DD12: Newton in the Dock
In August 1710 a jury of goldsmiths tested the gold coins of the Royal Mint at the Trial of the Pyx and judged them below standard. The Master of the Mint, the man answerable for every one of them, was Isaac Newton. He did not accept it. He tested the test, and found that the coins were sound and the standard was wrong. What he uncovered on the way is stranger than the quarrel: in 1710, nobody could say exactly what pure gold was.
The verdict
Newton was Master of the Mint from Christmas Day 1699 until his death in 1727. The Royal Mint Museum says he claimed "with justice" to have brought the coinage to a "much greater degree of exactness than ever was known before". Hence its dry line on what followed: "it is no wonder that he reacted angrily in 1710 to the mistaken judgement of the jury at the Trial of the Pyx that his gold coins were below standard."
The trial, described in The Jury That Tries the Money, compares coins against a trial plate, a piece of metal made to the legal standard. Newton's report to the Treasury, preserved in the World Gold Council's documentary history of the gold standard, dates the trial to 21 August 1710 and begins without ceremony: "Finding reason to suspect that the present indented trial piece is too fine; I have nicely examined it".
What fine meant
To see how a standard can be too good, you need Newton's own account of the assay. It survives in his own hand, transcribed by the Newton Project at the University of Oxford. The assayer refined a small sample and reported the weight it lost. If it lost nothing, the gold was 24 carats fine, "and this is fine Gold and fine Silver in the Sense of the Law". Fine gold, in other words, was not a fact of nature. It was whatever survived the customary assay, made "with two waters and noe more", two baths of acid to dissolve away the silver.
There lay the difficulty. A third bath, Newton noted in a draft memorandum, "will leave it still finer, & a fourth still finer". Refiners "seldome make it above 23 car : 3 gr : 3 qters : fine", a quarter of a grain short of 24 carats, yet goldsmiths "are generally of opinion that Gold cannot be above 24 carats fine not knowing that there are ways of making it finer then by the assay". His report states the consequence in seven words: "there are various degrees of fine gold". Make a trial plate from finer gold, and the plate comes out finer. So "the standard of gold is rendred uncertain".
A plate that was too good
That is what had happened. The plate was new, introduced in 1707, and Newton records how it was made. The jury "did not then refine the Gold & Silver but bought them" from his own deputy. "They know not how he refined them, and wondred that he could make them so fine: But being generally of opinion that no gold or silver could be too fine, they made the trial pieces thereof".
No gold could be too fine. It sounds like common sense, and it is the error. Honest 22 carat coin, laid against a plate made from unusually pure gold, looks short. Newton put numbers on it: "the Gold money was Standard full by the Assay and the Trial-piece a quarter of a grain better than the money", so the jury "represented the money a quarter of a grain worse than Standard". By his assays the 1707 plate was "five twelfths of a grain better than Standard", the 1688 plate "a sixth part of A grain better", and the plate of 1660 exact.
On the assayer's scale he was using, 24 carats of four grains each, a quarter of a grain is one part in 384. The rest is our arithmetic, not his: a shade under one gram of gold in every troy pound of metal, about 98 pounds at the 100.58 pounds a gram our markets feed returned at 07:05 UTC on 3 October 2026. Small, and enough to brand as deficient a coinage its maker thought the most exact ever struck.
What Newton asked for
The Royal Mint's account is that the fault "lay not in the coinage but in the new Trial Plate", which "contained more gold than required", and that "the older 1688 Trial Plate was reinstated". The Science Museum tells it the same way. Two honest notes. Every account we could find is the Mint's or Newton's own; we hold no record of the jury's reply. And the plate that came back was, by Newton's own figures, still a sixth of a grain too fine.
His remedy is the interesting part. He asked, in a set of queries, "whether there should be one common standard degree of fineness of fine gold" for all Great Britain, and whether a piece of the very gold used should "be kept in the Exchequer" to check every future plate. He saw the danger plainly: otherwise "the Goldsmiths will have it in their power to alter the Standard without Controul as often as they are to make a New Tryal-piece".
The mirror of Old Coppernose
Set this beside Henry VIII. In the 1540s the trial plates were remade below sterling and, on our reading, debased coin could match them. In 1707 a plate was made above the standard, and honest coin could not. Opposite errors, one lesson. Both times the jury did its job. A comparison is only ever as good as the thing compared against, and that thing is made by people, from the metal and the beliefs they have to hand.
That is why Newton wanted the standard to be an object, fixed once, and kept by somebody other than the people it judged. It is the idea behind a hallmark, an independent assay struck into the metal itself. And the standard he defended has outlived the argument: "eleven ounces of such fine Gold" to "one ounce of allay", 22 carats, is still the gold of the sovereign. It is the standard behind the guinea, and behind The Gold Standard and Its Fall. Newton knew better than anyone that a measure is a decision, and that somebody has to keep it honest.
Every piece we sell is hallmarked, or its exemption is evidenced, and the hallmark is photographed on the listing. You can read the marks yourself with our free identifier, see what is on the shelf, or sell with Chapman Gold. Nothing here is financial advice.