Among the Heliconiini, a tribe of tropical butterflies that share a recent common ancestor, one species lives 14 days and another lives 348. A 25-fold gap in lifespan within a single phylogenetic group. Jessica Foley and her colleagues at the University of Bristol and the Smithsonian Tropical Research Institute published the numbers in Nature Communications in June 2026, drawing on field studies, public butterfly houses, and mark-recapture programs across 10 species. It is the highest variation in lifespan ever recorded in non-fish animals with such close kinship.
The short-lived species, Dione juno, is ordinary. Most butterflies survive a few weeks. The long-lived ones, the Heliconius genus, are the anomaly. And what separates them from their short-lived cousins is, on its surface, a dietary quirk: Heliconius butterflies eat pollen.
The only butterflies that eat pollen
Among all known species of Lepidoptera, the butterflies and moths, Heliconius are the only ones known to collect and digest pollen as adults. Most adult butterflies survive entirely on nectar: sugar water that fuels flight but provides no amino acids, no building blocks for repair. Heliconius land on a flower, coil their proboscis through the pollen grains, and dissolve them in saliva over hours. The pollen releases amino acids and immunity-boosting lipids.
Pollen-feeding Heliconius species averaged 177 days of maximum lifespan across the study. Their non-pollen-feeding relatives averaged 58 days. The pollen feeders showed lower baseline mortality, slower rates of aging, and longer reproductive windows. The differences were not marginal. They were categorical.
The most striking case is Heliconius hecale. In grip-strength tests, a standard measure of physiological decline in small organisms, H. hecale showed no detectable deterioration with age. Older individuals gripped just as hard as young ones. Dryas iulia, a closely related butterfly that does not feed on pollen, showed clear age-related decline in the same tests.
Foley's conclusion was that the butterflies had evolved both "longer lifespans" and "slower ageing" simultaneously.
The distinction between longer-lived and slower-aging matters. A longer lifespan could come from fewer predators or a safer habitat. Slower aging means the organism's biology is maintaining itself at a different rate. Something in these butterflies is running a repair program that their relatives are not.
Take away the pollen
The obvious hypothesis: pollen feeding sustains longevity directly. Amino acids fuel protein maintenance. Remove the fuel, lose the maintenance. If that were the whole story, a pollen-deprived Heliconius should age at the same rate as any other butterfly.
Foley's team tested this. They raised Heliconius hecale on a pollen-free diet and compared them to Dryas iulia, which never eats pollen. If the pollen itself were doing the work, the two species should converge.
They did not converge. Pollen-deprived H. hecale retained a substantial longevity advantage. It still maintained body mass longer. It still held muscle function. The gap narrowed, but the advantage persisted.
The longevity program is heritable. It lives in the genome now, not in the diet. Pollen did not grant H. hecale its long life. Something else did.
The scaffold and the building
What pollen did was make longevity selectively profitable.
For most butterflies, investing in cellular maintenance is a losing trade. They will be eaten or freeze before the investment pays off. The metabolic budget goes to reproduction instead: fast maturation, rapid egg production, short life. This is the default program across Lepidoptera, and the genes for cellular maintenance sit in every species' genome, silent.
Heliconius butterflies stumbled into a different economy. Pollen feeding gave adults a steady supply of amino acids, which made protein repair cheap. Their warning coloration and chemical defenses lowered predation rates enough that long-lived individuals survived to benefit from the investment. In that shifted economy, selection could discover longevity. The proteostasis genes that sit idle in every butterfly's genome became worth activating. Over evolutionary time, the activation became heritable, hardwired into developmental regulation rather than dependent on dietary input.
The pollen was the scaffold. It was necessary for construction: without it, selection never discovers the longevity program because the program never pays off. But it is dispensable for the finished building. Once evolution wired the maintenance program into the genome, the scaffold could be kicked away. The building stands.
The capacity, not the gene
The longevity field spends most of its energy looking for the thing that causes long life. A gene, a pathway, a molecule, a diet. The Heliconius story suggests a different question. The capacity for long life may already be present, dormant, in species we assume are locked into short lives. What differs is not capability but the ecological accident that makes the capability worth running.
This echoes a finding from regenerative biology published the same month. Ken Muneoka's lab at Texas A&M showed that mammalian fibroblasts carry the full regenerative program for regrowing bone, joint, and tendon. They do not express it because the faster scar-formation program runs first. The capacity was never lost. It was overridden by a cheaper default.
In both cases, the received story is backwards. We say mammals traded regeneration for speed. We say most butterflies are short-lived because they lack the longevity machinery. Neither is true. The machinery is there. What is lacking is the set of conditions under which it pays to turn it on.
The scaffold is not the building. It is the set of conditions that lets the building be discovered. And the most interesting question about any scaffold is whether the building can stand without it.