Humans cannot reproduce asexually under natural conditions. Every person alive was produced by the fusion of an egg and a sperm, and the molecular architecture of our genome enforces this requirement with striking rigidity. Yet human eggs do occasionally begin dividing on their own, and recent laboratory breakthroughs in mice have shown that the genetic barriers to mammalian asexual reproduction can, with heavy engineering, be partially overcome. The gap between what nature permits and what technology might someday enable is where this question gets genuinely interesting.
Why Mammalian Reproduction Requires Two Parents
The core reason humans need both an egg and a sperm comes down to a phenomenon called genomic imprinting. In mammals, a subset of genes are chemically tagged during the formation of eggs and sperm so that only one parent’s copy is active after fertilization. Some genes work only when inherited from the mother; others work only when inherited from the father. These tags are laid down through chemical modifications to DNA, and they survive the massive reprogramming that happens after fertilization.1PubMed Central. Genomic imprinting in mammals An embryo with two maternal copies of the genome, or two paternal copies, ends up with the wrong dose of these critical genes. The result is developmental failure.
This is not a minor technicality. Imprinted genes regulate growth of the placenta, nutrient transfer between mother and fetus, and brain development. Getting them wrong does not produce a slightly imperfect offspring. It produces an embryo that cannot develop past the earliest stages, or one that forms a disorganized mass of tissue instead of a body. Biparental reproduction is necessary precisely because these parent-specific modifications make the maternal and paternal genomes non-interchangeable.2PubMed. Genomic imprinting is a barrier to parthenogenesis in mammals
When Human Eggs Start Dividing on Their Own
Although a full asexual pregnancy has never occurred in humans, human eggs do sometimes activate without sperm. This spontaneous parthenogenesis has been documented in fertility clinic settings, where eggs collected for IVF occasionally begin dividing before fertilization takes place. In one well-documented case, eggs repeatedly activated on their own across multiple treatment cycles, underwent cell division, and produced structures that looked like early embryos. Every single one arrested before reaching the blastocyst stage, and none could be transferred.3Reproductive BioMedicine Online. A case of recurrent spontaneous parthenogenetic oocyte activation In other documented cases, eggs have shown a single pronucleus or jumped straight to multi-cell stages without normal fertilization.4PubMed Central. Genetic analysis of recurrent parthenogenesis: A case report and literature review
This phenomenon may be more common than researchers once assumed. A review of the literature concluded that spontaneous human parthenogenesis is likely not as rare as case reports suggest, though it consistently fails to produce viable embryos.5PubMed Central. A new hypothesis may explain human parthenogenesis and ovarian teratoma: A review study The machinery to kick-start cell division exists in the egg. What is missing is the paternal half of the imprinting equation, and without it, development collapses.
What Happens When Parthenogenesis Goes Wrong in the Body
When an egg activates parthenogenetically inside the body rather than in a lab dish, the outcome is not a pregnancy but a tumor. The most familiar example is the mature ovarian teratoma, sometimes called a dermoid cyst. These growths arise from a single oocyte that began developing without fertilization and can contain an unsettling variety of tissues: skin, hair, teeth, even rudimentary neural tissue. They are considered parthenogenetic tumors because they carry only a maternal genome.6PubMed Central. Methylation profiles of imprinted genes are distinct between mature ovarian teratoma, complete hydatidiform mole, and extragonadal mature teratoma
There is also a mirror-image pathology. A hydatidiform mole, or molar pregnancy, arises when an egg loses its maternal genetic contribution and the embryo develops with only paternal DNA. This is called an androgenetic conceptus. Instead of forming a fetus, it produces a mass of abnormal placental tissue that can become cancerous if left untreated. Both ovarian teratomas and hydatidiform moles illustrate the same principle from opposite directions: an embryo with only one parent’s genome cannot build a body. It builds something, but that something is a tumor.7PubMed. On human parthenogenesis8PubMed. From human andro- and parthenogenesis (hydatidiform moles and benign ovarian teratomas) to cancer
Animals That Actually Pull It Off
The impossibility of asexual reproduction in mammals stands in sharp contrast to what happens in other vertebrates. Facultative parthenogenesis, the ability of a normally sexual species to occasionally reproduce without mating, has been confirmed in wild populations of several vertebrates. Smalltooth sawfish in Florida were the first wild vertebrate documented to have living parthenogenetic offspring in a sexually reproducing population.9PubMed. Facultative parthenogenesis in a critically endangered wild vertebrate Two species of pit viper, the copperhead and the cottonmouth, have also produced parthenogenetic offspring in the wild, suggesting this ability is more widespread in reptiles than anyone previously thought.10PubMed Central. Facultative parthenogenesis discovered in wild vertebrates
Birds can do it too, at least under unusual circumstances. California condors in a captive breeding program produced two chicks whose DNA showed no paternal contribution. Both were male (ZZ sex chromosomes, as expected for avian parthenogenesis) and were homozygous for all markers inherited from their mothers. The remarkable detail is that these females were housed with fertile males they had previously bred with. They produced parthenogenetic offspring even though sexual reproduction was available to them.11PubMed Central. Facultative Parthenogenesis in California Condors
Why can sharks, snakes, and condors do what mammals cannot? The answer circles back to imprinting. Reptiles and birds either lack genomic imprinting entirely or use it far less extensively. Without that molecular lock requiring contributions from both parents, an egg that spontaneously begins developing has a fighting chance of producing a viable organism. Mammals evolved imprinting roughly 150 to 200 million years ago, and with it came a hard requirement for biparental reproduction that no mammal has naturally overcome.
