"How Does Evolution Actually Work? From Darwin's Observations to Modern Synthesis"

# Evolution We Can Actually Watch in Real Time

One of the persistent public misunderstandings about evolution is that it's too slow to observe. The London Underground mosquito from the last chapter starts to answer this. But there are more direct, more controlled, and more dramatic examples of evolution happening on timescales from days to decades.

## The longest-running evolution experiment: E. coli at 35+ years

In 1988, Richard Lenski at Michigan State University started a now-famous experiment: 12 populations of *Escherichia coli* in glucose-limited minimal media, maintained continuously, every day, transferred to fresh media. As of 2026, this is over 80,000 generations of continuous evolution. That's roughly equivalent to 1.5 million years of human evolution in terms of generation count.

The results are remarkable and specific. All 12 populations evolved improved fitness in the glucose-limited environment. Several convergently evolved the same solutions — similar mutations in the same genes across independently evolving populations, suggesting certain paths are more likely than others.

The most dramatic event occurred around generation 31,500 in one population: it evolved the ability to aerobically consume citrate, a carbon source present in the medium that *E. coli* cannot normally use aerobically. This had never been observed before in the species. The key mutations happened in a specific sequence — a structural gene change wasn't enough; a prior "potentiating" mutation had to occur first, which only then made the citrate-metabolizing mutation viable. This is historical contingency in action: evolution's outcome depends partly on what happened before.

## Antibiotic resistance: evolution with immediate consequences

Antibiotic resistance is evolution made visible through medical cost. The mechanism is straightforward:

A population of bacteria includes natural variation in susceptibility. Antibiotics kill susceptible individuals; resistant individuals survive and reproduce. If the resistant trait is heritable (and bacterial resistance is highly heritable — plasmids carrying resistance genes can even transfer between different bacterial species), resistance spreads in the population.

The timescale is alarming. Penicillin was introduced clinically in 1943. Penicillin-resistant Staphylococcus aureus was reported by 1945 — two years. Methicillin (developed specifically to counter penicillin resistance) was introduced in 1959; MRSA appeared in 1961.

This is evolution operating at the speed of bacterial generations: 20 minutes per generation in fast-growing conditions. The hospital environment is a selection pressure; resistant strains are the selected phenotype.

## The LTEE citrate experiment in a visual

In 2008, a team at Harvard created a spectacular visual demonstration of evolution in real time. They placed bacteria in a large petris dish with areas of increasing antibiotic concentration, from zero at the edges to 1,000x the lethal dose in the center.

Over 11 days, they filmed bacterial colonies growing outward. Each time a colony hit an antibiotic zone, almost everything died. But the small number of resistant mutants survived, colonized the new zone, and provided the source population for the next, higher-concentration zone.

By day 11, bacteria had evolved resistance to 1,000 times the lethal dose. The film of this experiment — showing colonies progressing across a giant petri dish through zones of increasing concentration — is one of the more viscerally convincing pieces of evolution education produced in recent decades.

## Darwin's finches: real-time measurement

Peter and Rosemary Grant spent 40+ years (from 1973 onward) on Daphne Major, one of the Galápagos Islands, banding and measuring every Darwin's finch on the island. Their data captured natural selection happening in measurable, quantitative terms.

During a severe drought in 1977, large seeds were the only food remaining. Larger-beaked finches could crack them; smaller-beaked birds starved. The average beak depth in the surviving *Geospiza fortis* population increased by 0.5mm in a single generation. When normal rainfall returned and small seeds were available again, beak size drifted back toward smaller.

This is the clearest direct documentation of natural selection changing a measurable trait in a wild population in response to an identifiable environmental pressure. It's not a theory; it's a dataset going back 50 years.

## Why these examples matter

The "we can't observe evolution" argument is factually wrong. We have:
- Directly watched bacterial populations evolve new metabolic capabilities in controlled laboratory conditions
- Tracked beak morphology changes in wild populations correlated with specific selection events
- Watched antibiotic resistance emerge in clinical settings in real time
- Created and characterized speciation events in laboratory populations of Drosophila

The scale of evolutionary change that requires deep time is the accumulation of enough speciation events and morphological innovation to produce the diversity visible in the fossil record. The *mechanism* is observable in systems with short generations.

> We can watch it happen. But evolution is also one of the most persistently misunderstood scientific theories in public discourse. The next chapter addresses the most common misconceptions — and why they're wrong.

"Evolution We Can Actually Watch in Real Time"

// COMMENTS

ON THIS PAGE