The longest-running lab experiment on Earth has produced nine drops in 98 years, and nobody has ever seen one fall

 Most experiments are designed to produce a result within the working life of the person running them. This one was not, though its creator does not appear to have realized that at the time.

In 1927, Thomas Parnell, the first professor of physics at the University of Queensland in Brisbane, wanted to demonstrate something to his students about the nature of matter. He took a quantity of pitch — a tar derivative, the same material once used to waterproof wooden ships — heated it until it softened, and poured it into a glass funnel with a sealed stem.

His point was that pitch is not what it appears to be. At room temperature it looks and behaves like a solid. It is hard, it feels brittle, and you can shatter it with a hammer. Parnell wanted to show that despite all of that, it is a fluid, just an extraordinarily slow one.

He let it settle for three years. In October 1930 he cut the bottom of the funnel and let gravity take over. Ninety-odd years later it is still going, it holds a world record, and it has developed a devoted following. Here is the story of the slowest experiment in science.

How Slow Is Slow

The number that makes sense of everything else is viscosity, the measure of how much a fluid resists flowing. Water is low; honey is substantially higher; pitch is on a different scale entirely.

The University of Queensland describes pitch as the thickest known fluid, and the experiment has been used to establish that it is on the order of 100 billion times more viscous than water, with some calculations putting the figure closer to 230 billion times. Either way, the practical consequence is the same: it flows, but at a rate that is invisible to human observation.

The first drop took eight years to fall. The next several took roughly seven to nine years each. Nine drops have fallen in total, the most recent in April 2014, and a tenth is currently forming.

That works out to an average of roughly one drop per decade. The apparatus is not sitting in a climate-controlled chamber; it lives in a display cabinet, which means seasonal temperature changes affect the flow rate. When air conditioning was installed in the building, the intervals lengthened noticeably, from eight or nine years out to twelve or thirteen. The eighth drop took over twelve years, and nobody was entirely certain why, though the reduced weight of the diminishing pitch above may play a part.

Nine Drops, Zero Witnesses

The detail that turned a teaching demonstration into a minor global obsession is that in nearly a century, no human being has ever watched a drop fall.

This is not because nobody tried. It is because the event is both extremely rare and extremely brief: the pitch stretches for years, narrows at the neck over months, and then the actual separation happens in an instant that nobody has managed to be present for.

Parnell himself saw two drops fall in the sense that he saw the result, but never witnessed either in motion. He died in 1948, and the apparatus was briefly forgotten, sitting in storage until it was rediscovered.

Its rescuer was John Mainstone, a physicist who took over as custodian in 1961 and looked after it for 52 years, moving it into public view and turning it into one of the best-known objects in any laboratory anywhere. His run of misses is close to comic in its consistency. He missed a drop by a day in 1977. He missed the 1988 drop by about five minutes, having stepped away for a cup of tea. And in 2000, a webcam had been set up specifically to capture the event — and suffered a power interruption of roughly twenty minutes during precisely the window in which the drop detached.

Mainstone spent more than half a century watching the funnel and never saw it happen. He died in 2013, months before the ninth drop fell.