Home About Us Advice & Welfare Get Involved News and Events Blog Contact Us
Aquarium Welfare Association - Working together with: Follow us: Info@aquariumwelfare.org Join Us!
Keeping cephalopods in captivity Dr Gavan Cooke

There are approximately 700 extant cephalopods [3], ranging in size from a few millimetres to more than 20 meters. Of course you find far larger cetaceans, most of which are arguably too big for captivity, requiring tanks prohibitively large. The smallest cephalopod (Idiosepius notoides, the pygmy squid), however, might intuitively do very well in a controlled captive environment. In fact most cephalopods do not get larger than 2000 mm (mantle length), which is smaller than many captive sharks. Does that lead, therefore, to the statement that all cephalopods would do well in captivity? No. Tank size is but one facet of captive life. In order to survive, never mind thrive, animals require other things, like a balanced diet suited to them. Of the species we know, all but one (Vampyroteuthis infernalis, the fantastically misnamed vampire squid, which eats marine snow and isn’t a squid at all) are thought to be obligate predators. A very small number of species (e.g. some cuttlefish and octopus) have been trained to eat dead food. These adaptable few exhibit far fewer of their natural behaviours [4] (I’ll come on to this more later) when eating prepared foods. Presently we must assume that the majority will starve without the stimuli to initiate feeding behaviours that only live prey can provide. To compound matters, the cephalopods that do respond to dead foods rarely do so before they are trained or reach a certain age – there is no immediate innate acceptance of dead foods as you find with many captive fish predatory species e.g. some haplochromine cichlids. This means that at some stage live food is a must even for them and the costs in time, money and knowledge associated with providing live food can be immense. In fact, although the cephalopod aquaculture industry is growing rapidly a major bottleneck remains with how to feed them efficiently and effectively [5]. To compound matters In addition, many cephalopods possess a paralarvae stage, which at present no one has any idea how to feed. Cephalopods are now protected in many countries when used in science [6] and a special licence is required , taking a considerably amount of time to obtain, simply to keep animals you suspect may starve to death when born in captivity. For those who do not fall under this legislation I would hope their ethical review process (which public aquaria have but is up to the individual hobbyist to debate within themselves) would prevent cephalopods breeding when you have no way to feed the offspring.

So, to recap, as you can get the right food for the right species, sometimes, is it acceptable to put appropriately sized cephalopods in appropriately sized tanks if they can be fed appropriately?  Again, the answer is no. Cephalopods are found in a wide variety of marine habitats and have a wide variety of marine lifestyles. A small number (e.g. juvenile cuttlefish, bobtail squid and some octopus) spend a good proportion of their lives on or in the substrate [3], hiding from predators and ambushing prey. So these, if the tank was large enough, might do OK for a time in tanks found in universities, public aquaria and even at home. This period of being OK is finite as many migrate vertically/horizontally to feed, breed or become otherwise much more active when they reach sexual maturity. But what about the others, what about squid? Don’t they have huge ranges and swim a lot? Yes, they do. Many true squids (not including the taxonomically misleading bobtail squid – [7]) need very large (long rather than tall) tanks and very high water flow just to exhibit their basic swimming behaviour. Aquaculture facilities and even some universities are able to offer bespoke tank design that would be suitable for some squid. However few public aquaria and probably no amateur hobbyist should attempt to keep many squid in regular aquaria tanks. Nautili are another example of a cephalopod unsuited to anything but well designed (and expensive!) captive environments. They require a decent amount of vertical movement and very tall tanks are not easy to buy and maintain. All cephalopods are very fragile, have no skeleton to protect them and do not always heal from injuries [4]. Many tanks have sharp edges or enrichment that can permanently damage cephalopods. Cuttlefish in particular suffer from what is known as ‘butt burn’ or ‘bubble butt’, which is a bacterial infection after damaging their posterior mantle region. Damage occurs when fleeing a threat and may retard growth and even kill them if the cuttlebone breaks. Cuttlefish are also prolific ‘inkers’ when threatened and these black out episodes put huge pressure on life support systems, may trigger stress in other cephalopods (the ink has been shown to act as an alarm substance in some species [8]) and frequently lead to damage because of severely reduced visibility. Providing live crabs, a staple of many cephalopods, causes huge demands on life support systems too. In fact, dead crabs can be just as bad: providing a crab to a cuttlefish is like sticking a hand grenade up the crabs backside and pulling the pin – the mess is immense! Even dead or alive shrimp, although less smelly after the explosion, still need considerable cleaning up. You might think “Well why not provide live or dead fish?” Live fish will do their best to escape being eaten and it might not be ethical to provide them. Also, fish do not have the proteins/amino acids that cuttlefish need for healthy growth, these are only found in crustaceans [9]. So again few will have sufficient resources to begin to think about keeping cuttlefish even though they are easy to obtain as eggs and thought of as an ‘easy’ cephalopod to keep.

