The two paleontologists dissolving rock cores more than 200 million years old were looking for freshwater algae. Instead, fragments of insect scales caught their eye — remnants that a report published recently in Science Advances identifies as the oldest evidence of butterflies and moths.
A series of fortunate events led to this discovery, which dates the insects to around 70 million years earlier than previously known, before there were flowers that they could pollinate.
In the fall of 2012, Paul K. Strother of Boston University, an expert in prehistoric pollen and spores, traveled to Germany to the lab of Bas van de Schootbrugge, a fellow microfossil paleontologist. There they dissolved cores dating to the late Triassic and early Jurassic periods by exposing the material to an acid that erased everything but fossilized organic material.
What grabbed Strother’s attention were tiny scales. “It struck me that these looked like butterfly scales,” he recalled.
Butterfly and moth wings are covered in tiny scales that overlap like shingles on a roof. The scientists contacted experts studying modern insects, who deflated their hopes of identifying butterflies. The scales were described as “not diagnostic,” Strother said — meaning such parts did not belong only to a specific insect group. Some mosquitoes and flies have scales, too.
About a year later in Paris, Strother found himself seated at a dinner near a man named Torsten Wappler. The University of Bonn scientist, an expert in extinct insects, had just published an extensive phylogeny, or family tree, describing 479 million years of insect evolution. Strother, who kept images of the curious scales on his computer, whipped out the photos. Wappler examined them and told Strother that it would be possible to classify the insects. It would just take a lot of grunt work down the barrel of a microscope.
Van de Schootbrugge enlisted an undergraduate student named Timo J. B. van Eldijk for the task. “Timo is the guy that did all the work,” Strother said.
The acid reduced the rock cores to what van Eldijk called “black organic smudge.” Out of this smudge he had to isolate the scales, which to the naked eye look just like a pile of dust. He embedded the dust in a mixture of glycerol and water. Then, using a needle tipped with a human nose hair, van Eldijk managed to prod the scales into view beneath an electron microscope.
His investigation revealed that the scales were divided into two types. One set of scales was solid all the way through and compact as steamrolled almonds — “primitive,” van Eldijk called them. The other scales were hollow, which proved to be the critical discovery — “the real shocker,” he said.
Previous studies of insect phylogeny had shown the earliest moth and butterfly families with solid scales, van Eldijk said. They also had mandibles for chewing food. The insects that later split off the family tree developed hollow scales in their wings and also grew proboscises: long sucking tubes for drinking plant nectar.
In the textbook example of coevolution, butterflies developed proboscises in response to plants that developed flowers.
But plants did not evolve flowers until 130 million years ago, according to the earliest fossil flowers. “There’s two possible scenarios if these [insects] really are pollinators” with proboscises, Strother said. Maybe there is a missing record of Triassic or early Jurassic flowers. Or maybe the proboscis came first — the scenario that the study authors hypothesize is more probable.
“During the Jurassic, the most dominant group of plants were the gymnosperms, like your classic pine tree,” van Eldijk said. Conifer cones have indentations to catch male pollen. Insects might have drunk this pollen, if they had tubular mouths.
