A study of the Comparative Morphology of Cephalopod Armature

Table of Contents:

There have been plenty of studies on cephalopods (Phylum Mollusca), including their hard structures. Unfortunately this has hitherto been restricted mostly to the buccal mass and beak of animals such as squids (teuthids), octopuses and cuttlefishes (sepioids). There is relatively little literature on the subject of the sucker pads on the arms and tentacles of cephalopods. There is even less that displays the information in a comprehensive and focussed document, solely intended for that particular subject. Most of the information is spread across vast amounts of scientific papers and books, where any text on suckers is usually found to be at best fragmentary.

I chose to study the armature of cephalopods in order to perhaps shed some more light on such a little-documented topic. In order to do this I had to find scientific literature. Because there is scant recorded information, I also had to observe preserved specimens directly, of which I took photographs and detailed diagrams.

It was discovered that cephalopods exhibit a great deal of variation in the architecture of their armature. The chief structure investigated was the familiar sucker pad, but other structures, which could perhaps be called unique, are also discussed; such as the cirri in deep-sea cirrate octopods, the strange adhesive ridges in nautiluses, and the very basic, if impressive, hooks found on the arms of fossil belemnites. A brief but concise description of the functional morphology of the structures is also included.

Introduction

We are all familiar with cephalopods; their amazing abilities and behaviours. Their endeavours, such as the octopus' ability to squeeze through the smallest of gaps, and its ability to blend into the surrounding substrata almost instantaneously, or the cuttlefish's use of vivid colour changes to communicate its emotions and intentions, never fail to astound us. The class Cephalopoda has even captured the human race's imagination so much as to feature prolifically in our mythology and folklore, from the Norse Kraken to H. P. Lovecraft's Cthulhu, and other folktales. Despite this familiarity with them, we still know only a small portion of their form, their history and even their intellect, which renders an already alien animal even more bizarre and exotic. Another frequently omitted fact is that most of the cephalopods we see and know today, bar one genus, are only a portion of the class. These are the Coleoidea. The one extant genus that is not a coleoid is the Nautilus, from the subclass Nautiloidea.

My focus is on the comparative morphology of the armatures of these cephalopods, which although would appear to be well known, is in fact rather poorly documented in the history of scientific literature, which is unfortunate as it proves to be a very deep and interesting subject, time and resources allowing. All known information on the subject is scattered across several pieces of literature, which for the most part are sketchy at best. True literature on the subject is indeed very difficult to come by as the morphology of suckers and the like have not been studied in any great detail. Also, half a year's study would not even scratch the surface of the subject owing to the diversity of the Class, and the probability that every paper on a new discovery would have to be studied in order to piece the facts together. As a result, I have had to come to several conclusions by looking at the evidence from what literature I have found, but also from direct observation of preserved specimens.

In order to give an idea of the immensity of the subject, with relation to just how long the cephalopods have been around, a brief history would seem appropriate.

The earliest cephalopods recorded have been described from Late Cambrian deposits from NE China. These primitive animals are typified by Plectronceras (such as P. cambria) and were very primitive, consisting of a small internally-chambered conical shell connected with a simple siphuncle. It is thought that they most probably lived a benthic lifestyle similar to marine gastropods.

It was not until the Ordovician that the cephalopods saw a huge evolutionary radiation resulting in the appearance of the orthoconic nautiloids. It is generally accepted that there were eight orders of orthocone, distinguished by the variations in the internal structure of their conical shells. These nautiloids enjoyed a long reign from the Ordovician to the Silurian almost unchallenged, filling many of the predatory niches that had been occupied by now long-extinct animals such as the Cambrian anomalocarids.

During the early Devonian, the cephalopods radiated even further, creating new forms such as the early nautilda, and a new taxonomic group in the form of the ammonoids, of the subclass Ammonoidea. These were highly chambered, mostly with a spiral shell. These ammonoids survived for approximately three hundred and forty million years, until the end of the Cretaceous, when they became extinct. The Late Devonian also witnessed the emergence of the Coleoida as an offshoot of the ammonoids which comprise of all modern cephalopods except Nautilus, the sole the surviving remnant of the subclass Nautiloidea.

It probably wasn't until after the Cretaceous extinction that the Coleoidea became more abundant and diverse; they were probably overshadowed by the ammonites until their disappearance, they were then able to diversify into some of the empty niches left by their externally shelled relatives.

The Carboniferous saw the beginnings of the diversifications amongst the Coleoida as well as the beginnings of a (coherent?) recognisably modern arm structure and the development of armature. Belemnoids showed a definite resemblance to modern squid, although from a different cohort, or group. They had ten arms displaying distinct armature, indicative of the basic coleoid plan, which would later be developed by the modern Decapodiformes.

It was at this time that what we recognise as modern coleoid cephalopods began to emerge, and diversify. Since belemnites are a separate branch on the taxonomic tree it is probable that suckers and hooks did not evolve linearly (one being the result of the other), as one might imagine, but very much at the same time. Because of the tiny amount of actual soft-bodied fossil evidence, owing to coleoid cephalopods being fragile creatures, it is very difficult to say for certain what happened. There are also questions raised such as those related to arm number, and the like.

Method; the Alternatives to Literature Study

As well as searching for scientific papers, using journals such as the Journal of Experimental Biology, Integrated Comparative Biology, and Acta Plaeontologica Polonica, and books such as Kir Nesis' Cephalopods of the World, I had to supplement what I found with direct observation of specimens. I contacted the St. Andrews Aquarium (formerly Sea-life Centre), and acquired a well preserved specimen of Eledone cirrhosa Lamarck 1798; the Northern, or Curled, Octopus; a native of the British coast. This would be the specimen I would use as my typical model, since octopuses' suckers do not display much structural variation, and really only differ in their number and size across species.

I also travelled to the National Museum of Scotland in Edinburgh, on the 5th December, 2005 to study several other specimens. The specimens were kept in a depot in West Granton Road where I spent the best part of a day studying specimens of Loligo forbesii Steenstrup 1856 (veined squid), Todaropsis eblanae Ball 1841(lesser flying squid), Sepia officinalis Linnaeus 1758 (common cuttlefish) and Architeuthis dux Steenstrup 1857 (giant squid).

Several photographs were taken during the examination of the specimens, the best of which have been included for reference. Please note in that the photographs a degree of discolouration in the animal has occurred due to the use of Formalin as a fixing agent which coagulates the cells of the animal. The specimens were safe to handle as the formalin was bound to the cells and therefore not capable of any further fixing. The specimens from the Museum were kept in ethanol. I had to drain my acquired Eledone specimen and clean it of formalin before I was able to examine it.

Basic Overview of the Cephalopod

All Cephalopods are ocean-dwelling invertebrates of the phylum Mollusca. As stated previously, the majority are of the subclass Coleoidea, with the exception Nautilus, the one surviving member of the Nautiloidea. The Coleoidea are split into two subdivisions; the Neocoleoidea and the Belemnoidea, the prefix Neo- is used to describe the extant group of coleoids, the belemnoids being extinct. Modern coleoids consist of two superorders: the Decapodiformes and the Octopodiformes. These are distinguished by the number of arms they possess. In modern decapods, such as squids and cuttlefishes, two specialised arms have been developed for prey capture, as shown in fig 1b.

