Every cell alive today descended from another cell. This principle, often stated in Latin as “omnis cellula e cellula,” has held up across every branch of biology for more than 150 years. No one has ever documented a cell spontaneously assembling from scratch under natural modern conditions. Yet the rule carries a deep paradox: the very first cells on Earth could not have had cellular parents. That tension, along with a growing list of biological phenomena that blur the neat boundaries of classical cell theory, makes the full answer far more interesting than a simple yes.
The Fall of Spontaneous Generation
For most of human history, people assumed living things routinely sprang from non-living matter. Maggots appeared on rotting meat, mice seemed to emerge from grain stores, and microbes materialized in broths left out overnight. The idea felt obvious. It took a series of clever experiments in the 1800s to dismantle it. Louis Pasteur’s swan-neck flask experiments were the most famous: he showed that broth left open to air but shielded from dust-borne microbes stayed sterile indefinitely, while broth exposed to airborne particles quickly teemed with life.1PubMed. Louis Pasteur (1822-1895) Those experiments overthrew the millennia-old assumption of spontaneous generation and cemented the idea that all life comes from preexisting life as biological dogma.2Frontiers in Synthetic Biology. Historical paradigm shifts in defining life: from spontaneous generation and vitalism to the Pasteurian Wall and the quest for artificial creation
This finding became the third pillar of cell theory, joining the observations that all organisms are composed of cells and that cells are the basic unit of life. It was a powerful statement, and it reshaped medicine, food preservation, and sanitation. But it also drew what some researchers now call the “Pasteurian wall,” a conceptual barrier between chemistry and biology that implied life could never arise from non-life. The wall holds perfectly well in everyday biology. Where it gets complicated is at the very beginning of life’s history, and at the experimental frontiers of modern science.
How Cells Actually Make New Cells
The standard route is division. A cell copies its DNA, grows, and splits into two daughter cells. In organisms with a nucleus, this process involves a carefully choreographed dance where chromosomes line up and separate so that each daughter gets a complete set of genetic instructions. But DNA is only part of what needs to be inherited. A new cell also needs working copies of its internal machinery, and how that happens is less tidy than textbooks sometimes suggest.
Mitochondria, the structures that generate most of a cell’s energy, cannot be built from scratch. They carry their own small genomes, a relic of their origin as free-living bacteria that were engulfed by an ancestral cell billions of years ago.3PubMed. Endosymbiotic theory for organelle origins During cell division, mitochondria are actively distributed to each daughter cell using the cell’s internal skeleton. Cells have evolved specific mechanisms to regulate this partitioning, including cycles of mitochondrial fusion and splitting, active transport along structural filaments, and quality-control systems that destroy defective mitochondria before they get passed along.4PubMed Central. Mitochondrial dynamics and inheritance during cell division, development and disease
The Golgi apparatus, which packages and ships proteins around the cell, presents its own inheritance puzzle. As a cell enters division, the Golgi breaks apart into smaller clusters and a fine haze of tiny vesicles. These fragments are divided between the two daughter cells, which then reassemble functional Golgi stacks from the inherited pieces.5PubMed. Division of the intermediate compartment at the onset of mitosis provides a mechanism for Golgi inheritance The point is that “a cell comes from a preexisting cell” is not just about copying DNA. It means inheriting an entire operating system of membranes, energy factories, and protein-processing equipment that has been continuously maintained since the dawn of cellular life.
Organelles That Can Be Built From Scratch
Not every cellular component strictly follows the “only from preexisting copies” rule. Peroxisomes, small compartments involved in breaking down fatty acids and neutralizing toxic byproducts, can proliferate by growth and division of existing peroxisomes, but they can also form fresh, without a pre-existing peroxisome template. In this alternative pathway, specialized proteins are first routed through the cell’s internal membrane network, the endoplasmic reticulum, then packaged into small precursor vesicles that eventually fuse together and mature into functional peroxisomes.6PubMed Central. De novo peroxisome biogenesis: Evolving concepts and conundrums7PubMed Central. The role of the endoplasmic reticulum in peroxisome biogenesis
Even more surprisingly, work in human cells lacking a key peroxisome-assembly protein called PEX3 showed that when the missing protein was reintroduced, it first appeared on mitochondria, was then pinched off in small mitochondria-derived vesicles, and those vesicles eventually matured into working peroxisomes capable of importing their own cargo proteins.8Current Biology. Peroxisome Biogenesis: Multiple Pathways to Peroxisomes So at the organelle level, “from preexisting” is not always the whole story. Some components can be rebuilt from other cellular structures when the originals are absent. The cell itself still comes from a preexisting cell, but its internal parts have more flexibility than the strict inheritance model would suggest.
