Journal of Experimental Marine Biology and Ecology 447 (2013) 4664

Contents lists available at SciVerse ScienceDirect

Journal of Experimental Marine Biology and Ecology

journal homepage: www.elsevier.com/locate/jembe

Special issue: Cephalopod Biology

The identication and management of pain, suffering and distress in cephalopods,

including anaesthesia, analgesia and humane killing ,
Paul L.R. Andrews a, b,, Anne-Sophie Darmaillacq c, Ngaire Dennison d, Ian G. Gleadall e, Penny Hawkins f,
John B. Messenger g, Daniel Osorio h, Valerie J. Smith i, Jane A. Smith j
a
Division of Biomedical Sciences, St George's University of London, Cranmer Terrace, London, SW17 0RE, UK
b
Stazione Zoologica Anton Dohrn, Villa Comunale, 80121, Napoli, Italy
c
Groupe Memoire et Plasticite comportamentale, University de Caen Basse-Normandie, F-14032, Caen cedex, France
d
Home Ofce, Animals in Science Regulation Unit, Dundee, DD1 9WW, Scotland, UK
e
Department of Applied Biosciences, Faculty of Agriculture, Tohoku University, Amamiya 1-1, Sendai, 981-8555, Japan
f
RSPCA Research Animals Department, Southwater, RH13 9RS, West Sussex, UK
g
University of Cambridge, Department of Zoology, Cambridge, CB2 1TN, UK
h
School of Life Sciences, University of Sussex, Brighton, BN1 9QG, UK
i
Scottish Oceans Institute, Gatty Marine Laboratory, School of Biology, University of St. Andrews, St. Andrews, Fife, KY16 8LB, Scotland, UK
j
The Boyd Group, Hereford, UK

a r t i c l e i n f o a b s t r a c t

Keywords: From January 2013 scientic projects involving cephalopods became regulated by Directive 2010/63/EU, but at
Anaesthesia present there is little guidance specically for cephalopods on a number of key requirements of the Directive,
Analgesia including: recognition of pain, suffering and distress and implementation of humane end-points; anaesthesia
Cuttlesh and analgesia, and humane killing. This paper critically reviews these key areas prior to the development of guide-
Nociception lines and makes recommendations, including identifying topics for further research. In particular: a) Evidence on
Octopus
how cephalopods might experience pain is reviewed; and a draft scheme of behavioural and physiological criteria
Pain
for recognising and assessing pain, suffering and distress in cephalopods used in scientic procedures is presented
and discussed. b) Agents and protocols currently used for general anaesthesia and analgesia are evaluated. Magne-
sium chloride, ethanol and clove oil are the most frequently used agents, but their efcacy and potential for induc-
tion of aversion need to be systematically investigated, according to the species of cephalopod and factors such as
body weight, sex and water temperature. Means of sedating animals prior to anaesthesia should be investigated.
Criteria for assessing depth of anaesthesia, including depression of ventilation, decrease in chromatophore tone
(paling), reduced arm activity, tone and sucker adhesiveness, loss of normal posture and righting reex, and loss
of response to a noxious stimulus, are discussed. c) Analgesia should be provided for cephalopods used in scientic
procedures, whenever this would be the case for vertebrates. However, research is needed to evaluate effective
agents and administration routes for cephalopods. d) Techniques for local anaesthesia need to be dened and eval-
uated. e) Currently used methods of killing and criteria for conrmation of death in cephalopods are evaluated.
Based on present knowledge, a protocol for humane killing of cephalopods is proposed. However, further evalua-
tion is needed, along with development of humane methods of killing that will not compromise study of the brain.
On humane grounds: i. mechanical (as opposed to chemical) methods of killing should not be used on conscious
cephalopods (unless specically authorised by the national competent authority); and ii. hatchlings and larvae
should be killed by overdose of anaesthetic and not by immersion in tissue xative.
Key gaps in current knowledge are also highlighted, so as to encourage research that will contribute to the
evidence base needed to develop guidelines to the Directive.
2013 Elsevier B.V. All rights reserved.

This article is part of a special issue on Cephalopod Biology published under the auspices of CephRes-ONLUS (www.cephalopodresearch.org).
All authors made an equivalent contribution to the manuscript.
Corresponding author at: Division of Biomedical Science, St George's University of London, Cranmer Terrace, London, SW17 0RE, UK.
E-mail address: pandrews@sgul.ac.uk (P.L.R. Andrews).

