# DNA Half-Life Argument

<!-- type: argument | created: 2026-11-17 | updated: 2026-11-17 -->

## Intro

DNA falls apart. After an animal dies, the long chain of its genetic code starts breaking into shorter and shorter pieces, the same way a necklace loses beads once the string starts to rot. In 2012 a team of scientists measured how fast that happens. They tested 158 bones of the moa, a giant flightless bird from New Zealand, whose ages were already known from carbon dating. They found that DNA in bone has a half-life of about 521 years: every 521 years, half of the remaining DNA pieces of a set length break.

That number matters for dinosaurs. Several research teams have reported structures inside dinosaur bone cells that light up with stains and antibodies that bind DNA. If those signals are the dinosaur's own DNA, they sit in bone said to be 75 to 80 million years old, which is far beyond any survival window the decay rate allows. The young-earth reading is simple: DNA traces are still there because the bones are thousands of years old. The deep-time reading has to explain the traces some other way, as contamination, as something other than DNA, or as preservation far beyond anything measured.

![Engraving of a giant moa (Dinornis giganteus) skeleton standing beside a Maori man for scale](/codex/assets/moa-dinornis-giganteus-skeleton-buller.jpg)

_Skeleton of the giant moa, Dinornis giganteus, from Walter Buller's 1905 catalogue of New Zealand birds. The 2012 DNA half-life study measured decay in 158 moa leg bones. Image: public domain, via [Wikimedia Commons](https://commons.wikimedia.org/wiki/File:Skeleton_of_Dinornis_giganteus.jpg), resized._

## In full

Allentoft et al., "The half-life of DNA in bone: measuring decay kinetics in 158 dated fossils" (*Proceedings of the Royal Society B* 279, 2012, 4724-4733), quantified mitochondrial DNA copy number in 158 radiocarbon-dated moa tibiotarsi from three sites of near-identical preservation conditions. The team estimated a per-nucleotide fragmentation rate of k = 5.50 × 10⁻⁶ per site per year at an effective burial temperature of 13.1°C, which gives a half-life of 521 years for a 242-base-pair mtDNA target. Modelled at a permafrost-like -5°C, the same kinetics give a half-life of about 158,000 years for a 30-base-pair fragment and an average fragment length of one base pair after roughly 6.8 million years. Age explained 38.6 percent of the sample-to-sample variation in preservation, so burial environment matters a great deal.

Against that baseline stand reports of DNA-reactive signals in Mesozoic tissue. Schweitzer, Zheng, Cleland and Bern (*Bone* 52, 2013, 414-423) found that osteocyte-like microstructures from *Tyrannosaurus rex* and *Brachylophosaurus canadensis* bound anti-DNA antibodies and reacted to the intercalating stains propidium iodide and DAPI, localized inside the cells. Bailleul et al. (*National Science Review* 7, 2020, 815-822) reported chondrocyte-like cells with nucleus-like and chromosome-like structures in calcified cartilage of a *Hypacrosaurus* nestling from the Two Medicine Formation, again staining with PI and DAPI inside the cells and not in the surrounding matrix. Neither team recovered a dinosaur DNA sequence, and both said so. The abductive argument holds that these signals, read with the measured kinetics, fit a young chronology without auxiliary assumptions, while the deep-time chronology must either reclassify the signals or posit a preservation mechanism several orders of magnitude beyond any demonstrated. This page is debate prep; each premise carries its positive case, steel-manned objections, numbered rebuttals, a live-cite kit, and tactical notes.

## Cheatsheet

**The 30-second reply:**

> Mainstream scientists measured it on 158 dated moa bones: DNA in bone has a half-life of about 521 years at the burial temperature they studied. Even frozen at -5°C, their own model has the average strand down to a single letter after about 6.8 million years. Now two research teams, one led by Mary Schweitzer, have found structures inside dinosaur bone cells that react to DNA stains and DNA antibodies, in bones dated at 75 to 80 million years. Either those signals are something else, or the decay chemistry was wrong by a factor of ten or more, or the bones are not that old. The young-earth reading needs none of those rescues.

