A Brief History of Birds

Summary

Living birds represent a narrow surviving subset of ancient avian diversity. Through the Cretaceous, multiple distinct bird-grade lineages coexisted alongside the ancestors of modern birds — including enantiornithines, the most diverse and abundant Mesozoic avian group, as well as hesperornithines, ichthyornithes, and others. None of those lineages have living descendants. The K-Pg mass extinction approximately 66 million years ago eliminated them entirely, along with non-avian dinosaurs. Modern birds (Neornithes) were present as a lineage before the extinction but were not the dominant group. What came after is a radiation from whatever neornithine lineages passed through that filter.

The K-Pg extinction was not a partial pruning of avian diversity. It was a severe ecological collapse that removed most lineages then present and left a narrow surviving base. The conditions that determined which neornithine lineages survived are only partially understood; proposed factors include ground-foraging capability, seed-eating, small body size, and reduced dependence on intact forest canopy, but none of these is confirmed as the primary driver. Post-K-Pg bird diversity is the product of radiation from that narrow surviving base, not of continuous unbroken diversification from the Mesozoic. The ancient lineages that were lost were not failed attempts at modern birds; they were separate evolutionary experiments with no modern successors.

The surviving Neornithes diversified rapidly during the Paleogene into the major groupings recognisable today: Palaeognathae (ratites and tinamous), Galloanserae (fowl and waterfowl), and Neoaves (the majority of living species). The timing and sequence of this diversification is contested. Molecular clock studies push many neornithine divergences into the Cretaceous, but fossil evidence for most living families begins in the Eocene or Oligocene. These two evidence types must be kept distinct. Cassowaries belong to Palaeognathae — specifically to Casuariiformes, the lineage shared with emus — but ratite evolution is not the straightforward Gondwanan vicariance story it was once treated as. The nested position of tinamous within ratite phylogenies complicates both the biogeography and the history of flightlessness in this group.

This document establishes the foundational evolutionary constraint for all bird-related reasoning in this repository: modern birds are survivors of a much larger avian radiation, not its representatives. Ancient birds must not be assumed to resemble their modern relatives. Modern families must not be projected unchanged into the Cretaceous. Cassowaries are later regional palaeognath outcomes, not models for Cretaceous avifauna or ancient bird life generally.

Metadata

Primary topic: Brief history of birds
Layer: Real-world reference
Topics: birds, evolution, extinction, K-Pg extinction, Neornithes, Palaeognathae, ratites, cassowaries
Regions: Global
Related species: Birds, avian dinosaurs, neornithines, palaeognaths, ratites, cassowaries, emus, tinamous

Core Reality

Before the K-Pg extinction

Avian-grade archosaurs (Avialae) appear in the fossil record from the Late Jurassic (~150 MYA); Archaeopteryx is among the earliest well-documented forms.
Through the Cretaceous, multiple distinct avian lineages diversified beyond the direct ancestors of living birds.
Enantiornithines were the dominant Mesozoic bird group globally: diverse, abundant, and occupying many ecological niches from small perching forms to larger species. They have no living descendants.
Hesperornithines were specialised diving birds with reduced wings. Ichthyornithes were toothed flying birds. Neither group is ancestral to living birds.
Neornithes were present in the Late Cretaceous but were not the dominant avian group. Their pre-K-Pg fossil record is sparse and uneven.
Mesozoic bird fossil recovery is biased toward specific depositional environments; total ancient avian diversity exceeds what current fossil evidence documents.
Modern bird families should not be projected backward into the Cretaceous unchanged. Ancient avian groups had body plans, ecologies, and relationships that do not map onto living families.

The extinction filter

The K-Pg mass extinction approximately 66 MYA, caused by the Chicxulub impactor, eliminated non-avian dinosaurs, enantiornithines, hesperornithines, ichthyornithes, and all non-neornithine avian lineages.
Living birds descend exclusively from neornithine lineages that crossed the K-Pg boundary. The extinction was a severe bottleneck, not a partial pruning.
Survival was lineage-specific under conditions of ecological collapse: forest canopy loss, disrupted food webs, and sustained reduction in primary productivity filtered which neornithine lineages persisted.
Proposed survival factors — ground-foraging and seed-eating, small body size, rapid reproductive turnover, reduced canopy dependence — are plausible hypotheses, not confirmed drivers. Their relative weight is not resolved.
The extinct lineages were not failed versions of modern birds. They were independent evolutionary outcomes with no modern successors.
Post-K-Pg bird diversity radiates entirely from the narrow surviving neornithine base. It does not reflect continuity with Mesozoic avian abundance.

The surviving modern-bird radiation

All living birds belong to Neornithes.
Neornithes divides into two major groups: Palaeognathae (ratites and tinamous) and Neognathae (all others).
Neognathae divides into Galloanserae (Anseriformes plus Galliformes) and Neoaves (the majority of living bird species).
Post-K-Pg Neoaves diversification was rapid. The speed of early radiation makes resolving deep internal relationships difficult; different genomic datasets produce conflicting phylogenetic topologies.
Molecular clock studies push divergence dates for several neornithine lineages into the Cretaceous; those dates reflect lineage separation, not the origin of recognisable modern families.
Fossil evidence for most living families begins in the Eocene or Oligocene. Fossil dates and molecular divergence estimates must be kept distinct: they answer different questions and depend on different assumptions.
Modern bird families are outcomes of post-K-Pg and Cenozoic diversification. None can be assumed to have existed in their present form before the K-Pg boundary.

