Note that I’m not referring to regeneration in the ecological sense (i.e. recovery of communities). I will also only talk about animals in this post, for the simple reason that regeneration in plants is commonplace and a routine part of their physiology. Of course, any animal with eyes also has regeneration ability, as the photoreceptors have to regenerate after the reaction with light. I will also be ignoring this. Basically, only regeneration of actual body parts in animals will be discussed here (see picture above; Bely, 2010). You know, the kind you know from lizard tails (or, even more interestingly, rabbit ears: if you punch a hole through them, the hole regenerates!). And as the myriad stories of resurrections and the dead coming back to life from folklore testify, humans have long found regeneration fascinating, most probably because we’re shitty at it, so it’s a genuinely interesting topic.

Related reading is my post on stem cells. To avoid the typical meandering I subject my readers to, I will not be talking about individual organs – for example the regeneration of snail brains (Matsuo et al., 2010). I will not go into the genetic pathways involved for the same reason; that can be requested though.

The picture above shows the taxonomic distribution of regeneration among animals (Bely, 2010).

When it comes to the topic of regeneration, planarian flatworms (Platyhelminthes: Planaria) are the most prominent taxon (Salo, 2006). The picture above (Reddien & Sánchez Alvarado, 2004) shows some of their regeneration potential: the line shows the amputation, grey regions are original tissues and white regions are regenerated, and it’s been known for over 2 centuries (e.g. Randolph, 1897) that they can regenerate a complete body from just a fragment of their original body (1/297th of the original body!). Their regeneration ability is granted to them by large populations of mesenchymal totipotent stem cells (Gschwenter et al., 2001), termed neoblasts, that make up 20-30% of the entire cell numbers of an average adult; even if you isolate a small population of these cells, a fully functioning flatworm adult will grow, simply because it is very likely that these neoblasts are within that part of the animal, since all that happens is that the neoblasts head to the wound site, forming a colourless structure called a blastema (basically, it’s just neoblasts wrapped in epifermal cells), out of which the rest of the animal regrows (Reddien & Sánchez Alvarado, 2004). The only two places of the animal with no neoblasts and that therefore cannot regenerate when isolated are an area in front of the photoreceptors and the pharynx (the elongate white area at the center of every planarian above). Anywhere else is fine though, and this is quite outstanding because planarians are relatively complex organisms, with a head complete with cerebral ganglia, several sensory organs (for light, chemosensation and water flow), as well as separate digestive and excretory systems, an epidermis and a mesenchymal gastrovascular system (the parenchyma, where the neoblasts are housed) – in other words, derivatives of all three germ layers are well-developed.

A sidenote on neoblasts. While it is now accepted that neoblasts are original stem cells, in accordance with the neoblast theory of Wolff & Dubois (1947), this was not always the case, with another hypothesis stating that these neoblasts are actually de-differentiated cells, i.e. differentiated cells that secondarily became undifferentiated (Flickinger, 1964). That hypothesis has been thoroughly crushed by the weight of opposing evidence (as well as ambiguity in Flickinger’s methodology). Neoblasts are readily recognisable, having few mitochondria, free ribosomes and a nucleus consisting mostly of uncondensed chromatin, and they are now easier than ever to label for detailed histological studies (Newmark & Sánchez Alvarado, 2000).

Sponges also have cells that act similarly to neoblasts called archaeocytes. These are amoeboid cells with multiple roles, including food and waste product transport, but are also totipotent and involved in regeneration and oogenesis. It might seem strange that what are nothng more than biological water pumps would need to regenerate, but sponges are actually quite active animals and regeneration is crucial for them to maintain ecological dominance wherever they’re found: if they get broken up, they can grow back and reattach themselves to the substrate (Wilkinson & Thompson, 1997), and when they get disfigured by predators, they can reconstruct their original shape to get back to optimal pumping (and thus feeding) efficiency (Walters & Pawlik, 2005). It goes without saying that sponge regeneration is not a simple process, and the overall capability differs depending on the species and the nature of injury. Some wounds will heal completely, while others will not be healed, especially if the injury is likely to happen again (often the case when encrusting predators are nearby, such as brittlestars). Of course, some injuries cannot be recovered from, e.g. getting crushed during a storm or something, but that should be obvious (sponges are awesome, but not invincible). The fact that this discretion is applied hints that regeneration is not free in sponges and that using the archaeocytes for this purpose entails a high energetic cost.

