Types of Pearls Explained

The Four Cultured Types, Compared

The four types differ by mollusc, by where they are farmed, and by the size and colour that mollusc can produce. Everything else on a ticket is grading language. Here is the spine of the subject in one place, using GIA's cultured pearl FAQ for the size figures and GIA's own wording for colour.

Type	Mollusc	Farmed in	GIA size range	GIA colour range
Akoya	Pinctada fucata, also written martensii	Japan, China	Typically less than 9.00 mm	White or cream, pink overtone, high luster
Tahitian	Pinctada margaritifera	French Polynesia	8.00 mm to 14.00 mm	Most commonly greenish gray; also gray, black and brown, with blue, green, purple or pink overtones
South Sea	Pinctada maxima, silver-lipped and gold-lipped	Australia, Indonesia, the Philippines, Myanmar	8.00 mm to 18.00 mm	White or varying shades of yellow, called golden in the trade
Freshwater	Hyriopsis cumingii, also Cristaria plicata and Anodonta woodiana	China; Deqing County, Zhejiang is the historic centre	None published - GIA says only smaller sizes, near-round or oval	Wide range of pastel colours, chiefly pink and orange

One caution about those size figures, and it is the kind of thing this page exists for: GIA publishes two sets and they do not agree. The cultured pearl FAQ, quoted in the table, gives akoya as "typically less than 9.00 mm", Tahitian "typically range in size from 8.00 mm to 14.00 mm" and South Sea "typically range in size from 8.00 mm to 18.00 mm". GIA's 4Cs "Meet the Mollusks" page gives akoya "generally between 2 and 11 mm" with "the average being 6 to 8 mm", Tahitian "generally between 9 to 14 mm", and South Sea "from 8 to 20 mm" with "the average is 13 mm". Those are not roundings of one another. We quote the FAQ as the primary set because it is the more conservative, but a seller quoting either is quoting GIA, and neither is a specification.

The freshwater blank is itself a finding: GIA gives no millimetre range on either page, so every neat freshwater range you will read elsewhere is a trade number. Note too that GIA sorts shape into seven categories - round, near-round, button, drop, oval, semi-baroque and baroque - and that the hierarchy putting freshwater at the entry tier and South Sea at the premium end is a market convention, not a graded standard. No institution publishes a price ranking of the four types.

Which Type Should You Buy

The question behind this page is usually a purchase, so here is the short version - flagged for what it is, a reading of the market rather than a graded ranking, because no institution publishes one.

If you want the thing most people picture when they hear pearl necklace, buy akoya: round, white to cream with a pink overtone, and the high luster GIA describes for the type. It is also the type where processing is most routine, so ask about bleaching and maeshori rather than about grade.

If budget is the binding constraint, freshwater does more for the money than anything else here,. What you give up is predictability: with no institutional size range and no grading standard, two strands sold as the same grade can look very different.

If you want colour rather than white, Tahitian is the only one of the four with a naturally dark body colour, and GIA's description - most commonly greenish gray, with blue, green, purple and pink overtones - is a better guide than the word black. Ask whether the colour is natural or dyed, in writing.

If size is the point, South Sea runs the largest range GIA publishes and produces golden colour naturally; this is the tier where a laboratory report is worth paying for. Across all four, what separates a good strand from a bad one at any price is luster, which GIA says might be the most important of its seven value factors. That you can see for yourself. The letter on the card you cannot.

Four Checks You Can Run Before You Pay

Loupe the drill hole. This is where coatings chip and flake, and where the edge of a bead nucleus becomes visible under thin nacre. GIA's own laboratory note on shell-pearl imitations describes exactly that: coating chipped near the drill holes, revealing white shell beneath.

Look at the surface under magnification. Nacre shows overlapping platelets; imitations show, in GIA's description of one examined sample, "numerous minute particles with a glittering effect... as well as a lack of the obvious overlapping nacre platelets". GIA also gives the one unaided check worth having: "If the nucleus is visible under the nacre, or if the pearl has a dull, chalky appearance, you can assume that the nacre is thin."

Ask which laboratory issued any certificate. Hanadama is issued by particular Japanese laboratories to their own criteria, and those criteria differ between them, so the name of the lab is the information - not the word on the card.

