Two new species of entomopathogenic fungi — Paleoophiocordyceps gerontoformicae and Paleoophiocordyceps ironomyiae — preserved in 99-million-year-old Kachin amber from Myanmar, associated with an ant pupa and a fly respectively. They share morphological traits with modern Ophiocordyceps, the genus famous for hijacking ant nervous systems and compelling infected hosts to climb vegetation before death.

Divergence time analysis pushes the origin of Ophiocordyceps back to approximately 133 million years ago — 33 million years earlier than previous estimates. But the more revealing finding is the host-switching pattern reconstructed from the phylogeny. The ancestral Ophiocordyceps parasitized beetles. It jumped to ants and moths during the Cretaceous, and these host shifts coincided precisely with the diversification of Hymenoptera and Lepidoptera.

The fungi did not develop new infection mechanisms for new hosts. The molecular machinery for penetrating insect cuticle, colonizing hemolymph, and consuming internal tissues was already in place — it worked on beetles. What changed was the availability of targets. As ants diversified and built colonies, they created dense populations of immunologically similar individuals in enclosed spaces — ideal conditions for a pathogen that spreads through physical contact. The infection of the ant pupa found in the amber likely began inside the nest, since larvae do not leave it.

The through-claim: the parasite’s host range was determined by what existed, not by what it could infect. The capacity to parasitize ants preceded ants’ ecological dominance. The jump happened not when the fungus evolved the ability, but when the hosts became numerous enough to sustain the relationship. Opportunity, not mechanism, gates the transition. The same pattern appears whenever a generalist pathogen encounters a newly abundant host: the infection was always possible, but the epidemic required density.