Neocoleoid cephalopods have suckers on their limbs. These are the main structures that allow the cephalopod to grab and hold onto things, and one of the main structures that distinguish them from other orders. The suckers are always situated along the oral surface of the limb (the ventral or "under-" side) and they aid in the passage of food to the mouth.

fig. 1: Display depicting arm arrangement and number. These numbers are actually used in describing the arm, and correspond to both left and right sides. Notice how the octopod has lost arm 2, but still retains the corresponding numbers!

Decapods tend to live in the water column and have very active lifestyles. Virtually all cephalopods are short lived, with a high metabolism. Because of this they need to feed regularly. The best method of prey capture for a pelagic animal is to seize its prey, and decapods do this with specially designed arms called tentacles. These are the fourth arm pair on the animal (fig.1), and consist of an array of muscle called the "muscular hydrostat" (Kier & Van Leeuwen 1997), which allows the tentacles to be launched by elongation of this muscle. Once the prey is contacted the suckers adhere to the animal and it is dragged back to the mouth where the other arms manipulate that prey to be eaten. Some squids, however, use their tentacles more like a fishing line and just let them dangle in the water column, waiting for prey to pass by. This strategy is mostly employed by deep-sea squids (such as Chiroteuthis and Mastigoteuthis).

Decapod suckers are commonly described as having a horny ring of chitin; although Nixon and Dilly (1977) claim that the structure does not contain chitin (referencing Rudall, 1955, as the source of this information). Nixon and Dilly fail to mention what the chemical composition of the ring is, in absence of chitin, and in light of many papers published after this one that refer to the ring as chitinous, it is safely assumed, for the purpose of this study at least, that the ring is made of chitin. Suckers are usually applied in two or four rows (also called series) along the arm and run pretty much along most of the length. The tentacle has suckers only on the club which is the large, spear or spoon-shaped part at the very end. These are arranged usually in four rows although as many as fifty rows of very tiny suckers have been documented (eg Sepiolidae, Nesis).

Minute suckers can be found on the carpus of the tentacle, and in several squids have unique structures. Suckers with smooth rings on one tentacle will have corresponding tubercles (or knobs) on the other. This allows the tentacles to be fastened together in a similar fashion to "popper" fasteners. It has been observed that squids do this while swimming rapidly which would suggest that it is employed as either a way to stop the tentacles from trailing behind or flailing about upon movement, or as a streamlining device. Since squids when swimming using water propulsion from the siphon are oriented mantle-first, and their arms to the rear, the former would seem the more plausible theory. This could also suggest that squids at least do not have as much motor control in their tentacles as they do with their arms, which would make sense as the tentacles are mainly used for prey capture only. Sepioids do not have these structures as they can fully retract their tentacles into pockets at either side of the head.

The main distinguishing feature of octopods is obviously the number of limbs. Essentially squids and cuttlefishes have eight arms as well; the retractile limbs usually being called tentacles, and have a different structure. The other difference with in the suckers of the animals being that octopuses also only have one or two rows, or series, of fleshy suckers running along their arms. Octopuses of the genus Eledone typically have only one row.

Basic Limb Terminology

The arms of cephalopods, as shown in fig. 2, have pretty uniform terms describing the regions. The base (or basal end) of the arm is that part which is closest to the main body, usually the widest part of the arm and proximal to the mouthparts. The median is the middle of the arm and the furthest part of the arm, and the tip, is the distal end.

The terminology of the tentacle is slightly more unique and specialised. The most proximal part of the club to the head is called the carpus. The carpus is the region just before the swelling of the club, and usually has a random number and distribution of suckers. The next region is called the manus and is essentially the bulk of the club. This is the widest region of the tentacular club, and houses the largest suckers. Sometimes variation among suckers is found such as the presence of hooks. The final region of the club is called the dactylus and is narrower than the manus.

fig. 2: the differences in the terminology.

The use of Suckers

The function in suckers, as adhesive structures, is to create an area of pressure inside the cavity of the sucker that is lower than that of the surrounding ambient pressure. This essentially "sucks" the object, or substratum, towards the sucker, creating a seal where the rim contacts the substratum. This rim is integral to keeping the reduced pressure inside constant. A breach in the seal will mean failure of adhesion, which is obviously not what the cephalopod wants, be it a decapod or an octopod.

The differences between the limbs of the two superorders are not limited to their number, or indeed the presence/absence of chitinous rings. The morphology and structure are also completely different, and are described below. However, both types of suckers still face the problem of maintaining a pressure differential: that is, the difference between the pressures inside and outside the sucker when attached. Of course there is the potential danger of the pressure differential being too great, causing damage to, or even destruction of, the structure. In theory, too small a pressure could cause the sucker to collapse. However this may not be a problem, since the pressure differential is nearly always limited to a threshold determined by cavitation. Thus, the tenacity of the sucker (the measure of force per unit area) depends mostly upon the difference between the pressure inside and outside the sucker, which is in turn determined by depth.

A brief description of Cavitation

Cavitiation is the formation of bubbles of gas in a fluid when subject to a reduced pressure (Smith 1995). It actually causes failure of suction-based adhesive structures that are experiencing cavitation. This is important to cephalopods since they employ the use of suckers to adhere to surfaces, be it a prey item or a piece of substratum. So the cavitation threshold is essentially the highest amount of pressure a cephalopod sucker can attain before failure of adhesion, breaking the cephalopod's hold of the object. This limit is not set, however, and there are factors that can change the pressure differential such as depth and area of the sucker. Depth has an effect because of the relationship between depth and ambient pressure.

Typically, at sea level, the cavitation threshold limits a sucker to a pressure differential of between 100 and 200kPa. However, every 10m increase in depth allows an increase in pressure differential by about 100kPa, so at 10m the limit will be 200-300kPa, at 20m the limit would be 300-400kPa, and so on. With increasing depth, cavitation proves to be less of a problem.

The other factor that can determine the cavitation threshold is the area of the sucker. Smith (1995) proclaimed that an area of 7.5mm2 or greater showed had a "normal" pressure differential of 100kPa for both decapod and octopod when attached to a surface. However, an area less than 7.5mm2 showed a distinct increase in the ability to produce higher pressure differentials. How a smaller sucker achieves a greater pressure differential is unknown. It could be due to the small size being able to theoretically maintain a decent seal between the rim of the sucker and the surface. Smith also speculated that the smaller sized suckers behaved similarly to Laplace's law for pressurised containers, in that the stress observed in the wall of the container (in this case the sucker), while it maintains a reduced pressure, is proportional to the radius of the container. Admittedly, an area of 7.5mm2 does sound awfully small since this would mean a radius of about 2mm! Many cephalopods do have very small suckers, ranging in about this figure, but some of the larger squids may face problems with cavitation if they have a considerable amount of large suckers, at least at sea level.