Ask what the seller's own letter grade means. AAA has no governing body, so the answer tells you whether you are dealing with someone describing their own range honestly or repeating a marketing register. And be suspicious of a strand whose pearls look identical: genuine uniformity is expensive and rare.

Natural, Cultured, Imitation - and Where Cultured Pearls Came From

In the United States the three words are defined at 16 CFR 23.19. A natural pearl is "a calcareous concretion consisting essentially of alternating concentric layers of carbonate of lime and organic material", formed without human intervention. A cultured pearl is the "composite product" created when a human-planted nucleus is coated with nacre by the mollusc. An imitation simulates a pearl in appearance and is manufactured.

The labelling rules sit in the next section, 16 CFR 23.20, and the distinction matters if you go to the text, because most articles cite only 23.19. Under 23.20 the word pearl may not be used unqualified unless the pearl is natural; cultured or cultivated must immediately precede the word pearl rather than float elsewhere in the description; and imitations must carry artificial, imitation or simulated immediately preceding. So a listing that says simply pearl, with no qualifier near the noun, is either mislabelled or claiming something extraordinary. Almost always it is the former.

CIBJO, whose Pearl Book is the international nomenclature reference rather than a grading system, defines natural pearls as "formations secreted, without human intervention, in the interior of molluscs and within naturally formed pearl sacs". The two bodies disagree on one definition worth flagging rather than resolving: the FTC puts a seed pearl at approximately two millimetres or less, CIBJO at "generally less than two but no more than three millimetres in diameter". Natural pearls are now effectively an auction category - our characterisation, but GIA's history: "Within a hundred years, these natural pearl sources had declined due to overfishing, pearl culturing, plastic buttons, and oil drilling."

One historical correction belongs here. The story everyone tells is that Kokichi Mikimoto invented the cultured pearl and patented it in 1896. The date is right and the object is wrong: his first success, in July 1893, produced hemispherical cultured pearls grown against the inside of the shell, and the 1896 patent - confirmed in the Japan Patent Office record - covered that hemispherical method. The round pearl descends from elsewhere: the first fully spherical cultured pearl was observed in 1905, and in 1907 the method the modern industry uses, a shell bead nucleus implanted with mantle tissue into the gonad, was patented as the Mise-Nishikawa method by Tatsuhei Mise and Tokichi Nishikawa. Mikimoto's own round-pearl patent came in 1916, drafted around theirs, and he later acquired rights to their method. Mikimoto built the market; the technique was somebody else's.

Grading: There Is No Universal Standard

GIA states the problem itself, and it is the most useful sentence in the subject: "Agreeing on a universal approach to classify pearls has remained one of the main challenges facing the pearl industry." There is no pearl equivalent of the diamond 4Cs with a single body behind it, and any page presenting one is presenting a seller's system as an institution's.

The AAA to A scale is the clearest example. No institutional source defines it; GIA neither uses nor references it. Retailers admit it - one puts it as plainly as anyone could, that "There is no single, worldwide standard for pearl grading. Each jeweler uses their own scale, which means the grades are only meaningful within that particular store or brand." The trade also runs a separate A to AAAA scale for freshwater, Tahitian and South Sea pearls, which is itself evidence that nothing is standardised, and no source we examined names an originator or a date for either. AAA is a marketing register, not a grade.

GIA's own approach is a set of descriptors rather than a verdict. The seven pearl value factors are size, shape, colour, luster, surface, nacre and matching, and GIA says luster might be the most important of them - the hedge is GIA's own and worth preserving. The luster scale is unusually concrete: Excellent, "reflections appear bright and sharp"; Very Good, "bright and near sharp"; Good, "bright but not sharp, and slightly hazy around the edges"; Fair, "weak and blurred"; Poor, "dim and diffused". And here is the contrast with AAA: GIA does not issue a single overall letter or number grade at all. It describes each factor and stops, so a seller quoting you one grade is not summarising GIA's opinion.

