On this page
In the 1790s many scientists thought heat was a substance, a weightless fluid that flowed from hot bodies to cold ones. Benjamin Thompson, Count Rumford, noticed something that did not fit while watching cannon being bored in Munich. His experiment boiled water without any fire, and his reasoning about it is still a model of how to test an idea. This post covers who he was, what he did, how to calculate the power involved, and what you can learn from him.
Who was Count Rumford?
The Royal Society's page on the Rumford Medal describes him as "an American-born former soldier, spy, statesman and scientist who would go on to found the Royal Institution". Library catalogues give his dates as 1753 to 1814, as in this record of his Essays. The same page gives his title as Count Rumford of the Holy Roman Empire.
He also left two lasting institutions. The medal named after him was first awarded in 1800 and is given for discoveries about the thermal or optical properties of matter. The Royal Institution was founded at a meeting at Joseph Banks's house on 7 March 1799 to teach science to the public through lectures and demonstrations.
What happened at the Munich arsenal?
Rumford's paper, An inquiry concerning the source of the heat which is excited by friction, appeared in the Philosophical Transactions for 1798. You can read it in the Internet Archive scan of volume 88.
He wrote that while superintending the boring of cannon "in the workshops of the military arsenal at Munich", he was struck "with the very considerable degree of heat which a brass gun acquires, in a short time, in being bored". He asked where the heat came from. Was it given up by the chips as they were cut away? He tested this and found that the chips' capacity for heat had not changed, so that explanation failed.
For the main experiment he had a hollow cylinder of gun-metal cast with a cannon. A blunt steel borer was pressed against its bottom while horses turned the cylinder, about 32 times a minute. The cylinder sat in a wooden box holding 18.77 lb of water at 60 °F. His readings:
after 1 hour, 107 °F;
after 1 hour 30 minutes, 142 °F;
after 2 hours, 178 °F;
after 2 hours 20 minutes, 200 °F;
after 2 hours 30 minutes, the water "ACTUALLY BOILED", as he printed it.
He described the astonishment of the bystanders at seeing so much cold water made to boil without any fire.
His conclusion, in his own words: "any thing which any insulated body, or system of bodies, can continue to furnish without limitation, cannot possibly be a material substance". He found it hard to imagine anything that could be excited and passed on in this way "except it be MOTION".
How much power did the horses put into the water?
You can estimate it with Class 11 calorimetry, which is in the JEE (Main) 2026 syllabus.
Mass. 18.77 lb × 0.4536 kg/lb = 8.51 kg of water.
Temperature rise. 60 °F to 210 °F (boiling at Munich, he notes) is 150 F°. A Fahrenheit degree is 5/9 of a kelvin, so ΔT = 150 × 5/9 = 83.3 K.
Heat. Q = mcΔT = 8.51 × 4,186 × 83.3 ≈ 2.97 × 10⁶ J.
Time. 2.5 hours = 9,000 s.
Power. P = Q/t ≈ 2.97 × 10⁶ / 9,000 ≈ 330 W into the water alone.
Rumford also allowed for the heat taken up by the metal parts. He reckoned the total equalled heating 26.58 lb of ice-cold water to boiling, 180 F° or 100 K. That is 12.06 kg × 4,186 × 100 ≈ 5.05 × 10⁶ J, or about 560 W. He said he had not counted the wooden box or the heat lost to the room, so the true figure was higher still.
Common mistakes:
Using the Fahrenheit rise as kelvin. 150 F° is 83.3 K, not 150 K.
Forgetting the container. Rumford's water got less than two-thirds of the heat he could account for.
Treating a lower bound as exact. Any heat lost to the air makes the real power larger than your answer.
Rumford measured heat but not the work the horses did, so he could not give an exchange rate between the two. That step came decades later; see our post on James Joule and the mechanical equivalent of heat.
What can a student learn from Rumford's way of working?
Notice what does not fit. He saw something ordinary, hot cannon, and asked a question others did not.
Test each rival explanation. He checked the chips' capacity for heat before rejecting the idea that they released it.
Make the effect big and public. Boiling a large amount of water in front of witnesses was hard to argue with.
Separate what you measured from what you guess. He wrote that he was far from claiming to know how the motion produced heat, only that it did.
If you want to see heat as particle motion rather than read about it, the Fermi Simulation Lab has interactive simulations for ideas easier to see than read.
Your next step: redo the power calculation using Rumford's reading at 2 hours (178 °F) and compare the heating rate in the first and second hours.
