It is an odd quirk of the English language that there exists no accepted vernacular name for this, the most highly organised group of marine invertebrates and the fifth class of the phylum Mollusca. The circumlocution, ‘octopuses, squids, and their allies’ becomes so tiresome that the scientific name, Cephalopoda seems infinitely preferable. Originally this meant ‘head-foot’ in comparison with the ‘hatchet-foot’ and ‘stomach-foot’ of the Pelecypoda and Gastropoda respectively. In English-speaking countries the whole group has become so encompassed with prejudice that mention of them seems only to conjure up thoughts of loathsomeness and exaggeration of size. This, regardless of the fact that one member of the group, the pearly nautilus, has served as inspiration for verse, and that even early taxonomic zoologists, not always noted for romanticism or even humour, have seen fit to use Argonauta and Nautilus as generic names. (Originality then departed especially as regards squids, in an attempt to combine every Greek descriptive substantive with the suffix -teuthis). In other parts of the world, realisation of the economic value of the group has left little place for prejudice.

Man has found great difficulty in capturing specimens of the group, especially the squid and cuttlefish group. Other animals are, however, very successful. Many fish feed upon Cepalopods often in large numbers, and many of the toothed whales, large and small, and many petrels and albatrosses feed almost exclusively on squids. In many parts of the world man captures the common species of the coastal fringe for bait and for food. Small octopuses fried in batter are considered a delicacy in Italy and in this form was the writer's personal introduction to this type of sea food. The flavour is similar to crayfish, though somewhat more solid in consistency. Medium-sized octopuses are eaten raw by Italians though this appears to be an acquired taste. The large nerve fibres have proved to be an ideal subject for physiological experiments upon nervous tissue, and for this reason there has been a constant demand for fresh specimens by physiological laboratories.

Sea water has a propensity for enlarging the creatures that dwell therein, at least in conjunction with the human eye. Popular estimates of the cephalopod ultimate size seem to have been largely influenced by this peculiar physical phenomenon. However, octopuses with tentacle length of six feet do seem to occur in New Zealand waters, though none appear to have been measured by scientific observers. The largest octopus recorded by a scientist seems to be one from Alaska with a ‘radial spread of nearly page 92 twenty-eight feet’. Large squids are better authenticated. A series of very large animals were washed ashore at Lyall Bay, Wellington, in the 1880's and one of them, Architeuthis longimanus, described and measured by T. W. Kirk, had a body seven feet in length with an additional fifty feet of tentacular arm. Newspaper accounts of the strandings of large cephalopods along the New Zealand coasts usually record them as octopuses when they are probably decomposed squids, or in one case investigated, strips of old whale blubber.

As usual in all man-made systems of biological classification, the higher groupings within the Cephalopoda have become increasingly complex, as less and less agreement is reached. Living forms have been divided into two subclasses, the Tetrabranchia (ta) which possess two pairs of ctenidia and a large chambered external shell, the sole living representative being Nautilus, and the Dibranchia (ta) with a single pair of ctenidia which comprise all the other living Cephalopods. By analogy the extinct ammonites have been associated with the Tetrabranchiata although some modern workers have separated them as a third group.

The Tetrabranchiata are not represented in New Zealand waters although a few drift shells of Nautilus have been recorded and there are numbers of ammonites represented in our Mesozoic strata.

Some mention of the general morphology must be made before any understanding of the group can be achieved. The Dibranchiata alone will be considered here. It is unfortunate that out-of-date ideas of the orientation of the Cephalopod body have resulted in two different systems, one favoured by morphologists, the other by systematists. For once in such matters the systematists have been demonstrated to be right, and the animal is now envisaged as oriented in a normal swimming position with the head anterior and the funnel ventral. Such an orientation brings the Cephalopoda more into line with the rest of the Mollusca.