Hanadama sits in a third category. It is a certification issued principally by the Pearl Science Laboratory in Tokyo, as "Aurora Hanadama", to that laboratory's own criteria; the New Japan Pearl Laboratory also issues hanadama certificates under different criteria, with a nacre threshold nearer 0.25 mm against the PSL-aligned figure of roughly 0.8 mm in total. We could find no trademark or statutory protection for the term itself. That does not make hanadama meaningless; it makes it lab-specific.

So the tiers do different jobs. Regulation first: the FTC governs what may be said about origin and treatment. Nomenclature second: CIBJO sets the disclosure duty, requiring that "Full disclosure by the vendor to the purchaser of all material information shall take place whether or not the information is specifically requested." Value factors third, laboratory certifications fourth, seller letter grades last. One statutory system does exist - French Polynesia classifies Tahitian pearls by law - and a trade summary of the statute gives the grades as Top Gem, A, A/B, B+, B, B/C, C+, C, C/D and D. We could not obtain the statutory text, and given the territory's 2017 reform - a separate measure from the nacre repeal described below - we cannot confirm what force the scale currently has.

Keshi, Baroque, Mabe and Seed: Terms Sold Loosely

Keshi is the term most often misused, and both authorities are unambiguous. GIA describes them as "non-bead cultured pearls" that are "byproducts of the bead culturing process"; CIBJO's 2024 definition matches, "a non-beaded cultured pearl formed accidentally or intentionally by human intervention". So keshi are solid nacre with no bead at the centre, which is genuinely desirable - and they are a by-product of culturing, not natural pearls. Marketing that presents keshi as natural is inaccurate and, in the United States, potentially non-compliant, in the United States, with the FTC Jewelry Guides.

Mabe is where buyers most often assume they are getting more than they are. CIBJO's Pearl Book defines the assembled cultured blister at 5.14 as assemblages of purpose-grown cultured blisters cut from their shell, with the original bead upon which they grew removed, the cavity filled with various types of man-made materials, and the whole backed by a layer of shell. We are paraphrasing rather than quoting that sentence whole, because CIBJO writes it in the plural and it does not shorten cleanly inside quotation marks. The substance survives: a finished mabe is an assembly - dome, fill, shell backing - not a whole pearl. It also has a bonded interface, so it should be wiped rather than soaked, like the assembled opals in our guide to cleaning jewelry.

Baroque is the least informative of the four: CIBJO defines it as "an irregularly shaped natural or cultured pearl", a definition spanning both origins, so the word tells you about shape and nothing about human involvement. And seed pearl is where the regulator and the nomenclature body give different sizes.

Nacre, Treatments and How Imitations Are Caught

Nacre thickness is the quality factor with the most direct link to how long a pearl lasts. GIA notes that "Luster and nacre quality are closely related", and that its reports "will clearly note if the nacre coating is too thin and/or damaged" - though nacre measurements are optional on GIA reports rather than standard. The FTC treats it as a material fact: 16 CFR 23.22 makes it unfair or deceptive to misrepresent the thickness of the nacre coating.

One widely repeated fact about nacre needs correcting into the past tense, and with its missing qualifier restored. French Polynesia formerly imposed a minimum nacre thickness of 0.8 mm over 80 percent of the pearl's surface on exported Tahitian pearls; that requirement was repealed with effect around January 2017 and replaced by a farm and lagoon quota system. Many articles still state it as current, and most of those drop the surface qualifier too. We could not locate the original decree number, so treat the history as partially unverified - but do not buy on the assumption that a statutory nacre floor is protecting you today.

Treatments are governed by disclosure rather than prohibition, and CIBJO codes them: dyeing is code D and non-permanent, irradiation code R, bleaching code B and possibly non-permanent. Luster enhancement, known as maeshori, is "a multi-part chemical treatment, including exposure to heat and/or light, that temporarily enhances lustre" - temporarily being the word that matters. In the US, 16 CFR 23.24 requires disclosure where a treatment is not permanent, creates special care requirements, or significantly affects value. The gap between standard and market is the useful part: no institutional source designates any treatment as routine, yet the trade treats bleaching and light polishing in the akoya pipeline as processing rather than enhancement.

