Bayesian dating combines DNA evidence about evolutionary change with fossil evidence about geological time. A molecular-clock model links substitutions in DNA to elapsed time; fossil calibrations anchor parts of the evolutionary tree to dates. The result is a probability distribution of possible divergence times, not a single date known with certainty.
What DNA and fossils each contribute
DNA reveals patterns of change
Comparing DNA sequences across living species helps infer their relationships and how much molecular change occurred along different branches. In a simplified clock, more substitutions correspond to more elapsed time. But that relationship depends on the model: substitution rates can vary among DNA sites and among lineages, so sequence differences alone do not directly tell researchers how many years have passed.
As an Amazon Associate I earn from qualifying purchases.
Fossils connect the tree to geological time
Fossils provide age information for organisms or the groups to which they belong. Researchers use that evidence to constrain when parts of the evolutionary tree could have existed. The fossil’s age and its placement on the tree both matter; uncertainty in either affects the resulting timeline.
Recommended Free Tools
Without a fossil or another external time calibration, molecular rate and elapsed time are confounded: the same observed amount of DNA change could reflect a short period at a fast rate or a longer period at a slow rate. DNA can inform relative timing, but it cannot by itself establish absolute dates.
#1 Best Overall
How the Bayesian calculation works
A clock model links change to time
A molecular-clock model describes how DNA substitutions accumulate over time. A strict clock assumes the rate is constant across lineages. A relaxed clock allows rates to differ among branches, which can better represent some datasets but introduces additional assumptions about how those rates vary.
Likelihood and priors are combined
The sequence-evolution model gives a likelihood: how well a proposed tree, set of divergence times, and rates explain the observed DNA. Priors represent information or assumptions specified before considering those sequence patterns, including fossil calibrations, clock rates, and plausible tree histories. In broad terms, Bayesian inference combines the likelihood with these priors to produce a posterior distribution over possible trees, rates, and dates.
Rank #2
A posterior date is therefore conditional on the evidence and the chosen models. It expresses the range and relative support of dates under those assumptions; it is not a guarantee that the assumptions or fossil interpretation are correct.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsThree ways to include fossils
| Approach | Where fossil evidence enters | How fossil placement is treated | What the model represents |
|---|---|---|---|
| Node dating | As age-calibration distributions on selected internal nodes | The fossil is assigned to a calibration node | Fossil-informed age constraints combined with clock and tree priors |
| Fossilized birth-death dating | Through a model of fossils and living taxa sampled from a shared macroevolutionary process | Fossil sampling is modeled as part of the process | Fossil and living-taxon sampling; expanded versions can also estimate diversification and sampling patterns |
| Total-evidence or fossil tip-dating | As dated fossil tips, using fossil ages and morphological characters alongside molecular sequences from living taxa | Fossil placement can be inferred from character data rather than fixed in advance | Phylogenetic placement and divergence times together |
Node dating
For node dating, researchers place a probability distribution on the age of a selected internal node, using a fossil and relevant geological knowledge to define what ages are plausible. A calibration may have soft bounds: those limits express low or high probability rather than declaring every age outside them impossible.
Calibrations do not operate independently in the final tree. Their effects interact with ancestor–descendant constraints and the tree prior, which describes plausible branching histories. As a result, the effective joint time prior can differ from what a researcher might infer by inspecting each calibration distribution alone.
Fossilized birth-death dating
The fossilized birth-death (FBD) approach models fossil and living taxa as samples from the same macroevolutionary process. This offers a way to incorporate fossil information without relying only on separate, node-specific calibration distributions. Expanded FBD models can also estimate aspects of diversification and sampling. The approach still depends on assumptions about the fossil record and how fossils are sampled; modeling that process can be difficult.
Total-evidence or fossil tip-dating
Total-evidence dating includes fossils as taxa in the phylogeny. Their ages and morphological character data are analyzed alongside molecular sequences from living species. Because the fossil’s position can be inferred from its characters, the method can carry placement uncertainty into the dating analysis rather than treating every fossil assignment as settled beforehand.
Why the choice of model changes the timeline
The approaches differ in where fossil evidence enters, whether fossil placement is fixed or inferred, and how fossil sampling is represented. Clock assumptions matter too: a strict clock and a relaxed clock can distribute elapsed time differently when lineages have different molecular rates. Tree priors also influence how dates are distributed across the branches.
Best Value
These methods address related but distinct modeling needs; no approach is automatically best for every fossil assemblage or dataset. The useful choice depends on the quality of fossil identification and dating, available morphological and molecular data, taxon sampling, and whether the assumptions needed by the model are defensible.
How to interpret and check estimated dates
Read dates as distributions
Report the posterior uncertainty rather than treating a point estimate as the timeline. A narrow interval can still be misleading if the fossil calibration, clock, or tree assumptions are weak. Adding more DNA sequence data can sharpen the sequence-based component of an estimate, but it does not remove uncertainty in fossil ages or calibrations.
Inspect the prior’s effect
Researchers should examine the effective joint time prior, not just the individual calibration distributions. They should also check how posterior ages change under reasonable alternative calibration and tree-prior choices. Large changes indicate that the dating result depends strongly on those assumptions.
Free tools Windows power users keep installed
One-click scans. No signup required.
Check whether the analysis is adequate for the case
A specific study requires case-specific fossil identification and geological dating, appropriate taxon sampling, sequence and, where relevant, morphology data, model selection, prior checks, and convergence diagnostics. The general framework does not prescribe a software package or a universal setup for every organism.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
