considerably larger than those ofYoung’s specimen. If Young’s smaller female wasfully mature, 1 m in length, brooding, and the eggswere correctly (relatively) sized, then his specimendiffers considerably from the New Zealand female,and suggests that there could be more than one spe-cies in this genus. Several species of Haliphron havebeen proposed, but the material upon which theywere based was limited, and many of the characterstates cited as diagnostic for individual species havesince proven to be of dubious systematic value. Athorough review would require access to far morecomprehensive bathymetric, geographic, and on-togenetic collections of specimens of both sexes thancurrently exists.

Alloposus hardyi Robson, the small male type(ML 40 mm) and only known specimen, has 6 or 7lamellae per gill demibranch, similar to the 6lamellae per demibranch described for Alloposinaalbatrossi Robson, 1932 (considered to be a syno-nym of A. mollis (= H. atlanticus)); a comparablysized (42 mm ML) male attributed to A. mollis byVoss (1956, p. 169), however, has 9 lamellae per gilldemibranch. A large female (ML 115 mm) capturedsouth of Newport, Rhode Island (Young 1972), pos-sesses 18 lamellae per whole gill (probably 9 lamel-lae in the outer demibranch), as does a near-fullymature female from off Norway (Willassen 1986).

12 New Zealand Journal of Zoology, 2004, Vol. 31

The small female type of A. pacificus Ijima, 1902,the female attributed to H. atlanticus from off theTanimbar Islands (ML 73.2 mm, total length300 mm, fixed weight 165 g) (Norman et al. 1997),and the New Zealand specimen all have 10 lamel-lae per outer demibranch of the gill. Gill lamellaecounts may prove useful for preliminary differentia-tion of species.

The reproductive system of this female is unusualamongst pelagic octopuses in that both the proximaland distal oviducts are of comparable length—a con-dition similar to that described for the extensivelygelatinous Amphitretus, but in contrast to that de-scribed for other pelagic genera, e.g., Argonauta,Ocythoe and to a lesser extent Tremoctopus, whereinthe distal oviducts are very long relative to the proxi-mal pair (O’Shea 1999). Thore (1949, p. 71, fig. 67)illustrated the reproductive system of an immaturefemale of total length 400 mm, wherein the distaloviducts are marginally longer than the proximalpair, the proximal chamber of the oviducal ball islarger than that of the distal, and a pronounced pa-pilla is apparent in a position consistent with thegenital pore. These character states differ in the NewZealand specimen, perhaps because of the relativeimmaturity of Thore’s smaller female.

Beaks attributed to H. atlanticus (as A. mollis)retrieved from sperm whales caught off Durban andDonkergat (South Africa) were shorter in all meas-urements than the present specimen. Their lowerhood lengths were up to ~2.2 cm, and the weightattributed to one (with a lower beak rostrum length(LRL) of 0.46 cm and crest length 3.71 cm) waswrongly cited as 991 g post preservation (Clarke1980). Two specimens with dorsal mantle lengths of0.45 m and 0.3 m reported from off the coast ofNorway weighed 41 kg (post fixation) and 25 kg(frozen), respectively (Willassen 1986). The lowerhood length of the beak from the 41 kg individualwas 2.33 cm; the LRL was not given, but the lowercrest length was 4.22 cm, again all shorter than theNew Zealand female.

The Haliphron female described here is possiblythe largest of all known octopuses. As its lower beakrostrum length is marginally greater than that re-corded by Clarke (1986) for the same species, ameasure including the delicate membranous marginsnot present (digested) on Clarke’s beaks, it is possi-ble that H. atlanticus attains slightly larger propor-tions than hereby reported.

ACKNOWLEDGMENTS

I thank Paul Grimes, Colin Sutton, Di Tracey (all NIWA),and the skipper and crew of R.V. Tangaroa for retainingthis unique specimen, Niel Bruce for his suggestions onan earlier version of this manuscript, Chris Thomas forhis photographic expertise (both NIWA), two reviewersfor their valued editorial suggestions, the Foundation forResearch, Science and Technology (grant CO1X0026)that enabled this work to proceed, and funding fromDiscovery Communications Inc. (Discovery Channel,USA) to complete it.

REFERENCES

Alvarino, A.; Hunter, J. R. 1981: New records of Alloposusmollis Verrill (Cephalopoda, Octopoda) from thePacific Ocean. The Nautilus 95(1): 26–32.

Clarke, M. R. 1980: Cephalopoda in the diet of spermwhales of the Southern Hemisphere and their bear-ing on sperm whale biology. Discovery Reports37: 1–324.

Clarke, M. R. ed. 1986: A handbook for the identificationof cephalopod beaks. Oxford, Clarendon Press.Pp. 1–273.