When Cells Merge Instead of Dividing
Cell division gets all the attention, but cells also fuse. This is not some rare accident; it is a routine, essential part of how certain tissues form and function. Muscle fibers are built when individual precursor cells called myoblasts fuse together into long, multinucleated fibers. The placenta contains a layer of fused cells, called the syncytiotrophoblast, that forms when individual trophoblast cells merge.9PubMed Central. Dynamin and endocytosis are required for the fusion of osteoclasts and myoblasts Bone-remodeling cells called osteoclasts are also multinucleated, created through the fusion of blood-derived precursors. In each case, the resulting structure has multiple nuclei sharing a single continuous cytoplasm. The process of cell-surface signaling that triggers these fusions is tightly regulated, involving specific scrambling of membrane lipids that flags cells as fusion-ready.10PubMed Central. TMEM16F phospholipid scramblase mediates trophoblast fusion and placental development
Fungi take this even further. Many fungal species form syncytia, continuous networks of cytoplasm containing many nuclei, through two distinct mechanisms. One involves nuclear division without the cell actually splitting in two, yielding a single compartment packed with multiple nuclei. The other involves genuine cell-cell fusion, where two separate fungal cells grow toward each other, merge their membranes, and pool their nuclei into one shared space.11PubMed. Fungal syncytia These fusion events challenge the picture of cells as neatly bordered units that only ever multiply by splitting. Across diverse organisms, the formation of multinucleated structures through fusion, and even the reverse process of carving a syncytium back into individual cells, are well-documented parts of normal life cycles.12PubMed. Cell-cell fusions and cell-in-cell phenomena in healthy cells and cancer
Not All Division Produces Equal Daughters
When we say a new cell comes from a preexisting cell, it is easy to picture two identical copies. That is often not what happens. Stem cells frequently divide asymmetrically: one daughter stays a stem cell, and the other takes on a more specialized identity. This is how tissues maintain themselves throughout your life, balancing the need for a reserve of undifferentiated cells against the constant demand for replacement workers.13PubMed Central. Cellular and molecular mechanisms of asymmetric stem cell division in tissue homeostasis The unequal outcome is driven by the lopsided distribution of specific fate-determining molecules into the two daughter cells during division.14Frontiers in Hematology. Asymmetric cell division of hematopoietic stem cells
Some cells go further and skip division altogether while still copying their DNA. Trophoblast giant cells in the placenta, for instance, undergo repeated rounds of DNA replication without ever splitting. This process, called endoreduplication, is triggered when a specific protein inhibits the machinery needed to enter mitosis, so the cell keeps building up DNA content but never divides.15BioMed Central / Cell Division. Cip/Kip cyclin-dependent protein kinase inhibitors and the road to polyploidy The result is a massive, polyploid cell. It still came from a preexisting cell, but the “division” part of the story has been neatly sidestepped.
The Origin-of-Life Exception
The most fundamental challenge to “all cells come from preexisting cells” is that the principle cannot extend back infinitely. Earth is roughly 4.5 billion years old, and the earliest evidence of life dates to at least 3.5 billion years ago. At some point, the first cell-like structures assembled from non-living chemistry. Abiogenesis, the emergence of life from non-life, is not a violation of modern cell theory so much as the precondition for it. The principle describes how biology works now; abiogenesis describes how it got started.
Researchers studying this transition have built protocell models using simple fatty acid or phospholipid vesicles, inorganic nanoparticles, and even membrane-free droplets formed by mixing peptides and nucleotides. These represent early steps toward understanding how the shift from non-living chemistry to self-sustaining, replicating compartments could have happened.16PubMed. Systems of creation: the emergence of life from nonliving matter One leading hypothesis places this transition at alkaline hydrothermal vents on the ocean floor, where natural gradients in temperature, acidity, and chemical potential, along with networks of mineral-walled pore spaces, could have served as ready-made primitive compartments in which the chemistry of life first organized itself.17PubMed Central. Hybrid organic-inorganic structures trigger the formation of primitive cell-like compartments
Nobody has yet created a living cell from non-living ingredients in a lab. But the gap between “simple chemistry” and “self-replicating compartment” has been steadily narrowing as researchers learn to build increasingly lifelike protocells.
Synthetic Cells and the Minimal Genome
If you cannot yet build a cell from raw chemicals, can you at least design one from scratch? In 2016, a research team created JCVI-syn3.0, a bacterium with a completely synthetic genome containing only 473 genes across 531 kilobase pairs, smaller than the genome of any self-replicating cell found in nature.18PubMed. Design and synthesis of a minimal bacterial genome This near-minimal cell was built by iteratively stripping away non-essential genes from a larger synthetic genome to identify the smallest gene set still compatible with autonomous growth and division.19PubMed Central. Minimal Cells-Real and Imagined
The catch, and it is a big one, is that the synthetic genome was transplanted into an existing cell. The host cell’s membranes, ribosomes, and metabolic machinery did the work of reading the new instructions and booting up a living organism. So even the most radical synthetic biology achievement to date still relied on a preexisting cell as the chassis. The genome was designed by humans; the cell was not. More recent work has pushed further by encapsulating cell-free protein synthesis systems inside artificial vesicles made of folded proteins, creating protocells that can manufacture their own membrane components and grow autonomously.20PubMed. Cell-Free Protein Synthesis of Fusion-Protein Building Blocks Enables Autonomous Growth in Globular Protein Vesicle Protocells These are not cells in the full biological sense, but they edge closer to the threshold of self-sustaining, self-building compartments that do not depend on a pre-existing living cell.