0022-0981/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jembe.2013.02.010
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
1. Introduction
There are strong ethical, legal and scientic obligations to avoid, rec-
ognise, and alleviate any pain, suffering and distress caused to animals
used in scientic procedures.
The ethical obligation is reected in legal requirements to minimise
suffering, such as those enshrined in EU Directive 2010/63/EU
(European Parliament and Council of the European Union, 2010). This
new law replaces EU Directive 86/609 and, for the rst time, includes
all live cephalopods within its scope (in addition to vertebrates).
The scientic imperative to reduce animal suffering arises because
adverse effects such as pain and distress are likely to confound experi-
mental results, especially if they are unrecognised. There are many stud-
ies showing that physiological responses to discomfort, pain, stress and
distress can affect the quality of data obtained in studies involving verte-
brates (e.g.: Hurst and West, 2010; Karelina et al., 2009; Roedel et al.,
2006) and this is highly likely to be the case in cephalopod studies too.
Similar considerations apply when animals are humanely killed to
obtain tissue for in vitro studies where suffering should be minimised
on ethical, legal and also scientic grounds, as the physiological con-
dition of the tissue may be affected by suffering around the time of
killing.
A degree of suffering may be unavoidable in the experimental proto-
col (e.g. surgery, administration of substances or adverse reactions in
toxicology procedures), or may itself be the object of the study, for ex-
ample in some aspects of psychology or pain research. Whatever the
cause, suffering should be recognised and assessed in order that it can
be minimised at all times and so that pre-set humane end points
can be implemented (see below).
This paper reviews some of the evidence on whether cephalopods
experience suffering, proposes an objective approach for assessing
suffering in an experimental setting, and reviews approaches to anaes-
thesia, analgesia and humane killing.
It should be emphasised that there is still much work to be done in all
of the areas discussed in this paper. We anticipate that implementation of
Directive 2010/63/EU will stimulate research to address some of the ques-
tions raised in relation to pain, suffering, distress, anaesthesia and eutha-
nasia as well as questions discussed by Smith et al. (2013this volume).
2. Evidence of the capacity for cephalopods to experience pain
A review of the published literature for evidence of nociceptors and
for the perception of pain in cephalopods can be summarised as absence
of evidence rather than evidence of absence. As pointed out by Smith et
al. (2013this volume), the time for debate about the rationale for includ-
ing cephalopods within the scope of EU legislation on animal experi-
ments has passed and the new law assumes that these animals can, and
do, experience pain, suffering and distress. Nevertheless, there is a pauci-
ty of robust data in many key areas relevant to suffering in cephalopods,
and so there is merit in examining the arguments put forward by the Sci-
entic Panel on Animal Health and Welfare (AHAW) on the Revision of the
Directive (EFSA Panel on Animal Health and Welfare, 2005), which advo-
cated inclusion of cephalopods in the legislation, and other published lit-
erature discussing this topic in order to highlight the major knowledge
gaps that limit the ability to make objective assessments about some as-
pects of cephalopod welfare.
It must be remembered that there are many causes of suffering, in-
cluding stress, distress, hunger, fear, anxiety, sensory deprivation and
frustration (e.g. resulting from physical connement), as well as pain
(Andrews, 2011). However, in this section we focus on pain, for two rea-
sons. First, the capacity of cephalopods to experience psychological dis-
tress has not been evaluated to a signicant extent. Second, AHAW
(EFSA Panel on Animal Health and Welfare, 2005) regarded pain as a
particularly important form of suffering, which is pivotal to many discus-
sions of welfare, and pain caused by animal experiments using inverte-
brates is an area of longstanding concern and controversy (Fiorito, 1986).
47
In considering evidence for an animal's capacity to experience
pain, it is important to recognise that there are two components to
pain responses: i. nociception (the physiological detection and response
to painful stimuli); and ii. the experience of pain (a conscious, emotional
sensation, which, in humans, is mediated by the highest part of the
brain, the cerebrum); and that nociceptive (reex) responses can
occur without any conscious sensation of pain.
A Working Party of the Institute of Medical Ethics (Smith et al.,
1991) identies seven criteria that, taken together, might provide evi-
dence for pain experience in animals. The seven criteria are listed
below, and we discuss evidence relevant to cephalopods against each
item in turn (taking criteria 6 and 7 together). In essence, the criteria
cover: possession of appropriate nervous receptors and pathways for
nociception; involvement of brain centres that could have the capacity
to generate pain sensation, with the possibility of modulation by endog-
enous and/or exogenous opioids; behavioural responses to noxious
stimuli that are functionally similar to those of humans; and evidence
of learning in relation to painful stimuli. Note that the AHAW report
(EFSA Panel on Animal Health and Welfare, 2005) utilised these criteria
in a slightly modied form.
It should be noted that over the last 10 years the sh research
community has dealt with a number of the issues now confronted
by the cephalopod research community. For example, the sensory ex-
perience of sh exposed to a noxious stimulus is contentious (for con-
trasting views see: Braithwaite and Boulcott, 2007; Rose, 2007), but
in contrast to cephalopods there is good neurophysiological evidence
for the existence of nociceptors in sh (Mettam et al., 2012; Sneddon,
2009; Sneddon et al., 2003).
2.1. Criterion 1: Possession of receptors sensitive to noxious stimuli, located
in functionally useful positions on or in the body and connected by nervous
pathways to the lower parts of the nervous system
2.1.1. Presence in cephalopods: likely but not proven
The International Association for the Study of Pain (IASP) denes a
nociceptor as a high-threshold sensory receptor of the peripheral so-
matosensory nervous system that is capable of transducing and
encoding noxious stimuli (IASP Task Force on Taxonomy, 2011). A
noxious stimulus is one which is potentially or actually damaging to
the organism, but note that the denition implies nothing about the
conscious sensation (what the animal may feel) resulting from ac-
tivation of a nociceptor. Nociceptors are usually identied by neuro-
physiological studies, but indirect evidence may also come from
ultrastructural studies (e.g. unspecialised afferent nerve terminals
may be a characteristic of nociceptors) and behavioural studies
(e.g. withdrawal of the animal in response to a noxious stimulus,
possibly combined with a blunting of the response by an analgesic).
A start has been made in identifying molecular signatures for nociceptors
(e.g. TRPV1, ASICs; Smith and Lewin, 2009) and this may be an approach
to resolving this question for cephalopods. An understanding of the
properties of nociceptors (mechanical threshold, chemical sensitivity,
capacity for sensitisation) in cephalopods will contribute to assessment
of the severity classication of experimental procedures.
There is good neurophysiological and behavioural evidence for
nociceptors sensitive to mechanical stimulation in the mollusc Aplysia
californica (Illich and Walters, 1997). There is also evidence for
nociceptors in the land snail Cepaea nemoralis (Kavaliers et al., 2000)
and the sea-slug Tritonia diomedia (Getting, 1976). Although there
have been no comparable studies on cephalopods, from a phylogenetic
perspective this would make it likely that they possess a nociceptive
system. The evidence that a receptor structure and its associated affer-
ent axon is a nociceptor is essentially based on characterisation of
the afferent discharge in response to graded stimuli. The EFSA (2005)
report concluded that cephalopods have nociceptors in their skin
(Section 2.4.5, p. 34 in the report) citing Wells (1978) to support this
(but see below).
48 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
In the words of Broom Cephalopods have nociceptors that respond
maximally to an injurious stimulus but not to an innocuous stimulus,
and increase in sensitivity after tissue has been injured to help the animal
avoid further injury. There is also some indication that the nociceptor rate
of ring or sensitivity is related to the sensitivity of the tissue that they
protect (Mather, 2004). (Broom, 2007, p. 102) This quotation implies
that the afferent discharge from a nociceptor has been recorded in a ceph-
alopod, but the Mather (2004) paper cited by Broom does not appear to
present such evidence. A recent review of nociception in molluscs
commented that, nociceptors have not yet been described in any ceph-
alopod (Crook and Walters, 2011).
Weak circumstantial evidence for nociceptors comes from electron
microscopy of the sucker of Octopus vulgaris (Graziadei, 1962), demon-
strating a variety of receptor structures, suggesting a diversity of func-
tion. Recordings from peripheral nerves and longitudinal tracts in the
arm of decerebrate O. vulgaris have revealed afferents sensitive to
sucker stimulation, with some results suggestive of the presence of
high threshold receptors (Rowell, 1966). In relation to mechanical
stimulation of the suckers Rowell commented that the units are not es-
pecially sensitive (1966, p. 593), and There is a considerable range of
threshold. Most units were stimulated by gentle stroking of the skin
with a needle. Some did not respond except to blows or pinches of con-
siderable force, enough to produce behavioural response from the other
arms (1966, p. 597). A more recent study recording from isolated nerve
cord of the octopus arm found a few mechanosensory afferents in the
external oblique muscle (Gutfreund et al., 2006), whose properties do
not however resemble nociceptors. Recordings from afferents in the
stellar nerves in Octopus bimaculoides/bimaculatus have revealed bres
sensitive to mechanical stimulation of the skin but the threshold was
not determined (Gray, 1960). Boyle (1976) reported afferents in a
at mantle preparation (O. vulgaris and Eledone cirrhosa) responsive
to either light touch or considerable mechanical force (pinching
the mantle). It is reported that cephalopods have a large number of
free nerve endings in the skin (Bullock, 1965), and this is the type of
afferent terminal typical of nociceptors in vertebrates.
Taken together these studies provide some circumstantial evidence
for afferents in octopus suckers, skin and mantle with a high threshold
consistent with nociception. However, it must be emphasised that this
requires further investigation combining afferent nerve recording
with carefully quantied noxious stimuli (mechanical, chemical, ther-
mal) in parallel with behavioural studies using the same stimulus inten-
sities. Recently, it has been demonstrated using immunohistochemistry
that peptide transmitter FMRFamide is present in primary sensory
neurones in the skin and gills of several cephalopod species, i.e.: Loligo
opalescens, Octopus rubescens, and O. vulgaris (Mackie, 2008). Combin-
ing FMRFamide immunohistochemistry with afferent nerve record-
ings may enable the morphological correlate of functionally dened
nociceptors to be identied.
A particular problem in dening a nociceptive system in
Octopoda is the complex nature of the nervous system in the arms,
which in O. vulgaris contain more than 60% of the total number of
neurones (Young, 1963). Rowell (1963) showed that isolated arms of
O. vulgaris have reex responses at the individual sucker and the
whole arm levels. Of particular relevance is the observation that
pricking the skin of the arm with a needle evoked local inching of
the skin and movement of the whole arm withdrawing it from the
stimulus. Rowell also reported that a painful stimulus (needle prick
or electric shock) to a sucker caused immediate withdrawal and, if con-
tinued, there was a general movement of the arm away from it. Using
a similar preparation Hague et al. (2011, 2013) demonstrated reex
withdrawal of an isolated arm on exposure of the tip (~1 cm) to acetic
acid (15%), hypotonic tap water and a strong pinch with forceps. Al-
though the stimulus was applied at the tip, the arm withdrew by con-
traction of the proximal muscles forming a joint in the proximal arm
as is observed in intact animals during withdrawal of the arm (Sumbre
et al., 2005). These studies support the existence of nociceptors in
cephalopods but also illustrate the difculty in identifying behav-
ioural indices of nociceptor stimulation in the whole animal.
According to Messenger (1973), cuttlesh learn to stop attacking a
prawn enclosed in a transparent tube, suggesting that this response is
due to pain induced by the forcible impact of the tentacular club with
the tube; but more recent alternative interpretations have been provided
(Agin et al., 2006). This is a potentially important observation as the me-
chanical stimulus is one which is likely to be encountered in the wild, in
contrast to the electrical shock more frequently used in these older
studies. Currently, positive reinforcement (reward) techniques would
be preferred to the use of aversive stimuli on ethical and welfare grounds.
2.2. Criterion 2: Possession of higher brain centres [in the sense of integration
of brain processing], especially a structure analogous to the human cerebral
cortex
2.2.1. Presence in cephalopods: present
There is no dispute that cephalopods have a complex brain structure
in comparison with other invertebrates (for review see for example Zullo
and Hochner, 2011). The organisation into lobes is unique amongst in-
vertebrates with evidence for a hierarchical organisation (e.g. sub- and
supra-oesophageal lobes) cf. vertebrates. A considerable body of data is
available comparing the brains of different cephalopod species (Nixon
and Young, 2003), which could give some insight into different cognitive
abilities. Studies of the roles of the vertical and frontal (inferior and supe-
rior) lobes in learning and memory have led to the proposal that they
may serve similar functions to the mammalian cerebral cortex (Nixon
and Young, 2003). More recently it has been argued that they could
also be involved in consciousness (Edelman and Seth, 2009).
Higher cognitive abilities are considered important for the presence
of sentience and the capacity to experience pain, suffering and distress.
Edelman and Seth (2009) make a general case for cephalopod con-
sciousness based upon behavioural and brain complexities, for which
there is a wide range of evidence (for review see, Borrelli and Fiorito,
2008), including: complex memory, discrimination and generalisation,
social learning, conditioned suppression, reversal learning, spatial
awareness, and operant conditioning (e.g.: Darmaillacq et al., 2004;
Fiorito et al., 1998; Fiorito and Chichery, 1995; Fiorito and Scotto,
1992; Karson et al., 2003; Mather, 1995; Robertson et al., 1995).
Note that evidence is from a variety of cephalopod species, in partic-
ular O. vulgaris and Sepia ofcinalis, but little is known about the relative
abilities of different species in each category. Moreover, care must be
taken in drawing comparisons between cephalopod and vertebrate
brains, as the last common ancestor of vertebrates and cephalopods
existed over 500 million years ago.
2.3. Criterion 3: Possession of nervous pathways connecting the nociceptive
system to the higher brain centres
2.3.1. Presence in cephalopods: uncertain
There is a considerable body of evidence describing the central pro-
jections of peripheral afferent pathways in cephalopods in general and
O. vulgaris and S. ofcinalis in particular (Nixon and Young, 2003;
Young, 1971). The pathways exist by which sensory afferent signals
from the arms, mantle and viscera could reach the higher levels of
the nervous system via relays. In fact, the literature contains several
diagrams showing the pathways by which signals resulting from an as-
sumed painful stimulus (electric shock, but see above) applied to an
arm could reach the lower and higher lobes of the brain, and result
in both simple reex and complex behavioural responses (see for exam-
ple Figs. 10 and 11 in Young, 1966). Caution needs to be exerted as
many of the diagrams where a pathway is labelled pain qualify this
either by putting pain in quotation marks or adding a question mark
(as in Fig. 12.22 and legend of Fig. 12.33 in Wells, 1978). In describing
the major inputs to the vertical lobe Wells comments that one of
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
these is generally assumed to signal pain though there is no proof of
this and it might well carry pleasure or any other signal (1978, p. 364).
In view of the strong evidence for the involvement of the vertical
and frontal (superior and inferior) lobes in memory, probably involving
a process analogous to Long Term Potentiation (Shomrat et al., 2008),
which is well characterised in the mammalian hippocampus, a key
question is whether information from nociceptors, should they exist,
reaches these lobes. An insight into the processing of an unpleasant
but not painful stimulus arises from the cuttlesh. Using cytochrome
oxidase histochemistry, Darmaillacq (pers. comm.) showed that higher
brain centres (inferior frontal, superior frontal and vertical lobes) could
be involved in short- and long-term memory of a distasteful prey which,
although not painful, is an aversive stimulus (Darmaillacq et al., 2004).
2.4. Criterion 4: Receptors for opioid substances found in the central nervous
system, especially the brain
2.4.1. Presence in cephalopods: probable
The opioid system is highly conserved in evolution (Zhu and
Stefano, 2009), so on phylogenetic grounds alone it appears likely
(but not inevitable) that it exists in cephalopods. However, it should
be noted that the opioid system is not exclusively involved in
nociception. Studies are required to examine the effect of morphine
(as a prototypical opioid) and related substances on the behavioural re-
sponse to a noxious stimulus and to examine whether this effect is
blocked by receptor antagonists such as naloxone. Although the opioid
system is a focal point for studies of pain, there is evidence from other
species for other endogenous modulatory systems (such as steroids
and cannabinoids, see below). A prerequisite for such studies is the
identication of well characterised nociceptive responses.
There is evidence for endogenous opioids, opioid receptors and sensi-
tivity to antagonists of opioid receptors in a range of invertebrate species.
However there is little published evidence to indicate whether or not
cephalopods have opioid or other receptor/transmitter systems (e.g. can-
nabinoid, steroid, prostanoid) that could modulate pain perception.
Enkephalin-like peptides are present in neurones from the
palliovisceral lobe supplying the neurosecretory system of the vena
cava in O. vulgaris (Martin et al., 1979; Martin et al., 1986). Enkephalins
have a potent chronotropic effect on the isolated systemic heart of
O. vulgaris, but the effect was not blocked by naloxone, which calls
the involvement of opioid receptors into question (Voigt et al.,
1981). Of more relevance is a study of the impact of morphine and
met-enkephalin on dopamine release from the brain of Octopus
bimaculatus (Stefano et al., 1981) which shows that this effect is present
in vertical, basal and frontal lobes and, signicantly, is blocked by
naloxone. In addition, there is evidence for a morphine receptor in
the squid axon (Frazier et al., 1973) and more recently Sha and col-
leagues (Sha et al., 2007; Sha et al., 2012) used immunohistochemistry
to demonstrate the presence of met-enkephalin and leu-ENK and
opioid receptors in mantle and arm tissue from Octopus ocellatus.
2.5. Criterion 5: Analgesics modify the animal's response to stimuli that
would be painful for a human
2.5.1. Presence in cephalopods: uncertain
There appear to be no specic studies of analgesic effects in cephalo-
pods, though EFSA (2005) cites a paper by Agnisola et al. (1996) which
suggests that the cold water anaesthetic technique used in O. vulgaris
(Andrews and Tansey, 1981) has analgesic effects. However, whilst
the physiological effects of cold water on Octopus can be debated, the
paper by Agnisola et al. (1996) is not a study of analgesia. In addition,
although AHAW report states that this paper cites several other studies
investigating analgesia in octopuses (EFSA Panel on Animal Health and
Welfare, 2005, p. 31), this does not appear to be the case.
Studies of potential analgesic substances in cephalopods are re-
quired especially because, in contrast to surgery in vertebrates, it has
49
not been routine practice to administer analgesics during surgery and
post-operatively to cephalopods.
The evidence discussed above (Criterion 4) seems sufcient to sup-
port investigations into the analgesic effects of opioids in cephalopods,
but care should be taken to control for general suppressive effects of
non-selective opioids (such as morphine) on behaviour, as opposed to
specic analgesic actions. On this basis, evidence suggesting that mor-
phine has an analgesic effect in shore crabs has recently been challenged
(Barr and Elwood, 2011). Analgesics and local anaesthetics are discussed
further below.
2.6. Criterion 6: An animal's response to stimuli that would be painful for a
human is functionally similar to the human response (that is, the animal
responds so as to avoid or minimise damage to its body) & Criterion 7:
An animal's behavioural response persists and it shows an unwillingness
to resubmit to a painful procedure; the animal can learn to associate
apparently non-painful with apparently painful events
2.6.1. Presence in cephalopods: present
Some caution needs to be exercised when comparing behavioural
responses to stimuli presumed to be painful in cephalopods and verte-
brates. For example, a human nipped on the nger by a crab will with-
draw the hand, but an octopus in a similar situation will usually
continue an attack (depending upon the size of the crab).
For evaluations of behavioural responses of pain to be valid, they
need to be relevant to the species and the threats it may encounter in
its environment. No studies from cephalopods were presented in
discussing this aspect in the AHAW report, although a paper on the
role of nitric oxide in touch learning using electrical shocks in O. vulgaris
is cited in a summary table (Robertson et al., 1995; cited in Table 2, p. 37
of EFSA, 2005).
Evidence suggesting a persistent response to pain comes from a
study of skin wound healing in E. cirrhosa (Polglase et al., 1983) which
reported that the animal covered a wound site with an arm, or stroked
it with the tip of an arm. Similar arm placing behaviour has been ob-
served in octopus following surgery to the optic capsule or the cranium
(I. Gleadall, pers. obs.). Further studies are required to try to understand
the functional role of guarding type behaviours as they may be useful
as criteria to assist in recognising suffering.
A recent study of arm injury in the squid Loligo pealei (Crook et al.,
2011) noted that animals did not exhibit wound-directed behaviour al-
though there was evidence of nociceptive sensitization (see below).
However, the presence of guarding is not an unequivocal indicator of
pain. Placing the arm on the wound also allows secreted antimicrobial
peptides (AMPs) from the skin surface of the guarding arm to disinfect
the wound on the damaged arm. Many aquatic animals (such as sh,
amphibians and ascidians) secrete copious amounts of AMP-enriched
mucus from and onto the body surface and the octopus seems to be
no exception (Malham et al., 1998; Nair et al., 2011). Self-disinfection
is also common in organisms that can manipulate limbs or appendages,
e.g. the licking of wounds by cats and dogs delivers lysozyme and AMPs
directly to the site of injury. A guarding cephalopod would have to be
aware that a wound was present, but the level of discomfort or pain
necessary to motivate this behaviour is not known. This highlights the
importance of a good level of understanding of cephalopod biology
when interpreting behaviour.
There is ample evidence that cephalopods can learn to associate
aversive with non-aversive stimuli, especially from studies of learning
and memory using an electrical shock as a negative reinforcer. Note
that such shocks are assumed to cause pain, but responses might not
be mediated by nociceptors (or could involve a variety of receptors)
and hence might be a different kind of aversive sensation, when com-
pared with pain experience in humans.
Behaviour in the eld also indicates that cephalopods learn to avoid
painful events; for example O. vulgaris will adopt cautious behaviour
when attacking hermit crabs carrying a sea anemone or in close
50 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
proximity to anemones (Boycott, 1954). Furthermore, there is ample
evidence that learning plays a role in the predatory behaviour of cuttle-
sh when attacking potentially hazardous prey such as crabs (Hanlon
and Messenger, 1996). The cuttlesh avoid the crabs' claws after
being pinched, eventually acquiring the highly successful method of cir-
cling around to attack the crab from behind. Such learning occurs by
trial-and-error and does not improve by observation (Boal et al.,
2000). Cuttlesh are able also to discriminate between crab species
and to avoid the most aggressive ones again after having been severely
pinched (Portunus sp.; A-S. Darmaillacq, pers. obs.).
Recently, Crook et al. (2011) have provided evidence for protracted
(at least 48 h) tactile sensitization close to the site of an arm wound in
squid. However, the authors are careful to point out that their observa-
tions do not provide evidence either for or against the existence of the
emotional or motivational aspects of pain in cephalopods.
Despite the clear behavioural responses to stimuli likely to induce
acute pain in the eld it is possible that cephalopods have endogenous
anti-nociceptive mechanisms, as in mammals, where transmission of
information in central nervous system pathways activated by noxious
stimuli can be modulated by descending inputs such as those from the
periaqueductal grey matter. Such a mechanism may be advantageous
in suppressing pain incurred when an octopus attacks a crab and is
gripped by the pincers but decides to persist with the attack.
2.7. Conclusions on the experience of pain in cephalopods
On the basis of the currently available evidence, cephalopods possess
(or are likely to possess) a number of the attributes that are believed to
be necessary for the ability to suffer pain. However, reviewing these attri-
butes serves to demonstrate where further research is needed to better
understand how cephalopods might experience pain, where efforts to re-
duce suffering due to procedures might be directed, and to identify good
practice. The following research would be benecial:
a) Investigations into potential nociceptor function, using afferent
nerve recordings in parallel with behavioural studies.
b) Investigation of the brain processing of signals arising from somatic
and visceral afferents (including nociceptors if present). These stud-
ies would identify areas of interest for future studies of anaesthesia
and analgesia.
c) Evaluation of opioid systems, or their equivalent in cephalopods.
d) Tests of putative analgesic agents and evaluation of their efcacy.
e) Characterisation of behavioural responses to noxious stimuli and
exploration of the use of behavioural techniques (e.g. cognitive
bias) to obtain additional information on the possible perception
of pain and other unpleasant sensations.
We note that this research may involve procedures with the potential
to cause pain, suffering and distress, which creates an ethical dilemma
given that the ultimate aim would be to reduce suffering for cephalopods
used in research. It would also benet other cephalopods under human
care, as the resulting information could also be used to rene veterinary
procedures. When making these recommendations, we assume that the
Three Rs (replacement, reduction and renement of procedures involv-
ing animals) would be fully implemented and results disseminated and
used without delay.
3. Objective signs of pain, suffering and distress in cephalopods
Objective indicators that an animal could be experiencing pain,
distress or other suffering are essential if appropriate action is to be
taken to avoid, relieve and reduce suffering and to implement hu-
mane end-points (see below). Hawkins et al. (2011) provide practical
guidance on dening and implementing welfare assessment schemes.
In the experience of many researchers, there are a number of
behavioural and physiological signs that can indicate that something
is wrong with cephalopods. These observations could form the basis
for a scheme to assess and classify the severity of adverse effects ex-
perienced by cephalopods in scientic procedures. Note that when
applying behavioural indices of potential suffering it is not necessary
to make assumptions about what, exactly, the animal might be
thinking or feeling although assumptions are frequently made
about sentience.
Drawing on experience of caring for and using cephalopods in the
laboratory and informed by the relevant literature, it is possible to
make an initial list of behavioural and physiological signs of potential
suffering, as shown in Table 1. The approach to assessing suffering illus-
trated in Table 1 is based on a well-established method used for verte-
brates, rst proposed by Morton and Grifths (1985). The Table is
intended to be a starting point, which acknowledges where some of
the many uncertainties lie, and aims to provoke further discussion
and research into welfare assessment for cephalopods.
3.1. Appearance and behavioural indicators of pain in cephalopods
As with vertebrate species, the key point when assessing and mon-
itoring appearance and behaviour is to understand what is normal for
the particular species and individual animal concerned, as any deviation
from the norm could indicate a welfare problem (see Hawkins et al.,
2011). Where there is doubt about what an individual animal is
experiencing, the precautionary principle should be applied for both
animal welfare and scientic reasons.
The monitoring system should detect the absence of normal, desir-
able, behaviours such as locomotion, foraging or hunting, reproductive
behaviour, territory inspection and social behaviours including commu-
nication. It should also identify the presence of damaging or otherwise
abnormal behaviours, including self-mutilation or injuring behaviours,
stereotypies (see below), withdrawal, anxiety-related behaviours or
aggression (Mellor et al., 2009). These will often be species-specic,
e.g. an apathetic or withdrawn octopus may have arms that resemble
corkscrews or are curled into a ball, whereas cuttlesh and squid may
have their arms hanging loose. Examples of abnormal body positions in-
clude octopus oating mantle-upwards or positive buoyancy (oating)
in cuttlesh. An example of the latter is reported by Agin et al. (2003)
where a toxic intravenous dose of cycloheximide in cuttlesh produced
a state of positive buoyancy that interfered with normal behaviours
such as prey catching.
Reduced feeding behaviour is often an early indicator of incipient
suffering in terrestrial animals, but its signicance is less certain for
cephalopods. For example O. vulgaris will normally resume attacking
crabs within 3075 min after surgery (Agnisola et al., 1996; Shomrat
et al., 2008), whereas Crook et al. (2011) report behavioural alterations
persisting for at least two days following injury to the arm in the squid
L. pealei. Normal changes in food consumption must also be taken into
account also when using this criterion as an indicator of health or wel-
fare in cephalopods; for example female O. vulgaris and Octopus maya
reduce their food intake just before egg laying (Walker et al., 1970).
Stereotypic behaviour, consisting of repetitive actions with no obvi-
ous goal or function, is widely recognised as an indicator that an animal
is, or has been, unable to cope with their environment (Mason, 1991). In
some vertebrates it is recognised that if any animal in a group is engag-
ing in such repetitive behaviour, it should be assumed that the environ-
ment is sub-optimal for all the animals. Some examples of stereotypies
have been reported in cephalopods; for example, cuttlesh may repeat-
edly jet against the side of tank, leading to skin damage, commonly
known butt burn (J. Rundle, pers. comm.). Stereotyped pacing activ-
ity has been reported in Enteroctopus doeini, which stopped when en-
vironmental enrichment was provided (Anderson and Wood, 2001).
3.2. Physiological indicators of pain in cephalopods
Although behaviour will be the primary method by which welfare
is routinely assessed in the laboratory, physiological indices are also
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
important. In some circumstances it may be possible to use physiologi-
cal indicators to detect incipient suffering before overt behavioural
changes occur, and so enable early intervention to prevent further
suffering (e.g. by providing analgesia or implementing a humane
endpoint).
Physiological indicators are often used to help assess welfare in ver-
tebrates (Hawkins et al., 2011) but, to our knowledge, there have been
no studies aimed at evaluating the use of such parameters to detect suf-
fering in cephalopods. However, some studies suggest indicators that
may be worthy of further investigation as markers of suffering in ceph-
alopods, and these are discussed below:
3.2.1. Cardiorespiratory system
There is evidence to suggest that heart rate, and respiration rate and
depth, might be indicators of arousal (and hence possibly welfare) in
cephalopods, but further research is needed.
Cardiovascular responses to exercise, sex and some drugs
(e.g. scopolamine) have been studied in E. doeini and O. vulgaris
using implanted vascular catheters connected directly to recording de-
vices (Fiorito et al., 1998; Johansen and Martin, 1962; Wells, 1978). In an
undisturbed animal the heart rate and pulse pressure appear relatively
stable (see for example Fig. 1 in Fiorito et al., 1998), but Wells (1978)
comments that with prolonged periods of recording in O. vulgaris sponta-
neous changes in pulse pressure occur and the heart may stop beating for
several seconds in the absence of any overt stimulus. He also notes that
the appearance of an observer, or a strong tap on the side of the tank,
causes an animal to miss heart beats, sometimes associated with a
dymantic display (deimatic sensu Hanlon and Messenger, 1996) and
probably linked to body muscle tensing. If this is the case, then missed
beats or bradycardia might be expected in response to a noxious stimu-
lus, as this is likely to induce tensing. However, this conjecture requires
conrmation.
It is likely that ventilation rates will increase when the animal is
aroused, as demonstrated by Boyle (1983) using crab extract, but it
has also been noted that, like heart rate, ventilation frequency and
depth can increase in the absence of any overt stimulus (Boyle, 1983;
Trueman and Packard, 1968). Boyle (1983) describes the relationship
between ventilation rate and activity level in O. vulgaris, underlying
the importance of correlating physiological indicators with behaviour.
Boal and Ni (1996) showed that, in cuttlesh, ventilation rate increased
after general disturbance (being moved to a new tank) and to a lesser
extent in response to visual stimuli such as prey items or conspecics.
Ventilation rate was also altered by chemical cues from conspecics,
potential predators or food (Boal and Golden, 1999).
Ventilation rate can be counted by an observer and depth can be
assessed qualitatively, but the observer should not be visible to the an-
imal as this may cause arousal. Ideally, an automated method should be
developed, building on the work of Boyle (1983) using impedance mea-
surement, but this may only be feasible in a small tank. Oxygen con-
sumption has been used in sh to assess responses to potential
environmental stressors (Folkedal et al., 2012) and might be adopted
for use in cephalopods.
Monitoring such physiological parameters in cephalopods presents
a number of technical difculties. Ideally, physiological measurements
should be made remotely in unrestrained animals (using biotelemetry),
as is often the case in laboratory vertebrates (e.g. Pierce du Sert et al.,
2009). However, application of biotelemetry to cephalopods will re-
quire identication of sites at which small devices can be surgically im-
planted so that they cannot be removed, but also located where they
will not cause the animal distress. Subcutaneous and intraperitoneal
sites are commonly used in vertebrates, and whilst the subcutaneous
site may be an option in cephalopods (but possibly not octopus as
their arms could access any site on the body), there is no equivalent
to the intraperitoneal site. The interior mantle may be an option if
sufciently small devices are available, but this would still be accessible
in O. vulgaris. Acoustic telemetry will be required in order to transmit
51
data from cephalopods because it is not possible to transmit radio
waves through salt water.
3.2.2. Inking, jetting and skin colour and pattern changes
Although inking is one of the most prominent aspects of cephalopod
behaviour, and the circumstances under which it occurs in the wild are
well described (Hanlon and Messenger, 1996), relatively little is known
about its control. In the wild, inking is a component of a defensive trilo-
gy (colour change, inking, jetting) to enable escape from a predator. In
the laboratory, inking may be induced by surprising the animal, han-
dling (especially repeatedly) and immersion in anaesthetics such as
urethane and ethanol that are presumed to be irritant (see below).
Inking in the laboratory is clearly a response to a stressful stimulus
and could be used as an indication that an experimental procedure is
stressful. However, although inking can be induced by handling and im-
mersion in some anaesthetic agents it is unclear whether inking can be
induced by pain. Electrical stimulation of an arm has frequently been
used in O. vulgaris as an aversive stimulus in studies of learning and
memory but inking is rare (I.G. Gleadall, J.B. Messenger, pers. obs.).
However, escape jetting and inking have been observed following sur-
gical amputation of the distal third of one arm in unanaesthetised
squid (Crook et al., 2011).
The peripheral innervation of the ink sac, its sphincter and associ-
ated adductor muscle have been described in O. vulgaris (Young,
1971), but the central pathways regulating inking have not been in-
vestigated. It seems likely that the decision to ink requires involve-
ment of higher parts of the brain, rather than being a simple
reex (Langridge, 2009; Langridge et al., 2007), and, if so, inking
could be an index of suffering. However, there may also be individual
variation in inking frequency that will need to be taken into account.
Skin colour and pattern changes would provide an ideal indication
of pain or suffering but species specic changes have not yet been
identied. However, overall skin paling, disruption of normal pattern-
ing and inability to match colour and pattern to background are likely
indications of some underlying problem.
3.2.3. Defaecation
Defaecation in response to stress has been described in a number
of vertebrate species and is commonly used as an indicator of anxiety,
for example in mice (e.g.: Machado et al., 2006; Savignac et al., 2011).
In cephalopods, however, it appears that the normal pattern of
defaecation is unknown, in both the laboratory and eld. Moreover,
the signicance of evoked defaecation as a marker of suffering in ceph-
alopods is not clear, so experimental procedures which appear to evoke
defaecation should be documented. Several authors have commented
that cephalopods will defaecate when placed in anaesthetic solutions,
but the impact of handling alone on defaecation is not documented so
it is difcult to determine what, exactly, has triggered the defaecation.
The use of defaecation as a possible indicator of pain, suffering or
distress will be confounded in situations where animals are not being
fed or are not feeding. Inappetance is a commonly used indicator of pos-
sible pain, distress or other suffering in vertebrates, and it is logical
to consider using this criterion as part of the assessment of suffering
in cephalopods too.
Although the above section focuses on defaecation as an indication
of pain suffering or distress, it is also possible that constipation could
indicate that the animal's welfare is compromised, but requires care-
ful observation to detect.
3.2.4. Vomiting and regurgitation
In vertebrates, vomiting (the forceful ejection of upper gastrointesti-
nal tract contents via the mouth) is used as a sign of suffering or distress
and can be induced by noxious stimuli (Stern et al., 2011). Regurgitation
(the return of undigested food to the mouth from the stomach or
oesophagus, without the forceful ejection associated with vomiting) is
also used as an indicator of suffering in vertebrates. Regurgitation of
52 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664

