Robert FitzRoy was the captain of The Beagle—yes, that Beagleand, as a captain, he was particularly concerned about the weather. Who wouldn't be? Judge a barometer reading inaccurately, and your ship might nearly overturn in a storm. Predict the storm's coming, and by preparing properly, you can increase the chances the ship will make it.

In 1854, years after he and Charles Darwin sailed around the world, he was appointed head of what would become Britain's Met Office. His job was supposed to be analyzing years of data about wind, collected by Britain's naval fleet, but he added "weatherman" to his job description. Weather watchers had been making predictions about which way the wind would blow for centuries, but FitzRoy was one of the first people to try to make it into something of a science. "With a barometer, two or three thermometers, some instructions, and an attentive observation, not of instruments only, but the sky and atmosphere, Meteorology may be utilized," he'd write a few years later.

But his most important tool didn't collect data at all. As a weatherman, FitzRoy had one huge advantage over his predecessors: the telegraph.

To create a weather forecast, FitzRoy first had barometers placed at ports dotting Britain's coast, and every morning, he'd have barometer readings, along with information about wind and temperature, telegraphed to his office. These data points, so quickly collated in one place, allowed him to predict with some lead time, when storms would hit and where.

Even with his data, though, FitzRoy was depending more on rules of thumb that had highly specific predictions. (It wasn't until almost a century later that computers were able to model the movements of the atmosphere quickly and accurately enough to make consistently dependable forecasts a few days in advance.) The barometers he installed along the coast came with instructions such as:

When rising: In winter the rise of the barometer presages frost. In wet weather if the mercury rises high and remains so, expect continued fine weather in a day or two. In wet weather if the mercury rises suddenly very high, fine weather will not last long.

And over the course of his weather-predicting career, critics attacked his predictions for their variability. But he started off well. In 1861, he published his first newspaper weather forecast:

North—Moderate westerly wind ; fine.

West—Moderate south-westerly ; fine.

South—Fresh westerly ; fine.

This forecast, it happened, was largely accurate.