On the morning of July 25, 1610, Galileo pointed his telescope at Saturn and was surprised to find that it appeared to be flanked by two round blobs or bumps, one on either side. Unfortunately, Galileo’s telescope wasn’t quite advanced enough to pick out precisely what he had seen (his observations are now credited with being the earliest description of Saturn’s rings in astronomical history), but he nevertheless presumed that whatever he had seen was something special. And he wanted people to know about it.

Keen to announce his news and thereby secure credit for whatever it was he had discovered, Galileo sent letters to his friends and fellow astronomers. This being Galileo, the announcement was far from straightforward:

SMAISMRMILMEPOETALEUMIBUNENUGTTAUIRAS

Each message that Galileo sent out contained little more than that jumbled string of letters, which when rearranged correctly spelled out the Latin sentence, “altissimum planetam tergeminum observavi”—or “I have observed that the highest planet is threefold.”

As the outermost planet known to science at the time, Saturn was the “highest planet” in question. And unaware that he had discovered its rings, Galileo was merely suggesting to his contemporaries that he had found that the planet was somehow divided into three parts. Announcing such a discovery in the form of an anagram might have bought Galileo some time to continue his observations, however, but there was a problem: Anagrams can easily be misinterpreted.

One of those to whom Galileo sent a letter was the German scientist Johannes Kepler. A keen astronomer himself, Kepler had followed and supported Galileo’s work for several years, so when the coded letter arrived at his home in Prague he quickly set to work solving it. Unfortunately for him, he got it completely wrong.

Kepler rearranged Galileo’s word jumble as “salve, umbistineum geminatum Martia proles,” which he interpreted as “be greeted, double-knob, children of Mars.” His solution was far from perfect (umbistineum isn’t really a grammatical Latin word, for one thing), but Kepler was nevertheless convinced that, not only had he correctly solved the riddle, but Galileo’s apparent discovery proved a theory he had been contemplating for several months.

Earlier in 1610, Galileo had discovered the four so-called “Galilean moons” of Jupiter: Io, Europa, Ganymede and Callisto. Although we now know that Jupiter has several dozen moons of varying shapes, sizes, and orbits, at the time the announcement of just four natural satellites had led Kepler to presume that there must be a natural progression in the heavens: the Earth has one moon; Jupiter, two places further out from the Earth, has four; and sat between the two is Mars, which Kepler theorized must surely have two moons, to maintain the balanced celestial sequence 1, 2, 4 and so on (his only question was whether Saturn had six or eight).

Kepler got the anagram wrong, and the presumption that Jupiter only had four moons had been wrong. Yet as misguided as both these facts were, the assumption that Kepler made based on both of them—namely, that Mars had two moons—was entirely correct. Unfortunately for Kepler, his theory would not be proved until long after his death, as the two Martian moons Phobos and Deimos (named after Ares’s sons in Greek Mythology) were not discovered until 1877, by the American astronomer Asaph Hall.

Nevertheless, a misinterpretation of the anagram had accidentally predicted a major astronomical discovery of the 19th century, nearly 300 years before it occurred.

So remarkable was Kepler’s misguided—yet ultimately true—misinterpretation of Galileo’s puzzle that other stories and tall tales regarding his misinterpretations of Galileo’s anagrams soon began to emerge. According to the legend, Galileo sent a second coded message to Kepler later in 1610 when he discovered that Venus, like the Moon, has phases and therefore does not emit light itself but merely appears to shine due to reflected sunlight as it orbits nearer to the Sun than the Earth. The message this time read:

Haec immatura a me iam frustra leguntur oy

… or, “I am now bringing these immature things together in vain, oy!”

The correct solution was “Cynthiae figuras aemulatur mater amorum,” meaning “The Mother of Love (an epithet for the planet Venus) copies the forms of Cynthia (an epithet for the Moon).” Kepler however misinterpreted the message this time as “macula rufa in Jove est gyratur mathem,” or, “there is a red spot in Jupiter, which rotates mathematically.” Kepler had done it again. Entirely by mistake, he predicted the discovery of Jupiter’s Great Red Spot more than two centuries before it was officially discovered.

Unlike the earlier account of Galileo and Kepler’s word play, however, this second story might be apocryphal. Galileo certainly sent out this second anagram to Kepler later in 1610, but the idea that he (albeit unintentionally) managed to make a second groundbreaking astronomical prediction entirely by accident appears to be a step too far. One accidental discovery, it seems, is more than enough.