**Fast facts:**

- **521 years** is the half-life of a **242-base-pair** mitochondrial DNA fragment at an effective **13.1°C**. It is a figure for one fragment length at one temperature, and it should always be quoted that way.
- **6.8 million years** is the paper's modelled time for the **average** fragment to shrink to **one base pair at -5°C**, the most favorable case they modelled. Hell Creek and Two Medicine burial temperatures were far warmer.
- At the moa sites' own temperature, the published rate constant puts that one-base-pair point at about 180,000 years (1/k), a figure derived directly from the paper's k.
- The oldest **sequenced** DNA on record is environmental DNA from Greenland's Kap København Formation, dated about **2 million years** (Kjær et al., *Nature* 612, 2022), and mammoth DNA from Siberian permafrost dated over **1 million years** (van der Valk et al., *Nature* 591, 2021). Both are from frozen ground.
- **Dinosaur DNA signals** reported: *T. rex* and *Brachylophosaurus* osteocytes (Schweitzer et al., *Bone* 52, 2013); *Hypacrosaurus* cartilage (Bailleul et al., *National Science Review* 7, 2020). Stain and antibody evidence, no sequence.
- The 1994 "Cretaceous DNA" sequence (Woodward et al., *Science* 266) turned out to be human contamination; Mary Schweitzer co-authored the paper that showed it (Hedges and Schweitzer, *Science* 268, 1995).

**Counter-moves:**

1. **Quote the number exactly.** "521 years for a 242-base-pair fragment at 13.1°C." Precision closes the "you misread the paper" exit.
2. **Separate DNA from collagen.** The half-life is a DNA figure. Collagen and other proteins have their own chemistry and their own case ([Soft Tissue in Dinosaur Fossils Argument](/codex/soft-tissue-in-dinosaur-fossils-argument/)).
3. **Ask for the mechanism and its size.** Any rescue has to bridge from about 2 million years (the frozen-ground record) to 75 million years in warm sediment.
4. **Point to who found the signals.** The stain and antibody results come from mainstream labs publishing in mainstream journals.

**Concessions worth granting, each with its return:**

- *No dinosaur DNA sequence has been published.* Granted, and the teams say so. **Return:** the deep-time framework also has no demonstrated mechanism for DNA-reactive material lasting 75 million years, so both sides are reading the same stain evidence, and only one side needs an undemonstrated mechanism to read it.
- *DAPI and PI can bind things other than DNA, and microbes live in buried bone (Saitta et al., 2019).* Granted. **Return:** in Bailleul's cartilage the signal sits inside the cells and not in the surrounding matrix or sediment, which is the pattern microbial contamination should not produce, and Schweitzer's team added anti-DNA antibodies to the stains.
- *Decay rates vary with temperature.* Granted, that is what the paper measured. **Return:** the variation runs the wrong way for the deep-time case; Mesozoic burial temperatures were warmer than the moa sites, which shortens survival.

**Closing line:** *"The half-life was measured by the people who hold the deep-time view, in their journals, on bones they dated. All I am asking is that we apply their own number to the cells their own colleagues stained."*

## Argument structure

| # | Premise |
|---|---|
| **P1** | DNA in bone decays by measurable first-order kinetics; the best-calibrated study gives a 521-year half-life for a 242 bp fragment at 13.1°C, and even at -5°C the modelled fragments are reduced to single bases within about 6.8 million years. |
| **P2** | Mainstream research teams have reported DNA-reactive signals (anti-DNA antibodies, PI and DAPI staining, chromosome-like structures) localized inside dinosaur cells from bone and cartilage dated 75 to 80 million years. |
| **P3** | No demonstrated preservation mechanism extends DNA survival from the measured ceiling of about 1 to 2 million years in frozen ground to tens of millions of years in temperate sediment. |
| **C** | **The DNA signals in dinosaur tissue are best explained by a chronology in which the tissue is thousands of years old; the deep-time chronology can accommodate them only by reclassifying the signals or positing an undemonstrated preservation mechanism.** |

