fall 1997 gems & gemology - gia · louderback’s docu-mentation of the ditrigonal-dipyramidal...

EDITORIAL165 Symposium 1999: Meeting the Millennium

William E. Boyajian

FEATURE ARTICLES166 Benitoite from the New Idria District,

San Benito County, CaliforniaBrendan M. Laurs, William R. Rohtert, and Michael Gray

188 Tairus Hydrothermal Synthetic SapphiresDoped with Nickel and ChromiumVictor G. Thomas, Rudolf I. Mashkovtsev, Sergey Z. Smirnov, and Vadim S. Maltsev

204 Multicolored Bismuth-Bearing Tourmalinefrom Lundazi, ZambiaMary L. Johnson, Cheryl Y. Wentzell, and Shane Elen

REGULAR FEATURES

212 Gem Trade Lab Notes220 Gem News231 1997 Challenge Winners233 Book Reviews235 Gemological Abstracts

ABOUT THE COVER: Commercial quantities of gem-quality benitoite are found atonly one location in the world, the Benitoite Gem mine, in San Benito County,California. This historic locality continues to produce fine-quality gemstones and miner-al specimens, and an extension of the deposit was found in the spring of 1997. The fea-ture article in this issue looks at the history of the Benitoite Gem mine, the geology of theNew Idria district and current mining there, and the gemological properties of this dis-tinctive material. The benitoite and diamond necklace illustrated here is shown with thelargest faceted benitoite to date, a 15.42 carat stone from the Benitoite Gem mine. Anartistic rendering, by GIA designer Lainie Mann, features this stone mounted in a pen-dant. Jewelry and gemstone from the collection of Michael M. Scott.

Photo © Harold & Erica Van Pelt––Photographers, Los Angeles, California.

Color separations for Gems & Gemology are by Pacific Color, Carlsbad, CA. Printing isby Cadmus Journal Services, Richmond, VA.© 1997 Gemological Institute of America All rights reserved. ISSN 0016-626X

T A B L E O F C O N T E N T S

pg. 185

pg. 214

pg. 205

GEMS&GEMOLOGYFALL 1997 VOLUME 33 NO. 3

pg. 189

1999

Editorial GEMS & GEMOLOGY Fall 1997 165

1999S Y M P O S I U M

Meeting the Millennium

In June 1991, GIA celebrated its 60th Anniversary with the Second International

Gemological Symposium in Los Angeles, California. Many of you came to that event and

have since told us that these were the most exciting and memorable four days of your career.

Having successfully completed the building of our new world headquarters and campus in Carlsbad,

California, we are now turning our attention to hosting the Third International Gemological Symposium,

June 21–24, 1999, in beautiful San Diego, California.

For GIA, this event will mark a seminal moment in the future of the gem and jewelry industry.

Focusing on the theme Meeting the Millennium, the Third International Symposium will feature more than

100 speakers and panelists, a “poster session” of reports on cutting-edge research, world-renowned opening

and closing keynote speakers, and a special day designed to stimulate the members of our industry to move

beyond traditional boundaries to meet the challenges of the new century. We anticipate more than 2,000 par-

ticipants in this unique event—a mixture of academic enrichment, gemological insight, cultural experience,

and networking opportunities that will be unlike any seen before in our industry. For those who have not yet

had an opportunity to visit our new world headquarters in Carlsbad, a full day of tours and other activities is

being planned immediately following the Symposium.

Although the event is still many months away, I suggest that you mark your calendars now.

Reserve your place for the experience of a lifetime. Come celebrate with us and meet the millennium at the

Third International Gemological Symposium, June 21–24, 1999.

William E. Boyajian

President, Gemological Institute of America

166 Benitoite GEMS & GEMOLOGY Fall 1997

ABOUT THE AUTHORS

Mr. Laurs is senior editor of Gems & Gemology,Gemological Institute of America, Carlsbad,California ([email protected]). Mr. Rohtert isManager of Gemstones, Kennecott ExplorationCompany, Reno, Nevada ([email protected]).Mr. Gray is owner of Graystone Enterprises, and apartner in Coast to Coast Rare Stones; he lives inMissoula, Montana ([email protected]).

Please see acknowledgments at end of article.

Gems & Gemology, Vol. 33, No. 3, pp. 166–187© 1997 Gemological Institute of America

em-quality benitoite has been mined intermittent-ly from only one region in the world, the New Idriadistrict of San Benito County, California. Benitoite

was first discovered there in 1907; it was named after thecounty, as well as the San Benito River, which runs throughthe property, and the nearby San Benito Mountain(Louderback, 1909). Stones are strongly dichroic, typicallyvioletish blue and colorless. Exceptionally rare gemstonesare colorless and pink. A few orange benitoites have beenproduced by heat treatment of colorless material. Benitoitehas high refractive indices, moderate birefringence, andstrong dispersion (exceeding that of diamond in some direc-tions), which render the faceted gems comparable in appear-ance and price to fine sapphire and tanzanite (figure 1).Although production has been sporadic throughout the lifeof the deposit, hundreds of carats of faceted material are cur-rently available, in sizes up to 2–3 ct.

Benitoite is a barium titanium silicate (BaTiSi3O9) thatforms under unusual conditions. At the New Idria district,benitoite is found exclusively in bodies of blueschist withinserpentinite. During 1995, the senior author conducted adetailed study of blueschist bodies throughout the NewIdria district on behalf of the Kennecott ExplorationCompany. The main occurrence, the Benitoite Gem mine,was mapped in detail, and other known benitoite occur-rences at the Mina Numero Uno and Victor prospects wereexamined. Microscopic benitoite crystals were detected forthe first time in a blueschist body on Santa Rita Peak. Afternearly a decade of inactivity, a small prospect containingminute amounts of gem-quality benitoite was rediscoveredalong a tributary of Clear Creek; this locality is now knownas the Junnila claim. More recently, an extension of theBenitoite Gem mine deposit was found in spring 1997, dur-ing exploration down slope of the historic pit.

The Benitoite Gem mine was referred to as the “Dallasmine” until the mid-1960s, but it has been variously calledthe “Benitoite mine,” “Dallas Benitoite mine,” and “Dallas

G

BENITOITE FROM THE NEWIDRIA DISTRICT, SAN BENITO

COUNTY, CALIFORNIABy Brendan M. Laurs, William R. Rohtert, and Michael Gray

Commercial quantities of gem-quality beni-toite are known from a single location in theworld, the Benitoite Gemmine in the NewIdria district of San Benito County, Californ-ia. A barium titanium silicate, benitoite istypically colorless to blue, and is noteworthyfor its high refractive indices, moderate bire-fringence, and strong dispersion. Benitoiteoccurs in altered blueschist within serpenti-nite. The gem crystals formed within frac-tures as a result of the alteration of blueschistby hydrothermal fluids derived from theregional metamorphism of the serpentinite.An extension of the historic deposit at theBenitoite Gemmine was discovered in thespring of 1997, which should contribute tocontinued stability in price and supply. Thegemological properties of benitoite readilyidentify it from similar-appearing gems.

Benitoite GEMS & GEMOLOGY Fall 1997 167

Gem mine” (Wise and Gill, 1977). Today the locali-ty is designated as the Benitoite Gem mine by themine owners, after the notation on topographicmaps of the U.S. Geological Survey which identifythe site simply as the “Gem mine.”

HISTORYAlthough there is general agreement in the litera-ture that benitoite was discovered in early 1907,there is considerable controversy about who actual-ly discovered it. Louderback (1907) originallyattributed the discovery to L. B. Hawkins and T. E.Sanders. However, two years later, he identifiedJ. M. Couch (figure 2), a prospector who was grub-staked by R. W. Dallas, as having first found a num-ber of mineral deposits that merited further investi-gation (Louderback, 1909). Dallas “induced”Hawkins to accompany Couch back into the moun-tains. According to Louderback (1909, p. 333),“While out to examine some copper deposits theyhappened upon the benitoite deposit and eachclaims to be responsible for the discovery.” In 1961,Couch’s eldest son, Oscar, published The BenitoiteStory, in which he outlined the reasons why hisfather should be given sole credit finding thedeposit. Austin (1988) provides further evidence forCouch being the sole discoverer of what, at thetime, he thought was a sapphire deposit. Frazier and

Frazier (1990a) reviewed the controversial historysurrounding the discovery and development of themine without making any judgment, and provided acomprehensive bibliography (1990b).

The first documented piece of benitoite to befaceted was brought to a lapidary in the SanFrancisco Bay area by Sanders’ brother. While anexpert from the Los Angeles area proclaimed thenew material “volcanic glass,” this lapidary calledthe stone spinel: “It was too soft for a sapphire, so Idecided it must be a spinel; [since] that’s the onlyother stone there is of that color” (Marcher, 1939).The lapidary showed some of the stones to GeorgeEacret, the manager of Shreve & Co., one of thelargest jewelry stores in San Francisco at that time.According to Marcher (1939), Eacret checked thestone with a dichroscope and found that it was dou-bly refractive, so he delivered a sample to Dr.George Louderback, a mineralogist at theUniversity of California (Berkeley) who determinedthat it was a new, undescribed mineral.

In the company of Mr. Eacret, Dr. Louderbackmade his first visit to the mine on July 19, 1907,and confirmed the discovery. He revisited thedeposit on October 11 of that year to study the geol-ogy. When Louderback returned to the mine withEacret on August 12, 1908, he took the first pho-tographs of benitoite for use in his 1909 full-length

Figure 1. Because of itsbeauty, rarity, and

singularity of occurrence,benitoite was declared the

official California StateGemstone in 1985. This

assortment of stones fromthe Benitoite Gem mine(1.10–4.77 ct) shows therange of colors that are

typically encountered inbenitoite. All stones were

faceted by Elvis (Buzz)and Michael Gray. Fromthe collection of MichaelM. Scott; photo © Harold

& Erica Van Pelt.


article describing the mineral. Louderback’s docu-mentation of the ditrigonal-dipyramidal crystalhabit finally proved the natural existence of thiscrystal form, which had been predicted mathemati-cally 77 years earlier (Hessel, 1830).

Soon after the discovery, the Dallas MiningCompany was formed to finance the developmentof the deposit, which was then called the “Dallasmine.” Cabins and corrals were built nearby, andmining equipment was hauled in by horse andwagon over tortuous roads from Coalinga, the near-est town. According to a diary of the Dallas MiningCompany, active mining began in July 1907. Anopen cut and a series of underground workings weredeveloped during the first few years of mining (fig-ure 3), and numerous plates of gem-bearing rockwere recovered.

Most of the benitoite crystals were coveredwith natrolite, which could have been dissolved inacid without harming the benitoite crystals.However, it was decided that this process was tootime-consuming, so the “knobs” formed by beni-toite crystals lying underneath the natrolite wereinitially broken off the rock matrix by use of ham-mer and chisel or a punch press. As a result, fewwell-crystallized mineral specimens were recoveredduring the early period of mining, as the main focuswas on the production of gem rough. The DallasMining Company diary records quantities of roughmeasured in “quart jars” and “cigar boxes” thatwere sent to Dallas’s offices in Coalinga.

The Dallas Mining Company continued opera-

tions until 1912, when it ceased being economical(Bradley et al., 1917). In October 1913, the miningequipment was auctioned off and the property wasvacated. The mine was issued mineral patent papersin 1917, five years after active mining ceased. Theownership of the mine remained in the Dallas fami-ly until 1987.

Between 1920 and 1940, little activity tookplace at the locality, except for occasional unautho-rized mining by various parties, including enterpris-ing teenagers Edward Swoboda and Peter Bancroft(Bancroft, 1984). Miller Hotchkiss, from the nearbySan Joaquin Valley town of Firebaugh, leased themine in the 1940s. Hotchkiss was the first to take abulldozer to the mine, which he used to rework themine tailings. From 1952 to 1967, a lease was heldby Clarence Cole, a mineral dealer from Oakland,California. Cole used a bulldozer and dynamite toenlarge the historic pit. In 1966, Cole granted a sub-lease to Gerold Bosley, of San Diego, California(Sinkankas, 1976). Backed by noted mineral collec-tor Josephine Scripps, Bosley used a bulldozer toexpose more of the open pit near the entrance of theoriginal tunnels. Bosley ultimately found very littlegem rough, but he did recover some notable mineralspecimens.

Cole died in 1967, and the lease subsequentlywas transferred to William “Bill” Forrest, of Fresno,California, and Elvis “Buzz” Gray, currently of Mis-soula, Montana. They purchased the mine from theDallas family in 1987, and they remain the soleowners today.

Figure 2. Although there hasbeen considerable controversysurrounding who actually dis-covered benitoite, much of theevidence supports the claim ofJ. M. Couch, shown here at a

campsite near the Benitoite Gemmine. Photo from the DallasMining Company archives.


LOCATIONANDACCESSThe Benitoite Gem mine is located 32 km (20miles) northwest of the town of Coalinga,California, at approximately 1,380 m (4,520 feet)above sea level. Paved county roads provide accessto the New Idria district from the southwest viaCoalinga, or from the northeast via Panoche Roadalong Interstate 5 in the San Joaquin Valley (figure4). These roads lead 63 km (39 miles) and 85 km (53miles), respectively, to a network of four-wheel-drive trails that are infrequently maintained by theU.S. Bureau of Land Management. Both routesrequire an additional 30 to 40 km of travel overrough dirt roads to reach the mine.

The Benitoite Gem mine is located on 16.2hectares (40 acres) of private patented mining prop-erty, which is secured by a locked gate. The site ispatrolled regularly by law enforcement officers, andvisitors must obtain written permission from theowners to enter the mine area. Four other benitoiteprospects in the district are claimed as follows: (1)The Junnila claim, owned by Leza Junnila of Fresno,California; (2) The Mina Numero Uno claim, ownedby Sharon and Eugene Cisneros of San Jose,California; (3) The Victor claim, owned by CraigStolberg of San Jose, California; and (4) The SantaRita Peak property, controlled by KennecottExploration Company. All of these mineral prospectsare closed to the public.

WORLDWIDEOCCURRENCEOF BENITOITEBenitoite has been confirmed from nine locationsaround the world, but only the historic BenitoiteGem mine and the Junnila claim have producedgem-quality material. All of the commercial gemproduction has come from the Benitoite Gem mine.At the three other deposits in the New Idria district,benitoite has been found only as small, platy (non-gem quality) crystals, up to a few millimeters indiameter (see descriptions of the Mina Numero Unoclaim, Victor claim, and Santa Rita Peak property inthe Geology section, below).

Outside the New Idria district, benitoite hasbeen found in situ at four areas. At Big Creek–RushCreek––in the Sierra Nevada foothills of easternFresno County, California––small grains of beni-toite are found in gneissic metamorphic rocks neara type of igneous rock known as granodiorite (Alforset al., 1965; Hinthorne, 1974). This occurrence islocated nearly 160 km (100 miles) northeast of the

Benitoite Gem mine, and is not geologically relatedto the New Idria district. In Japan, benitoite wasnoted from albite-amphibole rock in a serpentinitebody, along the Kinzan-dani River, at Ohmi in theNiigata Prefecture (Komatsu et al., 1973; Chihara etal., 1974; Sakai and Akai, 1994). At Broken Hill,New South Wales, Australia, Dr. Ian R. Plimerreported benitoite as a rare mineral in high-gradegranite gneiss (Worner and Mitchell, 1982). Mostrecently, crystals of colorless, blue, and pink beni-toite averaging 1–2 mm in diameter were detectedin gas cavities in syenite at the Diamond Jo quarryin Hot Springs County, Arkansas (Barwood, 1995;H. Barwood, pers. comm., 1997).

Detrital grains of benitoite (not in situ) wererecovered from Pleistocene lake sediments at theLazzard estate, a few kilometers west of Lost Hills,in the San Joaquin Valley, Kern County, California(Reed and Bailey, 1927). On the basis of geologicconsiderations, sediments from this location couldnot have been derived from the New Idria district orfrom the Big Creek–Rush Creek area.

Two previously reported benitoite locationsshould be discredited. Anten (1928) probably

Figure 3. This photo, taken around 1910, showsbenitoite miners and an adit at the BenitoiteGem mine. Photo from the Dallas MiningCompany archives.


to Panoche

to Coalinga

T18S

R12

ER

13E

R11

E

R12

E

T18S

T19S

San

Ben

itoC

ount

yF

resn

oC

ount

y

Idria

Idria

Rd..

Clear Creek Rd.

T17S

TERTIARYSedimentsSyenite

CRETACEOUSMoreno Formation

JURASSICFranciscan GroupSerpentiniteAltered serpentinite

Panoche FormationAsbestos pits

FaultsStrike and dip

Benitoite occurrence

N5 km3 mi.

3080

85

70

60

60

60

70

50

10

18

Santa Rita Peak

Santa Rita Peak

MontereySalinas

Hollister

Big Sur

Pacific

Ocean

Bitterwater

King City

San Lucas Coalinga

Idria

SanJoaquin

Valley

Los Gatos Creek Road

Benitoite Gem MineBenitoite

Gem Mine

F

Mina Numero

UnoMina

Numero Uno

JunnilaJunnila

VictorVictor

Clear Creek

DiabloRange

Figure 4. This simpli-fied geologic map ofthe New Idria dis-trict shows the distri-bution of benitoitedeposits. Benitoite isfound within Fran-ciscan blueschistbodies that havebeen tectonicallyincorporated into theserpentinite (afterEckel and Myers,1946; Coleman, 1957;Dibblee, 1979; andLaurs, 1995).

misidentified benitoite in a thin section fromOwithe Valley, Belgium (Petrov, 1995). Lonsdale etal. (1931) tentatively identified benitoite in sedi-ments from the Eocene Cook Mountain formationof southwest Texas; Smith (1995) suggested that thelocality should be discredited because the authorsconfused the spelling of bentonite (a clay mineral)with benitoite, but we recommend discreditationbecause the available data provided by Lonsdale etal. suggest that they misidentified grains that wereactually sapphirine. Rumors of benitoite from Koreahave not been confirmed.

GEOLOGYGeologic Setting. The New Idria district is locatedin the southern Diablo Range of the California Coast

Range geologic province. Since 1853, the district hasbeen mined and prospected for numerous mineralresources, including mercury, chromium, gold,asbestos, gems, and mineral specimens. The districtencompasses a serpentinite body that was tectoni-cally emplaced into surrounding sedimentary andmetamorphic rocks. During the late Jurassic, whentwo of the earth’s plates collided (see, for example,Hopson et al., 1981), the relatively low-density ser-pentinite rose through the overlying layers of rock,which included the Franciscan Formation. Slices ofthe Franciscan Formation were incorporated intothe highly sheared serpentinite body during itsemplacement; these are called tectonic inclusions(see Coleman, 1957). The serpentinite breached thepaleo-surface in the mid- to late-Miocene (Coleman,


1961). It is now exposed along the crest of theDiablo Range (figure 5) over an area 23 km by 8 km,elongate to the northwest.

Tectonic inclusions of Franciscan rocks withinthe serpentinite consist dominantly of blueschistand graywacke, with lesser mica schist, greenstone,and amphibolite schist (Coleman, 1957; Laurs,1995). All of these rocks are derived from oceaniccrust and overlying sediments that were accretedonto the continental margin during the Jurassicperiod. Two blueschist protoliths (parent rocks)may be differentiated on the basis of texture andcomposition: (1) metavolcanic, which are largelybasaltic lavas and volcaniclastics; and (2) metasedi-mentary, which are largely marine sediments.Depending on the composition of the protolith, theblueschist contains variable amounts of very fine-grained albite, glaucophane-crossite, actinolite-tremolite, aegerine-augite, and titanite, with orwithout stilpnomelane, quartz, K-feldspar, epidote,and apatite. Benitoite is found in blueschist derivedfrom both protoliths (Laurs, 1995).

Prior to breaching the surface, the New Idria ser-pentinite was intruded by small bodies of igneousrocks, predominantly syenite (Coleman, 1957; again,see figure 4). The mid-Miocene age of the syenitecorrelates well with Neogene magmatism in theCalifornia Coast Range, which is associated with theplate tectonic reconfiguration of western California(Van Baalen, 1995). Metamorphism associated withthis reconfiguration is probably responsible for form-ing calc-silicate vein assemblages that are scatteredthrough the serpentinite, as well as benitoite-bearingveins in the blueschist bodies (Van Baalen, 1995).Tentative age data suggests that the benitoite crys-tallized about 12 million years ago (M. Lanphere,pers. comm., in Van Baalen, 1995), so it is muchyounger than the enclosing blueschist, whichformed about 100 to 160 million years ago (see, forexample, Lee et al., 1964).

Benitoite Gem Mine. Previous Work. Arnold (1908)and Sterrett (1908, 1911) summarized the early min-ing activity. Coleman (1957), Van Baalen (1995), andWise and Moller (1995) wrote district-wide geologi-cal reports, which include discussions of theBenitoite Gem mine. Geologic maps of the minewere prepared first by Coleman (1957), and later byRohtert (1994) and Laurs (1995). Laird and Albee(1972) made physical, chemical, and crystallographicmeasurements on benitoite and associated mineralsfrom the district, and Wise and Gill (1977) provided

a detailed description of the complicated mineralo-gy.

Geology. The mine is situated on a low hill (figure6), which is underlain by Franciscan rocks emplacedinto serpentinite (Coleman, 1957; Wise and Gill,1977). These rocks consist of blueschist and green-stone, which are locally sheared (figure 7). Theblueschist is dark bluish gray where unaltered, andlighter blue-green within the mineralized zone.This zone is at least 60 m long, strikes N60°W, dipsmoderately northeast, and is about 3 m thick.Recent field observations by the authors suggestthat the deposit is offset along two north- to north-west-trending faults (again, see figure 7). All of thehistoric lode production was obtained from the cen-tral section of the deposit. However, in the spring of1997 an extension of the mineralized zone was dis-covered in the western offset portion. The faultingapparently down-dropped a portion of the mineral-ized blueschist, which lay buried beneath up to 10m of unconsolidated eluvium and dump material.

Mineralization. Benitoite mineralization is confinedentirely to the blueschist, in a hydrothermally

Figure 5. This low-altitude oblique aerial photo-graph, looking northeast, shows the Benitoite Gemmine workings (small brown cut). The large openpit on the left is an active asbestos mine beingworked by the King City Asbestos Corp. SantaRita Peak, where benitoite has also been found, isvisible above the Benitoite Gem mine and formsthe highest point on the skyline. Photo courtesy ofKennecott Exploration Company.


altered zone, which is characterized by: (1) recrystal-lization of fibrous amphibole and pyroxene, (2) localalbite dissolution, and (3) veining and pervasiveinfiltration by natrolite. Benitoite formed during thefirst two stages of the alteration (inasmuch as manycrystals contain amphibole and pyroxene inclu-sions), but prior to the formation of natrolite––which coats benitoite in the veins (figure 8). Thenatrolite veins average less than 1 cm wide, andcontain benitoite only locally, commonly where theveins narrow or terminate.

The benitoite forms as euhedral crystals (figure 9)up to 5.6 cm in diameter, 1–1.5 cm on average, thatare attached to the vein walls. Other minerals on theblueschist vein walls include neptunite, silica pseu-domorphs after serandite, joaquinite group minerals,apatite, albite, jonesite, and the copper sulfidesdjurleite, digenite, and covellite (Wise and Gill, 1977),as well as traces of other minerals described by VanBaalen (1995) and Wise (1982). Natrolite coats all ofthe vein minerals and, in most cases, completely fillsin and closes the veins. Natrolite also has infiltratedthe altered blueschist adjacent to the veins, fillinginterstitial space that was previously occupied byalbite.

Two types of benitoite were noted by Wise andGill (1977): (1) gem-quality crystals attached to thewalls of cross-cutting veins; and (2) disseminated,euhedral, non-gem benitoite with abundant amphi-

HISTORIC MINE AREA

1997 DISCOVERY

O pen pit high wall

30

60

65

60-80

40

50

3050

40 ft 12 m

N

60

*arrow shows dip; U=up, D=down

50

Colluvium and eluvium

Sheared greenstone Greenstone Altered blueschist Blueschist Serpentinite

Fault* Strike & dip of foliation

Figure 6. In this 1997 photo (by Michael Gray) of the Benitoite Gem mine, a backhoe sits in the main open cut,and the equipment processing colluvial material can be seen in the center, just below the single pine tree. Twosettling ponds are visible below this equipment. Water is pumped from the San Benito River at the base of theworkings into the ponds, from where it is recirculated through the processing equipment. The inset (afterLouderback, 1909) shows a view of this same area in 1907, shortly after the discovery of the deposit.

Figure 7. This simplified geologic map (top) andcross-section (bottom) of the Benitoite Gem mineshow the altered blueschist where benitoite isfound, in lenses that are tectonically incorporat-ed into the sheared greenstone. After Coleman(1957), Rohtert (1994), and Laurs (1995).


bole and pyroxene inclusions that formed within thealtered blueschist. Some crystals show both charac-teristics, since they apparently grew into both thevein and the adjacent host-rock substrate. Benitoitein the veins commonly forms simple triangular crys-tals with dominant π {0111} faces (Wise and Gill,1977; figure 10). The crystals typically show a frostyappearance on the π faces due to natural etching,whereas the µ faces are mirror smooth. Color zoningis common, with a milky white or colorless coregrading outward into transparent blue corners.However, gem-quality blue crystals recently discov-ered in the western offset portion of the deposit typi-cally lack color zoning and show only π faces (A infigure 10). In general, the relatively small size offaceted material is due to the abundant cloudy areasand the flattened morphology of the crystals.

