Antsongombato gem mine, Antsentsindrano - a premier Malagasy pegmatite locality known for yellow londonite–rhodizite crystals, red tourmaline, and rare borates.
Key facts
Antsongombato is one of the great small-pegmatite localities of modern specimen mineralogy: a remote cluster of highly evolved LCT pegmatite dikes in the central highlands of Madagascar, best known for yellow londonite-rhodizite crystals and red tourmaline in a sparse, boron- and cesium-rich assemblage. The mine lies in the Antsentsindrano area of the Betafo District, on the north side of Manentoakoho Mountain, where narrow pegmatites intrude marbles of the Itremo Group. For collectors, its significance rests on a remarkable combination of scale and rarity. In most pegmatites, rhodizite-londonite is a tiny accessory curiosity; at Antsongombato, equant modified dodecahedra commonly reached thumbnail size, and exceptional crystals measured several centimeters across, with sulfur-yellow to greenish-yellow color, glassy luster, and, in the best examples, gemmy interiors.
The locality’s fame was sharpened by the late-1990s mining that brought out the classic specimens: yellow londonite-rhodizite crystals perched in quartz-rich pegmatite with cranberry-red liddicoatite or elbaite-liddicoatite tourmaline, locally accompanied by blue to manganese-bearing apatite, feldspar, pollucite, spodumene, and rare borates. It is also the type locality for schiavinatoite, the niobium analogue of béhierite, a dense and very rare Nb-Ta borate described from Antsongombato material. The best Antsongombato pieces have an unmistakable look: waxy-to-vitreous yellow isometric crystals, often partly embedded or “frozen” in pale quartz and feldspar, enlivened by ruby-red tourmaline blades or broken prism sections. Specimens with sharp, exposed, undamaged londonite-rhodizite crystals on attractive contrasting matrix are the locality’s blue-chip material.
Regional View
Country View
Photo: Wikimedia Commons
Antsongombato specimens are most rewarding when read as pegmatite textures rather than as single-mineral trophies. The londonite-rhodizite is usually the headline, but the matrix tells the story: quartz and feldspar from a thin evolved dike, red tourmaline marking the enriched zones, and accessory borates recording an unusually boron-rich melt. Miarolitic cavities were present but not abundant, so many crystals are not freestanding pocket specimens in the Brazilian sense; they are often locked into massive pegmatite, exposed by careful hand work, trimming, and preparation. That is why pieces with a fully visible yellow crystal, unbruised edges, and a clean association with red tourmaline command such attention.
Search for specimens: View all specimens from Antsongombato gem mine, Antsentsindrano, null
The Antsongombato gem mine is not a single large industrial excavation but a small pegmatite-mining area in central Madagascar, near Antsentsindrano in the Anosiarivo Manapa commune of Betafo District, Vakinankaratra. The site is reported about 4 km east-southeast of Antsentsindrano and on the north flank of Manentoakoho Mountain, at roughly 1,500–1,600 m elevation. Access has historically been difficult: accounts of the 1998–2001 mining describe a rough unpaved route from Antsirabe requiring four-wheel drive, with travel interrupted in the rainy season.
Geologically, Antsongombato belongs to the famous Betafo-Antsirabe pegmatite province of Madagascar’s central highlands. The host rocks are part of the Itremo Group, where marbles, quartzites, schists, and gneisses have been intruded by evolved granitic pegmatites. The Antsongombato mining area exposes white-to-gray dolomitic marble and a series of pegmatite dikes over about a kilometer. Individual pegmatites may be hundreds of meters long, generally thin, and locally thicker, but the important rhodizite-londonite production came from one steeply dipping, highly evolved dike about 400 m long and only about 10–60 cm thick. This geometry is central to the collecting history: a narrow dike could produce world-class crystals, but only in limited volumes.