## Form

**Abductive, with a conditional core.** If the signals are endogenous DNA, the measured kinetics and the conventional dates cannot both stand. The argument weighs three readings of the same data: (a) the signals are contamination or non-DNA artifacts, (b) the signals are endogenous and an unknown mechanism preserved them for 75 million years, (c) the signals are endogenous and the tissue is young. Reading (a) carries the burden of the intracellular localization; reading (b) carries the burden of a mechanism; reading (c) carries the burden of the radiometric chronology, which is handled on its own pages ([Carbon-14 in Deep-Time Specimens Argument](/codex/carbon-14-in-deep-time-specimens-argument/), [Young Earth Creationism](/codex/young-earth-creationism/)).

---

## P1, DNA decays at a measurable rate that leaves no strand standing over tens of millions of years

### Affirmative case (second-order arguments)

1. **The moa study is the cleanest calibration available.** Allentoft and colleagues chose 158 bones of one taxon (moa), one element (the tibiotarsus), from three sites with nearly identical burial conditions, all radiocarbon-dated. Holding the environment constant isolates time as the variable. The result was a per-site fragmentation rate of 5.50 × 10⁻⁶ per year and a 521-year half-life for a 242 bp mtDNA target (*Proceedings of the Royal Society B* 279, 2012).
2. **The kinetics are first order, so they extrapolate.** Fragmentation by depurination and strand breakage proceeds as a random-cutting process; average fragment length falls in inverse proportion to elapsed time. That lets the team model other fragment lengths and other temperatures, which is how they reached the -5°C figures: a 158,000-year half-life for a 30 bp fragment, and one-base-pair average length at about 6.8 million years.
3. **Warmer burial shortens every number.** The paper's temperature dependence means a burial warmer than 13.1°C decays faster. Late Cretaceous North America was warm, and the dinosaur formations at issue (Hell Creek, Two Medicine, Judith River) were not permafrost.
4. **The independent ancient-DNA record agrees with the ceiling.** The oldest sequenced DNA comes from frozen ground: about 2 million years in Greenland sediment (Kjær et al., *Nature* 612, 2022) and over 1 million years in Siberian mammoth teeth (van der Valk et al., *Nature* 591, 2021). Nothing older has been sequenced, and nothing from warm sediment comes close.

### Anticipated objections

1. **"The 521-year figure is a 242 bp fragment at 13.1°C. You cannot use it as 'the half-life of DNA'."**
2. **"Age explained only 38.6 percent of the variation. The model is too noisy to extrapolate across millions of years."**
3. **"The Greenland and mammoth results prove DNA lasts far longer than people once thought, so the ceiling keeps moving."**

### Rebuttals

1. **Correct, and the argument uses it that way.** The load-bearing number is the rate constant and the modelled -5°C ceiling, both from the same paper. Quoting "521 years for a 242 bp fragment at 13.1°C" is the precise form, and every version of the figure, precise or loose, puts 75 million years out of reach. Failure mode: **correcting the phrasing without changing the conclusion**.
2. **Noise widens the error bars by factors, and the gap is a factor of tens.** Unexplained variance means some bones keep DNA better than the model predicts, perhaps several times better. The distance between the most favorable modelled case (6.8 million years at -5°C) and the dinosaur dates (75 to 80 million years, at warmer temperatures) is more than an order of magnitude. Failure mode: **treating scatter around a trend as permission to ignore the trend**.
3. **The ceiling moved from hundreds of thousands to about 2 million years, in permafrost.** Both records come from the coldest, most stable burial settings on earth, and both still sit some forty times short of the dinosaur dates. A moving ceiling that moves only in frozen ground does not reach temperate Cretaceous sediment. Failure mode: **extrapolating a cold-storage record to warm burial**.

