Feeding, diets, & growth

photograph of octopus Enteroctopus dolfleini in its den
Octopuses are opportunistic predators and eat many different types of molluscan and crustacean prey, while squids, being mainly pelagic, eat mostly shrimps, fishes, and other squids. This section is divided into diets, prey capture, handling & drilling (octopuses only), and growth. Cannibalism is uncommon in west-coast octopuses (only a single report, for O. bimaculatus) but common in squids (many reports for the feisty Humboldt squid Dosidicus gigas and for Doryteuthis opalescens).  See review by Ibáñez & Keyl 2010 Rev Fish Biol Fisheries 20: 123.




Octopus Enteroctopus dolfleini in its den. A few prey items
such as bivalves can be seen on the octopus' left 0.3X

  black dot


  The section on feeding & growth is divided into diets, considered here, and PREY CAPTURE, PREY HANDLING & DRILLING, and GROWTH, considered in other sections.  The topic of diets is further subdivided into a section on octopuses, and one on squids, both presented below.
  black dot


Research study 1

Observation of octopuses Octopus bimaculatus held in aquarium tanks at Scripps Institution of Oceanography with California scorpionfishes Scorpaena guttata reveals that while the fishes readily eat small octopuses, they go out of their way to avoid large ones, especially large ones Octopus bamaculatus stalks a scorpionfishattempting to stalk them.  In response to other threatening stimuli, the fishes will raise their venomous spines, but not against octopuses, where retreat is the preferred strategy.  In total, the authors record 5 successful attacks by Octopus on scorpionfishes (5-33cm in length).  During attacks, the scorpionfishes raise their spines, but whether they penetrate the octopus is unknown.  The authors interpret their observations to suggest that in the field, octopuses likely prey on scorpionfishes, especially on the juveniles.  Taylor & Chen 1969 Pac Sci 23: 311.

NOTE  additionally, 2 specimens of Enteroctopus dofleini, obtained from Puget Sound, Washington, are observed to attack scorpionfishes of 26 and 30cm length, successfully killing and eating one of them

Octopus bimaculatus (50cm arm spread) initiates an attack
on a watchful scorpionfish Scorpaena guttata (25cm length)
Research study 2

photograph of midden remains after feeding of an octopus Enteroctopus dolfleiniOctopuses live in spaces under rocks and in crevices, known as dens because they are semi-permanent homes for their owners.  Not only are the dens of Enteroctopus dofleini in British Columbia adorned with dead and uneaten parts of prey, but many living fishes and invertebrates are present as den associates.  These are far too numerous to list here, but include about 10 species of fishes, 6 species of crabs, several molluscs, and half a dozen types of echinoderms.  Some of these are prey that have been caught and carried back to the den uneaten, such as scallops, and others are scavengers looking for food, such as fishes, sunflower stars, and snails Amphissa columbianaHartwick & Thorarinsson 1978 Ophelia 17: 163.

Midden remains around a den of Enteroctopus dofleini,
consisting mainly of parts of a red rock crab Cancer
. An octopus will sometimes tuck the crab's
appendages neatly back into the upturned carapace 0.7X

  black dot
Research study 3

table of prey items of octopus Enteroctopus dolfleini outside a den in Barkley Sound, British ColumiaThe following are % composition of prey remains in den middens of octopuses Enteroctopus dofleini in the Barkley Sound, British Columbia and Prince William Sound, Alaska. 

British Columbia: 3492 prey remains collected at 117 dens (see Table above Right)  Note that over 60% of the prey are bivalves, represented by 5 species, with the cockle Clinocardium nuttallii ranking first in preference.  This suggests that foraging occurs more commonly on sand/mud than on rock substrata.  Prey that are too large (e.g., moon snails), fastened too tightly (rock scallops, limpets, chitons, abalone), or buried too deeply (geoducs) are less favoured prey for octopuses. Hartwick et al. 1981 Veliger 24: 129.