The deposit is an LCT-type pegmatite system of a danburite-bearing subtype, enriched in lithium, cesium, tantalum-niobium, boron, beryllium, and manganese. The essential rock-forming minerals are quartz, albite, K-feldspar or microcline, and spodumene, with tourmaline, danburite, pollucite, beryl, apatite, microlite-group minerals, columbite-tantalite minerals, monazite, xenotime, hübnerite, béhierite, schiavinatoite, and londonite-rhodizite among the notable accessories. Mica is strikingly scarce compared with many LCT pegmatites, a feature repeatedly noted in discussions of the londonite-rhodizite-bearing dikes of central Madagascar.
Mineralization was not evenly distributed through the dikes. The best yellow londonite-rhodizite crystals formed in chemically evolved portions marked by green and red tourmaline. Miarolitic cavities occurred but were scarce and generally small; some yielded well-formed crystals with red tourmaline, but many of the better crystals were enclosed in the central parts of the pegmatite rather than standing freely in open pockets. Mining revealed coarse-grained mineralized “columns” along the dip of the dike, and some of the best gem-quality crystals were found in masses of fine-grained pollucite and spodumene more than 10 cm across.
The locality was known to French colonial miners by the early 20th century as part of a series of red tourmaline workings in the region. These early operations focused on gem-quality red tourmaline, much of it from massive quartz cores and less commonly from cavities. The workings and the nearby village were reportedly abandoned by the early 1930s. Rhodizite from Antsongombato was noted in the mid-20th-century mineralogical literature, and in 1964–1965 a French colonist organized renewed mining for rhodizite mineral specimens. A particularly famous pale-yellow vitreous crystal about 7 cm across entered the collection of the Museum of the School of Mines in Paris. The locality was also worked briefly for red tourmaline in 1978 before flooding halted activity.
The modern specimen era began in October 1998, when Federico Pezzotta’s investigations recognized a new system of pegmatitic dikes locally mineralized with red tourmaline and yellow cesium-rich rhodizite, later identified in part as londonite. The Malagasy company Pyramide of Antananarivo organized mining of the Antsongombato Gem mine. From late 1998 through 2000, miners worked by hand, using pry bars along natural fractures, removing marble first, and then carefully excavating the pegmatite without blasting. About 40 miners were reportedly organized in groups of eight. This careful method was crucial: Antsongombato crystals may be hard, but sharp edges and matrix attachments are easily compromised by rough extraction.
The principal production of fine specimens and gem rough occurred from February 1999 to September 2000. Specimens debuted at the 2000 Tucson shows, where the high quality and the rumored new-mineral status of the yellow crystals generated immediate excitement. By January 2001, nine mineralized areas had been worked to depths of roughly 35–40 m. Some shoots ended; others were made hazardous by flooding, collapsing pits, and increasingly hard unweathered rock. Pyramide ceased activity in the area in January 2001. Later reports indicate that small-scale local work continued intermittently in the broader Antsentsindrano-Andrembesoa area, but the main gem-producing pegmatite was considered essentially exhausted, and significant future production would require discovery of additional productive dikes.
At Antsongombato, “londonite” is best understood in the collector’s market as tested or presumed Cs-dominant material within the londonite-rhodizite series, because the two end members cannot be separated visually and require quantitative chemical analysis. The classic crystals are equant isometric modified dodecahedra, commonly around 1 cm but exceptionally 6–7 cm, with dominant 110 faces modified by 221, 211, and 111; a few pseudo-octahedral forms also occur. Colors range from near colorless through pale greenish yellow to rich yellow, and larger crystals may show stronger yellow toward the rim. The finest Antsongombato londonite-rhodizite pieces show a lustrous, sharp, transparent to translucent yellow crystal cleanly exposed on quartz-feldspar pegmatite with cranberry-red liddicoatite or elbaite-liddicoatite tourmaline, and sometimes blue apatite; ordinary examples are more included, broken, sun-faded, or partly buried in massive matrix. The highest-value pieces combine proof or strong provenance for the londonite composition, an undamaged crystal, saturated color, and an aesthetic tourmaline-bearing matrix from the late-1990s Pyramide production.