### Live-cite kit

- **Scripture:** [Romans 8:20-22](/codex/romans-8-20-22/) (creation subjected to the bondage of decay); [Psalm 102:25-27](/codex/psalms-102-25-27/) (the works of creation wear out like a garment)
- **Scholarly:** Morten E. Allentoft et al., *Proceedings of the Royal Society B* 279 (2012); Kurt H. Kjær et al., *Nature* 612 (2022); Tom van der Valk et al., *Nature* 591 (2021)
- **Aphorism:** "The clock was built by deep-time scientists, calibrated on dated bones, and published in their journal. It reads the same no matter who looks at it."

### Tactical notes

- **Lead with the precise figure** and the 158 dated bones. Precision is what makes the number unanswerable.
- **Never call 521 years "the" half-life of all DNA.** That is the one error an informed opponent is waiting for.
- **What not to defend:** that the moa study itself dated any dinosaur. It did not. It measured a rate; the argument applies the rate.

---

## P2, DNA-reactive signals have been reported inside dinosaur cells

### Affirmative case (second-order arguments)

1. **Schweitzer and colleagues, 2013.** In *Bone* 52 (2013), 414-423, the team freed osteocyte-like microstructures from *T. rex* (MOR 1125) and *Brachylophosaurus canadensis* (MOR 2598) bone. Anti-DNA antibodies bound in localized regions inside the cells, and the same structures reacted positively with propidium iodide and DAPI, two stains that intercalate with DNA. Antibodies to actin, tubulin, PHEX and histone H4 also reacted, and mass spectrometry supported several of these proteins. Histone H4 is a protein that packages DNA in the nucleus.
2. **Bailleul and colleagues, 2020.** In *National Science Review* 7 (2020), 815-822, the team reported calcified cartilage from a nestling *Hypacrosaurus stebingeri* containing cells with nucleus-like bodies and what appear to be chromosomes, including two cells joined as if caught at the end of division. Isolated cells stained with PI and DAPI inside the cell and not in the surrounding matrix, and the matrix reacted with antibodies to collagen II, the cartilage-specific collagen.
3. **Localization is the key datum.** Contaminating microbes and their DNA spread through porous bone and sediment. A stain signal confined to cell interiors, at the place a nucleus would be, is the distribution endogenous DNA would produce.
4. **These are mainstream results.** Schweitzer is a North Carolina State University paleontologist; Bailleul worked at the Chinese Academy of Sciences' Institute of Vertebrate Paleontology and Paleoanthropology with Schweitzer and Jack Horner as co-authors. The data were published in peer-reviewed journals with no young-earth framing.

### Anticipated objections

1. **"Stains are not sequences. No one has sequenced dinosaur DNA, and every claim to have done so has collapsed."** The 1994 Woodward sequence was human contamination.
2. **"Saitta et al. (*eLife* 8, 2019) dug a *Centrosaurus* bone aseptically and found the organic material was recent and microbial. Buried bone is a microbial habitat."**
3. **"DAPI and PI bind other molecules, and fossilization chemistry can mimic cell structures."**

### Rebuttals

1. **Granted that no sequence exists, and the young-earth case does not need one.** The argument runs on the stain and antibody evidence as published. The 1994 failure is a point in the field's favor: Schweitzer herself co-authored the paper that exposed it (Hedges and Schweitzer, *Science* 268, 1995), so the team reporting the 2013 signals is the team most alert to contamination. Failure mode: **using an old failure to dismiss a newer, more cautious result from the people who caught the old one**.
2. **Saitta's bone and Bailleul's cartilage are different specimens with different results.** Saitta's finding shows that some fossil bone hosts recent microbes; it does not show that all DNA-reactive signals everywhere are microbial. Bailleul's signal sat inside cells and not in the matrix where a microbial community would be. Failure mode: **generalizing one specimen's contamination to every specimen**.
3. **The alternative-binding objection has to be specified.** Schweitzer's team used three independent DNA probes (antibody, PI, DAPI) that converge on the same intracellular location, alongside a histone signal. "Something else" must explain why three probes agree and why the histone is there. Failure mode: **an unspecified alternative offered as if it were an explanation**.

