The Founder of the Smithsonian Institution Figured Out How to Brew a Better Cup of Coffee

Almost two hundred years ago, James Smithson devised a method for better brewing. We recreated it.

James Smithson
Founder James Smithson (1765-1829) published a paper in search of better way to brew coffee and then considered how his method might work with hops to make beer. NPG/ Wikimedia Commons / Flickr user Marcelo Braga / Flickr user sammydavisdog

Englishman James Smithson is best known for leaving his personal fortune to the United States government for the creation of the Smithsonian Institution. But Smithson, who died in 1829, was more than just a wealthy philanthropist. He was an accomplished scientist who published research papers on many subjects, including how to make the best cup of coffee.

Smithson published his paper on coffee in 1823 in a monthly publication called Thomson's Annals of Philosophy, which was sort of a combination between a scientific journal and a modern popular science magazine. Smithson "enters chemistry when it's just beginning," says Pamela Henson, director of the Institutional history division of the Smithsonian Institution Archives. "And they have no idea of all the things they are going to be able to do. For example, science is much more generalized back then. You were looking at everything in the world. You don't have the broken down disciplines like you do now."

In an age before automatic drip coffee machines, Smithson was trying to solve several problems at once on his way to the perfect cup of coffee. Smithson wanted the coffee to be properly hot; economically used; and above all else he was striving for “the preservation of the fragrant matter.”

He had probably noticed the same thing that generations of later coffee drinkers would figure out. The better the smell of coffee brewing, the less flavor the coffee will have. When aromatic compounds are driven out of the coffee during brewing, less flavor remains to the coffee drinker. Smithson wanted to find a way of keeping those aromatic compounds in the coffee.

Smithson instructed the reader to put coffee grounds in a glass bottle. Then, to pour cold water over the grounds and put a cork loosely in the mouth of the bottle before placing the bottle in a pot of boiling water. When the coffee is done, the bottle is removed from the boiling water and allowed to cool without removing the cork. This gives those aromatic compounds time to condense from their gaseous form and seep back into the liquid of the coffee. Next, Smithson's method called for pouring the coffee grounds and liquid through a filter, then quickly reheating the sieved coffee to drink it.

Would this brewing system work? Was Smithson really keeping any extra flavor in his coffee? And would this same idea make beer any better, as he also suggested? To find out, I recreated and taste-tested Smithson's long-forgotten idea. But first I had to fill in some gaps.

James Smithson
Turns out the founder of the Smithsonian James Smithson, a scientist by training, figured out how to brew a pretty good cup of coffee. National Portrait Gallery

Most recipes written before the 20th century are short on details and exact measurements. Smithson doesn't say what volume of water to use, how much ground coffee to add, or what shape and volume of glass vessel to select. Not very scientific. But people's taste in coffee probably varied as much in 1823 as it does today.

Some prefer a strong brew and others like something weaker. How many cups do you intend to drink at once? There was no point in getting specific about the recipe. Smithson was offering a method that he knew everyone would adapt to their own taste.

I selected a clear wine bottle to brew in because a tinted glass would make it difficult to judge when the coffee was ready. Out of concern that the bottle could explode under pressure, I decided to leave about a third of its volume empty so that a small amount of steam could build up.

While a bottle of Smithson coffee was warming up on one burner, I heated an identical volume of water on another burner in order to prepare my control group. I needed to compare Smithson's system to something, so I chose the popular pour-over method using a Chemex. (The Chemex-style pour-over method was not popular during the early 19th century, but I chose it for the control group because it is the favored method of most modern coffee connoisseurs.)

In Smithson's era, he was comparing his method against two types of coffee preparation that are no longer common in either his native England or the United States.

The most common method was to heat a pot of water over a fire and toss coffee grounds into the pot. When the grounds sank to the bottom, the coffee would be poured into cups and served. Beginning around 1800, there was also a preparation known as percolation, which was not the same thing as the tall, cylindrical percolators that were popular in the U.S. until the late 1970's before drip coffee makers became state-of-the-art. Percolation of Smithson's era involved pressing coffee grounds into a short, even cylinder and pouring boiling water through a metal filter.

The idea of approaching coffee as a subject of serious scientific inquiry began with the 1813 publication of an essay entitled “Of the Excellent Qualities of Coffee,” by Sir Benjamin Thompson, Count Rumford. Thompson also designed Munich's famous English Garden; as well as a furnace to produce quicklime; and he invented thermal underwear.

In his essay, he outlined the problems with making the perfect cup of coffee and offered an early method of percolation to counter them (Thompson is very precise in his recipes, measurements and instructions for making novel coffee roasting and brewing equipment. Any reader interested in diving deeper into the recreation of coffee history should start there).