In all Dibranchiate Cephalopods the body may be divided into head and mantle. The head bears a pair of highly organised eyes and a circlet of eight arms surrounding the mouth. These arms bear rows of suckers on page 93 the inside surfaces. In the Decembrachiata there are in addition, an additional pair of arms known as the tentacular arms, which are divided into stalk and an expanded club. In this group the circlet of eight arms are then distinguished by the term sessile arms. The tentacular arms arise inside the eight sessile arms and are usually retractile to a certain extent within a pair of pockets. The tentacle stalks may often be bound together by a series of interlocking suckers and knobs so that both arms may be shot out as one unit. The tentacle clubs bear suckers on their inner faces. These suckers (and also the suckers of the sessile arms in the Decembrachiata) are usually strengthened by chitinous rims which often bear strong incurved sharp teeth, or the sucker and ring may be modified to form strong hooks. This combined tentacular structure then forms a most efficient mechanism for catching and holding prey. In many Cephalopods a web extends between the arms to a greater or lesser degree, forming the umbrella. In the centre of the circlet of arms the mouth is situated. Just inside the mouth a pair of horny jaws (similar in shape to a parrot's beak) tears the food into conveniently sized blocks for swallowing. A somewhat degenerate radula is present, the sole function of which seems to be to grip food passing between the beaks and ensuring its entry into the oesophagus. There are two pairs of ‘salivary glands’ the secretion of one pair at least having a toxic effect on many marine animals. The alimentary canal passes backwards to open into a crop and then a stomach. The rectum then passes forward to open into the mantle cavity near the base of the funnel (Fig. 18). The major associated gland is the so-called ‘liver’ or hepato-pancreas, which opens into the alimentary canal near the junction of crop and stomach. The hepato-pancreas is relatively immense and lies ventral to the oesophagus. The gonad is situated posteriorly to the stomach and, especially in mature females becomes terrifically enlarged displacing and distorting all other organs. The genital ducts pass forwards to open into the mantle cavity. Lying close against the ventral side of the hepato-pancreas, but not in organic connection is the ink sac, which opens by a duct either into, or very close to the anus.

The mantle is closely attached to the body dorsally, posteriorly and laterally. Ventrally it encloses the mantle cavity which is bounded anteriorly by the funnel. The funnel is usually fitted with valve-like flaps so that water cannot pass into the mantle cavity through it but only in the reverse direction. By drawing the mantle away from the funnel the mantle cavity may be placed in direct connection with the surrounding sea. By a cyclical series of movements, raising the mantle from the funnel and thus enlarging the mantle cavity so that sea water enters, and then closing the mantle tight against the funnel and forcing water through the funnel a Cephalopod can achieve a continuous circulation of water through the mantle cavity (Fig. 25 A-C). This current of water is utilized in a number of ways. Firstly by directing the funnel in a suitable direction the Cephalopod can achieve jet-propelled movement either forwards or backwards. Secondly a stream of fresh sea-water is passed over the ctenidia which lie free within the page 94 mantle cavity. Then the products of excretion and the fine powdery material from the ink sac duct can all be successfully eliminated from the mantle cavity through the funnel. In many of the squids the union of funnel and mantle flap is made more secure by an interlocking cartilaginous device known as the mantle locking apparatus (Figs. 5, 7). In a few forms this aperture is completely sealed. The mantle cavity is usually divided in two longitudinally by a thin partition, the interpallial septum.

All the Decembrachiata have a pair of fins, usually situated posteriorly. The shape, position and extent of these is a useful taxonomic character.

External sexual dimorphism is slight, but detectable in many of the Cephalopoda, though within the group a series can be established from forms such as Histioteuthis cookiana where no difference may be detected, through the majority of forms where sex is indicated by modification of one of the arms or at most of a single pair of arms, to Argonauta where the male is very small and the modification of the arms is great. This modification of the arms in the male is known as hectocotvlization and the modified portion of the arm is called a hectocotylus. The term really applies to the situation found in the Octobrachiata where its function is perhaps more easily explained, but has been extended to include the peculiar and individual cases found in the squids. It is a peculiar example of the blindness of scientists that it is only comparatively recently that hectocotylization has been observed and recognised in the group. Early figures and descriptions do not mention such structures and one is forced to the belief that, not only did early workers not see the structure because they were not looking for it, but even ignored its presence and drew what they considered to be the ideal condition in the animal before them. In Octopus and related genera one of the arms is usually shortened in the male to terminate in a slightly expanded cup, the ligula (Figs. 2-4). There is a fold of skin running along the side of the hectocotylized arm which is known as the sperm groove. In the male Cephalopod the sperm are packed tightly together into the head of a complicated spermatophore. Under certain stimuli the spermatophore ruptures and a complicated ejectory apparatus pushes the mass of sperms rather violently out of the enveloping sheaths. How the spermatophores are transported from the ejector of the male reproductive system which opens into the mantle cavity, to the hectocotylus is not known but the presence and structure of the sperm groove along the side of the hectocotylized arm would indicate that they are probably transported by this means. Fertilization is effected by the male pushing the hectocotylized arm into the mantle cavity of the female and depositing the spermatophores there. In Argonauta the whole hectocotylized arm breaks off and is left in the mantle cavity of the female. The writer has found up to six detached arms in the mantle cavity of a single female. When the early naturalists discovered this detached arm it was thought to be a parasitic Trematode and was described as such under the generic name Hectocotylus. Elucidation of the actual nature of this ‘trematode’ resulted in the discovery that hectocotylization was a general page 95 feature of the Octobrachiata. In the Decembrachiata the so-called ‘hectocotylus’ is often much more complicated than the structure described above and there is even some doubt whether the structures are homologous in the two groups. Little is known of the function of these complicated adaptations in the squids. The structure and position of the hectocotylus has been found to be a valuable systematic feature.

