Tsavorite

Most buyers arrive at tsavorite after emerald has disappointed them. They have looked at oiled goods and read a clarity comment they did not expect, and what they are shown next is a garnet. Tsavorite is grossular — calcium aluminium garnet — coloured by vanadium, chromium, or both, and almost all of it comes from one Neoproterozoic metamorphic belt that runs down East Africa and on into Madagascar. It is a young stone in this trade, and it behaves unlike the corundum and beryl beside it.

Pure grossular is colourless. Trace vanadium and chromium in the structure produce the green; iron, where it is present, drags the colour toward olive and takes the life out of it. The stone is singly refractive, index near 1.74, and has no cleavage. Strain shows up as anomalous double refraction between crossed polars, which alarms the people who look for it and means nothing.

Campbell Bridges found green grossular in north-eastern Tanzania in the late 1960s, and again a few years later in southern Kenya, near Tsavo. Henry Platt at Tiffany named it tsavorite after the park, and the house promoted it. Bridges was killed in Kenya in a dispute over mining ground, and Kenyan production has been erratic since, though it has not stopped and the family ground is still worked. Merelani in northern Tanzania now supplies at least as much, out of the same graphitic gneiss that hosts tanzanite. Both fields sit in one old metamorphic belt, and the line between them was drawn by governments.

Nothing is done to it

There is no commercial heat treatment for tsavorite, and no diffusion. Filling has been reported on badly broken goods, but it is not routine and it is not what you meet in the better material. Nothing in the standard repertoire improves the colour, so nothing is applied to it. That counts in the stone’s favour, though not for the reason the trade usually gives. With corundum you pay a premium for an absence, and that premium for the untreated stone is now one of the largest variables in the price. With tsavorite, untreated is the condition of the material rather than a choice anyone made.

The corollary is unforgiving: a mediocre tsavorite cannot be rescued, and you buy exactly what the pocket gave up.

Why size is the whole story

Tsavorite grows in isolated nodules and pockets in the gneiss, not in long clean channels. The host rock has been folded and stretched, and the crystals carry that history as fractures. Mining is shaft work, then hand work along a reef. It does not scale. The rough arrives small, and cutters lose much of it working around the breaks. Graphite from the host turns up as inclusions in Merelani goods.

A clean stone of a couple of carats is ordinary and five carats is a serious stone. Above ten carats there is no market to speak of, only individual objects. Tsavorite is the clearest case I know of why price per carat climbs in steps rather than a slope, and the steps are steeper here than almost anywhere, because the top of the range barely exists.

Against emerald

Optically the comparison flatters the garnet. An index near 1.74, against roughly 1.58 for emerald, returns more light to the eye, and dispersion is around double emerald’s, though saturation buries most of it. There is no cleavage and no oil, so it takes a brilliant or a mixed cut and holds up in a ring. At 7 to 7.5 it is the softer of the two and will abrade in daily wear, but it does not need the protection a step cut gives emerald.

The green is not the same green. Emerald at its best is cooler and slightly blue, and part of what collectors pay for is that softness. Tsavorite is more direct, warmer, and at full saturation it can run dark under a weak light. Anyone offering one as a substitute for the other is selling something. The two greens do different jobs, and emerald reaches sizes tsavorite does not.

Undervalued, with conditions

I think tsavorite is underpriced, for reasons commercial rather than gemmological. The trade pays for narrative, and this one is about fifty years old. There is no colour appellation and no auction history of consequence. Origin premium is effectively nil: separating Kenyan from Tanzanian material is contested, no laboratory offers it as routine, and no buyer is paying for the opinion. All of that describes the trade’s habits rather than the stone.

The other side of it is worth stating. The buyer pool is thin, and the largest sizes are illiquid because they are rare. A stone that is hard to replace is hard to sell in a hurry, which is the standing problem with treating coloured stones as an asset, and tsavorite does not escape it.

My rule here is the same as anywhere. Buy the stone for what it does in daylight, at a price that still looks sensible on the day the trade stops calling it undervalued.