Compare calibration-sensitive clocks, dated whole plants, regional phytolith assemblages and animal-diet isotopes without turning them into one false origin or dominance timeline.
Each step names its evidence claims, with references below. Interactive Explorer states are available in the full application; the narrative remains an editorial synthesis.
A 2026 analysis integrates 25,685 fossil occurrences with molecular data, but its crown-angiosperm interval changes radically with calibration treatment; the narrow Late Jurassic result is the authors' preferred model, not a fossil flower.
The atlas displays the preferred hard-bound result while exposing the much older alternatives beside it.
claim:event:crown-angiosperm-calibration-sensitivityA separate metacalibrated time tree places several major flowering-plant crown lineages near 135–130 Ma. Those estimates neither validate the atlas's legacy ranges nor record fossil first appearances.
The navigation tree is an educational subset: monocots and eudicots do not exhaust Angiospermae, and Magnoliopsida is not an exact eudicot catalogue root.
claim:event:angiosperm-expansionMore than one thousand Montsechia specimens combine leafy axes, closed single-seeded fruits and depositional evidence for a submerged freshwater plant. They do not make the flowering-plant ancestor aquatic.
A fossil habit, a phylogenetic placement and an ancestral ecological state are different claims.
claim:event:barremian-montsechiaLeefructus links diagnostic leaves and reproductive organs in a 125.8–122.6 Ma Yixian bed, supporting an early eudicot record while leaving Ranunculaceae affinity provisional.
One locality cannot define the global origin, first appearance or diversity of eudicots.
claim:event:yixian-leefructusCratolirion preserves roots, leaves, anatomy and reproductive organs together at about 115 Ma. Its crown-monocot placement does not turn it into a modern lily or the first monocot.
Whole-plant preservation strengthens character association, not claims beyond the analysed specimen and placement.
claim:event:crato-cratolirionFourteen Great Plains palaeosol assemblages record an 8–5.5 Ma increase in potentially C4 grass phytoliths. The 50–60% value applies within grasses, not to all vegetation.
Regional assemblages and diagnostic limits prevent a synchronous global C4-grassland claim.
claim:event:great-plains-c4-phytolith-transitionTooth-enamel carbon isotopes record widespread, regionally asynchronous C4 biomass expansion around 8–6 Ma. They do not identify the exact C4 plant lineages or directly measure open-grassland cover.
Dietary isotope proxy, plant identity, vegetation cover and faunal adaptation must remain separate.
claim:event:c4-grassland-expansion