None of the Octobrachiata have more than mere vestiges of internal shells though some form of internal shell skeleton is found in most of the Decembrachiata. In one group of the Octobrachiata, the genus Argonauta, the female produces an external shell (Fig. 23). This is unchambered, produced by a modified pair of arms and has no organic connection with the animal which can leave the shell and return at will. In the shell the eggs are laid and it serves as a brood chamber for the developing young. Two major forms of internal ‘shell’ are found in the Decembrachiata. Cuttlefish produce an elongate-oval shell heavily reinforced with calcified material which is known as a sepion (Fig. 9). Most of the squids have a very thin feather shaped gladius or pen, which reinforces the body dorsally just under the mantle (Figs. 8, 10). Spirula (a Decembrachiate) has a small spiral, chambered shell, whose coils are separated (Fig. 15).

Chromatophores (extensible pigment cells) produce much of the colour of the living animals and physiological control of chromatophore activity results in rapid changes of general colour. Over recent years it has become apparent that Decembrachiata probably possess as a group the highest development of light production of any animal group. There is much variety in the structure of light producing organs or photophores throughout the group, from simple discharging glands to complex bull's eye lanterns and mirrored searchlights. The colour of light produced in a single animal is varied. In one species the colour of light produced by different photophores is ultramarine, sky blue, ruby-red and pearly white. A survey made by Berry in 1920 showed that of 397 described species of Decembrachiata 126 had been demonstrated to be photogenic. At least one of our New Zealand species Histioteuthis cookiana from deep water in Cook Strait must be a brilliant animal in life. Unfortunately no living specimens have as yet been captured though the species must be comparatively common. For additional information on light production in the Cephalopoda the reader is referred to surveys by Cotton and Godfrey (1940) and Berry (1920).

Literature on the New Zealand Cephalopoda is very scattered. The Octobrachiata of the world were monographed by Robson in two volumes (1929 and 1932) and reference to this work will give some idea of the difficulties of Cephalopod systematics. The body form of the octopuses is so variable and so difficult to describe that recourse has been made to a series of measurements and indices (measures of proportion) to differentiate species. Suter's (1913) account of the group in New Zealand is a compilation of previous work and presents little new information. Benham, in a series of papers in Trans. Roy Soc. N.Z. from vol. 72 to vol. 75, gives a fuller account of some of the local species of Octobrachiata. The writer has in the press a monograph of New Zealand Cephalopoda which will be published by the Dominion Museum.

Key to the New Zealand Cephalopoda

Key

^* Argonauta bottgeri and A. hians have been recorded from New Zealand waters on the basis of single specimens in each case. The distinction between these two species and juvenile A. nodosa has not been thoroughly analysed.

^* The interocular index is obtained by expressing the interocular width (i.e. the distance between the outside surfaces of the eyes) as a percentage of the mantle length (i.e. the distance from the posterior end to the line joining the centres of the eyes).