Table 1
Potential indicators of mild, moderate and severe suffering in cephalopods. The table is based upon the concepts available in Morton and Grifths (1985), and illustrates an approach to
objective routine welfare assessment. Considerable work (including original research) is required to validate the signs indicated and to develop the severity scales. Although the table
refers to cephalopods in general, it will be necessary to develop separate tables for the main laboratory species.

Types of sign1 Mild Moderate Severe

Appearance
Skin colour (e.g. pallor), abnormal patterning ? ? ?
(e.g. excessive/inappropriate ashing
in squid, wandering clouds in Octopus)
or display (deimantic)
Skin texture (e.g. papillae), unexplained ? ? ?
swellings (e.g. granuloma, cyst) and
integrity (e.g. tears, ulcers, cysts)2
Abnormal body posture (e.g. mantle above ? ? ?
the level of the head in an octopus on
the oor of a tank)

Behaviour
Unprovoked Apathetic and/or withdrawn (e.g. refusal to ? ? ?
leave den or refuge area even for food)
Abnormal body position in the tank or the ? ? ?
water column (e.g. inability of cuttlesh
to maintain position near the oor of a tank)
Stereotypic behaviour Occasional Regular Very frequent high % of
animal time
Abnormal motor or locomotor coordination ? ? ?
(e.g. uncoordinated swimming in spinner
cephalopods3)
Grooming and possible guarding behaviour Reduced grooming ? Absent grooming
in Octopus4
Changes in social interactions, for
social species

Feeding Changed urge/speed to attack prey and Reduced urge/increased ? No desire to attack or unable
possibly time to subdue live prey time to subdue live prey
Other provoked Defaecation (e.g. increased defaecation on ? ? ?
handling but could also include
uncontrolled defaecation)
Inking (e.g. when placed in an anaesthetic ? ? ?
solution but could also include
uncontrolled inking5)
Willingness to come out of cover ? ? ?
At rest, preference for smooth substrate6 ? ? ?
Change in response to other external stimuli ? ? ?
(e.g. unresponsive to a gentle tap on the
side of the tank or a shadow passing
over the tank)

Clinical signs
Digestive Food intake Reduced Very reduced Absent
Faecal output (consider in combination Reduced Very reduced Absent
with food intake)7
Vomiting/regurgitation7 Rare Often occurs following Always occurs following
ingestion of food ingestion of food
Ventilation/heart rate Rate, depth and/or coordination of Mild rate change, but ? Slow, laboured, poorly
ventilation8 remaining coordinated coordinated
Heart rate ? ? ?
Blood biomarkers e.g. Increased concentration of ? ? ?
catecholamines and phagocytes9
Body weight10 Reduction in body weight over specic ? ? 20% over one week?
time periods