### Live-cite kit

- **Scripture:** [Job 40:15-24](/codex/job-40-15-24/) (Behemoth, read by many young-earth writers as a large dinosaur-like creature); [Genesis 1:24-25](/codex/genesis-1-24-25/)
- **Scholarly:** Mary H. Schweitzer et al., *Bone* 52 (2013); Alida M. Bailleul et al., *National Science Review* 7 (2020); S. Blair Hedges and Mary H. Schweitzer, *Science* 268 (1995)
- **Aphorism:** "Three different DNA probes lit up in the same spot inside a dinosaur cell. Three probes agreeing on one spot inside a cell is a nucleus asking to be explained."

### Tactical notes

- **Cite the stain papers by journal and author.** The opponent expects a creationist source and gets Schweitzer.
- **Concede the sequence point quickly**, then move to localization and the three-probe agreement.
- **What not to defend:** "dinosaur DNA has been sequenced." It has not, and the case is stronger without the overstatement.

---

## P3, No demonstrated mechanism carries DNA from about 2 million years to 75 million

### Affirmative case (second-order arguments)

1. **Iron cross-linking is a protein mechanism.** Schweitzer et al. (*Proceedings of the Royal Society B* 281, 2014) showed that iron from hemoglobin slowed degradation of ostrich blood vessels over two years. The mechanism concerns cross-linked proteins and vessel walls. It has not been shown to protect the phosphodiester backbone of DNA, which breaks by hydrolysis and depurination.
2. **The best natural refrigerator reaches about 2 million years.** Permafrost and polar sediment are the most favorable settings measured, and they set the ceiling. Any mechanism for dinosaur DNA must outperform permafrost by a factor of about forty in warmer ground.
3. **Mineral binding has been shown for proteins at the few-million-year scale**, as in proteins bound to ostrich eggshell calcite from Laetoli dated about 3.8 million years (Demarchi et al., *eLife* 5, 2016). That is still a protein result, and still twenty times short of the dinosaur dates.

### Anticipated objections

1. **"Absence of a known mechanism is an argument from ignorance. Molecular paleontology is young; mechanisms will be found."**
2. **"The radiometric dates for Hell Creek and Two Medicine are tied to volcanic ash beds dated by several isotope systems. A single anomaly cannot override them."**

### Rebuttals

1. **The argument runs from what is known.** Measured kinetics are positive knowledge about how DNA behaves. The young chronology fits that knowledge directly; the deep-time chronology fits it only with a mechanism that does not yet exist. Promissory notes are allowed in science, but they are not evidence, and an inference to the best explanation weighs the explanations on the table. Failure mode: **treating a hoped-for mechanism as present evidence**.
2. **The chronology question is real and is argued on its own ground.** Young-earth geologists contest the assumptions of radiometric dating (initial conditions, closed systems, constant decay) and point to measurable carbon-14 in material assigned to deep time ([Carbon-14 in Deep-Time Specimens Argument](/codex/carbon-14-in-deep-time-specimens-argument/)). The DNA argument adds an independent molecular line that converges with that case. Which line yields when two lines conflict is the question, and it cannot be settled by declaring one of them fixed. Failure mode: **holding one dataset immune and making every other dataset bend to it**.

### Live-cite kit

- **Scripture:** [Jeremiah 33:25](/codex/jeremiah-33-25/) (the fixed patterns of heaven and earth, which is why decay rates are trustworthy at all)
- **Scholarly:** Mary H. Schweitzer et al., *Proceedings of the Royal Society B* 281 (2014); Beatrice Demarchi et al., *eLife* 5 (2016)
- **Aphorism:** "A rescue that has never been observed is a hope, and hopes do not outweigh measurements."