DECAPOD SUCKER MORPHOLOGY

Functional Morphology:
The suckers of decapods display a great deal of structural variation in the form of the chitinous rings, but the basic morphology is similar. Decapods have suckers located on stalks. The sucker is essentially a rigid cylinder with a muscular piston housed within it (the acetabulum). This is all connected to the tentacle by a stalk. The chitinous ring is housed at the rim of the sucker (the infundibulum). Upon examining the preserved specimens I noticed that the suckers could be pivoted and manipulated using forceps, albeit in a fairly limited arc of movement. They could be twisted slightly and "wiggled" which seemed to suggest that decapod suckers are far from sessile, and can be moved and rotated.

To generate a low pressure of the water inside the sucker, the muscular piston is pulled away from the cavity, towards the arm. Smith (1996) discovered that pulling on the stalk in turn pulled the piston, which reduced the pressure of the water in the sucker. Presumably, therefore, the squid can engage the suckers by pulling the muscular pistons and disengage by relaxing them.

As was described, cephalopods are limited to a pressure differential of 100-200kPa at sea level, but the threshold increases with depth, as well as increasing inversely with a reduction in sucker area.

What is interesting about decapod suckers is that with this decrease in area, it is not a slight increase in pressure differential that is observed, but an exponentially large increase anywhere up to 800kPa!

Structural Morphology:
It must be made clear that decapod suckers are employed for prey capture only, except in some special cases where sexual dimorphism occurs. Sexual dimorphism is a term used to describe traits that distinguish the male and female of the species. In this case, a prime example would be the Grimaldi scaled squid Lepidoteuthis grimaldii Joubin 1895, which displays a grossly enlarged sucker and hook, found only in the males (Jackson and O'Shea, 2003), which would suggest a reproductive function as opposed to a feeding one.

Most squids and cuttlefish that have a neritic (coastal) lifestyle, or at least stay close to surface waters, tend to appear to have slightly smaller suckers. Cuttlefish are generally associated with neritic zones or benthic upper-slopes as opposed to pelagic zones.

The larger squids, including the giant squid, Architeuthis dux (fig.3) have very large suckers, as would be expected, reaching a diameter of nearly 20mm. Since cavitation is not so much a problem at depth, and indeed giant squid are reported live only at extreme depths, it can afford to have large suckers, although quite why it would need to have such large armature is mysterious. The only reason that springs to mind would be predation on large prey or as protection against predation itself. Since most of the giant squid's diet consists (at least in part) of prey items not considered to be overly large (eg. E. cirrhosa, and Norway Lobster Nephrops norvegicus, although the Atlantic Horse Mackerel Trachurus trachurus can achieve a considerable 70cm total length!), it seems odd that they display suckers of such size. Bolstad and O'Shea (2004) discovered in the gut of an individual the remains of another giant squid, which would suggest at least cannibalism in the species. They also proposed that Architeuthis dux is in fact a pelagic animal rather than bentho-pelagic. This would suggest that the giant squid would probably live at much lesser depths than is commonly believed, since most trawled specimens were found at a depth of between 400 and 600m. Still, the reason for such a large diameter (the area would be roughly 314mm²!) is still a conundrum, unless the squid relies more upon the huge horny rings around the rim of the sucker.

fig. 3: Suckers on the tentacle of Architeuthis dux. Note the clearly visible piston musculature at the base of the sucker cavity. Inset: side view of the chitinous ring.

The specimen of Sepia officinalis displayed relatively large suckers on their tentacle clubs in the second outermost row. The diameter is 4mm, so the area will be about 12.6 mm2 which is quite considerable for a coastal animal! According to Smith, the highest pressure differential the cuttlefish could achieve would be 100kPa.

What distinguishes the decapods, though, is the variation in the formations surrounding the edge of the sucker cavity. Surrounding the cavity of the sucker pad is a horny ring. This ring typically consists of chitin, a polysaccharide. It is usually circular, and quite often has structures that protrude from its base. These denticles, or teeth, display a large variation across species, even families, of squids and cuttlefish; the most variation evident in squids. In coastal families such as the Loliginidae, and in most sepioid cuttlefish, circularis muscles surround the base of this chitinous ring, although information on the function of this muscle was unobtainable. Presumably it is just to hold the ring in place, or may facilitate some movement. The ring surrounding the sucker has many proposed functions; the most obvious being to aid the sucker in prey capture. Of course, the sucker is the primary weapon for seizing prey, but the horny ring would aid in latching onto the prey item initially in he strike.

In the Loliginidae, the teeth are relatively small and can be described as acuminate (Nesis); the term used to describe claw-like or conical teeth. Another feature in the rings of Loliginid squids is the tendency for larger and smaller teeth to alternate round the circumference. Indeed, the Loligo forbesii specimen I examined showed this characteristic, although the size difference was little in the alternating denticles there was still variation. The arms of the specimen showed two rows of suckers, the diameters being relatively similar until quite far along the distal part of the arm. The tentacles were fairly typical in that there were four distinct rows, with the largest suckers present on the Manus; the middle section of the club (fig.4).

fig. 4: a shows the club, with diameter of the largest sucker, as well as a sucker of similar diameter to those on the medial part of the arms. Notice the dark brown ring; this is the chitin! B is a line drawing of the sucker. The ring is slightly exaggerated to show its formation.

Another structural pattern in the sucker ring is that of a crenellate shape; being square, like that of a castle or tower rampart. In fact, it very much resembles this defensive structure, although the function is completely different. An interesting structural form is that of a completely smooth ring surrounding the edge of the sucker, which is found in the family Onychoteuthidae (Nesis, 1982). These muscular squids display the smooth ring in both rows along their arms. The tentacles of the family are markedly different from the arms, in terms of the armature, due to the presence of hooks running in two rows. Onychoteuthidae typically only have two rows on their clubs from the manus to the dactylus, with a circle of the inter-locking "popper" suckers on the carpus, as described above.

It is generally accepted that the coastal Myopsid squids (the suborder that comprises the Loliginidae and the Australiteuthidae) have simple ring structures. These structures do vary in the manner described, from pointed to square, alternate to all the same size to even slightly uneven rings.

The Oegopsid squids, including the onychoteuthidae already mentioned, are a large suborder that contains the most dominant squids in the pelagic regions. These squids are very much oceanic, which is reflected by a variety of structural differences to the myopsids. One difference is that they lack the circularis muscle found in most loliginids. The other, and perhaps the most obvious, is the presence of actual tentacle hooks, found commonly in the oegopsids.

The hooks are formed by specialisation of the chitin ring, usually housed in distinguishably larger suckers. It would seem (Nesis 1982, Engeser & Clarke 1988) that the hooks develop during the growth of the young animal, and are not present at birth. Thus hooks, or more specifically a full quota, can be associated with a fully mature animal. The denticles (Engeser & Clarke) on the ring of such a sucker are lost save for the central, most distal, one. The distal end with the one remaining denticle then begins to extend, which draws the sucker edges together. The edges then fuse, which causes the proximal edge of the ring to form basal lobes. Usually a sheath accompanies the hook, which may play a protective role, either for the hook itself or to minimise the risk of the squid impaling itself!