Engineering Parthenogenesis in Mice
If imprinting is a lock, researchers have been hunting for the key. In 2022, a Chinese team reported the birth of live mice from single unfertilized eggs. They accomplished this by using gene-editing tools to rewrite the chemical tags at seven imprinting control regions in the egg, essentially mimicking the presence of a paternal genome where none existed. The modified parthenogenetic embryos were transferred to surrogate mothers and developed to full term, producing viable offspring.12PubMed Central. Viable offspring derived from single unfertilized mammalian oocytes
This was a proof of concept, not a recipe for human application. The number of imprinting edits required was substantial, the efficiency was low, and the health outcomes of the resulting mice were not fully characterized over their lifespans. But the experiment demolished the idea that mammalian parthenogenesis is inherently impossible. It showed that the barrier is specific and identifiable: a defined set of epigenetic marks that can, with enough precision, be overridden.
Two Fathers, No Mother
Even more exotic than motherless reproduction is fatherless reproduction involving two males. In 2018, a team at the Chinese Academy of Sciences produced bimaternal mice (two mothers, no father) by deleting three imprinting regions in certain stem cells. They also produced bipaternal mice (two fathers, no mother) by deleting seven imprinting regions. The bimaternal mice grew to adulthood and were fertile. The bipaternal mice were born alive but died within days.13PubMed. Generation of Bimaternal and Bipaternal Mice from Hypomethylated Haploid ESCs with Imprinting Region Deletions
Subsequent work pushed the bipaternal approach further. A 2025 study described the generation of fully adult bipaternal mice by implementing 20 genetic modifications affecting hundreds of imprinting-related genes. These animals survived to adulthood through multiple technical routes, including somatic cell nuclear transfer. The sheer scale of the engineering confirms how deeply imprinting resists single-sex reproduction in mammals, while also confirming that the resistance can be beaten with sufficient effort.14Cell Stem Cell. Generation of viable bi-paternal mice through targeted imprinting modifications
An even earlier proof-of-concept used a different strategy entirely. Researchers took male mouse cells, reprogrammed them into pluripotent stem cells, and then selected for rare subclones that had spontaneously lost the Y chromosome. These XO stem cells were used to generate female chimeras, which were then bred with a second male. The resulting offspring carried genetic material from two fathers and no mother, although the route required going through a female intermediary body.15PubMed Central. Generation of viable male and female mice from two fathers
Cloning and Its Limits
Somatic cell nuclear transfer, the technique that produced Dolly the sheep in 1996, is another form of reproduction without sexual combination. A donor cell’s nucleus is placed into an egg whose own nucleus has been removed. The egg’s cytoplasm reprograms the donor DNA, and the resulting embryo is genetically nearly identical to the donor. In 2018, researchers successfully cloned macaque monkeys this way, marking the first time primates had been cloned by SCNT. They achieved blastocyst rates of roughly 45 to 73 percent depending on donor cell type, though the rate without their epigenetic optimization step was around five percent.16Cell. Cloning of Macaque Monkeys by Somatic Cell Nuclear Transfer
Even in species where SCNT works, the process is plagued by epigenetic errors. A quarter-century of research since Dolly has shown that the oocyte can reprogram some epigenetic information in the donor nucleus but not all of it. Canonical imprinting marks, the DNA-methylation-based tags from the donor’s own parents, survive transfer intact, which is actually helpful. But another layer of imprinting based on histone modifications gets erased, leading to abnormal gene expression in the placenta and other tissues.17PubMed. 25th ANNIVERSARY OF CLONING BY SOMATIC-CELL NUCLEAR TRANSFER: Epigenetic abnormalities associated with somatic cell nuclear transfer Cloned embryos also show aberrant methylation patterns at repetitive DNA elements compared to normally fertilized embryos.18PubMed Central. Epigenetic reprogramming in the mammalian embryo: struggle of the clones
No one has cloned a human, and it remains illegal in every jurisdiction that has addressed it. But the technical trajectory from sheep to monkeys to the editing tools now available means the barrier is increasingly ethical and legal rather than purely biological.