No discussion of animal regeneration is complete without a mention of Hydra. These are small (1-3 mm), solitary polyps. Being cnidarians, they only have two germ layers, separated by an acellular mesoglea (the jellyfish homologue is the ‘bell’). They’re very simply built: a gastric cavity within a stalk, and tentacles around the single opening. They have no sophisticated organ systems except for a diffuse nerve net. However, they are remarkable for one feature: for all practical intents and purposes, they are immortal. In other words, they have a near unlimited regeneration potential. This is achieved by continuously renewing stem cell populations (Galliot & Schmid, 2002), allowing them to, like the planarians, regenerate completely even when cut up into tiny pieces (Böttger & Alexandrova, 2007). As the picture above (put together from Watanabe et al., 2009) shows, there are three stem cell types found evenly distributed along the body (those green dots are different stem cells), that give it this regeneration potential.

This is actually a feature of most cnidarians. Their epithelium, despite consisting of differentiated cells, can always grow back. This has been observed even in tissues where no mitosis occurs, such as the tentacles (jellyfish: Lesh-Laurie et al., 1991; Chapman, 1966; anemones: Chia, 1976). In fact, cnidarians depend on their ability to regenerate for their reproductive biology: their gametes are released out into the water through the rupture of the body wall, which then regenerates. The same is true for ctenophores (which are not cnidarians, in case you read that somewhere).

Similarly, in some ascidians, regeneration is a routine part of the vegetative life cycle, whereby the solitary individual will bud off a part of itself, which will then regenerate a series of zooids elsewhere. Bryozoans undergo a similar procedure when reproducing asexually: stem cells there can either remain totipotent at budding, or they differentiate into germ cells for sexual reproduction.

Another group of animals that has impressive regeneration capabilities is the crinoids (Candia Carnevalli, 1993). Their arms are easily decapitated and regenerate readily. In one genus, Linckia, a single arm can regenerate a complete animal.

Their echinoderman relatives, the ophiuroids, also have numerous representatives capable of regeneration, being able to regenerate their discs when lost. This can lead to some taxonomic confusion: the regenerated discs differ in their morphology and the individual could be mistaken for a new species. This problem isn’t unqiue to echinoderms though.

Interestingly, marine polychaete, especially nereidids, regeneration has shown some very tight negative correlations with lunar cycles. Reproduction in these polychaetes is related to the moon, so this hints that extreme gamete production might have a negative effect on regenerative ability (Lawrence & Soame, 2009), most probably due to energetic costs.

Of course annelids in general are commonly known for their regeneration: I’m sure any child who’s played with earthworms is aware of it. But it’s also widespread among all annelids. For example, serpulids can regenerate their operculum when it’s damaged (ten Hove, 1970).

Some insects are also capable of regenerating appendages (first reported by Bodenstein, 1933). In some, it’s even vital to their survival. For example, immature stick insects can sever their leg (autotomy) and then regain it after 2-3 moults (Carlberg, 1992). However, this amputation comes at the cost of smaller wings and therefore a reduced flight ability in the adult, as the energy that would have gone into building the wing is used to building the new leg (Maginnis, 2006).

Repair during moulting can also be seen in the fossil record! Most trilobite fossils are not of the actual dead organism, but merely its moulted exoskeleton (this is also why we can reconstruct trilobite ontogenetic series with such confidence). Take the picture of Greenops widderensis above (McNamara & Tuura, 2011) as an example. There are two clear signs of regenerating injury here: in 2, compare the spines on the segments numbered 2-5. One side has spines that are just starting to regrow. In 3, you have to look to the head. See that spine that extends backwards over the side of the body? That’s the genal spine, and on one side, it’s noticeably shorter, again because it broke off and is being grown back with every moult. The most probably cause of these injuries is predation, although trilobites are also notorious for having a rather faulty moutling mechanism where any mistake or disturbance can be costly, so the spines could have been lost during moulting too.

One arthropod group where regeneration needs to be more thoroughly studied is the spiders. Just as lizards cut off their tail to escape predators, spiders can cut off a leg for the same purpose. Unlike the phasmids though, this entire leg can be regenerated in a single moult. How this happens is currently unknown. I’m sure there’s a lab somewhere that is looking for postdocs/Ph. D. candidates to study this.