So it seems that tank and husbandry requirements for cephalopods probably exceed those of teleost fish and elasmobranchs, as do their feeding demands. At the beginning of this article I introduced a comparison between keeping cetaceans in captivity with keeping cephalopods in captivity.  In one area of their behaviour a comparison may be more appropriate than any of the others I mentioned: their intelligence. Whilst none would say cephalopod intelligence equals that of cetaceans they are plausibly second in line when comes to aquatic animal intelligence. Most people accept that even the ‘simple’ fish ought to have some environmental enrichment, many argue cetaceans should be excluded from captivity often solely on impossible provision of appropriate stimulation. Cephalopods have recently been shown to have impressive cognitive ability in all manner of metrics: problem solving [10], memory [11]; learning [11] and even tool use [12]. The evidence suggests they might feel pain and even suffer psychologically [5]. But are all cephalopods equally intelligent? Or close to being equally intelligent? This is very difficult to answer. The studies on intelligence, pain and psychological suffering are limited to a few core ‘models’ (e.g. O. vulgaris, S. officinalis being the two most popular). Only one species has been shown to use tools (the veined Octopus, Amphioctopus marginatus), one or two species are possibly what you might consider social (e.g. the Reef Squid, Sepioteuthis sepioidea [13]) and the vast majority we simply have no idea about. Most palaeontologists and taxonomists agree that the nautiluses are very ancient compared with the rest of the cephalopods and not as derived. In fact the now extinct ammonites and belemnites are more closely related to extant non-nautilus cephalopods [14]. Many observations of Nautiluses seem to suggest they are not as intelligence as their more dynamic cousins but this is subjective and we should probably begin looking at how consistent across taxa cephalopod intelligence really is. Either way saying all cephalopods are too intelligent for keeping in captivity is not appropriate, for now, but adopting the ‘precautionary principle’ [15], as we do in science regarding animal welfare, suggests that before we know for sure we have to assume they are intelligent enough for us to be very careful when considering  keeping them in captivity.

One of the best ways, in my experience, of providing enrichment is to provide foods that have to be obtained by the animal themselves i.e. foraging or predating. As mentioned above, cephalopods exhibit a far greater repertoire of behaviours when presented with live food rather than dead food. Unfortunately this raises its own ethical issues – do we subject prey, who cannot escape, to an inevitable and messy death? Cephalopods are thought to nearly always inject their prey with a neurotoxin, which paralyses the prey - we cannot know for sure whether or not the prey feels pain or experiences other suffering. “But wait!” you say “…surely the lowly shore crab cannot suffer?!” Recent research suggests  that decapod crustaceans might feel pain  [16] and can therefore possibly suffer (a debate for another time!) and if they do is it right that one group of animals suffers so another does not? Why not remove both from the equation? Important questions, in my opinion.

Another essential and incredibly enriching behaviour is breeding. Most of us will have seen the overwhelming desire in animals to reproduce, leading to life or death risks simply to pass on their genes. An important but tough-to-answer question is “Do animals that breed only once in their lives before dying have an even stronger desire to procreate than animals that have many chances?” Is it more frustrating for a cuttlefish not to find a mate when it has one chance compared with a fish that breeds seasonally for up to 20 years? Can the latter ‘know’ to be patient whilst the former ‘knows’ it must do all it can before senesce sooner or later stops it? In my experience the severe pacing exhibited by a fully mature cuttlefish male is clearly a stress response, I cannot say for sure it is down to a lack of reproductive opportunities but it sure does coincide in an otherwise very enriched environment. Again, in my experience, I have never seen the same levels of ‘frustration’ in the iteroparous species I have worked with as I have seen in the semelparous ones after the onset of sexual maturity.

Where does that leave us? As a mealy mouthed scientist it gives me the classic get out of not having a definitive answer. Should we keep any cephalopods in captivity? It depends: on space available, on food available, conspecific availability, type of institution (or if you keep them as pets), species, life span, etc. etc. ad nauseam. I will stick my neck out a little bit, I will say nearly all cephalopods are unsuitable in nearly all common captive circumstances. Most public aquaria offer fantastic tanks, have resources for live food and have to offer amazing enrichment if only to entertain the public. They also frequently provide breeding opportunities, if only to maintain their own stocks. It is here where the trade-off between being in captivity and providing the opportunities for studying them (and educating the public) is possibly most acceptable. Universities and aquaculture facilities often have the space and resources (the former also have strictest welfare laws and globally many countries include cephalopods within these laws) but have a historical propensity for sterile environments lacking in enrichment (although this changing for pragmatic reasons). It is at home that concerns me most. Notable aquatic animal dealers are selling more and more cephalopod species, many of which we don’t know how to keep effectively. I see many YouTube videos of pet cephalopods exhibiting stereotypic or other stress related behaviours, which their owners see as interesting or exciting when in fact the animal is distressed. In science there are strict housing requirements for many animal species, it would be very nice if legislation enforced restrictions like this at home too. Much like the ‘Big Fish Campaign’ [17], I would like people to think very hard before keeping any cephalopod, no matter what their size, based on their need for enriching food, high husbandry demands, their intelligence, likelihood of eating some of their tank mates, frustration at not being able to breed and the likelihood of their offspring dying of starvation if they are able to breed. There is huge commercial pressure on public aquaria and aquatic animal dealers to provide new exciting animals that the public hasn’t seen before. It is vital we learn from old mistakes and do so before rushing to keep these charismatic and fascinating animals.