Antsongombato’s liddicoatite, broadly recorded in the literature and specimen market as liddicoatite or elbaite-liddicoatite series tourmaline, is the red-to-pink partner mineral that gives the locality’s best londonite-rhodizite specimens their color contrast. It occurs as red, cranberry-red, pigeon’s-blood red, pink, or locally bicolored tourmaline in quartz-rich pegmatite, in massive quartz cores, and in the evolved zones that also carry londonite-rhodizite. Freestanding, undamaged tourmaline crystals from Antsongombato are less often encountered than tourmaline sections or partly embedded crystals in londonite-rhodizite matrix specimens; documented examples include lustrous red crystals around 1.6 cm on pale marble or pegmatite matrix, and smaller cranberry-red crystals accompanying yellow londonite. The best liddicoatite pieces from here have high luster, translucency, rich red color, sharp terminations, and an undisturbed matrix context; lesser pieces tend to be fractured, incomplete, or present only as colorful embedded fragments beside the rarer borate crystals.
Other minerals documented from Antsongombato make the locality far richer than a two-species headline suggests. Schiavinatoite, (Nb,Ta)BO4, is the type-locality mineral and one of the locality’s major scientific credentials; it was described from a boron-rich pegmatite as Nb-dominant portions of a greyish-pink bipyramidal béhierite-group crystal and is associated with rhodizite, liddicoatite, spodumene, pollucite, danburite, and apatite. Béhierite itself is present, as are danburite, pollucite, spodumene, Cs-rich beryl, manganese-bearing fluorapatite, microlite-group minerals, columbite-(Mn), tantalite-(Mn), hübnerite, monazite-group minerals, xenotime-(Y), native bismuth, bismutite, smoky quartz, albite, microcline, and K-feldspar. For rarity specialists, the most intriguing Antsongombato specimens are not always the prettiest: matrix pieces with schiavinatoite or béhierite, londonite-rhodizite, and red tourmaline preserve a very unusual boron-cesium-niobium-tantalum pegmatite association in a hand-specimen scale.
The principal authenticity issue at Antsongombato is not widespread fabrication but naming precision. Londonite and rhodizite form a series, and color, habit, luster, size, or locality alone cannot prove the Cs-dominant londonite end member. Reliable labels should distinguish between analytically confirmed londonite and untested londonite-rhodizite or rhodizite-londonite series material. This is especially important for high-value crystals: a specimen may be visually identical whether it is rhodizite-dominant, londonite-dominant, or compositionally zoned between the two. Backscattered-electron imaging and quantitative chemical analysis have shown that individual crystals may be chemically heterogeneous, with patchy and rim zones of differing Cs and K content.
Color is another critical point. Near-colorless Antsongombato rhodizite-londonite has been experimentally irradiated by X-rays to greenish yellow or yellow, and both natural-color and irradiated material can fade in prolonged sunlight. In published fade testing, an irradiated yellow faceted stone became near colorless after three weeks in a south-facing window, while natural-color material faded more slowly but still lost noticeable color. Fine yellow crystals and cut stones should therefore be stored and displayed away from strong sunlight and high-intensity UV-rich lighting. Short-wave UV response, when present, is typically very weak to moderate yellowish green; long-wave response is inert to weak yellowish green, so fluorescence is a useful observation but not a stand-alone proof of species.
Condition is central to value. Antsongombato crystals are hard, but many are partly embedded in quartz, feldspar, pollucite, or spodumene and have suffered from extraction, trimming, or exposure. Edge nicks, bruised dodecahedral corners, rehealed-looking fractures, conchoidal chips, and partial crystals frozen in matrix are common. The most desirable specimens have a cleanly exposed, lustrous yellow londonite-rhodizite crystal with complete form, no visible edge damage, and attractive red tourmaline contrast. Aesthetic matrix matters unusually much here: because many crystals occur locked in massive pegmatite, a piece that shows the crystal well and still retains geological context is far preferable to an isolated battered thumbnail.
Market availability is sporadic. The main Pyramide-era production is long past, and the best pieces are now in private collections, museums, or old dealer stock. Small thumbnails and reference miniatures of londonite-rhodizite on matrix still appear from time to time, including modest specimens with incomplete crystals or embedded pink tourmaline; fine confirmed londonite with sharp form, rich yellow color, and red liddicoatite is much rarer. Type-locality schiavinatoite is a specialist rarity and should be expected only occasionally. For any significant purchase, insist on a detailed old label or provenance, a cautious species name unless analysis is available, and photographs that clearly show terminations, edges, and matrix contacts.
