Module 0: Accipitriformes — Evolution & Systematics

Around 60 million years ago (Mya), in the Palaeogene recovery following the Cretaceous–Palaeogene (K–Pg) extinction, several independent lineages of diurnal birds of prey arose. The best-supported phylogenetic framework (Jetz et al., 2012; Prum et al., 2015; Jarvis et al., 2014) places the order Accipitriformes within the clade Afroaves, alongside owls (Strigiformes), mousebirds, trogons and woodpeckers. In contrast, the order Falconiformes belongs to the sister clade Australaves, with parrots and passerines as its closest relatives.

This means that the remarkable morphological similarity between a peregrine and a sharp-shinned hawk — sharp talons, hooked beak, forward-facing eyes, cere at the base of the bill — is the product of convergent evolution, not shared ancestry. The most recent common ancestor (MRCA) of Falco peregrinus and Accipiter gentilis lived approximately 80 Mya, whereas the MRCA of Falco peregrinus and Psittacus erithacus (African grey parrot) lived only ~65 Mya.

The Accipitridae family

Within Accipitriformes, the family Accipitridae is by far the largest, with approximately 250 recognised species (Clements checklist, IOC). It contains all true eagles, hawks, kites, harriers, Old World vultures and fish-eagles. Its sister families are Cathartidae (New World vultures) and Sagittariidae (the monotypic secretary bird), together forming the crown Accipitriformes.

Representative eagle genera

The “eagles” do not form a single clade but are distributed across several lineages within Accipitridae:

• Aquila (true or “booted” eagles) — 11 species including Aquila chrysaetos (golden eagle), A. heliaca (eastern imperial), and A. audax (wedge-tailed).
• Haliaeetus (sea and fish eagles) — 8 species including H. leucocephalus (bald eagle), H. albicilla (white-tailed), and the giant H. pelagicus (Steller’s sea eagle, up to 9 kg).
• Harpia harpyja — the Neotropical harpy eagle, a canopy mammal specialist.
• Stephanoaetus coronatus — the African crowned eagle, a dietary super-specialist taking up to 20 kg monkeys (see below).
• Pithecophaga jefferyi — the critically endangered Philippine eagle.
• Spizaetus (hawk-eagles) — Neotropical crested eagles, e.g. S. ornatus and S. tyrannus.

Key Eocene–Oligocene fossils

Direct palaeontological evidence supports a mid-Palaeogene crown for Accipitriformes. Notable fossils include:

• Miosurnia diurna (Late Miocene, China) — a diurnal owl that illustrates the ecological convergence with accipitrids (Li et al., 2020).
• Proictinia gilmorei (Early Oligocene, Nebraska) — an early kite-like accipitrid.
• Aquila bullockensis (Miocene, Australia) — ancestral to the wedge-tailed eagle.
• Messelastur gratulator (Eocene, Germany) — previously placed as a stem accipitriform, now regarded as stem Cathartidae.

Molecular clock estimates (Mindell et al., 2018; Lerner & Mindell, 2005) place the crown Accipitridae at 30–35 Mya, with major generic divergences clustered in the mid-Miocene (15–25 Mya), concurrent with the global spread of grasslands and C₄ photosynthesis.

Phylogenetic distance is usually expressed as time to most-recent common ancestor (MRCA), estimated from calibrated molecular substitution rates. For two sequences \( s_i, s_j \) of length \( L \) with \( k_{ij} \) observed differences, the Jukes–Cantor corrected distance is:

Given a molecular clock rate \( \mu \) (substitutions per site per Myr), the divergence time is simply \( t_{ij} = d_{ij} / (2\mu) \). For avian nuclear UCEs, \( \mu \approx 2 \times 10^{-3}\,\text{Myr}^{-1} \) is typical.

Griffiths 1994 DNA–DNA hybridisation

Before whole-genome data, Griffiths (1994) used DNA–DNA hybridisation to measure the thermal stability \( \Delta T_{50\mathrm{H}} \) of heteroduplex DNA between species. The melting-temperature shift calibrates to divergence time via \( t \approx 4.5\,\Delta T_{50\mathrm{H}} \) Myr. This study was the first to strongly reject a sister relationship between Falconidae and Accipitridae and instead found falcons closer to passerines. Whole-genome phylogenomics (Jarvis et al., 2014) confirmed this conclusion 20 years later.