The presence of a hook reduces the sucker to virtually nothing, except maybe a small opening in the structure. Although there are many squids that employ both hooks and suckers, as in the case of L. grimaldii which has, for the majority, suckers, there are many squids that employ the use of hooks only, at least on the tentacles of the onychoteuthidae. Because of this, it is safe to assume that these squids are not relying on negative pressure to adhere to a prey item, but impaling the prey in order to catch it. It was suggested above that the hooks in L. grimaldii were not to aid in predation, but in reproduction. The main apparent reason for this is that the hooks are not located on the tentacles, the primary hunting tool, but on the second arms. They also occur only in the males. According to photographs taken by Jackson and O'Shea, sucker ring is still very much present, and the huge hook resembles a very pronounced single denticle, albeit looking particularly like a miniature claymore or broadsword! It was proposed that the hooks were embedded directly into the female's skin or lock into her scales (dermal cushions that give the species its name). It is also possible that the hooks are employed in male-male agonistic mating displays. The scaled squid does not have a hectocotylised arm, and instead has a relatively long penis (or terminal organ), which probably means that it implants the spermatophores directly into the female's mantle cavity, which would mean that hooks which allow purchase of the female's body during this mode of reproduction would indeed be favoured. In this case, reproduction would probably take place "head to feet" as it were, with the partners facing opposite directions. Technically this means that a female with damage to either her mantle or the scales would indicate the probable use of the hooks, where a male has held the female during reproduction.

The formation of the ring in the lesser flying squid is rather notable in that the denticles appear to protrude from fleshy ridges, which resembled sockets, the denticles, or at least what was visible of them, were very small, and slightly rounded (fig.5).

fig. 5: a, clockwise from left: the full club, one of the larger suckers on the manus, the dactylus. Part b shows the tentacle- It looks like only 1 series of suckers, but it is in fact 2. Part c is a line drawing showing the folds with the small, rounded denticles.

The cuttlefish, of the order Sepiida, displayed slightly different superficial sucker architecture. Although the suckers were obviously virtually identical in their functional morphology, the sepioid sucker did look a little different. For one thing, in the typical teuthid sucker, at least from the research and observations conducted for this paper, the rim of the sucker cavity (and the ring) was similar to the area of largest diameter on the sucker body, which gave the structure a defined bulb-like appearance, whereas the cuttlefish sucker appeared more "enclosed" in that the rim of the sucker cavity looked as if it melded with the rest of the structure behind it. This can be seen by comparing figs. 5 and 6.

The common cuttlefish specimen exhibited denticles that resembled tiny, curved claws. The actual ring itself was hidden in the rim of the sucker. This order also has the circularis muscles that the loliginids possess. The suckers did not appear to be located on stalks, but they did move quite freely upon gentle manipulation with a pair of forceps, which would suggest the presence of a stalk. The outside of the sucker was very broad giving an upturned bowl shape, which is what probably rendered the stalk invisible.

fig. 6: a, left: the tentacle club of S. officinalis and right: the triangular arm, b; schematic of the large (4mm) tentacular sucker.

This appears to be fairly typical of sepioid cuttlefish suckers. The suckers on the arm were in four tightly packed rows. The arms themselves were interesting in that they were actually short and stumpy, tapering sharply to form almost triangular arms. The presence of the fully retractable tentacles means that the cuttlefish doesn't always have to employ them to hunt. Cuttlefish are mostly neritic animals living near the sea bed, and the common cuttlefish is a model of this. It seems that when hunting benthic animals the cuttlefish merely hangs above its prey and lunges at it using the eight arms, and not the tentacles (Nixon & Dilly). The tentacles therefore, are probably reserved for fast swimming, pelagic animals like prawns or fish, as opposed to crabs. All of these comprise part of the common cuttlefish's diet, and probably that of other species, since most live in coastal waters.

OCTOPOD SUCKER MORPHOLOGY

Functional Morphology:
Octopod suckers are very much identical save for a few special cases. They are also functionally and physically different from decapod suckers. The most obvious difference is the lack of a hard structure in the octopus' sucker. Since octopuses use their suckers for a wider range of activities than decapods, such as locomotion, adhering to substratum and even picking up, manipulation and collecting attractive objects, they are more tactile. The basic design revolves around two chambers, as described by Kier and Smith (1990 and 2002), the acetabulum and the infundibulum as shown in fig.7 and fig.8.

fig. 7: the cross section of a sucker from the Eledone cirrhosa specimen. The structure is typical of octopuses, and the only distinguishing characteristic is that this species only has one row of suckers as opposed to 2.

fig. 8: diagram of an octopus sucker distinguishing between the tissue types.

The principles of octopod suckers are similar to that of decapods: in order for the structure to adhere to an object, the aim is to generate a reduced pressure. While squids and cuttlefish employ the use of a muscular piston, octopus suckers use two cavities. A seal is formed between the object and the sucker, by the fleshy rim. The radial muscles in the acetabular wall and roof contract so that the acetabulum increases in size. This is because the radial wall and roof become thinner and longer making the acetabulum wider and taller. If this were to happen in the water column then water would just rush in to fill the larger area. However, the sucker is pressed against an object, prey item or piece of substratum, so no excess water can enter. The water inside the sucker resists expansion and so there is a decrease in pressure. This water essentially begins to act like a solid whilst under pressure, allowing adherence to the surface. In the acetabular wall, the circular muscle bundles are the antagonists of the radial muscle, and reduce the circumference and height of the acetabulum upon contraction, which in turn releases the sucker.

Structural Morphology:
As I have already described, octopus suckers do not show much variation, save in a few very specialised cases. It seems to be that generalisation limits the variation in structure, whereas specialisation encourages unique features at the expense of how many functions it can perform. Although the basic octopus sucker doesn't vary much across species, except for maybe slight superficial changes, such as size and shape, it can perform many extraordinary feats such as being able to literally fold in half in order to hold onto an object as thin as a sheet of paper (Kier and Smith 2002)! Octopus suckers are also able to adhere to irregular surfaces, presumably due to them not having hard parts anywhere. They can also be used as chemotactile organs (Kier & Smith, and also Chase & Wells, 1985). Octopuses do not just forage using eyesight alone, although they are clearly visually-oriented animals, and frequently search for food using their arms to probe where they can't see. Perhaps a cross between touch and taste (or smell!) would be the best description of the term "chemotactile". Octopuses are also attracted to items and objects in a similar way to magpies. Frequently it will pick up, for example, a piece of glass that is glinting in the sunlight and carry it back to its den where it keeps a collection or "horde" of shells and debris. As well as adhering to the smooth edge, it would probably grasp the thin edge by bending a sucker in close proximity to it. An analogy could be a party game where the participant has to pick up an object using only a folded paper plate.

