The first eight wolves arrived through the Roosevelt Arch on the morning of 12 January 1995, in a horse trailer escorted by two park service patrol cars.
The wolves had been live-trapped in three different packs in Jasper National Park and the surrounding wilderness of Alberta, Canada, weighed, fitted with radio collars, and flown south. Wyoming Farm Bureau Federation lawyers had obtained a stay from a federal appeals court before the plane landed, and the wolves spent the next several hours confined in their transport crates while the legal status of the project was resolved. The stay was lifted just after midnight.
The wolves were transferred, still in their crates, to three acclimation pens at Crystal Creek, Rose Creek, and Soda Butte Creek in the Lamar Valley. Six more wolves arrived later in the same month, bringing the first year’s total to fourteen. A second cohort of seventeen wolves arrived in 1996. By the end of that year, thirty-one wolves had been released into the park.
The last verified wolf killing in Yellowstone before the reintroduction had occurred in 1926, as part of a federal predator-control programme that had operated continuously across the park for the previous fifty years. The Greater Yellowstone Coalition’s published timeline of the reintroduction records the seventy-year gap as one of the longest deliberate ecological absences of an apex predator in American conservation history.
What happened in the thirty years after 1995 has become one of the most-cited and most-contested case studies in contemporary ecology.
What Yellowstone looked like without wolves
During the seven decades of wolf absence, the elk population in the park’s northern range grew approximately tenfold, reaching peaks of around 19,000 animals by the early 1990s. The elk browsed continuously on the woody plants of the riverine valleys, particularly willows, aspens, and cottonwoods. Streams in the Lamar Valley and elsewhere that had supported beaver colonies in the 1920s had, by the 1950s, lost most of their beavers entirely. The beavers depended on willows and aspens for both food and dam construction; the elk had eaten the willows faster than the willows could grow back. Without beaver dams to slow the water, the streams ran faster, cut deeper channels, and lowered the local water table further. The drier banks, in turn, made it harder for the surviving willows to recover.
By 1995, large stretches of the park’s northern range had been transformed from willow-dominated wetland-and-stream complexes into shorter grass-and-sage steppe. The elk grazed it. The few beavers that remained survived in marginal habitats. The cottonwood and aspen stands were ageing without new recruitment, because every young shoot that emerged was eaten by elk before it could grow above browse height.
This was the baseline against which the wolves’ return would be measured.
To understand better what happened, check out this short video on how the wolves changed everything.
The Ripple and Beschta thesis
The hypothesis that wolves had triggered a continent-scale ecological recovery was developed primarily by William Ripple and Robert Beschta at Oregon State University, in a series of papers published between 2004 and 2012. Their 2012 paper in Biological Conservation, titled “Trophic cascades in Yellowstone: The first 15 years after wolf reintroduction”, set out the most complete synthesis of their argument. The wolves, they wrote, had directly reduced the elk population through predation and indirectly reduced the elk’s browsing pressure on willows and aspens by altering elk behaviour, driving the surviving animals out of riverine corridors where they were most vulnerable to ambush. The released browsing pressure had allowed young willows and aspens to grow back. The recovering vegetation had stabilised stream banks, allowed beavers to return, and through their dam-building had begun to restore the hydrology of the northern range. The cascade flowed downward through the food web from the apex predator to the very water flow of the rivers.
The argument was published, widely cited, and ultimately translated for general audiences in a 2014 video produced by Sustainable Human, narrated by the British journalist George Monbiot. “How Wolves Change Rivers” has now been viewed more than 45 million times across YouTube and other platforms and remains one of the most-shared ecology videos in the history of the internet. The video’s claim is unambiguous. The wolves changed the elk. The elk left the rivers. The willows came back. The beavers returned. The rivers themselves, by the video’s closing claim, narrowed, meandered less, and ran clearer.
The video did not present the claim as contested. Almost none of the popular coverage of the Yellowstone reintroduction has.
The challenge from Colorado State
The published peer-reviewed challenge to the Ripple and Beschta thesis has come most consistently from a research team at Colorado State University led by Thomas Hobbs and David Cooper. Their work, beginning with a 2013 paper in Proceedings of the Royal Society B and culminating in a major 2024 paper in Ecological Monographs, has argued that the recovery story is substantially more complicated than the popular trophic-cascade account suggests.
The Hobbs and Cooper team conducted a 20-year experimental study on the northern range, beginning in 2001, that involved building artificial beaver dams on selected streams and fencing off some willow stands to exclude elk while leaving others open. The experiment was designed to test whether the absence of browsing alone could restore willow growth, or whether other factors, particularly stream hydrology and the height of the local water table, were the limiting variables.
