
Many flowers make nectar to attract visits from pollinating insects – a classic example of a mutually beneficial relationship refined by millions of years of evolution. But a reanalysis of a Jurassic fly suggests pollinators may have been flying around on Earth long before the first flowers bloomed.
Alexander Khramov, of the Borissiak Palaeontological Institute in Moscow, Russia, and his colleagues have been studying a specimen of Archocyrtus kovalevi, a late Jurassic fly first described in 1996.
When the team zoomed in on a long, straight structure lying underneath the fly’s compressed body with microscopes, they realised it wasn’t a piece of vegetation, as previously assumed, but rather a part of the fly. The structure isn’t enriched in carbon, as would be expected for vegetation, and a food canal running through its middle is clearly visible.
Advertisement
They report that A. kovalevi was about a centimeter long with a large, drinking straw-like proboscis nearly twice its body-length. As it doesn’t have any piercing structures, the team think the proboscis was for sucking nectar, not blood. But at about 160 million years old, the fly is over 40 million years older than the emergence of the angiosperms, the group of flowering plants.
“Finding a fossil insect of that age with such a wonderfully long proboscis is like finding a caveman with an AK-47,” says Khramov.

Since A. kovalevi lived in a flowerless world, Khramov thinks the fly’s giant straw was for reaching into the cones of gymnosperms, the group of plants that includes conifers. One candidate is Bennettiales, a now-extinct type of plant with almost flower-like cones that may have lured pollinators with hidden sugar drops. The fly’s proboscis is about the right length for the job, and fossils of these plants’ cones were found in the same layer of rock.
“For that time and place, members of Bennettiales are the most logical host plant for this fly,” says Carol Hotton, at the Smithsonian National Museum of Natural History in Washington D.C.
The discovery suggests the earliest pollinators visited gymnosperms and switched to flowers later on, shaping their evolution.
Alternatively, flowers may have evolved earlier than the fossil record suggests. But Khramov doesn’t think this is likely, as most known primitive flowers were shallow and simple, and unlikely to require probing mouthparts.
Gondwana Research