Synthetic Embryos and Lab-Made Gametes
Beyond parthenogenesis and cloning, a third frontier is emerging: building embryo-like structures from stem cells entirely outside the body. Researchers have coaxed human skin cells into forming blastoid structures, sometimes called iBlastoids, by activating four reprogramming factors. These structures resemble human blastocysts in shape and gene activity, and some have shown early signs of implantation-like behavior.19PubMed Central. A new era of stem cell and developmental biology: from blastoids to synthetic embryos and beyond Other groups have produced peri-gastruloids from human stem cells, structures that self-organize into both embryonic and extraembryonic tissue layers and model stages of development just after implantation.20PubMed. Modeling post-implantation stages of human development into early organogenesis with stem-cell-derived peri-gastruloids
Separately, in vitro gametogenesis aims to produce functional eggs and sperm from induced pluripotent stem cells. If successful, this could allow someone to generate gametes from their own skin or blood cells. In mice, the approach has already yielded functional eggs that produced live pups. In humans, progress has been much slower, but the research has advanced considerably in the past decade.21PubMed Central. Modelling in vitro gametogenesis using induced pluripotent stem cells: a review If someone could produce both eggs and sperm from their own cells, self-fertilization would become theoretically possible, though it would be closer to extreme inbreeding than to true asexual reproduction.
Parthenogenetic Stem Cells and Medicine
While parthenogenesis cannot produce a viable human, it turns out to be useful for medicine in a different way. Researchers have derived stem cell lines from human parthenogenetic blastocysts, embryo-like structures grown from chemically activated eggs. Six such lines were reported in one study, all showing normal female chromosomes, the ability to form all three embryonic tissue types, and crucially, a genetic match to the egg donor’s immune profile.22PubMed. Patient-specific stem cell lines derived from human parthenogenetic blastocysts
The appeal here is immunological. Because parthenogenetic embryos carry only maternal DNA, the stem cells derived from them tend to be homozygous for immune-compatibility markers. A bank of such cells from a relatively small number of donors could provide immunologically matched tissue for transplantation to a large segment of the population, sidestepping the rejection problem that plagues organ and tissue grafts.23PubMed Central. Concise review: parthenote stem cells for regenerative medicine: genetic, epigenetic, and developmental features The very feature that makes parthenogenesis fail as reproduction, missing half the genome’s diversity, turns into an advantage for regenerative medicine.
Legal and Ethical Boundaries
Regulatory frameworks around the world have generally been written to prohibit human reproductive cloning, but many are broad enough to cover parthenogenesis and related technologies as well. Hong Kong’s Human Reproductive Technology Ordinance explicitly bans embryo cloning, nucleus substitution, and parthenogenesis as a reproductive method.24PubMed Central. Human embryo cloning prohibited in Hong Kong Australia’s legislation targets the creation of a “genetic copy” but has been critiqued for lacking clarity about whether edge cases like parthenogenesis fall within its scope.25PubMed Central. What exactly is an exact copy? And why it matters when trying to ban human reproductive cloning in Australia
Most countries that have addressed the issue treat any attempt to bring a cloned or parthenogenetic human embryo to term as prohibited. The ethical objections are numerous: safety risks to the resulting child, the unknown long-term health consequences of epigenetic manipulation, concerns about commodification of reproduction, and deeper philosophical questions about genetic identity and diversity. Even among researchers pushing the boundaries of what is technically possible in mice, there is broad agreement that translating these experiments to human reproduction would be premature and irresponsible at this stage.
Sesquizygotic Twins and Other Genetic Curiosities
The boundary between sexual and asexual reproduction gets blurry in some rare human genetic oddities. Sesquizygotic twins, documented only twice, appear to result from a single egg being fertilized by two separate sperm, followed by division into two embryos. These twins share all of their mother’s DNA but only part of their father’s, placing them somewhere between identical and fraternal.26PubMed. The New Sesquizygotic Twins and More: Exotic Twin Types This is not asexual reproduction, but it illustrates that human fertilization is not always the tidy one-egg-one-sperm process textbooks describe. The cellular machinery of the egg is capable of surprising behaviors when the usual controls slip.
Chimeras, individuals carrying cells from two genetically distinct embryos that fused early in development, represent another way human reproduction occasionally departs from the standard script. None of these oddities amount to asexual reproduction, but they reveal a flexibility in early human development that complicates any simple story about how we reproduce. The machinery is more versatile than the outcomes suggest, constrained not by a lack of capability but by the imprinting system that channels that capability in one specific direction.