The results, set out in their 2024 paper, were unambiguous in the opposite direction from the popular account. Willows grew taller where elk were excluded, but only modestly. They grew dramatically taller where beaver dams raised the local water table, regardless of whether elk were excluded. The willows that recovered most fully were those growing in stream sections where dams had restored the wetland conditions that had existed before the wolves were removed. Stream hydrology, in the Hobbs and Cooper interpretation, was the dominant variable. The reduction in elk browsing was real but secondary. The conclusion of the 2024 paper, in plain language, was that the restoration of apex predators to Yellowstone “failed to restore riparian plant communities on Yellowstone’s northern range,” and that the system had moved into what the authors call an “alternative stable state” caused by the original loss of beaver-driven hydrology, which the return of wolves had not been sufficient to reverse.
Ripple, Beschta, and several colleagues published a formal comment in Ecological Monographs in early 2026, responding to the Hobbs paper. They argued that the Colorado State experimental design had underestimated wolf effects by focusing only on a subset of the northern range, that the willow-recovery measurements Hobbs and Cooper had used were too narrow to capture the broader cascade, and that the comparison between their sampled streams and the wider park was not representative. The exchange has continued through 2026.
What both sides agree on
Within the active debate, several findings are no longer disputed.
The wolves did come back. By 2024, the park population had stabilised at approximately 100 wolves in ten to twelve packs, depending on the year. The elk population on the northern range has fallen from its 1995 peak of approximately 19,000 to current figures of approximately 4,000 to 6,000, a decline of roughly 70 to 80 per cent. The decline was driven by multiple factors including hunting outside park boundaries, drought, the return of cougars and grizzly bears alongside the wolves, and disease, but the wolves were a substantial contributor. Some willow and aspen stands have grown taller. Beaver numbers on some streams have recovered. As Live Science set out in its December 2025 summary of the current state of the debate, the question is not whether changes occurred but how much of the change is attributable specifically to wolves rather than to the broader recovery of multiple predators and to hydrological factors that had nothing directly to do with predation.
The debate is, in this sense, narrower than the popular framing suggests. It is not about whether the trophic cascade exists in some form. It is about whether wolves are the dominant driver of it, or whether their role has been overstated in a way that obscures the contribution of beavers, of bison, of cougars, of bears, and of the simple hydrology of the streams themselves.
The wolf called 21
The human interest of the reintroduction has, over the past three decades, attached itself most consistently to a single animal. Wolf 21 was a black male born in the spring of 1995 in the Rose Creek pen, the offspring of two of the original Canadian wolves. He left the natal pen with his mother and siblings in March of that year, was recaptured for radio-collaring as a yearling, and joined the Druid Peak Pack in 1997 as the new alpha male after the previous alpha was killed. He led the Druid Peak Pack for the next seven years, fathered approximately twenty pups across multiple litters, never killed another wolf in any of the dozens of inter-pack confrontations he was observed in, and died in his sleep on a mountainside in June 2004 at the age of nine.
The naturalist Rick McIntyre, who arrived at Yellowstone in 1995 as a seasonal Park Service employee and stayed for the next twenty-five years, watched 21 for almost every day of his life. McIntyre’s 2020 book on the wolf, The Reign of Wolf 21, sets out a 250-page biography assembled from field notebooks accumulated over McIntyre’s continuous daily observation. By the measure of recorded observation, 21 is the most documented non-domesticated wolf in history.
What the next decade may show
The Yellowstone wolves are now in their fourth generation since the 1995 reintroduction. The popular story has run substantially ahead of the science it depended on. The peer-reviewed science is, in the form of the Hobbs and Beschta exchanges, openly working through the implications of a model that may have been simpler than the system it described.
What has emerged, on the available evidence, is something other than what the 45-million-view video promised. The ecosystem has not been straightforwardly restored to its pre-1920s condition. Streams that had been deeply incised during the elk-dominated decades have, in many cases, not returned to their previous channel forms. Willow stands recover in some places and not in others. Beaver populations are higher than they were in 1995 but lower than they were in 1925. The wolves are part of the system. So is everything else.
The reintroduction did not, on the strongest current reading of the evidence, change the course of the rivers in the way the popular account suggests.
What it did do, on the same evidence, was something more incremental, more complicated, and more honest. The wolves came back. The system began to move. The direction of the movement is still being measured.