Alaska: 235 prey remains collected at 42 dens (see Table lower Right).  Note that greater than 75% of the diet are crustaceans, with the majority being helmet crabs Telmessus cheiragonus. Dodge & Scheel 1999 Veliger 42: 260. Photo courtesy Dave Cowles, Walla Walla University, Washington rosario.wallawalla.edu.

table of prey items of octopus Enteroctopus dolfleini outside a den in AlaskaNOTE  the authors note that because about 30% of the prey is eaten elsewhere (e.g., crabs), den remains provide a less than complete picture of an octopus’ diet photograph of crab Telmessus cheiragonus courtesy Dave Cowles, Walla Walla University, Washington

  black dot

photograph of cockle Clinocardium nuttalliiWhat is it about cockles in comparison with other bivalves that makes them a preferred prey of Enteroctopus? Think about the answers provided, then CLICK HERE for explanations.

They are easy to catch. 

They taste good and are large and meaty.

They are easier to open than other bivalves. 

They are easier for the octopus to see. 

Research study 4

photograph of of Octopus bimaculoides courtesy Roger Hanlon, Marine Biological Laboratory, Woods Hole, MassachusettsOctopus bimaculoides in Santa Catalina, California consumes 55 prey species from 3 phyla.  In laboratory preference tests, crabs are the most highly preferred prey, followed by snails, chitons, and limpets, and then by bivalves.  Field diets are greatly influenced by prey availability, and highly preferred dietary items may actally be rare in the field.  In a 4-yr study in which prey remains are collected from over 100 dens and their frequencies of occurrence compared with densities available in the vicinity of the dens, snails are found to comprise 78% of the diet.  These are mostly Chlorostoma (Tegula) aureotincta, but also included are Norrisia norrisi and Astraea (Pomaulax?) undosa.  Another species, T. eiseni, is actually more abundant than Chlorostoma aureotincta in the field, but is bypassed by the octopus for the other species.  Sedentary grazers, such as chitons, abalones, and limpets, comprise only 7% of the diet, while crustaceans (crabs, shrimps, hermit crabs) comprise 7%, bivalves (several species) 6%, and miscellaneous prey 2%.  The author notes that Octopus bimaculoides is a selective predator and not all potential prey encountered may be eaten.  Moreover, an octopus may bypass isolated, but more preferred prey, to go to an area with high availability of a less preferred prey, such as Chlorostoma aureotincta.  An individual may live in the same den for over a month at a time, and with its good memory capabilities it may learn the distributions of its prey species and modify its foraging accordingly. Ambrose 1984 J Exp Mar Biol Ecol 77: 29; Ambrose 1988 Malacologia 29: 23. Photo of Octopus bimaculoides courtesy Roger Hanlon, Marine Biological Laboratory, Woods Hole, Massachusetts.

NOTE  nearby shelled prey are identified by the characteristic drill-holes made by the octopuses

Research study 5

photograph of octopus Enteroctopus dolfleini in its denA more recent study on diets of octopuses Enteroctopus dofleini in Prince William Sound and Port Graham, Alaska shows that 6 prey species make up 80% of den litter, with nearly 30 species of other taxa making up the remainder.  Four of the 6 favoured species are crabs (Cancer oregonensis, Telmessus cheiragonus, Lophopanopeus bellus, and Pugettia gracilis), with the remaining 2 species being scallops (Chlamys hastata and C. rubida).  Overall, 77% of prey remains are crabs and 19% are bivalves.  The remaining 4% is made up of limpets, barnacles, chitons, gastropods, and sea urchins.  No cockles are recorded from Alaskan den-middens.  Sonar-tracking studies show that the octopuses stay within a 150m radius over 44d, and move readily between intertidal and subtidal locations.  Octopuses are more likely to locate their dens in areas where crabs are common.  Interestingly, no prey species is significantly preferred over its estimated abundance in the Alaskan habitats, and only one species, the crab Acantholithodes hispidus, is significantly avoided.  In comparing their results with those from other studies in British Columbia, the authors note that while diet breadth is similar for E. dofleini in Alaska and British Columbia, the identity of major hard-bodied prey is quite different.  They suggest that E. dofleini includes various bivalve and crab species in its diet corresponding with local abundances, while at the same time exercising an element of selectivity.  Overlapping distributions in Alaska of octopuses and sea otters, with some similarity in dietary preferences, suggests to the authors that competition between them may be occurring.  The study is the first on E. dofleini that compares diet selectivity with the availability of live prey species near their dens.  Vincent et al. 1998 Marine Ecology 19: 13.

