In a study appearing in The American Naturalist, scientists in China and the United States report the discovery of a symbiotic relationship between a tropical tree and a tiny moth which is one of the most intricately interconnected relationships between a plant and its pollinator ever described.

Examining dried specimens of the leafflower tree Glochidion lanceolarium collected 80 years earlier, botanist Shixiao Luo noticed something bizarre: dried fruits that burst open to reveal not only seeds, but tiny adult moths. Intrigued, Luo hunted down some live G. lanceolarium trees (the nearest ones were, conveniently, growing wild in the South China Botanical Garden where he works) and, with his colleagues Gang Yao, Ziwei Wang, and Dianxiang Zhang, set about observing these trees’ flowers and fruit over the course of a year to unravel their relationship with the tiny moths.

The moths were the species Epicephala lanceolaria, part of a genus which, as adults, pollinates the flowers of Glochidion trees and, as larvae, consumes seeds of the same trees. The relationship between Glochidion trees and Epicephala moths (leafflower moths) had been known to science since 2003. But the relationship between G. lanceolarium and E. lanceolaria was even more intimate and interconnected than that between other known species of leafflower trees and moths. E. lanceolaria caterpillars not only eat the seeds of their host tree, but they also spin their cocoons inside hollow chambers inside the fruit. The adult moths emerge inside these chambers right before the mature fruits split open—explaining the tiny dried moths Luo had found on the herbarium specimens. Instead of producing flowers and fruit continuously throughout the growing season like most of their relatives, G. lanceolarium take a year to develop their flowers into fruit. All fruits ripen at the same time in April, and over the course of a few nights, the fruits split open and the adult moths emerge, mate, and fly to the next year’s newly opened flowers to pollinate them and lay their eggs.

In collaboration with David Hembry (University of Arizona), Luo and Wang sequenced DNA from E. lanceolaria and compared it to all known species in the same genus. This analysis revealed an evolutionary enigma: these moths were not part of the same clade that pollinates all other species of Glochidion trees. Rather, they were an unknown, distantly related lineage with no known close relatives—possibly, in fact, more closely related to leafflower moths that pollinate other genera of tropical leafflower trees in Asia (Phyllanthus and Breynia). How G.  lanceolarium and its moth came to be associated is mysterious. Most likely, E. lanceolaria represents a lineage that, although ancient, shifted relatively recently onto the ancestor of G.  lanceolarium. The moth probably had other close relatives which are either now extinct, or not yet discovered. This work highlights the extent to which highly intimate and coevolved associations may be more evolutionarily dynamic than had been previously realized, and how much remains to be discovered about the biodiversity and natural history of tropical Asia.