De-extinction

Although the term 'de-extinction' was coined around 2012, the underlying aspiration of reversing species loss has deeper roots. In the 1920s and 1930s, German brothers Heinz and Lutz Heck attempted to recreate the aurochs (Bos primigenius) and the tarpan (Equus ferus ferus) through selective back-breeding of domestic cattle and horses. The resulting 'Heck cattle' and 'Heck horses' bear superficial resemblance to their wild ancestors but are widely regarded as phenotypic look-alikes rather than genuine restorations. In 1987, the Quagga Project in South Africa began selectively breeding plains zebras (Equus quagga burchelli) to recover the distinctive coat pattern of the extinct quagga subspecies (Equus quagga quagga), which had disappeared by 1883. By 2005, fifth-generation animals displayed recognisable quagga-like markings.

The modern era of de-extinction began in earnest with advances in molecular biology and ancient DNA (aDNA) research. The milestone cloning of Dolly the sheep in 1996 demonstrated that somatic cell nuclear transfer could produce viable mammals, sparking speculation that the same technique might be applied to recently extinct species possessing preserved tissue samples.

The Pyrenean ibex (Capra pyrenaica pyrenaica), commonly called the bucardo, became the first—and as of early 2025, the only—extinct animal to be brought back via cloning. The last living bucardo, a female named Celia, died on 6 January 2000 in the Ordesa National Park in Spain. Prior to her death, researchers had preserved skin biopsies and cultured fibroblast cells by cryopreservation. In experiments described in a 2009 paper published in Theriogenology (Folch et al.), 285 embryos were constructed using SCNT, of which 57 were transferred to surrogate mothers (domestic goats and ibex–goat hybrids). Only one pregnancy was carried to term, and the resulting kid was born alive on 30 July 2003. The clone died within minutes of birth due to a severe lung defect—an extra lobe in the left lung caused fatal atelectasis. Although the bucardo has sometimes been called 'the first de-extinction,' the organism was technically a faithful genetic replica of the extinct subspecies rather than a proxy, which has prompted debate over whether cloning should properly be classified as de-extinction or as a form of assisted species recovery.

Back-breeding (Selective Breeding): This approach selects individuals from extant populations that display phenotypic traits resembling those of a target extinct species and breeds them over multiple generations to concentrate these traits. Back-breeding can only work when living descendants or close relatives carry ancestral trait variation. The Quagga Project and the Taurus Programme (for aurochs restoration) are ongoing examples. Limitations include the inability to recover genetic sequences lost since domestication and the risk of inbreeding depression.

Cloning via Somatic Cell Nuclear Transfer (SCNT): Cloning requires intact viable cells from the extinct organism. The nucleus from a preserved somatic cell is transferred into an enucleated egg cell of a closely related living species, which is then implanted in a surrogate mother. Because it requires preserved cells, SCNT is only applicable to very recently extinct species or subspecies for which cryopreserved tissue exists. The bucardo experiment is the sole completed example. The resulting organism carries the nuclear genome of the extinct species but the mitochondrial DNA of the surrogate's egg donor, creating a mito-nuclear mismatch.

Genome Editing (Precise Hybridisation): The most broadly applicable de-extinction technique involves editing the genome of a closely related living species to express key alleles and phenotypic traits of the extinct target. CRISPR-Cas9 is the primary tool. Researchers sequence the paleogenome of the extinct species, identify functional genetic differences from the living template species, and use multiplexed genome editing to introduce those changes. This produces a hybrid proxy—an organism genetically derived from the template species but expressing selected phenotypes of the extinct species. George Church's laboratory at Harvard Medical School has been a leader in this approach, having edited more than 40 loci in Asian elephant cell lines to introduce woolly mammoth cold-adaptation traits.

The feasibility of genome-editing-based de-extinction depends critically on the quality of ancient DNA. Allentoft et al. (2012) empirically demonstrated that DNA in bone decays according to first-order kinetics with a half-life of approximately 521 years for a 242-base-pair mitochondrial DNA fragment at an effective burial temperature of 13.1 °C. Under ideal preservation conditions (−5 °C), all bonds in a DNA strand would be destroyed after approximately 6.8 million years. In practice, short DNA fragments have been recovered from specimens as old as 700,000 years (a horse bone from Yukon permafrost). This means that de-extinction via genome editing is theoretically possible for species of the late Pleistocene and Holocene but is categorically impossible for species that disappeared in deep geological time—rendering dinosaur de-extinction, as popularised by Jurassic Park, purely fictional. Nuclear DNA degrades at least twice as fast as mitochondrial DNA, further constraining the window of feasibility.