NOTE  193 den middens at depths of -31 to +1m are examined, yielding 1059 prey items

NOTE  as noted in Research Study 1 above, in British Columbia 29% of litter remains are respresented by cockles Clinocardium nattallii (0% in Alaskan den litter), 27% by Cancer spp. (mostly C. productus, a species comprising only 2% of Alaskan den litter), and 16% by clams Protothaca staminea (only 3% in Alaskan den litter).  Conversely, the crab Telmessus cheiragonus, which makes up 20-30% of Alaskan den litter, occurs in less than 1% of remains in British Columbia

Research study 6

photograph of Octopus rubescens in a bottle courtesy Roland Anderson and Seattle AquariumIn Puget Sound, Washington small-sized red octopuses Octopus rubescens can be found inhabiting discarded bottles on the sea floor.  Up to the time of the publication presented here, natural dens have not been described for this species, nor has natural prey been identified. SCUBA-divers routinely examine discarded containers for interesting things, and the finding of octopuses in shoes, snorkel tubes, bottles, and other containers (e.g., empty barnacle tests) is not unusual.  By placing out empty beer bottles for 66d, then comparing potential prey occupants in these with prey remains in bottles1 inhabited by O. rubescens, researchers are able to identify 5 predominant prey types plus 15 other less-abundant prey types.  The 5 main prey types2 are shown in the Table below.table showing prey types of Octopus rubescens in Puget Sound, Washington

The 15 less-abundant prey types are crustaceans and other molluscs. The authors note that for the octopuses to obtain Alia sp., they would have to travel at least 100m inshore to eelgrass beds to find the snails and then return with them to their bottles.  The possibility that the Alia shells may have contained hermit crabs is considered then dismissed by the authors, as no crab remains are identified in the bottles.  In this geographical area, then, O. rubescens appears to favour a diet of gastropods.  Anderson et al. 1999 Malacologia 41: 455; see also Anderson 2001 The Festivus 33 (2): 17. Photographs of O. rubescens and Kurtziella sp. courtesy Roland Anderson and Seattle Aquarium; photograph of Alia sp. courtesy Linda Schroeder, Pacific Northwest Shell Club, Seattle, Washington PNWSC.

NOTE1   16 bottles occupied by O. rubescens of 8g mean live mass are examined in the study

NOTE2  determined by the fact that there are significantly more of these prey items in occupied bottles than in unoccupied bottles

NOTE3   this indicates in how many of the 16 bottles a certain prey item is found; thus, 13 bottles contain Olivella baetica

Research study 7

lithograph of octopus attacking the Nautilus in Jules Verne's "20,000 leagues under the sea"Attacks by octopuses Enteroctopus dolfleini on humans are rare, but they do happen.  After describing 4 attacks on SCUBA-divers in Washington/British Columbia by E. dolfleini, a consortium of scientists does not discount the possibility that the octopuses involved may have been actually looking for food.  Thus far, however, there are no records of anyone being even wounded by E. dolfleini.  Anderson et al. 2007 The Can Field-naturalist 121: 423. Lithograph from Jules Verne's 1870 novel 20,000 leagues under the sea. Alphonse de Neuville & Edouard Riou.






Not so lucky. An octopus attacks the
with great loss of life

Research study 8

photograph of Octopus rubescens crawling on the hand of an experimentorA study by researchers at Rosario Beach Marine Laboratory on prey selection in Octopus rubescens shows that crabs Hemigrapsus oregonensis are preferred in a ratio of 3:1 over clams Nuttallia obscurata in simple cafeteria-style laboratory tests.  However, because of greater mass in a clam than in a crab, total prey mass consumed turns out to be virtually identical on the 2 diets.  Handling time is significantly greater, by a factor of about 4, when an octopus eats a  crab than when it eats a clam. Overall, an octopus could obtain 10-fold more energy per unit handling time by choosing clams rather than crabs, suggesting that factors other than simply energy input are involved, perhaps taste, digestibility, or nutritional content.  With respect to the last, the researchers show that lipid-absorption efficiency is significantly higher on a diet of crabs (95%) than on a diet of clams (75%).  Moreover, they find that crabs contain more than twice the lipid content than clams.  Since lipids are not an important metabolisable energy source for cephalopods, the authors think that lipids may be primarily needed for manufacturing egg yolk.  In comparison, neither overall energy absorption nor protein absorption differed much between the 2 diets (overall variability of 94-97%).  The authors determine energy budgets for octopuses on each type of diet, but high variability between individuals and inconsistent growth rates lessen the confidence in the data.  The authors present an interesting discussion on optimal foraging in octopuses, both from the customary standpoint of energy requirement, and from the more interesting and likely more important standpoint of nutritional requirement.  Onthank & Cowles 2011 Mar Biol 158: 2795. Photograph courtesy Dave Cowles, Walla Walla University, Washington wallawalla.edu.