The arrangement of the suckers is also interesting. In general, the suckers begin very close to the mouth, and increase in diameter while continuing out along the arms in either one (usually associated with the Eledone genus) or two (the more common number) rows before decreasing in diameter as the arm thin out to a point. The difference in the basal diameters is clear in fig.9. The mouth can also be seen in fig 9, which is the opening in the centre of the suckers in the distance.

fig. 9: oral view of E. cirrhosa. Note the distinct increase of sucker diameter. The diameters were relatively constant until quite far past the median in this specimen, only decreasing once they started reaching the tips. Also notice how flexible the suckers look. It is also possible to see the acetabulum through the holes of a few of the suckers.

The suckers can also rotate, and move quite freely (as demonstrated by the example of it folding), which is how the octopus is able to manipulate prey and move it towards the mouth.

Most octopuses have a hectocotylus. This takes the form of a specialised arm for use in reproduction and includes a sperm channel. The hectocotalised arm in incirrate octopods is one of the third arms (Arm 3), and is inserted into the female's mantle cavity to allow transfer of the males' sperm packets.

The lining of the infundibulum, a thin pellicle, is frequently cast off. During this activity the octopus is commonly said to be "shedding its suckers" (Nesis). Presumably this is to ensure that the condition of the infundibulum is kept optimal, and free of dead skin or other debris that could impair its adhesive abilities. Since octopuses spend a great deal of time attached to surfaces that are potentially colonised by algae and other small colonial or film- producing organisms, a good way to keep the sucker free of unwanted hitchers and to ensure reliable function, would be to shed the layer that's in contact with these conditions the most!

Another reason could be that the octopus would want to keeps its suckers as flexible and supple as possible, and if the lining inside gets too rigid or firm, then the sucker will become limited in its movements and its usefulness impaired. Essentially, frequent shedding will keep the pellicle fresh.

PREHISTORIC COLEOIDS: BELEMNOIDEA

Belemnoids are prehistoric coleoids that were most abundant during the Mesozoic era, and became extinct at the end of the Cretaceous. They came from the same ancestral stock as the neocoleoids, and lived alongside them from the early Carboniferous right up until their demise at the KT boundary.

Belemnoids typically display the ancestral arm pattern (fig. 1a), comprising ten subequal arms, with no distinct specialisation. This is very much speculative, since we only know of belemnoids from fossil evidence. There have been some well preserved individuals, however (Fig. 10). Of course they do have hard parts, but even the soft parts, such as the arms, and mantle, have been preserved in the rock remarkably well on rare occasions, possibly implying a very muscular, and not gelatinous, animal. On the other hand it could have just been the composition of the sediment in which the animal came to rest.

A remarkable feature of the belemnoids, and one that fossilises very well, is the presence of distinct, large hooks running in two series along the oral surface of the animal (Engeser & Clarke, 1988). Figs.10 and 11 show photographs of fossilised belemnoid remains.

The basic belemnoid hook probably comprised the same components as that of extant coleoids; namely chitin. It is possible that the hooks contained more components, such as harder, possibly mineral, materials, but then the only evidence for this really is that the hooks are so well preserved, compared to fossil neocoleoid suckers rings. The hook was positioned curving inwards, and in two series running along each arm, in pairs of between 20 and 50.

fig. 10: Acanthoteuthis from Solenhofen, Germany, currently residing in the collection of U. S. National Museum of Natural History. Of particular recognition is the clear detail of the animal's body, including the hooks clearly radiating off the arms!

fig. 11: belemnoid hooks (possibly Belemnoteuthis) taken from Lyme Regis in Dorset, England, also in the collection of U. S. National Museum of Natural History. Both photographs used with kind permission by Richard E. Young, University of Hawaii.

The hook itself is split up into several sections, each with its own distinct nomenclature as shown in fig.13. The presence of this small hairline groove known as the orbicular scar indicates that the hook was probably not exposed, and was housed within a sheath of soft tissue. Because the tissue attachment point is quite far up the shaft, only the uncinus would be visible, or at least naked. Engeser and Clarke proposed that the hook was stalked, and that it was attached by muscles that could have allowed lateral movement. Since the belemnoid did not have an extensile pair of limbs, like the tentacles of squids and cuttlefish, or the flexible, muscular sucker discs like octopuses, it would have had to manipulate the prey for eating by using its whole arm. If the hooks were sessile, bearing in mind that it may have had up to fifty, then the belemnoid would be severely limited in its arms' movement. Having hooks that could be moved, even slightly, would allow the animal more movement. The manner in which the hooks were positioned, recurved, would have also helped, in that the trapped prey, which was obviously caught by being impaled, could only move in one direction in order to free itself from the hook: towards the mouth. According to Engeser and Clarke, there are three groups of belemnoids that display arm hooks, and one that does not. The Phragmoteuthids are reported as far back as the late Permian and are described as having more slender, slighter hooks than what is described as the "normal" or archetypal hook (fig.13) although variations including enlarged bases and more curved uncini are reported in species. The Belemnoteuthids have more typical hooks, although Chondroteuthis wunnenbergi Bode 1933, exhibited prominent spurs on the proximal lower third. The Belemnitids, most well known as the belemnites, also display a deal of variation. Acrocoelites (Toaricibelus) raui Werner 1912 appears to have two different types of hook, differentiating from the typical hook to those with an internal spur towards the distal end of the arm. Two fossil specimens of Passaloteuthis paxillosa Schlotheim 1820 showed the same structural oddities of having no hooks whatsoever towards the distal parts of the arms. This suggests that this species may well have lacked hooks at the ends of their arms. Unfortunately no other specimens were described or even mentioned, so a conclusion can only be drawn from these two fossils alone.

fig. 13: schematic of a typical belemnoid hook, adapted from Engeser & Clarke

Ideally a more thorough investigation into prehistoric cephalopod arm structure would have been desirable, but there is so little fossil evidence of cephalopods, being a soft-bodied animal naturally, that such a task is nigh on impossible at present. Most of the neocoleoid fossils are little more than faint outlines with maybe the presence of one or two organs, or surfaces, but this can be sketchy at best, since the fossils are interpreted differently. For this reason, only belemnoids have been included as a representative of prehistoric cephalopods since the available information on fossil cephalopods with regard to armature is pretty much restricted to the belemnoids. It is curious just how well these animals have preserved compared to the other prehistoric cephalopods, given that this class has been in the world's oceans since the Late Cambrian.

SPECIAL CASES

The case of Stauroteuthis syrtensis:
Stauroteuthis syrtensis Verrill 1879 (Johnsen et al., 1999) is a deep sea cirrate octopod that sports very unique armature. It is similar morphologically to other cirrate octopods in that it possesses two rows of cirri along the oral surface (the underside) of each arm, and also has a long web that connects each arm with a single series of structures analogous to the suckers of a typical octopus such as E. cirrhosa or Octopus vulgaris. However, these structures only vaguely resemble their adhesive counterparts, and can even be described as vestigial. Indeed these suckers no longer perform any adhesive role whatsoever, instead they exhibit Bioluminescence! Although bioluminescence is common among the decapods (Johnsen et al.) it is incredibly rare in octopods, being described in that manuscript as only being present in "2 of the 43 genera", and even then only found in breeding females. In this species, however, it appears that both females and males possess the light organs (photophores). Robison and Young (1981) observed that the light emitted from these modified suckers were of a definite green hue, compared to the blue-green light more commonly found in bioluminescent animals. This does set it apart from other inhabitants of the deep with such similar abilities!