Woolly Mammoth (Mammuthus primigenius) — Colossal Biosciences / Revive & Restore: The woolly mammoth is the flagship de-extinction candidate. Colossal Biosciences, co-founded in 2021 by entrepreneur Ben Lamm and geneticist George Church, has raised over $400 million to create a cold-adapted proxy of the mammoth by editing Asian elephant (Elephas maximus) genomes. In March 2024, Colossal announced the first-ever creation of induced pluripotent stem cells (iPSCs) from Asian elephant tissue—a critical milestone, as iPSCs can theoretically be differentiated into any cell type, including germ cells. In March 2025, the company unveiled the 'Colossal Woolly Mouse,' a mouse with seven simultaneously edited mammoth-derived genes conferring woolly fur and cold-adaptation traits, demonstrating the viability of multiplex editing. Colossal has publicly stated its goal of producing its first mammoth calf by 2028. The ecological justification centres on the 'Pleistocene Park' hypothesis proposed by Sergey Zimov: reintroduced mammoth-like megaherbivores could trample snow and restore grassland ecosystems across the Arctic tundra, increasing albedo and slowing permafrost thaw.

Passenger Pigeon (Ectopistes migratorius) — Revive & Restore: Revive & Restore, the nonprofit founded by Stewart Brand and Ryan Phelan in 2012, has been working to recreate the passenger pigeon, which went extinct in 1914 when the last individual, Martha, died at the Cincinnati Zoo. The project uses the band-tailed pigeon (Patagioenas fasciata) as the template species and plans to introduce passenger pigeon alleles via interspecies primordial germ cell transplantation (iPGCT), since birds cannot be cloned through standard SCNT.

Thylacine (Thylacinus cynocephalus) — Colossal Biosciences / University of Melbourne: Colossal, in collaboration with the TIGRR Lab (Thylacine Integrated Genetic Restoration Research) at the University of Melbourne, is working toward recreating the Tasmanian tiger, which went extinct in 1936. The thylacine genome has been sequenced, and the fat-tailed dunnart (Sminthopsis crassicaudata), a small marsupial, serves as the template species.

Dodo (Raphus cucullatus) — Colossal Biosciences: In 2023, Colossal announced its third de-extinction programme targeting the dodo, which went extinct in the late seventeenth century in Mauritius. The Nicobar pigeon (Caloenas nicobarica) is the closest living relative, and its genome serves as the editing scaffold. In September 2025, Colossal reportedly achieved significant breakthroughs in dodo primordial germ cell research.

Dire Wolf (Aenocyon dirus) — Colossal Biosciences: In April 2025, Colossal announced the birth of three genetically modified wolf pups—Romulus, Remus, and Khaleesi—bearing traits associated with the dire wolf, which went extinct approximately 10,000 years ago. While Colossal described this as 'the world's first de-extinction,' some biologists have questioned whether the resulting organisms are sufficiently distinct from modern grey wolves to constitute genuine de-extinction, and the claim has provoked debate.

Other Programmes: Additional active or proposed de-extinction efforts include the Galápagos tortoise hybrid-backbreeding programme (to recover the Floreana Island giant tortoise, Chelonoidis elephantopus), the heath hen project (Revive & Restore), and the Genetic Rescue Foundation's plans for New Zealand moa species.

In 2016, the IUCN Species Survival Commission published 'Guiding Principles on Creating Proxies of Extinct Species for Conservation Benefit,' the first official framework for de-extinction governance. The document defines de-extinction as the generation of functional proxies—organisms that fulfil the ecological roles of extinct species without being genetically identical to them. The guidelines emphasise that de-extinction should be pursued only when it provides measurable conservation benefit, that proxy species should be reintroduced only within the former range of the extinct species, and that the welfare of surrogate organisms (template and gestational hosts) must be protected. The guidelines also caution that de-extinction must not divert resources from conventional conservation priorities.