Of course, some insects can also regenerate a complete appendage in a single moult, but this is unique to certain species. For example, the cockroaches Leucophaea maderae can regenerate a leg in one go (Bohn, 1970), as can Blattella americana (O’Farrell et al., 1956), but Blabera craniifer needs several moults (Bullière, 1967). As with the trilobite, in such cases, each moult makes a slightly larger appendage, eventually getting one that is more correctly-sized (but generally not the length of a regular appendage).

From a history of biology perspective, arguably the most significant regenerating animal is the newt. The zoologist Hans Spemann (1869-1941) performed groundbreaking transplantation research into the regeneration of newt extremities. The most significant of these was when he transplanted a severed arm back onto the body until it was functioning as normal. He then cut this arm again, and arms regenerated both from the cut-off stump and from the part that stayed on the body. This, and other such experiments, layed the foundations for research into the molecular mechanisms underlying development, and such regeneration experiments are still used nowadays in developmental biology.

Amphibians in general are quite remarkable with their regeneration abilities. Salamanders can regenerate the eye lens using cells from the iris. The surprising thing here is that the iris is a muscle, i.e. a different tissue type. The process involved here is dedifferentiation and redifferentiation into the target cell type (Brockes & Kumar, 2005). Needless to say, this is under intense research as it could have potential therapeutic applications. Lizards and snakes can only do this for their tails. Iguanas are an exception – when they sever an appendage, only a small bud can grow back. Similarly, chameleons will never cut off their highly-specialised tail, since it’s more or less their livelihood.

In those lizards and amphibians that do drop their tail, the mechanisms may simply be a mechanical one. However, some have tails specialised for breaking off, with shortened vertebrae and muscular architecture allowing the break the be constricted to minimise the amount of tail to be cut off.

In mammals, regeneration of body parts is, to my knowledge, unknown. At most, they can regenerate tissue. The reason may be due to the different pathways that are activated by a wound. In the lizards and amphibians, the syncytial (i.e. have multiple nuclei) muscle cells dedifferentiate and form a blastema at the wound site and begin regenerating the limb. In mammals, no dedifferentiation occurs. Instead, special stem cells called satellite cells are awakened and go into muscle tissue, helping it regenerate at the wound site (Collins et al., 2005). Salamanders have these as well (Morrison et al., 2006), but can regenerate the entire limb because of the ability of their muscle cells to dedifferentiate.

Of unique regeneration, i.e. singular species that have the ability to regenerate this not being a feature of their taxon, we can look to Loxosomella antarctica, which can regenerate and thus achieve a very long life-span. L. antarctica is a kamptozoan, and no other regenerating kamptozoans have been documented yet.

To conclude, I want to say a couple of words about the origin of regeneration. While it may be easy to simply chalk regeneration up as an adaptation that evolved specifically, such an explanation is a bit short-sighted. As I said with the newt example, we use animals’ natural ability to regenerate as a source of evidence for how morphogenesis in the embryo works. The reason we can even do this is because regeneration and morphogenesis make use of the same/very similar pathways, using stem cells; regeneration is not a new ability evolved in the adult, but one that is kept from the embryonic days. At first glance then, it may seem absurd that many animals have lost this ability, if they had it naturally all along – but remember that regeneration is not free and other life history traits suffer (such as wing size, therefore flight ability, in phasmids, etc.).

Besides purely energetic costs, there may also be sexual costs. This is best seen in various arthropods. Male wolf spiders have large legs with tufts of hair they use in competition and mate attraction. However, regenerated legs don’t have these colourful hairs and they’re smaller, meaning a much smaller chance of reproduction (Uetz et al., 1996).

The case of the male fiddler crab (Uca spp.) is even more drastic. One of their claws is grossly enlarged – up to 40% of their body mass (Crane, 1975) – and is used both as a weapon in male-male fights and as a signal for females (by waving it around while standing next to their burrows). The larger the claw, the more likely the male will win the fight (Reaney & Blackwell, 2007). However, the large claw is obviously a burden. Complications during the fight or during moulting can always occur, as can predator attacks, and the claw is the first thing to get cut off in such cases. The regenerated claw is noticeably smaller and thinner (Yamaguchi, 1973) and much less adequate for its use as a display and competitive structure (Blackwell et al., 2000). The root cause of this smaller claw (and non-decorated legs in the wolf spider) is the sheer amount of energy required to rebuild an original appendage – it’s best to do damage control and get a small chance at a mate than risk remaking an original one and die before mating.