1 (28)	Cephalopods with eight arms.	Octobrachiata.
2 (7)	With uniserial suckers (i.e. suckers in single row).
3 (4)	Short optic nerve, eyes directed laterally, male without a hectocotylus. Radula with multicuspid medians and laterals. Small, planktonic, transparent.	Japatella diaphana
4 (3)	Long optic nerve.
5 (6)	Male with hectocotylus. Liver and stomach in normal position. Small, planktonic.	Eledonella pygmaea
6 (5)	Stomach anterior to the liver. Liver long, narrow and pointed. Small planktonic.	Vitreledonella richardi page 97
7 (2)	With biserial suckers.
8 (17)	Hectocotylus involving the whole arm which is autonomous.
9 (10)	One aquiferous pore on each side of funnel and pair on head. Medium sized, no shell. Range same as for Argonauta.	Tremoctopus violaceus
10(9)	No aquiferous pore, external shell in female. North of Marlborough Sounds and Chatham Islands.	G. Argonauta
11 (12)	Keel narrow, less than 12% of length of aperture. Ribes close and numerous, not broken into nodules. Rather rare in N.Z.	A. argo
12 (11)	Keel rather broad, more than 20% of aperture length.
13 (14)	Shell large, over 100mm. total length, reaching 242mm. Ribs broken into fairly coarse tubercles. The commonest species of this genus in N.Z. waters. (Figs. 23, 24.)	A. nodosa
14 (13)	Shell of medium size, up to 108mm. total length, surface finely granular.
15 (16)	Shell up to 58mm. total length, no auricles, no colouring on carinal knobs.	A. bottgeri *
16 (15)	Shell up to 108mm. total length, auriculate forms occur, carinal knobs red-brown.	A. hians *
17 (8)	Hectocotylus involving the tip of the third right arm.
18 (19)	A broad fin along either side of the body. This species was described 1832. It has not been collected since and no preserved specimens exist.	Pinnoctopus cordiformis
19 (18)	No lateral fin.
20 (25)	Hectocotylus long with transverse ridges on floor of ligula.	G. Octopus
21 (22)	Gill filaments 13-14. The common N.Z. octopus found along the coastline and down to moderate depths.	O. maorum
22 (21)	Gill filaments less than 13.
23 (24)	Gill filaments 9, animal large, up to 210mm. mantle length. Vagina of female flask-shaped. Aperture of oviduct situated on interpallial septum. Known from a single specimen taken off Moeraki.	O. zealandicus
24 (23)	Gill filaments 8, animal small, up to 46mm. mantle length. Whole of distal oviduct wide and swollen. (Fig. 21.) Aperture of oviduct in front of base of gills. Known from two specimens from Port Chalmers.	O. adamsi
25 (20)	Hectocotylus short, with no transverse ridges on floor of ligula. (Fig. 2.)	G. Robsonella
26 (27)	Mantle length up to 36mm. Second diverticulum of penis comparatively small and globular. Head comparatively wide. (Interocular Index * average 62.) The common, small shore octopus, found also in Australia.	Robsonella australis page 98
27 (26)	Mantle length up to 56mm. Second diverticulum of penis large and cylindrical. Head comparatively narrow (Interocular index average 44). Southern N.Z. and Auckland Is.	R. huttoni
28 (1)	Cephalopods with 10 arms.	Decembrachiata
29 (30)	An internal spiral shell (Fig. 15). Shells washed ashore in countless numbers on northern beaches. The animal appears to be bathypelagic in habit.	Spirula spinila
30 (29)	No internal spiral shell.
31 (40)	Eye covered with a continuous membranous lid.	Myopsida
32 (35)	Body short, rounded, mantle fused to the head dorsally. Shell a gladius, or missing. Fins marginal, not extending the whole length of body.
33 (34)	Shell lanceolate, left dorsal arm hectocotylized.	G. Sepiola
34 (33)	No shell, left ventral arm hectocotylised. Fig. 13.) Lives in schools in coastal waters.	Sepioloidea pacifica
35 (32)	Body oval or elongate, shell a gladius or ‘cuttlebone’, fins marginal, extending whole length of body.
36 (37)	Internal shell a gladius or ‘pen’ (Figs. 10, 11).	Sepioteuthis bilineata
37 (36)	Internal shell a hard cuttle bone. (Fig. 9.)	Sepiidae
38 (39)	Sepion large, up to 465mm. in length, no posterior spine in adult. (Fig. 9.) Drift shells only in N.Z.	Amplisepia apama
39 (38)	Sepion small, up to 110mm. in length, prominent posterior spine. Drift shells only in N.Z.	Solitosepia plangon
40 (31)	Eye with a perforated lid.	Oegopsida
41 (46)	Some of the suckers modified into hooks. (Fig. 6.)
42 (43)	Sessile arms with suckers only, some hooks on tentacle clubs. Gladius with no end cone.	F. Onychoteuthidae
43 (42)	Apex of gladius slender, strongly compressed, spoon slightly developed. Buccal membrane with 7 lobes and 6 pits. Two photophores in buccal cavity. (Fig. 6.)	Onychoteuthis banksi