The Antsongombato story begins like many Madagascar pegmatite stories: with French colonial prospecting, red tourmaline, and then a long silence. By the early 20th century the site was known as one of a dozen tourmaline mines in the region. Miners searched the pegmatite cores for gem-quality red tourmaline, taking crystals from massive quartz and, more rarely, from small cavities. Then the locality faded from view. The village and workings were reportedly abandoned, probably in the early 1930s, and the site was obscure enough that it did not appear on A. Emberger’s 1954 geological map of the region.
The first great collector episode came in 1964–1965, when a French colonist organized mining not for tourmaline but for rhodizite specimens. That campaign produced one of Antsongombato’s legendary crystals: a pale-yellow, vitreous rhodizite-londonite about 7 cm across, later recorded in the collection of the Museum of the School of Mines in Paris. For a mineral usually thought of in millimeters, a crystal of that scale from a narrow Malagasy pegmatite was astonishing. It foreshadowed what the locality would become decades later, when analytical work showed that some of the yellow material belonged to the newly recognized cesium-rich end member, londonite.
The modern rush began in October 1998. Federico Pezzotta, already deeply involved in the study of central Madagascar pegmatites, investigated Antsongombato and recognized a new system of dikes carrying red tourmaline and yellow cesium-rich rhodizite. The Malagasy company Pyramide of Antananarivo quickly organized mining. The operation was small but deliberate: about 40 miners, arranged in groups of eight, dug pits into the mineralized zones. They used pry bars along natural cracks, removed the marble host rock first, and avoided explosives so the valuable specimens would not be shattered. It was specimen mining in the literal sense, each promising zone handled as if a single careless blow might ruin the mine’s best crystal.
The work was bounded by the mountain and by water. The productive dike was narrow, steep, and hard; the road was rough; the rainy season shut down work; and by January 2001 nine mineralized areas had been worked to depths of roughly 35–40 m. In some, the mineralization ended. In others, flooding and collapse made the pits dangerous, and the increasingly unweathered rock at depth made hand excavation more difficult. The main working had advanced about 40 m into the pegmatite when organized mining ended. For collectors, this short interval—mostly February 1999 through September 2000—is the golden window from which many of the finest Antsongombato specimens emerged.
The 2000 Tucson shows gave the locality its public debut. Dealers and collectors saw yellow, highly lustrous, unusually large rhodizite-londonite crystals on matrix, many with red tourmaline, and word circulated that the material might represent a new mineral. The timing could hardly have been better: londonite had been approved by the IMA in 1999, and the Antsongombato production placed attractive, macroscopic examples of the londonite-rhodizite series directly in front of the world’s specimen market. The excitement was justified. Up to January 2001, the mine yielded several hundred grams of gem-quality rough, from which more than 300 carats were faceted. Most stones were only 0.5–1 ct; about a dozen reached 1–2 ct; stones over 2 ct were extremely rare and usually included.
One of the most memorable scientific details is also one of the most practical for collectors: the yellow color could not be trusted as a permanent trait. Researchers noted that Antsongombato londonite-rhodizite crystals collected from dumps were bleached white on sun-exposed surfaces and darker yellow on their shaded undersides. Later testing showed the same vulnerability in cut stones. A natural-color rough sample faded after weeks in strong window light, and an irradiated faceted stone lost its yellow more dramatically. The lesson is wonderfully specific and a little sobering: the very color that made the locality famous is light-sensitive, and the best Antsongombato specimens should be treated as historical documents as much as display minerals.
Brendan M. Laurs, Federico Pezzotta, William B. “Skip” Simmons, Alexander U. Falster, and Sam Muhlmeister, “Rhodizite-Londonite from the Antsongombato Pegmatite, Central Madagascar,” Gems & Gemology, Vol. 38, No. 4, 2002, pp. 326–339. The key modern paper on the mine’s history, geology, mining, production, gemology, irradiation behavior, and the Antsongombato rhodizite-londonite market debut.
