Aquatic insects are among the most reliable sentinels of freshwater health, and few groups are as sensitive as stoneflies. Now, a monumental compilation of nearly 50,000 occurrence records spanning almost 170 years has delivered the most complete picture ever assembled of the stonefly fauna of the US Midwest, and the picture is sobering. The dataset, published in the Biodiversity Data Journal, covers Illinois, Indiana, Iowa, Michigan, Minnesota, Ohio and Wisconsin, and documents 157 species from 41 genera across nine of the ten Nearctic stonefly families. It also confirms that every single state in the region has lost stonefly species, with Illinois suffering the heaviest toll.
The effort drew on specimens held in 26 regional natural history museums, a review of published literature records stretching back to the early 1860s, state agency biomonitoring data, and decades of targeted fieldwork by the authors since 1987. The resulting dataset contains 49,046 occurrence records representing 221,334 specimens, standardised to Darwin Core Archive format so that the data can be indexed by the Global Biodiversity Information Facility and reused by researchers, conservation planners and water quality agencies. Records range from the earliest archived specimens collected in 1860 through collections made as recently as June 2025.
The taxonomic payoff is substantial. The compilation adds 21 new state records, including three species recorded for the first time anywhere in the Midwest. Minnesota gained nine new state records, Wisconsin seven, Ohio five, and Illinois, Indiana, Michigan and Iowa one or two each. These additions raise the known state richness to 85 species in Illinois, 93 in Indiana, 47 in Iowa, 71 in Michigan, 65 in Minnesota, 109 in Ohio and 68 in Wisconsin. Even so, richness estimates derived from non-parametric species accumulation models suggest the true regional fauna lies between roughly 164 and 172 species, meaning eight to 16 species may still await discovery in undersampled drainages or in museum drawers holding specimens of recently revised genera.
The geographic pattern of the data is striking. Collections have covered 91.4 percent of the region’s 623 counties and 91.1 percent of its 316 HUC-8 drainage basins, yet the density of records is far from uniform. Illinois alone contributed 11,072 records, followed by Ohio with 9,667 and Wisconsin with 9,520, while Iowa and Minnesota trail with 1,932 and 3,496 respectively. All 28 unsampled drainages and counties sit in western Iowa and Minnesota, a bias the authors attribute to the long history of taxonomic expertise stationed at eastern institutions and the remoteness of the region’s western reaches. Species richness itself declines along both east-west and south-north gradients, with Ohio hosting the most species and Iowa the fewest.
Family-level patterns reveal where the region’s diversity is concentrated. Perlidae leads with 42 species, followed by Perlodidae with 38, Capniidae with 21, Chloroperlidae and Nemouridae with 15 each. These five families account for 84 percent of the regional fauna. The most diverse genera are Isoperla with 24 species, Allocapnia with 16, Perlesta with 12, Acroneuria and Alloperla with 11 each, and Neoperla with 10. Ohio stands out as the only state where all nine eastern Nearctic families occur, including the single regional representative of Peltoperlidae, while Iowa lacks both Chloroperlidae and Peltoperlidae. Twenty-three species were found throughout all seven states, but 37 species are known from just one state, with Ohio claiming 24 of these regional endemics.
Identifying so many specimens to species demanded exacting laboratory work, particularly for the largest and most difficult families. Pinned adults, some dating to 1883, made up 2.3 percent of records and had to be relaxed in humid chambers for one to two days before examination. For males of Perlidae and Perlodidae, confident identification often requires everting the aedeagus, an internal structure that resembles a rolled-up sock within the abdomen. The researchers clipped abdomens at the second or third segment, macerated soft tissue in cold potassium hydroxide, and used probes and fine forceps to apply hydraulic pressure that pushed the aedeagal sac through the genital pore, revealing the spinule patches and sclerite patterns essential for species determination. Fluid-preserved specimens, which account for over 93 percent of records, were cleared in hot KOH using a double-boiler of beaker and shell vial.
Fieldwork between 2006 and 2025 prioritised historical localities and the best remaining habitat, alongside degraded streams where losses were suspected. Adults were gathered with beating sheets, sweep nets and hand-picking from rocks, logs and bridges, and attracted with ultraviolet lights along streams. Because at least 60 percent of regional species cannot be identified from nymphs alone, the team also reared mature nymphs in modified Styrofoam cups suspended in Living Stream tanks that mimicked source stream temperatures, then extruded aedeagi from freshly darkened adults under dissecting microscopes. Newly collected specimens are now preserved first in 70 to 80 percent ethanol to keep tissues supple before transfer to 95 percent ethanol for molecular work.
The conservation headline is unambiguous. Thirteen species, eight percent of the regional fauna, are known solely from historical collections, and ten have not been recollected since 1970, nine of them not since before 1960. Four species, including two Isoperla described by Theodore Frison and Philip Harden’s team, are presumed extinct. Neoperla mainensis has not been recollected anywhere across its broader distribution since 1944. Illinois fared worst, with 16 species absent from collections made after 1970, followed by Indiana with nine, Michigan and Ohio with eight each, Minnesota with seven, Wisconsin with two and Iowa with one. Measured against a less conservative 1980 threshold, Minnesota’s losses climb to 13 species, or 20 percent of its fauna.
The extirpations are not evenly distributed across the tree of life. Perlidae lost 17 species from at least one state by 1970, some 40 percent of its regional richness, and Perlodidae lost eight, reflecting the vulnerability of these large, often cold-water predators to warming, sedimentation and degraded water quality. By contrast, every species of Peltoperlidae and Taeniopterygidae was recovered in contemporary collections, offering a sliver of hope that intact habitat can still sustain sensitive fauna where it survives.
The authors emphasise that none of this would have been possible without natural history museums and the specimens archived within them, which preserved the evidence of species’ historical presence from which modern comparisons could be drawn. The Midwest dataset is framed as the first step toward a continental understanding of North American stonefly richness, with a parallel treatment of the northeastern states already in preparation. By identifying where species once lived and where they have disappeared, the dataset is intended to guide conservation prioritisation and to underpin future distribution and occupancy modelling, ensuring that the quiet decline of these clean-water indicators does not go unnoticed.
Subject of Research: Historical and contemporary stonefly species occurrences and extirpations in the Midwestern United States
Article Title: Stoneflies ( Insecta , Plecoptera ) of the Midwest, USA: species checklist derived from nearly 170 Years of collection records
Article References: Hogan, P., DeWalt, R., & Grubbs, S. (2026). Stoneflies (Insecta, Plecoptera) of the Midwest, USA: species checklist derived from nearly 170 Years of collection records. Biodiversity Data Journal, 14, Article e208358. https://doi.org/10.3897/bdj.14.e208358
Image Credits: AI Generated
Keywords: stoneflies, Plecoptera, biodiversity, freshwater insects, extirpation, Midwest USA, natural history museums, species checklist, biomonitoring, aquatic insects, conservation, Darwin Core Archive
News Source: Gavin Prescott. (October 17, 2026). Stoneflies Vanishing Across the US Midwest as 170-Year Dataset Reveals Steep Losses. Scienmag.