And finally, a small word on what one should not look for when looking for animals that can regenerate: eutely, i.e. a constant number of cells (e.g. the adult hermaphroditic Caenorhabditis elegans has exactly 959 cells, male has 1031, with no variations and exceptions). The reason is obvious: these animals do not have stem cells left. All aschelminths are eutelous, and therefore cannot regenerate; at most, they can heal wounds. As proof of this principle, take the annelids: all annelids have extensive regenerative abilities… except the leeches – the leeches are also the only annelids with a determinate cell number.

I hope this post gave you an idea of just how widespread regeneration is and to see that regeneration, while sounding like a magical ability, does not equal automatic evolutionary success. I made a purposeful effort to constrain the arthropod section, without going into the whole history of research that has gone on with regenerating insect legs – which includes things as cool as insects with supernumary legs! It pains me when I have to do that, but the upside is that I won’t get complaints about Gouldian sidetracks.


Bely AE. 2010. Evolution of animal regeneration: re-emergence of a field. Trends in Ecology & Evolution 25, 161-170.

Blackwell PRY, Christy JH, Telford SR, Jennions MD & Passmore NI. 2000. Dishonest signalling in a fiddler crab. Proceedings of the Royal Society B 267, 719-724.

Bodenstein D. 1933. Beintransplantationen an Lepidopterenraupen. II. Zur Analyse der Regeneration der Brustbeine von Vanessa urticae-Raupen. Wilhelm Rouc’s Archiv für Entwicklungsmechanik der Organismen 130, 747-770.

Bohn H. 1970. Interkalare Regeneration und segmentale Gradienten bei den Extremitäten von Leucophaea-Larven (Blattaria) – I. Femur und Tibia. Archiv für Entwickulungsmechanik der Organismen 165, 303-341.

Böttger A & Alexandrova O. 2007. Programmed cell death in Hydra. Seminars in Cancer Biology 17, 134-146.

Brockes JP & Kumar A. 2005. Appendage regeneration in adult vertebrates and implications for regenerative medicine. Science 310, 1919-1922.

Bullière D. 1967. Étude de la régénération chez un insecte blattopteroide, Blabera craniifer Burm. (Dictyopère) – I. Influence du niveau de la section sur le régénération de la patte métathoracique. Bulletin de la Société Zoologique de France 92, 523-536.

Candia Carnevalli MD, Lucca E & Bonasora F. 1993. Mechanisms of arm regeneration in the feather star Antedon mediterranea: Healing of wound and early stages of development. Journal of Experimental Zoology 267, 299–317.

Carlberg U. 1992. Cost of autotomy in the Phasmida (Insecta). I. Species with low autotomy frequency. Zoologischer Anzeiger 228, 229-237.

Chapman DM. 1966. Evolution of the Scyphistoma. In: Rees WJ (ed.). The Cnidaria and Their Evolution.

Chia FS. 1976. Sea anemone reproduction, patterns and adaptive radiations. In: Mackie GO (Ed.). Coelenterate Ecology and Behaviour.

Collins CA, Olsen I, Zammit PS, Heslop L, Petrie A, Partridge TA & Morgan JE. 2005. Stem cell function, self-renewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell 122, 289-301.

Crane J. 1975. Fiddler Crabs of the World. Ocypodidae: Genus Uca.

Flickinger RA. 1964. Isotopic evidence for a local origin of blastema cells in regenerating Planaria. Experimental Cell Research 34, 403-406.

Galliot, B & Schmid V. 2002. Cnidarians as a model system for understanding evolution and regeneration. The International Journal of Developmental Biology 46, 39-48.

Gschwentner R, Ladurner P, Nimeth K & Rieger R. 2001. Stem cells in a basal bilaterian. Cell & Tissue Research 304, 401-408.

ten Hove TA. 1970. Serpulinae (Polychaeta) from the Carribean: I – the genus Spirobranchus. Studies on the fauna of Curaçao and other Carribean Islands 32, 1-57.