Notes:
1
All criteria should be assessed in terms of degree of departure from the norm for the particular species and individual animal concerned. All signs should be taken in combi-
nation to assess the overall severity of adverse effects experienced by the animal.
2
Could also indicate infection. See Harms et al. (2006) for granuloma description in cuttlesh and Wodinsky (1977) for head cyst and infection in Octopus hummelincki.
3
See Colmers et al. (1984) for description of spinner cephalopods.
4
For example the type of arm curling behaviour reported in octopus (Borrelli et al., 2006; Mather, 1998) and possible wound guarding behaviour reported by Polglase et al.
(1983).
5
See Andrews and Tansey (1981) for description of inking on exposure to some anaesthetic agents. Note that uncontrolled spontaneous inking and defaecation may also indicate
a loss of sphincter control (see Young, 1971 for description of ink duct and rectal innervation).
6
Anecdotal evidence that senescent or ill octopuses avoid rough substrates.
7
May be very difcult to assess unless the animal is under continuous observation.
8
Measured non-invasively.
9
See Malham et al. (2002).
10
Methods for minimally traumatic weighing need to be developed and the welfare impact of repeated weighing assessed. There is some evidence that repeated weighing in
Octopus vulgaris may reduce growth (Nixon, 1966).
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
stomach contents is reported in the squid Sepiotuethis sepiodea im-
mersed in magnesium sulphate (3% or 4%) but the signicance of
this observation is unclear (Garca-Franco, 1992). Recently captured
E. doeini have been observed to regurgitate food stored in the crop as
a result of net-feeding (I.G. Gleadall, pers. obs.). The possible welfare
implications of regurgitation and/or vomiting require more systematic
investigation in a range of cephalopods species.
3.2.5. Blood biomarkers
There is preliminary evidence that both endocrine and immune
markers of stress in vertebrates increase in situations that are potential-
ly stressful for cephalopods. For example, Malham et al. (2002) showed
an increase in blood noradrenaline and dopamine in E. cirrhosa removed
from sea water for 5 min, and Abbott and Miyan (1995) reported that in
S. ofcinalis, agitation and hypoxia in animals difcult to remove from
the home tank resulted in a marked elevation of white blood cells
when the animals were subsequently anaesthetised.
Obtaining routine blood samples from cephalopods is not easy with-
out prior implantation of indwelling catheters. However, with the pos-
sible exception of the arm veins in Octopoda, there are no convenient
supercial blood vessels in cephalopods. Sampling from the branchial
hearts may be a possibility in octopus (Agnisola et al., 1996) and sam-
pling using chronic indwelling catheters has been used in the large
octopus, E. doeini (Harrison and Martin, 1965). Some experimental
protocols may require catheterization, in which case blood biomarkers
may be monitored without causing additional suffering. However,
catheterizing animals solely for the purpose of monitoring welfare
may introduce harms that outweigh the scientic benets and this
ethical weighing will need to be considered carefully in deciding
whether or not catheterization is justied.
Plasma cortisol has been used an indicator of stress in a number of
species including sh although caution must be exercised in interpreting
the results to differentiate changes resulting from physiological vs psy-
chological stimuli (see Ellis et al., 2012 for discussion) and the same argu-
ments would apply in cephalopods. Cortisol is excreted via the gills in
sh enabling measurement in the water (Zahl et al., 2010) and in mam-
mals it can also be detected in faeces. Both routes may provide non inva-
sive methods for obtaining cortisol measurements in cephalopods.
3.3. Recommendations
a) Further work is needed to dene objective systems for assessing
pain and distress in cephalopods, using both behavioural and
physiological indicators of suffering.
b) The draft scheme set out in Table 1 for recognising and assessing
suffering in cephalopods should be evaluated by scientists and
animal care staff for a range of procedures and situations, so that
it can be augmented and rened according to feedback from users.
c) Research should be undertaken into the feasibility of using heart
rate and respiration rate and depth as indicators of suffering.
d) The potential to use inking, jetting, skin colour and pattern change,
defaecation (increased or decreased) and regurgitation and vomiting
(if present) as indicators of pain, suffering or distress should also be
investigated.
e) Techniques for obtaining routine blood samples, such as indwelling
catheters, should be developed, along with guidelines on when the
use of such techniques is likely to benet the animal.
f) Consideration should be given to measuring biomarkers of animal
welfare in aquarium water or faeces.
4. Humane endpoints in cephalopod studies
Directive 2010/63/EU requires that death as an end-point of a proce-
dure should be avoided as far as possible and that humane end-points
should be used to minimise the duration and intensity of animal
suffering.
53
Humane end-points describe the circumstances in which a proce-
dure will be stopped for animal welfare reasons, so as to limit the
level of suffering experienced by the animal, regardless of whether fur-
ther scientic results could be achieved (for details see Smith et al.,
2013this volume). Humane endpoints should be dened at the project
planning stage and form part of the information that must be provided
in applications for project authorisation under the new Directive (see
Smith et al., 2013this volume for further discussion). When the hu-
mane endpoint is reached, immediate action must be taken to reduce
suffering, such as humanely killing the animal, taking the animal off
the procedure, or administering analgesia. For further information and
guidance, see http://www.humane-endpoints.info/eng/.
Schemes for recognising and assessing suffering, such as that shown
in Table 1, can be used to ensure appropriate implementation of hu-
mane end-points, by selecting in advance specic criteria and associated
levels of severity that those responsible for monitoring the animals
agree will form a stop-point for the scientic procedure, on welfare
grounds.
Where cephalopods are concerned, natural senescence may be a
complicating factor in recognising suffering and implementing humane
end-points. Most species of cephalopod are relatively short-lived. Fe-
males of many species die naturally after their rst egg laying or egg
brooding, and males may die after mating. The process of senescence,
leading to death, can last for months in some species (Anderson et al.,
2002). Signs of senescence may be similar to many of the signs listed
in Table 1 e.g. for octopus: reduced appetite/food intake and associat-
ed weight loss; white skin lesions; undirected/uncoordinated activity
(Anderson et al., 2002) and greying of the eyes (i.e. cataracts that
make it difcult for the animals to capture food, so that they require
hand-feeding).
This raises questions about assessing the severity of procedures
and implementation of humane end-points, such as: will it neces-
sarily be clear whether observed signs are a result of experimental
intervention or due to natural senescence?; what adverse effects
might cephalopods experience as a result of natural senescence?;
and, has the researcher a moral responsibility to implement hu-
mane end-points solely in response to the senescence process (i.e.
should s/he interrupt a natural process by humanely killing the an-
imal, as part of the responsibility towards captive animals in human
care)?
We propose that the precautionary principle be invoked, in
that animals showing signs of senescence should be humanely killed,
unless there is a clear scientic or animal welfare justication for
keeping them alive. However, we also recognise that there is no
easy-to-apply formula for such a decision, and that good communi-
cation will be needed between all parties in deciding appropriate
action involving, e.g. researchers, the designated veterinarian
(see Article 25 in the new Directive), local Animal Welfare Body (Ar-
ticles 26 and 27), competent authority at the policy level, and others.
See Example 3 in Smith et al. (2013this volume) for further discus-
sion regarding senescence.
4.1. Recommendations
a) Humane endpoints should be determined for all studies involving
cephalopods, taking their natural senescence into account wherever
applicable.
b) Effective communication processes should be set up with all those
concerned with overseeing cephalopod use and caring for the
animals, to ensure that humane endpoints are implemented appropri-
ately.
c) The researcher should take responsibility for cephalopod welfare
during the senescence process, ensuring that animals are humanely
killed unless there is scientic or animal welfare justication for
their remaining alive.
54 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
5. General anaesthesia for cephalopods
The new EU Directive 2010/63/EU requires that anaesthesia must be
used during scientic procedures, unless it is judged more traumatic
than the procedure itself and/or is incompatible with the purpose of
the procedure (Article 14).
It may be necessary to anaesthetise an animal for a number of rea-
sons including:
i. performing a minor or major experimental surgical procedure,
e.g. to lesion part of the nervous system surgically (e.g. Andrews
and Tansey, 1983), chemically or electrolytically (e.g. Graindorge
et al., 2008);
ii. implantation of catheters (e.g. Andrews and Tansey, 1981) or re-
cording devices (e.g.: Brown et al., 2006; Zullo et al., 2009);
iii. to treat a veterinary problem (e.g. repair of damaged skin, Harms
et al., 2006);
iv. to facilitate handling for investigation such as repeated ultrasound
measurement of brain size (e.g. Grimaldi et al., 2007);
v. for transport (e.g.: Garca-Franco, 1992; Sen and Tanrikul, 2009);
vi. to facilitate drug administration into the branchial hearts
(e.g. Agnisola et al., 1996);
vii. to sample body uids (e.g. Malham et al., 2002) or tissue
(i.e. biopsy, Polglase et al., 1983).
The depth and duration of anaesthesia required will depend upon
the nature of the procedure, species under study and the individual
animal's ability to clear the anaesthetic agent either by metabolism
(probably by the hepatopancreas) or by excretion (e.g. gills, kidney).
The vast majority of studies to date involving anaesthetising cephalo-
pods have required general anaesthesia for only 15 to 30 min, with
rare examples of longer periods, achieved using repeated exposure
to anaesthetic, as for the case of the squid Doryteuthis pealei (up to
5 h, Mooney et al., 2010).
Techniques for longer lasting anaesthesia would need to be devel-
oped to permit complex surgical procedures to be undertaken with re-
covery or for long-term in vivo physiological studies, particularly of the
brain, where recovery may not be required. In both cases procedures for
monitoring and maintenance of cardiovascular and respiratory system
function in anaesthetised cephalopods and effective analgesia will
need to be developed.
This section reviews commonly used agents claimed to induce
general anaesthesia in cephalopods and procedures for induction,
maintenance, monitoring and recovery from anaesthesia. This topic
has also recently been reviewed by Lewbart and Mosley (2012) and
by Sykes et al. (2012), in relation to aquaculture research.
5.1. Commonly used general anaesthetic agents
A general anaesthetic agent should, at an appropriate dose, render
an animal into a reversible state of unconsciousness and insensibility,
particularly to painful stimuli. This state should last for at least the du-
ration of the period when a potentially painful or distressing procedure,
such as surgery, is performed. The agent should be non-aversive, or
minimally aversive at the appropriate dose rate and there should also
be no memory of the procedure (amnesia). It is important to note that
some substances have only a neuromuscular blocking effect, leading
to muscle relaxation or paralysis. This produces a state that supercially
resembles general anaesthesia in that there is no obvious response to
external stimuli but animals are not under general anaesthesia and
are capable of experiencing pain and distress. For this reason, neuro-
muscular blocking agents (NMBAs) should never be used without ap-
propriate anaesthesia or analgesia. This is reected in Directive 2010/
63/EU, which forbids the use of any drug to stop or restrict [animals]
showing pain without an adequate level of anaesthesia or analgesia
(Article 14.2b).
Systematic study of general anaesthesia in cephalopods is in its in-
fancy, with few investigations of the relative efcacy or potency of can-
didate agents, and a paucity of studies investigating mechanisms of
anaesthesia. There is also little information available on both the criteria
for assessing the depth of general anaesthesia (see below) and the
physiological status of animals in a presumed state of general anaesthe-
sia. Although we use the terms anaesthetic and anaesthesia in this
review we do so with caution in relation to cephalopods and the sub-
stances used until more detailed studies of mechanisms are undertaken.
Investigations of anaesthetic efcacy in cephalopods (see discussion
of criteria for anaesthesia below) have included a diverse range of
agents (sometimes in combination) and species (Abbott et al., 1985;
Andrews and Tansey, 1981; Berk et al., 2009; Estefanell et al., 2011;
Garca-Franco, 1992; Gleadall, 2013this volume; Goncalves et al.,
2012; Ikeda et al., 2009; Joll, 1977; Messenger et al., 1985; Mooney et
al., 2010; Sen and Tanrikul, 2009; Westermann et al., 2002).
The following agents have been utilised: benzocaine (ethyl p-amino-
benzoate); carbon dioxide; chloretone; chloroform; chloral hydrate; cold
sea water; clove oil; ethanol; magnesium chloride; magnesium sulphate;
menthol; metomidate; MS222 (ethyl-m-aminobenzoate, also known
as tricaine mesylate, tricaine methylsulphonate, Metacaine, Finquel);
nicotine sulphate; 2-phenoxyethanol (phenoxetol); propoxate and ure-
thane (ethyl carbamate).
Species in which anaesthetic agents have been investigated include:
Allotethis subulata, Amphioctopus fangsiao, Argonauta argo, D. pealei,
E. cirrhosa, Eledone moschata, E. doeini, Idiosepius notoides, Loligo forbesi,
Loligo vulgaris, Nautilus pompilius, O. vulgaris, Octopus tetricus,
Pinnoctopus variabilis, S. ofcinalis, Sepioteuthis lessoniana, Sepioteuthis
sepioidea and Watasenia scintillans.
The agents investigated are predominantly those used in other
aquatic species, and all have been administered by immersion of the an-
imal. It is therefore surprising that there has been little consideration of
the potential that these substances have for skin irritation, their pH or
the osmolarity of solutions in relation to sea water. As the animal is to
be completely immersed in the anaesthetic solution it is essential that
its chemical properties are considered so that any potential irritant or
damaging effects can be assessed. Studies on isolated arms in O. vulgaris
have shown reex withdrawal in response to acidic and hypotonic solu-
tions applied to the tip (Hague et al., 2011, 2013), but nothing is known
of the sensitivity of the eye or the gills to such solutions. Although there
are only rare reports of eye or skin damage caused by immersion in pu-
tative anaesthetics, the potential for such damage should not be
overlooked (see Gleadall, 2013this volume). For example, Callan
(1940) commented that a solution of MgCl2 stronger than 2.5% pro-
duced severe skin damage in O. vulgaris, but no details are given and in-
vestigators who have exposed the same species to 3.5% MgCl2 and kept
them for at least 5 days post exposure have not reported any skin le-
sions (G. Fiorito, pers. comm.). In general, provided respiration restarts
if it has stopped, recovery as assessed by willingness to take food
is rapid (e.g. for O. vulgaris see Agnisola et al., 1996). However, there
are no longer-term studies of the impact of anaesthesia on behaviour.
From the above extensive list we focus on magnesium chloride
because it is commonly used and relatively well studied and clove
oil because of a growing interest in its usage.
5.1.1. Magnesium chloride
Magnesium chloride appears to be the anaesthetic most frequently
used in octopus, cuttlesh and squid. In some instances it has been also
combined with ethanol as, for example, in S. ofcinalis (Graindorge
et al., 2008) or O. vulgaris (Shomrat et al., 2008). In comparison to
other agents immersion trauma is not reported for magnesium chloride,
and it has a rapid onset of action and very low post-anaesthetic mortal-
ity (Messenger et al., 1985; Mooney et al., 2010). Kier et al. (1989) used
both magnesium chloride and ethanol in their study of n musculature
in S. ofcinalis and commented that magnesium chloride was found
more reliable although criteria were not given for this assessment.
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
The rationale for combining magnesium chloride and ethanol and
the concentrations used are not usually stated in papers. However,
Shomrat et al. (2008) describe a combination of 55 mM MgCl2 (approx.
0.5% w/v) and 1% ethanol which induced deep anaesthesia (criteria not
stated) after around 25 to 45 min, with rapid recovery, and critically did
not block the formation of Long Term Potentiation (LTP) in the vertical
lobe of the brain. Graindorge et al. (2008) used 2% ethanol and 17.5
MgCl2 (equivalent to 1.75%MgCl2 in the nal solution) and commented
that the magnesium chloride was to prevent any muscular contraction
during surgery.
The comment by Graindorge et al. (2008) reects a suspicion that
magnesium chloride is acting as an NMBA or muscle relaxing agent
rather than as a general anaesthetic (Gunkel and Lewbart, 2008). If
true, the use of MgCl2 would be unacceptable on grounds of an animal
welfare. This concern may arise from long standing use of MgCl2 as an
agent to relax marine animals prior to xation (Pantin, 1946). For ex-
ample, Phillips (1979) used isotonic MgCl2 to relax the mantle tentacles
in the prosobranch mollusc Notoacmea scutum. In Aplysia MgCl2 is used
as an anaesthetic acting to inhibit transmitter release at chemical syn-
apses probably via an impact on calcium ion inux (Liao and Walters,
2002). It also reduces axon membrane excitability via effects on sodium
and calcium channels (Baker et al., 1971; Katz and Miledi, 1969; Liao
and Walters, 2002) with injections of isotonic MgCl2 into the tail of
Aplysia being used to block axonal conduction, generator potentials
and synaptic transmission (Walters, 1987).
Messenger et al. (1985) present evidence supporting the action of
magnesium chloride as a general anaesthetic and some direct evidence
that it does not block neuromuscular transmission, using a concentra-
tion and formulation adapted from Pantin (1946), as mentioned
above. However, there is limited concentrationresponse data for
the effects of magnesium chloride in squid (two concentrations,
Garca-Franco, 1992) and no published data for octopus or cuttlesh.
Further studies are required to investigate its effect on brain function
and in particular to ensure that it induces unconsciousness. This will re-
quire use of neurophysiological techniques to demonstrate for example
that the sensory input from lower to higher parts of the brain is re-
duced; as in the case of the action of general anaesthetics on thalamic
relay nuclei in mammals (Angel, 1993) and that somatomotor outow
is reduced. The data presented by Shomrat and coworkers (see supple-
mental data in Shomrat et al., 2010), using vertical lobe slices from
O. vulgaris previously anaesthetised using 55 mM MgCl2-1% ethanol
