Rosa Rubicondior: Refuting Creationism - A Fruit-Eating Dinosaur

Life reconstruction of the Hell Chicken Infernogallus munching on early fruit, while a T. rex lurks in the background.

Rare fossil discovery reveals new dinosaur species that were specialized fruit-eaters | Royal Ontario Museum

A dinosaur nicknamed the “hell chicken” might sound like an especially unpleasant predator, but its remarkable skull suggests rather different eating habits. Infernogallus jamesi, a turkey-sized dinosaur from the approximately 66-million-year-old Hell Creek Formation of North America, appears to have specialised in processing soft plant foods, probably including fleshy fruit. Behind the entertaining name lies a useful illustration of evolution: a lineage descended from carnivorous ancestors acquiring a feeding apparatus suited to a very different diet.

The discovery is described in a paper in Nature Communications by Gregory Funston and colleague. Infernogallus belonged to the caenagnathids, a group of toothless, feathered oviraptorosaurs whose feeding habits have been difficult to establish. Its unusually complete skull has now provided evidence that isolated bones could not supply.

Using CT scans and anatomical reconstruction, the researchers investigated how its jaws would have worked. The arrangement of the palate and jaw muscles indicates that the lower jaw could move backwards and forwards, while opposing ridges and grooves helped process food inside the mouth. Rather than an apparatus for tearing flesh, this was suited to mashing softer material. Particularly revealing is the resemblance to feeding adaptations in living fruit-eating parrots: an example of convergent evolution, in which similar functional demands produce comparable adaptations in separate evolutionary lineages.

There is an important distinction here between reconstructing a feeding mechanism and finding the remains of an animal’s last meal. The evidence supports specialised herbivory and makes fleshy fruit a strong candidate, but it does not establish an exclusively fruit-based diet. Nor does it demonstrate that these dinosaurs dispersed seeds or drove the evolution of fruit. The authors identify a potentially significant relationship between dinosaur feeding specialisations and the diversification of flowering plants, while acknowledging that the ecological details remain uncertain.

For creationism, the difficulty extends beyond the familiar problem of a dinosaur living tens of millions of years before the supposed biblical Creation Week. This animal belongs within an evolutionary history of changing anatomy, changing diets and changing ecosystems. Its feeding apparatus was built from the inherited structures of a theropod skull, modified for a new way of obtaining food. Calling it a separately created “kind” supplies neither an explanation for those anatomical relationships nor a testable account of how its specialisations arose.

What makes this discovery especially interesting is the glimpse it offers of an ecological role that has been easy to overlook. The world of the last non-avian dinosaurs contained more than enormous predators and bulk-feeding herbivores. It also supported smaller animals with specialised ways of exploiting plant foods — a diversity that science can reconstruct, cautiously and progressively, from the evidence preserved in their bones.

How can we tell what an extinct dinosaur ate? We cannot watch an extinct dinosaur feeding, but its fossils can preserve several kinds of evidence about its diet. Palaeontologists combine these clues to distinguish what an animal could process, what it probably ate, and, in exceptional cases, what it actually swallowed.

Skulls and jaws: reconstructing the feeding apparatus

The shape of a skull, the surfaces of its teeth or beak, and the position of its jaw joints constrain how an animal could bite and process food. Muscle attachment sites help researchers reconstruct the arrangement and action of muscles that have long since disappeared. CT scans can reveal otherwise inaccessible structures and help scientists reconstruct fragile or distorted fossil bones.

These reconstructions allow researchers to investigate questions such as whether a jaw was suited to forceful biting, slicing, crushing or repeated movements that mashed food. However, a feeding apparatus rarely identifies a single food with certainty: several foods can have similar mechanical properties.

Living animals provide comparisons

Animals whose diets and feeding movements can be observed provide a way to interpret fossil anatomy. If an extinct animal possessed a feeding mechanism resembling that of a living fruit-eater, fruit becomes a plausible dietary candidate. The comparison becomes stronger when several relevant features agree, rather than just the overall appearance of the beak.

