For most of biology's history, inheritance has had a single picture: a parent passes its DNA to its offspring, the offspring grows up, has its own offspring, and so on down the line. Genes go vertically — from parent to child. Mutations introduce small variations along the way, and natural selection sorts what works.
That picture is correct for animals like us. It is catastrophically incomplete for the bacteria that make up the bulk of life on Earth.
Bacteria — and to a lesser extent archaea, viruses, fungi, and even plants — engage in something fundamentally different. They share genes horizontally: across species, across genera, across vast evolutionary distances. A bacterium of one species can pick up genes from a bacterium of an entirely different species and start using them within hours. The phenomenon is called horizontal gene transfer (HGT), and it is one of the most underappreciated facts about how life actually works.
The Three Mechanisms
Horizontal gene transfer happens through three primary routes, all well-characterized.
Transformation. A bacterium absorbs naked DNA from its surroundings — DNA that has leaked out of a dead bacterium, for example — and incorporates it into its own genome. Frederick Griffith first observed this in 1928 with pneumococcus, in the experiment that pointed Avery, MacLeod, and McCarty toward proving DNA was the genetic material in 1944. Some bacteria are naturally competent to do this; others can be made competent under stress.
Conjugation. Two bacteria physically connect through a structure called a pilus, and one transfers a copy of a circular DNA fragment called a plasmid to the other. Joshua Lederberg and Edward Tatum demonstrated this in 1946. Plasmids carry many of bacteria's most strategically important genes — including, notoriously, antibiotic resistance.
Transduction. A virus that infects bacteria — a bacteriophage — accidentally packages a piece of bacterial DNA inside its own protein coat instead of viral DNA. When that virus infects the next bacterium, it injects the bacterial gene as well. Transduction was discovered by Norton Zinder and Joshua Lederberg in 1952.
Each of these mechanisms can move DNA across species lines that vertical inheritance would never cross. The result is a genetic web, not a tree — at least at the bacterial level.
Why "Like Email" Is Not a Bad Analogy
If you imagine bacterial life as a community where any bacterium can attach a "useful gene" to a metaphorical email and send it to anyone — friend, stranger, a different species entirely — you are not far off how things work. Plasmids are essentially portable gene packages that can be copied and transferred independently of the organism's main chromosome.
A plasmid carrying a gene for resistance to penicillin can move from a strain of E. coli in someone's gut to a strain of Staphylococcus aureus on their skin within hours. The receiving bacterium does not need to evolve the resistance through random mutation. It simply receives it, fully formed, and begins expressing it.
This is why antibiotic resistance has spread so explosively in modern medicine. We are not waiting for the right mutation to evolve in each species independently. The bacteria are exchanging blueprints. A resistance gene that emerges anywhere can, in principle, end up everywhere.
What This Does to the Tree of Life
The discovery of pervasive HGT has forced biologists to redraw the foundational picture of evolution. The traditional "tree of life," with a single trunk branching into all later species, was based on the assumption that genes flow only vertically. Once you accept that bacteria and archaea trade genes promiscuously, you cannot draw a clean tree at the bottom of the diagram. You get something closer to a net or a web.
For the first two billion or so years of life — when only single-celled organisms existed — most evolution was happening this way. The genes flowing through ancient microbial ecosystems were not lineage-bound. They moved.
This has practical consequences for how we read ancient evolution. Some of the most important transitions in life's history may have been horizontal acquisitions rather than vertical descent. The mitochondria in your cells almost certainly began as a free-living bacterium absorbed by an ancestor of all complex life — a story preserved in the field of endosymbiosis, but with strong horizontal-transfer overtones.
How Common Is It?
In bacteria, HGT is staggeringly common. By some estimates, up to 80% of the genes in some bacterial genomes have been acquired horizontally at some point in their evolutionary history. Different strains of the same bacterial species often share less than 60% of their genes, with the rest being a mix of horizontally acquired sequences. The "core genome" of a species is small; the pan-genome — the union of all genes ever found in any member of the species — can be many times larger.
In animals, HGT is much rarer but not absent. Researchers have identified probable horizontally transferred genes in insects, nematodes, and even some vertebrates — usually picked up from bacteria, fungi, or viruses, sometimes hundreds of millions of years ago. The famous case of bdelloid rotifers — small aquatic animals — shows that perhaps 10% of their genes were acquired horizontally from non-animal sources. The phenomenon is real even in the metazoan lineage, just unevenly distributed.
Why It Matters Beyond the Lab
Horizontal gene transfer is not a curiosity. It shapes problems that matter directly to human life.
Antibiotic resistance is the most consequential. The reason we are running out of effective antibiotics is in large part the speed of HGT among pathogenic bacteria. The same plasmids carry resistance to multiple drugs at once, so selecting for resistance to one antibiotic often selects for resistance to several. The arithmetic of HGT, more than the slow ticking of mutation, is what drives the modern antibiotic crisis.
Bacterial evolution in the human microbiome is happening continuously. The bacteria living in your gut are exchanging genes with each other and with bacteria you eat, breathe, or pick up from surfaces. The microbiome is not a fixed community — it is a churning marketplace of genetic information.
Bioengineering is built on HGT mechanisms. Plasmids are the workhorses of biotechnology — we use bacterial conjugation systems to move genes into engineered organisms, into crops, into therapeutic cells.
Defining "species" in bacteria becomes philosophically tricky. If two bacteria of "the same species" share only 60% of their genes while regularly trading the rest with bacteria of other species, what does the species concept even pick out? Microbiologists have largely accepted that bacterial species are looser, fuzzier categories than animal species — bounded more by ecology and behavior than by genetic uniformity.
What to Take From This
Most of life on Earth, by mass and by number, is bacterial. And most of bacterial evolution does not look like the tidy diagrams we learn in school. It looks like a vast, ancient, ongoing exchange of genetic information across whatever boundaries the textbooks try to draw.
Vertical inheritance gives us continuity. Horizontal transfer gives bacteria the ability to acquire wholesale, ready-made solutions to problems — sometimes problems we created.
It is a humbling picture, and a more accurate one than the simple branching tree we inherited from the 19th century. The deeper biology goes, the more it reveals that life is connected — by genes that travel, by lineages that cross, by a network older and stranger than the trunk-and-branches picture would suggest.
