The study of insect pollinator communities is at the heart of a great paradox: today, characterizing the diversity of these species implies sacrificing them (trapping and then morphological identification). Methods based on the amputation of a member (segment of antennae, piece of wings or legs) have been used in honey bees (Madella et al., 2021) or bumblebees (Holehouse et al., 2003) but they lead to mortality rates depending on the size of the specimens. Recently, different laboratories around the world are mobilizing to try to develop new non-lethal identification methods such as acoustics (Heise et al., 2020), deep learning: mathematical algorithms capable of identifying specimens from insect photos (Høye et al., 2021), and sampling from faeces (Scriven et al., 2013). At the same time, with rapid advances in sequencing methods, environmental DNA-based approaches (i.e., identification of individuals from traces left in the environment) have opened up promising prospects for inventorying aquatic or terrestrial biodiversity while preserving the integrity of specimens (Banerjee et al., 2022).We proposed to develop a non-lethal protocol to identify wild bees from the traces they can leave while foraging on flowers. It is based on the enclosure or exposure to the open air of strawberry plants and the collection of flowers for the extraction of insect DNA traces. The experimental protocol is composed of 4 strawberry plants placed in 3 conditions: a cage with insects, a cage without insects and exposed to the open air.Initially we extracted DNA from insect tracks left on flowers only in the "bees introduced into the enclosure" condition and by testing two types of extraction kits. The extracted DNAs were then amplified by PCR with the insect 16S minibarcode (Clarke et al., 2014). We had previously constructed a barcode baseline for wild bees with this minibarcode (results in publication). Several PCR conditions allowed us to select the DNA Extraction Kit that yielded sequencable amplifiates. Sequencing of 4 samples (2 duplicates) and bioinformatics processing of the data allowed us to find the sequences of the bees that were introduced but also sequences corresponding to 16S DNA of strawberry plants