This does not mean that the extinct animal necessarily ate exactly the same foods. Similar structures can serve more than one purpose, and living species may vary their diets with season, age or availability.

More direct evidence: meals and droppings

Gut contents: Food remains preserved within an animal’s digestive tract can provide unusually direct evidence of a meal. Such finds are rare, and researchers must distinguish genuine gut contents from material introduced during burial.

Coprolites: Fossilised faeces may contain fragments of bone, plant tissues, seeds or other dietary remains. These can reveal what was eaten, although identifying the animal that produced a particular coprolite can be difficult.

Tooth wear: Where teeth are present, patterns of wear can help identify how food was processed and the kinds of material encountered. They do not necessarily identify particular food species.

Even a preserved meal is only a snapshot. One animal’s last meal does not establish the complete diet of its species.

What does this mean for the “hell chicken”?

In Infernogallus jamesi, the reconstructed jaw musculature and opposing ridges and grooves indicate a mechanism for processing soft food through backwards-and-forwards movements of the lower jaw. Comparisons with fruit-eating parrots support the interpretation that fleshy fruit formed an important part of its diet.

The conclusion is therefore an inference from functional anatomy and comparison with living animals, rather than the identification of fruit in preserved gut contents. It supports probable fruit-eating, but does not establish an exclusively fruit-based diet.

Eating fruit is not the same as dispersing seeds

A fruit-eater might swallow seeds, discard them, or destroy them while feeding. Effective seed dispersal requires seeds to survive and reach places where they can germinate. Evidence that an animal could process fruit therefore does not, by itself, demonstrate that it dispersed seeds or influenced the evolution of fruit-bearing plants.

Recognising these distinctions is part of the strength of palaeontology: researchers can propose a well-supported explanation while identifying which parts remain open to further evidence.

The published paper is accompanied by a Royal Ontario Museum news release:

Rare fossil discovery reveals new dinosaur species that were specialized fruit-eaters
Among the ravenous carnivores and lumbering herbivores that roamed the Cretaceous landscape, another kind of dinosaur was evolving that started to eat something else: the plants that began to flower and produce fruit.
The discovery by paleontologists from Stony Brook University, Royal Ontario Museum (ROM), and the North Dakota Geological Survey reveals a new species, Infernogallus jamesi – a type of small oviraptorosaur – had chewing muscles for mashing soft food, making it the first known dinosaur that specifically ate fruit.

Led by Dr. Greg Funston, Assistant Professor in the Department of Anatomical Sciences at Stony Brook University, the research – which analyzed the chewing muscles of the new species, showing that it had a unique arrangement only seen in fruit-eating parrots today – was published in the journal Nature Communications.

We knew dinosaurs evolved to fill so many different roles in ecosystems in the Cretaceous, but we never expected to discover fruit-eating dinosaurs. For more than 100 years, we didn’t know what the skulls of these animals looked like. Now, we know they were even weirder than we could have predicted.

Assistant Professor Dr. Greg Funston, lead author
Department of Anatomical Sciences, Stony Brook University Stony Brook, NY, USA.

Infernogallus, meaning “hell chicken,” was so named because the fossils – including a nearly complete, articulated skull and partial skeleton, currently on display at the Royal Ontario Museum – were found in the Hell Creek Formation in the U.S. (stretching across Montana, North Dakota, South Dakota, and Wyoming), where it would have lived alongside T. rex and Triceratops.

Belonging to a group of toothless, feathered dinosaurs called caenagnathids, Infernogallus was a small dinosaur the size of a turkey, and would have weighed about 12 kg (25 lbs) – making it one of the smallest members of its group even as a fully grown adult.

Curiously, Infernogallus and its relatives evolved from the meat-eating side of the dinosaur tree. Detailed analysis of its chewing mechanics shows that its skull was specialized for mashing stone fruits to remove the flesh from the main seed. This came as a surprise, because paleontologists had previously struggled to show any clear links between dinosaur diversity and the evolution of flowering plants. However, this discovery underscores parallels between the evolutionary radiations of caenagnathids and fruit, with caenagnathids evolving their specialized chewing capabilities as fruits became bigger and more common.

