Rocks older than 541 million years contain C30 and C31 steranes — molecular fossils that are the geologically stable descendants of sterols, the cholesterol-like compounds found in cell membranes. For decades, these compounds have been tentatively linked to demosponges, which would make sponges the earliest animals. But the link was circumstantial: the steranes looked like what sponge sterols would become after hundreds of millions of years of diagenesis. “Looked like” is not proof. Abiotic chemistry, bacterial metabolism, or other organisms could potentially produce the same end products.
MIT researchers resolved the question by synthesis. They made all eight possible sterol precursors that could theoretically produce the target steranes through geological transformation. They then subjected each synthetic sterol to simulated diagenesis — the chemical conditions of heat, pressure, and time that convert biological molecules into their fossilized forms. Of the eight candidates, only two produced compounds matching the C31 steranes found in Ediacaran rocks. Both of those two are produced by modern demosponges.
The six failures are more informative than the two successes. If only two of eight possible precursors yield the observed product, the probability of a non-sponge origin drops sharply. The argument is not “sponges make these compounds.” The argument is “almost nothing else does.” The negative results constrain the space of possible explanations far more than the positive results expand it.
This is a general pattern in chemical paleontology, and in science more broadly. A positive detection tells you what is present. A systematic negative detection tells you what is excluded. The exclusion is usually more powerful because the space of alternatives is large and the space of correct answers is small. Showing that six of eight candidates fail eliminates six hypotheses simultaneously. Showing that two succeed confirms two possibilities but does not distinguish between them — for that, you need additional evidence, which in this case comes from the known biochemistry of living demosponges.
The technique also addresses the persistent worry about contamination — that modern biological molecules somehow infiltrated ancient rocks. Synthetic chemistry provides a clean comparison: the researchers know exactly what they started with because they built it. The match between synthetic products and geological signals is independent of any assumption about what organisms were present. The chemistry is the argument, not the biology.
What the work establishes is that sponges were present in the oceans before the Cambrian explosion, during the Ediacaran Period. It does not establish when they first appeared — only that the rocks sampled so far contain their chemical signatures. The actual origin could be earlier, potentially much earlier, but the molecular fossils degrade with time, and the geological record becomes sparser. The evidence says sponges were there. It does not say when they arrived. The synthesis tells you what made the compound. The stratigraphy tells you when the rock formed. The gap between the two — between chemical identity and temporal precision — is where the remaining uncertainty lives.