The disseminated non-gem benitoites closelyresemble the host rock in color, because of the abun-dant fibrous inclusions from the altered blueschist.As described by Wise and Gill (1977), such crystalscommonly show dominant c faces, as well as vari-ably developed pyramid (π and p) and prism (µ andm) faces, resulting in tabular crystals with triangularto hexagonal outlines (D and E in figure 10). Rarely,the non-gem crystals form star-shaped twins causedby penetration twinning due to a Dauphiné-type180° rotation about the c-axis (W. S. Wise, pers.comm., 1997; figure 11). The mine owners know of

only nine twinned crystals recovered thus far (E.Gray, pers. comm., 1997).

Junnila Claim. The Junnila claim is located 7 kmnorthwest of the Benitoite Gem mine, on a low hilloverlooking a tributary of Clear Creek (again, see

Figure 9. This specimen, which was mined dur-ing the late 1980s, shows an unusually transpar-ent benitoite crystal on a matrix of white natro-lite and light green altered blueschist. The natro-lite originally covered the benitoite crystals aswell, but it was removed with dilute hydrochlo-ric acid. Photo © Harold & Erica Van Pelt.

Figure 8. On May 1, 1997, this remarkable boulderof benitoite-bearing blueschist was excavated fromthe newly discovered extension of the BenitoiteGem mine. Knobs on the surface of the natrolite(which covers 40 cm × 70 cm of the boulder) indi-cate the presence of benitoite and neptunite crys-tals, which are attached to the vein wall and over-grown by natrolite. Several gem-quality pieces wereremoved from the outside edge of the vein. Photoby Michael Gray.

π

π

π

µ

µ µ

π

π

π

πp

p

pmp

c

cc

π

π

A B

C

Figure 10. The common crystal habits of benitoiteare redrawn from Louderback (1909) and Wise andGill (1977). The π face decreases in importance inthe sequence from (A) to (E). The morphology of thegem-quality crystals generally resembles (C), exceptfor crystals from the recently discovered westernextension of the deposit, which resemble (A). Non-gem crystals typically resemble (D).

ED

–


figure 4). The claim was worked intermittently byA. L. McGuinness and Charles Trantham from1982 to 1986. The deposit lay forgotten after thedeath of Mr. McGuinness in the late 1980s, but wasrediscovered in 1995 by the Kennecott ExplorationCompany and almost simultaneously by L. Junnilaof Fresno, California. At the time of rediscovery,thick brush covered the area and only a small beni-toite-containing prospect pit was found (Laurs,1995). In fall 1995, the property was excavated usinga D-8 tractor (figure 12) in a cooperative effort byJunnila and several mineral dealers (Moller, 1996;Ream, 1996). Although only small amounts of beni-toite were found, a few crystals (figure 13), up to 2.6cm in diameter (Ream, 1996), reportedly containedmaterial suitable for faceting; it is from this materi-al that the two stones (0.42 ct and 0.21 ct) shown infigure 14 were cut.

As at the Benitoite Gem mine, benitoite at theJunilla claim is found in a body of blueschist withinthe serpentinite. The blueschist crops out over anarea approximately 50 m by 20 m; it is brownishgray and weathered into platy fragments (Laurs,1995). Near the center of the body is a zone ofaltered blueschist that contains benitoite. This ver-tical zone is resistant to weathering, massive in tex-ture, and has a distinctive light grayish blue color; itstrikes northeast and ranges from 1 to 2 m thickover a length of 6 m. Sparse amounts of benitoiteform along vertical tension gashes (commonly lessthan 1 cm wide), and also along foliation and cross-fractures. In contrast to the natrolite gangue presentat the Benitoite Gem mine, the benitoite-bearingveins at the Junnila claim are filled with a calcium-rich mineral assemblage consisting mostly of thom-sonite, pectolite, calcite, and stevensite; neptuniteand joaquinite-group minerals are present locally(Laurs, 1995).

The benitoite crystals are translucent (rarelytransparent), tabular (see, e.g., figure 10D), and aver-age 1 cm in their longest dimension. The crystalsare intergrown with other vein minerals, and there-fore they rarely show more than two or three faces.Blue, colorless, or color-zoned crystals (typicallywith colorless cores and blue rims, as at theBenitoite Gem mine) have been found.

Mina Numero Uno Claim. Located along upperClear Creek, 7.8 km northwest of the BenitoiteGem mine, the Mina Numero Uno claim consistsof boulders and rare outcrops of blueschist scat-tered through dense brush over an area approxi-mately 20 m by 100 m, elongate to the east-north-east. Benitoite mineralization was noted in boul-ders at opposite ends of the deposit (Laurs 1995).The boulders at the eastern end contain rare vugs,up to 6 cm long and 3 cm wide, that are elongatedparallel to foliation. The vugs are commonly linedwith albite and silvery gray fibrous amphibole; inplaces, they contain benitoite, neptunite, or joaqui-nite-group minerals (Wise, 1982; Laurs, 1995).Benitoite forms pale blue, platy, euhedral crystalsthat are up to 5 mm in diameter. Less common aretan grains, previously reported as “pink” byChromy (1969).

In the boulders at the western end, benitoiteand neptunite form along narrow, randomly ori-ented fractures less than 1 mm wide (Laurs, 1995).These fractures typically contain no gangue min-erals except for minor amounts of quartz locally(there is no albite). Benitoite forms pale blue orcolorless, tabular, pseudo-hexagonal plates androsettes up to 10 mm in diameter, with most crys-tals averaging 2 mm in diameter. The crystalsoccur with neptunite and lay flat against the frac-ture walls.

p p

c

cππ

Figure 11. Twinned benitoite crystals show a 180° rotation about the c-axis. In this drawing (courtesy ofWilliam S. Wise), the c faces of the two crystals are merged and indistinguishable from one another, resultingin flat, star-shaped faces. The photo shows some of the twinned benitoite crystals (2–3 cm wide) recoveredfrom the Benitoite Gem mine. Note the characteristic opaque appearance, resulting from abundant inclusionsof fibrous amphibole and pyroxene. Photo © Harold & Erica Van Pelt.


Victor Claim. The Victor claim is located along atributary 1 km south of Clear Creek, and 9.2 kmnorthwest of the Benitoite Gem mine. According todocumentation at an old claim marker on the prop-erty, the Victor claim was first staked in 1974 as theFranciscan mine by Steven M. Dwyer of PalmDesert, California. During the early 1980s, theclaim was owned by Ed Oyler of the San FranciscoBay area; in 1991, the deposit was acquired by CraigStolberg of San Jose, California. Outcrops ofblueschist up to 3 m in diameter are surrounded bya sheared tremolite-chlorite zone within the serpen-tinite (Laurs, 1995). Millage (1981) inferred that theblueschist body measures 150 m × 100 m. The cen-tral blueschist outcrops contain small (up to 5 mmin diameter) colorless benitoite platelets androsettes, in thin veinlets along the foliation.

Santa Rita Peak Property. This occurrence is located1.2 km north-northeast of the Benitoite Gem mine,on the southeast flank of Santa Rita Peak (elevation

5,165 feet, 1,574 m). The blueschist is exposed as afield of boulders, at least 30 m wide, that is sur-rounded by a sheared tremolite-chlorite-jadeite zonewithin the serpentinite (Laurs, 1995). Some of theboulders are hydrothermally altered and cut bynatrolite veins up to 2 cm thick. In places, thealtered blueschist shows a nodular texture similar

Figure 13. This is one of the largest benitoite crys-tals (2.2 cm wide) recovered from the Junnila claimduring the 1995 excavation. The benitoite is inter-grown with thomsonite and actinolite. The irregu-lar surface texture is due to intergrowth with cal-cite, which was removed from the specimen byacid dissolution. Specimen courtesy of WilliamLarson; photo © Jeffrey Scovil.

Figure 14. These two benitoites (0.42 and 0.21 ct)were reportedly faceted from material found at theJunnila claim. Stones courtesty of William Larson;photo © GIA and Tino Hammid.

Figure 12. The Junnila claim was excavated withheavy equipment in 1995. Benitoite was recoveredfrom the narrow, vertical zone of resistant blue-schist that is visible here beneath the slender pinetree. Photo by William Larson.


to that observed at the Benitoite Gem mine; elec-tron microprobe analysis revealed microscopic crys-tals of benitoite in this altered blueschist (R. L.Barnett, pers. comm., in Laurs, 1995).

ORIGINOF BENITOITEEver since benitoite’s discovery, its origin has beena persistent enigma. The formation of benitoite atthe New Idria district has been generally ascribed tohydrothermal processes that caused the unusualcombination of barium (Ba) and titanium (Ti).Coleman (1957) suggested that Ti was derived fromfluids associated with the crystallization of smallsyenite intrusions in the district; Ba was liberatedfrom the alteration of blueschist. However, Ti is rel-atively immobile in hydrothermal fluids (VanBaalen, 1993), and the closest syenite to any of the

benitoite deposits in the district is 1 km. Wise andGill (1977) proposed that fluids derived Ti, Fe, andrare-earth elements from the serpentine, and Bafrom the blueschist, to form benitoite and the asso-ciated vein minerals. Most recently, Van Baalen(1995) proposed that benitoite formed as a result ofmetamorphism localized along the contact betweenblueschist and greenstone, in the presence of sodi-um-rich, low-silica metamorphic fluids. He suggest-ed that more than enough Ba and Ti were present inthe blueschist and greenstone to produce theinferred amount of benitoite and neptunite.

We generally concur with Van Baalen’s (1995)model, subject to some modification. Van Baalen(1995) stated that benitoite formed along the con-tact between blueschist and greenstone at theBenitoite Gem mine; it is actually contained entire-ly within the blueschist, regardless of proximity tothe greenstone. Greenstone is not required for beni-toite formation, as indicated by its absence at theJunnila and Mina Numero Uno claims, although itis possible that it contributed to Benitoite formationat the Benitoite Gem mine. Throughout the NewIdria district, there is also a strong correlationbetween the amounts of blueschist alteration andbenitoite mineralization. We suggest that benitoiteformed at the Benitoite Gem mine when Ba and Tiwere released by the alteration of blueschist, andpossibly greenstone, in the presence of magnesium-and calcium-rich fluids generated during the region-al metamorphism of serpentinite (rather than in thepresence of sodium-rich fluids, as postulated by VanBaalen [1995]). This model is based on detailed geo-logic mapping and the collection of chemical dataon blueschist minerals and whole-rock samples atall the known benitoite occurrences in the district(Laurs, 1995).

MODERNMININGThe present owners of the Benitoite Gem mine,Forrest and Gray, initially worked the open pit formineral specimens, including neptunite and jone-site. One boulder produced a small quantity of pinkbenitoite, which yielded gems of one-quarter caratto just over one carat, but overall the amount ofgem material recovered from the lode was not sub-stantial. In 1970, Forrest and Gray began washingthe mine-dump material by pumping water uphillthrough a fire hose. Then (as now) they worked onlyduring the spring months, to take advantage of thefavorable weather as well as the water availablefrom the headwaters of the San Benito River, which

Figure 15. With 52 faceted benitoites weighing atotal of about 33 ct (the largest stone is 2.84 ct), thisbenitoite and diamond necklace is unique in theworld. All of the benitoites were cut and matchedby E. (Buzz) Gray, and the necklace was designedby William McDonald of McDonald’s Jewelry,Fresno, California. Necklace courtesy of Michael M.Scott; photo © Harold & Erica Van Pelt.


runs through the property. After washing the dumpmaterials, they picked out mineralized blueschistand loose pieces of benitoite by hand (Gray, 1986).Using this very simple method during the early1970s, they recovered significant quantities offaceting rough that produced some fine stones.Among these was the flawless 6.53 ct pear-shapedbrilliant that was the center stone of the pendant tothe famous benitoite and diamond necklace thatwas stolen in Europe in 1974. The necklace (figure15) was recovered in 1975, but the pendant wasnever found.

After working through most of the mine dump,the two owners processed the underlying colluviumand eluvium using the same mining methods. Thenatrolite coating protected many of the benitoiteand neptunite crystals during weathering and trans-port, so many fine mineral specimens were recov-ered. The gem rough recovered during this miningoperation was generally of high quality, since thecrystals that weathered out of the veins had typical-ly broken along existing fractures to isolate thegemmy tips and nodules.

Since 1982, the owners have used a more sys-tematic and mechanized recovery method (Gray,1992). A front-end loader feeds dirt and rockthrough a grizzly, to separate out boulders, into ahopper (figure 16), where water is used to wash thematerial down a chute to a layered screening appa-ratus fitted with high-pressure water jets (figure 17).Material larger than 25 mm (1 inch) moves over thetop screen, where it is again cleaned by water jetsand checked for specimen potential. Material small-er than 25 mm falls onto a lower screen with a 3

mm (1/8 inch) mesh. Pieces smaller than 3 mm aredischarged into settling ponds, and materialbetween 3 and 25 mm is washed into a gravitationalseparation jig, where the heavier benitoite is trappedin several parallel trays. Rough gem benitoite (figure18) is removed from the trays by hand at the end ofeach work day. In the late 1980s, Forrest and Grayrecovered a piece of rough from which the 10.47 ctgem pictured in figure 19 was cut. The largest,

Figure 16. The mine owners at the Benitoite Gemmine process colluvial and alluvial material usingmechanized equipment. Here a loader dumpsmaterial through a grizzly and into a hopper. Rockslarger than 15 cm roll off the grizzly and are exam-ined individually; stones smaller than 15 cm arewashed down a chute to the apparatus pictured infigure 17. Photo by Brendan M. Laurs.

Figure 17. At the BenitoiteGem mine, material is fed firstinto a layered screening appa-ratus fitted with high-pressure

water jets, and then into agravitational separation jig.

Pieces over 25 mm are separat-ed by the screening apparatus

and checked for specimenpotential, while materialbetween 3 and 25 mm is

washed into the jig where theheavier benitoite is trapped inseveral parallel trays. Photo by

Brendan M. Laurs.


finest rough found to date was recovered in theearly 1990s; it was cut into a 15.42 ct gem (see coverof this issue).

By the end of the 1996 mining season, the collu-vium and eluvium were largely exhausted. Usinglarger machinery, mine owners Forrest and Grayfound a new productive area down slope of the openpit during spring 1997 (figure 20). Excavation duringthe 1997 season (March through May) revealed min-eralized lode material that contained fine mineralspecimens, as well as colluvium and eluvium withabundant gem rough. Mining of this new area willcontinue in 1998.

PRODUCTIONThe Benitoite Gem mine is the only commercialsource of gem-quality benitoite in the world. Fromthe time of its discovery in 1907 until 1967, whenForrest and Gray began working the mine, it is esti-mated that about 2,500 carats of faceted benitoitewere produced (E. Gray, pers. comm., 1997). Of thatamount, nearly 1,000 carats were produced duringthe period of active mining between 1907 and 1911,based on the amount of rough reported in the DallasMining Company’s account books. It was duringthat time that the gem rough that produced the 7.6ct gem pictured in Louderback’s 1909 report wasrecovered; this was subsequently recut to the 7.53ct stone that is now in the Smithsonian Institution.Few other stones exceeded 3 ct, however, and mostof the gems weighed less than 1 ct.

The balance of 1,500 carats believed to havebeen recovered between 1911 and 1967 was esti-mated from verbal information provided by previ-ous miners (E. Gray, pers. comm., 1997). Most ofthese stones were faceted by cutters in the UnitedStates, and weights of up to 5 ct were obtained.Clarence Cole retrieved most of the rough duringthe interval from 1952 to 1962.

Since 1967, Forrest and Gray have producedapproximately 2,000 carats of faceted benitoite.According to E. Gray (pers. comm., 1997), thisinventory can be divided into six weight classes: (1)About 2,000 pieces were cut in commerical factor-ies, and generally finished to less than 0.25 ct each;(2) another 1,500 pieces ranged between 0.25 and 1ct; (3) some 500 stones weighed between 1 and 2 ct;(4) a total of 50 stones ranged between 2 and 3 ct; (5)only 25 stones weighed between 3 and 4 ct; and (6)the 15 largest stones exceeded 4 ct. Thus, 89% ofthe stones were under 1 ct; 9% were between 1 and2 ct; and 2% were over 2 ct. All but the commer-cially cut stones were faceted by the families of thecurrent owners or in their own facilities. Forrest andGray have not marketed cuttable rough except, in afew cases, as mineral specimens.

Forrest and Gray also estimate that about 500carats of finished goods have entered the marketthrough informal channels and other sources, suchas the cutting of older rough and the faceting ofmineral specimens. Thus, a total of about 5,000carats of faceted benitoites have been produced overthe life of the mine. Because of this small produc-

Figure 18. Rough benitoite gem nodules and crys-tal fragments, such as these, are removed fromthe gravitational separation jig by hand. Thelargest stone on the upper right is 1.0 cm wide.Photo by Maha DeMaggio.

Figure 19. This exceptional faceted benitoite weighs10.47 ct. From the collection of Michael M. Scott;photo © Harold & Erica Van Pelt.


tion, benitoite is a collector’s gem, one of the rarestin the world.

PHYSICAL ANDCHEMICAL PROPERTIESMaterials and Methods. We examined a total of 139benitoite samples, of which 83 were faceted stonesand four were cabochons. With the exception of twofaceted stones reportedly from the Junnila claim, allof the benitoites examined were from the BenitoiteGem mine. The rough stones (52) were all of gemquality, and were provided by Forrest and Gray fromtheir gravity-separation apparatus. The fashionedsamples were selected for their various internal fea-tures; consequently, they were not as clean as beni-toites typically seen in the marketplace.

The faceted stones in our study ranged from0.10 to 1.46 ct (see, e.g., figure 21), and the cabo-chons ranged from 1.06 to 5.31 ct. All of the sam-

ples––both rough and fashioned––were examinedwith magnification to locate and describe inclu-sions and color zoning. Refractive index measure-ments were made on the 21 faceted stones thatwere larger than 0.20 ct. Specific gravity, fluores-cence, and visible absorption spectra were alsodetermined for these faceted stones, as well as forthe four cabochons.

Refractive indices and birefringence were mea-sured on a GIA GEM Instruments Duplex II refrac-tometer, using a 1.815 contact liquid and amonochromatic sodium-equivalent light source.Because the upper R.I. of benitoite (1.804) is close tothe upper limit of the Duplex II (1.810), we alsomeasured the four benitoites pictured in figure 21with a specially designed refractometer fitted with acubic zirconia hemicylinder. Specific gravity wasmeasured by the hydrostatic method (three mea-surements per sample) with a Mettler CM1200 digi-tal balance. All of the samples were examined witha standard gemological microscope, a LeicaStereozoom with 10× to 60× magnification; therough specimens were immersed in methyleneiodide. Absorption spectra were observed with aBeck spectroscope on a GIA GEM Instruments base.

Electron microprobe analyses of some of themineral inclusions in benitoite were performed atthe University of Manitoba in Winnipeg, Canada,on two gem-quality pieces of rough that containedinclusions visible with 10× magnification.

Figure 20. The backhoe is excavating the newly dis-covered western extension of the deposit at the Be-nitoite Gem mine. Boulders of altered blueschist,some bearing benitoite, are piled near the left rim ofthe pit. The high wall of the main pit can be seen onthe skyline. Photo by Brendan M. Laurs.

Figure 21. These faceted benitoites from theBenitoite Gem mine, which show a range of toneand saturation, were among the samples studiedfor this investigation. Clockwise from the top, thestones weigh 0.95 ct, 0.87 ct, 0.49 ct, and 0.54 ct.Photo by Maha DeMaggio.


Visual Appearance. The samples ranged from blueto violetish blue in color, and from very light tomedium dark in tone. This color range is represen-tative of the stones commonly marketed (figure 22).

Dark colors were seen in melee (less than 0.20 ct),as well as in larger stones. Most of the gems weremedium to medium-dark in tone, with strong satu-ration, a color range sometimes referred to as “corn-flower” blue. A green tinge was evident in some ofthe paler cabochons due to inclusions of fibrousgreen amphibole or pyroxene (discussed below).Strong dichroism (figure 23), from colorless to blueor violetish blue, was evident when the stones wereviewed through a dichroscope in any direction. Tothe unaided eye, pleochroism was visible as lightand dark tones that mingled with the brilliance anddispersion. The dispersion is masked somewhat bythe blue body color. Color zoning was common.

As described previously, colorless and pinkfaceted stones are extremely rare; consequently,they were not included in this portion of the study.Heat treatment of lighter colored material mayresult in an orange hue similar to that associatedwith Imperial topaz (see, e.g., figure 24); the dichroiccolors of this material are pink and orange. Theorange color has not been observed in untreatedbenitoite. Heat treatment has been successful foronly a small portion of the material treated, andsome of the crystals have exploded in the furnacebecause of differential expansion of inclusions. Theheat-treatment procedure is proprietary, and experi-ments were not performed as part of this study.

Physical Properties. The standard gemological prop-erties of benitoite are shown in table 1 and dis-cussed below.

Refractive Indices. Most of the stones showed nω =1.757 to 1.759, and nε = 1.802 to 1.804. The accura-

Figure 22. Weighing from 0.30 to 1.70 ct, these ben-itoites show a range in size, color, and facetingstyle that is representative of commercially avail-able benitoite. Courtesy of Edward Swoboda;photo © Harold & Erica Van Pelt.

Figure 23. This benitoite cabochon (5.31 ct) displays strong dichroism and contains fibrous inclusions ofamphibole and/or pyroxene. The dark reddish brown inclusion on the lower left is neptunite. Darkfield illu-mination; photomicrograph by John I. Koivula.


cy of the upper measurements was compromised bytheir proximity to the upper limit of the refractome-ter. Nevertheless, these measurements comparefavorably to the published values of nω = 1.757 andnε = 1.804 (Webster, 1994). In addition, the fourstones tested with the cubic zirconia refractometershowed nε = 1.804 to 1.805. The measured birefrin-gence ranged from 0.043 to 0.047, which comparesfavorably to the published value of 0.047 (Webster,1994). Because of the high birefringence, doubling ofpavilion facet junctions is easily seen through thetable with 10× magnification.

Specific Gravity. The measured values for thefaceted stones ranged from 3.65 to 3.80. In most ofthe stones larger than 0.40 ct, the S.G. values werenear 3.70. This differs somewhat from the publishedmeasured value of 3.65, but it is close to the pub-lished calculated value of 3.68 (Anthony et al.,1995). The largest scatter in measured values wasnoted for the smaller stones, which is due to mea-surement error. The S.G. of the cabochons rangedfrom 3.31 to 3.69; the lighter measurements wereobtained for stones containing abundant fibrousinclusions (described below).

Ultraviolet Fluorescence. Similar to the descrip-tions by Wise and Gill (1977) and Mitchell (1980),the sample blue benitoites fluoresced strong blue toshort-wave UV radiation, and appeared inert to

long-wave UV; the colorless benitoite fluorescedslightly stronger blue to short-wave, and showeddull red fluorescence to long-wave. Some light-col-ored samples examined in this study also fluoresceddull red to long-wave UV. Moderate chalkiness ischaracteristic of the fluorescence to both short- andlong-wave UV. Because of benitoite’s conspicuousfluorescence, miners have recovered some crystalsby scanning the ground surface at night with aportable UV lamp.

Internal Features. Inclusions. Microcrystals, “fin-gerprints,” and fractures are commonly seen in ben-itoite. Many of the 83 faceted stones examined forthis study contained white needles as dissemina-tions and intergrown clusters (figure 25). We did notnote any preferred orientation of the needles. Theneedles are probably actinolite-tremolite, as sug-gested by petrography and by electron microprobeanalyses of similar crystals in the host rock (Laurs,1995). Because the needles are so small (typicallyless than 0.5 mm long), they are rarely visible withthe unaided eye unless they are abundant. With 10×magnification, they resemble lint particles inappearance and texture. Also present in heavilyincluded samples were dark blue and green needlesthat formed elongate tufts on the order of severalmillimeters long (again, see figure 23). Electronmicroprobe analyses of benitoite inclusions showedthe presence of aegerine-augite and diopside, which

Figure 24. Shown here, clockwisefrom the top left, are some of the

more unusual benitoites that havebeen found at the Benitoite Gem

mine: a 0.79 ct light blue round bril-liant, with unusually large two-

phase inclusions; a 1.29 ct pinkishblue round brilliant; a 1.07 ct nearlycolorless lozenge-shaped mixed cut,

with a slight blue tint; a 0.34 ct darkpurplish pink round brilliant; a 0.53ct heat-treated pinkish orange oval

mixed cut; a 0.61 ct heat-treated par-ticolored (pinkish orange and blue)rectangular modified emerald cut;and a 0.91 ct heat-treated particol-ored shield shape mixed cut. Theparticolored stones resulted from

heat treatment of material that wasoriginally colorless and blue. All

stones were faceted by Elvis (Buzz)and Michael Gray. From the collec-

tion of Michael M. Scott; photo ©Harold & Erica Van Pelt.


are both pyroxenes (F. Hawthorne, pers. comm.,1997).

Silica pseudomorphs after serandite form pris-matic euhedral inclusions in benitoite, but these

are seldom seen in faceted stones (see the centerright photomicrograph in Gübelin and Koivula,1986, p. 416; these crystals were apparently mis-identified as neptunite). Other inclusions seen onlyrarely were anhedral to euhedral dark reddish brownneptunite and euhedral honey-colored joaquinite-group minerals (specific species not determined).Albite, apatite, and metallic greenish gray djurleitecrystals have been reported as inclusions in beni-toite (Wise and Gill, 1977), but they were notobserved in this study. Two-phase (liquid and gas)inclusions (figure 26) were noted in only two sam-ples, along healed fractures. The “fingerprints” arecommonly planar, but they appeared curved in sev-eral stones. Unhealed fractures typically showedstep-like conchoidal break patterns.

