Исследование доказало, что ископаемые остатки создают искаженную картину эволюции

Most of what we know about past life comes from fossils. But the fossil record isn’t a census of ancient life. Most animals don’t get fossilized, and the fossil record is just what survives after nature deletes most of the evidence. A new study turned that problem into a clever experiment. Researchers started with thousands of living land mammals, whose species, relationships, body sizes and diets are already known, then simulated which ones might make it into a future fossil record. When they tried to reconstruct evolution from this record, they ended up with a very different picture of life than what we see around us. Basically, if someone in the future were to look at today’s fossil record, they’d be misled. So then, aren’t we also misled when thinking about the past? Some things are more likely to become fossils Becoming a fossil requires a remarkable chain of events. An animal must die somewhere its remains can escape destruction long enough to become buried, usually in suitable sediment. Geological processes must preserve those remains for potentially millions of years. The rocks containing them must survive erosion and eventually become accessible. Then somebody has to look in the right place, recognize the fossil and study it. Each step removes information and reduces the likelihood of a future observer (us) having an accurate picture of what’s going on. The losses are also not random. A large animal with a broad geographic range has more opportunities to leave remains than a tiny species confined to a small patch of habitat. Some rock formations are exposed and heavily explored; others remain buried or inaccessible, covered by vegetation. Researchers have also studied some groups much more intensively than others. Then, some animals with hard body parts (like shells) are much more likely to become fossilized than others (like soft worms, for instance). That’s the main idea behind this experiment, but it’s turned upside down. Instead of starting from the fossil record and imagining what life looked like, the researchers knew what the complete answer should look like before they started deleting evidence. Making the cut They began with a family tree containing 3,710 living land-mammal species. Then they progressively removed species according to several familiar fossil-record biases: whether suitable sediments existed within their ranges, the size of their ranges, their body size, where fossils have historically been collected and which mammal families have received more attention. They tested three levels of filtering and repeated the process hundreds of times, producing 4,200 simulated samples. After each round, they asked standard evolutionary methods to reconstruct patterns in mammal body mass and diet from whatever species remained. In other words, instead of having the puzzle and looking for the key, they started with the key and tried to see how they’d do the puzzle. The fake fossils rewrote mammal history The team found that if someone were to look at a fossil record from today’s time, they’d have a very skewed idea of what’s going on. Animal size is a big part of it. Most living mammals are small. Because smaller animals were less likely to survive the researchers’ simulated preservation filter, the resulting “fossil” communities became disproportionately rich in medium-sized mammals and herbivores, while small-bodied and animal-eating species disappeared. So, a future researcher would know the larger animals far better than the smaller ones. Bats, rodents and the group containing shrews, moles and hedgehogs were particularly vulnerable. Under the most severe sequence of filters, the final simulated samples retained only about 1.3 to 3 percent of the original 3,710 species. In some scenarios, entire major groups effectively vanished. And once those species disappeared, the evolutionary reconstruction changed. Using the complete modern dataset, the researchers estimated the body mass of the ancestral mammal in their tree at roughly 790 grams — around the weight of a small rabbit. But judging by what would survive in the fossil records, the reconstructed ancestor would be over 10 kilograms (around the weight of a Corgi). The inferred diet could flip too. With the full dataset, the mammal ancestor was most likely a faunivore — an animal that ate other animals. After filtering, analyses could instead favor an omnivorous or herbivorous ancestor. What does this mean? Paleontologists have known for generations that the fossil record is incomplete. That’s not the problem. The unsettling part is that a fragmented record doesn’t necessarily produce a blurrier or incomplete version of the truth. It can produce a very coherent but false story. Add this train of thought to the age of the dinosaurs, and you realize we may have a very skewed picture of what kinds of animals dominated, and how body size and diet changed. If small-bodied animals, rare species or entire branches of a family tree are systematically less likely to survive as fossils, the species we do recover may create patterns that were much weaker or didn’t exist at all. The same principle could matter whenever scientists use fossils to reconstruct ancient ecosystems or long-term evolutionary trends: the surviving fossils may be a biased sample that tells a coherent story, but not necessarily the right one. So, ultimately, this doesn’t show that fossil analysis or reconstructions are wrong. Rather, they show that they’re quite likely reaching distorted conclusions because the available information is incomplete. Hopefully, they’ll find ways to account for this. The study was published in Nature Communications.