A sharptail sunfish can weigh close to two tons, so watching one shatter like dropped glass is not something marine biologists see twice in a career. Off the coast of San José del Cabo in Mexico's Gulf of California, that is exactly what happened, not once but in two separate encounters recorded a little over a year apart. A new study published this week in the journal Frontiers in Ethology gives the behavior a name for the first time: hold-to-ram.

What the footage actually shows

The mechanics are almost theatrical. One orca clamps its jaws around the sunfish's stubby tail fin, called a clavus, and holds the carcass steady in open water. A second orca then accelerates toward it at speed. Just before impact, the orca holding the fish lets go, and the second animal slams into the body like, as one witness described it, a freight train hitting a watermelon. The tissue does not just tear, it fragments into thousands of pieces that drift through the water column.

Marine biologist Kathryn Ayres, lead author of the study and a scientist with the ocean conservation nonprofit Beneath the Waves, filmed the first encounter in July 2024 while working as a dive guide. A second, independent sighting came in September 2025, captured by ecotourism operator Héctor Franz in the same stretch of water. Both events involved the sharp-tail sunfish, a species that can stretch beyond three meters and weigh up to 2,000 kilograms, making it one of the heaviest bony fish in the ocean.

The kill happened before the ramming

Here is the detail that surprised researchers most: the ramming isn't what killed the sunfish. In both cases, the fish had already been killed and its internal organs removed by the time the hold-to-ram sequence began. That rules out the obvious explanation, that this is simply how orcas subdue prey. Instead, the researchers concluded it is almost certainly a butchery technique, a way of breaking down an otherwise unwieldy, tough-skinned carcass into pieces small enough for smaller mouths to manage.

That distinction matters biologically. It reframes the behavior away from predation mechanics and toward something closer to food processing, a step that in many species is culturally learned rather than instinctive.

Feeding the young, or just having fun

Once the carcass broke apart, the pattern of who ate what was consistent across both sightings. Juvenile orcas present in the group fed on the smaller floating fragments, while the adults who had done the ramming focused on the larger remaining chunk of the body. Ayres and her co-authors offer two competing explanations, and notably decline to pick one. The behavior may function as a form of parental investment, deliberately breaking food into pieces small enough for calves to handle on their own. Or it may simply be play. Orcas are well documented as animals that manipulate and toss prey for no clear nutritional reason, and the researchers note that entertainment cannot be ruled out.

Either explanation would fit what is already known about orca society, where killer whales are famous for developing hunting methods so specific to one population that they function almost like local dialects, passed down rather than instinctively wired in.

A pattern that fits a much bigger picture

Orcas already have a well-documented reputation for inventing region-specific hunting cultures. Pods in Norway coordinate to herd herring into tight balls before slapping them with their tails. Groups near New Zealand have learned to flip stingrays onto their backs to immobilize them. In Antarctica, orcas generate coordinated waves to wash seals off ice floes. Hold-to-ram now joins that catalog as a Gulf of California innovation, and because it has now been observed twice by two independent witnesses more than a year apart, researchers see it as a candidate for an emerging local tradition rather than a one-off fluke.

What makes the Gulf of California population particularly relevant to watch is that sunfish already rank among the most frequently hunted and eaten prey species in those waters, according to the study's authors. If hold-to-ram is becoming a standard part of how this pod processes large prey, it suggests killer whale culture in the region is still actively evolving, observable in close to real time rather than locked in place generations ago.

Why a niche behavior study matters more than it looks

It would be easy to file this away as a striking video clip and move on. But behavioral novelty like this carries weight for conservation science, because it demonstrates that even one of the most heavily studied predators on the planet still has undocumented layers to its social and hunting life. Every confirmed new behavior adds pressure on researchers and policymakers to treat orca populations not as a single uniform species, but as a patchwork of distinct cultures, each of which could be vulnerable to different threats, from prey depletion to noise pollution disrupting the coordination these hunts clearly require.

For a species this intelligent, the more precisely scientists can describe how a population actually behaves, the harder it becomes to manage or protect that population using generic, one-size-fits-all rules.

---

*Sources cited in this article include the peer-reviewed study published in Frontiers in Ethology, reporting and quotes from ScienceDaily, CNN, CBC News, and ABC News covering the July 2026 publication, and firsthand witness accounts from researchers Kathryn Ayres and Erick Higuera. All figures reflect reporting available as of July 25, 2026.*