Color and Growth Zoning. Color zoning was mostapparent in emerald-cut stones, which commonlydisplayed a gradation from blue to colorless whenviewed face-up with the unaided eye. When thestones were viewed from different angles at 10×magnification, sharp planar boundaries between theblue and colorless areas were eye-visible in one ormore directions. These boundaries are situated nearthe core of the rough crystals.

Growth zoning was seen at 10× magnificationin some stones as multiple bands that showed vari-able tones and spacing (figure 27). Such growth zon-

TABLE 1. Properties of benitoite.a

Composition BaTiSi3O9

Color Colorless, and blue to violetish blue(commonly zoned); rarely pink.Heat treatment may (rarely) causecolorless → orange hue.Light to medium dark in tone, typicallystrong in saturation.

Pleochroism Strongly dichroic; colorless (ω) andblue to violetish blue (ε)

Clarity Translucent to transparentRefractive indices nω = 1.757, nε = 1.804b

nω = 1.757–1.759*, nε = 1.802–1.805*Birefringence 0.047b; 0.043–0.047*Optic character Uniaxial positiveSpecific gravity 3.65 (measured); 3.68 (calculated)c

3.65–3.80 (measured, this study)Dispersion ε = 0.039, ω = 0.046d (diamond is 0.044)Hardness 6–6.5c

Polish luster Vitreous to sub-adamantinee

Fracture luster Vitreous on conchoidal to unevensurfacese

Cleavage NoneToughness Fair; brittleUV FluorescenceShort-wave (254 nm) All stones––moderately chalky,

strong blueLong-wave (365 nm) Blue stones––inert

Pale blue to colorless stones––moderately chalky, dull red

Cathodo-luminescence Intense bluef

Morphology Hexagonal system; crystals are tri-angular, flattened on the c-axis, andpyramidal or tabular

Inclusions Minerals (actinolite-tremolite,aegirine-augite, diopside, seranditepseudomorphsg, neptuniteg,joaquiniteg, albiteg, apatiteg,djurleiteg), “fingerprints,”h fractures,two-phaseh

Typical size range Cut stones, 10 points to 1 ct;rough crystals, 1 cm in diameter

Largest cut stone 15.42 ctStabilitye Sensitive to rapid temperature changes

and ultrasonic vibration. Insoluble inhydrochloric and sulfuric acids. Easilyattacked by hydrofluoric acid.

May be Sapphire, tanzanite (depending onconfused with orientation), blue diamond, iolite,

blue tourmaline, blue spinel, sap-phirine, blue zircon

aProperties as obtained in this study unless otherwise noted.bWebster (1994). cAnthony et al. (1995). dPayne (1939). eGemologicalInstitute of America (1993). fLaird and Albee (1972). gWise and Gill (1977).hGübelin and Koivula (1986).*Values obtained in this study from measurements taken on the tables offaceted stones.

Figure 25. Lint-like needles of amphibole (probablyactinolite-tremolite) are the most frequentlyencountered mineral inclusions in benitoite. Dark-field illumination, magnified 20×; photomicro-graph by John I. Koivula.


ing is much less regular or apparent than that typi-cally seen in corundum. Less commonly than pla-nar growth zoning, our samples showed irregular,patchy, diffuse areas of color.

A distinctive characteristic of benitoite, notedin more than half the samples examined for thisstudy, were sharp planar features that closelyresemble the internal graining seen in some dia-monds. This “graining” appeared as shadow-likeplanes that, at 10× magnification, were commonlyseen along the boundaries of color zones. It alsotruncated color zones at oblique angles (again, seefigure 27). The graining was almost always uniformand straight, but in some stones it was step-like orjagged. No surface graining or other polishing irreg-ularities were noted where the internal grainingreached the surface of the stone.

Chemistry and Spectroscopy. A tabulation of previ-ous data and recently acquired compositions of ben-itoite is shown in table 2. No new analyses wereobtained for this study, because routine electronmicroprobe analyses have shown a constant compo-sition, and the data reported by Laird and Albee(1972) are considered very reliable (G. Rossman,pers. comm., 1997). The measured amounts of Ba,Ti, and Si are remarkably constant, regardless ofcolor or locality, and fall near the ideal values thatare calculated from the chemical formula. The

greatest chemical variation in benitoite was report-ed by Gross et al. (1965) for samples from the BigCreek–Rush Creek area in Fresno County,California, which contained up to 4.1% tin oxide. Azirconium-bearing blue benitoite from Japan wasfound to contain 1.77 wt.% ZrO2; and up to 1.51wt.% ZrO2 was measured in a colorless benitoitefrom Arkansas (H. Barwood, pers. comm., 1997). Itis not surprising that Sn and Zr show the greatestvariations, since these elements form barium sili-cate minerals analogous to benitoite that are calledpabstite and bazirite, respectively (Hawthorne,1987). Other elements show minor variations, mostnotably Nb, Na, Ca, K, and Fe.

Numerous investigators have attempted to linkthe chemical composition of benitoite to its col-oration. Louderback (1907) initially attributed theblue color to traces of reduced Ti, but in 1909 hereported that no reduced Ti could be measured.Coleman (1957) suggested that trace impurities ofV, Nb, and Cu might cause the blue coloration.Laird and Albee (1972) noted slightly higher concen-trations of Fe in blue than in colorless benitoite.Burns et al. (1967) suggested that the coloration iscaused by crystal defects due to an oxygen deficien-cy. Burns (1970) suggested that the pleochroism ofbenitoite and other minerals might be caused byFe2+→Fe3+ charge transfer. Millage (1981) also pro-posed metal-metal charge transfer, such as

Figure 26. Two-phase inclusions are not commonlyseen in benitoite. The larger inclusions in thehealed fracture of this 0.95 ct stone contain vaporbubbles of variable shape and size. Darkfield illu-mination, magnified 25×; photomicrograph byJohn I. Koivula.

Figure 27. Planar color zoning is commonly visi-ble at some angles within benitoite. The colorzones appear curved in this 0.81 ct stone, becauseof refraction through adjacent facets. Shadow-likeinternal graining bounds and truncates the colorzones, and is visible as the stone is tilted fromside to side. Darkfield illumination, magnified15×; photomicrograph by John I. Koivula.


TABLE 2. Chemical compositions of benitoite from California, Arkansas, and Japan.a

Benitoite BenitoiteGem mineb Gem mineb Victorc Diamond Jod Diamond Jod Diamond Jod Ohmie Ohmie OhmieCalifornia California California Arkansas Arkansas Arkansas Japan Japan Japan

Oxide Colorless Blue Colorless Colorless Blue Pink Colorless Blue Blue(wt.%) (6) (6) (1) (3) (6) (3) (1) (1) (1)

SiO2 42.62 43.10 43.66 43.52 43.55 43.49 43.55 43.16 41.05TiO2 19.44 19.51 19.29 17.90 18.47 18.99 19.30 19.11 19.82ZrO2 0.108* 0.016* nd 1.16 0.33 0.08 nd nd 1.77Al2O3 0.04* 0.20* 0.30 0.29 0.22 0.21 0.18 0.10 ndFeO 0.01* 0.05* 0.01 0.15 0.04 0.01 0.081† 0.033† ndMnO nd nd 0.03 0.03 0.04 0.03 0.05 0.11 ndMgO trace* trace* nd nd nd nd 0.12 0.14 0.16CaO 0.006* 0.10* 0.11 nd nd nd 0.28 0.24 0.12BaO 37.27 37.23 36.48 36.70 36.21 36.36 34.82 36.04 36.78SrO 0.003* 0.004* nd nd nd nd nd nd ndNa2O 0.14 0.13 0.20 0.23 0.22 0.18 0.84 0.75 0.27K2O nd nd nd 0.01 0.20 0.13 0.44 0.21 ndNb2O5 0.20* <0.10* nd bdl 0.74 0.34 nd nd ndV2O5 0.002* 0.003* nd nd nd nd nd nd ndS nd nd nd nd nd nd 0.06 0.02 ––Cl nd nd nd nd nd nd 0.28 0.10 ––

Total 99.84 100.34 100.08 99.99 100.02 99.82 100.00 100.01 99.97

aComments: Ideal formula concentrations (wt.%): SiO2 43.59, TiO2 19.32, BaO 37.09 (Laird and Albee, 1972). Values determined by electron micro-probe, except: * – emission spectroscopy, and † – atomic absorption. Total iron is reported as FeO. Number of samples in parentheses.Looked for by Laird and Albee (1972), but not detected: Ag, As, Au, B, Be, Bi, Cd, Co, Cr, Cu, Ga, La, Mo, Mn, Nd, Ni, Pb, Pt, Sb, Sc, Sn, Ta, Th,TI, W, Y, Yb, and Zn. The abbreviation “nd” = not determined; “bdl” = below detection limit.bLaird and Albee (1972). cMillage (1981). dH. Barwood, pers. comm. (1997).eSakai and Akai (1994).

Fe2+→Ti4+ or Fe2+→Fe3+, as a coloring agent, butthen negated this possibility because these atomsare separated too far from one another in the beni-toite lattice. Instead, Millage attributed the blue col-oration to traces of Zr in the Ti site. The composi-tions of colorless and blue benitoite from Arkansasand the Benitoite Gem mine contradict this, how-ever, since more Zr is present in the colorless sam-ples than in the blue material (again, see table 2).

In this study, no spectral characteristics couldbe resolved in any of the samples tested, regardlessof color, with a desk-model type of spectroscope.Spectral data have been collected by means of aspectrophotometer for blue, pink, and heat-treatedorange benitoite (Burns et al., 1967; Rossman, 1997).Blue benitoite shows a broad peak at about 700 nm(figure 28), most of which appears in the near-infrared region. Infrared spectra have also been col-lected, but the data in this region are inconsistentand of little use (G. Rossman, pers. comm., 1997). Inspite of much effort to interpret the spectral data,the cause of color in benitoite remains elusive.

Synthesis and Scientific Use. Synthetic benitoitehas been successfully grown in the laboratory (Rase

and Roy, 1955a and b; Hinthorne, 1974; Christopheet al., 1980), but only in minute crystals that arecolorless and too small to facet. Natural benitoitehas an important industrial application in electron

500 1000 1500 2000

ABSO

RPTI

ON

WAVELENGTH (nm)

Figure 28. The absorption spectra for blue beni-toite from the Benitoite Gem mine, shown here inthe range from UV to near-infrared at orienta-tions parallel (red line) and perpendicular (blueline) to the c-axis, do not provide any clues to thecause of color in this material. Courtesy ofGeorge R. Rossman.


microprobes: Because it fluoresces strongly to anelectron beam, benitoite is used to align and adjustthe beam size. Benitoite is also employed as an ana-lytical standard for Ba and Ti, because of the con-stancy of its chemical composition. Webster (1994)suggested that benitoite could be used as a knownstandard for measuring dispersion in gemstones.

Identification. Benitoite may be distinguished fromsapphire by its greater birefringence, strong disper-sion, and its strong blue-and-colorless pleochroism.When set in a piece of jewelry, benitoite also willabrade more easily than sapphire. Natural and irra-diated blue diamonds can also give the appearanceof the lighter shades of benitoite because of similari-ties in color and dispersion, but the separation iseasily made because diamond is singly refractiveand has a much higher R.I. Benitoite can be distin-guished from tanzanite on the basis of pleochroism,since benitoite lacks the purple trichroic compo-nent that is almost always apparent at some anglewithin tanzanite. In addition to the differences ingemological properties, sapphirine is too dark to beconfused with benitoite, and zircon does not occurin the same hue. Iolite and blue tourmaline havemuch lower dispersion, refractive indices, and bire-fringence than benitoite. As described previously,benitoite also commonly has distinctive growth-zoning patterns and lint-like mineral inclusions.

CUTTINGBenitoite is one of the easiest gemstones to facet.There are no directional weaknesses (e.g., cleavage).At 6.0–6.5 on the Mohs scale, benitoite is about ashard as tanzanite and peridot, harder than sunstoneand other feldspars, and softer than quartz and tour-maline. Benitoite can be oriented so that color zon-ing is rarely evident in the face-up direction (figure29). Since the dichroic colors of blue benitoite areblue and colorless, there are no ancillary hues tomix through optical orientation. As a result, thealignment of the rough crystal for faceting requiresonly the placement of the desired color within theculet of the finished stone. In most cases, however,this is not important, since blue coloration is evi-dent regardless of the direction in which the stoneis oriented.

One of the authors (MG) has cut thousands ofbenitoites. Because of the typically small finishedsizes, faceting can generally be accomplished usinga single lap––in most cases, 1200 grit. Polishing is sorapid that the lapidary has to be careful not toenlarge a facet by overpolishing. The best grit is 0.04micron aluminum oxide, generally known as LindeB, which is a hundred times finer than the morecommon Linde A. It works very well on a tin ortin/lead polishing lap. The fineness of the powderhelps eliminate the polishing lines that are typicallyimparted to softer gemstones.

Figure 29. This 1.16 ctmodified pear shapebenitoite mimics theshape of the perfect

1-cm-wide rough crys-tal; both are from theBenitoite Gem mine.

The specimen wasrecovered from the

upper portion of themain pit by the pres-

ent owners. It is one ofonly three found inwhich the benitoite

apparently crystallizedcontemporaneouslywith albite crystals.

From the collection ofMichael M. Scott;photo © Harold &

Erica Van Pelt.


The critical angle of benitoite is 34.7°. Success-ful makes have crown angles that range from 30° to40°, with optimum results obtained at 38°. Thepavilion angle range is narrower, from 40° to 42°with the optimum at 41°. Benitoite is amenable toall styles and makes (see again, figure 22), but ittends to show maximum dispersion when finishedas a round brilliant or a trilliant.

CONCLUSIONIt is likely that future production of faceting-qualitybenitoite will come solely from the Benitoite Gemmine. The recent discovery of additional mineral-ized material down slope from the historic open pitis highly encouraging for the stability of price andfuture supply. Although rare, faceted stones andmineral specimens of benitoite are commerciallyavailable, and mining is ongoing. The formation ofrelatively large, locally abundant, gem-quality crys-tals at the Benitoite Gem mine resulted from acombination of unusual geologic processes. The dis-tinctive gemological properties of benitoite make itreadily identifiable from other gem materials.

Acknowledgments: The authors thank W. C. Forrestand E. (Buzz) Gray for their cooperation and support,and the Kennecott Exploration Company for permis-sion to use proprietary data. This manuscript benefit-ed from constructive reviews by three referees. Fortechnical assistance, we are grateful to Harold andErica Van Pelt, John Koivula, Tino Hammid, andJeffrey Scovil (photography); and Dr. Frank C.Hawthorne, University of Manitoba, Canada (elec-tron microprobe analysis). Shane McClure, managerof Identification Services at the West Coast GIAGem Trade Laboratory, provided specific gravitydata, as well as measurements of the refractiveindices with a cubic zirconia refractometer. Dr. W. S.Wise, of the University of California at SantaBarbara, and Dr. G. R. Rossman, of the CaliforniaInstitute of Technology, are thanked for several use-ful discussions. Preliminary versions of figures 4 and7 were drafted by Sue Luescher of Graphic Harmonyin Reno, Nevada. We thank L. Junnila, C. Stolberg,and S. Cisneros for information and permission tovisit their claims.

REFERENCESAlfors J.T., Stinson M.C., Mathews R.A., Pabst A. (1965) Seven

new barium minerals from eastern Fresno County, California.AmericanMineralogist, Vol. 50, pp. 314–340.

Anten J. (1928) Sur la composition lithologiques des psammites

du Condroz. Société Géologique de Belgique, Annales, Vol.51, pp. B330–331.

Anthony J.W., Bideaux R.A., Bladh K.W., Nichols M.C. (1995)Handbook of Mineralogy, Vol. 2: Silica, Silicates; Part 1.Mineral Data Publishing, Tucson, AZ.

Arnold R. (1908) Notes on the occurrence of the recentlydescribed gem mineral, benitoite. Science, Vol. 27, No. 686,pp. 312–314.

Austin D.H. (1988) The Benitoite Story. Independently pub-lished, 80 pp.

Bancroft P. (1984) Gem and Crystal Treasures. WesternEnterprises, Fallbrook, CA, pp. 94–95.

Barwood H. (1995) Benitoite and joaquinite in Arkansas. MineralNews, Vol. 11, No. 5, pp. 2, 5.

Bradley W.W., Huguenin E., Logan C.A., Waring C.A. (1917)Gems. Mines and Mineral Resources of the Counties ofMonterey, San Benito, San Luis Obispo, Santa Barbara,Ventura. California State Mining Bureau, San Francisco, CA,pp. 42–43.

Burns R.G. (1970) Mineralogical Applications of Crystal FieldTheory. Cambridge University Press, Cambridge, England.

Burns R.G., Clark R.H., Fyfe W.S. (1967) Crystal field theory andsome geochemical applications. In A.P. Vinogradov, Ed.,Chemistry of the Earth’s Crust, Vol. 2, Israel Program forScientific Translations, Jerusalem, pp. 93–112.

Chihara K., Komatsu M., Mizota T. (1974) A joaquinite-like min-eral from Ohmi, Niigata Prefecture, central Japan.Mineralogical Journal, Vol. 7, No. 4, pp. 395–399.

Christophe M., Gouet G., Wyart J. (1980) Synthése hydrother-male de la bénitoite. Bulletin de Minéralogie, Vol. 103, pp.118–119.

Chromy B. (1969) Pink benitoite. Gems and Minerals, No. 279,p. 32.

Coleman R.G. (1957) Mineralogy and petrology of the New Idriadistrict, California. Ph.D. dissertation, Stanford University,Stanford, CA.

Coleman R.G. (1961) Jadeite deposits of the Clear Creek area,New Idria district, San Benito County, California. Journal ofPetrology, Vol. 2, pp. 209–247.

Couch O. (1961) The Benitoite Story. Independently published,10 pp.

Dibblee T.W. Jr. (1979) Geologic map of the central Diablo Rangebetween Hollister and New Idria, San Benito, Merced andFresno Counties, California. U.S. Geological Survey OpenFile Map 79–358, scale 1:125,000.

Eckel E.B., Myers W.B. (1946) Quicksilver deposits of the NewIdria district, San Benito and Fresno counties, California.California Journal of Mines and Geology, Vol. 42, pp. 81–124.

Frazier S., Frazier A. (1990a) A rare bit of history. LapidaryJournal, Vol. 44, No. 8, pp. 46–58.

Frazier S., Frazier A. (1990b) A benitoite bibliography. LapidaryJournal, Vol. 44, No. 8, pp. 61–68.

Gemological Institute of America (1993) Gem Reference Guide.Gemological Institute of America, Santa Monica, California.

Gray M. (1986) Benitoite mining today. Canadian Gemmologist,Vol. 7, No. 3, pp. 82–83.

Gray M. (1992) Recent developments at the Benitoite mine.Canadian Gemmologist, Vol. 13, No. 4, pp. 118–119.

Gross E.B., Wainwright J.E.N., Evans B.E. (1965) Pabstite, the tinanalogue of benitoite. American Mineralogist, Vol. 50, pp.1164–1169.

Gübelin E.J., Koivula J.I. (1986) Photoatlas of Inclusions inGemstones, ABC Edition, Zurich, Switzerland.

Hawthorne F.C. (1987) The crystal chemistry of the benitoitegroup minerals and structural relations in (Si3O9) ring struc-tures. Neues Jahrbuch für Mineralogie, Monatshefte, Vol. 1,pp. 16–30.

Hessel J.F.C. (1830) Krystallometrie. Reprinted as: Nr. 88.Ostwald’s Klassiker der exakten Wissenschaften (1897),


Wilhelm Engelmann, Leipzig.Hinthorne J.R. (1974) The origin of sanbornite and related miner-

als. Ph.D. dissertation, University of California at SantaBarbara.

Hopson C.A., Mattinson J.M., Pessagno E.A. (1981) Coast Rangeophiolite, western California. In W.G. Ernst., Ed., TheGeotectonic Development of California, Prentice Hall,Englewood Cliffs, NJ, pp. 418–510.

Komatsu M., Chiara K., Mizota T. (1973) A new strontium-tita-nium hydrous silicate mineral from Ohmi, NiigataPrefecture, central Japan. Mineralogical Journal, Vol. 7, No. 3,pp. 298–301.

Laird J., Albee L. (1972) Chemical composition and physical,optical, and structural properties of benitoite, neptunite, andjoaquinite.AmericanMineralogist, Vol. 57, pp. 85–102.

Laurs B.M. (1995) Progress report: Benitoite exploration in theNew Idria mining district, San Benito County, California.Report submitted to the Kennecott Exploration Company,Reno, NV.

Lee D.E., Thomas H.H., Marvin R.F., Coleman R.G. (1964)Isotopic ages of glaucophane schists from the area ofCazadero, California. U.S. Geological Survey ProfessionalPaper 475, pp. D105–D107.

Lonsdale J.T., Metz M.S., Halbouty M.T. (1931) The petrographiccharacters of some Eocene sands from southwest Texas.Journal of Sedimentary Petrology, Vol. 1, pp. 73–81.

Louderback G.D. (1907) Benitoite, a new California gem mineral.University of California, Department of Geological SciencesBulletin, Vol. 5, pp. 149–153.

Louderback G.D. (1909) Benitoite, its paragenesis and mode ofoccurrence. University of California, Department ofGeological Sciences Bulletin, Vol. 5, pp. 331–380.

Marcher G.H. (1939) Story of a gemstone christening. TheMineralogist, pp. 200, 217–219.

Millage A.H. (1981) Mineralogy of the Victor claim, New Idriadistrict, San Benito County, California. M.S. thesis, StanfordUniversity, Stanford, CA.

Mitchell R.K. (1980) The fluorescence of benitoite. Journal ofGemmology, Vol. 17, No. 3, p. 149.

Moller W.P. (1996) A new discovery of benitoite. MineralogicalSociety of Southern California Bulletin, September, pp. 3–4.

Payne C.J. (1939) Dispersions of some rarer gemstones.Gemmologist, Vol. 9, pp. 33–35.

Petrov A. (1995) More benitoite locality information, anothernew one and another discredited. Mineral News, Vol. 11, No.8, p. 9.

Rase D.E., Roy R. (1955a) Phase equilibria in the system BaTiO3-

SiO2. Journal of the American Ceramic Society, Vol. 38, pp.389–395.

Rase D.E., Roy R. (1955b) On the stability and hydrothermal syn-thesis of benitoite. American Mineralogist, Vol. 40, pp.542–544.

Ream L.R. (1996) Benitoite from the Junnila mine. MineralNews, Vol. 12, No. 3, p. 3.

Reed R.D., Bailey J.P. (1927) Subsurface correlation by means ofheavy minerals. Bulletin of the American Association ofPetroleumGeologists, Vol. 11, pp. 359–368.

Rohtert W.R. (1994) Preliminary geologic map of the BenitoiteGem mine. Report submitted to the Kennecott ExplorationCompany, Reno, NV.

Rossman G.R. (1997) Benitoite visible spectra (350–1100 nm).Web page, http://minerals.gps.caltech.edu/files/benitoit/.

Sakai M., Akai J. (1994) Strontium, barium and titanium-bearingminerals and their host rocks from Ohmi, Japan. ScientificReports of the Niigata University, Series E: Geology andMineralogy, Vol. 9, pp. 97–118.

Sinkankas J. (1976) Gemstones of North America, Volume II.Van Nostrand, Princeton, NJ, pp. 280–288.

Smith A.E. (1995) The reported benitoite occurrence inTexas––doubtful?Mineral News, Vol. 11, No. 5, p. 5.

Sterrett D.B. (1908) Benitoite. U.S. Bureau of Mines Minerals1907 Yearbook, Part 2: Non-metallic Products. Washington,DC, pp. 798–799.

Sterrett D.B. (1911) Benitoite. U.S. Bureau of Mines Minerals1909 Yearbook, Part 2: Non-metals. Washington, DC, pp.742–748.

Van Baalen M.R. (1993) Titanium mobility in metamorphic sys-tems: A review.Chemical Geology, Vol. 110, pp. 233–249.

Van Baalen M.R. (1995) The New Idria serpentinite. Ph.D. disser-tation, Harvard University, Cambridge, MA.

Webster R. (1994) Gems: Their Sources, Descriptions, andIdentification, 5th ed. Rev. by P.G. Read., Butterworth Hein-emann, London.

Wise W.S. (1982) Strontiojoaquinite and bario-orthojoaquinite:Two new members of the joaquinite group. AmericanMineralogist, Vol. 67, pp. 809–816.

Wise W.S., Gill R.H. (1977) Minerals of the Benitoite Gem mine.Mineralogical Record, Vol. 8, pp. 7–16.

Wise W.S., Moller W.P. (1995) Geology and mineralogy of theNew Idria district, San Benito and Fresno Counties,California. Rocks andMinerals, Vol. 70, No. 1, pp. 27–35.

Worner H.K., Mitchell R.W., Eds. (1982)Minerals of Broken Hill.Australian Mining & Smelting Ltd., Melbourne, pp. 54, 87,and 202.

