The Measurement That Measures Itself

You can't measure microplastics without adding microplastics.

That's the finding from a University of Michigan study that would be hilarious if it weren't so disorienting. Researchers discovered that the nitrile and latex gloves scientists wear to keep their samples clean are shedding thousands of particles that look, under a microscope, exactly like the microplastics they're trying to count. The instruments of measurement are participating in the measurement. The observer is contaminating the observation.

Somewhere, Werner Heisenberg is laughing.

Madeline Clough, the study's lead author, noticed the problem when her particle counts came back orders of magnitude higher than expected. Not slightly off. Not within an interesting margin of error. Wildly, impossibly inflated. The kind of result that makes you check your equipment, then check it again, then stare at your gloves and have a very bad afternoon.

The culprit: stearates. Manufacturers coat disposable gloves with these soap-like salt particles to make them slide easily off production molds. When a gloved hand touches lab equipment — substrates, filters, sample containers — it deposits roughly 2,000 false positive particles per square millimeter of contact. These stearates are indistinguishable from microplastics under both light-based spectroscopy and scanning electron microscopy — the standard detection methods.

The tools we built to see the contamination are blind to the contamination we're introducing by using them.

The study, published in Analytical Methods, tested seven different glove types. Standard nitrile and latex gloves — the kind worn in virtually every environmental science lab on the planet — were the worst offenders. Cleanroom gloves, which lack the stearate coating, released far fewer particles. The fix exists. But most researchers didn't know they needed one.

"We may be overestimating microplastics," said Anne McNeil, the study's senior author and professor of chemistry at Michigan, "but there should be none."

That sentence contains the entire tension. Yes, the data may be inflated. No, that doesn't mean the problem isn't real. The microplastics are still there. The question is whether we've been counting our own fingerprints alongside them — and if so, how much of what we've measured is signal and how much is the noise we brought with us.

Clough and McNeil developed methods to differentiate stearate particles from genuine microplastics, which means previously contaminated datasets might be recoverable. That's the good news. The unsettling news is that this contamination pathway existed for years across potentially thousands of studies before anyone caught it. Every researcher wore gloves. Every pair of gloves shed particles. Every particle looked like the thing being measured.

There's a pattern here that extends well beyond chemistry labs. Every measurement system interacts with what it measures. Every framework shapes the data it collects. Every tool carries assumptions — sometimes literally coating them in invisible particles — that become part of the result.

The microplastics are real. The crisis is real. But our picture of how much and where has been drawn, in part, by the hand holding the pencil. Which is either a devastating methodological failure or a reminder that observation is never passive.

The universe doesn't let you watch without participating. Even your gloves have opinions.

Sources:

• Nitrile and latex gloves may cause overestimation of microplastics, U-M study reveals — University of Michigan News, 2026-03-28 — Analytical Methods (RSC Publishing), 2026-03-26