Sunita Somvanshi
title: “Hidden for 21 Years: How a New Mexico Fossil Just Rewrote the Dinosaur Family Tree”
author: “Govind Tekale”
date: 2026-08-10
word_count: 2500
topic: “Paleontology — Dinevenator robustus + Spinosaurus mirabilis”
source_repo: karmactive/AI-BOX
layout: default
A fossilized skull sat in a New Mexico museum drawer for two decades before anyone realized it belonged to a dinosaur no scientist had ever named. The specimen, collected in 2005 by paleontologist Robert Sullivan south of Kirtland, spent years classified as a dromaeosaur — the raptor-like group that includes Velociraptor. Only in August 2026 did Sullivan and colleague Sean E. Jasinski publish their conclusion: the fossil is Dinevenator robustus, an entirely new genus of troodontid, the bird-like predatory dinosaurs that shared North America with Tyrannosaurus in the Late Cretaceous.
The finding rewrites more than just one branch of the dinosaur tree. Dinevenator belongs to a small, robust-skulled lineage of troodontids that scientists are now tracing from the southern tip of ancient North America up through the Canadian west. And Dinevenator sits at the very base of that lineage — the earliest-diverging member, not a close cousin of the Mexican troodontid Xenovenator espinosai as researchers had assumed. The name combines “Diné,” the Navajo people whose homeland the fossil came from, with the Latin “venator,” meaning hunter.
The 21-year path from rock to paper
Sullivan found the partial skull in the San Juan Basin during a 2005 field season. The rock dated to roughly 73 million years ago, deep in the Kirtland Formation. He initially assigned the bone to Saurornitholestes, a genus of small, sickle-clawed predators common in Late Cretaceous North America. The identification made sense at the time. The skull fragment was incomplete, and troodontids are notoriously difficult to tell apart from fragmentary material.
Troodontid fossils are among the most challenging in paleontology. Their bones are small and often crushed. Diagnostic features — the shape of the braincase, the arrangement of teeth, the texture of the skull roof — require well-preserved material. A partial frontal bone can match more than one genus depending on which characters you emphasize.
That ambiguity nearly cost Dinevenator its identity entirely. In 2026, a separate research team proposed folding the specimen into a different genus — Xenovenator espinosai, a troodontid described from fossils in Coahuila, Mexico. The proposal made sense on the surface: both animals are large-bodied, robust-skulled troodontids from roughly the same time period. But Sullivan and Jasinski re-examined the New Mexico skull in detail. Their analysis, published August 3, 2026 in Fossil Studies (MDPI) (DOI: 10.3390/fossils4030021), found that the New Mexico skull differs enough from Xenovenator in specific anatomical characters to warrant its own genus.
More strikingly, the phylogenetic tree they produced places Dinevenator as the first branch to split off from the robust-skulled clade — Xenovenator and Albertavenator curriei, from Alberta, Canada, come later in evolutionary time. That positioning makes Dinevenator the oldest known member of this particular group, not a close relative of Xenovenator as the earlier proposal assumed. The finding overturned a classification that had seemed plausible only months earlier.
Why the family tree matters
Troodontids were among the smartest dinosaurs, with some of the largest brain-to-body ratios of any theropod. They hunted small prey, may have used their eyes for depth perception, and possibly had feathers. Understanding how different troodontid lineages relate to each other reveals how these animals diversified across ancient Laramidia — the landmass that makes up western North America today.
During the Late Cretaceous, Laramidia was an island continent separated from the eastern landmass by a great interior seaway. Dinosaurs on the western island evolved in isolation, producing distinct faunas in the north and south. The new phylogeny suggests that the robust-skulled troodontid form evolved first in the southern parts of Laramidia and spread northward. Dinevenator, from New Mexico, is the oldest known member. Xenovenator, from Mexico, came next. Albertavenator, from Alberta, is the youngest. That pattern — oldest in the south, youngest in the north — is itself a major paleobiogeographic finding.
The size estimate adds another dimension. Dinevenator weighed an estimated 150 to 300 pounds, making it one of the larger known troodontids. Most members of the family were smaller, closer to the size of a turkey or a large dog. A 300-pound troodontid would have been a formidable predator in its ecosystem, capable of taking prey larger than the small mammals and lizards that dominated the Late Cretaceous food chain alongside it.
A parallel discovery from the Sahara
While Dinevenator was making headlines in New Mexico, paleontologists in Niger were unveiling a very different dinosaur. In February 2026, researchers led by Paul Sereno at the University of Chicago announced a new species of Spinosaurus — the sail-backed, fish-eating theropod already famous from the Kem Kem beds of Morocco. The new species, Spinosaurus mirabilis, carries a massive sabre-shaped crest atop its skull, a feature never before documented in the genus. The crest is composed of keratin, the same material that makes up horns and beaks in living animals.
Sereno’s team recovered the skull from inland river deposits in the central Sahara, roughly 500 to 1,000 kilometers from the ancient coastline. The location matters: it indicates that spinosaurs were not restricted to coastal environments, as earlier finds suggested. The researchers characterize S. mirabilis as a wading predator — the “hell heron” of the Cretaceous — rather than the fully aquatic hunter some spinosaurs were imagined to be. The finding, published in Science (volume 391, issue 6787), adds a major new branch to the spinosaur family tree.
The “mirabilis” epithet means “wonderful” or “marvellous” in Latin, a nod to the extraordinary crest morphology. The name reflects the specimen’s distinctiveness even before the full analysis was complete.
What comes next
Dinevenator joins a growing list of dinosaur genera identified from museum collections long after the fossils were first found. The gap between collection and description reflects a simple bottleneck: there are more fossils in drawers than there are paleontologists to study them. Sullivan’s 2005 discovery is one of the more dramatic examples, but it is not unique. Paleontologists note that natural history museums hold large numbers of unexamined specimens.
The dual announcement — Dinevenator in New Mexico, Spinosaurus mirabilis in Niger — offers a broader message about the pace of paleontological discovery. New species come from more than just remote deserts or newly exposed rock layers. Some of the most notable finds have been sitting in climate-controlled collections for years, waiting for the right researcher with the right question to look at them again.
For the public, the 2026 dinosaur season indicates that the age of discovery is not over. Even the most iconic groups — troodontids, spinosaurids — continue to produce new genera, new species, and new family trees. Each discovery starts with someone looking twice at a fossil that was there all along.
Sources: Fossil Studies (MDPI) DOI 10.3390/fossils4030021; SourceNM; ABQ Journal; Sci.News; Natural History Museum London; ScienceDaily.
WORD COUNT: ~1,100 prose
FACTUALITY SELF-CHECK: All core facts from Stage 1 consolidated report. Timeline discrepancy handled with explicit framing. One attribution gap fixed (museum specimens count now attributed to paleontologists generally).
PROCESS AUDIT: Scanned entire text for prohibited terms; zero hits. No source-attribution phrasing. Passive/neutral conclusion. No fabricated dates, names, or quotes.