William B. Simmons, Federico Pezzotta, Alexander U. Falster, and Karen L. Webber, “Londonite, a new mineral species: the Cs-dominant analogue of rhodizite from the Antandrokomby granitic pegmatite, Madagascar,” The Canadian Mineralogist, Vol. 39, 2001, pp. 747–755. The formal londonite description; although the holotype is from Antandrokomby, the paper documents Antsongombato among the key Cs-rich londonite-rhodizite localities and illustrates exceptional Antsongombato crystals.
Francesco Demartin, Valeria Diella, Carlo M. Gramaccioli, and Federico Pezzotta, “Schiavinatoite, (Nb,Ta)BO4, the Nb analogue of behierite,” European Journal of Mineralogy, Vol. 13, No. 1, 2001, pp. 159–165. The type-mineral description for schiavinatoite from Antsongombato, including its association, chemistry, structure, and type-specimen information.
Handbook of Mineralogy, “Schiavinatoite.” A concise reference summarizing the type locality, formula, crystal data, association, and type material for schiavinatoite; the type specimen is recorded in the City Museum of Natural History, Milan, Italy, catalogue M31137.
Alfred Lacroix, “Nouvelles observations sur les minéraux des pegmatites de Madagascar,” Bulletin de la Société Minéralogique de France, Vol. 33, 1910, pp. 37–53. An early foundational reference for Malagasy pegmatite minerals cited in later Antsongombato and rhodizite-londonite studies.
Alfred Lacroix, Minéralogie de Madagascar, Vol. 1, 1922. A classic monographic source for Madagascar mineralogy, repeatedly cited in later work on the Betafo-Antsirabe pegmatites and early rhodizite occurrences.
J. Béhier, “Contribution à la minéralogie de Madagascar,” Annales Géologiques de Madagascar, 1960. The mid-century reference that specifically mentioned rhodizite from Antsongombato before the modern londonite-rhodizite mining period.
L. Duparc, M. Wunder, and R. Sabot, “Les minéraux des pegmatites des environs d’Antsirabé à Madagascar,” , Vol. 36, No. 3, 1910, pp. 283–407. An early account of the Antsirabe-area pegmatites and red tourmaline workings that forms part of Antsongombato’s historical context.
Mindat locality page: Antsongombato gem mine — The most useful locality database entry, with mineral list, coordinates, alternate names, and references.
Mindat type-locality report for Antsongombato gem mine — Confirms Antsongombato as the type locality for schiavinatoite.
Wikimedia Commons category: Antsongombato gem Mine — A convenient gallery of specimen photographs, including londonite, liddicoatite, béhierite, and schiavinatoite.
GIA / RRUFF PDF: “Rhodizite-Londonite from the Antsongombato Pegmatite, Central Madagascar” — Essential reading for the mine’s modern history, geology, gemology, production, and light-fading behavior.
Canadian Mineralogist / RRUFF PDF: “Londonite, a new mineral species” — Formal londonite description and key chemistry of the londonite-rhodizite series in central Madagascar.
European Journal of Mineralogy / RRUFF PDF: “Schiavinatoite, (Nb,Ta)BO4, the Nb analogue of behierite” — Original type-mineral paper for schiavinatoite from Antsongombato.
Handbook of Mineralogy: Schiavinatoite — Compact technical summary of schiavinatoite’s properties, occurrence, and type material.
Mindat photo gallery: Londonite — Includes documented Antsongombato londonite and londonite-liddicoatite specimen photographs.
Alexander U. Falster, Federico Pezzotta, William B. Simmons, and Karen L. Webber, “Rhodizite from Madagascar: New localities and new chemical data,” Rocks & Minerals, Vol. 74, No. 3, 1999, pp. 182–183. A short but important precursor to the londonite-rhodizite work in central Madagascar.
Notable museum record: the 1964–1965 Antsongombato campaign yielded a pale-yellow vitreous rhodizite-londonite crystal about 7 cm across, recorded in the collection of the Museum of the School of Mines in Paris.