Clarke, M. R.; MacLeod, N. 1982: Cephalopod remainsfrom the stomachs of sperm whales caught in theTasman Sea. Memoirs of the National Museum ofVictoria 43: 25–42.

Hochberg, F. G.; Nixon, M.; Toll, R. B. 1992: Octopoda,in “larval” and juvenile cephalopods: A manualfor their identification. In: Sweeney, M. J.; Roper,C. F. E.; Mangold, K. M.; Clarke, M. R.; Boletzky,S. ed. Smithsonian Contributions to Zoology 513:1–281.

Ijema, I.; Ikeda, S. 1902: Notes on a specimen ofAmphitretus obtained in the Sagami Sea.Annotationes Zoologicae Japonenses 4(3):85–101.

Imber, M. J. 1992: Cephalopods eaten by wanderingalbatrosses (Diomedia exulans L.) breeding at sixcircumpolar localities. Journal of the Royal Soci-ety of New Zealand 23: 243–263.

Joubin, L. 1895: Cephalopods from the scientific expedi-tions of Prince Albert I of Monaco: Contributionto the study of cephalopods of the North Atlantic.English translation (1995) of: Contribution àl’étude des céphalopodes de l’Atlantique Nord.Résultats des Campagnes Scientifiques Prince deMonaco 9: 1–63. Smithsonian InstitutionLibraries and The National Science Foundation,Washington DC: 1 (1): 1–48.

O’Shea—The giant octopus Haliphron atlanticus 13

Kristensen, T. K.; Knudsen, J. 1983: A catalogue of thetype specimens of Cephalopoda (Mollusca) in theZoological Museum, University of Copenhagen.Steenstrupia 9: 217–227.

Nesis, K. N. 1987: Cephalopods of the world (Englishtranslation). Neptune City, Tropical Fish Hobbiest(TFH) publications. Pp. 1–352.

Norman, M. D.; Hochberg, F. G.; Lu, C. C. 1997: MolluscaCephalopoda: mid-depth octopuses (200–1000 m)of the Banda and Arafura Seas (Octopodidae andAlloposidae). Memoirs du Muséum Nationald’Histoire Naturelle 16: 357–383.

O’Shea, S. 1997: Status of three Octopoda recorded fromNew Zealand, based on beaks recovered fromlong-distance foraging marine predators. New Zea-land Journal of Zoology 24: 265–266.

O’Shea, S. 1999: The marine fauna of New Zealand:Octopoda (Mollusca: Cephalopoda). NIWABiodiversity Memoir 112: 1–280.

Robson, G. C. 1930: Cephalopoda, I. Octopoda. Discov-ery Reports II: 371–402.

Robson, G. C. 1932: A monograph of the RecentCephalopoda. Part 2. The Octopoda (excludingthe Octopodinae). British Museum (Natural His-tory), London. Pp. 1–359.

Sasaki, M. 1929: A monograph of the dibranchiatecephalopods of the Japanese and adjacent waters.Journal of the College of Agriculture, HokkaidoImperial University 20 (Suppl.): 1–356.

Spencer, H. G.; Willan, R. C. 1995: Index to the fauna: 3.Mollusca. New Zealand Oceanographic InstituteMemoir 105: 1–125.

Steenstrup, J. 1861: Overblik over de I KjøbenhavnsMuseer opbvarede Blæksprutter fra det aabne Hav.Oversigt over det Kongelige Danske Videns-kabselskaps Forhandlinger 1860–61. Pp.69–86.

Thore, S. 1949: Investigations of the “Dana” Octopoda,Part 1. Bolitaenidae, Amphitretidae, Vitrele-donellidae, and Alloposidae. Dana Report 33:1–85.

Verrill, A. E. 1880: Notice on the remarkable marinefauna occupying the outer banks off the southerncoast of New England. American Journal of Sci-ence, Series 3, 20: 390–403.

Verrill, A. E. 1881: The cephalopods of the North-easternCoast of America, Part 2: the smaller cephalopods,including the “squids” and the octopi, with otherallied forms. Transactions of the Conneticut Acad-emy of Sciences 5(6): 259–446.

Voss, G. L. 1956: A review of the cephalopods of theGulf of Mexico. Bulletin of Marine Science of theGulf and Caribbean 6(2): 85–178.

Willassen, E. 1986: Haliphron atlanticus Steenstrup(Cephalopoda: Octopoda) from the coast of Nor-way. Sarsia 71: 35–40.

Young, R. E. 1972: The systematics and areal distribu-tion of pelagic cephalopods from the seas offsouthern California. Smithsonian Contributionsto Zoology 97: 1–159.

Young, R. E. 1995: Aspects of the natural history ofpelagic cephalopods of the Hawaiian mesopelagic-boundary region. Pacific Science 49(2):143–155.