mixture, showed a reduction in evoked pre- and post-synaptic poten-
tials, providing indirect evidence for a central action of this combina-
tion. Similar studies using MgCl2 alone are now required, as are
proper assessments of other candidate anaesthetics such as ethanol
and clove oil.
5.1.2. Clove oil
Clove oil appears to have sedative/anaesthetic properties in Octopus
minor (Seol et al., 2007) and Sepia elongata (added to ethanol;
Darmaillacq and Shashar, 2008), in which cases it seems to have charac-
teristics similar to magnesium chloride with minimal immersion trau-
ma, relatively rapid induction, good recovery and minimal mortality.
Its use has also been investigated in O. vulgaris, although an anaesthetic
concentration was not identied (Estefanell et al., 2011), so further
studies are required to assess its potential as an anaesthetic agent in
species of octopus, either a single agent or as an adjuvant to magnesium
chloride. However, in D. pealei Mooney et al. (2010) report jetting,
inking, chromatophore ashing and death in 4 min at higher doses of
clove oil, suggesting that responses are both dose- and species-
specic. Additional studies are therefore required to assess the suitabil-
ity and appropriate doses of clove oil for different species of cephalopod.
Consistency should be improved by using pharmaceutical grade clove
oil with a known, constant level of its active ingredient (eugenol), and
none of the impurities or excipients which may be present in the natu-
ral product.
55
5.1.3. Commonly used general anaesthetic agents: nal comments
The varying responses of different cephalopods to candidate anaes-
thetic agents emphasises the need for systematic comparison of these
compounds, as it is conceivable that different general anaesthetic proto-
cols may need to be developed for each species. It will be necessary
to investigate appropriate concentrations with respect to efcacy and
aversiveness; the effect of duration of exposure on depth of anaesthesia
and recovery time; and the duration of anaesthesia once the animal has
been removed from the solution in order to perform surgery. Overall
there is a paucity of such systematically conducted studies of candidate
general anaesthetic agents in cephalopods, and further research will be
needed to evaluate the level of aversion and establish optimal regimes
from both scientic and welfare aspects.
5.2. Induction of anaesthesia in cephalopods
The induction process should be designed to reduce stress and trau-
ma to the animals as far as possible, so as to safeguard animal welfare as
far as possible, and in particular to: reduce the likelihood of violent re-
actions in squid and cuttlesh that could lead to mantle/cuttlebone/
pen damage by collision with the side of the chamber; avoid inking
(most likely to occur in cuttlesh); and help to ensure effective anaes-
thesia, as there is anecdotal evidence suggesting that stressed animals
are more difcult to anaesthetise.
The usual practice is to remove the cephalopod from the home tank,
by hand or using a net, then place it directly in a vessel containing a so-
lution of the anaesthetic. There are, however, several ways in which this
procedure might be rened, so as to reduce adverse effects on the
animals.
It may be less traumatic to carry out the transfer to the anaesthetic
chamber within a sea water lled tank and so avoid removing the ani-
mals from the water (as in Walker et al., 1970). Alternatively, as octopus
and cuttlesh can easily be trained, it may be possible to train them to
enter a transport box. For octopus, it is important that the anaesthetic
chamber is completely lled and has a secure lid, as animals will often
try to escape so delaying the onset of anaesthesia and requiring han-
dling to ensure that the gills are immersed (as it is assumed that anaes-
thetic enters the systemic circulation predominantly via the gills rather
than the skin).
Adverse effects may also be reduced by placing the animals in a
chamber lled with oxygenated sea water (rather than directly into an-
aesthetic), allowing them to acclimatise, and then gradually replacing
the sea water with the anaesthetic solution. The acclimatisation period
will allow the animal to recover from the transfer, and the gradual ex-
posure to anaesthetic should reduce trauma. Note that this procedure
is similar to that used for humane killing of mammals with carbon diox-
ide, in which a chamber is gradually lled with the gas rather than plac-
ing animals directly into a high concentration of CO2.
As a further renement, pre-induction sedation or tranquilisation
could be administered to the animal in its home tank so that it is
transferred to the anaesthetic chamber in a sedated state.
Pre-anaesthetic sedation is used in sh to minimise stress during
general anaesthesia (Zahl et al., 2009; Zahl et al., 2012). Electrosedation
has also been investigated in sh and when compared to chemical
agents produced a smaller rise in cortisol (e.g. Trushenski et al., 2012).
There are no studies of candidate classical sedative agents (such as ke-
tamine) in cephalopods, but Gleadall 2013this volume reports the use
of low concentrations of calcium as a sedative for transporting rey
squid and MgCl2 has been proposed for use as a sedative for shipping
squid (Berk et al., 2009). Further research is needed on appropriate
methods for sedating cephalopods.
5.3. Criteria for assessing depth of anaesthesia in cephalopods
Effective assessment of the depth of anaesthesia is pivotal to the iden-
tication of anaesthetic protocols for cephalopods. It is essential that
56 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
criteria for anaesthesia establish that the animal is actually in a state
of general anaesthesia, and is not just appearing to be anaesthetised
(i.e. unresponsive but conscious). For example, lack of spontaneous loco-
motion and ventilation may be signs of general anaesthesia, but similar
effects are produced by a neuromuscular blocking agent, where the ani-
mal remains conscious but unable to show any motor reaction to noxious
stimuli. Similarly, paling of chromatophores, which occurs in response to
magnesium chloride, could be an indication of a decrease in regional
brain activity associated with a state of general anaesthesia or could be
a direct effect on the chromatophores and their innervation at peripheral
sites. Similarly, loss of tone to the arms and reduced sucker adhesion
(suction or stickiness) could be due to either central or peripheral
effects of a putative anaesthetic agent. An understanding of the pharma-
cological effects of anaesthetic agents is therefore vital in developing
criteria to ensure that they induce anaesthesia and not just motor
paralysis.
It is also desirable and it should be possible to dene different
planes of anaesthesia so that there is early recognition that the effect
is wearing off (becoming light) so that additional doses can be given
as required. In mammals, physiological functions generally return in
the reverse order in which they are lost (Flecknell and Flecknell, 2009).
It is common practice in research involving vertebrates (including
sh, Ross and Ross, 2008) to state the criteria by which general anaes-
thesia was assessed, but this does not appear to be widespread in the
cephalopod literature, in which comments such as the animals were
deeply anaesthetized without any qualication are not uncommon
(e.g.: Piscopo et al., 2007; Shomrat et al., 2008; Zullo et al., 2009).
The effect of temperature on anaesthetic efcacy in different species
is also a relatively neglected topic. For example, ethanol (2% or 3%) has
only a sedative effect on E. doeini at temperatures from 5 to 21 C (with
deeper sedation at warmer temperatures but insufcient for surgery;
Gleadall, 2013this volume). Ethanol appears to be effective on
A. fangsiao (914 C; Gleadall et al., 1993); whilst with O. vulgaris at
2225 C it has a traumatic effect (Andrews and Tansey, 1981). Infor-
mation on the temperature at which animals were anaesthetised is
therefore highly relevant and should be included in publications.
Review of relevant publications suggests a range of candidate
criteria for assessing the depth of anaesthesia in cephalopods, mostly
for species of octopus (Amphioctopus fangasio, O. vulgaris, E. moschata,
P. variabilis [often misidentied as O. minor]). These are briey reviewed
below.
5.3.1. Depression of ventilation (with notes on cardiac function)
Most authors report an initial increase in the frequency of ventila-
tion (as indicated by coordinated mantle and funnel contraction) after
immersion in putative anaesthetic solutions, sometimes accompanied
by powerful mantle contractions that are sufcient to eject solution
via the funnel out of the anaesthetic chamber. This stimulation of venti-
lation may be due to an increase in activity stimulated by the chemical
effects of the anaesthetic, or by the stress associated with handling, or
contact with a possibly irritant or unpleasant solution.
As anaesthesia sets in, the mantle and funnel activity become
uncoordinated in O. vulgaris, and as ventilation slows, an increase in
depth (laboured breathing) is reported before cessation (Andrews
and Tansey, 1981). Seol et al. (2007) report a change in mantle cavity
shape (uneven and elliptical) in O. minor that presumably reects a
change in motor drive to the mantle muscles.
It should be noted that neuromuscular blocking drugs decrease and
then eliminate respiration by paralysis of the muscles, so elimination or
reduction of ventilation is not in itself always a sign of anaesthesia. It is
also worth noting that under anaesthesia the mantle loses muscle tone
and becomes accid, collapsing under its own weight if the animal is re-
moved from water, whereas unanaesthetised animals transiently out of
water maintain their mantle tone and respiratory movements, showing
that the loss of tone is a consequence of the anaesthesia.
Ideally, when breathing is at a low level, the mantle cavity should be
perfused with fresh oxygenated sea water as a support measure, as al-
though gas exchange will continue to occur across the gills, the lack of
mantle movement means that the sea water in the mantle will have
its oxygen depleted whilst the level of excreted carbon dioxide will in-
crease leading to asphyxia.1 This assumes no circulatory arrest (see
below) so that perfusion of tissues continues, albeit with decreasing
oxygen and increasing carbon dioxide partial pressures. For this reason,
Boyle (1991) recommends that a limit of 10 to 20 min should be set for
the duration of procedures in octopus under anaesthesia. There is anec-
dotal evidence that immersion in magnesium chloride also produces
profound bradycardia, or even cardiac arrest, in O. vulgaris but this re-
quires further study. In mammals rapid intravenous infusion of magne-
sium salts can cause cardiac arrest (Close et al., 1996) and is consistent
with the description of magnesium as nature's physiologic calcium
blocker (Iseri and French, 1984).
Few authors comment in detail upon the recovery of ventilation
post anaesthesia except to note how long this takes. A protracted period
(e.g. 30 min) of hypoxia or asphyxia, possibly accompanied by circulatory
stasis, may affect scientic measurements made post-recovery, so any
such impact should be scientically evaluated. There may also be effects
on tissues removed from animals killed by prolonged immersion in an-
aesthetic solutions (Boyle, 1991). O. vulgaris has a remarkable ability to
recover rapidly (e.g. within half an hour) from protracted periods of
apnea. Wells (1978), discussing differences between the oxygen dissoci-
ation curves for octopus blood, cuttlesh and squid comments that these
could be related to the habit of octopuses of living in holes where local
oxygen deciencies may occur.
Cessation of respiration under anaesthesia is undesirable, although
Young (1971) described this as an end point2 for surgical anaesthesia,
and this was also adopted by Andrews and Tansey (1981) and Messen-
ger and coworkers in their studies of magnesium chloride (Messenger
et al., 1985). Joll (1977), using magnesium chloride to anaesthetise
O. tetricus for weighing, comments that cessation of ventilation permitted
the mantle to be fully drained presumably because mantle contractions
were blocked, but does not report criteria for anaesthesia. Seol et al.
(2007) include cessation of breathing (death) as the nal stage in
their list of anaesthetic stages although they appear to have used loss
of sucking intensity (see below) as the end point in their study of
clove oil in O. minor. It should be noted that the latter cites return of
regular breathing as a criterion for recovery, but does not describe
the breathing pattern under anaesthesia.
Where respiration is lost, its return often requires assistance using
manual compression of the mantle. It is important to ensure that air
does not become trapped in the mantle, as this will lead to an abnor-
mal posture post-recovery and the animal may nd it difcult to void
the gas. O. vulgaris has poor ability to recover respiratory drive
unaided once respiration has stopped, which suggests weak reex re-
sponses to either hypoxia or hypercapnia, but this has not been di-
rectly investigated. Wells (1978) reports that an O. vulgaris placed
in a small enclosed container with non-circulated water showed a
marked bradycardia (a few beats per min) before showing any signs
of respiratory distress. In view of this, and in the absence of any fur-
ther data, it would be wise to oxygenate any anaesthetic solution to
minimise the impact on the heart. Even if the solution is oxygenated,
local hypoxia could occur in the mantle once contractions cease.
It is possible that external mantle compression could also stimu-
late the heart(s) (both branchial and systemic). Observation of the
systemic heart in recently killed animals shows that mechanical stim-
ulation will evoke a contraction. The ability of the isolated systemic
heart to contract in response to distension has been noted (Wells,
1978), and so external compression might perform a similar role.
1
Publications that include surgery on cephalopods rarely report such perfusion,
however.
2
Not to be confused with humane endpoints.
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
Research into the effects of anaesthetics on the cardiac ganglion,
and on the key elements of the cardiovascular system required to
maintain cerebral perfusion (branchial and systemic hearts, vena
cava), will be important to future understanding of anaesthesia in
cephalopods.
In summary, whilst a reduction in respiration frequency is not unex-
pected in the presence of a general anaesthetic, cessation of respiration
is questionable as a criterion for anaesthesia. There is an urgent need
for anaesthetic techniques that preserve respiratory drive in cephalopods.
If this is not feasible, reliable methods of maintenance of blood gases,
preservation of cardiac function and post-anaesthetic ventilation resusci-
tation need to be developed. Irrespective of the signicance of respiration
frequency as an index of anaesthetic depth, it would be good practice to
monitor respiration frequency during anaesthesia as an index of the
physiological state of the animal. Blood colour (if visible in areas such as
the branchial hearts) may be a crude indication of blood gas status as
well-oxygenated blood is blue, whereas deoxygenated blood is colourless
(Abbott et al., 1985). Non-invasive techniques need to be developed for
measurement of blood oxygen (haemocyanin) saturation comparable
to infra-red pulse oximetry technique used in mammals.
5.3.2. Decrease in chromatophore tone (paling)
Paling is an early sign of anaesthesia and is accompanied by
smoothing of the skin (i.e. loss of its 3-D texture). This is because
the overall colour of a cephalopod is determined primarily by the
innervation to the chromatophores, which arises directly from the
sub-oesophageal lobes of the brain (for review see for example:
Hanlon and Messenger, 1996; Messenger, 2001). However, it is uncer-
tain whether this paling is a centrally-controlled or solely peripheral
effect, and so, as noted above, might not in fact indicate that the animal
is unconscious. If the chromatophore tone is an index of brain activity, it
would be expected that all chromatophores would be equally affected
by anaesthesia, however the effect of anaesthesia on the different
groups of chromatophores has not been investigated for chemical an-
aesthetics. Cold water anaesthesia has differential effects on chromato-
phores in O. vulgaris (Andrews et al., 1982) although this could also be
explained by peripheral actions.
Skin colour might be one of the most sensitive indicators of the
depth of anaesthesia, if it could be readily measured in real-time. In
assessing chromatophore tone, the colour of the background and light-
ing needs to be taken into account. In addition, when assessing recovery
it is important to distinguish spontaneous permanent recovery of nor-
mal colouration in an animal left undisturbed following anaesthesia,
from transient change in chromatophore activity caused by stimulating
the animal.
5.3.3. Reduced arm activity, tone and sucker adhesiveness
Overall, the spontaneous locomotor activity of the animal should re-
duce as general anaesthesia progresses. However, assessment of activity
may be confounded, at least initially, by reaction to the anaesthetic
chamber and the anaesthetic itself. Several authors report attempts to
escape from chambers containing urethane and ethanol (e.g. Andrews
and Tansey, 1981).
Loss of tone in the arms manifests as a gradual relaxation, most eas-
ily observed when the arms are initially curled as in the defensive pos-
ture often adopted on exposure to anaesthetic. The arms become
immobile and contact the bottom of the tank. In octopus attached to
the side of the tank at the onset of anaesthesia, the distal part of the
arm appears to become detached from the wall before the more proxi-
mal part. Manipulation of the arms conrms the accidity and also re-
veals reduced sucking intensity (Seol et al., 2007) which further
contributes to the inability to maintain normal posture. It is not
known whether the adhesiveness of all suckers along the arms is lost
at the same time (indicative of loss of central control) or if the loss is
progressive, e.g. distal to proximal. Several publications investigating
the effect of discrete brain lesions in O. vulgaris note changes in ability
57
of the suckers to grip (review in Wells, 1978). These papers may give
insights into central site(s) of action of anaesthetics. Seol et al. (2007)
report that sucking intensity returns early in the recovery process in
O. minor, suggesting a central action.
5.3.4. Loss of normal posture and righting reex
Overall posture is difcult to monitor objectively in cephalopods es-
pecially if they are in a relatively small container such as those com-
monly used for anaesthesia. Anaesthetised squid will fail to maintain
their normal place in the water column, but octopus and cuttlesh
may simply settle on the bottom of the anaesthetic chamber whilst
maintaining what appears to be a normal posture. In octopus, the rela-
tive positions of the arms, head and mantle may alter so that they are at
abnormal angles to each other, and in squid and cuttlesh the head/
arms may appear unsupported by the mantle collar muscles. In
O. vulgaris the head is normally elevated above the arms and the mantle,
but under anaesthesia the animal has a more attened appearance
when on the oor of the tank. The righting response is driven by the
statocyst system and is easier to assess objectively, but involves han-
dling the animal and so is difcult to test repeatedly during the induc-
tion phase, as handling may arouse the animal and delay the onset of
anaesthesia. Once anaesthetised, the animal can be inverted to check