188 Tairus Hydrothermal Synthetic Sapphires GEMS & GEMOLOGY Fall 1997

ABOUT THE AUTHORS

Dr. Thomas is a senior researcher, and Mr.Maltsev is a technological researcher, at Tairus (ajoint venture between the Siberian Branch of theRussian Academy of Sciences and Pinky TradingCo. Ltd. of Bangkok, Thailand). Dr. Mashkovtsevis a senior researcher, and Dr. Smirnov is aresearch scientist, at the Siberian GemologicalCenter, Novosibirsk

Gems & Gemology, Vol. 33, No. 3, pp. 188–202.

© 1997 Gemological Institute of America

he history of hydrothermally grown syntheticcorundum began nearly 40 years ago, whenLaudise and Ballman (1958) produced the first col-

orless synthetic sapphire. Their technique, when modified,also can be used to grow rubies (Kuznetsov and Shternberg,1967). Until recently, however, attempts to grow blue sap-phire by the hydrothermal method had failed. This problemis significant because flux-grown and flame-fusion syntheticsapphire crystals typically have uneven color distribution. Inblue flame-fusion material, an intense blue hue is usuallyconcentrated in thin outer zones of boules, while most ofthe bodycolor is much lighter. In blue flux-grown syntheticsapphires, color inhomogeneity is apparent as a sequence ofblue and colorless angular growth zones that conform to thecrystal shape. Such color inhomogeneity makes cutting dif-ficult and greatly decreases the yield of faceted material.

To address this market gap, the authors investigated theuse of hydrothermal growth techniques to produce bluecrystals that would be free from the defects of synthetic sap-phire grown by other methods. This problem wasapproached in two ways: (1) by doping hydrothermal syn-thetic sapphire with iron (Fe2+) and titanium (Ti4+) ions (asproposed by, for example, Burns, 1981); and (2) by usingother doping impurities that could reproduce the blue colorof sapphire. Research into the development of a commercialtechnique by the first approach is still ongoing. However,the second approach has led to success: The use of nickel(Ni2+) as a dopant has resulted in crystals of “sky” blue syn-thetic sapphire, from which thousands of faceted stones aslarge as 4 ct have been produced by the Tairus joint ventureand marketed internationally since 1995.

The next step in the development of this technique wasto produce sapphires of different colors by varying the con-centrations of Ni2+, Ni3+, and chromium (Cr3+). At thistime, the Tairus research group has created the technology

TResearchers of the Tairus joint venture havedeveloped the technology for the hydrother-mal growth of gem-quality crystals of syn-thetic sapphire. They have achieved a broadspectrum of attractive colors by varying theconcentrations of Ni2+, Ni3+, and Cr3+.Comprehensive testing of 18 faceted synthet-ic sapphires and 10 synthetic crystals, repre-senting the range of colors, revealed a typicalset of internal features that are specific tohydrothermally grown crystals (microscopy),as well as the presence of OH– and variouscarbon-oxygen groups in the sapphire struc-ture (infrared spectroscopy). The reactions ofthe test samples to ultraviolet radiation andvisible light were also found to be useful indistinguishing the Tairus synthetic sapphiresfrom their natural counterparts and othersynthetics.

TAIRUS HYDROTHERMALSYNTHETIC SAPPHIRES DOPEDWITH NICKEL AND CHROMIUM

By Victor G. Thomas, Rudolf I. Mashkovtsev, Sergey Z. Smirnov, and Vadim S. Maltsev

Tairus Hydrothermal Synthetic Sapphires GEMS & GEMOLOGY Fall 1997 189

to grow commercial quantities of sapphires in blueand a wide range of other colors (see, e.g., figure 1).This article presents the results of our research intothe growth of hydrothermal synthetic sapphires andtheir characterization, including the chemistry,gemological properties, and distinctive internal fea-tures of these new synthetics.

GROWTH TECHNIQUESThe autoclave arrangement used to grow Tairushydrothermal synthetic sapphire is illustrated in fig-ure 2. The growth process is carried out in a her-metically sealed “floating” gold ampoule with avolume of almost 175 ml inside the autoclave, atemperature of 600°C, and a pressure of about 2000bar. Growth takes place when the initial syntheticcolorless sapphire feed is transferred and crystal-lized onto seeds, via a vertical temperature gradientof 60° to 70°C, through a carbonate solution with acomplex composition (the details of which are pro-

prietary). Seed plates oriented parallel to the hexago-nal prism {1010} are used to grow cuttable crystals.For research purposes, we also grew two crystals ona spherical seed to investigate the ideal arrangementof the crystal faces. All of the seeds in the samplesused for this study were cut from Czochralski-grown colorless synthetic sapphire. The dopant(NiO and, for certain colors, Cr2O3) is introducedvia unsealed capsules placed at the bottom of theampoule. The concentrations of Ni and Cr in thegrowing crystal are controlled by the diameter ofthe opening in these capsules.

An oxygen-containing buffer plays a significantrole in the sapphire growth process. It is a powderedreagent that is placed at the bottom of eachampoule. This substance governs the valence stateof Ni in the solution, so that sapphires with differ-ent Ni2+:Ni3+ ratios can be grown. Various metalsand their oxides (none of which produces any colorin the synthetic sapphire) have been used for this

Figure 1. Tairus expertshave produced hydro-

thermal synthetic sap-phires in a broad rangeof colors, as illustratedhere. These Tairus syn-thetic sapphires range

from 1.00 to 4.74 ct.Photo © GIA and Tino

Hammid.

–


buffer (e.g., coupled Cu-Cu2O, Cu2O-CuO, etc. [seetable 1]). Synthetic sapphire crystals as large as 7 ×2.5 × 2 cm have been grown in a month, with agrowth rate of almost 0.15 mm per day.

MATERIALS AND METHODSTwo crystals of each color (10 pairs total) weregrown under uniform conditions as described above.One crystal in each pair was used to produce facetedgems (from one to five faceted stones, each weigh-ing approximately 1 ct). The other was cut intoplates or used whole for crystallographic measure-ments. Detailed crystallographic measurementswere also made on the two greenish blue crystals(each about 0.8 cm across) that were grown onspherical seeds. (These latter crystals were grownboth to illustrate the ideal arrangement of crystalfaces for corundum produced under the experimen-tal conditions described above, and because it waseasier to obtain accurate goniometric measure-ments on them.). We selected 18 faceted samples(round and oval mixed cuts, weighing about 1 cteach) that ranged from very light pink throughorange, yellow, and green to blue (see, e.g., figure 3)for gemological testing. More than 20 polished (on

both sides) plates were used for spectroscopy, inclu-sion study, and microprobe analysis.

Standard gemological testing was carried out atthe Siberian Gemological Center, Novosibirsk. Theface-up colors of the 18 faceted synthetic sapphireswere determined with a standard daylight-equiva-lent light source (the fluorescent lamp of a GIAGEM Instruments Gemolite microscope) and anordinary incandescent lamp. For each faceted sam-ple, the color was compared to standards from theGIA GemSet and described in terms of hue, tone,and saturation. Specific gravity was determinedhydrostatically. Refractive indices were measuredwith a GIA GEM Duplex II refractometer and a fil-tered, near-monochromatic, sodium-equivalentlight source. To examine internal features andpolarization behavior, we used a GIA GEMIlluminator polariscope and gemological micro-scope with various types of illumination.

Detailed observations of inclusions, color zon-ing, and growth sectors were made on crystal platesof all 10 color varieties with a Zeiss Axiolab polariz-ing microscope. Some samples were also immersedin methylene iodide for microscopic examination.Appearance in visible light was observed with theincandescent illumination provided by a GIA GEMInstruments spectroscope base, and UV lumines-cence was observed with the aid of a GIA GEMInstruments ultraviolet cabinet. Samples with weakluminescence were examined in a completely dark-ened room. Dichroism was determined with a cal-cite dichroscope and a GIA GEM InstrumentsFiberLite 150 fiber-optic illuminator.

The arrangement of crystal faces was studied bymeans of a Zeiss ZRG-3 two-circle goniometer. The

Figure 2. Hydrothermal synthetic sapphires aregrown by Tairus scientists in an autoclave that mea-sures 60 cm high and 8 cm in diameter. The compo-nents are as follows: (1) lid; (2) push nut; (3) auto-clave body; (4) seal ring; (5) gold ampoule; (6) syn-thetic sapphire seeds; (7) diaphragm; (8) crushed syn-thetic colorless corundum charge; (9) capsules withNi/Cr oxides; (10) powdered oxygen-containingbuffer. Crystal growth proceeds via dissolution of thecorundum charge (8) in a hydrothermal solution inthe lower (i.e., hotter) part of the autoclave, followedby convective transfer of the saturated solution tothe upper (i.e., cooler) part of the autoclave, wherethe synthetic sapphire crystallizes on the seeds (6).The autoclave is heated from the bottom and theside in a special furnace.


compositions (including major and trace elements)of 11 samples representative of the different colorswere determined with a Camebax-Micro electronmicroprobe. We observed visible-range spectra ofthe 18 faceted stones using a GIA GEM InstrumentsPrism 1000 spectroscope. In addition, UV-visiblespectra were obtained with a Zeiss Specord M40spectrophotometer, and infrared spectra were mea-sured with Zeiss Specord M80 and SF-20 spec-trophotometers, on 1-mm-thick plates. A processorbuilt into the M40 locates the position of absorptionband peaks. Color coordinates calculated from theunpolarized absorption spectra were plotted on aCIE chromaticity diagram, to compare color rangesof synthetic sapphires doped with Ni2+, Ni3+, Cr3+,Ni2+-Cr3+, and Ni3+-Cr3+. The accuracy of these cal-culations was checked by a catalogue of coloredglass (Catalogue . . ., 1967). One hydrothermal syn-thetic ruby (donated by Alexander Dokukin) andone flame-fusion synthetic ruby were used for com-

parison in the CIE diagram as examples of corun-dum doped with Cr3+ only.

Some of the very light greenish blue sapphiresdoped with Ni2+ were gamma irradiated (60Cosource, dose 0.1–1 Mrad) to convert part of the Ni2+

to Ni3+, which changed the color to yellow. Weinvestigated the ultraviolet and visible spectra onthese treated samples.

RESULTS AND DISCUSSIONCrystal Morphology. Crystals grown on flat andspherical seeds are shown in figure 4. These crystalshave several sets of major flat faces (i.e., forms) con-sisting of the hexagonal prism a {1120}, dipyramid n{2243}, and rhombohedron r {1011}. These forms arecommon on natural corundum crystals (Gold-schmidt, 1916; Yakubova, 1965). In addition, weobserved minor faces of the following forms: rhom-bohedra f {1014} and d {1012}, as well as rough flatsurfaces q {0.1.1.11} and c {0001} (again, see figure 4).

Figure 3. Although this photoshows the full color range (includ-

ing colorless and red) of thehydrothermal synthetic corundumproduced by Tairus researchers todate (here, 1.11–4.74 ct), the pre-

sent study focused on those pink,yellow, and green-to-blue synthet-

ic sapphires produced by dopingwith Ni2+, Ni3+, and/or Cr3+.Photo by Evgeny Krayushin.

TABLE 1. Colors and partial chemistry of the Tairus synthetic sapphires.a

Concentration (wt.%)

Sample Color Cr2O3 NiOb FeO MnO TiO2Oxygen buffer

1 Greenish blue bdlc 0.23 tr tr tr Cu-Cu2O2 Greenish blue 0.15 0.24 tr tr tr Cu-Cu2O3 Greenish blue 0.26 0.34 tr tr tr Cu-Cu2O4 Light reddish purple 0.07 0.04 tr tr tr Cu-Cu2O5 Pink 0.07 bdl tr tr tr Cu-Cu2O6 “Padparadscha” 0.08 bdl tr tr tr PbO-Pb3O47 Yellow trd tr tr tr tr PbO-Pb3O48 Light greenish yellowe bdl tr tr tr tr Cu-Cu2O9 Green bdl 0.35 tr tr tr Cu2O-CuO10 Blue-green bdl 0.23 tr tr tr not controlled11 Colorless bdl bdl tr tr tr not controlled

a Analyses by electron microprobe.bTotal NiO+Ni2O3.cbdl –– below the detection limit = <0.01 wt.%.dtr –– trace = about 0.02 wt.% for each element.eThis sample was gamma-irradiated to produce the yellow color (details in Materials and Methods section of text).

–

––

–

––


With the exception of c, these forms are not typicalof natural corundum crystals. The Miller indices ofthe f and q faces were calculated from goniometricmeasurements, but we could not find a reference toeither face in the available literature on corundumcrystal morphology.

The habit of the samples grown on sphericalseeds is transitional between some natural andsome flux-grown––e.g., Knischka (Gübelin,1982)––crystals. By comparison, the habit of crystalsgrown on flat seeds is similar to that seen on spheri-cal seeds, but the overall shape of the crystals (e.g.,the relative sizes of the different faces) is distorted,because of the different crystallographic orientationand size of the seed plate. Crystals grown on seedplates showed r, n, and q as major faces, with asmoother surface than their equivalents on crystalswith a spherical seed. Minor faces include thehexagonal prism a and scalenohedral faces forwhich indices have not yet been determined.

Chemistry. The electron microprobe analyses (table1) indicate that the chemical composition of thenickel- and chromium-doped hydrothermal syn-thetic sapphires was about 99.4–99.9 wt.% Al2O3.Special attention was paid to measuring the possi-ble chromophores Cr, Ni, Ti, manganese (Mn), andFe. (Ti, Mn, and Fe are the components of autoclavesteel, so they could have entered the growing sap-phire through the possible unsealing of the ampouleduring the run.) All of the samples including thecolorless synthetic sapphire (sample 11; table 1),contain traces of Ti, Mn, and Fe. Since these ele-ments were present in the colorless sample, theyapparently do not affect the color at the concentra-tions present in the Tairus synthetic sapphires.

Optical Absorption Spectra and Color. OpticalAbsorption Spectra. Ni-doped sapphire crystalsgrown by the Verneuil method are known to con-tain trivalent nickel (Ni3+). To convert the Ni to adivalent state (Ni2+), synthetic sapphire has beenannealed in a reducing (hydrogen) atmosphere(Muller and Gunthard, 1966). Minomura andDrickamer (1961) produced Ni2+- and Ni3+-dopedsapphire by annealing pure Al2O3 crystals in NiOpowder. For the samples in our study, however, theNi2+:Ni3+ ratio was set in situ during the growthprocess by controlling the oxidation-reductionpotential (see Growth Techniques section), and itvaried widely.

The absorption spectrum (in unpolarized light)of Ni2+-doped greenish blue Tairus synthetic sap-phire (sample 1, table 1) includes three intensebands––at 377, 599, and 970 nm––and two weakbands, at 435 and 556 nm. In polarized light (figure5A), each intense band is resolved into twobands––that is, 370 and 385 nm, 598 and 613 nm,and 950 and 990 nm (The last two bands are notshown in figure 5A but were detected with the SF-20 spectrophotometer.) These results differ some-what from the positions reported earlier for Ni2+

absorption bands in synthetic sapphires grown byother methods (Minomura and Drickamer, 1961;Muller and Gunthard, 1966), which in turn differnoticeably from one another. In particular, lines at995 and 1020 nm were observed by Minomura andDrickamer, but not by Muller and Gunthard. Thesedifferences are understandable given the radicallydifferent methods of sapphire growth.

Green synthetic sapphire (sample 9, table 1)grows under relatively high oxidizing conditions, soa greater ratio of Ni3+ to Ni2+ is to be expected in a

Figure 4. The photo on the left shows two large(5 2́ cm) crystals of greenish blue synthetic sap-

phire that were grown on seed plates, and a small(about 0.8 cm) crystal grown on a spherical seed.

(Photo by Evgeny Krayushin.) The line illustrationon the right shows the actual crystal faces

obtained on a crystal grown on a spherical seed.


green crystal. The absorption spectrum of this sap-phire (figure 5B) is similar to a Ni2+ absorption spec-trum, but with intense absorption between 380 and420 nm, rapidly increasing toward the ultraviolet.

The spectra in figure 5C were derived by sub-tracting the spectra in figure 5A from those in figure5B: They correlate well with the reported Ni3+ ionabsorption spectrum (Muller and Gunthard, 1966;Boksha et al., 1970). Thus, we believe that the greencolor of the hydrothermal synthetic sapphires iscaused by the simultaneous absorption of Ni2+ andNi3+ ions. As will be shown below, the absorptioncharacterized by the spectrum in figure 5C will pro-duce a yellow color in sapphire. In practice, we havefound that by increasing the oxidation potentialfrom a Cu2O-CuO to a PbO-Pb3O4 buffer, we cangrow yellow sapphire (sample 7, table1) with a spec-trum similar to those in figure 5C.

Another way to produce yellow sapphire dopedwith Ni3+ is by gamma irradiation of Ni2+-dopedpale greenish blue sapphire. The absorption spectraof one of the samples that was gamma-irradiatedwith 0.5 Mrad (sample 8 in table1) also resemblethose in figure 5C. Moreover, following irradiation,only a weak decrease in Ni2+ absorption is observed,whereas Ni3+ absorption increases to become com-parable to Ni2+ absorption.

The addition of small amounts of Cr (up to 0.3wt.% Cr2O3) to Ni-containing sapphires (NiO>0.2wt.%) results in a greenish blue color, and the sim-ple superposition of Ni and Cr spectra (figure 6). Thespectrum of a greenish blue Ni2+-Cr3+-doped sap-phire is similar to that of ruby, with characteristicCr lines at 469, 475 nm, and 690 nm. The band at510 to 620 nm is a result of the superposition of Ni(580–620 nm) and Cr (510–620 nm) bands. Since thisfeature is clearly seen in a gemological spectroscope,it can be regarded as a good identifying feature ofTairus greenish blue Cr-containing sapphires.

Color Varieties. There are two ways to study anddescribe color. The first is the “comparison”method, whereby the color of the sample is simplycompared to a standard with known color. Thismethod is used in most gemological investigationsand for gemological testing. The second way is tocalculate color characteristics from the absorptionspectra of the gemstones. This method allows usnot only to describe the color of these synthetic sap-

Figure 5. These optical absorption spectra are of (A)greenish blue (sample 1, table 1) and (B) green (sam-ple 9, table 1) Tairus synthetic sapphires doped withNi2+ and Ni2+-Ni3+, respectively. Spectrum C wasproduced by subtracting spectrum A from spectrumB and represents a hypothetical yellow syntheticsapphire doped only with Ni3+. Blue = parallel tothe c-axis; red = perpendicular to the c-axis.

Figure 6. The unpolarized absorption spectra of (A)Ni2+-doped and (B) Ni2+-Cr3+–doped Tairus greenishblue synthetic sapphires are quite different from theabsorption spectrum of ordinary blue Verneuil-grown Fe2+ + Ti4+ synthetic sapphire (C). SpectrumC is similar to that of natural Fe2+ + Ti4+ sapphires.


phires, but also to compare them quantitatively tonatural sapphires and other synthetics. To quantifythe different colors of Tairus hydrothermal synthet-ic sapphires, we converted their absorption spectrato color coordinates in the CIE color system. TheCIE system is fully described elsewhere (Wright,1964), but a brief explanation of the chromaticitydiagram, which is integral to this system, will bemade here.

A set of coordinates was developed to expressthe relationship among all colors in the spectrumon the basis of a “standard observer” response (theaverage of observations by a large number of people)to different color stimuli. The chromatic portions ofthese measurements were plotted on an x-y coordi-nate graph to produce the chromaticity diagram (fig-ure 7). The skewed parabolic curve on this diagramis the location of saturated hue coordinates. Thepoints representing saturated blue and violet colorsare located in the left corner of the region, whilepoints of saturated red and orange are on the right.The curved top corresponds to green and yellow sat-urated colors. The area along the straight linebetween the violet and red corners is the location ofnonspectral purple colors. The point for white coloris located in the approximate center of the region;point C marks the white daylight source.

Thus, any hue and saturation can be represent-ed by a set of x-y coordinates and plotted along orwithin the chromaticity curve. The points shownon the curve are the dominant wavelengths (innanometers) of each saturated color. Along any radi-al line drawn between white color and a saturatedhue lie colors with decreasing saturation as oneapproaches the white color point.

The synthetic sapphires doped with Ni2+ (point1 in figure 7) have a greenish blue hue. The synthet-ic sapphires grown under oxidizing conditions andcolored predominantly by Ni3+ (point 7) are yellow.The position of this point is close to hypotheticalNi3+ sapphire (point 12), which is calculated fromthe spectra in figure 5C, as well as greenish yellowsapphire (point 8) produced by gamma irradiation.Variations in the ratio of Ni2+ to Ni3+ can produceall of the colors in the region bounded by points 1,8, 9, and 10.

When significant concentrations of Ni2+ (morethan 0.2 wt.% oxide) are present, Cr content hasvery little impact on the hue of the greenish bluesynthetic sapphires. The saturation is slightly low-ered (points 2 and 3) and the tone is darkened in therelatively Cr rich samples. When both NiO andCr2O3 are present at concentrations below 0.1 wt.%each, a light reddish purple (e.g., points 4 and 5) is

Figure 7. On this CIE chromaticity diagram,points are plotted for the Tairus synthetic sap-

phires (diamond shapes) and for natural andother synthetic sapphires (circles). Points 1

through 10 correspond to the Tairus samplenumbers from table 1. Sample 11 is a

hydrothermally grown ruby; sample 12 (pureNi3+) was calculated from the spectrum for

hypothetical yellow sapphire in figure 5C; andsample 13 is a flame-fusion synthetic ruby.

Points 14 through 18 were calculated from rep-resentative spectra of natural sapphires by

Platonov et al. (1984 ): 14 is a yellow sapphirefrom the Ural Mountains, Russia; 15 is a green-

ish blue sapphire from Australia; 16 is a palegreenish blue sapphire from the Ural Moun-tains; 17 is a deep blue sapphire that is also

from the Urals; and 18 is a deep blue sapphirefrom Burma. Point “C” represents a standarddaylight source. The triangular region defined

by the dotted lines illustrates the approximaterange of colors produced in Tairus syntheticsapphires by various amounts of Ni2+, Ni3+,

and Cr. Note the overlap in color between theTairus synthetic sapphires and the natural and

other synthetic samples.


produced. The addition of small amounts of Cr3+ toNi3+-doped synthetic sapphire shifts the hue fromyellow (such as for point 7) to the orange part of thespectrum, producing a “padparadscha” type pinkishorange color (point 6). Thus, addition of Cr to eitherNi2+ or Ni3+ shifts the color coordinates toward pur-ple, near the lines connecting point 13 with points 1and 12.

The combination of different amounts of Cr3+,Ni2+, and Ni3+ permits the production of syntheticsapphires in virtually any color in the vicinity of tri-angle 1-12-13 (shaded triangle on figure 7). The hueand saturation are strongly dependent on the con-centrations of these elements. Thus, although thetriangle 1-12-13 does not represent absolute bound-aries (e.g., points 3, 9, and 10 do lie away from the

triangle), it helps to illustrate the range of colorsrepresented by the samples grown in our experi-ments and discussed in this article. In addition, fig-ure 7 shows that the colors of some Tairus synthet-ic sapphires overlap the colors of their natural coun-terparts, as well as the colors of synthetic sapphiresgrown by other methods.

Gemological Characteristics. The properties forgreenish blue, pinkish orange (padparadscha), andyellow Tairus hydrothermal synthetic sapphires arereported in tables 2, 3, and 4, respectively. Samplestones are illustrated in figures 8, 9, and 10, respec-tively. The blue and pinkish orange samples arecompared to Chatham flux-grown synthetic sap-phires and flame-fusion samples from an unknown

TABLE 2. Properties of Tairus greenish blue (Ni2+ and Ni2+-Cr3+) synthetic sapphires and some of their naturaland other synthetic counterparts.

Tairus ChathamProperty hydrothermal Flux-grown Flame-fusion Natural, from Natural, from Natural, from Natural, from

synthetic synthetic synthetic Vietnam Australia China Sri Lanka

Color Moderately Blue Medium-dark Predominantly Very slightly Dark blue “Fine” bluestrong, violetish blue blue to greenish greenish blue with areas medium to (diffusion blue to bluish to greenish blue, of blue and dark greenish treated) green, medium- and near-color- bluish greenblue light to very dark less to extremely

dark “royal” blue

Color Even Uneven; Uneven; tiny Uneven; distinct Uneven; color Uneven; very Uneven; verydistribution strong color curved color color banding banding (color- strong color strong color(eye-visible) zoning banding (colorless/blue) less or pale zoning zoning

yellow/blue)

Pleochroism Weak (Ni2+- Strong; Strong; Distinct to strong; Not reported Strong Moderate Cr3+) to violetish violetish blue blue or violetish greenishstrong green blue to to grayish blue to blue-green blue to inky blue to blue greenish blue or yellow-green blue(Ni2+); in some bluecases (Ni2+-Cr3+), none

R.I. (nw) 1.768–1.769 1.770 1.768 1.769–1.772 1.769–1.772 1.769–1.771 1.768R.I. (ne) 1.759–1.761 1.762 1.759 1.760–1.764 1.761–1.763 1.761–1.762 1.760 Birefringence 0.007–0.010 0.008 0.009 0.008–0.009 0.008–0.009 0.008–0.009 0.008

S.G. 3.98–4.03 3.974–4.035 Not 3.99–4.02 3.99–4.02 3.99–4.02 4.00determined

Fluorescence Typically inert, Variable, Inert Inert Inert Inert Inertto LWUV but in some inert to

cases (with strong in high Cr) yellow to purplish red orange

Fluorescence Typically inert, Variable, Chalky light Inert Inert Inert Light blueto SWUV but in some inert to blue

cases (with strong in high Cr) very green, yellow weak purplish to orange red colors

Reference Present study Kane (1982) SGC Smith et al. Coldham (1985) Furui (1988) Gunawardene Databanka (1995) and Rupasinghe

(1986)

aStones tested in the Siberian Gemological Center (SGC).


manufacturer; all three hue types are compared tonatural sapphires from various sources worldwide.