NOTE  interestingly, the researchers find that different octopuses have different and possibly varying individual preferences, an individual sometimes eating both prey types over a few feeding sessions, then changing to one or the other prey types for a time

NOTE  because crabs have nowhere to hide and clams need not be dug up in the aquarium test-environment, "handling" times will beless than they would be in the field. The authors get around this to some extent, however, by recording handling times from the time of capture to the time the uneaten bits are discarded (i.e, feeding time)

Research study 9

mapdiagramOne would expect large predators like octopuses to have diet specialisation relating strongly to size and tastiness of prey species available in a particular habitat.  A recent report by researchers from Alaska Pacific University and the Seattle Aquarium on midden content of Enteroctopus dolfleini at 5 locations from Puget Sound to the Aleutian Islands shows that large crabs comprise 75% of all prey items identified in the middens.  Indeed, a single species Cancer productus is preferred at 3 sites, but not at 2 of the Alaska sites (see map).  Diet specialisation is most evident at sites in Saanich Inlet where middens contain only 9 prey species, with C. productus predominating.  Here, populations rather than individuals appear to specialise. In contrast, middens in Prince William Sound contain fewer individuals of smaller prey species (52 prey species in total), and diets are generalised at both population and individual levels.  The illustration at Right shows the extent of changing species composition of middens in Prince William Sound over the years 1997-2011 (3037 prey individuals represented in 440 middens).  The authors conclude that most individual E. dolfleini are generalist feeders.  Scheel & Anderson 2012 Am Malacol Bull 30 (2): 267; for a comparison of diet specialisation of several world species of octopuses, including E. dolfleini, see Mather et al. 2012 Current Zool 58 (4): 597.

NOTE  Cancer productus, Cancer oregonensis, Telmessus cheiragonus, Pugettia gracilis, and Lophopanopeus bellus. Other commonly eaten prey items are clams, including Protothaca staminea and Pododesmus macrochisma



Research study 1

Identification of foods eaten by squids is made difficult by the often fragmented and well-digested state of the stomach contents.  Researchers in the Moss Landing Marine Laboratories, California therefore rely mostly on the presence of paired eye-lenses, mandibles, statoliths, otoliths, polychaete jaws, and so on, to identify prey items in the stomach contents of Doryteuthis opalescens captured by trawling in Monterey Bay, California.  To estimate prey preferences, the researchers compare frequencies of occurrence of prey in the diet with those in the habitat.  Results show that crustaceans predominate in the diets of both small and large squids, with euphausids being most preferred.  A few fishes and gastropods are also eaten.  The authors provide data on stomach emptying rates, not included here.  Karpov & Cailliet 1978 p. 45 In, Biological, oceanographic, and acoustic aspects of the market squid, Doryteuthis (Loligo) opalescens Berry 185pp. State of California, Dept Fish Game, Fish Bull 169.

NOTE  this produces an index of “relative importance”, which by the method used, is zero when frequency eaten is 0% and frequency in the habitat is 0%, and goes up to 10,000 when the 2 frequencies are each 100% (100 x 100, or a “mono-diet”)

Research study 2

A group of Russian scientists review the biology of jumbo squids Dosidicus gigas from 230 previously published articles and some original research. The paper covers a range of topics, but the emphasis here is on diet and growth. The species is large and aggressive and, with a life span of just a year (with carry-over sometimes into the second year; ageing done primarily by ring counts in the statoliths), growth is rapid. The researchers catch and analyse gut contents of 750 individuals. Dietary preferences change with age, with macroplanktonic invertebrates being eaten by juveniles, and mainly fishes and squids by adults. Feeding is mostly at dusk and dawn. Common fishes eaten by adults are photograph of jumbo squid Dosidicus gigas with a half-eaten tunadeep-dwelling lanternfishes and, among squids eaten by adults, other ommastriphid species including D. gigas (mostly juveniles). The squids swim in small schools (20-40 individuals) from surface waters down to about 50m depth. Normal swimming speed is from 5-10 km . h-1, but burst speeds can be up to 25 km . h-1. Nigmatullin et al. 2001 Fisheries Res 54: 9. Photograph courtesy Bob Cranston/GETTY IMAGES, London, Enland, & ARKIVE.