Lawrence AJ & Soame JM. 2009. The endocrine control of reproduction in Nereidae: a new multi-hormonal model with implications for their functional role in a changing environment. Philosophical Transactions of the Royal Society B 364, 3363-3376.

Lesh-Laurie GE, Hujer A & Suchy PE. 1991. Polyp regeneration from isolated tentacles of Aurelia scyphistomae: a role for gating mechanisms and cell division. Hydrobiologia 216/217, 91-97.

Maginnis TL. 2006. Leg regeneration stunts wing growth and hinders flight performance in a stick insect (Sipyloidea sipylus). Proceedings of the Royal Society B 273, 1811-1814.

Matsuo R, Kobayashi S, Murakami J & Ito E. 2010. Spontaneous recovery of the injured higher olfactory center in the terrestrial slug Limax. PLoS One 5, e9054.

McNamara KJ & Tuura ME. 2011. Evidence for segment polarity during regeneration in the Devonian asteropygine trilobite Greenops widderensis. Journal of Paleontology 85, 106-110.

Morrison JI, Lööf S, He P & Simon A. 2006. Salamander limb regeneration involves the activation of a multipotent skeletal muscle satellite cell population. The Journal of Cell Biology 172, 433-440.

Newmark PA & Sánchez Alvarado A. 2000. Bromodeoxyuridine specifically labels the regenerative stem cells of planarians. Developmental Biology 220, 142-153.

O’Farrell AF, Stock A & Morgan JA. 1956. Regeneration and the moulting cycle in Blattella germanica L. – IV. Single and repeated regeneration and metamorphosis. Australian Journal of Biological Sciences 9, 406-422.

Randolph H. 1897. Observations and experiments on regeneration in planarians. Archiv für Entwicklungsmechanik der Organismen 5, 352-372.

Reaney LT & Blackwell PRY. 2007. Temporal constraints and female preference for burrow width in the fiddler crab, Uca mjoebergi. Behavioral Ecology and Sociobiology 61, 1515-1521.

Reddien PW & Sánćhez Alvarado A. 2004. Fundamentals of planarian regeneration. Annual Review of Cell and Developmental Biology 20, 725-757.

Salo E. 2006. The power of regeneration and the stem-cell kingdom: freshwater planarians (Platyhelminthes). BioEssays 28, 546-559.

Uetz GW, McClintock WJ, Miller D, Smith EI & Cook KK. 1996. Limb regeneration and subsequent asymmetry in a male secondary sexual character influences sexual selection in wolf spiders. Behavioral Ecology and Sociobiology 38, 253-257.

Walters KS & Pawlik JR. 2005. Is there a trade-off between wound healing and chemical defenses among Carribean reef sponges? Integrative and Comparative Biology 45, 352-358.

Watanabe H, Hoang VT, Mättner R & Holstein TW. 2009. Immortality and the base of multicellular life: Lessons from cnidarian stem cells. Seminars in Cell & Developmental Biology 20, 1114-1125.

Wilkison CR & Thompson JE. 1997. Experimental sponge transplantation provides information on reproduction by fragmentation. In: Proceedings of the 8th International Coral Reef Symposium.

Wolff E & Dubois F. 1947. La migration des cellules de régénération et les facteurs qui la provoquent chez les triclades. Comptes Rendus de l’Academie des Sciences 224, 1387-1388.

Yamaguchi Y. 1973. Asymmetry and dimorphism of chelipeds in the fiddler crab Uca lactea De Haan. Zoological Magazine 82, 154-158.

Research Blogging necessities :)

Reddien, P., & Alvarado, A. (2004). FUNDAMENTALS OF PLANARIAN REGENERATION Annual Review of Cell and Developmental Biology, 20 (1), 725-757 DOI: 10.1146/annurev.cellbio.20.010403.095114

Galliot B, & Schmid V (2002). Cnidarians as a model system for understanding evolution and regeneration. The International journal of developmental biology, 46 (1), 39-48 PMID: 11902686

Collins, C., Olsen, I., Zammit, P., Heslop, L., Petrie, A., Partridge, T., & Morgan, J. (2005). Stem Cell Function, Self-Renewal, and Behavioral Heterogeneity of Cells from the Adult Muscle Satellite Cell Niche Cell, 122 (2), 289-301 DOI: 10.1016/j.cell.2005.05.010

Leave a Reply