The refractive indices of the Tairus syntheticgreenish blue samples were found to be near thelower limits of corundum (see, e.g., Feklichev, 1989).As shown in table 2, they overlap the R.I. valuesreported for Sri Lankan sapphires and flame-fusionsynthetic sapphires, and they are slightly below thevalues reported for the flux-grown synthetic sap-phires and natural sapphires from the other sources.The values obtained for the yellow and pinkishorange Tairus synthetic sapphires were somewhatlower still. Some of the pinkish orange Tairus sam-ples had values below the known range for corun-dum (see, e.g., Feklichev, 1989). The pink (nw=1.765,ne=1.758) and dark green (nw=1.768, ne=1.760) sam-ples––not reported in the tables––showed values thatwere between those of the yellow-to-orange (includ-ing padparadscha) and greenish blue samples. Therefractive indices of the pink samples were close tothose reported for Chatham flux-grown syntheticpink sapphires (nw=1.768, ne=1.759; Kammerling etal., 1994), and they fell within the ranges reported for

other hydrothermal lab-grown sapphires. As indicat-ed in tables 2–4, the values for birefringence and spe-cific gravity of the Tairus hydrothermal syntheticsapphires overlap those for these natural and othersynthetic sapphires.

Dichroism. Ni2+-doped synthetic sapphires (e.g.,sample 1 in table 1) showed a strong green-blue toblue dichroism. In the blue-green Ni2+-Ni3+–dopedsamples (e.g., sample 10), dichroism was strong vio-letish blue to blue-green; in the green (e.g., sample 9)Ni2+-Ni3+–doped samples, the colors were somewhatweaker brownish green and green. Increasing Cr con-tent in the greenish blue Ni2+-Cr3+–doped syntheticsapphires weakens their dichroism. In greenish blueCr-rich sapphires (e.g., sample 3), dichroism wasclearly visible only with a bright incandescent lightsource: A light purple tint was seen in one direction.This purple tint is due to chromium’s characteristicred luminescence to incandescent light. Even withan ordinary incandescent lamp, purple flashes can beseen in faceted samples viewed face-up. These fea-tures of greenish blue Cr-rich Tairus synthetic sap-

TABLE 3. Properties of Tairus pinkish orange “padparadscha” (Ni3+-Cr3+) synthetic sap-phires and some of their natural and other synthetic counterparts.

Tairus ChathamProperty hydrothermal Flux-grown Flame-fusion Natural, from

synthetic synthetic synthetic Sri Lanka

Color Medium-light Orange to Orange to yellowish Pinkish orangeamoderate reddish orange, orange, in medium-pinkish orange in moderate to light to medium-dark

vivid saturation tones, moderateto strong saturation

Color Even Color zoned Even Not reporteddistribution(eye-visible)Pleochroism Weak reddish Strong pink and Weak to moderate Orange-yellow and

orange and light orange to brown- orangy yellow and yellowish orangereddish orange ish yellow yellowish-orange

R.I. (nw) 1.765 1.770 1.768 1.768R.I. (ne) 1.757 1.762 1.759 1.760Birefringence 0.008 0.008 0.009 0.008S.G. 4.00 4.00 0.003 3.97 4.00Fluorescence Moderate Strong to very Weak orange-red Strong “apricot”bto LWUV orange strong orange-

red to yellowishorange

Fluorescence Weak orange Very weak to Inert Strong “apricot”bto SWUV weak in same

colors as LWReference Present study Kane (1982) SGC Databankc Gunawardene and

Rupasinghe (1986)

aCrowningshield (1983).bType of UV radiation was not specified.cStones tested in the Siberian Gemological Center (SGC).

±


phires can help to distinguish them from their natu-ral and other synthetic counterparts. The yellow sap-phires colored predominantly by Ni3+ showed nodichroism, whereas the addition of Cr to Ni3+ result-ed in weak dichroism in the pinkish orange (pad-paradscha) sapphires.

Reaction to Ultraviolet Radiation. Sapphire’s lumi-nescence to UV radiation strongly depends on Crcontent. The greenish blue Cr-rich synthetic sap-phires in this study fluoresced red to both short-wave (SW) and long-wave (LW) ultraviolet, whereasthe greenish blue Ni-rich samples were inert toboth. The fluorescence of the greenish blue Cr-richsamples was much weaker to SWUV than toLWUV. Red fluorescence has not been observed inmost natural (except some light blue Sri Lankan andsome rare dark blue stones [Gem Reference Guide,1995]) or other synthetic blue sapphires, so it can beused as a diagnostic feature for greenish blue Cr-richTairus synthetic sapphires.

“Padparadscha”-colored samples fluorescedmoderate orange to LWUV and weak orange toSWUV––reactions different from those of the naturaland other synthetic sapphires of similar color report-ed in table 3. However, the reaction of natural yel-low sapphires to UV radiation varies considerably, sothis feature cannot serve to separate Tairus syntheticsapphires from natural stones of this color.

Inclusions and Other Internal Features. Some of thesamples studied were internally flawless; otherscontained various amounts of inclusions. Theseinternal features consisted of fluid inclusions, crys-talline inclusions, and optical inhomogeneities.

Figure 9. The colors of these pinkish orange “pad-paradscha”(Ni3+-Cr3+) Tairus hydrothermal syn-thetic sapphires are similar to those seen in natureor in other synthetic sapphires. Photo © GIA andTino Hammid.

Figure 10. Yellow Tairus hydrothermal syntheticsapphires are colored predominantly by Ni3+. Photo© GIA and Tino Hammid.

Figure 8. Tairus researchers produced these green-ish blue hydrothermal synthetic sapphires by dop-ing them with Ni2+ and Ni3+ and, in some cases,Cr3+. Photo © GIA and Tino Hammid.


Two- and Three-Phase (Primarily Liquid) Inclu-sions. These could be divided into two major types.In both types, the inclusions consisted of two (gas-liquid) or three (a liquid, a gas, and a small crystal)phases. Two-phase inclusions were much morecommon than three-phase ones. Crystal phaseswere uncommon, but those present were easilyseen in polarized light.

The first type consists of large inclusions withan irregular, elongate form located along planes per-pendicular to the seed and running along planes par-allel to one pair of the prismatic a faces (figure 11).We believe that these are primary (see, e.g.,Roedder, 1984), formed when a small portion ofgrowth solution was trapped under a crystal faceduring growth. These inclusions were apparentlycaptured in flattened depressions on the surface ofthe growing crystal. Such depressions are typical ofhydrothermal crystals grown on a seed (Smirnov,1997). They were observed in only two facetedstones and a few crystals, which were blue, yellow,and colorless.

The second type are typical “fingerprints” (fig-ure 12), with regular to irregular patterns confined tohealed fractures, and therefore are secondary (see,e.g., Roedder, 1984). They also form perpendicular tothe seed plate and extend in both directions. “Finger-

prints” were found in many faceted stones and inmost of the crystals, in samples of almost all colors.

Crystalline Inclusions. Almost all of the green-to-blue samples examined contained crystalline inclu-sions. In some cases, these were single crystals thatcould be resolved only with high magnification andimmersion. Other samples contained abundantinclusions in the form of small, flake-like aggregatesof tiny crystals (figure 13); these inclusions could beclearly seen with a hand loupe. In some cases, whenexamined with bright, fiber-optic, oblique illumina-tion, most of these crystalline inclusions appearedto have a metallic luster. With high magnification,they appeared to be opaque isometric grains a fewmicrometers in diameter. Because the growth pro-cess used Cu-containing oxygen buffers, it is likelythat this metallic phase is copper.

Other crystals in these aggregates were transpar-ent, colorless or near-colorless, and isotropic with anoctahedral shape (figure 14). Because the individualcrystals were so small (less than about 12 mm), wecould not determine their composition with theavailable instrumentation. However, our recentexperiments have revealed that the addition of asmall amount of Mg to an alumina-bearing solutionleads to the formation of spinel (MgAl2O4) crystals.Trace amounts of Mg could occur as an impurity inthe charge used for the autoclave, so we suggest thatthese octahedra are spinel crystals.

Another type of transparent crystal was seen inthe green synthetic sapphires. These crystalsappeared as isolated platelets with regular hexago-nal outlines and no apparent color (figure 15). Theoptical properties of these crystals suggest that theyare gibbsite.

Optical Inhomogeneities and Color Zoning. Weobserved characteristic optical inhomogen-eities––swirl- and chevron-like patterns––in all thegreenish blue and green specimens, rough as well asfaceted (figure 16). These terms commonly refer towavy or angular features inside synthetic or naturalcrystals.

The swirl-like patterns resemble textures thatcan be seen in artificial glass. When the facetedsamples were examined face up, these patternsappeared as very faint, contrasting wavy lines.Chevron patterns were seen in the same samples,but at an angle to the table plane. These swirls andchevrons were much more visible in the darkersamples than in the lighter ones. They were less dis-tinct in the pinkish orange (padparadscha type) syn-

TABLE 4. Properties of Tairus yellow (Ni3+-Ni2+) syn-thetic sapphires and some of their natural and othersynthetic counterparts.

Tairus Property hydrothermal Natural, from Natural, from

synthetic Australia Sri Lanka

Color Yellow to Light yellow to Intense "golden" greenish strong “golden” yellow to light yellow, light yellow or pale yellowtones, strongsaturation

Color Even Occasionally Not reporteddistribution color zoned(eye-visible)Pleochroism None Not reported Orangy yellow

to grayish yellow

R.I. (nw) 1.767–1.765 1.774–1.772 1.768R.I. (ne) 1.758 1.765–1.763 1.760Birefringence 0.007–0.009 0.008–0.009 0.008S.G. 3.98 3.97–4.01 4.02Fluorescence Inert or weak Inert “Apricot” redto LWUV orangeFluorescence Inert or very Inert “Apricot” orangeto SWUV weak orangeReference Present Coldham Gunawardene

study (1985) and Rupasinghe(1986)


thetic sapphires and were not observed at all in theyellow or light pink samples.

Our studies revealed that the swirls andchevrons were typically seen in those samples thatwere grown on a seed plate with planar surfaces cutso as to be oriented at an angle to actual crystalfaces (i.e., on a nonsingular crystal surface). Theywere found to represent boundaries between growthsectors of subgrains that formed on the seed plate asthe crystal grew. Thus, the appearance of these pat-terns will be affected by the orientation of the tableof a faceted stone. They are distinctly different fromthe angular growth zones found in natural and influx-grown synthetic sapphires. Nor do they resem-ble the curved growth zoning usually seen inVerneuil-grown synthetic sapphires.

In several cross-sections of greenish blue Tairus

samples, we observed color zoning parallel to theplane of the seed plate. The portion of the crystalnearest the seed plate was yellow, while the restwas greenish blue. This thin (about 0.5 mm) sharpzone is usually cut away during pre-forming, so it isseen only rarely in faceted stones.

Infrared Spectroscopy. Hydrothermally grown crys-tals commonly contain water in different forms.This is also true for corundum, both natural (seeSmith, 1995, and references therein) and hydrother-mal synthetic stones (Peretti and Smith, 1993). Inrare cases, traces of water as OH– groups may befound in Verneuil-grown crystals (Volynets et al.,1974; Beran, 1991). The form of water incorporationinto sapphire and ruby has been the focus of manystudies, because it can help clarify the genesis of thegem. Today it is generally accepted, on the basis ofinfrared spectroscopy, that water as OH– groupsenters the corundum structure as a compensatingcharge due to either the doping of the corundumwith divalent cations or the formation of Al vacan-cies in the structure (Muller and Gunthard, 1966;Volynets et al., 1974; Beran, 1991). To examine thisissue and also further distinguish the Tairus syn-thetic sapphires from their natural and other syn-thetic counterparts, we conducted an IR-spectro-scopic study of the Tairus Ni- and Cr-dopedhydrothermal synthetic sapphires.

The IR spectra of the greenish blue and greensynthetic sapphires (samples 1 and 9, respectively,in table 1) are similar in the 4000-1600 cm-1 region(figure 17). This region is characterized by anintense wide band, with a maximum at 2980 cm-1; anarrow peak at 2750 cm-1; and five narrow peaks in

Figure 11. Irregular, elongate fluid-and-gas inclusionswere found in some blue, yellow, and colorless Tairussynthetic sapphires, located along crystallographicallyoriented planes. Photomicrograph by S. Z. Smirnov;magnified 100×.

Figure 12. “Fingerprints” are common in the Cr-and Ni-doped Tairus synthetic sapphires.Photomicrograph by S. Z. Smirnov; magnified 100×.

Figure 13. Flake-like crystalline inclusions are abun-dant in many of the green-to-blue Tairus hydrother-mal synthetic sapphires. Photomicrograph taken by S.Z. Smirnov, with sample between crossed polarizers,immersed in methylene iodide, magnified 25×.


the region between 2500 and 2000 cm-1. In the spec-trum of the greenish blue sapphire (A), there is ashoulder near 3300 cm-1. In the spectrum for thegreen sample (B), two narrow peaks are superim-posed on the wide band at 3400 and 3310 cm-1.

The presence of an intense wide band near 3000cm-1 is attibuted to OH– groups contained in alu-minum oxides. It is observed in the IR spectra ofdiaspore (AlOOH; Ryskin, 1974) and in the spectraof natural corundum containing divalent cations(Muller and Gunthard, 1966; Eigenmann andGunthard, 1971; Volynets et al., 1974). The IR spec-tra of the Tairus Ni-doped synthetic sapphires (seefigure 17) are similar to those for Ni-containingcorundum grown by the Verneuil method in areducing, hydrogen atmosphere (Eigenmann andGunthard, 1971). Following these authors, webelieve that the 2980 cm-1 band is the absorptionband of OH– groups that compensate the charge forthe Al3++O2-Æ Ni2++(OH)– replacement.

The narrow peaks observed between 3600 and3000 cm-1 are also related to the absorption of OH–

groups which compensate for charge deficiency dueto the formation of cation vacancies. These peakshave been reported both for natural (Smith, 1995,and references therein) and synthetic corundum(Belt, 1967; Eigenmann and Gunthard, 1971; Voly-nets, 1974; Beran, 1991; Peretti and Smith, 1993,1994). The peaks at 3310 and 3400 cm-1, observed inthe infrared spectrum of our green sample (see fig-ure 17), were also reported earlier (Smith, 1995).

Peaks near 2000 cm-1 may be observed in the IRspectra of other OH– containing aluminum oxides.However, the peaks at 2114 and 1984 cm-1 that arepresent in the spectrum of diaspore (Ryskin, 1974;Smith, 1995) were absent from the spectra of ourhydrothermal synthetic sapphires. Therefore, we

believe that the bands near 2000 cm-1 in our sam-ples are not related to diaspore inclusions, especiallyas the samples studied were free of inclusions (at100× magnification). The IR absorption lines for COmolecules may occur between 2133 and 2148 cm-1

(Hallam, 1973). Wood and Nassau (1967) observedthe absorption bands related to CO2 molecules inberyl structure channels at 2354 cm-1. Because theTairus crystals grew in carbonate solutions of com-plex composition, the incorporation of CO, CO2,CO3

2-, and other forms of carbon oxides into thegrowing crystal is very likely. Thus, we believe thatthe narrow bands at 2460, 2400, 2264, 2136, and

Figure 14. At high (500×) magnification, smalltransparent octahedral crystals (possibly spinel)can be seen inside some of the aggregates in green-ish blue sapphires shown in figure 13. Photomicro-graph by S. Z. Smirnov.

Figure 15. In the green hydrothermal syntheticsapphires, the transparent crystals appeared to beanisotropic hexagonal platelets, which may begibbsite. When viewed between crossed polariz-ers, they showed a bright interference color.Photomicrograph by S. Z. Smirnov; magnified 500×.

Figure 16. Swirl-like and chevron patterns werecommonly seen in the Tairus synthetic green-to-greenish blue sapphires. Immersed in methy-lene iodide, magnified 100×; photomicrographby S. Z. Smirnov.


2016 cm-1 could be related to the C–O bond in dif-ferent carbon oxides that were incorporated into thesapphire crystals. Strong pleochroism is another fea-ture of these bands, which suggests that these C–Obonds are part of the corundum structure and arenot caused by separate carbon oxide species (e.g.,CO or CO2 in fluid inclusions).

These bands were also reported for Tairushydrothermal synthetic rubies by Peretti and Smith.(1993). We have not found a description of suchbands in the literature for natural and other synthet-ic sapphires. Therefore, we believe that the five nar-row peaks between 2500 and 2000 cm-1 are charac-teristic of Tairus synthetic sapphires.

IDENTIFYING CHARACTERISTICS OF TAIRUSHYDROTHERMAL SYNTHETIC SAPPHIRES Previously, blue, green, and yellow sapphires weregrown only by flux- or melt-growth techniques, butthe synthetic sapphires studied for this article moreclosely resemble their natural counterparts becausethey have been grown from a hydrothermal solu-tion, as takes place frequently in nature. Never-theless, there are a number of features by which an

experienced gemologist can separate these sapphiresfrom their natural counterparts and other syntheticsapphires:

1. The swirl-like patterns observed in Tairussynthetic sapphires are common to mostsynthetic crystals grown on seed platesunder hydrothermal conditions and are rarefor natural stones.

2. The red reaction to UV radiation of thegreenish blue synthetic sapphires contain-ing Cr is different from that of most naturaland other synthetic sapphires.

3. The crystalline copper inclusions and flake-like aggregates seen in the Tairus syntheticsare not present in natural sapphires.Metallic inclusions are common in melt-and flux-grown synthetics, but they usuallyconsist of platinum-group metals.

4. Ni, an important chromophore in theTairus synthetic sapphires, is not found innatural sapphires or most synthetic ones.

5. The absorption spectra of the greenish blueNi-doped Tairus synthetic sapphires differfrom those of natural blue sapphires, whichowe their color to Fe and Ti. The spectrum(in unpolarized light) of a greenish blueTairus synthetic sapphire includes threeintense bands––at 377, 599, and 970 nm––and two weak bands, at 435 and 556 nm.

6. Five narrow peaks between 2500 and 2000cm-1 in the infrared spectra of Tairus syn-thetic sapphires clearly separate them fromtheir natural and other synthetic counter-parts.

CONCLUSIONThe technology to manufacture synthetic sapphirein a wide variety of colors by means of hydrother-mal growth has been developed by researchers ofthe Tairus joint venture. This was achieved by dop-ing synthetic corundum with different amounts ofNi2+, Ni3+ and Cr3+. Specific colors are also producedby controlling the oxidation-reduction environ-ment. Although these colors overlap those of mostnatural and other synthetic sapphires, the Tairussynthetic sapphires can be readily identified on thebasis of their distinctive internal features (swirl-likepatterns and copper inclusions), visible-range andinfrared spectra, and the presence of Ni as a traceelement.

To date, thousands of hydrothermal syntheticsapphires have been grown at the Tairus factory in

Figure 17. The infrared absorption spectra ofgreenish blue (A) and green (B) Tairus syntheticsapphires are very similar (they are shown hereoffset for purposes of comparison). The curveshave been corrected for a spurious absorptionattributed to CO2.


Novosibirsk (Russia) and cut at the Pinky TradingCo. facilities in Bangkok. Greenish blue, pinkishorange, and yellow sapphires are commerciallyavailable as cut stones up to 4 ct.

Acknowledgments: The authors are grateful to IvanFoursenko and Alexander Dokukin for the samplesof hydrothermal synthetic ruby donated for thisstudy. Special thanks to Evgeny Krayushin for thephotos of synthetic sapphires studied. This work hasbenefited from discussions with Dr. DmitryFoursenko, Tairus director of research, and Dr.Vladislav Shatsky, director of the SiberianGemological Center.

REFERENCESBelt R.F. (1967) Hydrothermal ruby: infrared spectra and X-ray

topography. Journal of Applied Physics, Vol. 38, No. 6, pp.2688–2689.

Beran A. (1991) Trace hydrogen in Verneuil-grown corundumand its color varieties: an IR spectroscopic study. EuropeanJournal of Mineralogy, Vol. 3, No. 6, pp. 971–975.

Boksha O.N., Grum-Grzymailo S.V., Pasternak L.B., PopovaA.A., Smirnova E.F. (1970) Synthesis and optical spectra ofcorundums doped with transition elements. In S.V. Grum-Grzymailo, B.S. Skorobogatov, P.P. Feofilov, and V.I.Cherepanov, Eds., Spectroscopy of Crystals, “Nedra,”Moscow, pp. 295–302 (in Russian).

Burns R.G. (1981) Intervalence transition in mixed-valence min-erals of iron and titanium. In G.W. Wetherill, A.L. Albee, andF.G. Stehli, Eds., Annual Review of Earth and PlanetarySciences, Vol. 9, Annual Reviews Inc., Palo Alto, CA, pp.345–383.

Catalogue of Colored Glasses (1967). “Mashinostroenie,”Moscow (in Russian).

Coldham T. (1985) Sapphires from Australia. Gems &Gemology, Vol. 21, No. 3, pp. 130–146.

Crowningshield R. (1983) Padparadscha: What’s in a name?Gems & Gemology, Vol. 19, No. 1, pp. 30–36.

Eigenmann K., Gunthard Hs.H. (1971) Hydrogen incorporation indoped a-Al2O3 by high temperature redox reactions.Chemical Physics Letters, Vol. 12, No. 1, pp. 12–15.

Feklichev V.G. (1989) Diagnostic Constants of Minerals.“Nedra,” Leningrad (in Russian).

Furui W. (1988) The sapphires of Penglai, Hainan Island, China.Gems & Gemology, Vol. 24, No. 3, pp. 155–160.

Gem Reference Guide (1995) Gemological Institute of America,Santa Monica, CA, 270 pp.

Goldschmidt V. (1916) Atlas der Krystalformen. Carl WintersUniverstatsbuchhandlung, Heidelberg.

Gübelin E. (1982) New synthetic rubies made by Professor P.O.Knischka. Gems & Gemology, Vol. 18, No. 3, pp. 165–168.

Gunawardene M., Rupasinghe S. (1986) The Elahera gem field incentral Sri Lanka, Gems & Gemology, Vol. 22, No. 2, pp.80–95.

Hallam H.E. (1973) Molecules, trapped in low temperaturemolecular matrices. In Hallam H.E., Ed., VibrationalSpectroscopy of Trapped Species: Infrared and RamanStudies of Matrix-Isolated Molecules, Radicals and Ions,Wiley-Interscience, London-New York, pp. 68–132.

Kammerling R.C., Koivula J.I., Fritsch. E. (1994) An examinationof Chatham flux-grown synthetic pink sapphires. Journal ofGemmology, Vol. 24, No. 3, pp. 149–154.

Kane R. (1982) The gemological properties of Chatham flux-grown synthetic orange sapphire and synthetic blue sapphire.Gems & Gemology, Vol. 18, No. 3, pp. 140–153.

Kuznetsov V.A., Shternberg A.A. (1967) Crystallization of rubyunder hydrothermal conditions. Soviet Physics-Crystallography, Vol. 12, No. 2, pp. 280–285.

Laudise R.A., Ballman A.A. (1958) Hydrothermal synthesis ofsapphire. Journal of the American Chemical Society, Vol. 80,pp. 2655–2657.

Minomura S., Drickamer H.G. (1961) Effect of pressure on thespectra of transition metal ions in MgO and Al2O3. Journal ofChemical Physics, Vol. 35, No. 3, pp. 903–907.

Muller R., Gunthard Hs.H. (1966) Spectroscopic study of thereduction of nickel and cobalt ions in sapphire. Journal ofChemical Physics, Vol. 44, No. 1, pp. 365–373.

Peretti H.A., Smith C.P. (1993) A new type of synthetic ruby onthe market: Offered as hydrothermal rubies fromNovosibirsk. Australian Gemmologist, Vol. 18, No. 5, pp.149–157.

Peretti H.A., Smith C.P. (1994) Letters to the editor. Journal ofGemmology, Vol. 24, No. 1, pp. 61–63.

Platonov A.N., Taran M.N., Balitsky V.S. (1984) The Nature ofGemstone Colors. “Nedra,” Moscow, 196 pp. (in Russian).

Roedder E. (1984) Fluid inclusions. Reviews in Mineralogy, Vol.12, 644 pp.

Ryskin Ya.I. (1974) The vibration of protons in minerals:Hydroxyl, water and ammonium. In Farmer V.C., Ed., TheInfrared Spectra of Minerals, Mineralogy Society Monograph,London, pp. 137–181.

Smirnov S. (1997) The inclusions of mineral-forming media insynthetic and natural gemstones (formation mechanisms andgenetic application). Ph.D. diss., Novosibirsk (in Russian).

Smith C.P. (1995) A contribution to understanding the infraredspectra of rubies from Mong Hsu, Myanmar. Journal ofGemmology, Vol. 24, No. 5, pp. 321–325.

Smith C.P., Kammerling R.C, Keller A.S., Peretti A., ScarrattK.V., Nguen Dang Khoa, Repetto S. (1995) Sapphires fromSouthern Vietnam. Gems & Gemology, Vol. 31, No. 3, pp.168–188.