### Tactical notes

- **Force-commit move:** "Name the mechanism that preserves DNA, not protein, for 75 million years in warm ground, and the study that measured it."
- **Do not over-argue the radiometric side here.** Hand it to the dating pages; keep this exchange on the molecule.

---

## Tensions

- **How the figures get quoted.** A common popular form says "DNA has a 521-year half-life, so nothing survives 6.8 million years." Both numbers are real, but they belong to different fragment lengths and temperatures (242 bp at 13.1°C; average length one base pair at -5°C). Precise quotation strengthens the case; the codex's [Soft Tissue in Dinosaur Fossils Argument](/codex/soft-tissue-in-dinosaur-fossils-argument/) has been brought into line with the paper's wording.
- **DNA versus protein.** The DNA half-life cannot be transferred directly to collagen. The protein case rests on its own chemistry and its own literature, and the two arguments are strongest kept side by side as two converging lines.
- **Sequence evidence.** The dinosaur DNA signals are stain, antibody and morphology evidence. Bailleul has said publicly that sequencing is not yet in reach. A future sequence would strengthen this argument decisively; a demonstrated non-DNA source for the stains would remove P2. Both are live research questions.
- **Evolutionary creationists and old-earth creationists** accept the conventional dates and expect the signals to be resolved as contamination or diagenesis. The codex presents the young-earth reading at full strength as one of the live in-house Christian positions ([Genesis Interpretation Spread](/codex/genesis-interpretation-spread/)).

## Conclusion

**The DNA signals in dinosaur tissue are best explained by a chronology in which the tissue is thousands of years old.** The decay rate was measured on dated bones by mainstream researchers; the DNA-reactive signals were reported by mainstream researchers; the gap between the oldest DNA ever sequenced and the dinosaur dates is more than an order of magnitude, in the wrong temperature direction. The deep-time reading remains open to anyone who can show the signals are not DNA or can demonstrate a mechanism of the needed size. Until then the young reading explains the data without a rescue.

## Master objections to the argument as a whole

1. **"You are building on unsequenced signals. Extraordinary claims need sequence evidence."** Reply: the same standard applies in both directions. The deep-time reading also needs evidence that the signals are not DNA, and for the *Hypacrosaurus* cartilage that evidence has not been produced. The argument states what the published data show and asks which chronology fits them without additions.
2. **"Christian scientists do not need this argument; God could preserve DNA however He likes."** Reply: the argument does not invoke a miracle. It invokes the regularity of decay that God upholds ([Jeremiah 33:25](/codex/jeremiah-33-25/)) and asks what that regularity implies.
3. **"Half-lives are probabilistic; a few molecules could survive by chance."** Reply: with first-order fragmentation, the expected number of surviving intact strands after many hundreds of half-lives is effectively zero. At the moa rate, 75 million years is roughly 144,000 half-lives for a 242 bp fragment.

## Tactical opening / closing

**Opening line:** "In 2012 a team of geneticists dated 158 moa bones and measured how fast DNA breaks down in them. I want to take their number seriously, and then look at what two mainstream labs found when they stained dinosaur bone cells for DNA."

**Closing landing strip:** "The decay clock is theirs. The stained cells are theirs. The gap between the oldest DNA anyone has read and the dinosaur dates is more than tenfold, in colder ground than any dinosaur lay in. The reading that needs no rescue is the one where these bones are young."

## Connection to Scripture

- [Romans 8:20-22](/codex/romans-8-20-22/), creation subjected to futility and the bondage of decay; the steady, measurable breakdown of matter is part of the world Paul describes.
- [Jeremiah 33:25](/codex/jeremiah-33-25/), the fixed order of heaven and earth, which grounds the reliability of measured rates.
- [Genesis 1:24-25](/codex/genesis-1-24-25/), the land creatures made after their kinds.
- [Job 40:15-24](/codex/job-40-15-24/), Behemoth, read on the young-earth view as a large creature of the dinosaur type known to Job's audience.