Which brings us to the test everybody has heard of. Rubbing a pearl against your teeth to feel for grit is folklore: no gemological institute lists it, and it fails in both directions, since imitations can be given gritty coatings and genuine pearls are routinely polished, waxed or lacquered smooth. We could not find an institution explicitly debunking it, so treat that verdict as our reasoning from the absence of endorsement rather than a citation. GIA's actual toolkit is X-ray radiography, "the most reliable non-destructive way of seeing the internal structure of pearls". Our roundups of pearl necklaces and pearl earrings apply these questions to live listings, and what a pearl necklace has symbolised covers the side of the subject that has nothing to do with grading.

Frequently Asked Questions

What are the four main types of pearls?

Akoya, Tahitian, South Sea and freshwater. Akoya are cultured in Pinctada fucata in Japan and China, typically less than 9.00 mm. Tahitian come from Pinctada margaritifera in French Polynesia, typically 8.00 to 14.00 mm. South Sea come from Pinctada maxima in its silver-lipped and gold-lipped varieties, farmed in Australia, Indonesia, the Philippines and Myanmar, typically 8.00 to 18.00 mm. Freshwater are cultured mainly in Hyriopsis cumingii in China. Note that GIA publishes a second, different set of ranges on its 4Cs mollusc page, and gives no millimetre range for freshwater on either.

Which type of pearl should I buy?

Akoya for the classic round white strand and the high luster GIA describes for the type. Freshwater if budget is the constraint, accepting less predictability between strands because there is no grading standard behind the letters. Tahitian if you want a naturally dark body colour, in which case ask whether the colour is natural or dyed. South Sea if size is the point. At every tier the property worth judging is luster, which GIA says might be the most important of its seven value factors.

What does AAA mean in pearl grading?

Less than it appears to. AAA is a trade convention with no governing body: no institutional source defines it, GIA neither uses nor references it, and retailers themselves acknowledge that their grades are only meaningful within their own store or brand. The trade also runs a separate A to AAAA scale for freshwater, Tahitian and South Sea pearls, which is itself evidence that nothing is standardised. Treat a letter grade as a position within one seller's own range, and ask about GIA's value factors instead.

Are keshi pearls natural pearls?

No. GIA describes keshi as non-bead cultured pearls that are by-products of the bead culturing process, and CIBJO defines a keshi as a non-beaded cultured pearl formed accidentally or intentionally by human intervention. What is true, and genuinely appealing, is that a keshi is solid nacre with no bead at its centre. What is not true is that it is natural: it exists because a farm implanted a nucleus in a mollusc. Marketing that calls keshi natural pearls is potentially non-compliant, in the United States, with the FTC Jewelry Guides.

Does the tooth test tell you if a pearl is real?

No, and no gemological institute lists it as a method. It fails in both directions: imitation pearls can be finished with deliberately gritty coatings, and genuine cultured pearls are routinely polished, waxed or lacquered to a smooth surface. GIA's actual identification tool is X-ray radiography, which shows a natural pearl's concentric ring structure, a bead-cultured pearl's solid core, or the irregular growth patterns and voids of a non-bead cultured pearl. At consumer level, a loupe on the drill hole is far more informative than your teeth.

What is a mabe pearl?

Not a whole pearl. CIBJO defines the finished article as an assembled cultured blister: a purpose-grown cultured blister cut from its shell, with the original bead it grew on removed, the cavity filled with man-made materials, and the whole thing backed by a layer of shell. So a mabe is a dome of nacre plus a fill plus a shell backing - an assembly. It also has a bonded interface, so it should be wiped rather than soaked.

How We Research

Our guides draw on manufacturer documentation, established industry practice, and what we learn researching products and listings for our buying guides. Where practice varies or opinion divides, we say so rather than pretending at consensus.

Questions worth asking a pearl seller: Four of them, in order of how much they reveal. Which species and which farm origin - a seller who cannot answer is reselling blind. Which treatments, given that CIBJO requires full disclosure of all material information whether or not the buyer asks for it. Which laboratory issued any certificate, since hanadama criteria differ between the labs that issue them and we could find no protection for the term itself. And what the seller means by their own letter grade, because AAA has no governing body and describes a position within one catalogue rather than a measured quality. A seller who answers all four straightforwardly is worth more than any grade on a card.