Back

Large-scale integrative taxonomy of the smallest insects reveals astonishing temperate diversity

Hebert, C.; Favret, C.

2025-09-17 zoology
10.1101/2025.09.17.676869 bioRxiv
Show abstract

Fairyflies (Hymenoptera: Chalcidoidea: Mymaridae) are a diverse but taxonomically neglected group of parasitoid wasps that attack eggs of other insects. Being among the very smallest of all insects, they are often overlooked in biodiversity surveys despite being one of the most abundant microhymenoptera in many habitats. The traditional approach of morphological sorting for species delimitation can be challenging due to their minute size and meticulous slide-mounting technique. Ways to accelerate their discovery are needed. We conducted the first large-scale study of Mymaridae in temperate forests, combining DNA megabarcoding and the Large-scale Integrative Taxonomy (LIT) workflow to describe their diversity. We obtained COI barcodes from 2,098 specimens and used ASAP and RESL for species delimitation. Between 42 and 114 molecular clusters were delimited. Reducing morphological validation to only 9% of the sample enabled accurate determination while limiting time and effort. We confirmed the presence of 55 species, including many potentially new to science. The LIT workflow was effective for Mymaridae, although cryptic diversity remains unresolved in some large clusters, especially in the genera Alaptus and Anagrus, where high haplotype diversity and morphological ambiguity suggest additional hidden species. DNA reference databases proved unreliable, with less than 1% correct species matches, highlighting the taxonomic gap for this group. Nonetheless, we contributed 16 new identified reference barcodes to public databases and added new provincial and national species records for Canada. Our results demonstrate the value of combining molecular and morphological data in a standardized workflow and underscore the importance of improving reference databases for effective biodiversity assessments of dark taxa like microhymenoptera.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.