It’s pretty rare for animals to become specialized for a diet of fruit, so when it happens, it comes with some remarkable adaptations. The anatomy of Infernogallus is bizarre among dinosaurs, and we quickly realized that we were dealing with something special.

Dr. Thomas Dudgeon, co-author
Department of Natural History
Royal Ontario Museum
Toronto, ON, Canada.
Infernogallus’s ability to chew fruit can be seen in the adaptations to its palatal bones – the parts of the skull that make up the roof of the mouth. Using CT scans to reconstruct the skull, the researchers found that these bones arched downwards into the mouth itself, which made certain muscles face front-to-back instead of up and down. This would have enabled a chewing motion where the beak slid back and forth, like in a human, instead of the rigid vertical bite of T. rex and other dinosaurs.

These forward-facing muscles are rare in the animal kingdom – though are present in mammals like humans who demonstrate complex chewing, and in some birds that specialize in fruit-eating, like the Vulturine Parrot (Psittrichas). The uniqueness of this muscle organization led the researchers to conclude that these adaptations likely evolved for the same reason: processing soft foods – such as fruit – in the mouth instead of the gut.

The revelation that Infernogallus was a fruit-eating dinosaur is additionally significant, as the discovery suggests that dinosaurs may have had an unrecognized impact on the evolution of flowering plants as seed-dispersers – an important role filled by birds, reptiles, and small mammals today. During the Cretaceous period, flowering plants took over from cone-bearing plants as the major vegetation on land – as seen in the fossil record, which includes fossilized fruits.

Complete remains of small dinosaurs like Infernogallus are very rare in the fossil record. The strange skull of Infernogallus is an exciting find because it reveals surprising insights into dinosaur evolution and feeding specialization. This discovery shows that we still have a lot to learn about the world of T. rex, and about dinosaurs more generally.

Professor Dr. David Evans, senior author
Department of Natural History
Royal Ontario Museum
Toronto, ON, Canada.

Recognizing that this species may have played a new ecological role during this period also indicates that we might not yet appreciate the full diversity of dinosaur species and behaviours at the end of the Cretaceous era – adding to the scientific debate about whether dinosaurs were in decline before their extinction.

Publication:
Funston, G.F., Dudgeon, T.W., Simon, D.J. et al.
Fruit-eating adaptations in a Late Cretaceous oviraptorosaur (Theropoda).
Nat Commun (2026). https://doi.org/10.1038/s41467-026-78343-w

Download PDF
Abstract
The diversification of herbivorous dinosaurs and the rise of angiosperms in the Late Cretaceous are each well established, but the link between them is debated. Broad-scale diversity patterns in the two groups appear not to closely align, and there is no direct evidence of dinosaur adaptation to angiosperm-based diets. Here we describe Infernogallus jamesi gen. et. sp. nov., a diminutive caenagnathid oviraptorosaur, and show that these theropods were specialised herbivores that originated alongside the rise of fleshy angiosperm fruits. The nearly complete, articulated skull of Infernogallus jamesi—the first for the clade—reveals drastic changes to the adductor musculature organisation, unique in non-avian theropods but converged upon by extant frugivorous parrots. These modifications enabled extensive oral processing of soft food using anteroposterior movement of the mandible and occlusion of ridges and grooves on opposed elements. Altogether, this attests to a highly specialised herbivorous diet in caenagnathids, which likely consisted in large part of fleshy angiosperm fruits, which increased in size and diversity in concert with the radiation of oviraptorosaurs. Caenagnathids thus provide one of the strongest candidates for angiosperm-based diet adaptations in dinosaurs, but their role in the Late Cretaceous proliferation of fruit—which shaped early Cenozoic flora—remains unclear.