that the righting response is lost. By the same token, return of the
righting response will signal recovery. In octopus, care needs to be
taken to ensure that animals are not orientating themselves by using
the suckers rather than the statocysts, although loss of suction makes
this less likely. Attempts to orientate the head and mantle may be a suf-
cient indication for return of orientation perception.
Messenger et al. (1985) showed loss of the righting response in
response to magnesium chloride anaesthesia in S. ofcinalis, L. forbesi,
O. vulgaris and E. cirrhosa, and noted that the response was lost before
cessation of ventilation. In their study of urethane, ethanol and cold
water anaesthesia in O. vulgaris, Andrews and Tansey (1981) noted
that with all three anaesthetics the temporal order of functional recov-
ery was: ventilation > chromatophores > posture. These observations
support the suggestion that functions return in reverse order to their
loss during induction, which in turn suggests that there is a common
brain target where the agents act to produce general anaesthesia.
5.3.5. Responsiveness to a noxious stimulus
This should be the gold standard criterion for assessing whether
an animal has reached a depth of anaesthesia sufcient for surgery,
but identifying a standardised technique for assessment is not simple.
The absence of visible responses to pressure on the eyeball (contraction
of the peri-orbital skin), pinching the supraorbital skin (known to be
particularly sensitive, but data are required on sensory nerve density),
the arms, mantle or n have all been used to assess anaesthesia in ceph-
alopods (Andrews and Tansey, 1981; Messenger et al., 1985).
Using ethanol (1.5%) as an anaesthetic agent in O. vulgaris, Miyan
and Messenger (1995) noted that loss of the response to a pinch of
the supraorbital skin generally occurred before cessation of respira-
tion. Pagano et al. (2011) have also investigated the use of a noxious
chemical stimulus (1% acetic acid) applied to the tip of the arm although
such a stimulus will also cause retraction of an arm in vitro (Hague et al.,
2011, 2013). In the absence of any neurophysiological data on the me-
chanical or chemical sensitivity of identied nociceptive afferents in
any cephalopod (see above) it is difcult to give guidance on the mag-
nitude of the stimulus that should be applied. For example, von Frey
hairs are commonly used as mechanical stimuli in mammals and have
been used to test the mechanical threshold of nociceptors in the gastro-
pod mollusc A. californica and the leech Hirudo medicinalis (Walters,
1996). Semmes-Weinstein laments have been used to test for sensiti-
zation to mechanical stimuli in the region of an arm wound in squid
(Crook et al., 2011). In selecting a stimulus and a response to gauge an-
aesthesia, it is essential that the response requires involvement of the
supraoesophageal lobes of the brain. In addition, the motor pathway
58 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
(e.g. chromatophores, arm musculature) being used as the read out for
the response should not be affected directly by the anaesthetic (see
above). Whilst the lack of an overt motor response to a noxious stimu-
lus is a good index of the depth of general anaesthesia, more sensitive
early indicators of change in depth of anaesthesia or sedation (such as
increase in blood pressure or heart rate used in mammals) may be
required.
5.4. Maintenance of general anaesthesia in cephalopods
The physiological state of cephalopods under general anaesthesia,
either with or without maintenance/support of respiration, has been
little studied and the fact that animals apparently recover normally
should not be taken as an indicator of well-being during this period.
Whenever procedures are undertaken in anaesthetised cephalopods
removed from water, it is essential that the mantle is perfused with ox-
ygenated sea water, to ensure that hypoxia and hypercapnia are elimi-
nated (see above). Cooling the animal could slow metabolism, which
should help to preserve brain functions, but the benets will be minimal
if the cardiovascular system is not sufciently functional to perfuse
the tissues.
As previously noted, commonly used anaesthetic protocols allow
maintenance of anaesthesia for up to about 30 min in cephalopods.
For longer-lasting procedures, there is a need for anaesthetic techniques
that allow maintenance of cardiorespiratory function and enable
prolonged, controlled anaesthesia. Effective methods to assess the
depth of anaesthesia are also essential, so that emergence is recognised
rapidly and supplementary doses can be given. All of this represents
a challenge.
Even so Mooney et al. (2010) have succeeded in maintaining the
squid D. pealei in a sedated state for up to 5 h by returning the animal
to magnesium chloride (0.15 M) each time they began to emerge
from anaesthesia. Interestingly, the duration of sedation increased
with each of four inductions reported, suggesting that there may be a
residual effect of each exposure, or that the magnesium chloride (or a
substance released by it) accumulates, perhaps in the brain. This indi-
cates that it may be possible in other cephalopods to maintain stable
level of anaesthesia, with preservation of adequate cardiorespiratory
function, by controlled exposure(s) to magnesium chloride.
Moreover, fty years ago, in a study of mantle innervation, Gray
(1960) was able to keep O. bimaculatus and O. bimaculoides under an-
aesthesia for several hours with the animal continuing to breathe.
He used 1 to 2% urethane to induce anaesthesia and 0.5% urethane to
maintain anaesthesia. Although urethane is no longer used because of
its toxicity, this study illustrates the type of anaesthetic protocol that
could be adopted with other agents depending upon their anaesthetic
properties (such as duration of action).
If the animal has undergone a surgical procedure, an appropriate an-
algesic regimen should be devised, including the administration of anal-
gesia at a time that will allow it to act prior to recovery of consciousness.
This is becoming increasingly common practice in vertebrate surgery,
where the aim is to prevent adverse sensitisation of the central nervous
system. See further discussion of analgesia below.
5.5. Recovery
Recovery will not always occur without intervention to restart
ventilation by massage of the mantle. The factors determining wheth-
er manual intervention will be required are not known. If it proves
impossible to uncouple the anaesthetic state from marked suppression/
cessation of respiration, then practical guidelines will be needed on
how to perform mantle massage. Once respiration has restarted then re-
covery of other functions (sucker adhesion, righting, chromatophore
tone) is relatively rapid and follows a similar pattern for all anaesthetics
studied. Many authors comment on the speed of behavioural recovery,
often indicated by the willingness to attack or take food when returned
to the home tank (Agnisola et al., 1996; Shomrat et al., 2008). However,
consideration also needs to be given to other indicators of recovery
such as return of normal colouration for camouage and in response to
mild threats.
Little is known of the time taken to recover fully from anaesthesia, or
the time after which it would be appropriate to commence an experi-
ment without the risk that residual effects of anaesthesia might be a
confounding factor. Many studies have been performed within 24 h of
anaesthesia when it is likely that physiological effects are still lingering.
5.6. Recommendations
a) Cephalopods should be anaesthetised for potentially painful or
distressing procedures, wherever the animal will benet.
b) Research should be undertaken into the effectiveness of different
anaesthetic agents, and their aversiveness to cephalopods. This
should be species-specic and special attention should be paid
to preserving cardiorespiratory function.
c) Techniques are needed for anaesthesia lasting more than 30 min.
d) Research should be undertaken into appropriate sedatives for
cephalopods, to help minimise adverse effects during induction of
anaesthesia.
e) Protocols for resuscitating cephalopods, e.g. mantle massage, should
be dened and disseminated.
f) Techniques should be developed for monitoring physiological func-
tions such as blood pressure, heart rate and respiratory rate, ideally
non-invasively.
6. Local anaesthesia and analgesia for cephalopods
We have been unable to nd any literature describing use of local
anaesthetics or analgesics in cephalopods for procedures where such
measures would be considered desirable if performed in a vertebrate.
This is an area requiring investigation, particularly in view of the many
types of surgical procedure with recovery (especially brain lesions)
performed on cephalopods.
6.1. Local anaesthetic and analgesic agents
Little is known about analgesia in cephalopods. For example, Gunkel
and Lewbart (2008) recommend that ketoprofen or butorphanol be ad-
ministered to cephalopods alongside anaesthesia (see above) whenever
procedures could cause pain, but their recommended doses are in-
formed by data for analgesia in sh and amphibians.
We are not aware of published evidence to indicate whether or not
cephalopods have opioid or other receptor/transmitter/autocoid sys-
tems (e.g. cannabinoid, steroid, prostanoid) that could modulate pain
perception. However, as previously noted endogenous opioid peptides
are highly conserved in evolution and are present in both invertebrates
and vertebrates, maybe originating as immune signalling molecules.
Stefano and Kream (2008) speculate that an immune stimulatory func-
tion could have become allied with a pain dampening function during
evolution. Levels of endogenous morphine change with surgical stress,
bacterial infection and following starvation in Mytilus edulis (as cited
in Zhu and Stefano, 2009), but biological signicance is still uncertain
here. CB1-like cannabinoid receptors have been found in invertebrates
as well as vertebrates; however, it is speculated that they may be present
only in the deuterostomes (which would rule out cephalopods; Elphick
and Egertov, 2001 but see Elphick, 2012 for recent review). However,
Brady and Carbone (1973) reported that the cannabinoid 11-hydroxy-
9-tetrahydrocannabinol reversibly blocked conduction in squid giant
axon. There is evidence for steroids in the brain of O. vulgaris (De Lisa
et al., 2012) and an antinociceptive effect of a neuroactive steroid
has been reported in the land snail C. nemoralis (Kavaliers et al.,
2000). Prostanoids have been reported in molluscs including
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
A. californica, but we are not aware of evidence from cephalopods
(Stanley-Samuelson, 1991).
Clearly, further research needs to be done to identify local anaes-
thetic and analgesic agents for cephalopods. As with general anaesthe-
sia, criteria are required to demonstrate the efcacy of candidate
agents so that appropriate dosing schedules can be produced. For local
anaesthetics, chromatophore and muscle tone would be useful indica-
tors of axon block provided that the nerve trunk targeted by the injec-
tion has both sensory and motor bres of similar diameter, and so
expected to be sensitive to the same concentration of local anaesthetic.
Preliminary indications that local anaesthetic agents with efcacy
in vertebrates are likely to be effective in cephalopods come from
studies of local anaesthetics on the squid giant axon (e.g. procaine,
Narahashi and Frazier, 1975). Blockade of tonic inhibitory discharge
from the gastric ganglion to the crop in O. vulgaris was achieved by
injection of 2% xylocaine into the gastric ganglion (Andrews and
Tansey, 1983) and locally applied xylocaine has also been shown to
block transmission in the arm nerve cord (Andrews et al., 1981).
6.2. Administration routes for local anaesthetics and analgesics
In octopus, analgesics when identied could be given via cardiac
(branchial hearts), arterial (dorsal aorta) or intramuscular routes
(e.g.: Agnisola et al., 1996; Fiorito et al., 1998; Robertson et al.,
1994; Robertson et al., 1996) or possibly by immersion if the animal
is in a closed circulation recovery tank. The subcutaneous route
could also be used but particular care needs to be taken to ensure
that the skin is not damaged. It has been suggested (Agnisola et al.,
1996) that the subcutaneous route may be slow and variable because
of the relatively low capillary density (Browning, 1982), but this
needs further investigation with a range of substances with different
chemical properties (e.g. lipophilicity).
The most suitable site for subcutaneous or intramuscular injections
has not been established, nor have optimal needle sizes, but useful prin-
ciples for rening the administration of substances can be found in
Morton et al. (2001). Care should be taken when injecting into the
arms to avoid damage to the central nerve cord and ganglia as this
could cause partial denervation of the arm. Care should also be taken
with solutions that may be irritant as this will cause distress and
could induce autophagy. Injecting into the mantle of cuttlesh should
be avoided because of the possibility of damage to the cuttlebone.
Agin et al. (2003) and Bardou et al. (2010) have used the side of the
neck at a depth of about 1 cm in cuttlesh (590900 g body weight)
for drug administration because it has a dense vasculature for absorp-
tion. Mooney et al. (2010) have given intramuscular injections into
the head, mantle and arms of the squid D. pealei. It is also worth noting
that a method for oral administration of active substances to squids
of the same species has been described (Berk et al., 2009), which
could be adapted for cuttlesh and possibly octopus.
6.3. Recommendations
a) As a matter of principle, local anaesthesia and analgesia should be pro-
vided for cephalopods whenever these would be given to vertebrates.
b) Research is needed into appropriate and effective local anaesthetic
and analgesic agents, and routes for administration.
7. Methods for humane killing
There may be a requirement to humanely kill animals for a variety
of reasons, e.g. at the termination of a study (if release to the wild or
rehoming to a collection are not viable options); when a pre-set hu-
mane end point has been reached; if tissues or organs are required
as part of a study; for veterinary reasons; or if animals are surplus
to requirements and there is no alternative use for them.
59
Article 6 of Directive 2010/63/EU requires that, Member States
shall ensure that animals are killed with minimum pain, suffering
and distress. Annex IV of the Directive sets out appropriate methods
of killing animals but cephalopods are not mentioned.
Ideally, standard methods for humane killing should be developed
to minimise adverse effects on the animals and facilitate comparison
of scientic data from different institutions. The methods should be
simple to learn, requiring minimal specialist training to achieve compe-
tence. This will require an agreed view on what constitutes a humane
death in cephalopods, along with agreed methods for conrmation of
death. The latter is particularly important in view of the ability of octo-
pus to recover after a prolonged period of apnoea.
Some of the above issues are discussed below and approaches sug-
gested, including a note about hatchling and juvenile animals.
To assist in evaluating the impacts of different methods of killing on
both science and animal welfare, it is important that publications de-
scribe the method of killing, and not just state that animals were killed.
7.1. Humane death: a note on welfare and scientic considerations
A consensus meeting on humane killing using CO2 in vertebrates
(Hawkins et al., 2006) concludes that it is more important to avoid
or minimise pain and distress than it is to ensure a rapid loss of
consciousness. Or, a gentle death that takes longer is preferable to
a rapid, but more distressing death 3. The same principle could be ap-
plied to the humane killing of cephalopods, and in practice, would
mean either using an instantaneous physical method or immersion
in an anaesthetic agent that is known to be minimally irritant or aver-
sive, introduced at a rate that will not cause discomfort, pain or dis-
tress. As yet, however, the welfare impacts of such techniques have
not been properly evaluated for cephalopods.
Applying the above principle could be problematic where there is
scientic justication for xing tissues rapidly and with minimal dam-
age, in which case a degree of compromise between a rapid and a
gentle death may be necessary. There is also a requirement for suitable
non-chemical methods to be developed for studies in which death by
anaesthetic overdose would potentially confound interpretation of
results, such as studies of brain neurochemistry. Development of a
non-chemical, non-mechanical humane killing method for the removal
of brain tissue will be a particular challenge and it is likely that electrical
methods such as those used for Crustacea (Neil, 2010)3 will need to be
explored.
7.2. Mechanical methods
Mechanical methods, such as cutting between the eyes (midline in-
cision, e.g. Boyle, 1976) to destroy the brain and/or decapitation (by
severing the head from the mantle collar) are difcult, requiring consid-
erable skill. This is especially the case in larger octopus, mainly due to
difculty in restraining the animals. Decapitation should be followed
immediately by destruction of the brain and there should be prior seda-
tion (see below) wherever possible, if a mechanical method is to be
used. It would also be good practice for killing not to be performed
within the sight of another animal as there is evidence, for example,
that O. vulgaris recognise conspecics (Tricarico et al., 2011), or in an
environment where blood or alarm signals could be sensed by other
animals (e.g. a recirculating sea water system).
In our opinion, mechanical methods of killing non-anaesthetised
cephalopods should not be included within Annex IV of the EU Direc-
tive, as considerable training and experience is needed to carry out
the methods competently and humanely, and they should be used
only when specically authorised by the relevant competent authority
(see: European Commission, 2011; Smith et al., 2013this volume).
3
Available at: http://crustastun.com/assets/les/Neil_2010_Glasgow_Report.pdf
(online, last visited, August 2012).
60 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
7.3. Overdose of anaesthetic
Overdose of a non-aversive or non-irritant anaesthetic (by an ap-
propriate route) appears to be the most humane method available,
and is recommended wherever possible (e.g. Boyle, 1991).
Of currently used anaesthetics (see above), likely candidate agents
would be MgCl2 (possibly in combination with ethanol, for example in
cuttlesh) and clove oil (or preferably its active constituent, eugenol),
but further work is needed to evaluate anaesthetic properties, humane-
ness and applicability to individual cephalopod species. In particular,
it will be important to demonstrate that unconsciousness intervenes
prior to any unpleasant sensations arising from reduction of ventilation
(e.g. asphyxia).
As is the case with induction of general anaesthesia, thought needs
to be given to the exact way in which the procedure is performed. On
the basis of the literature and guidelines for killing with carbon diox-
ide in terrestrial vertebrates, immersing the animal directly into a
high concentration of the euthanasia solution could cause avoidable
suffering and it would be more humane to place them in a chamber
containing sea water to which an incremental concentration of the
agent is added (European Parliament and Council of the European
Union, 2010; Home Ofce, 1997; see also Hawkins et al. 2006).
In O. vulgaris magnesium chloride (MgCl2, 3.5%) usually stops evident
breathing within 15 min of immersion, at which time there is also
