February 11, 2026
Maureen Searcy
UChicago paleontologists investigate how life entered and adapted to the deep sea.
The deep sea is a dark, cold place. It’s just a few degrees above freezing, subject to immense pressure, and beyond the reach of the sunlight needed for photosynthesis. It is an inhospitable habitat for most forms of marine life, which primarily rely on microscopic plants. The life that does survive in such a hostile place must find a different way to thrive.
To help understand how certain species adapted to deep-sea living, scientists can look at when their ancestors moved into those harsh depths and whether they were already fit for such conditions.
A new study led by scientists at the University of Chicago examined the lineages of two groups of bivalves—marine invertebrates that include clams, oysters, mussels, and scallops—that successfully inhabit the deep sea. They found that some lineages already well-suited to the harsh environment moved into that habitat in a sporadic way without diversifying much once they were there. For other lineages, a single
“The deep sea is the biggest habitat on the planet, but very few lineages have actually managed to penetrate that environment,” said David Jablonski, the William R. Kenan Jr. Distinguished Service Professor of Geophysical Sciences at UChicago and co-author on the paper. “You might imagine one key to fit that lock, but we found that there are different ways of getting through.”
Tracing adaptations
Two of the best-sampled groups of bivalves include mussels and a type of marine clam called hatchet shells. These ancient families include hundreds of species found all over the world in deep and shallow water.
Jablonski’s team used these lineages as case studies to trace how and when these organisms became suited to different environments, including adaptations that allow them to live deep under water.
There is a wealth of publicly available information on these lineages, including molecular data, such as genetic and protein sequences. This material can be used to build a phylogeny, or an evolutionary family tree that helps infer the relationship between related organisms.
The team then integrated multiple datasets including molecular data, depth ranges, geographical distribution, and fossil records.
“From there, we conducted statistical analyses to test for patterns among these lineages,” said co-author Ava Ghezelayagh, a postdoctoral scholar in the Department of the Geophysical Sciences and the Data Science Institute, who led the data analyses. “For example, are shallow-water species that are closely related to deep-sea lineages more likely to live in colder temperatures? That might have prepared them to enter the deep sea.”
These analyses helped the team trace which lineages failed to make it to the deep sea, which ones got there but died out, and which ones made the leap and established residence.
‘Dribs and drabs’ versus a breakthrough
Most bivalves feed on phytoplankton, which rely on sunlight—a diet incompatible with the deep sea. Certain lineages of hatchet shells and mussels are successful in the deep sea because they have both developed a symbiotic relationship with bacteria that derive energy from sulfur, methane, and other chemicals released by hydrothermal vents and cold seeps at the bottom of the ocean. These bacteria live in the bivalves’ gills and share their energy with their hosts.
The results of the study show that these bivalve lineages managed to acquire this ability in different ways,
The hatchet shells set up their partnership with the bacteria in shallow water very early on, with evidence dating back to the early Paleozoic, more than 450 million years ago. For 300 million years, they seemed content to remain in the shallows, cooperating with their symbiotic bacteria.
Then in the mid-Mesozoic, “they began invading the deep-sea in dribs and drabs,” said Jablonski. Because of their relationship to the bacteria, “they were ‘preadapted’ for the deep-sea, and so—we think—just opportunistically slid an occasional species into the deep when the chance arose, but they almost never diversified down there.” The team calls this a “piecemeal model” of entry to the deep sea.
Mussels, on the other hand, followed an “in-situ diversification model.” About 60 million years ago, one lineage of mussels acquired the same partnership with bacteria for their nutrition, allowing it to break into the deep-sea habitat. That one entry diversified into at least 70 species.
“Of course, there are nuances and complications, and when we look across all bivalve lineages, we find some that fall somewhere in between,” said Jablonski. “But the idea that lineages can trickle into the deep sea and not do much or can send one branch down there with a key adaptation and explode
3D evolutionary tree
Studies like this depend on the completeness of the evolutionary family tree. While hatchet shells, mussels, and other bivalve families are well-sampled, there are still gaps in the molecular DNA data.
“Bivalves aren’t a sexy group,” said Ghezelayagh. “There’s not a lot of people sequencing them, other than the ones that are important for the food industry.”
A related but separate project that Ghezelayagh and Edie are working on is creating a more comprehensive phylogeny using machine learning.
To build an evolutionary family tree, scientists can use molecular data, like their genetic sequence, or morphology, the form and structure of their bodies. Ghezelayagh and Edie are working on a hybrid bivalve phylogeny by marrying these two types of data.
Jablonski’s team has 3D micro-CT scanned images of 90% of living bivalve genera, which can show physical traits such as shell shape and texture, muscle attachments, and details of how the hinge fits the two shells together. Morphological data from fossils can also be integrated into the hybrid evolutionary tree. But gathering such data is a slow and painstaking effort, requiring someone to go through each taxon one by one. Machine learning can speed up the process considerably.
“We can include living and extinct taxa, which you normally can't do in a molecular tree,” said Ghezelayagh. “This project allows for a level of completeness that’s never been attained. Then we can ask some really great new questions.”
Citation: Ava Ghezelayagh, Stewart M. Edie, David Jablonski; Alternative pathways into the deep sea: patterns in Bivalvia. Proc Biol Sci 1 February 2026; 293 (2064): 20252002.
Funding: NSF, NASA, the University of Chicago Data Science Institute, and Smithsonian Institution