The cassowary line

Cassowaries belong to Palaeognathae, the lineage that also includes emus, ostriches, rheas, kiwi, tinamous, and the extinct moas and elephant birds.
Cassowaries (Casuariidae) are closely related to emus (Dromaiidae); together they form Casuariiformes, a lineage with its primary distribution in Sahul and nearby regions.
Palaeognathae is not a uniformly flightless clade. Tinamous (Tinamiformes) retain flight capability and, in current molecular phylogenies, are nested within ratites rather than as their outgroup. This arrangement requires either that flightlessness evolved independently in multiple ratite lineages after their divergence, or that the common ancestor of Palaeognathae was capable of flight and ratite flightlessness is convergent. Neither interpretation is fully settled.
The older Gondwanan vicariance model — ratites riding apart on separating landmasses as Gondwana fragmented — is not a sufficient explanation. Molecular divergence dates for several ratite lineages are younger than the relevant Gondwana breakup events; dispersal must be considered alongside or instead of vicariance.
Cassowary-emu divergence timing is uncertain; estimates vary across studies and calibration choices.
Cassowaries are later regional palaeognath outcomes constrained by Sahul ecology and history. They are not models for Cretaceous birds, ancient palaeognaths generally, or ancestral neornithines.

Other living bird groups

Galloanserae (fowl and waterfowl) is the sister group to Neoaves within Neognathae. It includes ducks, geese, swans, chickens, and relatives.
Neoaves encompasses the majority of living bird species — passerines, parrots, raptors, shorebirds, pigeons, and most other familiar groups.
Neoaves diversified rapidly during the Paleogene. Deep relationships within Neoaves remain contested; competing topologies from major phylogenomic studies have not fully converged.
No Neoaves family can be treated as representative of the ancestral neornithine condition; all living groups are late-diverging outcomes of post-K-Pg radiation.

Constraints

Surviving neornithines are not representative of all ancient bird diversity. Pre-K-Pg avian lineages with no living descendants must not be treated as early versions of modern birds, as predecessors to modern families, or as evidence of what ancient bird ecology looked like.
Earliest known fossils are not true origins. Fossil absence does not establish lineage absence; the Mesozoic bird fossil record underrepresents actual diversity.
Modern bird families are not unchanged Cretaceous entities. A family's present form is not equivalent to any ancestor that shared its broader lineage divergence.
Ratites are not simple unchanged Gondwanan relics. The nested position of tinamous within current ratite phylogenies requires that ratite biogeography treat dispersal as a viable mechanism alongside or instead of vicariance; repeated independent loss of flight cannot be excluded.
Cassowaries are not models for all early palaeognaths or for early birds generally. They are specific outcomes of palaeognath evolution within Sahul under Cenozoic conditions.
Fossil evidence, molecular clock estimates, and ecological inference are distinct evidence types. They must not be conflated or treated as interchangeable. Molecular divergence dates are not fossil observations; ecological inferences from modern birds are not direct evidence of ancient ecology.
Uncertainty must be preserved where lineages, divergence dates, or phylogenetic relationships are contested. Avian phylogeny at multiple levels continues to be actively revised.

System Implications

Extinction filtering constrains all deep bird ancestry reconstruction. Any claim about ancient bird ecology or diversity must account for the pre-K-Pg lineages that are now absent and cannot be inferred from modern birds.
Cassowary ancestry must pass through palaeognath uncertainty. Claims about cassowary deep history cannot bypass unresolved questions about ratite biogeography, repeated flightlessness, and Sahul lineage timing.
Lineage survival across the K-Pg boundary differs from modern-family presence. A lineage's post-K-Pg existence does not confirm that it maintained the same ecological role, body plan, or range as any hypothesised Cretaceous ancestor.
Bird evolution is branching diversification, not linear progress. Later diversity represents new outcomes radiating from a survival bottleneck, not improvements on or continuations of earlier avian experiments.

Known Variability

Fossil preservation is biased toward specific depositional environments; Mesozoic avian diversity in tropical or island environments is likely substantially underrepresented.
Molecular clock calibration choices affect divergence date estimates substantially; the same sequence data can yield estimates that differ by tens of millions of years across studies depending on fossil calibration and model choices.
Avian taxonomy continues to be revised as phylogenomic studies update family and order boundaries; classifications in older literature may not match current usage.
The timing of neornithine lineage divergences relative to the K-Pg boundary — how many lineages were already distinct before the extinction versus how many diverged after — remains debated and is sensitive to which molecular clock approach is used.
Models for ratite biogeography continue to be revised as molecular data and dispersal evidence accumulate; vicariance-only and dispersal-only models have both been proposed and neither is now accepted without qualification.

Open Questions

Which crown-bird lineages are confirmed as present before the K-Pg boundary on the basis of unambiguous fossil evidence?
Which proposed K-Pg survival traits are strongly evidenced versus plausible but unresolved?
How much of the apparent post-K-Pg radiation of Neoaves reflects true rapid diversification versus gaps in fossil preservation?
Was ratite flightlessness lost once in a common ancestor of all ratites, or independently in multiple lineages after dispersal? The nested position of tinamous makes this a live and unresolved question.
When did the cassowary-emu lineage diverge, and what constraints does that divergence impose on claims about cassowary deep history in Sahul?
Which pre-K-Pg avian lineages — particularly enantiornithines — occupied ecological roles now held by modern bird groups, and what does that mean for ecological inferences drawn from modern-bird behaviour?

Cassowary World