marked bradycardia and immobility (e.g. Grimaldi et al., 2007;
Messenger et al., 1985). After 30 min of immersion it is not possible to
revive the animals in fresh seawater even with mantle massage and it
is considered that this is the concentration and exposure time required
to kill the animal (Boyle, 1991; Grimaldi et al., 2007) but see below. A
0.15 M solution (~1.4%) of magnesium chloride was used to kill squid
D. pealei, leaving them in the solution for 3 to 4 min after ventilation
stopped (Mooney et al., 2010) and 7.14% MgCl2 in seawater was used
to anaesthetise embryo, hatchling and adult I. notoides, S. ofcinalis,
L. vulgaris, Eupymna scolopes and A. argo prior to xation (Wollesen
et al., 2009). In S. ofcinalis, both chemical and mechanical methods
are used at the same time for humane killing. Cuttlesh can be
anaesthetised by immersion for 10 min in sea water containing 2% etha-
nol by which time they become pale and immobile (Agin et al., 2001) and
then killed by rapid decapitation (Bardou et al., 2010; Boyer et al., 2007;
Groeger et al., 2006; Talbot and Marshall, 2011). If the use of ethanol is
precluded because of the type of immunohistochemical procedure to be
undertaken then cuttlesh have been killed by rapid decapitation alone
(Bardou et al., 2010; Boyer et al., 2007; Groeger et al., 2006; Talbot
and Marshall, 2011).
It must be emphasised that the use of magnesium chloride as an
agent for humane killing of cephalopods has not been studied systemat-
ically to investigate variables such as animal weight, sex, season and
water temperature. However, it is clear that immersion in magnesium
chloride is capable of stopping ventilation in exemplar species of octo-
pus, cuttlesh and squid, although the optimal concentration and expo-
sure time for humane killing requires investigation, and the time taken
to achieve death might make brains and other tissues with a high met-
abolic activity unsuitable for in vitro work. No studies have investigated
the use of clove oil specically for humane killing.
Cooling (particularly in tropical and warm temperate species) might
assist in the process of humane killing, with the possibility of potentiat-
ing anaesthetic action as well as reducing metabolic rate (so enhancing
tissue viability for subsequent studies). Cold seawater containing 2%
ethanol has been used to deeply anaesthetise O. vulgaris prior to brain
removal (Hochner et al., 2003) but we are not aware of studies where
this and other combinations with cold sea water have been used for
humane killing.
There may also be potential for using volatile anaesthetics, such
as isourane, by bubbling the agents into the seawater but special
apparatus will need to be designed to deliver a known partial pressure
of the anaesthetic and to remove residual vapourised anaesthetic.
Chloroform is the only volatile anaesthetic to have been investigated
in cephalopods: E. cirrhosa, O. vulgaris, S. ofcinalis (Alexandrowicz,
1964; Alexandrowicz, 1965), and S. sepioidea (Garca-Franco, 1992). In
squid it induced a violent reaction, with animals becoming immobile
in less than 1 min and dying even if subsequently placed in fresh sea
water (Garca-Franco, 1992). In this study animals were placed directly
in the solution, which was probably highly aversive, so it is possible that
exposure to an incremental rise in chloroform (or another volatile
agent) would avoid such a violent reaction, which is unacceptable
from an animal welfare aspect (cf. the incremental rise in CO2 used for
humane killing in mammals). Note that such a method would only be
suitable for killing and not for anaesthesia, as the animals will recover
quickly once removed from the water unless the mantle is perfused
with sea water containing the volatile agent (which, as noted, must
be administered using specialist equipment).
Garca-Franco (1992) investigated killing using carbon dioxide
(soda water) in S. sepioidea and reported that at 10% it was lethal, prob-
ably by causing tetanus of the mantle, producing asphyxia. This method
is likely to cause avoidable suffering and is not recommended. However,
it is possible that an incremental rise in CO2 may reduce such problems
and should be investigated.
Hatchlings or larvae can also be humanely killed by an overdose of
anaesthetic (e.g.: Mackie, 2008; Wollesen et al., 2009; see also Hu et
al., 2010) so it is not acceptable to kill them by direct immersion in
xative. The sensitivity of hatchlings and larvae to anaesthetics has
not been investigated e.g. they may be more resistant to hypoxia
than adults and so care must be taken in conrming death.
7.4. Methods for conrming death
Following any of the above methods, steps must be taken to con-
rm that death has occurred. Methods of conrming death are also
laid out in Annex IV of the Directive. Of these, for cephalopods:
i. Destruction of the brain is the most effective method for most spe-
cies, but the small size of the brain and difculty in location in
some (e.g. A. argo, N. pompilius) may make this hard to do without
training. Clearly, this method is not applicable if the brain is required
for study.
ii. Exsanguination by cutting the dorsal aorta or incision of the system-
ic heart might also be a suitable method for conrming death (but
not for humane killing per se). Incision of the systemic heart is per-
haps the most practical method for cuttlesh, where the dorsal
aorta is difcult to nd because of the cuttlebone.
Although cephalopod blood does not clot, haemorrhage is not always
rapid. For example, in O. vulgaris and E. cirrhosa transection of the arms
under light ethanol anaesthesia results in little bleeding from the stumps
because of an immediate potent vasoconstriction effect, whereas under
deep anaesthesia the vasoconstrictor response does not occur and the
animal haemorrhages to death (Abbott and Miyan, 1995). Although
this nding has not been systematically investigated, this observation
shows that some care will need to be taken when identifying appropriate
methods for conrming death.
Other methods listed in Annex IV are not suitable for conrming
death in cephalopods. They are:
i. Conrmation of permanent cessation of circulation. Not appro-
priate if performed by measuring elimination of the heart beat
as all three hearts are likely to beat for a prolonged time post
mortem, although at a low frequency. However, destruction of
the systemic heart would ensure that brain perfusion ceases.
ii. Conrmation of the onset of rigor mortis. Impossible as there is
no rigor mortis in cephalopods.
iii. Dislocation of the neck. Also impossible, for obvious reasons.
Maceration, which is not considered a suitable method of killing
conscious animals, might also be an appropriate method for conrming
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
death in cephalopods killed by other means, but is not listed in Annex IV4
as a conrmatory method. However, the Annex does allow for methods
of killing other than those listed in the relevant table on unconscious
animals, provided that they do not recover consciousness before death.
7.5. Conclusions on humane killing
The choice of killing method will mainly depend upon the reason for
the decision to kill the animal. For example, a chemical method may be
appropriate if the animal is being killed because a humane end point has
been reached, but if tissues are required for study in vitro then a me-
chanical or other non-chemical method may be required as the anaes-
thetic agent may interfere with tissue functioning. For magnesium
chloride it is likely that the effects on excitable tissue (e.g. brain and
heart) will reverse when the tissue is placed in anaesthetic free solution,
but the effects on neurochemistry are less easy to predict. In addition
there is evidence from both S. ofcinalis and E. cirrhosa that inadequate
brain perfusion leads to hypoxia and neuronal swelling (Abbott and
Miyan, 1995).
7.5.1. A proposed protocol
Based upon current evidence the following method is proposed for
humanely killing cephalopods such as O. vulgaris, E. cirrhosa, S. ofcinalis
and D. pealei:
At least 15 minutes immersion in MgCl2 (with a rising concentration
[optimal rates remain to be determined], ending with a nal concen-
tration of at least 3.5% in the chamber used for humane killing), pos-
sibly enhanced by using a chilled solution or with the clove oil active
constituent eugenol, followed by immediate mechanical destruction
of the brain.
If the brain is needed for study, the immersion period can be in-
creased to >30 min, which should kill the animal without a need for
mechanical destruction. However, the impact of this method upon
brain physiology and pharmacology will need to be investigated. It
may be necessary to use shorter periods of exposure to the anaesthetic
agent followed by decapitation once it has been ensured that the animal
is insensible.
It must be emphasised that the above are initial proposals and it is
likely that methods will need to be rened, possibly with the identi-
cation of species-specic protocols.
7.6. Recommendations
a) Criteria to dene what constitutes a humane death should be
developed and agreed for cephalopods.
b) Conscious cephalopods should not be killed using mechanical
methods alone, unless there is scientic justication and specic
authorisation.
c) Research is needed into appropriate agents and administration
protocols for the humane killing of cephalopods. A humane proto-
col is suggested above, for further consideration.
d) Protocols for conrming death in cephalopods should be dened
and disseminated.
e) Hatchlings and larvae should be killed by overdose of anaesthetic
and not by immersion in tissue xative alone.
4
Schedule 1 to the UK Animals (Scientic Procedures) Act 1986 permits mechanical
disruption as a means of conrming death, but requires conrmation that the animal is
insensitive and does not respond to a painful stimulus before it is placed in the macer-
ator (Home Ofce, 1997).
61
Acknowledgements
Some of the concepts discussed in this paper and in particular
Table 1 emerged from a Cephalopod Working Group of the Boyd
Group. We wish to thank colleagues from that group (G. Fiorito,
J. Patterson, G. Ponte, J. Rundle, and R. Williamson) who are not authors
on the current paper. PLRA wishes to thank ASSEMBLE (Association of
European Marine Biological Laboratories) for funding a preliminary
study of nociception in O. vulgaris at Stazione Zoologica Napoli during
which some of the ideas in this paper were developed. [SS]
References
Abbott, N.J., Miyan, J.A., 1995. Cerebrovascular organization and dynamics in cephalopods.
In: Abbott, N.J., Williamson, R., Maddock, L. (Eds.), Cephalopod Neurobiology:
Neuroscience Studies in Squid, Octopus and Cuttlesh. Oxford University Press,
Oxford, New York, pp. 459476.
Abbott, N.J., Bundgaard, M., Cserr, H.F., 1985. Brain vascular volume, electrolytes and
bloodbrain interface in the cuttlesh Sepia ofcinalis (Cephalopoda). J. Physiol.
368, 197212.
Agin, V., Chichery, R., Chichery, M.P., 2001. Effects of learning on cytochrome oxidase
activity in cuttlesh brain. Neuroreport 12, 113116.
Agin, V., Chichery, R., Maubert, E., Chichery, M.P., 2003. Time-dependent effects of
cycloheximide on long-term memory in the cuttlesh. Pharmacol. Biochem.
Behav. 75, 141146.
Agin, W., Chichery, R., Dickel, L., Chichery, M.P., 2006. The prawn-in-the-tube procedure
in the cuttlesh: habituation or passive avoidance learning? Learn. Mem. 13, 97101.
Agnisola, C., Castaldo, P., Fiorito, G., 1996. Octopus vulgaris (Mollusca, Cephalopoda) as
a model in behavioral pharmacology: a test of handling effects. Physiol. Behav. 59,
729733.
Alexandrowicz, J.S., 1964. The neurosecretory system of the vena cava in cephalopoda I.
Eledone Cirrosa. J. Mar. Biol. Assoc. U. K. 44, 111132.
Alexandrowicz, J.S., 1965. The neurosecretory system of the vena cava in Cephalopoda
II. Sepia ofcinalis and Octopus vulgaris. J. Mar. Biol. Assoc. U.K. 45, 209228.
Anderson, R.C., Wood, J.B., 2001. Enrichment for giant Pacic octopuses: happy as a
clam? J. Appl. Anim. Welfare Sci. 4, 157168.
Anderson, R.C., Wood, J.B., Byrne, R.A., 2002. Octopus senescence: the beginning of the
end. J. Appl. Anim. Welfare Sci. 5, 275283.
Andrews, P.L.R., 2011. Laboratory invertebrates: only spineless, or spineless and pain-
less? Introduction. ILAR J. 52, 121125.
Andrews, P.L.R., Tansey, E.M., 1981. The effects of some anesthetic agents in Octopus
vulgaris. Comp. Biochem. Physiol. 70C, 241247.
Andrews, P.L.R., Tansey, E.M., 1983. The digestive tract of Octopus vulgaris. The anatomy,
physiology and pharmacology of the upper tract. J. Mar. Biol. Ass. UK 63, 109134.
Andrews, P.L.R., Messenger, J.B., Tansey, E.M., 1981. Color changes in cephalopods after
neurotransmitter injection into the cephalic aorta. Proc. R. Soc. Lond. B 213, 9399.
Andrews, P.L.R., Packard, A., Tansey, E.M., 1982. A physiologically discrete population of
chromatophres in Octopus vulgaris (Mollusca). J. Zool. 198, 131140.
Angel, A., 1993. Central neuronal pathways and the process of anesthesia. Br. J.
Anaesth. 71, 148163.
Baker, P.F., Hodgkin, A.L., Ridgway, E.B., 1971. Depolarization and calcium entry in
squid giant axons. J. Physiol. 218, 709755.
Bardou, I., Leprince, J., Chichery, R., Vaudry, H., Agin, V., 2010. Vasopressin/oxytocin-
related peptides inuence long-term memory of a passive avoidance task in the
cuttlesh, Sepia ofcinalis. Neurobiol. Learn. Mem. 93, 240247.
Barr, S., Elwood, R.W., 2011. No evidence of morphine analgesia to noxious shock in the
shore crab, Carcinus maenas. Behav. Proc. 86, 340344.
Berk, W., Teperman, J., Walton, K.D., Hirata, K., Sugimori, M., Llinas, R.R., 2009. Oral ad-
ministration of pharmacologically active substances to squid: a methodological de-
scription. Biol. Bull. 216, 16.
Boal, J.G., Golden, D.K., 1999. Distance chemoreception in the common cuttlesh, Sepia
ofcinalis (Mollusca, Cephalopoda). J. Exp. Mar. Biol. Ecol. 235, 307317.
Boal, J.G., Ni, J.N., 1996. Ventilation rate of cuttlesh, Sepia ofcinalis, in response to vi-
sual stimuli. Veliger 39, 342347.
Boal, J.G., Wittenberg, K.M., Hanlon, R.T., 2000. Observational learning does not explain
improvement in predation tactics by cuttlesh (Mollusca: Cephalopoda). Behav.
Proc. 52, 141153.
Borrelli, L., Fiorito, G., 2008. Behavioral analysis of learning and memory in cephalo-
pods. In: Byrne, J.J. (Ed.), Learning and Memory: A Comprehensive Reference. Aca-
demic Press, Oxford, pp. 605627.
Borrelli, L., Gherardi, F., Fiorito, G., 2006. A Catalogue of Body Patterning in Cephalopoda.
Stazione Zoologica A. Dohrn. Firenze University Press, Napoli, Italy, pp. 1626.
Boycott, B.B., 1954. Learning in Octopus vulgaris and other cephalopods. Pubbl. Staz.
Zool. Napoli 25, 6793.
Boyer, C., Maubert, E., Charnay, Y., Chichery, R., 2007. Distribution of neurokinin A-like
and serotonin immunoreactivities within the vertical lobe complex in Sepia
ofcinalis. Brain Res. 1133, 5366.
Boyle, P.R., 1976. Receptor units responding to movement in octopus mantle. J. Exp.
Biol. 65, 19.
Boyle, P.R., 1983. Ventilation rate and arousal in the octopus. J. Exp. Mar. Biol. Ecol. 69,
129136.
62 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
Boyle, P.R., 1991. The UFAW Handbook on the Care and Management of Cephalopods
in the Laboratory. Universities Federation for Animal Welfare, Potters Bar, Herts,
UK, pp. 163.
Brady, R.O., Carbone, E., 1973. Comparison of effects of delta-9-tetrahydrocannabinol,
11-hydroxy-delta-9-tetrahydrocannabinol, and ethanol on electrophysiological
activity of giant axon of squid. Neuropharmacology 12, 601605.
Braithwaite, V.A., Boulcott, P., 2007. Pain perception, aversion and fear in sh. Dis.
Aquat. Org. 75, 131138.
Broom, D.M., 2007. Cognitive ability and sentience: which aquatic animals should be
protected? Dis. Aquat. Org. 75, 99108.
Brown, E.R., Piscopo, S., De Stefano, R., Giuditta, A., 2006. Brain and behavioural evi-
dence for rest-activity cycles in Octopus vulgaris. Behav. Brain Res. 172, 355359.
Browning, J., 1982. The density and dimensions of exchange vessels in Octopus pallidus.
J. Zool. 196, 569579.
Bullock, T.H., 1965. Mollusca: Cephalopoda. In: Bullock, T.H., Horridge, G.A. (Eds.),
Structure and Function in the Nervous Systems of Invertebrates. W.H. Freeman
and Company, San Francisco and London, pp. 14331515.
Callan, H.G., 1940. The absence of a sex hormone controlling regeneration of the
hectocotylus in Octopus vulgaris L Pubbl. Staz. Zool. Napoli 18, 1519.
Close, B., Banister, K., Baumans, V., Bernoth, E.M., Bromage, N., Bunyan, J., Erhardt, W.,
Flecknell, P., Gregory, N., Hackbarth, H., Morton, D., Warwick, C., 1996. Recommen-
dations for euthanasia of experimental animals: part 1. Lab. Anim. 30, 293316.
Colmers, W.F., Hixon, R.F., Hanlon, R.T., Forsythe, J.W., Ackerson, M.V., Wiederhold,
M.L., Hulet, W.H., 1984. Spinner cephalopods defects of statocyst suprastructures
in an invertebrate analog of the vestibular apparatus. Cell Tissue Res. 236, 505515.
Crook, R.J., Walters, E.T., 2011. Nociceptive behavior and physiology of molluscs: animal
welfare implications. ILAR J. 52, 185195.
Crook, R.J., Lewis, T., Hanlon, R.T., Walters, E.T., 2011. Peripheral injury induces long-
term sensitization of defensive responses to visual and tactile stimuli in the
squid Loligo pealeii, Lesueur 1821. J. Exp. Biol. 214, 31733185.
Darmaillacq, A.S., Shashar, N., 2008. Lack of polarization optomotor response in the cut-
tlesh Sepia elongata (d'Orbigny, 1845). Physiol. Behav. 94, 616620.
Darmaillacq, A.S., Dickel, L., Chichery, M.P., Agin, V., Chichery, R., 2004. Rapid taste
aversion learning in adult cuttlesh, Sepia ofcinalis. Anim. Behav. 68, 12911298.
De Lisa, E., Paolucci, M., Di Cosmo, A., 2012. Conservative nature of oestradiol signalling
pathways in the brain lobes of Octopus vulgaris involved in reproduction, learning
and motor coordination. J. Neuroendocrinol. 24, 275284.
Edelman, D.B., Seth, A.K., 2009. Animal consciousness: a synthetic approach. Trends
Neurosci. 32, 476484.
EFSA Panel on Animal Health and Welfare, 2005. Opinion of the Scientic Panel on
Animal Health and Welfare (AHAW) on a request from the Commission related
to the Aspects of the biology and welfare of animals used for experimental and
other scientic purposes. EFSA J. 292, 1136.
Ellis, T., Yildiz, H.Y., Lopez-Olmeda, J., Spedicato, M.T., Tort, L., Overli, O., Martins, C.I.M.,
2012. Cortisol and nsh welfare. Fish Physiol. Biochem. 38, 163188.
Elphick, M.R., 2012. The evolution and comparative neurobiology of endocannabinoid
signalling. Phil. Trans. R. Soc. B 367, 32013215.
Elphick, M.R., Egertov, M., 2001. The neurobiology and evolution of cannabinoid sig-
nalling. Philos. Trans. R. Soc. Lond. B 356, 381408.
Estefanell, J., Socorro, J., Afonso, J.M., Roo, J., Fernandez-Palacios, H., Izquierdo, M.S.,
2011. Evaluation of two anaesthetic agents and the passive integrated transponder
tagging system in Octopus vulgaris (Cuvier 1797). Aquac. Res. 42, 399406.
European Commission, 2011. National Competent Authorities for the implementation
of Directive 2010/63/EU on the protection of animals used for scientic purposes.
Working Document on Specic Articles in Directive 2010/63/EU. European Com-
mission, Brussels, pp. 112.
European Parliament, Council of the European Union, 2010. Directive 2010/63/EU of
the European Parliament and of the Council of 22 September 2010 on the Protec-
tion of Animals Used for Scientic Purposes. Concil of Europe, Strasbourg.
Fiorito, G., 1986. Is there pain in Invertebrates? Behav. Proc. 12, 383388.
Fiorito, G., Chichery, R., 1995. Lesions of the vertical lobe impair visual discrimination
learning by observation in Octopus vulgaris. Neurosci. Lett. 192, 117120.
Fiorito, G., Scotto, P., 1992. Observational learning in Octopus vulgaris. Science 256, 545547.
Fiorito, G., Agnisola, C., d'Addio, M., Valanzano, A., Calamandrei, G., 1998. Scopolamine
impairs memory recall in Octopus vulgaris. Neurosci. Lett. 253, 8790.