Jetz 2012 global bird phylogeny

Jetz et al. (2012) combined 6,663 bird species into a time-calibrated supertree using a backbone of molecular phylogenies and taxonomic placement for data-poor lineages. Their diversification analysis showed:

Old World vs. New World diversification

Two macro-evolutionary centres dominate modern accipitrid biogeography. The Old World produced the booted eagles (Aquila, Hieraaetus, Clanga), the Gyps vultures and the snake-eagles (Circaetus). The New World yielded the hawk-eagles (Spizaetus), the crested eagles (Morphnus), the harpy eagle, and the bird-of-prey’s most recent radiation: the neotropical Buteo. Dispersal events across the Bering land bridge and along the Central American corridor shaped present-day distributions.

In nearly all raptors, females are larger than males, often dramatically so. This reverses the pattern seen across most vertebrate orders, where males are typically the larger sex. The phenomenon is referred to as reversed sexual size dimorphism (RSD). Quantitatively, one defines:

Krüger 2005 hypotheses

Krüger (2005, Evolutionary Ecology) reviewed >20 competing hypotheses and grouped them into three broad categories:

1. Aerial-agility hypothesis: small-bodied males retain greater flight manoeuvrability and provision the female/nest during incubation; highly aerial, bird-hunting raptors should show the largest RSD.
2. Prey-size partitioning: dimorphism reduces intra-pair competition by letting the pair exploit a wider prey-size spectrum.
3. Nest-defence / female dominance: larger females successfully defend the nest and defend against infanticidal or cuckolding males; sexual selection favours female size.

Empirically, \( R \) strongly correlates with the aerial-agility index: species taking fast, evasive bird prey (Accipiter nisus, Falco columbarius) show \( R \approx 1.7\text{--}2.0 \), while scavengers (Gyps, Aegypius) and snake-eagles (Circaetus) hover near \( R \approx 1.05 \). The allometric pattern — RSD decreasing in very large raptors — is a form of Rensch’s rule operating in reverse.

African crowned eagle as dietary super-specialist

Sub-Saharan Africa’s Stephanoaetus coronatus is a striking illustration of dietary specialisation in Accipitridae. Males weigh ~3.2 kg, females ~4.7 kg, yet they routinely take prey up to 20 kg— ungulates such as suni antelope and large monkeys (colobus, vervets). The famous Taung Child fossil (Australopithecus africanus, ~2.8 Ma) bears puncture marks consistent with crowned-eagle predation (Berger & Clarke, 1995). Their combination of short, broad wings, immensely thick tarsi, and long hind talons reflects a grip-force-optimised body plan that will be revisited in Module 3.

The fossil record of Accipitriformes is patchy but informative. Notable Palaeogene localities include the Messel oil shale (47 Mya), the Green River Formation (52 Mya) and the London Clay (55 Mya). All three preserve small to medium-sized stem accipitriforms with the characteristic opposed hallux and recurved phalanges.

By the Oligocene (~30 Mya), crown Accipitridae were established. By the Mio-Pliocene (5–15 Mya), giant accipitrids had appeared in Pleistocene Eurasia and the Americas — most famously the Argentavis-like Teratornithidae, although these are technically stem Cathartiformes. The largest well-documented Accipitridae is the Holocene New Zealand Hieraaetus moorei (Haast’s eagle, 15 kg), the apex predator of an avifauna that included the moa. It went extinct ~1400 CE following Polynesian settlement.

Tempo of diversification

Using a birth–death process with rate heterogeneity, the temporal density of Accipitridae speciation is:

Integrating the expected number of lineages \( N(t) = N_0 \exp\!\int_0^t (\lambda(\tau)-\mu(\tau))\,d\tau \) over the last 30 Myr reproduces the modern species richness of ~250.