## Patristic / scholarly note

**Classical:** Molecular paleontology has no patristic antecedent. The literal-day reading of Genesis that the young chronology preserves runs through Basil the Great's *Hexaemeron* (c. 378).

**Modern:**
- Morten E. Allentoft, Michael Bunce and colleagues, the decay-kinetics study (2012).
- Mary H. Schweitzer, molecular paleontologist, the osteocyte signals (2013) and the iron hypothesis (2014); also co-author of the 1995 paper that exposed the 1994 contamination.
- Alida M. Bailleul, the *Hypacrosaurus* cartilage study (2020).
- Evan T. Saitta and colleagues, the aseptic *Centrosaurus* excavation (2019), the strongest contamination critique.
- Brian Thomas (Institute for Creation Research), young-earth tracking of the molecular-paleontology literature.

## Scholarship

Published academic works this page draws on, each matched to a publisher, index or library catalog record.

- Morten E. Allentoft et al., "The half-life of DNA in bone: measuring decay kinetics in 158 dated fossils," *Proceedings of the Royal Society B: Biological Sciences* 279 (2012). [doi:10.1098/rspb.2012.1745](https://doi.org/10.1098/rspb.2012.1745) [PubMed 23055061](https://pubmed.ncbi.nlm.nih.gov/23055061/)
- Alida M Bailleul et al., "Evidence of proteins, chromosomes and chemical markers of DNA in exceptionally preserved dinosaur cartilage," *National Science Review* 7 (2020). [doi:10.1093/nsr/nwz206](https://doi.org/10.1093/nsr/nwz206) [PubMed 34692099](https://pubmed.ncbi.nlm.nih.gov/34692099/)
- Beatrice Demarchi et al., "Protein sequences bound to mineral surfaces persist into deep time," *eLife* 5 (2016). [doi:10.7554/elife.17092](https://doi.org/10.7554/elife.17092) [PubMed 27668515](https://pubmed.ncbi.nlm.nih.gov/27668515/)
- S. Blair Hedges and Mary H. Schweitzer, "Detecting Dinosaur DNA," *Science* 268 (1995). [doi:10.1126/science.7761839](https://doi.org/10.1126/science.7761839) [PubMed 7761839](https://pubmed.ncbi.nlm.nih.gov/7761839/)
- Evan T Saitta et al., "Cretaceous dinosaur bone contains recent organic material and provides an environment conducive to microbial communities," *eLife* 8 (2019). [doi:10.7554/elife.46205](https://doi.org/10.7554/elife.46205) [PubMed 31210129](https://pubmed.ncbi.nlm.nih.gov/31210129/) [open access](https://elifesciences.org/articles/46205.bib)
- Mary H. Schweitzer et al., "A role for iron and oxygen chemistry in preserving soft tissues, cells and molecules from deep time," *Proceedings of the Royal Society B: Biological Sciences* 281 (2014). [doi:10.1098/rspb.2013.2741](https://doi.org/10.1098/rspb.2013.2741) [PubMed 24285202](https://pubmed.ncbi.nlm.nih.gov/24285202/) [open access](https://www.ncbi.nlm.nih.gov/pmc/articles/3866414)
- Tom van der Valk et al., "Million-year-old DNA sheds light on the genomic history of mammoths," *Nature* 591 (2021). [doi:10.1038/s41586-021-03224-9](https://doi.org/10.1038/s41586-021-03224-9) [PubMed 33597750](https://pubmed.ncbi.nlm.nih.gov/33597750/) [open access](https://researchprofiles.ku.dk/da/publications/f8d69292-4ad1-4a1f-b6b0-6abf168b6c52)
- Scott R. Woodward, Nathan J. Weyand and Mark Bunnell, "DNA Sequence from Cretaceous Period Bone Fragments," *Science* 266 (1994). [doi:10.1126/science.7973705](https://doi.org/10.1126/science.7973705) [PubMed 7973705](https://pubmed.ncbi.nlm.nih.gov/7973705/)