The problem with the information regarding this species, as pointed out by the team that produced said document, is that it lives at great depths. Indeed the three specimens were caught in a range of 750- 920m using deep trawling methods. They are understandably difficult to study in their natural environment not only due to the depth, or the mind-blowing vastness of its habitat, but also because the octopod can probably detect the submersible light long before contact, meaning that it is unclear whether the behaviour observed is natural or induced by the presence of the submersible. It is described as hanging in the water column with its arms and web outstretched to create a sort of bell-shaped posture. It is possible that this is a threat response, but it appears to be a natural position rather than that of defence. There are a few theories as to why such an animal would develop bioluminescence, which can be both an advantage and a disadvantage, as will be discussed.

S. syrtensis exhibits sexual dimorphism in that mature males have grossly enlarged suckers (or photophores) compared to those of a female of similar development. Since both have the bioluminescent ability, it was proposed that it was used as sexual displays. Sexual signalling is understandably common in a phylum that appears to rely so much on vision, and so this theory would appear to make sense. The evidence of a distinct light colour also enables what Johnsen, Balser, Fisher and Widder described as a "private line of communication" between members of the same species. Indeed, any clear visual cue that distinguishes you from another species, especially in the dark, is bound to be favourable! However, there are a few factors that may serve to contradict this theory. The first is that in this instance, all three specimens examined had these photophores, and yet they all appeared to be immature. Surely only mature individuals, with the ability to reproduce, would only need these? But then, we only know a small amount about this species, so anything is possible. Another problem is that the octoradial pattern (imagine eight twinkling spokes) would be distorted after a certain distance due to light dispersal. However, it is questionable that this would prove too much a problem since the photophores emit a fairly distinct green colour! Finally this colour would be clearly visible to potential predators. That said, although in a slightly different context, many male birds compromise cryptic camouflage in favour of conspicuous coloration and displays that give its presence away, in order to attract mates. Although sexual signalling probably isn't the primary reason for photophores, I am not, at such an early stage, wholly willing to dispel the possibility.

This being the case, a slightly more plausible hypothesis describes the light's use in prey capture. S. syrtensis feeds on small, planktonic crustaceans. It has been proposed that cirrate octopods secrete lines of mucous web which are handled by the cirri to trap the animal's prey. This kind of method has been suggested for three of the cirrate genera (all of which possess "non-functional" suckers), Stauroteuthis being one of them. In this case, the bell posture could be easily explained. It is possible, therefore, that the photophores are used as a lure, similar to the angler fish. These crustaceans have well developed eyes, and seem to be attracted to light sources, which would serve the S. syrtensis very well. The fact that the animal is usually observed in the bell or umbrella posture either oriented with its mouth facing upwards, or facing downwards could coincide with nutrient flows in the water column, since many planktonic and nektonic animals make vertical migrations every night in order to feed on phytoplankton. On the other hand it could just be, as Denton (1990) stated, because upwelling light intensity is a small percentage of downwelling light intensity, which Johnsen et al. proclaimed would make the octopod highly visible to potential prey higher up in the water column, although this statement I have difficulty in imagining. Surely if the light intensity of an octopod facing upwards was less it would be less visible to prey above it?

The case of the Vampyromoprh:
The Vampire squid, Vampyroteuthis infernalis Chun 1903 is the only known remaining member of the order Vampyromorpha (sometimes Vampyromorphida). There are fossils of these animals ranging back to the Mesozoic, which is rather remarkable given their gelatinous tissue composition! A resident of the deep oceans, this is yet another relatively unknown animal.

The vampire squid, although commonly referred to as a "squid", is a member of the Octopodiforme superorder. Its second arms (Arms 2; the ones octopods have lost) have been modified to retractile sensory filaments that actually look like they are not part of the limb structure since they are not connected by the web. Like the cirrates they only possess one series of suckers, which are "banked" by a row of cirri on either side, and in V. infernalis are only present of the distal half. The web is attached to the arms to about two thirds of the way.

Young (1967) showed that the filaments are innervated differently to the other eight arms, which suggests that they are derived from a source different to the "decapod ancestor". These filaments are used in feeding and are probably used like a fishing line. A prey item bumps into the filament in the dark waters. Upon sensing this stimulus, the vampire squid lunges toward the source in the hope of enveloping it with the arms and web. It is possible that the cirri play a part in this too. The vampire squid also displays unique behaviours, especially in response to a potential threat. One threat response is to invert its arms and web round its head, as if it's turning itself "inside out", and extending its cirri outwards so that they resemble spikes or hooks (fig.14) in what is known as a "Pineapple posture".

fig. 14: photograph of the vampire squid in its "pineapple" posture with cirri held rigid.

Another predator-avoidance strategy is the use of photophores in the skin, combined with ejection of mucous containing bubbles of bioluminescent light. Once a string has been completed the squid halts operation of its photophores and slips of, leaving the bioluminescent mucous thread decoy with the predator.

The suckers of V. infernalis lack a cuticular lining, which may suggest that they do not have the ability to seal the sucker to an object, meaning the sucker would be useless, if not extremely weak, as an adhesive structure.

The case of the Nautiluses:
The nautiloids are the oldest cephalopods on the planet. From the Cambrian to the Early Devonian they were the only cephalopods found in the oceans, and even continued to dominate right through to the Jurassic period, where they experienced a sharp decline, which saw the extinction of the orthoconic nautiloids. It is the early nautilda that gave rise to our modern nautiluses, of which there are only around six known species. The early nautilda showed many of the characteristics familiar with these modern species. The nautilus really is a relic of an ancient time!

What sets the nautiluses apart from the coleoid cephalopods is the absence of suckers. By definition, nautiluses do not have arms, and instead can have anywhere up to ninety tentacles which are kept in sheaths when not in use. The absence of suckers does not dictate a lack of adhesive ability however, and the nautilus still employs special structures to allow it to grip objects, albeit using a completely different strategy.

Nautiluses have, lining their tentacles, a series of alternating ridges, or cirri, and grooves. These allow nautiluses to grip substrata or prey with an effect similar to Velcro. In the emperor nautilus, Nautilus pompilius Linnaeus 1758, and most likely in the other five species (and both N. pompilius subspecies), the cirri are in turn lined with annular ridges that run round the circumference of each cirrus, and are more pronounced on the oral side (Muntz & Wentworth, 1995). The epithelium exhibits thickening on the proximal surface of these ridges, and houses electron-dense granules which are only found in this location on the cirrus. The granules contain a mucopolysaccharide which could be what facilitates adhesion.