Ben Novak (2018) proposed a new classification of endangerment: 'evolutionarily torpid species.' This category applies to species for which no living multi-celled individuals survive but whose reproductively competent single cells (cryopreserved fibroblasts, gametes, or stem cells) persist ex situ. Under this framework, species such as the bucardo, the gastric-brooding frog (Rheobatrachus silus), and the northern white rhinoceros (Ceratotherium cottoni) are not truly extinct but are in a state of evolutionary dormancy—alive at the cellular level but no longer reproducing or evolving. The northern white rhinoceros, with only two surviving females (as of early 2025) and cryopreserved cells representing a dozen genetic lineages, is widely regarded as the species most likely to be recovered from an evolutionarily torpid state through iPSC technology and stem-cell embryogenesis.

Restorative Justice: Some proponents argue that de-extinction represents a form of moral restitution for human-caused extinctions. If humans drove the passenger pigeon or woolly mammoth to extinction through overhunting and habitat destruction, we may have an obligation to restore what we destroyed. Critics counter that the individual organisms harmed are long dead, that the resulting proxies would be novel entities rather than true members of the extinct species, and that moral responsibility diminishes over generations.

Conservation Resource Allocation: Bennett et al. (2017, Nature Ecology & Evolution) argued that spending limited conservation funds on de-extinction could result in a net loss of biodiversity if those resources would otherwise have been directed at saving currently endangered species. Proponents respond that de-extinction attracts private capital (e.g., Colossal's venture funding) that would not otherwise flow into conventional conservation.

Animal Welfare: De-extinction inevitably involves experimental procedures on sentient animals—harvesting eggs, implanting embryos, and gestating potentially nonviable offspring in surrogate mothers of endangered species (such as the Asian elephant). The bucardo clone's brief, painful life illustrates the welfare costs. Browning (2018) argued that cloning can lead to 'miscarriage, stillbirth, early death, genetic abnormality and chronic disease.'

Hubris and the 'Playing God' Argument: Minteer (2014) contended that de-extinction reflects a 'Promethean spirit' that overestimates human capacity to predict and control ecological outcomes. Proponents argue that this objection would equally condemn all conservation technologies.

Authenticity and Identity: Philosophers of biology debate whether a de-extincted organism genuinely belongs to the extinct species. Under the 'historical entity' view of species (Hull, 1978), species are defined by unbroken genealogical lineages; a laboratory-created mammoth may lack such continuity. Others (Campbell and Whittle, 2017; Slater and Clatterbuck, 2018) argue that human-facilitated reproduction does not sever causal continuity any more than IVF severs human continuity.

Moral Hazard: Critics such as Stuart Pimm and Susan Clayton worry that the perceived reversibility of extinction could undermine public motivation to prevent ongoing species loss—a phenomenon sometimes called 'moral hazard' or the 'extinction is not forever' problem.

De-extinction via genome editing cannot produce a faithful replica of an extinct species. Paleogenomes are always fragmentary and must be scaffolded against a living species' reference genome, meaning the exact karyotype and chromosomal synteny of the extinct species are irretrievably lost. Furthermore, the genotype-to-phenotype relationship is mediated by epigenetics, developmental environment, and social learning, none of which can be inherited from a vanished species. A precisely hybridised 'mammophant' reared in captivity without a mammoth herd and mammoth steppe habitat may never express the full behavioural and physiological repertoire of its extinct model. Additionally, DNA synthesis technology remains orders of magnitude too limited for whole-genome construction of eukaryotic chromosomes; the longest synthesised DNA sequence as of 2018 was approximately 770 kilobases, while mammalian chromosomes can exceed hundreds of millions of base pairs.

Despite these constraints, de-extinction research has produced spin-off technologies with direct conservation applications: iPSC derivation protocols for non-model species, advanced multiplexed CRISPR editing, synthetic biology tools for genetic rescue of endangered populations, and recombinant alternatives to horseshoe crab blood for pharmaceutical testing.

De-extinction has captured the public imagination in ways few conservation topics have achieved. Michael Crichton's novel Jurassic Park (1990) and Steven Spielberg's 1993 film adaptation introduced the concept of DNA-based species revival to global audiences, albeit in a science-fiction context involving dinosaurs. The real-world de-extinction movement—particularly Colossal Biosciences' well-funded, media-savvy campaigns featuring the woolly mammoth, thylacine, dodo, and dire wolf—has generated widespread press coverage and sparked renewed public interest in conservation genetics. As of early 2025, Colossal's total funding exceeded $400 million, and CEO Ben Lamm's net worth was estimated by Forbes at $3.7 billion, reflecting the commercial momentum behind the field. Nonetheless, sceptics caution that hype may outpace science, and that the gap between editing genes in cell culture and producing a viable, ecologically functional organism in the wild remains vast.