NOTE these include copepods, euphausids, shrimps, pelagic crabs, heteropod molluscs, pelagic juvenile octopuses, fish fry, and other squids

NOTE at highest speeds, such as to escape a predator (adult predators include sharks, swordfishes, marlin, and sperm and pilot whales), jumbo squids are known to leave the water a glide for a time

Giant or jumbo squid Dosidicus gigas feeding on some
sort of tuna fish. It seems unlikely that such a fast-
moving prey could be captured (but not impossible), so
the fish may already have been dead or wounded 0.3X

Research study 3

map showing past and present distributions of Humboldt squids along the west coast of the AmericasUp until the mid-1980s, Humboldt squids Dosidicus gigas occupied a range from Baja California to the tip of South America.  Since then their range has expanded northwards to Alaska (see map).  Study of individuals involved in a mass stranding near Tofino on Vancouver Island, British Columbia in autumn 2009 provides interesting insights into their diets as well as levels of bioaccumulation of shellfish toxins.  Among undigested remnants of the squids’ favoured prey of sardines (Sardinops sagax) and herring (Clupea pallasii) identified by DNA “barcoding”, the authors report the presence of traces of Dungeness crab, sandfish, tomcod, sculpin, salmon, and kelp greenling.  Interspersed are plastic nurdles, monofilament line, various plant materials, rocks, sand, and even a guillemot feather.  As both herring and sardines are known to bioaccumulate paralytic shellfish toxins (PSTs), these and domoic acid (DA) are quantified in the dead squids as possible implicative factors in causing the strandings.  Results show low DA concentrations but high PST concentrations.  The authors do not know if PSTs are a causative factor in the strandings, but note that theirs is the first report of PSTs in Humboldt squids.  In view of squid’s preference for herring, the authors express concern that with continued warming trends and further northward incursions of the squids, commercial herring fisheries may be threatened. Braid et al. 2012 Mar Biol 159: 25. Photograph courtesy Josie Osborne and Raincoast Education Society, Tofino, British Columbia.

NOTE  this technique involves extracting and sequencing DNA in prey samples taken from the stomachs, followed by matching against reference barcodes.  In all, the researchers are able to stranded Humboldt squid Dosidicus gigas courtesy Josie Osbornematch 107 sequences with >99% similarity to species profiles in the Barcode Database

NOTE  these are small plastic pellets used (mainly) in the manufacture of plastic products.  Somehow they escape the manufacturing process and end up as floating marine debris and amongst trash on beaches.  Not only are the pellets eaten by various marine animals, sometimes clogging filtering apparatuses and gut tubes in the process, but they apparently absorb hydrophobic pollutants such as PCBs from seawater, possibly leading to toxicity



Stranded squid Dosidicus gigas on
Chesterman Beach, Tofino. Scale bar = 15cm




Research study 3

photograph of digestive system of a juvenile squid Dosidicus gigasphotographs of prey types eaten by juvenile squids Docidicus gigas in the Gulf of CaliforniaInformation on diets of juvenile squids Dosidicus gigas (11-121mm mantle length) is provided by a group of researchers from colleges and institutions in Baja California, Mexico. Of 150 guts examined, pelagic crustaceans such as euphausids, copepods, amphipods, and various crustacean parts represent the major component (16-52% numerical frequency; see photographs on Right), with fishes, cephalopods, pteropods, bivalves, and polychaetes also being eaten. Larger-sized individuals (60-121mm) have a greater proportion of identifiable remains in their digestive systems than smaller-sized ones. These larger individuals also have a slightly greater variety of prey in their guts as compared with smaller ones, as would be expected. While prey items in juveniles are mostly found in their stomachs versus other gut areas (esophagus, cecum, intestine, and rectum), food matter in paralarvae is mostly found in their cecae. The investigation provides valuable data for a little-studied phase in the life history of this commercially important species. Camarillo-Coop et al. 2013 J Mar Biol Ass UK 93 (7): 1903.

NOTE 36 paralarvae are also examined, but gut contents in all are unidentfiable. For a time after hatching, yolk and possibly a type of mucus-feeding found only in the paralarvae provide nourishment until the youngsters’ attacking skills become developed