1. http://en.wikipedia.org/wiki/Kingdom_(biology) (accessed 25th July 2015)

2. http://grammarist.com/usage/octopi-octopuses/(accessed 25th July 2015)

3. Hanlon, R.T., and Messenger, J.B. (1996). Cephalopod Behaviour. Cambridge University Press, Cambridge.

4. Tonkins, B.M, Tyers, A., M & Cooke, G.M, (2015)  Cuttlefish in captivity: an investigation into housing and husbandry for improving welfare. Accepted in Appl. Anim. Behav. Sci. April 2015. DOI: http://dx.doi.org/10.1016/j.applanim.2015.04.004

5. Sykes, A.V., Baptista, F.D., Gonc¸ alves, R.A., and Andrade, J.P., (2012) Directive 2010/63/EU on animal welfare: a review on the existing scientific knowledge and implications in cephalopod aquaculture research. Rev. Aquac. 4, 142–162

6. Andrews, P.L.R., Darmaillacq, A., Dennison, N., Gleadall, I.G., Hawkins, P., Messenger, J.B., Osorio, D., Smith, V.J., and Smith, J.A. ( 2013). The identification and management of pain, suffering and distress in cephalopods, including anaesthesia, analgesia and humane killing. J. Exp. Mar. Biol. Ecol. 447, 46–64

7. http://en.wikipedia.org/wiki/Bobtail_squid (accessed 25th July 2015)

8. Lucero, M.T., Farrington, H., and Gilly, W.F. (1994)  Quantification of l-dopa and dopamine in squid ink: implications for chemoreception. Biol. Bull. 187, 55–63

9. Domingues P, Sykes A, Sommerfield A, Almansa E, Lorenzo A,. Andrade  2004, Growth and survival of cuttlefish (Sepia officinalis) of different ages fed crustaceans and fish. Effects of frozen and live prey, Aquaculture, Volume 229, Issues 1–4, 12 pp 239-254, ISSN 0044-8486, http://dx.doi.org/10.1016/S0044-8486(03)00351-X

10. Fiorito G, von Planta C, Scotto P. 1990 Problem solving ability of Octopus vulgaris Lamarck (Mollusca, Cephalopoda). Behav Neural Biol. 53(2):217-30.

11. Brown ER1, Piscopo S. 2013Synaptic plasticity in cephalopods; more than just learning and memory? Invert Neurosci. 13(1):35-44. doi: 10.1007/s10158-013-0150-4.

12. Julian K. Finn, Tom Tregenza, Mark D. Norman, Defensive tool use in a coconut-carrying octopus, Current Biology, Volume 19, Issue 23, 15 December 2009, Pages R1069-R1070, ISSN 0960-9822, http://dx.doi.org/10.1016/j.cub.2009.10.052.

13. Byrne, R.A., U. Griebel, J.B. Wood & J.A. Mather 2003. Squids say it with skin: a graphic model for skin displays in Caribbean Reef Squid. Berliner Geowissenschaftliche Abhandlungen 3: 29-35.

14. Lindgren, A.R., G. Giribet, and M.K. Nishiguchi. 2004. A combined approach to the phylogeny of Cephalopoda (Mollusca). Cladistics 20(5):454-486.

15. http://en.wikipedia.org/wiki/Precautionary_principle

16. Stuart Barr, Peter R. Laming, Jaimie T.A. Dick, Robert W. Elwood, Nociception or pain in a decapod crustacean?, Animal Behaviour, Volume 75, Issue 3, March 2008, Pages 745-751, ISSN 0003-3472, http://dx.doi.org/10.1016/j.anbehav.2007.07.004.

17. http://www.bigfishcampaign.org/

I was recently asked whether it is right to keep cephalopods in captivity. At first this appears to be a reasonably straight forward question, similar to the topical question of whether or not we should keep cetaceans in captivity.  Cetacea is one “order” of the mammalian “class” [1]. It is a sensible, albeit artificial,  narrow taxonomic group; despite having a range of sizes , all cetaceans have the same specific body plan,  share a large number of other characteristics (apart from a range of feeding requirements and reproductive apparatus) and may have more or less of similar intelligence throughout the group (although this yet to be tested).Cephalopods, however,  are a higher taxonomic rank (being a ‘Class’, like Mammalia) and as such have a far greater degree of diversity. I do not think many would say the nautilus is as similar to an octopus as a porpoise is to a blue whale (ignoring the huge size differences, the morphology is arguably very similar). Furthermore, the octopuses [2] (Order Octopoda, the same taxonomic rank as Cetacea) form a remarkably diverse group with many weird and wonderful forms being discovered all the time. My point is that it’s easier to speak for all cetaeca in the context of captive welfare than it is for cephalopods.

Octopus enrichment. Credit: Sarah Borg