They EvolvedExplore

Correction · 2026 record

A list of things that have been observed to evolve.

The original plate is a creationist graphic: a textbook animal cell captioned as structures that have “never been observed to evovle.” Every label on it has a phylogenetic address. Several are still changing in living organisms. One new organelle was described in 2024.

Open the cellCases happening now

20

Structures on the plate

5

Living cases below

26+

Cited papers

The claim
Original meme: a labeled animal cell under the heading A list of things that have never been observed to evovle.
The correction
Corrected plate: A list of things that have been observed to evolve, with a eukaryotic cell.

The error

“Never observed” is a category mistake.

Nobody filmed a mountain range rise in real time either. We still know how mountains form: we have strata, faults, GPS, and intermediates. Organelles are the same kind of problem.

Observation in evolutionary biology is phylogeny, living intermediates, genomes mid-transfer, lab endosymbioses, and organelles being installed in organisms that are alive right now. Demanding a time-lapse of the Paleoproterozoic is not skepticism. It is a refusal to look at the notebook.

The misspelling is the tell. The biology underneath is not a mystery. It is a research program with dates, organisms, and papers.

Interactive plate

Every label, sourced.

Pins follow the original drawing. Open a structure for origin, evidence, living traces, and papers. On a phone, use the name chips under the plate.

Animal cell · numbered pins

0/20 opened

Cutaway illustration of an animal cell. Numbered pins mark each labeled structure from the original meme.

Happening in living systems

If the demand is “show me one,” here are five.

A new organelle, this decade

The nitroplast

Braarudosphaera bigelowii · Candidatus Atelocyanobacterium thalassa (UCYN-A)

2024

In April 2024, Science reported that a nitrogen-fixing cyanobacterium inside a marine alga has crossed the line from endosymbiont to organelle. Host proteins are imported into UCYN-A. Its division is synchronized with the host. It cannot be cultured free-living. That is the operational definition of an organelle — and it is happening in the modern ocean, not in a thought experiment.

If someone says ‘we have never seen a bacterium become an organelle,’ the reply is: we have. This year. It fixes nitrogen.

  • Coale TH et al. (2024). Nitrogen-fixing organelle in a marine alga.Science 384:217–222

Primary endosymbiosis, take two

The chromatophore

Paulinella chromatophora · Synechococcus-like cyanobacterium

~90–140 mya · still transferring genes

Plant chloroplasts come from one ancient cyanobacterial capture. Paulinella did it again, independently, in the Cretaceous–Paleogene window. The chromatophore genome is slashed. Photosynthetic genes have moved to the host nucleus and the proteins are shipped back in through an evolving import apparatus. It is a primary plastid caught mid-installation — a second origin of a photosynthetic organelle, still unfinished.

One origin of chloroplasts could be called a fluke. Two independent primary captures, plus secondary and tertiary plastids across the tree, is a mechanism.

  • Singer A et al. (2017). Massive protein import into the early-evolutionary-stage photosynthetic organelle of the amoeba Paulinella chromatophora.Current Biology 27:2763–2773
  • Nowack ECM, Weber APM (and related Paulinella work) (2022). Genomics-informed insights into chromatophore integration.see Nowack lab reviews, 2016–2022

Watched in a lab

Jeon’s amoebae

Amoeba proteus / discoides · Legionella-like X-bacteria

1966 → ~200 generations

In 1966 a bacterial infection nearly wiped Kwang Jeon’s amoeba cultures. Survivors became dependent on the bacteria. Nucleus-transfer experiments showed the host genome had changed: uninfected nuclei in infected cytoplasm died. A lethal infection became an obligate endosymbiosis in a scientist’s lifetime. That is not a just-so story. It is a lab notebook.

The original meme demands a time machine. Jeon used a pipette.

  • Jeon KW, Jeon MS (1972). Development of cellular dependence on infective organisms.Science 176:1122–1123

An organelle changing jobs

Hydrogenosomes and mitosomes

Trichomonas, Giardia, Entamoeba, microsporidia, and others

Multiple independent origins

Mitochondria did not freeze after they were captured. In anaerobic lineages they repeatedly evolved into hydrogenosomes (H₂-producing, usually no genome) or mitosomes (tiny, genome-less, Fe–S cluster factories). Phylogeny says they are mitochondria. Biochemistry says they have new jobs. That is organelle evolution observed in living clades, over and over, convergently.

The plate shows one cartoon mitochondrion. The tree shows a family of derived organelles with a documented conversion path.

  • Lewis WH et al. (2020). Convergent evolution of hydrogenosomes from mitochondria.Mol Biol Evol 37:524–539
  • Roger AJ et al. (2017). The origin and diversification of mitochondria.Current Biology

A nucleus being erased

Nucleomorphs

Cryptophytes and chlorarachniophytes

Secondary endosymbiosis, still running

When a eukaryote engulfs another eukaryote, you can end up with four membranes and a leftover nucleus — the nucleomorph. Its genome is a miniature, intron-packed remnant, still expressing a handful of genes. We are looking at a nucleus on the way out. Secondary and tertiary plastid captures (euglenids, dinoflagellates, stramenopiles, apicomplexans) are additional, independent organelle origins.