Nautiluses have incredibly poor vision; they're eyes being little more than pinhole eyes (Basil et al., 2000), meaning that they can probably only distinguish light and dark, and are most likely unable to clearly define objects. Clearly nautiluses do not rely on vision to hunt or scavenge; they tend to feed on dead matter but they are still active predators so they need to forage, and locate prey.

Their tentacles also house cells that resemble taste buds (Basil et al.), but were found not to allow the animal to track the odour, or chemical signal. For this the animal requires two specialised organs located below the eye called Rhinophores. While the tentacles were found to pick up chemicals in the water, it was the rhinophores that allowed them to locate the source. Test animals with their rhinophores temporarily plugged did show a distinct behavioural change upon the presence of a chemical in the water, but they could not effectively locate the source. This suggests that the tentacles could pick up preliminary stimuli which tell the animal that a food source is there. The emperor nautilus inhabits fairly deep regions (300m approx.) and migrates vertically at night to feed. Essentially it lives in a dimly-lit, or even completely dark, world. If the animal was using its rhinophores all the time, such a behaviour would surely be wasteful, and so a preliminary warning system that tells the animal when it can use its rhinophores to locate prey would be a huge benefit.

Aside from foraging, the nautilus also uses its tentacles to adhere to surfaces during stages of inactivity. According to O'Dor, 2002, nautiluses engage in a state of torpor akin to sleep. He also stated that an animal in this state, if not anchored to any substratum, would drift around at a velocity fairly close to its optimum speed, in a random fashion. This movement is induced by breathing alone. Since nautiluses achieve locomotion by "jetting". Since they move by expelling water from their mantle cavity, the same site where the gills are situated, random movement would be understandable during unconsciousness. This of course poses many hazardous implications for the nautilus, and the animal would quite rightly want to stay put!

CONCLUSION

Similarities and Differences:
The sucker of the decapods and octopods are analogous. They both perform exactly the same function; adherence to a surface, but the morphology is completely different. The basic uses are also the same, although the octopods tend to have more uses for their sucker pads.

It would appear that decapod suckers are very limited in their uses, being mostly for prey capture: to aid in the adherence of the sucker to the body of the prey, and in some cases, probably by impalement! Some hook structures, such as those found in sexually dimorphic species, especially in those in which the dimorphism is dictated by the hooks, there is the possibility of a use in reproduction, most likely in species that reproduce through direct implantation by the male of sperm into the female's mantle cavity. Aside from this, decapod suckers do not appear to have any other role. Octopuses use their suckers for more than just prey capture and manipulation, and the suckers are more flexible and muscular. The sucker structure itself is relatively sessile, but the rim of the infundibulum has a remarkable array of movement. Octopuses are regarded as highly intelligent animals, which could be a result of their multipurpose suckers, or even vice versa. The intelligence of decapods was not researched for this paper, but it is possible that squids at least cannot achieve the intellectual level of an octopus, on the evidence of their highly specific armature. This statement is completely subjective, however, but it is one conclusion that can be drawn from the observation of the different structures.

The decapods do show a lot more variety in terms of structure, such as the shape and size of the ring in relation to the sucker. This may be a reflection on the animal's prey. Since nearly every family, at least, have architectural differences in the ring the different shapes could prove optimal formations for catching particular prey species, or perform better at different velocities or depths. In a three-dimensional world, it would also be wise to include the decapod's angle in relation it its prey, which might also affect the success of the tentacle strike. Decapods typically have a very fast tentacle strike; the tentacle extending and contacting a prey item in about 15-35ms (Smith, referring to Kier, 1982), with an incredibly high success rate: normally 80-90%. The chitinous rings could play a large part in this success, allowing an initial grip which facilitates formation of a rim, in turn allowing adhesion by pressure differential. Owing to the highly specialised use of tentacles, it is possible that decapods do not have as much motor control over their tentacles as they do their arms.

Octopus capture methods may differ somewhat as the octopus is using one or more of its arms to grab a prey item, which it probably has more control over, compared to a pair of tentacles. As such it can probably allow for a much surer strike. Octopus arms also frequently coil up when they grab an object, which no doubt increases its hold over the item, which means that they may not require hard, claw-like formations. Octopuses also lunge at their prey with all their arms, usually from above, in the water column. This behaviour is also demonstrated by cuttlefish. Another reason why octopuses do not require hooks or rings is that they are ambush predators and are masters of camouflage, able to change their appearance and colour at will. This means that they can close in on a prey animal, and will only strike when they are sure of success. In a three-dimentional environment, such as those of pelagic squids, there is no such luxury of a complete surprise attack, and so the squid would have to rely more on a high-velocity mode of capture.

Indeed, hard parts may even prove detrimental to an octopus, given the majority of octopuses' habitats; the sea floor. Octopuses live in and around rocks, and frequently reside in dens. It is common knowledge of the octopus' ability to squeeze into the narrowest gaps, and its inquisitiveness and tendency to explore and probe. An octopus can go anywhere so long as its beak can get through. If an octopus was carrying around so many hard structures in its arms, then it would be severely limited in where it would go.

It seems to be the case generalisation demands a simpler anatomical structure (in this case the octopod's suckers) whereas specialisation begins to limit the range of uses, in order for that structure to be optimal in that one special activity. This appears to also be the case in cephalopods with regards to their structures.

Cephalopods are very visually oriented animals, and employ the use of complex signals. This does include very complex, and impressive colour displays using chromatophores to switch to a variety of colours to beautiful effect, but the use of arm signals is also important. Although this is more holistic, and may not even include the suckers in some cases, an interesting implication does arise when considering sexual dimorphism. As in the case of L. grimaldii it was presumed that the hooks were used physically in mating, but just as probable is role in display. Just as stags sport their impressive antlers during the rut, that are usually only employed as a grappling weapon if two competing males refuse to back down, so the male scaled squid may perform similar displays to impress a female. Until two scaled squid are found in the act of mating, we can only speculate.

A conclusion can also be drawn from the terminology derived from the observation of cephalopod limbs, namely the tentacles. It is found that in both the coleoid decapods and the nautilus both have limbs called tentacles. Which it was observed that tentacles were specialised limbs, what is consistent in both subclasses is that these limbs are fully retractable and extendible. Indeed nautiluses only sport this kind of limb, and do not have any unspecialised arm. Whether it was by coincidence, it can nevertheless possible that the term tentacle can refer to a limb which is extendable.

A possible Evolutionary Theory:
The existing Nautiloidea, of the genus Nautilus, exhibit what could be classed as a rather primitive array of armature. Providing the Nautiloidea did possess these adhesive ridges to begin with, and it wasn't just convergent evolution displayed by the nautiluses, this could very well explain the beginnings of the cephalopod sucker structure. It's doubtful that nautiluses alone would just happen to develop an adhesive system on their limbs in recent times without good reason, and so it would be expected that the ridges are the very ancestral progenitor of the sucker pad. This theory would be ideally backed up by the discovery of soft-bodied bactritid ammonoid fossils, these being the intermediary stage between nautiloidea and coleoidea, and ancestral to the later ammonoids with perfectly preserved arms that could show whether any structures were present. Since in ammonoids the only fossils reported have thus far only been the shell, conclusions have to be drawn.