Volynets F.K., Sidorova E.A., Stsepuro N.A. (1974) OH– groups incorundum crystals which were grown with the Verneuil tech-nique. Journal of Applied Spectroscopy, Vol. 17, No. 6, pp.1626–1628.

Wood D.L., Nassau K. (1967) Infrared spectra of foreignmolecules in beryl. Journal of Chemical Physics, Vol. 47, No.7, pp. 2220–2228.

Wright W.D. (1964) The Measurement of Colour, 3rd ed. VanNostrand, New York, 291 pp.

Yakubova V.V. (1965) Corundum. In F.V. Chukhrov and E.M.Bornschted-Kupletskaya, Eds., Minerals, Nauka, Moscow, pp.63–75 (in Russian).

204 Notes and New Techniques GEMS & GEMOLOGY Fall 1997

ABOUT THE AUTHORS

Dr. Johnson is manager of Research and Development, andMs. Wentzell is staff gemologist, at the GIA Gem TradeLaboratory, Carlsbad, California. Mr. Elen is a researchgemologist in GIA Research, Carlsbad.

Please see acknowledgments at end of article.

Gems & Gemology, Vol. 33, No. 3, pp. 204–211.© 1997 Gemological Insitute of America

of this article is to report on this new deposit and todocument the unusual composition and other prop-erties of this Bi-bearing elbaite.

LOCATION AND ACCESSThe multicolored tourmalines described in thisreport were found in a field in the Kalungabeba areaof the Lundazi district (figure 2), about 16 km (10miles) southwest of Lundazi itself, according to Mr.Sarosi, who visited the site in early 1997. Thestones were originally discovered by a person whowas digging a hole for an outhouse.

A rough dirt road leads to this site fromLundazi; it is best traveled by four-wheel-drive vehi-cle. During the rainy season, from about late De-cember until March, the area is often inaccessible.

GEOLOGYThe geologic feature known as the MozambiqueBelt forms a triangular area bounded by theLuangwa valley (eastern Zambia), the Zambezi val-ley (in northern Mozambique), and Lake Malawi (inMalawi; again, see figure 2). The main rock units inthe belt are high-grade gneisses and granitic rocks of

NOTES AND NEW TECHNIQUES

MULTICOLORED BISMUTH-BEARINGTOURMALINE FROM LUNDAZI, ZAMBIA

By Mary L. Johnson, Cheryl Y. Wentzell, and Shane Elen

At the 1997 Tucson show, one of the authors(CYW) saw some unusual tourmalines that werereportedly from recent workings in Zambia. MarcSarosi Co., of Los Angeles, was marketing thismaterial as “mixed-color” tourmaline: Colors avail-able include light-to-medium-toned desaturatedpink, orange, green, and yellowish green (see, e.g.,figure 1), as well as colorless, and individual fash-ioned stones frequently contain regions of two ormore of these colors. Coincidentally, we found thatthe chemistry of these tourmalines was unusual,with elevated contents of bismuth (Bi). The purpose

Tourmalines from an alluvial deposit near Lundazi in Zambia consist of color zoned pink/colorless/yellow-green “watermelon” nodules; manyfashioned stones contain patches and blends of all three colors. Althoughthese elbaite tourmalines are unusually rich in bismuth (up to 0.49 wt.%Bi2O3), the presence of bismuth apparently has no effect on their gemologi-cal properties. Approximately 10,000 carats of fashioned stones are knownto have been produced from October 1996 to mid-1997.

Notes and New Techniques GEMS & GEMOLOGY Fall 1997 205

the 2.05–2.68 billion-year-old Basement Complex,as well as metacarbonates, schists, metavolcanics,and quartzites of the 1.35–2.05 billion-year-oldMuva Supergroup (Kamona, 1994). The gneisses andschists are intruded by later granitic pegmatites,which commonly have quartz cores and outer zonesof microcline feldspars. Muscovite is usually con-centrated at the contact between these zones, andaquamarine may occur in either zone. The peg-matites may also contain gem tourmaline, as wellas concentrations of rare elements (Nb, Ta, Li, Cs,Y, La, and U). The Mozambique Belt also hosts goldmineralization, in sheared quartz veins occurring inschists and metavolcanics of the Muva supergroup;bismuth minerals (bismuthinite, bismutite) arefound in this gold ore.

Tourmaline has been mined in at least threeareas in eastern Zambia (figure 2): Lundazi, at thenortheast end of the Belt (e.g., the Aries deposit:Kamona, 1994); Chipata, in the center of the belt(see, e.g., Schmetzer and Bank, 1984a and b); andNyimba, in the south (e.g., the Hofmeyer deposit:Kamona, 1994). The Chipata area is known to pro-duce “intense” yellow tourmalines, (Schmetzer andBank, 1984a and b; Koivula, 1985); red to brownishred tourmalines (Koivula and Fryer, 1985); and

brownish orange and brownish green stones (men-tioned by Koivula and Kammerling, 1991). TheHofmeyer pegmatite near Nyimba is the source oftourmaline varieties rubellite and green “emeraldo-lite” (Kamona, 1994). Bank (1982) described yellow-green, green, red, brown, and violet tourmalinesfrom Zambia, but he provided no specific informa-tion about their source. The bicolored tourmalinementioned by Thomas (1982) may have come fromLundazi or Nyimba.

MINING AND PRODUCTIONThe Kalungabeba deposit is alluvial, and the tour-maline-bearing gravels also contain spessartine,aquamarine, and dispersed quartz. Both pink and“multicolored” tourmalines have been found, theformer occurring in a variety of crystal shapes, butthe latter always as nodules; all show wear fromalluvial transport. At the time of Mr. Sarosi’s visit,there were no known gem-bearing outcrops in theimmediate vicinity of this field, although Mr. Sarosireports that subsequent mining has unearthed spes-sartine garnet–bearing pegmatites.

Production at this locality started in October1996. At the time of Mr. Sarosi’s visit in early 1997,about 90% of the tourmaline was coming from a

Figure 1. This 9.44 ct pinkelbaite tourmaline and its11.48 ct yellow-green com-panion are among the thou-sands of carats of gem tourmaline that have beenfashioned from materialrecovered recently from adeposit near Lundazi,Zambia. Note the distinc-tive “checkerboard” cut.Courtesy of Marc Sarosi;photo © GIA and TinoHammid.


single pit measuring about 12 m (40 feet) in diame-ter and 6 m (20 feet) deep. Approximately 700 unli-censed miners worked the deposit during the initial“rush”; after five died, the Ministry of Mines closedthe area and then issued a number of licenses(around May 1997). At present, approximately 60miners are working the deposit by hand, haulingloads of gravel to an adjacent water-filled depres-sion, where wet-sieving is done to recover the gemrough. About 100 kg of facet-quality rough (and sev-eral tons of lower-quality material) had been recov-ered as of June 1997. By this time, Mr. Sarosi, whobelieves he purchased most of the available multi-colored rough, had fashioned some 10,000 carats ofgoods, ranging up to 20 ct.

MATERIALS AND METHODSNineteen faceted stones (3.04 to 11.48 ct) and 10nodules (10.28 ct [9 × 11 × 14 mm] to 20.61 ct [12 ×14 × 15 mm]) were examined; these were chosen torepresent the variety of material on hand. Thefaceted stones consisted of six pink, five yellow-green, and eight multicolored samples; the roughnodules were all multicolored green and pink. Welooked at the rough samples with diffuse transmit-ted illumination to characterize the color zoning,

but we did not test these further gemologically.Gemological characterization of the fashioned

samples was performed as follows on all 19 facetedstones, except where noted. We observed the face-up colors using daylight-equivalent fluorescent lightsources. Polarization behavior was noted using aGIA GEM Instruments Illuminator polariscope, andwe determined pleochroism using a dichroscopeand the polariscope light source. The colors seenwith a “Chelsea” filter were noted for 13 of the 19faceted stones, and we measured refractive indicesfor these same 13 stones using a Duplex II refrac-tometer with a sodium-equivalent light source and apolarizing eyepiece. Specific gravities were deter-mined by hydrostatic weighing (13 stones). Fluores-cence was observed in a darkened room using a con-trolled viewing environment and a GIA GEM short-wave/long-wave ultraviolet lamp. Absorption spec-tra were noted with a Beck spectroscope in a desk-model configuration and transmitted light.Microscopic properties were observed with a gem-ological microscope (Reichert Stereo Star Zoom)equipped with polarizing plates.

We investigated trace-element chemistry byqualitative energy-dispersive X-ray fluorescence(EDXRF) spectroscopy, using a Tracor Xray (now

MOZAMBIQ

UE

BELT

Lusaka

TANZANIA

MOZAMBIQUE

ZIMBABWE

ZAIRE

ANGOLA

MALAWI

ZAMBIA

Lundazi

ChipataLake

Ma

lawi

Ta nganyika

Nyimba

200 mi.

300 km

0N

LUAN

G WA

RIV

ERZAM

BEZI RIVER

ZAMBEZI RIVER

Figure 2. A newdeposit of multicol-ored tourmalines hasbeen found in a fieldin the Kalungabebaarea of Zambia’sLundazi district,about 16 km south-west of the city ofLundazi.


Spectrace Instruments) Spectrace 5000 instrumentwith a rhodium-target X-ray tube. Complete spectrawere acquired for six faceted stones, including abicolored stone for which both pink and greenregions were analyzed. It is unusual to see bismuthin tourmaline (Paraíba stones being a notable excep-tion; see, e.g., Fritsch et al., 1990), and we werequite surprised to see evident Bi peaks in theEDXRF spectra of the first stones we examined.Therefore, for eight other stones, data were collect-ed only in the bismuth region (8–18 keV) of theEDXRF spectrum.

Electron microprobe analyses were obtainedand processed at the University of Manitoba,Canada, for three samples (3 points each on a 9.44 ctpink cut stone and an 11.48 ct yellow-green cutstone, and 8 points in a traverse along a polishedslice from a multicolored nodule). The followingelements were quantitatively measured: Na, Mg,Al, Si, P, F, K, Ca, Ti, V, Cr, Mn, Fe, Zn, and Bi(Bi2Se3 was used as a Bi standard for analyses, andthe detection limit was 0.17 wt.% Bi2O3). The ana-lytical conditions and methods for calculating thetourmaline formulas are given in Burns et al. (1994).

APPEARANCE AND GEMOLOGICAL PROPERTIESVisual Appearance. Rough Material. The ninerough samples (some of which are illustrated in fig-ure 3) seemed typical for tourmaline, showingthree-sided symmetry in cross-section. The roundedtop and bottom surfaces showed conchoidal frac-ture, creating an overall nodular shape. Some paral-lel striations were evident on the sides of the nod-ules, which is also typical for tourmaline; however,all samples showed worn surfaces.

With diffuse transmitted light, we saw that therough was zoned “watermelon” fashion, typicallywith pink triangular cores and green rinds thatapproached hexagons in outline. Paler (green to col-orless) zones were sometimes evident between thepink cores and green rinds (again, see figure 3).

Cut Stones. We examined one round, one cushion,two rectangular, four square, and 11 oval cut stones,all modified brilliants (see, e.g., figure 4). All of thefashioned stones had been modified so that thecrown area was covered with many facets(“checkerboard” cut). The colors of these samplesare summarized in table 1. Color distributionranged from even through uneven to multicolored.Most of the multicolored stones were fashioned to

show patches of color throughout (figure 5), ratherthan the regular bars of color that are commonlyseen in a bicolored or tricolored tourmaline.

According to Mr. Sarosi, this “mixed-color”tourmaline presents some unusual challenges to thegem cutter. To obtain the desired multicoloredappearance, Mr. Sarosi noted, the green rind on therough must be retained in the faceted stone. In addi-tion, the rind also seems to impart physical stabilityto the stone; if all of it is removed, the stone oftenbreaks during fashioning. Regardless, the tourma-line rough cracks into pieces if sawn. A larger thanexpected number of stones break during faceting,and a few have even broken after cutting was com-pleted. Because of this, the largest fashioned stoneshave only reached 20 ct in size, and the rough has ayield of about 20% (in comparison to an expectedyield of 30%–33%). Mr. Sarosi finds that the“checkerboard” cut (again, see figure 1) works bestto minimize breakage during cutting.

Physical and Optical Properties. Since we mightexpect the physical and optical properties of tour-maline to vary with color, we arranged the datainto three classes: pink (including red), yellow-green, and multicolored stones. The properties foreach of these groups are summarized in table 1. Adesaturated “blend” of pleochroic colors was seenin some of the multicolored stones. All 13 stones on

Figure 3. These four tourmaline nodules from thenew Lundazi deposit show “watermelon” colorzoning. Total weight is 55.73 ct, with the largeststone measuring about 15 × 14 × 12 mm. Courtesyof Marc Sarosi; photo by Maha DeMaggio.


TABLE 1. Gemological properties for various colors of Lundazi tourmaline.

Property Pink Yellow-green Multicolored

Weight 3.04–9.44 ct 4.75–11.48 ct 3.38–8.14 ctColor Purplish pink to red-purple to Yellowish green to greenish yellow Pink and yellowish green

orangy redPleochroism Purplish red to red––orangy red Yellowish green to green––yellow to Mixtures of the colors seen in

to yellow green to green-brown the pink and yellow-greenstones

Optic character Uniaxial Uniaxial UniaxialColor-filter reaction Gray-green Slightly grayish green to green Gray-green to grayish greenR.I. (nε) 1.619–1.620 1.619–1.620 1.620R.I. (nω) 1.637–1.638 1.638–1.639 1.639Birefringence 0.018 0.018–0.020 0.019Specific gravity 3.041–3.043 3.040–3.049 3.045–3.055Reaction to long- Inert Inert Inertand short-wave UVAbsorption spectrum Lines at 451 and 458 nm, weak One stone: lines at 460 and 500 nm, Mixture of pink and yellow-

group of 4 or 5 lines to 500 nm; general absorption above 640 nm. green spectra535 nm line, 490–540 nm band, The rest: weak general absorptionabsorption below 415 nm below 500 nm, faint to weak diffuse

500 nm lineInclusions “Pinpoints” of tiny colorless Clean, or scattered “pinpoints” only; Curved or planar clouds of

crystals, parallel growth tubes, “pinpoints” lying in growth planes; thin “pinpoints”; growth tubes,one small transparent plate; growth tubes near perimeter of stone primarily in green areas; filledfilled growth tube growth tubes

Growth structures Minimal to angular Wavy and/or angular Straight, wavy, and/or angular

Figure 4. The multicol-ored tourmalines arefashioned as either a

single color or as amixture of colors in a

single stone. Thesefour elbaite tourma-

lines, which rangefrom 3.38–8.14 ct,

were part of the study sample. Courtesy of

Marc Sarosi; photo byMaha DeMaggio.

which refractive index readings were taken hadR.I.’s in the range of nε = 1.619–1.620 and nω =1.637–1.639, with birefringences of 0.018–0.020.None of the stones tested showed any reaction toUV radiation. These properties are not unusual fortourmalines of these colors.

Spectroscopy. The samples showed typical spectrafor tourmalines of similar colors (again, see table 1).

Microscopy/Inclusions. The main inclusions seenwith magnification were: “pinpoints” (very smallminerals), growth tubes (figure 6) or needles, andfractures. All of these inclusions are common intourmaline, regardless of its source, and there wasno indication of possible included bismuth minerals(e.g., metallic bismuthinite or native bismuth, orhigh-relief near-colorless bismutite). Small opaquereddish orange spheres were noted in the ends of


growth tubes in four samples.The prominent growth features we identified

were growth bands––sharp boundaries betweenregions of slightly different refractive index, similarto the graining seen in diamonds. These bands werestraight, angular, or wavy. Angular growth bandswere seen in pink, yellow-green, and multicoloredstones; wavy growth bands in combination withangular growth bands were seen in yellow-greenand multicolored stones; and straight growth bandsin combination with angular growth bands wereseen in multicolored stones. In one pink stone, wesaw only minimal growth banding. The growthbands were pronounced (eye-visible) in nearly allstones; however, no strain was obvious in any stonewhen viewed between crossed polarizers.

CHEMICAL ANALYSISEDXRF. We performed EDXRF analyses on six ofthe faceted tourmalines: two pink, one yellow-green, two multicolored (analyzed without regard tocolor orientation), and one pink-and-yellowishgreen bicolor in which both color regions could beseparately examined. Selected-energy-regionEDXRF analyses for bismuth were performed oneight additional stones––three pink, two yellow-green, and three multicolored. These analyses couldbe roughly quantified on the basis of the microproberesults (see below). We found major amounts

(greater than 1 wt.% oxide) of Al and Si, and minor(greater than 0.1 wt.% oxide) Ca and Mn. The Fecontent could be classified as major, minor, or trace(less than 0.1 wt.% oxide) in these stones, depend-ing on color: The amount of absorption due to Fe isappreciably greater in the spectra of yellow-greenstones (or regions) than in those of pink stones (orregions). Bismuth occurred at minor to trace levels.Trace amounts of Cu, Zn, Ga, and Ge were found inall the EDXRF spectra. Ti was seen in the yellow-green region of one multicolored stone and inanother multicolored stone at the trace level. Pbwas detected in one multicolored stone.

Each of the eight spectra taken using conditionsto detect only Bi revealed relatively high amounts ofthat element (i.e., at the minor-element level),regardless of the color of the material. Raw countsin the EDXRF analyses suggest that more Bi is pres-ent in the early-forming pink zones (i.e., the cores ofthe multicolored nodules) than in the later-formingyellow-green rinds.

Electron Microprobe. Three samples were examinedin more detail using an electron microprobe (thetwo cut stones shown in figure 1 and the multicol-ored nodule shown in figure 7); the results are givenin table 2. The compositions indicate that thesetourmalines are all elbaite, the sodium-, lithium-,and aluminum-rich species in the tourmaline min-eral group. As with the EDXRF analyses, the greenrind of the tourmaline nodule showed less bismuth

Figure 5. The multicolored nature of the fashionedstones comes from the cutter’s incorporation of thedifferent color zones in the original tourmalinerough. Attractively mixed in the face-up view ofthis 5.87 ct Lundazi tourmaline, the color zones areclearly evident in the side view of a 4.94 ctLundazi stone. Photo by Maha DeMaggio.

Figure 6. Some of the growth tubes in the green rindof this 12.02 ct multicolored tourmaline nodule arepartially filled with an orange material. Photo-micrograph by John I. Koivula; magnified 20×.


than the pink core. The most bismuth (0.49 wt.%)was found in the pink stone.

DISCUSSIONThe gemological properties of the Zambian tourma-lines we studied are consistent with those of thetourmalines from the Lundazi area described byThomas (1982). One reason for extensive past inter-est in Zambian tourmalines was the possibility thatspecimens from the Chipata area might have beenthe hypothetical manganese-rich species tsilaisite,although none proved to be so (see, e.g., Schmetzerand Bank, 1984a and b; Shigley et al., 1986). All thestones examined for the present study have low Mncontents.

The unusual feature of these elbaites is theirhigh Bi contents. Traces of Bi are sometimes foundin rocks crystallized from late-stage silicic melts(i.e., in granitic pegmatites: see, e.g., Mintser, 1979);in Zambia, a possible source for this element is the nearby gold deposits (Kamona, 1994). However, bis-muth does not readily substitute for common ele-ments in silicates (see, e.g., Kupcík, 1972), so itspresence in tourmaline is unexpected.

Examples of Bi-bearing tourmaline are relative-ly uncommon. In an exhaustive literature search,Dietrich (1985) noted trace amounts of Bi in dravite-uvite and schorl tourmalines, but not in elbaites.Peretyazhko et al. (1991) described rubellites(elbaites) from miarolitic granitic pegmatites in theMalkhan Range in Central Transbaikalia, Russia,that contained both Bi (up to 0.55 wt.% Bi2O3) andlead. The elbaites from São José da Batalha, Paraíba,Brazil, are best known for their copper contents, butthey also contain up to 0.83 wt.% Bi2O3 (on thebasis of microprobe analyses by Fritsch et al., 1990;Henn et al., 1990; and Rossman et al., 1991). Inaddition, according to unpublished microprobe data,a color-change tourmaline from East Africa(described by Koivula and Kammerling, 1991) con-tained trace amounts of Bi.

One reason for the paucity of information onbismuth in tourmaline is that most modern analy-ses are performed with the electron microprobe,which has a relatively high detection limit (e.g.,0.17 wt.% Bi2O3 in our analyses; however, EDXRFis considerably more sensitive for Bi. Further, unlessbismuth is looked for in microprobe analyses, it will

TABLE 2. Electron microprobe data for three Lundazi tourmalines.

Multicolored nodule

Oxide Pink Yellow-green Green rind Pink core(wt.%) (3 points) (3 points) (4 points) (4 points)

SiO2 38.00–38.50 38.00–38.90 38.00–39.10 38.60–39.10TiO2 nda 0.03–0.05 0.05–0.34 nd–0.02Al2O3 41.00–41.30 39.20–40.60 37.10–40.00 41.30–41.90V2O3 nd–0.01 nd–0.02 nd ndCr2O3 nd–0.05 nd–0.12 nd–0.09 nd–0.12Bi2O3 0.34–0.49 nd–0.13 nd–0.02 0.16–0.25MgO nd 0.04–0.07 0.21–0.61 ndCaO 0.51–0.58 0.51–1.07 0.09–0.34 0.38–0.47MnO 0.18–0.26 0.30–0.48 0.36–0.91 0.26–0.38FeO nd–0.07 0.69–0.84 1.70–4.19 0.03–0.13ZnO nd–0.02 0.01–0.03 nd–0.05 ndNa2O 1.66–1.74 1.74–1.83 2.03–2.49 1.70–1.82K2O nd–0.01 0.01–0.02 nd–0.02 nd–0.01P2O5 nd–0.01 nd nd–0.01 nd–0.03Li2Ob 2.14–2.19 2.04–2.31 1.80–2.16 2.11–2.27F 0.01–1.12 0.88–1.00 0.95–1.12 0.85–1.07B2O3

c 11.03–11.13 11.03–11.09 10.87–11.19 11.18–11.25H2Oc 3.30–3.84 3.35–3.39 3.29–3.40 3.37–3.46-O=F (-0.47)–nd (-0.42)–(-0.37) (-0.47)–(-0.40) (-0.45)–(-0.36)

Total 99.01–99.84 99.39–100.09 100.06–101.15 100.27–101.39

and=none detected.bDetermined by assuming there are 3 Y cations per formula unit, among them Ti, Al, V, Cr, Mg, Mn, Fe, Zn, and Li (see Burns et al., 1994).cDetermined from tourmaline stoichiometry: 3 B per formula unit,4 (OH + F) per formula unit.

Figure 7. Microprobe analyses across the primarycolor zones––pink and green––in this Lundazi tour-maline nodule showed higher Bi contents in thepink core. Courtesy of Marc Sarosi; photo © GIAand Tino Hammid.

v


not in general be found. We would not be surprisedif tourmalines from other localities are also found tocontain significant amounts of bismuth.

It is not apparent what effect, if any, bismuthhas on the gemological properties of tourmaline.Bismuth is not a chromophore (see, e.g., Kupcík,1972). As a heavy element, its presence shouldincrease both specific gravity and refractive index(see, e.g., Bloss et al., 1983), but neither of theseproperties appears to have been affected at the con-centration levels noted in our samples.

CONCLUSIONSignificant amounts of tourmaline from a new allu-vial deposit in the Lundazi area of Zambia are nowavailable. These attractive tourmalines are notablefor their chemical composition (Bi-bearing elbaite)as well as for the unusual “mixed-color” appearanceof the faceted stones, a consequence of the durabili-ty problems inherent in the rough. About 10,000 ctof faceted stones are now available, and tons oflower-quality material have been mined.

Bismuth is not usually present at detectablelevels in gem tourmaline. In the Lundazi stones,

more Bi was measured in the pink cores than in thegreen rinds, and individual analyses give concentra-tions of up to 0.49 wt.% Bi2O3. Nevertheless, thepresence of Bi in these tourmalines does not appearto influence their gemological properties. The mostdistinctive property of these stones is the growthzoning; otherwise, the gemological properties areconsistent with those recorded for elbaites of simi-lar colors from other localities.

Acknowledgments: The authors thank Marc Sarosi ofMarc Sarosi Co., Los Angeles, for making this materi-al available to us, and for responding so willingly toour repeated queries. Dr. Frank Hawthorne and JulieSelway of the University of Manitoba, Canada, pro-vided electron microprobe analyses. Staff gemologistDino DeGhionno, GIA Gem Trade Laboratory,Carlsbad, confirmed the gemological observations;and Sam Muhlmeister, GIA Research, Carlsbad,aided in the interpretation of the EDXRF spectra. Dr.Eugene Foord, of the U.S. Geological Survey, broughtthe Malkhan material to our attention, and providedtranslation of some of the relevant Russian results.

REFERENCES

Bank H. (1982) Turmaline diverser Grün- und Rottöne ausSambia. Zeitschrift der Deutschen GemmologischenGesellschaft, Vol. 31, No. 1/2, pp. 91–92.

Bloss F.D., Gunter M., Shu C.S., Wolfe H.E. (1983) TheGladstone-Dale constants: A new approach. CanadianMineralogist, Vol. 21, No. 1, pp. 93–99.

Burns P.C., MacDonald D.J., Hawthorne F.C. (1994) The crystalchemistry of manganese-bearing elbaite. CanadianMineralogist, Vol. 32, No. 1, pp. 31–41.

Dietrich R.V. (1985) The Tourmaline Group. New York, VanNostrand Reinhold Co., 300 pp.

