It’s not astonishing; it’s convergent evolution and it points to design in nature:
Whether nectar-sucking butterflies or blood-sucking mosquitoes — the ingestion of liquid food has long been known for many insects and other arthropods. A research team from Germany and Switzerland, led by the Leibniz Institute for the Analysis of Biodiversity Change (LIB) and the University of Bonn, now shows that millipedes also use a sucking pump to ingest liquid food. A sucking pump has thus evolved independently in different groups of organisms over several 100 million years. In the process, astonishingly similar biomechanical solutions for ingesting liquid food have evolved in widely distant animal groups. The study results have now been published in the journal Science Advances.
University of Bonn, “February 17, 2022” at ScienceDaily (February 17, 2022)
Millipedes, like insects, are arthropods but that doesn’t, in itself, point to much detailed similarity.
Millipedes in general:
Using high-resolution tomography as well as histological methods and electron microscopy, the researchers discovered a sucking pump in millipedes that is strikingly similar to those of insects. It consists of a chamber that is widened by strong muscles to suck in liquid food. “Together with the protractible mouthparts the sucking pump enables these millipedes to ingest more or less liquid food,” explains Leif Moritz, a doctoral student at the University of Bonn and the LIB.
The research team was thus able to show that the functional tools for a diet with liquid nutrients have evolved several times independently in all major subgroups of arthropods. “The biomechanical-morphological similarities between the groups of organisms indicate the strength of selection as soon as a food source provides even a slight evolutionary advantage,” elaborates Alexander Blanke head of the working group for evolutionary morphology at the University of Bonn.
University of Bonn, “February 17, 2022” at ScienceDaily (February 17, 2022)
Wait. There is a fundamental conceptual error in that last remark by Alexander Blanke (though it may have been something he felt forced to say): The question is not whether a sucking pump would be an advantage but how it could have arisen independently twice by natural selection acting on random mutations within the time available. And no, “natural selection” is not supposed to be a synonym for “hocus pocus.” The situation should be researchable, if not now, at least at some point. The probability of a non-design origin can become calculable once we assign specifics.
The paper is open access.
You may also wish to read: Evolution appears to converge on goals—but in Darwinian terms, is that possible?