Fig. 1 Infernogallus jamesi (ROM VP 65050). Photograph (a) and line drawing (b) of ROM VP 65050. c Skeletal reconstruction of Infernogallus jamesi based on ROM VP 65050 and other eucaenagnathines (shaded areas are preserved in ROM 65050). d Palaeohistological thin section of the femur of ROM VP 65050, showing lines of arrested growth (arrows) and an external fundamental system indicative of advanced age and skeletal maturity. cv cervical vertebrae, dv dorsal vertebrae, efs external fundamental system, fe femur, ga gastralia, hu humerus, il ilium, ma manus, pb pubis, pp plane-polarised light, sc scapulacoracoid, sk skull, tb tibia, xp cross-polarised light. Scale bars as indicated.

Fig. 2 Cranial anatomy of Infernogallus jamesi gen. et. sp. nov. Photograph (a) and line drawing (b) of the skull of ROM VP 65050 in left lateral view. c µCT reconstruction of the skull and mandible, with each bone segmented separately (colours) and retrodeformed (see “Methods”). d µCT reconstruction with the mandible removed to show the palatal bones, which protrude ventrally below the cheek margin and the mandibular articulation point of the quadrate. e idealised reconstruction of the skull and mandible as they would have appeared in life, showing distinctive reduction of infratemporal fenestra (blue arrow), and rearrangement of pterygoid musculature (orange line) compared to archetypal oviraptorid Citipati osmolskae (f). Dotted blue lines in (e, f) outline internal features obscured by overlying mandibles. bo basioccipital, ect ectopterygoid, fr frontal, j jugal, lac lacrimal, lts laterosphenoid, mvp maxillovomeral process, mPTv pterygoideus ventralismuscle, mx maxilla, p parietal, pal palatine, pbs parabasiphenoid, pmx premaxilla, po postorbital, pt pterygoid, q quadrate, qj quadratojugal, sq squamosal, v vomer. Scale bar as indicated.

Fig. 3 Cranial muscle organisation in caenagnathids. a µCT model of the skull and mandible of Infernogallus, with three-dimensional adductor muscle reconstructions when the mandible is positioned in the middle of its anteroposterior articular range (see Supplementary Fig. 8 for other configurations). Points are coloured by muscle, colours do not represent additional data. b decomposed muscle vector orientations for Infernogallus, showing strongly anteroposterior (rostrocaudal) actions of the pterygoideus muscles. c principal components analysis of total mandibular adductor muscle organisation, summarising muscular vectors for each taxon. Note the close clustering of non-avian theropods, except for Infernogallus, which, compared to other oviraptorosaurs (e.g., Citipati osmolskae), emphasises anteroposterior motions at the expense of dorsoventral motions. A similar pattern of change (dotted grey arrows) is evident between Psittacus and Psittrichas, likely related to different oral processing behaviours. Triangles are nonoviraptorosaur theropods, squares are oviraptorosaurs, and circles are avians; points with an asterisk possess a mandibular articulation that enables anteroposterior motion; convex hulls group oviraptorosaurs (orange) and avians (blue). Points are coloured by species, colours do not represent additional data. d–f posterior view of mandibular articular regions of Infernogallus, Dinosaur Park eucaenagnathines, and Psittrichas, showing ridge and saddle arrangement enabling anteroposterior sliding of the mandible; images not to scale. mAMEM m. adductor mandibulae externus medialis, mAMEP m. adductor mandibulae externus profundus, mAMES m. adductor mandibulae externus superficialis, mAMP m. adductor mandibulae posterior, mPSTp m. pseudotemporalis profundus, mPSTs m. pseudotemporalis superficialis, mPTd m. pterygoideus dorsalis, mPTv m. pterygoideus ventralis, PCA principal component axis, R ridge, S saddle.