Fritsch E., Shigley J.E., Rossman G.R., Mercer M.E., MuhlmeisterS.M., Moon M. (1990) Gem-quality cuprian-elbaite tourma-lines from São José da Batalha, Paraíba, Brazil. Gems &Gemology, Vol. 26, No. 3, pp. 186–205.

Henn U., Bank H., Bank F.H. (1990) Transparent bright blue Cu-bearing tourmalines from Paraiba, Brazil. MineralogicalMagazine, Vol. 54, pp. 553–557.

Kamona A.F. (1994) Mineralization types in the MozambiqueBelt of eastern Zambia. Journal of African Earth Sciences, Vol.19, No. 3, pp. 237–243.

Koivula J.I. (1985) Gem news: Tucson 1985. Gems & Gemology,Vol. 21, No. 1, p. 60.

Koivula J.I., Fryer C.W. (1985) Interesting red tourmaline fromZambia. Gems & Gemology, Vol. 21, No. 1, pp. 40–41.

Koivula J.I., Kammerling R.C. (1991) Gem News: Tourmaline

with unusual “color change.” Gems & Gemology, Vol. 27,No. 3, pp. 184–185.

Kupcík V. (1972) Bismuth: Crystal chemistry. In K. H. Wedepohl,Ed., Handbook of Geochemistry, Vol. II/1, Springer-Verlag,Berlin, pp. 83-B-1-83-M-1, 83-O-1.

Mintser E.F. (1979) The geochemical properties of the behavior ofbismuth in hypogenic processes. In E.E. Angino and D.T.Long, Eds., Benchmark Papers in Geology, Vol. 49:Geochemistry of Bismuth, Dowden, Hutchinson & Ross,Stroudsburg, PA, pp. 268–327.

Peretyazhko I.S., Zagorskiy V.Ye., Bobrov Yu.D. (1991) First findof bismuth- and lead-rich tourmaline. Doklady AkademiiNauk USSR, Earth Sciences Section, Vol. 307, No. 1–6, pp.175–179 [In Russian].

Rossman G.R., Fritsch E., Shigley J.E. (1991) Origin of color incuprian elbaite from São José da Batalha, Paraíba, Brazil.American Mineralogist, Vol. 76, pp. 1479–1484.

Schmetzer K., Bank H. (1984a) Crystal chemistry of tsilaisite(manganese tourmaline) from Zambia. Neues Jahrbuch fürMineralogie Monatshefte, No. 2, pp. 61–69.

Schmetzer K., Bank H. (1984b) Intensive yellow tsilaisite (man-ganese tourmaline) of gem quality from Zambia. Journal ofGemmology, Vol. 19, No. 3, pp. 218–223.

Shigley J.E., Kane R.E., Manson D.V. (1986) A notable Mn-richgem elbaite tourmaline and its relationship to “tsilaisite.”American Mineralogist, Vol. 71, pp. 1214–1216.

Thomas A.E. (1982) Zambian tourmaline. Journal ofGemmology, Vol. 18, No. 1, pp. 4–6.

v

v

212 Lab Notes GEMS & GEMOLOGY Fall 1997

DIAMOND

Cabochon CutBlack opaque materials, usually fash-ioned as cabochons or beads, are sub-mitted regularly to the Gem TradeLaboratory for identification; manyrequire advanced testing (see, e.g., M. L. Johnson et al., “Some Gem-ological Challenges in IdentifyingBlack Opaque Gem Materials,”Gems& Gemology, Winter 1996, pp. 252–261). Nevertheless, the identity of thetwo black cabochons shown in figure 1came as quite a surprise to us.

The refractive index was over thelimits of a standard refractometer, andthe specific gravity, measured by thehydrostatic method, was 3.44. Withmagnification, the cabochons showed

a poor polish and an aggregate struc-ture, with several transparent-to-translucent, near-colorless areas seenon the base of each piece (figure 2). Thenear-colorless areas showed weak yel-low fluorescence to long-wave ultravio-let radiation, but the rest of each cabo-chon was inert. These properties sug-gested that the material was diamond,so we performed a hardness test on aninconspicuous portion of the basaledge of each piece, by rubbing itagainst a faceted piece of syntheticcorundum and observing the result at10∞ magnification. Together, the highdegree of friction between the twomaterials, plus the scratch producedon the synthetic corundum, indicateda hardness of 10––and confirmed the

identification of these cabochons asdiamond.

We have reported on many un-usual diamond cuts in this section, butthese are the first diamonds we haveseen fashioned as cabochons. We notedthat the transparent areas on the bot-toms of the two pieces were similar insize, shape, and placement. When weasked our client (who actually cutthese gemstones) about this, he told usthat the rough material had beenshaped first into a bead, then sawed inhalf to make the two cabochons.

IR

Editor's note: The initials at the end of each itemidentify the editor(s) or contributing editor(s) whoprovided that item.Gems & Gemology, Vol. 33, No. 3, pp. 212–218”1997 Gemological Institute of America

Figure 1. Unlike most black opaque materials, these cabochons (3.55and 3.56 ct, approximately 9.26 mm in diameter and 4.56 mm indepth) did not require advanced testing for identification becausestandard gemological tests proved them to be diamond.

Figure 2. Fiber-optic illumina-tion revealed the aggregatenature of the cabochons shownin figure 1, as seen here throughthe base of one of them.Magnified 15×.

Lab Notes GEMS & GEMOLOGY Fall 1997 213

Rare Fancy Vivid OrangeIn the Fall 1996 issue (pp. 206–207),we reported on a diamond of rarecolor: a 3.40 ct heart shape that wasgraded Fancy Intense pinkish orange.At that time, we noted the unusualhue as well as an exceptionally strongsaturation. A 5.54 ct cushion-shapeddiamond (figure 3), which was recent-ly submitted to the East Coast lab,shared certain gemological propertieswith that heart shape and gave us anopportunity to expand our reportingof rare colors.

The current stone was more sat-urated and more of a “pure” orangethan the previously described heart

shape; it received a color grade ofFancy Vivid orange. While we en-counter diamonds in the orange huerange from time to time, we seldomencounter one that we can describesimply as “orange,” with no modi-fiers. The intensity of the stone’scolor also adds to its rarity. In the sys-tem used by the GIA Gem TradeLaboratory to describe colored dia-monds, the Fancy Vivid grade repre-sents those light- to medium-toneddiamonds of the strongest saturationrange (if the stone was darker in tone,the description would be Fancy Deep).

The rough was reported to be ofSouth African origin and to haveappeared predominantly brown, withonly a hint of orange. Like the FancyIntense pinkish orange diamonddescribed earlier, this stone alsoproved to be a type IIa diamond (asdetermined by infrared spectroscopy).The diamond fluoresced moderateyellow to long-wave UV radiation andmoderate orange to short-wave UV.

John King and TM

With Surface Droplets of

Filling MaterialThe most prominent visual feature ofmost fracture-filled diamonds—thebrightfield and darkfield flash-effectcolors—should be familiar to G&Greaders (see, e.g., R. C. Kammerling etal., “An Update on Filled Diamonds:Identification and Durability,” Gems& Gemology, Fall 1994, pp. 142–177;

and S. F. McClure and R. C. Kammer-ling, “A Visual Guide to the Ident-ification of Filled Diamonds,” Gems& Gemology, Summer 1995, pp.114–119 plus chart). In recent months,however, staff members in both theEast and West Coast laboratories haveseen fracture-filled diamonds inwhich these flash-effect colors wereless pronounced. Examples from theEast Coast lab were shown in theSpring 1997 Lab Notes section (pp.56–57). More recently, the WestCoast lab examined a fracture-fillednear-colorless marquise brilliant (0.49ct) that had extremely subtle flashcolors—and surface features that pro-vide a plausible explanation for thissubdued appearance.

With magnification, one surface-reaching “feather” in the stoneshowed low relief over most of itsextent, with extremely subtle yellowand purple flash colors (figure 4).However, a portion of this fracturenear the surface of the stone had veryhigh relief, with no evidence of fillingmaterial. Transparent droplets werevisible on the surface of the diamondnear this apparently “unfilled” por-tion of the feather (figure 5). The useof energy-dispersive X-ray fluores-cence (EDXRF) spectroscopy con-firmed the presence of lead (Pb) in thisstone, which we have found in alleffective diamond-filling materialsthat we have tested to date. In fact,we would not expect EDXRF todetect Pb in a gem diamond of thisappearance unless some sort of fillerwas present.

In the Fall 1994 article byKammerling et al. (cited above), wereported that heating a fracture-filleddiamond could cause the flash-effectcolors to be more subdued. The fea-tures we observed in this stone mightwell have been created if the diamondhad been heated with a torch after itwas filled. Although the filling mate-rial deep within the fracture probablywas damaged by heating, the fillingmaterial nearer the surface of thestone had actually boiled out andthen condensed as small droplets.

MLJ and SFM

Figure 3. This 5.54 ct FancyVivid orange diamond is rare forits intensity of color. Diamondsin the orange hue range are typi-cally darker, less saturated, andoften modified by brown.

Figure 4. The yellow darkfield (left) and purple brightfield (right)flash-effect colors in this fracture-filled 0.49 ct marquise brilliantare very subtle, as is typically seen in a filled diamond that has beensubjected to heat. Magnified 30×.


A Faceted Cat’s-Eye EMERALD

Most chatoyant stones are fashionedas cabochons. In general, the cat’s-eyephenomenon is due to the presence ofparallel bands of included materials(or hollow tubes) in a stone that hasbeen shaped so that light is concen-trated along a line perpendicular tothese parallel bands. In the rare caseswhere a potentially chatoyant stonehas been fashioned as somethingother than a cabochon, or with thelong axis of the cabochon orientedparallel to the inclusions, the resultmay be an overall sheen, but it is usu-ally not a sharp “eye.”

In summer 1997, a 3.19 ct semi-transparent green stone was submit-ted to the West Coast lab for identifi-cation. Although faceted, the stoneshowed an obvious and sharp “eye”(figure 6). The gemological propertiesof this emerald-cut stone were consis-tent with emerald, including: R.I.’s of1.570–1.578, an S.G. of 2.68 (mea-sured with a DiaMension noncontactmeasurement system, manufacturedby Sarin Technologies Ltd.), and typi-cal “chrome lines” seen with a desk-model spectroscope. It was inert toboth long- and short-wave UV radia-tion. With magnification, we couldsee small, dark, opaque crystals—probably chromite—as well as frac-tures and curved bands of parallelwhite fibrous inclusions resembling

the “silk” commonly seen in rubiesand sapphires. These inclusions con-firmed the identification of the stoneas a natural emerald, and no evidenceof clarity enhancement was present.

The observations through themicroscope also made it clear that theorientation of the white fibers wasresponsible for the chatoyancy. Thecurvature of the “silk” itself—not ofthe stone’s surface—and the orienta-tion of the curve to the table of thestone concentrated the light into asharp “eye.” Although cat’s-eye emer-alds are known from both Colombia(see Winter 1996 Gem News, pp.284–285) and Brazil (see, e.g., GemNews: Spring 1992, p. 60; Spring1995, pp. 60–61; and Fall 1995, p.

206), this is the first faceted cat’s-eyeemerald we have seen.

MLJ, Dino DeGhionno, and Philip Owens

JADEITE JADE

Impregnated, with ExceptionalTransparencyPolymer impregnation in jadeite jadeis an identification challenge fre-quently encountered at the GIA GemTrade Laboratory. While we rely oninfrared spectroscopy for conclusiveevidence of impregnation, standardgemological testing can reveal fea-tures suggestive of this treatment.These features have been documentedin the article “Identification ofBleached and Polymer-ImpregnatedJadeite” (Fritsch et al., Gems &Gemology, Fall 1992, pp. 176–187)and in a number of Lab Notes (e.g.,Spring 1995, p. 55).

An 8.25 ct variegated green-and-white carving was submitted to theEast Coast laboratory for identifica-tion. Standard gemological testingrevealed properties typical of jadeitejade, particularly a refractive index of1.66 by the spot method, and the pres-ence of chromium lines in the spec-troscope (which also indicates naturalgreen color). The bulk of the carvingwas inert to both long- and short-wave UV.

Usually, a distinctive surface tex-ture is the first indication of impreg-nation; in this case, however, it wasthe unusual internal structure thatinitially raised our suspicions. Thecarving had near-colorless areas ofexceptional transparency surroundingisolated green grains of jadeite (figure7). Although untreated jadeite is oftenvariegated green-and-white, it is quiterare to see the “white” areas with thisdegree of transparency (see, however,Gem Trade Lab Notes, Summer 1995,pp. 123–124). When we examined thepiece further, we discovered residualpolymer, most noticeably in a drilledhole (figure 8), but also in depressedareas of the carving. These areas fluo-resced strong yellow to long-wave UV

Figure 6. This 3.19 ct (11.05 ×10.73 × 3.69 mm) faceted emer-ald shows a sharp “eye.”

Figure 5. Small droplets on the surface of the filled diamond in figure 4are probably condensed filler material that boiled out of the filled frac-ture when the stone was heated. Magnified 40×.


radiation, while the rest of the pieceremained dark. Visible outlines ofgrain boundaries in both the whiteand green areas further supported oursuspicion that the piece had beenimpregnated; this was subsequentlyconfirmed by infrared spectroscopy.

The infrared spectrum revealedyet another unusual feature in thisthin (and apparently very heavilyimpregnated) carving: The polymerpeaks were so strong that the typicaljadeite spectrum was dwarfed in com-parison, and the absorption peaks typ-ical of jadeite were not recognizableuntil the spectrum was greatly magni-fied. Elizabeth Doyle

Resembling Nephrite JadeOne of the unavoidable temptationsfor any gemologist is the “Sight ID”:the identification of a stone withoutthe use of a refractometer, a spectro-scope, a microscope, or any otherinstrument. In many cases, the mate-rials to be identified are indeed whatwe suspected them to be, which rais-es the risk of increasing our hubris tounacceptable levels; then another,more challenging sample arrives torestore necessary humility.

In spring 1997, the West Coastlaboratory examined a sample thatillustrates the hazards of attemptinginstrument-free identifications. Thepierced and carved lid (to an urn thatwas not submitted) measured 97.8 ×97.7 ∞ 49 mm. The carving wastranslucent, with a glassy luster and a

mottled brownish green color.Perhaps a better descriptive term than“mottled” for the distribution of colormight be “cloudy” (nebulous), sincethe color patches appeared to have dif-fuse margins, and there were nostrongly saturated areas. At this point,the practitioner of the “Sight ID”technique would have proclaimed itto be nephrite on the basis of its lus-ter, diaphaneity, and desaturatedcolor.

Gemological properties immedi-ately confounded this expectation:We obtained an R.I. of 1.66; observeda line at 437 mm (characteristic ofjadeite) and chrome lines with the

spectroscope; and resolved an aggre-gate optic character with the polar-iscope—as expected for jadeite,nephrite, and many other aggregatesand rocks. The piece was inert to bothlong- and short-wave UV radiation; itwas too large for us to obtain specific-gravity measurements.

With magnification, we observedfeatures expected for jadeite, notnephrite, as well as the reason for thenephrite-like appearance. Jadeite jadeand nephrite jade, both aggregatematerials, owe their toughness to dif-ferent microstructures: In jadeite, thematerial consists of intergrownblocky jadeite crystals with interfin-gering margins; whereas nephrite is afelted (compressed and intertwined)aggregate of actinolite-to-tremoliteamphibole fibers. This lid was carvedfrom material consisting of ratherlarge (millimeter-size) jadeite grainswith translucent-to-semitranslucentgreen cores and near-colorless trans-parent margins (figure 9); such cloudy-cored “phantom” jadeite crystalswere new to our experience. Theappearance of the overall piece wasdue in part to the green cores, as wellas to the thin, iron-stained fracturesthroughout it. As with all jadeitejades, we checked this piece for poly-mer impregnation using FourierTransform infrared spectroscopy(FTIR); no absorptions indicative ofimpregnation were seen.

MLJ

Synthetic Green PERICLASE

In the Spring 1996 Gem Trade LabNotes (pp. 48–49), we reported on a5.49 ct colorless periclase that mighthave been confused with grossulargarnet. At the end of that entry, wenoted that green synthetic periclase,as reported in the literature, mightcause even more concern because ofits resemblance to green garnet.

As fate would have it, the WestCoast laboratory was recently askedto identify the 115.32 ct transparentyellowish green piece of rough shownin figure 10, which had been repre-sented as a natural periclase from

Figure 8. Excess polymer is visi-ble in this hole drilled in thecarving shown in figure 7.Magnified 20×.

Figure 7. The “white” areas inthis polymer-impregnated green-and-white jadeite jade carvingwere extremely transparent.Magnified 20×.

Figure 9. Jadeite “phantom”crystals, with translucent greencores and near-colorless rims,were seen with low magnifica-tion in this carved lid. The cen-ter crystal measures about 0.5mm across.


Ghana. Gemological properties were:R.I.—1.736; S.G.—3.59; optic charac-ter—singly refractive with weakanomalous double refraction; no fluo-rescence (inert) to either long- orshort-wave UV radiation; and noabsorption features seen with thehandheld spectroscope. When astrong parallel beam of light wasdirected into the piece from a fiber-optic light source, we saw strong redtransmission. With magnification, wenoted numerous blocky crystals inaddition to well-developed cleavagealong cube planes. This cleavage actu-ally dictated the shape of the rough(and proved that the piece could notbe grossular garnet, since garnets donot show cleavage in any direction).Chemical analysis by EDXRF spec-troscopy showed a high magnesiumcontent, with small amounts ofchromium, calcium, and iron; thisconfirmed the identity of the materialas periclase (pure periclase is magne-sium oxide, MgO).

Periclase has not been reported asoccurring in nature in anything closeto this size as facetable single-crystalmaterial. In fact, we could not find

any reference to a known naturalfaceted periclase of any size (we couldnot conclusively identify the emerald-cut stone described in the Spring 1996issue as natural or synthetic)—or tonatural periclase colored green bychromium. However, large transpar-ent pieces of synthetic periclase havebeen reported as by-products of themanufacture of refractory magnesiaby a company in Australia (see G.Browne, “Australian SyntheticPericlase,” Australian Gemmologist,November 1993, pp. 265–269). Thisprocess uses magnesite nodules—magnesium carbonate—from theKunwarara deposit near Rock-hampton, Queensland, which arecrushed and heated to produce cal-cined magnesia. This material canthen be electrofused to form ingots offused synthetic periclase. Theseingots have a 2-m-diameter centralcore of cryptocrystalline syntheticpericlase, which is surrounded by tworims of differing structure. The innerrim, where the gem-quality crystalsform, is approximately 10 cm wide.One of the habits of these crystals, asdescribed by Browne, is “pseudocubicmasses bounded by cleavage planes,”which exactly describes the rough weexamined.

The process for producing syn-thetic periclase differs greatly from allother known growth processes forsynthetic gems, and the internal char-acteristics generated by electrofusionare not well documented. Square,plate-like negative crystals wereobserved in some samples of knownsynthetic periclase seen in the GemTrade Laboratory in 1969, a materialmarketed under the name Lavernite(“Developments and Highlights at theGem Trade Lab in Los Angeles,”Gems & Gemology, Spring 1969, p.22.) However, since natural periclaseis so rarely found in gem-quality crys-tals of any size, the inclusions weobserved could not assist in identify-ing this piece as natural or synthetic.In this instance, we reasoned thatsince the size and color of this speci-men were so far removed from anypericlase that has been found in

nature, this item must be synthetic.As discussed in the earlier entry, wehave not found definitive criteria toestablish whether a smaller facetedpiece of such material would be natu-ral or synthetic.

Maha DeMaggio and SFM

Editor's note: An 18.98 ct yellowishgreen pear-shaped brilliant, examinedin the East Coast lab at press time,also turned out to be synthetic peri-clase. It had the same R.I., S.G., andchemical constituents describedabove; was inert to long-wave UV, butfluoresced an extremely weak orangyyellow to short-wave UV radiation;and showed weak red transmission tothe fiber-optic light. With magnifica-tion, we saw a pinpoint cloud and asmall round bubble. IR

QUARTZ

Quench-Crackled and Dyed toImitate AmethystLast winter, a client called the EastCoast lab to ask us about an allegednew treatment that uses lasers toinduce a purple color in rock crystal.The client stated that this treatedmaterial had been offered for a priceeven lower than that of syntheticamethyst. Although we were awarethat radiation might be able to inten-sify the purple color in amethyst(G. R. Rossman, “Color in Gems: TheNew Technologies,” Gems & Gem-ology, Summer 1981, pp. 60–71), wedid not know of any mechanismwhereby a laser could induce purplecolor in near-colorless quartz. At ourrequest, then, the client sent us twobeads for examination (figure 11).

The beads were light to mediumpurple, but even to the unaided eyethe color looked uneven. With magni-fication, we observed that the materi-al was fractured throughout, with thepurple color concentrated in thesefractures (figure 12). Small flat sur-faces on the beads allowed us to mea-sure the refractive indices at 1.54 and1.55, indicating quartz. When the

Figure 10. The shape of this115.32 ct piece of rough synthet-ic periclase was dictated bywell-developed cubic cleavage.


beads were exposed to both long- andshort-wave UV radiation, the frac-tures fluoresced a weak to mediumorange, whereas the solid areas of thematerial were inert. The mid-infraredspectrum showed several overlappingpeaks between 3000 and 4000 cm-1,which are seen in all varieties ofquartz, plus additional structurearound 2900 cm-1, which is similar tothe absorptions seen in polymer-impregnated materials. These proper-ties led us to the conclusion that thebeads had been quench-crackled anddyed. The pervasive cracks effectively

hid any other internal features, so wecould not determine whether thestarting material was natural or syn-thetic quartz.

Over the years, we have reportedon quench-crackled quartz in bothgreen and red, imitating emerald andruby (see, for example, Winter 1981Lab Notes, pp. 229–230, and GemNews: Winter 1989, p. 247, and Fall1992, pp. 205–206). However, this isthe first time we have seen quartzdyed purple to simulate amethyst.

IR

TOPAZ

Fashioned to ImitateDiamond RoughIn the hands of a clever lapidary,almost any near-colorless rough mate-rial can be fashioned into an objectresembling a diamond crystal. Wehave reported on a number of cubiczirconia imitations of diamond rough(see Lab Notes, Winter 1988, pp.241–242, and Fall 1996, p. 205; andGem News, Spring 1994, p. 47). Thehigh refractive index of cubic zirconiaresults in a good-looking imitation,but such pieces feel suspicious in thehand because of their high specific

gravity. Recently we reported on animitation made from topaz (LabNotes, Spring 1997, p. 57). With a spe-cific-gravity range that includes thevalue for diamond, such an imitationhas the appropriate heft, but the lowerrefractive index gives a suspectappearance.

In late spring 1997, the EastCoast lab received two specimens foridentification, both of which appearedto be diamond rough in the form ofdistorted octahedra with characteris-tic dodecahedral grooves. The larger(63.65 ct) “crystal” was near-colorless,while the smaller (26.88 ct) stone hadnumerous dark inclusions and anuneven light blue color (see figure 13).The stones did not have an adaman-tine luster, but they did show agreasy-looking patina similar to thatsometimes seen in waterworn dia-mond crystals. Although the specificgravities, measured hydrostatically,were 3.55 and 3.54, respectively, noneof the other properties was consistentwith those of diamond.

In fact, the optical and physicalproperties revealed that the true iden-tity of these two specimens wastopaz. Both were doubly refractive,and we found a biaxial optic figure inthe larger piece. Although the rough-ened surfaces were not ideal for R.I.measurement, we obtained values of1.60 and 1.61 by the spot method.Both pieces showed weak yellow fluo-rescence to long-wave UV radiationand no reaction to short-wave UV.Hardness tests on discreet areas ofeach piece of rough showed them to

Figure 11. These purple beads, each about 10 mm in diameter, wereoffered as amethyst. They are actually colorless quartz that has beenquench-crackled and dyed purple.

Figure 12. Magnification readilyrevealed the extensive networkof fractures and the concentra-tion of purple color in the frac-tures of the beads shown in fig-ure 11. Magnified 10×.

Figure 13. These two “crystals,”fashioned to look like diamondrough, were identified as topaz.They weighed 63.65 and 28.88 ct.


be between 7 and 9 on the Mohsscale. Neither piece showed any fea-tures in the hand spectroscope.

Using magnification, we identi-fied the dark inclusions in the smallerpiece by their crystal habit and blue,gray, and green colors as tourmalineand chlorite (figure 14), which are not

known to occur in diamond. Whenthis same stone was examined withdiffused light, it was apparent thatmost areas were near-colorless andthe overall blue color was caused bythe inclusions. The trigons on bothpieces also showed a number of char-acteristics that are not seen in dia-

mond: They were raised above thesurface rather than depressed into it,were different in shape from the faceson which they occurred, and (asshown in figure 15) were not inverted(as described for natural diamond inY. L. Orlov, The Mineralogy of theDiamond, John Wiley and Sons, NewYork, 1973, especially pp. 72–106). Inreflected light, we saw weak polishinglines on some areas of the 26.88 ctpiece, which proved that these sur-faces were not naturally occurring.

It is evident that a great deal oflapidary artistry was used to manufac-ture these specimens. They serve toremind us that a gemologist mustnever be swayed by the “obvious”appearance of an unknown specimenbefore reaching a conclusion.

GRC and IR

PHOTO CREDITSNicholas DelRe supplied the pictures used in fig-ures 1, 2, 7, 8, and 11–15. Shane McClure provid-ed figures 4 and 5. Maha DeMaggio supplied thephotos used in figures 6, 9, and 10; and the pictureused in figure 3 is courtesy of Sotheby’s.