Giant Viruses and the Blurry Edge of Cellularity
Viruses are not cells, and the standard textbook line is that they need a host cell to reproduce. But giant viruses of amoebae have shaken up the neat division between cellular and non-cellular life. Mimivirus, the first giant virus discovered, has a particle packed with more than 100 proteins and messenger RNA, and a gene repertoire that rivals some bacteria, including genes for translation components that were previously thought to belong exclusively to cells.21PubMed Central. Mimivirus: leading the way in the discovery of giant viruses of amoebae After entering an amoeba by being swallowed, Mimivirus sets up an enormous cytoplasmic structure called a viral factory, a dedicated zone where viral DNA replicates and new virus particles are assembled in a process that visually resembles cell division more than it does a conventional viral infection.22PubMed Central. Ultrastructural characterization of the giant volcano-like virus factory of Acanthamoeba polyphaga Mimivirus
Giant viruses do not reproduce independently, so they are not true cells. But they blur the boundary enough to unsettle the clean categories that cell theory was built on. If a virus can carry its own translation machinery and build a factory that functions almost like a second nucleus inside its host, the question “what counts as a cell?” gets harder to answer than it used to be.
Replicators That Are Not Cells at All
Prions offer perhaps the strangest challenge to tidy biological categories. A prion is a misfolded protein that can force normally folded copies of the same protein to adopt its abnormal shape. This propagation is autocatalytic: one misfolded molecule converts others, and the newly converted molecules do the same, creating an exponentially growing chain of misfolded protein. Laboratory experiments have confirmed this cascade, showing that misfolded prion protein can be serially amplified through fresh brain tissue, with newly formed misfolded protein converting normal protein just as efficiently as the original infectious material, yielding roughly 300-fold amplification over 100 rounds.23PubMed Central. Autocatalytic self-propagation of misfolded prion protein
Prions are not cells. They contain no DNA, no membranes, and no metabolic machinery. But they self-propagate, they carry heritable information in their three-dimensional shape, and they cause diseases like scrapie and Creutzfeldt-Jakob disease. They remind us that the principle “all cells from cells” is a statement about cells specifically, not a universal rule about self-replicating biological entities.
Extracellular Vesicles and Intercellular Smuggling
Cells also produce smaller non-cellular packages that carry biologically active cargo between cells. Exosomes, a type of extracellular vesicle averaging around 100 nanometers in diameter, originate inside a cell’s internal membrane compartments and are released into the surrounding environment. They carry nucleic acids, proteins, lipids, and metabolites that reflect their cell of origin, and they can be taken up by other cells, altering those cells’ behavior.24PubMed Central. The biology, function, and biomedical applications of exosomes Extracellular vesicles do not reproduce on their own and are not cells. But they represent a way that cellular material, including functional genetic information, can move between cells without cell division. In this sense, cells share parts of themselves in ways that do not fit neatly into the division-based inheritance model.
Horizontal Gene Transfer and the Tangled History of Cells
Even if every cell comes from a preexisting cell, the genetic content of those cells does not always follow a tidy parent-to-offspring path. Microbes routinely pick up DNA from unrelated organisms through horizontal gene transfer, a process so pervasive that it has constantly reshaped genomes throughout evolutionary history and thrown into confusion the traditional tree-of-life model, which assumes genetic information flows mainly downward from parent to descendant.25PubMed Central. Horizontal Gene Transfer and the History of Life
The effects go beyond bacteria swapping resistance genes. In eukaryotic microbes, endosymbiotic gene transfer, where genes move from an engulfed organism into the host’s own genome, has been a major force. Analysis of diatom genomes, for example, found that roughly a quarter of the genes encoding membrane transport proteins appear to have originated from ancient algal endosymbionts rather than being inherited vertically.26PubMed Central. Endosymbiotic and horizontal gene transfer in microbial eukaryotes A cell may come from a preexisting cell, but large chunks of its instruction manual can come from entirely different lineages. The cellular lineage is continuous; the genetic lineage is a web.
Could Life Elsewhere Start Differently?
The principle that cells come from cells is grounded in Earth biology. Researchers in astrobiology have begun questioning whether life elsewhere would necessarily use the same molecular building blocks. The concept of contingency, the idea that different starting conditions could lead to fundamentally different outcomes, suggests that non-standard molecules could have played essential roles in the origin of life, producing self-replicating systems that look nothing like terrestrial cells.27PubMed Central. Alternative Pathways in Astrobiology: Reviewing and Synthesizing Contingency and Non-Biomolecular Origins of Terrestrial and Extraterrestrial Life Whether such hypothetical life would organize into compartments we would recognize as cells, and whether those compartments would reproduce by dividing, remains entirely open. The “all cells from cells” rule may turn out to be a local fact about Earth rather than a universal law of biology, if biology beyond Earth exists at all.