The meme’s plate is one cell. Endosymbiosis has happened enough times that we have leftover nuclei in the wreckage.

  • Archibald JM (2015). Endosymbiosis and eukaryotic cell evolution.Current Biology 25:R911–R921

Deep time

From LUCA to a 2024 organelle.

  1. ~4 bya

    LUCA

    A membrane, ribosomes, a genetic code, and metabolism. Plasma membrane, cytosol, and the ribosome on the plate are already in place.

  2. ~2.2–1.6 bya

    Mitochondrial endosymbiosis

    An Asgard-related archaeon and an alphaproteobacterium enter a hydrogen-based partnership that becomes permanent. Eukaryogenesis begins.

  3. ~1.8–1.2 bya

    LECA

    The last eukaryotic common ancestor already has a nucleus, ER, Golgi, peroxisomes, endo-lysosomes, actin and microtubule cytoskeletons, a cilium/basal body, and a mitochondrion.

  4. ~1.5 bya

    Primary plastid

    A cyanobacterium becomes the chloroplast of Archaeplastida (glaucophytes, red algae, green plants). One capture, three living lineages.

  5. Many times since

    Secondary and tertiary plastids

    Euglenids, chlorarachniophytes, cryptophytes, dinoflagellates, haptophytes, stramenopiles, apicomplexans. Organelles, repeatedly.

  6. ~90–140 mya

    Paulinella chromatophore

    A second primary photosynthetic organelle. Gene transfer still in progress.

  7. 1966–present

    Jeon amoebae

    Lab-observed transition from lethal infection to obligate endosymbiosis.

  8. 2024

    Nitroplast

    A nitrogen-fixing organelle described in a living marine alga. The list keeps growing.

Method

What “observed” actually means.

Phylogenomics

We do not guess ancestry. We align thousands of proteins across bacteria, archaea, and eukaryotes. Mitochondria sit inside the alphaproteobacteria. The host sits inside Asgard archaea. That is a measurement.

LECA reconstruction

If a machine is present in all major eukaryotic lineages, it was in their last common ancestor. Richards et al. (2024) recover a nucleus, ER, Golgi, peroxisomes, cytoskeleton, and mitochondrion in LECA. The plate is not a wish list. It is a descendant.

Living intermediates

Nucleomorphs, chromatophores, hydrogenosomes, mitosomes, and the nitroplast are organelles caught mid-installation or mid-conversion. Evolution does not owe anyone a complete cartoon in one species.

Experiment

Jeon’s amoebae turned a lethal infection into obligate endosymbiosis in ~200 generations. Peroxisomes regenerate from ER in yeast. Orthogonal ribosomes are evolved in labs. The ‘never in the lab’ clause is false even on its own terms.

Backup

Core papers, 1967–2026.

This is a working set, not a bibliography of everything. Landmark statements of endosymbiosis, LECA, Asgard eukaryogenesis, and the 2024 nitroplast. Open an organelle on the plate for the local citations.

  1. 1.Sagan L (Lynn Margulis). 1967. On the origin of mitosing cells. J Theor Biol 14:225–274.
  2. 2.Jeon KW, Jeon MS. 1972. Development of cellular dependence on infective organisms. Science 176:1122–1123.
  3. 3.Spang A et al. 2015. Complex archaea that bridge prokaryotes and eukaryotes. Nature 521:173–179.
  4. 4.Archibald JM. 2015. Endosymbiosis and eukaryotic cell evolution. Current Biology 25:R911–R921.
  5. 5.Roger AJ, Muñoz-Gómez SA, Kamikawa R. 2017. The origin and diversification of mitochondria. Current Biology 27:R1177–R1192.
  6. 6.Singer A et al. 2017. Massive protein import into the chromatophore of Paulinella chromatophora. Current Biology 27:2763–2773.
  7. 7.Imachi H et al. 2020. Isolation of an archaeon at the prokaryote–eukaryote interface. Nature 577:519–525.
  8. 8.Richards TA et al. 2024. Reconstructing the last common ancestor of all eukaryotes. PLoS Biology 22:e3002917.
  9. 9.Coale TH et al. 2024. Nitrogen-fixing organelle in a marine alga. Science 384:217–222.
  10. 10.Zhang J et al. 2025. Deep origin of eukaryotes outside Heimdallarchaeia within Asgardarchaeota. Nature 642:990–998.
  11. 11.Tobiasson V et al. 2026. Dominant contribution of Asgard archaea to eukaryogenesis. Nature 650:141–149.