Thompson identified the most aromatic chemical part of the coffee that he believed was lost through boiling. “. . .This aromatic substance, which is supposed to be an oil, is extremely volatile,” wrote Thompson, “and is so feebly united to the water that it escapes from it into the air with great facility.”

Preventing the loss of this aromatic oil was a focus of both Thompson and Smithson's research into coffee. Smithson's paper was almost certainly intended in part as a belated response to Thompson's essay.

That lack of clear lines between disciplines was why scientists of the early 19th century were able to move between subjects as far-ranging as Smithson's coffee experiments and his better-known work on chemistry and geology.

"There isn't the distinction between academic science and practical science back then," says Henson. "So it's not that unusual for him to be interested in the coffee. At that time coffee is a very precious substance. So you wanted to get the maximum effect from whatever coffee beans you had. By doing it with that closed vessel, you got the maximum effect and it didn't just go up in the air through steam."

Smithson's best-known scientific work was on the subject of a group of minerals called calamines. Calamines contain varying amounts of zinc, a valuable metal. Miners "would go after these veins of calomine not knowing how much zinc they were going to get out of it," Henson says. But often the effort would be wasted when they later found that a particular deposit of calomine was low in zinc. "He came up with this method for finding out how much zinc was in there before they began mining. So you see all of those zinc rooftops in Paris, Smithson really enabled that."

As my bottle sat in the boiling water for eight minutes I was surprised to find that the water within it never came to a boil and so the cork was never in danger of being blown off. I removed it from the pan of water when the color looked suitably dark enough.

Four cups of coffee from each method were prepared in identical glasses marked only with a number. Number one was made in the Chemex and number two used Smithson's method. The tasters had no idea which they were about to drink.

“Number one is more robust,” said Dale Cohen, one of my taste-testers. “Number two is smoother, lighter.”

“It's a very stark difference to me,” said Stefan Friedman, another taster. “I want to say there is less bitterness and acidity in number two.”

There was no question that each type of coffee tasted different. But including myself, half of the my taste-testing subjects preferred the modern pour-over method and the other half preferred Smithson's coffee.

Sitting among a group of colleagues discussing scientific ideas over coffee, as we did while experimenting with Smithson's method, would have been a very recognizable scenario to James Smithson. 

"He is a part of what is called coffee house culture," says Henson. "Very early on he's at Oxford he's hanging out with [British scientist] Henry Cavendish and people like that. And he's hanging out in these coffee houses and this is where you discuss your scientific ideas. He's the youngest member of the royal society. . . He has this focus on practicalities."

During the following month, I experimented more with Smithson's method. Leaving the bottle in boiling water for 15 minutes instead of eight minutes yielded better results. I noticed more flavor in the coffee. When I was in a hurry, I tended to use the pour-over method. But if I had plenty of time to wait for the coffee to cool before removing the cork, I found myself gravitating towards using Smithson's method.

One more line in Smithson's paper intrigued me as my experiments came to a close.

“Perhaps [this method] may also be employed advantageously in the boiling of hops, during which, I understand, that a material portion of their aroma is dissipated,” Smithson wrote.

As a life-long homebrewer, I decided to apply Smithson's corked bottle method to the brewing of beer. At C'Ville-ian Brewing Company in my home town of Charlottesville, Virginia, I talked the manager into allowing me to appropriate his brewing system in order to make an experimental 30-gallon batch of 1820's styled India pale ale at the brewery.

In a dozen glass bottles, I placed all of the boiling hops that are used to make beer bitter. In place of the plain water used in the coffee experiment, I used a mixture of water and malt in the bottles (some of the desirable chemicals in hops are not fully soluble in water that does not also contain malt). My hope was that the aromatic compounds that are usually driven off during the 90-minute boiling process would be retained in the beer, making it more flavorful. After the dozen bottles had been heated for 90 minutes in their water baths, I decanted them into the fermentation vessel along with the rest of the beer.

The result is an interesting beer that is worth drinking but does not resemble what would have been recognized as an India pale ale either in the 1820's nor today. I had hoped that this would produce some sort of super-IPA, but the beer tastes lighter and less bitter than a conventional IPA.

If I were going to try this experiment again, I would use Smithson's method for the finishing hops towards the end of the boil rather than for the boiling hops. But regardless of the outcome, I like to think that James Smithson would appreciate the effort that a reader had made to finally test his ideas, 193 years later.

Get the latest on what's happening At the Smithsonian in your inbox.