Flecknell, P.A., Flecknell, P.A., 2009. Laboratory animal anaesthesia, 3rd ed. Elsevier/
Academic Press, Amsterdam.
Folkedal, O., Torgersen, T., Olsen, R.E., Ferno, A., Nilsson, J., Oppedal, F., Stien, L.H.,
Kristiansen, T.S., 2012. Duration of effects of acute environmental changes on
food anticipatory behaviour, feed intake, oxygen consumption, and cortisol release
in Atlantic salmon parr. Physiol. Behav. 105, 283291.
Frazier, D.T., Ohta, M., Narahash, T., 1973. Nature of morphine receptor present in squid
axon. Proc. Soc. Exp. Biol. Med. 142, 12091214.
Garca-Franco, M., 1992. Anaesthetics for the squid Sepioteuthis sepioidea (Mollusca:
Cephalopoda). Comp. Biochem. Physiol. C Comp. Pharmacol. 103, 121123.
Getting, P.A., 1976. Afferent neurons mediating escape swimming of the marine mollusc
Tritonia. J. Comp. Physiol. A: Neuroethol. Sens. Neur. Behav. Physiol. 110, 271286.
Gleadall, I.G., 2013. The effects of prospective anaesthetic substances on cephalopods:
Summary of original data and a brief review of studies over the last two decades.
J. Exp. Mar. Biol. Ecol. 447, 2330 (this volume).
Gleadall, I.G., Ohtsu, K., Gleadall, E., Tsukahara, Y., 1993. Screening pigment migration in
the octopus retina includes control by dopaminergic efferents. J. Exp. Biol. 185,
116.
Goncalves, R.A., Aragao, C., Frias, P.A., Sykes, A.V., 2012. The use of different anaes-
thetics as welfare promoters during short-term human manipulation of European
cuttlesh (Sepia ofcinalis) juveniles. Aquaculture 370371, 130135.
Graindorge, N., Jozet-Alves, C., Chichery, R., Dickel, L., Bellanger, C., 2008. Does kainic
acid induce partial brain lesion in an invertebrate model: Sepia ofcinalis? Compar-
ison with electrolytic lesion. Brain Res. 1238, 4452.
Gray, J.A.B., 1960. Mechanically excitable receptor units in the mantle of the octopus
and their connexions. J. Physiol. 153, 573582.
Graziadei, P., 1962. Receptors in suckers of octopus. Nature 195, 5759.
Grimaldi, A.M., Agnisola, C., Fiorito, G., 2007. Using ultrasound to estimate brain size in
the cephalopod Octopus vulgaris Cuvier in vivo. Brain Res. 1183, 6673.
Groeger, G., Chrachri, A., Williamson, R., 2006. Changes in cuttlesh retinal sensitivity
during growth. Vie Milieu 56, 167173.
Gunkel, C., Lewbart, G.A., 2008. Anesthesia and analgesia of invertebrates, In: Fish, R.E.,
Brown, M.J., Danneman, P.J.A.A., Karas, A.Z. (Eds.), Anesthesia and Analgesia in Lab-
oratory Animals, Second ed. Academic Press, San Diego, pp. 535545.
Gutfreund, Y., Matzner, H., Flash, T., Hochner, B., 2006. Patterns of motor activity in the
isolated nerve cord of the octopus arm. Biol. Bull. 211, 212222.
Hague, T., Florini, M., Brown, E., Fiorito, G., Andrews, P.L.R., 2011. Preliminary evidence
for a nociceptive reex to mechanical. Osmotic and acidic stimuli in the isolated
arm of Octopus vulgaris. J. Shellsh. Res. 30, 1007.
Hague, T., Florini, M., Andrews, P.L.R., 2013. Preliminary in vitro functional evidence for
reex responses to noxious stimuli in the arms of Octopus vulgaris. J. Exp. Mar. Biol.
Ecol. 447, 100105.
Hanlon, R.T., Messenger, J.B., 1996. Cephalopod Behaviour. Cambridge University Press,
Cambridge, pp. 1232.
Harms, C.A., Lewbart, G.A., McAlarney, R., Christian, L.S., Geissler, K., Lemons, C., 2006.
Surgical excision of mycotic (Cladosporium sp.) granulomas from the mantle of a
cuttlesh (Sepia ofcinalis). J. Zoo Wildl. Med. 37, 524530.
Harrison, F.M., Martin, A.W., 1965. Excretion in cephalopod Octopus doeini. J. Exp. Biol.
42, 7198.
Hawkins, P., Playle, L., Golledge, H., Leach, M., Banzett, R., Coenen, A., Cooper, J.,
Danneman, P., Flecknell, P., Niel, L., Raj, M., 2006. Newcastle Consensus Meeting
on Carbon Dioxide Euthanasia of Laboratory Animals. 17 pp. Available at:
http://www.nc3rs.org.uk/downloaddoc.asp?id=416&page=292&skin=0 (online,
last visited, August 2012).
Hawkins, P., Morton, D.B., Burman, O., Dennison, N., Honess, P., Jennings, M., Lane, S.,
Middleton, V., Roughan, J.V., Wells, S., Westwood, K., 2011. A guide to dening
and implementing protocols for the welfare assessment of laboratory animals.
Eleventh report of the BVAAWF/FRAME/RSPCA/UFAW Joint Working Group on Re-
nement. Lab. Anim. 45, 113.
Hochner, B., Brown, E.R., Langella, M., Shomrat, T., Fiorito, G., 2003. A learning and
memory area in the octopus brain manifests a vertebrate-like long-term potentia-
tion. J. Neurophysiol. 90, 35473554.
Home Ofce, 1997. The Humane Killing of Animals under Schedule 1 to the Animals
(Scientic Procedures) Act 1986: Code of Practice. TSO, London, pp. 116.
Hu, M.Y., Sucre, E., Charmantier-Daures, M., Charmantier, G., Lucassen, M., Himmerkus,
N., Melzner, F., 2010. Localization of ion-regulatory epithelia in embryos and
hatchlings of two cephalopods. Cell Tissue Res. 339, 571583.
Hurst, J.L., West, R.S., 2010. Taming anxiety in laboratory mice. Nat. Methods 7,
825826.
IASP Task Force on Taxonomy, 2011. IASP Taxonomy Nociceptor. International Associa-
tion for the Study of Pain http://www.iasp-pain.org/AM/Template.cfm?Section=
Pain_Deisplay.cfm&ContentID=1728#Nociceptor (online; last visited: August,
2012).
Ikeda, Y., Sugimoto, C., Yonamine, H., Oshima, Y., 2009. Method of ethanol anaesthesia
and individual marking for oval squid (Sepioteuthis lessoniana Ferussac, 1831 in
lesson 18301831). Aquac. Res. 41, 157160.
Illich, P.A., Walters, E.T., 1997. Mechanosensory neurons innervating Aplysia siphon
encode noxious stimuli and display nociceptive sensitization. J. Neurosci. 17,
459469.
Iseri, L.T., French, J.H., 1984. Magnesium: nature's physiologic calcium blocker. Am.
Heart J. 108, 188193.
Johansen, K., Martin, A.W., 1962. Circulation in Cephalopod, Octopus doeini. Comp.
Biochem. Physiol. 5, 165176.
Joll, L.M., 1977. Growth and food-intake of Octopus tetricus (Mollusca-Cephalopoda) in
aquaria. Aust. J. Mar. Freshwat. Res. 28, 4556.
Karelina, K., Norman, G.J., Zhang, N., Morris, J.S., Peng, H.Y., DeVries, A.C., 2009. Social iso-
lation alters neuroinammatory response to stroke. Proc. Natl. Acad. Sci. U. S. A. 106,
58955900.
Karson, M.A., Boal, J.G., Hanlon, R.T., 2003. Experimental evidence for spatial learning in
cuttlesh (Sepia ofcinalis). J. Comp. Psychol. 117, 149155.
Katz, B., Miledi, R., 1969. Spontaneous and evoked activity of motor nerve endings in
calcium ringer. J. Physiol. 203, 689706.
Kavaliers, M., Perrot-Sinal, T.S., Desjardins, D.C., Cross-Mellor, S.K., Wiebe, J.P., 2000.
Antinociceptive effects of the neuroactive steroid, 3 alpha-hydroxy-5 alpha-
pregnan-20-one and progesterone in the land snail, Cepaea nemoralis. Neuroscience
95, 807812.
Kier, W.M., Smith, K.K., Miyan, J.A., 1989. Electromyography of the n musculature of
the cuttlesh Sepia ofcinalis. J. Exp. Biol. 143, 1731.
Langridge, K.V., 2009. Cuttlesh use startle displays, but not against large predators.
Anim. Behav. 77, 847856.
Langridge, K.V., Broom, M., Osorio, D., 2007. Selective signalling by cuttlesh to preda-
tors. Curr. Biol. 17, R1044R1045.
Lewbart, G.A., Mosley, C., 2012. Clinical anesthesia and analgesia in invertebrates. J. Ex-
otic Pet Med. 21, 5970.
Liao, X.G., Walters, E.T., 2002. The use of elevated divalent cation solutions to isolate
monosynaptic components of sensorimotor connections in Aplysia. J. Neurosci.
Methods 120, 4554.
 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664
Machado, M.S., Rosa, R.M., Dantas, A.S., Reolon, G.K., Appelt, H.R., Braga, A.L., Henriques,
J.A.N., Roesler, R., 2006. An organic selenium compound attenuates apomorphine-
induced stereotypy in mice. Neurosci. Lett. 410, 198202.
Mackie, G.O., 2008. Immunostaining of peripheral nerves and other tissues in whole
mount preparations from hatchling cephalopods. Tissue Cell 40, 2129.
Malham, S.K., Runham, R.W., Secombes, C.J., 1998. Lysozyme and antiprotease activity
in the lesser octopus Eledone cirrhosa (Lam.) (Cephalopoda). Dev. Comp. Immunol.
22, 2737.
Malham, S.K., Lacoste, A., Gelebart, F., Cueff, A., Poulet, S.A., 2002. A rst insight into
stress-induced neuroendocrine and immune changes in the octopus Eledone
cirrhosa. Aquat. Living Resour. 15, 187192.
Martin, R., Frosch, D., Weber, E., Voigt, K.H., 1979. Met-enkephalin-like immunoreactiv-
ity in a cephalopod neurohemal organ. Neurosci. Lett. 15, 253257.
Martin, R., Haas, C., Voigt, K.-H., 1986. Opioid and related neuropeptides in molluscan
neurons. In: Stefano, G.B. (Ed.), CRC Handbook of Comparative Opioid and Related
Neuropeptide Mechanisms, vol. 1. CRC Press, Boca Raton, FL, pp. 4964.
Mason, G.J., 1991. Stereotypies a critical review. Anim. Behav. 41, 10151037.
Mather, J.A., 1995. Cognition in cephalopods. Adv. Stud. Behav. 24, 317353.
Mather, J.A., 1998. How do octopuses use their arms? J. Comp. Psychol. 112, 306316.
Mather, J.A., 2004. Cephalopod skin displays: from concealment to communication.
Evol. Commun. Syst. Comp Approach 193213.
Mellor, D.J., Stafford, K.J., Patterson-Kane, E., 2009. The Sciences of Animal Welfare
(UFAW Animal Welfare). Wiley-Blackwell, Oxford, pp. 1224.
Messenger, J.B., 1973. Learning in cuttlesh, Sepia. Anim. Behav. 21, 801826.
Messenger, J.B., 2001. Cephalopod chromatophores: neurobiology and natural history.
Biol. Rev. 76, 473528.
Messenger, J.B., Nixon, M., Ryan, K.P., 1985. Magnesium chloride as an anesthetic
for cephalopods. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 82,
203205.
Mettam, J.J., McCrohan, C.R., Sneddon, L.U., 2012. Characterisation of chemosensory tri-
geminal receptors in the rainbow trout, Oncorhynchus mykiss: responses to chem-
ical irritants and carbon dioxide. J. Exp. Biol. 215, 685693.
Miyan, J.A., Messenger, J.B., 1995. Intracellular recordings from the chromatophore
lobes of Octopus. In: Abbott, N.J., Williamson, R., Maddock, L. (Eds.), Cephalopod
Neurobiology: Neuroscience Studies in Squid, Octopus and Cuttlesh. Oxford Uni-
versity Press, Oxford; New York, pp. 415429.
Mooney, T.A., Lee, W.J., Hanlon, R.T., 2010. Long-duration anesthetization of squid
(Doryteuthis pealeii). Mar. Freshwat. Behav. Physiol. 43, 297303.
Morton, D.B., Grifths, P.H.M., 1985. Guidelines on the Recognition of pain, distress and
discomfort in experimental-animals and an hypothesis for assessment. Vet. Rec.
116, 431436.
Morton, D.B., Jennings, M., Buckwell, A., Ewbank, R., Godfrey, C., Holgate, B., Inglis, I.,
James, R., Page, C., Sharman, I., Verschoyle, R., Westall, L., Wilson, A.B., 2001. Ren-
ing procedures for the administration of substances. Lab. Anim. 35, 141.
Nair, J.R., Pillai, D., Joseph, S.M., Gomathi, P., Senan, P.V., Sherief, P.M., 2011. Cephalopod
research and bioactive substances. Indian J. Geo-Mar. Sci. 40, 1327.
Narahashi, T., Frazier, D.T., 1975. Site of action and active form of procaine in squid
giant-axons. J. Pharmacol. Exp. Ther. 194, 506513.
Neil, D., 2010. The Effect of the Crustastun on Nerve Activity in Crabs and Lobsters.
University of Scotland and Scientic Report to Studham Technologies Ltd.,
Glasgow, UK, pp. 121.
Nixon, M., 1966. Changes in body weight and intake of food by Octopus vulgaris. J. Zool.
150, 19.
Nixon, M., Young, J.Z., 2003. The Brains and Lives of Cephalopods. Oxford University
Press, Oxford, pp. 1392.
Pagano, E., Ponte, G., Andrews, P.L.R., Fiorito, G., 2011. A comparative analysis of different
anaesthetics in octopus: towards true anesthesia? J. Shellsh. Res. 30, 1016.
Pantin, C.F.A., 1946. Notes on Microscopical Technique for Zoologists. Cambridge
University Press, Cambridge, pp. 191.
Phillips, D.W., 1979. Ultrastructure of sensory cells on the mantle tentacles of the gas-
tropod Notoacmea scutum. Tissue Cell 11, 623632.
Pierce du Sert, N., Chu, K.M., Wai, M.K., Rudd, J.A., Andrews, P.L.R., 2009. Reduced
normogastric electrical activity associated with emesis: a telemetric study in fer-
rets. World J. Gastroenterol. 15, 60346043.
Piscopo, S., Moccia, F., Di Cristo, C., Caputi, L., Di Cosmo, A., Brown, E.R., 2007. Pre- and
postsynaptic excitation and inhibition at octopus optic lobe photoreceptor termi-
nals; implications for the function of the presynaptic bags. Eur. J. Neurosci. 26,
21962203.
Polglase, J.L., Bullock, A.M., Roberts, R.J., 1983. Wound-healing and the hemocyte re-
sponse in the skin of the lesser octopus Eledone cirrhosa (Mollusca, Cephalopoda).
J. Zool. 201, 185204.
Robertson, J.D., Bonaventura, J., Kohm, A.P., 1994. Nitric oxide is required for tactile
learning in Octopus vulgaris. Proc. R. Soc. Lond. B 256, 269273.
Robertson, J.D., Bonaventura, J., Kohm, A., 1995. Nitric-oxide synthase inhibition blocks
octopus touch learning without producing sensory or motor dysfunction. Proc. R.
Soc. Lond. B 261, 167172.
Robertson, J.D., Bonaventura, J., Kohm, A., Hiscat, M., 1996. Nitric oxide is necessary for
visual learning in Octopus vulgaris. Proc. R. Soc. Lond. B 263, 17391743.
Roedel, A., Storch, C., Holsboer, F., Ohl, F., 2006. Effects of light or dark phase testing on
behavioural and cognitive performance in DBA mice. Lab. Anim. 40, 371381.
Rose, J.D., 2007. Anthropomorphism and mental welfare of shes. Dis. Aquat. Org. 75,
139154.
Ross, L.G., Ross, B., 2008. Anaesthetic and Sedative Techniques for Aquatic Animals, 3rd
ed. Wiley-Blackwell, pp. 1240.
Rowell, C.H.F., 1963. Excitatory and inhibitory pathways in arm of octopus. J. Exp. Biol.
40, 257270.
63
Rowell, C.H.F., 1966. Activity of interneurones in arm of octopus in response to tactile
stimulation. J. Exp. Biol. 44, 589605.
Savignac, H.M., Dinan, T.G., Cryan, J.F., 2011. Resistance to early-life stress in mice:
effects of genetic background and stress duration. Front. Behav. Neurosci. 5 (13),
112.
Sen, H., Tanrikul, T.T., 2009. Efcacy of 2-phenoxyethanol as an anaesthetic for the
musky octopus, Eledone moschata (Lamarck 1799), (Cephalopoda: Octopodidae).
Turk. J. Vet. Anim. Sci. 33, 463467.
Seol, D.W., Lee, J., Im, S.Y., Park, I.S., 2007. Clove oil as an anaesthetic for common
octopus (Octopus minor, Sasaki). Aquac. Res. 38, 4549.
Sha, A.L., Sun, H.S., Wang, Y., 2007. Immunohistochemical localization of met-enkephalin
in mantles and feet of octopus Octopus ocellatus Grey. Fish. Sci. 26, 622624.
Sha, A.L., Sun, H.S., Wang, Y.Y., 2012. Immunohistochemical study of leucine-
enkephalin and delta opioid receptor in mantles and feet of the octopus Octopus
ocellatus Gray. Int. J. Pept. Res. Ther. 18, 7176.
Shomrat, T., Zarrella, I., Fiorito, G., Hochner, B., 2008. The octopus vertical lobe modulates
short-term learning rate and uses LTP to acquire long-term memory. Curr. Biol. 18,
337342.
Shomrat, T., Feinstein, N., Klein, M., Hochner, B., 2010. Serotonin is a facilitatory
neuromodulator of synaptic transmission and reinforces long-term potentiation
induction in the vertical lobe of Octopus vulgaris. Neuroscience 169, 5264.
Smith, E.S.J., Lewin, G.R., 2009. Nociceptors: a phylogenetic view. J. Comp. Physiol. A
195, 10891106.
Smith, J.A., Boyd, K.M., Institute of Medical Ethics (Great Britain), 1991. Lives in the bal-
ance. The Ethics of Using Animals in Biomedical Research: The Report of a Working
Party of the Institute of Medical Ethics. Oxford University Press, Oxford, pp. 1352.
Smith, J.A., Andrews, P.L.R., Hawkins, P., Louhimies, S., Ponte, G., Dickel, L., 2013. Ceph-
alopod research and EU Directive 2010/63/EU: Requirements, impacts and ethical
review. J. Exp. Mar. Biol. Ecol. 447, 3145 (this volume).
Sneddon, L.U., 2009. Pain perception in sh: indicators and endpoints. ILAR J. 50,
338342.
Sneddon, L.U., Braithwaite, V.A., Gentle, M.J., 2003. Do shes have nociceptors? Evi-
dence for the evolution of a vertebrate sensory system. Proc. R. Soc. Lond. B 270,
11151121.
Stanley-Samuelson, D.W., 1991. Comparative eicosanoid physiology in invertebrate ani-
mals. Am. J. Physiol. Regul. Integr. Comp. Physiol. 260, R849R853.
Stefano, G.B., Kream, R., 2008. Endogenous opiates, opioids, and immune function:
evolutionary brokerage of defensive behaviors. Semin. Cancer Biol. 18, 190198.
Stefano, G.B., Hall, B., Makman, M.H., Dvorkin, B., 1981. Opioid inhibition of dopamine
release from nervous-tissue of Mytilus edulis and Octopus bimaculatus. Science 213,
928930.
Stern, R.M., Koch, K.L., Andrews, P.L.R., 2011. Nausea Mechanisms and Management.
Oxford University Press, Oxford, pp. 1480.
Sumbre, G., Fiorito, G., Flash, T., Hochner, B., 2005. Motor control of exible octopus
arms. Nature 433, 595596.
Sykes, A.V., Baptista, F.D., Gonalves, R.A., Andrade, J.P., 2012. Directive 2010/63/EU on
animal welfare: a review on the existing scientic knowledge and implications in
cephalopod aquaculture research. Rev. Aquac. 4, 142162.
Talbot, C.M., Marshall, J.N., 2011. The retinal topography of three species of coleoid
cephalopod: signicance for perception of polarized light. Philos. Trans. R. Soc.
Lond. B 366, 724733.
Tricarico, E., Borrelli, L., Gherardi, F., Fiorito, G., 2011. I know my neighbour: individual
recognition in Octopus vulgaris. PLoS One 6, e18710.
Trueman, E.R., Packard, A., 1968. Motor performances of some cephalopods. J. Exp. Biol.
49, 495507.
Trushenski, J.T., Bowker, J.D., Gause, B.R., Mulligan, B.L., 2012. Chemical and electrical
approaches to sedation of hybrid striped bass: induction, recovery, and physiolog-
ical responses to sedation. Trans. Am. Fish. Soc. 141, 455467.
Voigt, K.H., Kiehling, C., Frosch, D., Schiebe, M., Martin, R., 1981. Enkephalin-related
peptides direct action on the octopus heart. Neurosci. Lett. 27, 2530.
Walker, J.J., Longo, N., Bitterman, M.E., 1970. Octopus in laboratory handling, main-
tenance, training. Behav. Res. Methods Instrum. 2, 1518.
Walters, E.T., 1987. Site-specic sensitization of defensive reexes in Aplysia a
simple-model of long-term hyperalgesia. J. Neurosci. 7, 400407.
Walters, E.T., 1996. Comparative and evolutionary aspects of nociceptor function. In:
Oxford, U. (Ed.), Neurobiology of nociceptors. Oxford University Press, pp. 92114.
Wells, M.J., 1978. Octopus: Physiology and Behaviour of an Advanced Invertebrate.
Chapman and Hall, London, pp. 1417.
Westermann, B., Ruth, P., Litzlbauer, H.D., Beck, I., Beuerlein, K., Schmidtberg, H., Kaleta,
E.F., Schipp, R., 2002. The digestive tract of Nautilus pompilius (Cephalopoda,
Tetrabranchiata): an Xray analytical and computational tomography study on the
living animal. J. Exp. Biol. 205, 16171624.
Wodinsky, J., 1977. Hormonal inhibition of feeding and death in octopus control by
optic gland secretion. Science 198, 948951.
Wollesen, T., Loesel, R., Wanninger, A., 2009. Pygmy squids and giant brains: mapping
the complex cephalopod CNS by phalloidin staining of vibratome sections and
whole-mount preparations. J. Neurosci. Methods 179, 6367.
Young, J.Z., 1963. The number and sizes of nerve cells in Octopus. Proc. Zool. Soc. London
140, 229254.
Young, J.Z., 1966. The Memory System of the Brain. University of California Press,
Berkeley, Calif., pp. 1128.
Young, J.Z., 1971. The Anatomy of the Nervous System of Octopus vulgaris. Oxford
University Press, London, pp. 1690.
Zahl, I.H., Kiessling, A., Samuelsen, O.B., Hansen, M.K., 2009. Anaesthesia of Atlantic cod
(Gadus morhua) effect of pre-anaesthetic sedation, and importance of body
weight, temperature and stress. Aquaculture 295, 5259.
64 P.L.R. Andrews et al. / Journal of Experimental Marine Biology and Ecology 447 (2013) 4664

Zahl, I.H., Kiessling, A., Samuelsen, O.B., Olsen, R.E., 2010. Anesthesia induces stress in Zullo, L., Hochner, B., 2011. A new perspective on the organization of an invertebrate
Atlantic salmon (Salmo salar), Atlantic cod (Gadus morhua) and Atlantic halibut brain. Commun. Integr. Biol. 4, 2629.
(Hippoglossus hippoglossus). Fish Physiol. Biochem. 36, 719730. Zullo, L., Sumbre, G., Agnisola, C., Flash, T., Hochner, B., 2009. Nonsomatotopic organi-
Zahl, I.H., Samuelsen, O., Kiessling, A., 2012. Anaesthesia of farmed sh: implications for zation of the higher motor centers in octopus. Curr. Biol. 19, 16321636.
welfare. Fish Physiol. Biochem. 38, 201218.
Zhu, W., Stefano, G.B., 2009. Comparative aspects of endogenous morphine synthesis
and signaling in animals. Ann. Rev. N.Y. Acad. Sci. 1163, 330339.