Berlese vs. Winkler: sift once, extract twice

We present results of a comparison between Winkler (Moczarski) eclector and the Berlese-Tullgren funnel methods for extracting forest litter‐dwelling Coleoptera from sifted substrates, with a focus on relevance for taxonomic studies. A review of the history and methodologies of leaf litter arthropod extraction is provided. Repeated, timed trials were performed to compare the relative effectiveness of the two methods using litter samples from various sites across the southern United States. Extraction rates for Winkler and Berlese funnels were evaluated based on temporal and taxonomic criteria by comparing accumulations of Coleoptera specimens and species at timed intervals. In addition, Curculionidae, Staphylinidae, and “rare” species were targeted and evaluated separately. No clear differences in extraction efficiency, either temporally or taxonomically, between Winklers and Berleses were documented. But, contrary to current extraction practices, extraction patterns for both Winklers and Berleses suggested that intervals at the beginning of trials (3-24 hrs) and later in the trials (60-144 hrs) are most productive. The typical 24-48 hr run duration currently used in most studies fails to recover a substantial fraction of the total abundance contained in samples. An extraction protocol for litter-dwelling Coleoptera is suggested based on these results.