<!-- SCHOLARSHIP:END -->

## See also

- [Soft Tissue in Dinosaur Fossils Argument](/codex/soft-tissue-in-dinosaur-fossils-argument/), the companion case from collagen, blood vessels and proteins
- [Carbon-14 in Deep-Time Specimens Argument](/codex/carbon-14-in-deep-time-specimens-argument/), the chronology case from measurable radiocarbon
- [DNA](/codex/dna/), the molecule itself and the information argument
- [Young Earth Creationism](/codex/young-earth-creationism/), the position this argument supports
- [Old Earth Creationism](/codex/old-earth-creationism/), the in-house Christian alternative on chronology
- [Genesis Interpretation Spread](/codex/genesis-interpretation-spread/), the live Christian readings of Genesis
- [Dinosaurs in Ancient Art](/codex/dinosaurs-in-ancient-art/), alleged dinosaur depictions in ancient art
- [Origins](/codex/origins-arguments/), category master

<!-- COMMON-QUESTIONS:START -->

<div data-pagefind-weight="5">

## Common questions this page answers

**Q: What is the half-life of DNA?**

The best-calibrated measurement, from 158 radiocarbon-dated moa bones (Allentoft et al., *Proceedings of the Royal Society B*, 2012), gives a half-life of about 521 years for a 242-base-pair mitochondrial DNA fragment at an effective burial temperature of 13.1°C. Shorter fragments and colder ground last longer; warmer ground shortens survival.

**Q: Does the DNA half-life prove dinosaurs are young?**

It shows that DNA cannot survive tens of millions of years under any condition yet measured, since even at -5°C the modelled fragments are down to single bases in about 6.8 million years. If the DNA-reactive signals reported inside dinosaur cells are the dinosaurs' own DNA, the young-earth chronology explains them directly, while the deep-time chronology needs a preservation mechanism no one has demonstrated.

**Q: Has dinosaur DNA been found?**

No dinosaur DNA sequence has been published. Mary Schweitzer's team (*Bone*, 2013) and Alida Bailleul's team (*National Science Review*, 2020) reported structures inside dinosaur cells that react with DNA stains (PI and DAPI) and, in Schweitzer's work, with anti-DNA antibodies and a histone protein. A 1994 claimed sequence turned out to be human contamination.

**Q: What is the oldest DNA ever found?**

Environmental DNA from the Kap København Formation in northern Greenland, dated about 2 million years (Kjær et al., *Nature*, 2022), and mammoth DNA from Siberian permafrost dated over 1 million years (van der Valk et al., *Nature*, 2021). Both come from frozen ground, the best possible preservation setting.

**Q: Is DNA the same as the soft tissue found in dinosaur bones?**

No. The soft-tissue finds are mostly proteins such as collagen, plus vessel-like and cell-like structures, and proteins decay by different chemistry from DNA. The DNA half-life applies to DNA; the protein evidence is a separate, converging line covered in [Soft Tissue in Dinosaur Fossils Argument](/codex/soft-tissue-in-dinosaur-fossils-argument/).

**Q: Could the DNA signals in dinosaur bone be contamination?**

Some fossil bone does host recent microbes, as Saitta and colleagues showed in a *Centrosaurus* bone in 2019. In the *Hypacrosaurus* cartilage, though, the stain signal sat inside the cells and not in the surrounding matrix, and Schweitzer's team found three independent DNA probes agreeing on the same intracellular spot. Contamination remains the main alternative reading, and it has to explain that localization.

</div>

<!-- COMMON-QUESTIONS:END -->