Fig. 4 Evolution of Caenagnathidae alongside the rise of fruit. a Bayesian phylogeny of Oviraptorosauria, scaled a posteriori using a morphological-clock informed fossilised birth-death (FBD) model. Mean node ages were derived from 100 trees output by the model to scale the branches. Line diagrams of oviraptorosaur pterygoids in right lateral (top) and ventral (bottom) views illustrate early acquisition of derived features in the caenagnathid lineage. Scale bars are 1 cm. b Size and diversity of Cretaceous angiosperm fruit (increasing to the left), using data from Naware and Benson⁸. Grey dots show estimated volume of fruit for fossil occurrences through time, enveloped line shows moving average of fruit size through time, with 95% confidence interval, and coloured points show the mean volume and number of lineages producing drupe-like fruit in 5-million-year time bins. c lineage-through-time plots (increasing to the right) of Oviraptorosauria based on a posteriori FBD-scaled Bayesian (blue) and parsimony (red) tree topologies, showing a drastic increase in the late Campanian, coincident with mean size increases in drupes and in fleshy fruit overall (b). Blue boxes show ages estimated by the morphological clock analysis for key nodes numbered in (a). Alb Albian, Apt Aptian, Barr Barremian, Ber Berriasian, Camp Campanian, Ceno Cenomanian, Con Coniacian, Hau Hauterivian, Maas Maastrichtian, Sant Santonian, Tur Turonian, Val Valanginian.

Funston, G.F., Dudgeon, T.W., Simon, D.J. et al.
Fruit-eating adaptations in a Late Cretaceous oviraptorosaur (Theropoda).
Nat Commun (2026). https://doi.org/10.1038/s41467-026-78343-w

Copyright: © 2026 The authors.
Published by Springer Nature Ltd. Open access.
Reprinted under a Creative Commons Attribution 4.0 International license (CC BY 4.0)

Infernogallus jamesi illustrates how the Theory of Evolution provides a framework for understanding a discovery. Its unusual skull is placed within a family tree, compared with the anatomy of its relatives, and examined for changes in the function of inherited structures. Its probable fruit-processing adaptations make sense as modifications within a theropod lineage whose distant ancestors were carnivorous. The resemblance to fruit-eating parrots supplies a further evolutionary comparison: separate lineages arriving at similar functional solutions.

That framework also allows scientists to investigate a wider ecological question. Could the evolution of specialised fruit-eating dinosaurs have been associated with the increasing availability and diversity of fleshy fruits produced by flowering plants? The researchers identify a possible connection between the diversification of these dinosaurs and that of an increasingly important food resource. This is precisely the sort of relationship evolutionary theory leads scientists to investigate: changes in one group of organisms creating opportunities and potentially altering selection pressures in another.

However, a correlation is not, by itself, proof of a particular causal relationship. Fruit-bearing angiosperms might have provided opportunities that favoured dietary specialisation among dinosaurs, while fruit-eating animals might, in turn, have influenced plants through their feeding behaviour. Demonstrating reciprocal evolutionary change would require further evidence. In particular, the ability to process fruit does not establish that Infernogallus dispersed viable seeds, still less that it drove the evolution of larger or fleshier fruits. The study opens these questions to investigation rather than declaring them settled. Co-evolution does, however, offer a plausible explanation for the rise of fleshy-fruit bearing plants before the evolution of fructivorous birds and mammals. This distinction between an explanatory framework and a predetermined answer is central to science. Evolutionary theory helps researchers formulate questions, compare alternatives and identify evidence that could support or challenge their interpretations. Uncertainty about the precise diet or ecological role of one dinosaur is an invitation to further research. It is not evidence that the evolutionary relationships revealed by its anatomy disappear whenever a detail remains unresolved.

For young-Earth creationism, meanwhile, the chronological problem remains unavoidable: this specialised animal inhabited a complex ecosystem approximately 66 million years ago. For intelligent design, simply attributing its feeding apparatus to a designer adds no testable explanation for its inherited anatomical relationships or its resemblance to independently evolved feeding mechanisms. Science can use those relationships to reconstruct a changing world and investigate how its inhabitants influenced one another. Creationism supplies a conclusion in advance; the fossils supply an evolutionary history.

Advertisement

Amazon