One of the oldest answers to this question is not a gemstone at all. A Spanish book of 1604 gives Virgo a magnet, picked for something it visibly does. A shop today gives moss agate, and nothing on a modern Virgo shelf is chosen that way.
Lodestone comes out of the ground already magnetised, and specimens often reach a buyer with small iron particles clinging to them. Minerals.net describes it as the only mineral that is a natural magnet. Whatever else it is, it is a piece of rock that demonstrates something, and it will do so on a kitchen table for anyone who doubts it.
That is the Virgo entry in a treatise printed at Valladolid in 1604, reproduced three centuries later by George Frederick Kunz with a proper citation and described by him as probably representing Arab tradition. Ten other signs in that list get gems. Pisces gets a blank line. Virgo gets a working object.
What a magnet was doing on a list of stones
A modern crystal list is understood as a set of symbolic or aesthetic choices, stones that suit a temperament or a colour. Run that reading back over the 1604 entry and it does not fit. A magnet is not a mood. It is closer to a piece of equipment, and a compiler who reaches for one is working somewhere that a stone was expected to have a property a person could show you.
The same material turns up in the other surviving system, doing the same kind of work. Agrippa’s 1533 chapter on what falls under Saturn ends its list of stones with the loadstone, among the dark, weighty, earthy things, and the Capricorn page takes that clause apart at length. A magnet was therefore in circulation as a stone in both of the older sources, seventy years apart, landing on Virgo in one of them and on Saturn in the other. Neither writer treats it as an oddity worth explaining.
Magnetite has no romance about it in the reference books. Iron oxide, cubic, Mohs 5.5 to 6.5 in the Handbook of Mineralogy, and strongly magnetic. It is common, and about as undecorative as a mineral gets.
Whatever the 1604 compiler thought he was doing, he was not compiling a set of pretty things with meanings attached, because a lodestone is not a pretty thing and its meaning is not attached, it is inherent and it is demonstrable. That is a different category of object from anything a shop sells for Virgo now. The break in the record is not that the stones changed. It is that the kind of thing being listed changed, and no list of stones for the signs survived the change intact.
Emerald in 1533, carnelian in 1913
The other two old answers are gemstones, and they do not agree either. Agrippa’s table of twelve signs and twelve stones gives Virgo emerald. Kunz’s own zodiacal list of 1913 gives Virgo carnelian. Set the three side by side and no two of them overlap anywhere. Virgo is not unusual in that. No two of those three lists agree on a single sign anywhere in the twelve.
Kunz’s entry is the weakest of the three on its own terms. He printed that zodiacal list with no reference of any kind, in a chapter where nearly every other assertion carries one, and carnelian is not what a shop reaches for now. The contrast is inside one book. He gave the 1604 treatise a footnote naming Morales and Valladolid, then set down a list of his own of the same kind with nothing behind it at all. The compilation modern accounts trace back to therefore disagrees with both older sources and was itself quietly dropped.
The planet is the part that holds. Ptolemy assigned Mercury to Gemini and to Virgo in the second century, and that is still the standard assignment more than eighteen centuries later. He argued for it rather than asserting it, which is more than any of the stone lists manage.
Moss agate, and the branches in it
Moss agate is the modern answer, and it appears in none of the three old lists. What it has instead is a resemblance. Pale translucent chalcedony threaded with dark branching inclusions looks like growing things caught in ice, and a sign associated in practice with gardens and patient improvement finds that picture without needing a text to point at. Practitioners say as much when asked, and recommend it for patience and for anything a person wants to grow rather than fix. Those are claims about what the practice holds, not about the stone, and they are the more honest for being stated in the open. How a stone gets onto a sign’s list by that route, and by the other routes available, is worked through on the Taurus page.
The branches are not plants and never were. USGS describes Montana moss agate as grayish-white to gray translucent chalcedony containing dendrites, and states the rest without hedging: the black, brown or red tree-like or scenic dendritic growths are actually included minerals of manganese and iron. Chalcedony is the fine-grained form of quartz, so moss agate is quartz with a pattern in it, and it has no entry of its own in the Handbook of Mineralogy and no standing as a species. The pattern is the whole of what a buyer is choosing between one slab and the next, and it is also the one thing about the material that no mineral reference will describe for them.
The rest of the shelf
Sapphire is the one Virgo stone with an institution behind it, and the institution is a trade body rather than an astrologer. Sapphire is September’s birthstone and Virgo runs into September. The list that says so is the one Jewelers of America has maintained since a trade meeting in 1912. It is the blue variety of corundum and second only to diamond on the Mohs scale, which means it will outlast everything else on the shelf without being asked to.
Amazonite is a feldspar. The Smithsonian’s mineral sciences collection calls it the blue-green variety of microcline and attributes the colour to small quantities of lead and water in the feldspar, which is a better origin story than most trade names manage. Microcline is potassium aluminium silicate, triclinic, Mohs 6 to 6.5, with perfect cleavage in two directions that meet at about a right angle. It scuffs, and a tumble that shares a pocket with a set of keys will show it inside a week. Nothing in its history points at Virgo, and practice reaches for it on the Mercury association and on its colour.
Peridot arrives with August, the other month the sign covers, and it brings a naming problem the Libra page works through at length, since peridot is gem olivine and olivine is a series rather than a species.
Search the Crystalance Mineral Library for moss agate and nothing comes back, because the Library is built on species and a pattern is not one. Sapphire is in there under corundum and amazonite under microcline. Magnetite has a page under its own name. A mineralogist and a shopkeeper would spell it the same way, which is not true of anything else on this page. What the compiler in 1604 thought a magnet would do for somebody born in September is not written down anywhere either of them could check.
Sources
- George Frederick Kunz, The Curious Lore of Precious Stones, 1913, chs. IX and X, for the zodiacal list and the 1604 Valladolid list: gutenberg.org
- Minerals.net, magnetite and lodestone: minerals.net
- Cornelius Agrippa, Three Books of Occult Philosophy, Book I ch. XXV on what falls under Saturn, in Peterson’s edition of the 1651 English translation of the 1533 original: esotericarchives.com
- Cornelius Agrippa, Three Books of Occult Philosophy, Book II ch. XIV, the Scale of the Number Twelve: esotericarchives.com
- Handbook of Mineralogy, magnetite: handbookofmineralogy.org
- Ptolemy, Tetrabiblos I.17, Robbins translation, on the houses of the planets: penelope.uchicago.edu
- US Geological Survey, Gemstones of the United States, on chalcedony and moss agate dendrites: apps.usgs.gov
- Jewelers of America, birthstones by month and the 1912 origin of the list: jewelers.org
- Smithsonian National Museum of Natural History, amazonite: naturalhistory.si.edu
- Handbook of Mineralogy, microcline: handbookofmineralogy.org
- Handbook of Mineralogy, corundum: handbookofmineralogy.org