Assuming that the nautiloids did have adhesive ridges, it can be suggested that these were carried and further developed by the ammonoidea: more specifically the aforementioned bactritids, which gave rise to the coleoids.

Assuming also that the bactritids (at least) had some sort of ancestral structure, such as a hook, makes the presence of the belemnoid hook more plausible as well as the neocoleoid hook. These structures could not have just sprung up out of nowhere.

It is possible, therefore, that the hook structure was possessed by a common ancestor to belemnoids and neocoleoids. This common ancestor could have had an ancestral hook. When the belemnoids and neocoleoids diverged, the belemnoids may have kept the ancestral hook (or indeed developed it in a completely different way!), while the neocoleoids developed the hook into a sucker pad. A short diagram showing the development from hook to sucker is shown in fig. 15.

fig. 15: a depicts development from an ancestral hook similar to that of a belemnoid's to that of a modern decapod, including the formation of the cavity (acetabulum) and a stalk, and the transition from a single hook to a ring of denticles. Included is b which depicts what could happen, carrying on from a in the same direction.

For sake of completeness, further development into the octopods sucker, lacking a ring, is included. If this were to happen, then because octopods did not branch off of decapods, and both more or less diverged in the same manner as with the belemnoids, theoretically the first octopods would have had some hard structures until these vanished. Of course the internal morphology would have to change again, but slight modification of the infundibulum and acetabulum, and the development of the piston into the acetabular roof could facilitate this. This is highly speculative as there is not a lot of evidence of fossil neocoleoids, but I'd like to hope that it fits together as a sound theory. One interesting fact is that the "paper nautilus" Argonaut argo, an incirrate octopus that secretes a very thin shell, appears to have hooks between the rim of the sicker and infundibulum (Nixon & Dilly); could this be a throwback to a day when all coleoid cephalopods had hard armature?

Thanks:
I would like to thank the following people for their advice and help throughout the duration of my study, in no particular order:

I thank the display manager at St. Andrews Aquarium for the specimen of Eledone cirrhosa as well as giving advice and showing an interest in the study. I also thank William Kier for providing me with his two papers on octopus suckers and having an interest from the offset.

Special thanks goes to the staff and forum members of TONMO, for the endless help, ideas, advice and critical views, including Phil Eyden for proof reading this paper and providing extensive information on fossil cephalopods, Dr Steve O'Shea of Auckland University of Technology in New Zealand, and especially Kat Bolstad, also of Auckland University, who provided endless help and sources. I would have struggled to obtain some sources otherwise.

Finally, within the department, I would also like to thank several people. Obviously I would like to thank my supervisor, Dr Allan Jones, for suggesting the topic that has dominated my life for the last half year, as well as providing valuable advice and moral support. Also of note is the university library, who managed to negotiate an extended inter-library loan of "Cephalopods of the World" for me. I would also like to thank a fellow student, Peter Rendle, for posing very interesting questions, as well as good, in-depth discussion regarding the subject.

REFERENCES

Anderson F. E. 2000, Phylogenetic relationships among loliginid squids (Cephalopoda: Myopsidia) based on analyses of multiple data sets. Zoological Journal of the Linnean Society 130: 603-633

Bolstad K. S., O'Shea S. 2004, Gut contents of a giant squid Architeuthis dux (cephalopoda: Oegopsida) from New Zealand waters. New Zealand Journal of Zoology 31: 15-21

Chase r., Wells M. J. 1986, Chemotactic behaviour in Octopus. Journal of Comparative Physiology 158/3: 375-381

Christensen W. K. 2002, Fusiteuthis polonica, a rare and unusual belemnite from the Maastrichtian. Acta Palaeontologica Polonica 47: 679-283

Dzik J. 1980, Origin of the cephalopoda. Acta Palaeontologica Polonica 26: 161-191
Engeser T. S., Clarke M. R. 1988, Cephalopod hook, both recent and fossil. The Mollusca 12: 133-151

Fiorito G., Gherardi F. 1999, Prey-handling of Octopus vulgaris (Mollusca, Cephalopoda) on Bivalve preys. Bejhavioural Processes 46: 75-88

Förch E. C., Uozumi Y. 1990, Discovery of a specimen of Lycoteithis lorigera (Steenstrup, 1875) (Cephalopoda: Teuthiodea) from New Zealand and Additional notes on its morphology. New Zealand Journal of Marine and Freshwater Research 24: 251-258

Jackson G. D., O'Shea S. 2003, Unique hooks in the male scaled squid Lepidoteuthis grimaldii Joubin 1895. Unpublished but available through the TONMO website

Johnsen S., Balser E. J., Fisher E. C., Widder E. A. 1999, Bioluminescence in the deep-sea cirrate octopod Stauroteuthis syrtensis (Mollusca: Cephalopoda). Biological Bulletin 197: 26-39

Keir W. M., Smith A. M. 1990, The morphology and mechanics of octopus suckers. Biological Bulletin 178: 126-136

Keir W. M., Smith A. M. 2002, The structure and adhesive mechanism of octopus suckers. Integrated Comparative Biology 42: 1146-1153

Kier W. M., Van Leeuwen J. L. 1997, A kinematic analysis of tentacle extension in the squid Loligo pealei. The Journal of Experimental Biology 200: 41-53

Lindgren A. R., Giribet G., Nishiguchi M. K. 2004, A combined approach to the phylogeny of Cephalopoda (Mollusca). Cladistics 20: 454-486

Lipinski M. R. 2001, Preliminary description of two new species of cephalopods (Cephalopoda: Brachioteuthidae) from South Atlantic and Antarctic waters. Bulletin of the Sea Fisheries Institute 1 (152) : 3-14

Lukeneder A. 2005, First nearly complete skeleton of the Cretaceous duvaliid belemnite Conobelus. Acta Geologica Polonica 2: 147-162

Muntz W. R. A., Wentworth S. L. 1995, Structure of the adhesive surface of the digital tentacles of Nautilus pompilius. Journal of the Marine Boilogical Association of the United Kindom 75 (3): 747-750

Nixon M., Dilly P. N. 1977, Sucker surfaces and prey capture. Symposium, Zoological Society, London 38: 447-511

Robison B. H., Young R. E. 1981, Bioluminescence in pelagic octopods. Pac. Sci. 35: 39-44

Smith A. M. 1996, Cephalopod sucker design and the physical limits to negative pressure. The Journal of Experimental Biology 199: 949-958

Vecchione M., Galbraith J. 2001, Cephalopod species collected by deepwater exploratory fishing off New England. Fisheries Research 51: 385-391

Nesis K. 1987, Cephalopods of the World. Originally published in Russia, 1982, this translated edition T.F.H Publications ISBN 0-86622-051-8

Tree of Life web project: www.tolweb.org/tree

Cephbase: www.cephbase.utmb.edu

Waikiki Aquarium website: www.waquarium.org

The Cephalopod Page: www.thecephalopodpage.org

Paleos: www.palaeos.com

All sites are operational as of July 2007.