page break

(Figures are not to scale)
Fig. 1: Ventral view of male specimen of Octopus maorum (Hutton) with mantle cavity opened to show disposition of organs. A, anus; C, gill; E, ejector of male genital apparatus; F, funnel; S, interpallial septum. Fig. 2: Hectocotylus of Rohsonella australis (Hoyle). Figs. 3, 4: Hectocotyli of Octopus maorum (Hutton) to show variation in development. Fig. 5: Details of locking apparatus of Nototodarus sloanii sloanii (Gray). Fig. 6: Tentacle club of Sepiotheuthis banksii (Leach). Fig. 7: Details of anterior portion of mantle cavity of Sepioteuthis bilineata Q. & G. showing cephalic and pallial elements of locking apparatus. Fig. 8: Gladius of Nototodarus sloanii sloanii (Gray). Fig. 9: Sepion of Amplisepia apama (Gray). Fig. 10: Gladius of Sepioteuthis bilineata Q. & G. Fig. 11: Outline of mantle and fins of Sepioteuthis bilineata Q. & G. Fig. 12: Outline of mantle and fins of Nototodarus sloanii sloanii (Gray). Fig. 13: Outline of mantle and fins of Sepioloidea pacifia (T. W. Kirk). Fig. 15: Shell of Spirula spirula (L.). Fig. 16: Horny ring from sucker of Nototodarus sloanii sloanii (Gray).

page 100

44 (45)	Apex of gladius very long, terete, vane missing. Spoon slightly developed. No photophores.	Moroteuthis ingens
45 (44)	Sessile arms with hooks, few suckers near tips, tentacle club with suckers only. Gladius with end cone.	Pterygioteuthis giardi
46 (41)	None of the suckers modified into hooks.
47 (52)	Funnel articulating with inner surface of mantle by a cartilaginous joint on either side.
48 (49)	Animals of very large size, mantle of 2-3 metres, fixing apparatus on tentacles extending along whole length of arm. Gladius feather-shaped, with terminal cone.	G. Architeuthis *
49 (48)	Animals of moderate size, mantle not exceeding 0.7 metres.
50 (51)	Siphonal element of mantle locking apparatus consisting of a T -shaped pit. (Fig. 5.) Fins saggitate, broader than long. Both ventral arms modified in the male. (Figs. 8. 12.) The common N.Z. squid.	Nototodarus sloanii sloanii
51 (50)	Siphonal elements of mantle locking apparatus an elongate pit with no transverse groove. Numerous photophores on ventral aspect of body, siphon, head and arms. Left eye much larger than right. The whole family to which this squid belongs have this latter peculiar modification for which no satisfactory explanation has been offered. (Fig. 22.)	Histioteuthis cookiana
52 (47)	Funnel fused with inner surface of the mantle on either side. Mantle fused with the head in the nuchal region. Tentacles (if present) without hooks.	F. Cranchiidae
53 (54)	Mantle beset with crystalline tubercles arranged in two rows on ventral surface.	Pyrgopsis pacificus
54 (53)	Mantle smooth.	Subfam. Taoniinae

^* Four species, each described from single stranded specimens, have been recorded, i.e. Architeuthis longimanus T. W. Kirk, A. stockii (T. W. Kirk). A. verrilli T. W. Kirk, and A. kirki Robson. At our present stage of knowledge stranded specimens of these giant squids must be identified from the original descriptions.

page 101

(The figures are not to scale)
Fig. 17: Mcgalocranchia pardus (Berry). Fig. 18: Diagrammatical representation of major points of cephalopod anatomy. (A, anus; B, beaks; C, crop; CT, ctenidium; F, funnel; G, gonad; H, hepato-pancreas; IS, ink sac; M, mantle; MC., mantle cavity; R, radula; S, stomach.) Fig. 19: Female genital duct of Octopus zealandicus (Benham) after Benham. Fig. 20: Female genital system of Octopus maorum (Hutton). O. oviduct; OG, oviducal gland; OS, ovarian sac. Fig. 21: Female genital duct of Octopus adamsi Benham. After Benham. Fig. 22: Histioteuthis cookiana Dell. Fig. 23: Shell of Argonauta nodosa Solander. Fig. 24: Female animal of Argonauta nodosa Solander. Fig. 25: A-C, Diagrammatic representation of inhalent and exhalent cycle through mantle cavity showing relationship of free mantle edge, funnel and funnel valve.