Figure 14. The many long, narrowinclusions in this topaz are tour-maline and chlorite crystals.They give an overall blue color tothe otherwise near-colorlesstopaz host. Magnified 15×.

Figure 15. The shape, orienta-tion, and raised relief of the“trigons” on this fashionedtopaz do not reproduce theappearance of actual trigons ondiamond crystals. Magnified 5×.

220 Gem News GEMS & GEMOLOGY Fall 1997

DIAMONDSExtreme wear on diamonds. A turn-of-the-century broochwas brought to one of the contributing editors (EF) by M.Bouchet––a Nantes, France, jeweler who specializes inestate pieces––to confirm that the gems in the piece wereindeed diamonds. The brooch contained several near-col-orless round brilliants, all of which showed extreme wearon the facet junctions surrounding their tables. Using athermal probe diamond tester, Drs. Fritsch and B. Lasnier,of the University of Nantes, easily proved that all thestones were natural diamonds. The center stone,unmounted, is shown in figure 1. This stone is even moredamaged than the one mentioned in the Fall 1983 GemTrade Lab Notes (p. 172). Note that the wear is muchgreater on one side, where it extends almost halfwaydown the star facet. Most remarkably, all the stones inthe brooch showed the same pattern of abrasion, with themost abraded parts of each in the same orientation rela-tive to the brooch. Damage was also visible on the tablesof most of the stones as scratches, nicks, and small feath-ers extending into the diamonds. The worn-down surfaceswere quite smooth.

The jeweler told Dr. Fritsch that sometime after thepurchase the customer had returned the brooch, claimingthat the stones were not diamonds and asking to be reim-bursed. The customer had shown the piece to another jew-eler, who stated that faceted stones with so much wearcould not possibly be diamonds, but were most likely zir-con imitations. Aside from the issue of hardness, the abrad-ed edges along the table of the diamond in figure 1 make itdifficult to discern the table reflections in this stone. Thus,the important visual "test" for diamonds––that is, thatmany diamond imitations do not "look right" because thereflection of the table in the pavilion facets is the wrongsize (for stones with different refractive indices) or evendoubled (in most orientations of birefringent materials,such as zircon)––would not be reliable in this instance.

The reason for this extreme wear remains a mystery,as no information was available on the history of thebrooch.

How many diamonds are there in Arkansas? Professor A.A. Levinson, of the University of Calgary, Alberta,Canada, has provided the following analysis of develop-ments at the Crater of Diamonds State Park, nearMurfreesboro, Arkansas.

Although diamonds were found in a pipe nearMurfreesboro in 1906, various attempts to mine themfrom 1907 to 1932 always ended in financial loss. In1972, the locality became the Crater of Diamonds StatePark, a tourist attraction where members of the publiccould keep all the diamonds they found. Over the past 20years, about two stones––with an average weight ofapproximately 0.2 ct––have been found daily (see, e.g.,“Arkansas's Diamond Park: Challenge of the Hunt ver-sus Chance for Big Profits” by R. Shor, Jeweler'sCircular-Keystone, November 1993, pp. 56–57). In thisera of frantic exploration for new diamond deposits, itwas therefore natural to reconsider the Murfreesborolamproite pipe for its economic potential. However,there had never been a reliable estimate of the ore grade(carats per ton) of the pipe, nor an accurate determinationof the value of the stones.

In 1987, the Arkansas state legislature passed a lawallowing commercial mining at the site. In September1996, after approval was received from the U.S. NationalPark Commission, and after the failure of litigation byvarious third parties seeking to halt the project, a groupunder the auspices of the State of Arkansas Departmentof Parks and Tourism began to collect an approximately10,000 ton bulk sample of the lamproite for evaluation ofthe ore. The final results of this evaluation were given inan Ashton Mining of Canada Inc. press release datedFebruary 10, 1997.

Citing information that was provided in this andprevious press releases, Professor Levinson noted that thebulk sample was collected from 18 trenches in thoseparts of the pipe that were the historical sources of mostof the recorded diamonds. From a total of 8,819 tons ofore, 207 diamonds were recovered (total weight 45.75 ct,for an average weight per stone of 0.22 ct). This calcu-

Gem News GEMS & GEMOLOGY Fall 1997 221

lates to an average diamond content of 0.005 carat perton of ore.

These results fall significantly short of the diamondcontent required to support a commercial mining opera-tion (e.g., typical economic kimberlite pipes average0.3–0.5 ct/ton). Clearly, this pipe can remain a touristattraction without the complication of concurrent min-ing.

Professor Levinson suggests that the tourists aremining more diamonds than the bulk sample indicatesthe source should provide naturally. From the informa-tion given, he notes, about 0.40 ct of diamonds are recov-ered daily by tourists. This is equivalent to the diamondcontent of 80 tons(!) of lamproite at this deposit (i.e., 0.40ct divided by 0.005 ct/ton), leading to an interesting ques-tion as to how some of these diamonds reached the site.

COLORED STONES AND ORGANIC MATERIALS

Prospecting for beryl in Saudi Arabia. There are manyinstances in the Arabic literature of emeralds and otherberyls being found in Saudi Arabia; these were compiledin a 1989 report by B. A. Al-Fotawi (“A Brief Review ofthe Mines and Mineral Occurrences in the ArabianPeninsula (Hamad Al-Jasear),” Saudi Arabian Ministryfor Mineral Resources Open File Report DGMR-of-09-18,Kingdom of Saudi Arabia, 19 pp.). While working for theMinistry of Petroleum and Mineral Resources, Faisal M.Allam investigated some of these occurrences, andreported his findings to the Saudi Arabian government in1992 (“Gemological Investigation for Gemstone––Emerald/Aquamarine in the Arabian Shield,” Technical

Report DGMR-TR-91-9 of the Ministry of Petroleum andMineral Resources). Mr. Allam recently became a GIAstudent, and generously agreed to share the results of thishard-to-find report with our readers.

Mr. Allam studied the Hadiyah area, which report-edly was a site of ancient emerald mining; however, theonly green mineral he found was epidote, not emerald.Other varieties of beryl had been described from 12 local-ities (pegmatite dikes) in the Precambrian Arabian Shieldrocks in Saudi Arabia between Riyadh and the Red Sea;these localities were explored in more detail. Althoughnone showed potential for large-scale gem mining, somegem-quality aquamarine was obtained from the KarathWell pegmatite, and small faceted aquamarines could befashioned from gemmy crystals (up to 8 cm long by 1.5mm wide) found at Jabal Tarban. Specimen-quality aqua-marine crystals may be recovered from both of thesesites, as well as from Sarat Bishah and Jabal al Hawshah.Two other sites had smaller or scarcer beryl samples,while no beryl was found in the remaining six historicoccurrences. Three good-quality aquamarines in the 1–2ct range were cut from Karath Wells material.

Lab Alert: Radioactive cat's-eye chrysoberyls. Severalhundred carats of radioactive cat's eye chrysoberyl arebeing offered on the Bangkok market, according toKenneth Scarratt, president of the Centre for GemstoneTesting (the successor of the AIGS Laboratory) in thatcity. The material was discovered in late August 1997,when a Bangkok dealer (J. Bergman, of Gem Essence Co.),requested that some brown cat's-eye chrysoberyl be test-ed for radioactivity.

The Centre for Gemstone Testing examined one 3.5ct stone and found it to be highly radioactive, with anactivity level of 52 nCi/g. This is significantly higherthan the legal release levels set by the relevant authori-ties in the United States (1.0 nCi/g), United Kingdom (2.7nCi/g), and Asia (2.0 nCi/g). Using a Geiger-Muellercounter, Mr. Scarratt found that the stone had a contactradiation level of about 11 milliroentgens per hour.Subsequent tests showed that it had a half-life of approxi-mately 103 days, indicating that this particular stonewould not reach the legal release level in Asia beforeJanuary 1999. Until then, it must be kept in a properlyshielded radioactive materials storage container.

The chrysoberyl in question was described as havingan “unusual dark brown” color (figure 2), which wasfound to be stable to a quick “fade test”––heating in thewell of a microscope (until too hot to hold) for a period ofone hour. The original material reportedly came fromOrissa, India, and Mr. Scarratt believed that it was bom-barded with neutrons in a nuclear reactor somewhere inAsia and then released illegally. According to one sourcewith whom Mr. Scarratt spoke, “hundreds” of thesestones have been available in Indonesia since April 1997,and by early September still more material had changedhands at the gemstone marketplace in Chantaburi,

Figure 1. Note the extreme and uneven wear on thisdiamond (about 4.5 mm in diameter), which was thecenterpiece of a brooch in which all the diamondsshowed similar abrasion. Photo by Bernard Lasnier.


Thailand. Robert E. Kane, director of the GübelinGemmological Laboratory in Lucerne, Switzerland,reported seeing several examples of this material at theSeptember 1997 Hong Kong Show. Mr. Scarratt advisedthat members of the gem trade immediately check allcat's-eye chrysoberyls of an unusual dark brown color forexcessive radioactivity.

Demantoid garnets from Russia . . . The greater availabil-ity of fine Russian demantoid was described by W. R.Phillips and A. S. Talantsev in the Summer 1996 Gems& Gemology (“Russian Demantoid, Czar of the GarnetFamily,” pp. 100–111). The editors saw an excellentdemonstration of the return of this material earlier thisyear. Bear and Cara Williams of Bear Essentials, JeffersonCity, Missouri, showed us several bright green deman-toids that were reportedly from the Bobravka River, nearNizhniy Tagil. These stones had been fashioned in thenearby Ekaterinburg region. We borrowed two stones,

weighing 1.34 and 1.18 ct (figure 3), for gemological char-acterization.

Gemological properties were as follows (largeststone first, where different): shape––oval brilliant, ovalmixed cut; color––yellowish green; color distribu-tion––even; pleochroism––none; optic character––singlyrefractive; Chelsea color filter reaction––deep orangy red,weak orangy red; R.I.––greater than 1.81; S.G.––3.87,3.86; inert to both long- and short-wave ultraviolet radia-tion; no visible luminescence (“transmission lumines-cence”); absorption spectrum––460 nm cutoff, 620 nmline, 638 nm line, weak 690 nm line (difficult to see inthe large stone), 680 nm cutoff (large stone only); inclu-sions––”horsetails” (both stones), fractures (both stones),large cavity on girdle (large stone only). An energy-dis-persive X-ray fluorescence (EDXRF) spectrum taken onthe largest stone showed major Si, Fe, and Ca, minor Cr,and trace V.

. . . . and from Namibia. A number of people havebrought to our attention the fact that demantoid gar-nets––long thought to occur almost exclusively in Russia(see, e.g., the Phillips and Talantsev article cited in theprevious entry)––are now commercially available from anew locality in the southern African nation of Namibia.Contributing editor Henry Hänni first brought thesestones to our attention. In spring 1997, he received twodemantoid garnet crystals that were reportedly from anew source in Namibia; they were sent by Mr. HilmarBosch, Hilton, Natal, South Africa. The two crystalsweighed 13.52 and 3.87 ct (Mr. Bosch reported seeingcrystals as large as 30 ct). They had the following proper-ties (largest first, where different): shape––both were par-tial rhombic dodecahedra (figure 4); surface tex-ture––stepped growth in small steps, with featuresresembling slight corrosion; color––light yellowish green,

Figure 2. Material similar to the pale green cat's-eyechrysoberyl on the left was irradiated to produce the3.5 ct radioactive dark brown cat's-eye chrysoberyl onthe right. Photo courtesy of Kenneth Scarratt.

Figure 3. These twodemantoid garnets, 1.18

and 1.34 ct, were recentlyrecovered from Russian

deposits. Stones courtesyof Bear Essentials; photo

by Maha DeMaggio


grayish green (both resembling peridot in color); colordistribution––yellower at the core (in the first stone),even (in the second); absorption spectrum––sharp band at425; reaction to Chelsea filter––reddish. With magnifica-tion, we saw open fissures in both crystals and some tinyfluid inclusions in the smaller piece; however, no “horse-tail” inclusions were seen in either crystal. EDXRF spec-troscopy gave Si, Ca, and Fe––consistent with andra-dite––but very little chromium (the green chromophore),which was estimated at 0.05 wt.% Cr2O3. An article byThomas Lind et al. in a recent issue of the Zeitschrift derDeutschen Gemmologischen Gesellschaft (“NeuesVorkommen von Demantoid in Namibia,” Vol. 46, No.3, 1997, pp. 153–160) reported between 0.02 and 0.13wt.% Cr2O3 in this material.

Marc Sarosi, a gem dealer in Los Angeles, California,shared seven fashioned demantoids (0.71–3.42 ct), twosmall demantoid crystals (8.91 and 9.54 ct), and one 1.11ct yellowish brown andradite from this locality with theGem News editors (some of these stones are shown infigure 5). Gemological properties of the fashioned deman-toids were as follows: color––green to yellowish green;

refractive index––over the limits of our standard refrac-tometer (greater than 1.81); optic character––singlyrefractive with strong anomalous double refraction; inertto both long- and short-wave ultraviolet radiation;Chelsea filter reaction––orange to red; specific gravi-ty––3.83–3.85; absorption spectrum (using a handheldspectroscope)––band at 445 nm, general absorption below450 nm, and (one stone only) diffuse 580 and 630 bands.Magnification revealed strong dispersion, “fingerprints,”crystals (some of which could be resolved as prismatic inshape), a negative crystal with a two-phase inclusion, andneedles in two samples. No sample contained “horse-tail” inclusions, although the 1.11 ct yellowish brownandradite contained yellow needles (some of whichcurved slightly); Lind et al. also found no “horsetails” inthe Namibian demantoids they examined. The stonesshowed pronounced angular or straight transparentgrowth zoning, and, in one case, iridescence along thegrowth planes (so-called “rainbow graining”). EDXRFanalysis of the green fashioned stones revealed major Ca,Fe, and Si, and minor Mn and Cr. The two rough speci-mens we examined showed parallel growth of smallercrystals, with smooth dodecahedral faces and etchedtrapezohedral faces; one piece had grown on a 5 mmwhite “scepter” quartz crystal.

A published report in Jewellery News Asia (August1997, p. 36) states that the source for these demantoidsand other andradites is the Usakos mine in Namibia, andthat the largest stone fashioned to date weighs 9.89 ct.The deposits are reportedly alluvial in nature.

A gem-quality ettringite group mineral, probably stur-manite. Each February, the gem and mineral shows inTucson provide many interesting, sometimes important,gemological discoveries. Although some of these are rela-tively easy to analyze and describe, others are much more

Figure 4. These two crystals (3.87 and 13.52 ct) aredemantoid garnets from a new locality in Nami-bia. Photo courtesy of Henry Hänni.

Figure 5. The newly dis-covered Namibian

deposit reportedly pro-duced these rough andfashioned demantoidsand yellowish brown

andradite from the newfind in Namibia. Thelargest crystal (right)

measures 13.91 × 12.06× 7.27 mm, and thelargest pear-shaped

demantoid weighs 1.95ct. Stones courtesy of

Marc Sarosi, LosAngeles, CA; photo by

Maha DeMaggio.


challenging. Such was the case with the 3.82 ct bright yel-low translucent oval cabochon shown in figure 6.

The original 10.12 ct piece of rough came fromKevin Lane Smith, a Tucson gem and mineral dealer. Mr.Smith also had a fist-size, highly translucent piece of thismaterial; a large, well-polished translucent cabochon;and a small faceted stone. The material was reportedlyfrom the Kuruman District in South Africa, and it wasrepresented as ettringite, a member of the mineral groupof the same name. Ettringite-group minerals are hydrousmixed sulfates/carbonates/borates/hydroxides of calciumand other elements (e.g., Al, Cr, Fe, Mn, and Si); identify-ing specific minerals in this group is difficult, so manyspecimens are simply labeled “ettringite group.” Thus,we thought that gemological identification of this mate-rial would be a good exercise in gem-testing technique.

Gem Trade Lab staff gemologist Phil Owens fash-ioned the rough into a 14.00 × 10.22 × 5.12 mm ovalcabochon (figure 6); he commented that the material wasvery soft (ettringite has a hardness of 2–2.5) and difficultto polish. In part, this latter difficulty was caused bynumerous inclusions of tiny, transparent-to-translucent,near-colorless crystals, which were appreciably harderthan the host material. As a result, the softer mineralundercut around the surface-reaching inclusions.

The basic gemological properties obtained from thecabochon were of little help in determining whether itwas ettringite or sturmanite. The spot R.I. reading of 1.50was consistent with a reading for either mineral. Thespecific gravity, determined by the hydrostatic method,was 1.86, slightly higher than the 1.847 value for stur-manite (see D. R. Peacor et al., “Sturmanite, a Ferric Iron,Boron Analogue of Ettringite,” Canadian Mineralogist,Vol. 21, 1983, pp. 705–709) and considerably higher thanthe 1.77 value for the type specimen of ettringite. Theslightly elevated value (relative to sturmanite) could becaused partly by the many tiny inclusions previouslymentioned, which contributing editor Dino De-Ghionno––using X-ray powder diffraction analysis––iden-tified as calcite.

EDXRF analysis, performed on the finished cabo-chon by former GTL Research Technician DijonDouphner, showed that the yellow material containedAl, Ca, Fe, Mn, and S, with possible traces of As, Cu, Pb,K, and Sr. The iron and manganese content appeared toohigh for ettringite, which nominally does not containthese elements. (Note that GIA's EDXRF system doesnot detect oxygen, hydrogen, or boron.) However, thequalitative chemistry fit very well with the chemical for-mula for sturmanite.

A small powder sample, scraped from the rough byMr. DeGhionno, was used for X-ray powder diffractionanalysis. The pattern that we obtained matched the linepositions and relative intensities listed for sturmanitemore closely than it matched those for the mineralettringite.

Thus, much of our data pointed to sturmanite as thecorrect identity of this bright yellow cabochon. However,because the ettringite group is a solid solution (as is, forexample, garnet), the chemistry and nature of the materi-al can vary significantly, even within a single sample.Consequently, our conclusion on a GTL report wouldread: “A member of the Ettringite mineral group, proba-bly Sturmanite.”

Bicolored grossular-andradite garnets from Mali. When“Gem-Quality Grossular-Andradite: A New Garnet fromMali” (M. L. Johnson et al., Fall 1995 Gems & Gemology)was written, we had seen these garnets in three colortypes: orange to brown, somewhat desaturated yellow-green, and bright green (see the cover of the Fall 1995issue for examples). In 1996, Bank et al. (“GemmologieAktuell,” Zeitschrift der Deutschen GemmologischenGesellschaft, Vol. 45, No. 1, pp. 1–4) predicted that bicol-ored yellowish green and dark brown grossular-andraditegarnets were also possible. Coincidental with the publi-cation of that article, in fact, two bicolored emerald-cutgrossular-andradite garnets from Mali were loaned to theGem News editors by David Knecht of RKG,Minneapolis, Minnesota. The larger, 2.57 ct, stone was a“classic” bicolor––brownish yellow on one side andgreenish yellow on the other. The smaller, 1.22 ct, stonehad a yellowish green core (figure 7) surrounded on allsides by green-yellow garnet; the core was especially dis-tinct in polarized light. The gemological properties were(larger stone first, where different): optic character––singly refractive, with medium-order (2.57 ct) or high-order (1.22 ct) strain; color-filter reaction––brownish red,none; R.I.––1.770 (2.57 ct stone, greenish yellow region),1.768 (2.57 ct stone, brownish yellow region), 1.761 (1.22ct stone); S.G.––3.66, 3.64; fluorescence––inert to bothlong- and short-wave UV; absorption spectrum––415 nmcutoff, 440–450 nm band, faint absorption at 500 nm(2.57 ct stone); faint 415 nm band, 440–450 nm band(1.22 ct stone). According to EDXRF results, both stonescontained Al, Si, Ca, Fe, Ti, Cr, and Mn. Green cores(like the one in the 1.22 ct stone) also have been seen in

Figure 6. This 3.82 ct yellow cabochon is a mineralin the ettringite group, probably sturmanite. Photoby Maha DeMaggio.


some near-colorless grossular garnets from Asbestos,Quebec (see, for instance, W. L. Roberts, G. R. Rapp Jr.,and J. Weber, Encyclopedia of Minerals, 1st ed., VanNostrand Reinhold Co., New York, 1974, plate 60).

More on opal from Shewa, Ethiopia. Dr. Don Hoover (for-merly of the U.S. Geological Survey) recently sharedwith us two samples of Ethiopian opals that have some-what unusual properties, as compared to other opals andespecially to those described in the Johnson et al. articleon this locality (Gems & Gemology, Summer 1996, pp.112–120). As illustrated in figure 8, thin “tubes”––suchas the ones shown in figure 10 of the 1996 paper––per-vaded the approximately 2 cm chunk of rough opal. Thetubes occupied the same volume as patches of play-of-color in this opal; however, the play-of-color regions didnot appear to be bounded, warped, or disturbed in anyway by the presence of the tubes. This relationshipimplies that the tubes were older than the final aggrega-tion of opal spheres into organized regions that causedthe play-of-color diffraction effect.

The second piece, an approximately 5 ct chunk ofwhite opal that also showed play-of-color, was interestingbecause of its very low apparent density. However, thespecific gravity could not be measured by standard hydro-scopic techniques, as the material rapidly took up water.With Dr. Hoover's permission, we attempted to slice acube of the material in order to determine density bydirect comparison of weight and volume (weight dividedby volume equals density). Although the material crum-bled while being sliced, we ended up with four fairly regu-lar pieces of 1.90, 0.88, 0.45, and 0.41 ct. GTL weights-and-measures coordinator Christopher Lewis then ran eachpiece on the Sarin DiaMension, a noncontact measure-ment device (used for determining the proportions of

faceted diamonds) that can calculate the volume of anyappropriately sized sample having a fairly simple convexshape (that is, one having no holes or dimples; for instance,a heart-shaped brilliant is not convex). The four pieces hadcalculated densities of 0.81, 0.79, 0.68, and 0.67 grams percubic centimeter. Such low densities have not beenrecorded for gem opals, but they have been seen in“tabasheer,” opaline siliceous material formed in thestems of bamboos and other grasses (see, e.g., C. Frondel,The System of Mineralogy of James Dwight Dana [etc.],7th ed., Vol. 3, 1962, pp. 287–306). The 0.45 ct piece wasthen placed in a beaker of water to see whether it sank orfloated: It floated for about 30 seconds, then sank rapidlyin a cloud of small bubbles. The piece became somewhatmore translucent and grayer after half an hour in water,but it maintained its play-of-color and was not damagedby its soaking. It reverted to “normal” once it dried out.

Gem rhodonite from Australia. Many minerals have verysimilar gemological properties, and can only be separatedwith great difficulty. Examples include: the rare mineralgenthelvite, which requires X-ray diffraction or chemicalanalysis to distinguish it from pyrope-almandite garnet(Fall 1995 Gem News, pp. 206–207); members of theamphibole group, which often cannot be distinguishedfrom one another even with X-ray diffraction and EDXRFanalysis; and members of the tourmaline group, whichare usually identified simply as “tourmaline” instead ofby the species name of elbaite, dravite, liddicoatite, andthe like, because of the difficulty of separating the vari-ous species.

In August 1997, Randy Polk of Phoenix, Arizona,sent Gems & Gemology editor Alice Keller a parcel ofboth rough and fashioned examples of an intense pink

Figure 7. This 1.22 ct garnet from Mali has an intenseyellowish green core. Photo by Maha DeMaggio.

Figure 8. The tubular structures did not disturb theplay-of-color in this rough opal from ShewaProvince, Ethiopia. Photomicrograph by John I.Koivula; magnified 15×.


material (figure 9). This material came from the Woodsmine in a remote area of New South Wales, Australia.When it was last seen on the market, about the early1970s, it was believed to be rhodonite. However, in thepast 20 years, new minerals have been charac-terized––including marsturite, nambulite, and natronam-bulite––which can be separated from rhodonite only withdifficulty. In addition, certain previously identified miner-als (pyroxmangite, bustamite) were commonly mistakenfor rhodonite, or believed to be varieties of rhodonite, inthe past. Thus, the question arose as to whether thismaterial––especially the fine-grained crystalline aggregate(figure 10) constituting the darker pieces––was actuallyrhodonite or some other mineral nearly indistinguishablefrom it. Adding impetus to our examination was a state-ment from Mr. Polk that this material had been analyzedand tentatively identified as natronambulite.

Staff gemologist Philip Owens determined gemologi-cal properties for six (3.01–7.78 ct) cabochons. All sixstones were translucent pink; four were evenly colored,and two (4.69 and 7.78 ct) were mottled with paler pinkregions. All showed an aggregate optic character, and allwere inert to the color filter. The four evenly coloredstones had R.I. values of 1.73 (1.728 on a flat face of arose cut stone) and S.G.'s between 3.41 and 3.72. Thetwo mottled stones had spot R.I.'s of 1.54 and S.G. valuesof 3.00 and 3.17 (such low S.G. values are typical of mas-sive rhodonite, which can have a significant admixture ofquartz). The six stones were inert to both long- and short-wave UV, and they showed no luminescence to visiblelight (“transmission luminescence”). Five of the sixstones showed absorption bands at 420 and 548 nm, aswell as weaker bands at 440 and 455 nm; in the sixth, the455 nm band was not observed. All of these propertieswere consistent with the material being rhodonite.

With magnification, all six stones showed a mottledtexture, consisting of fine to medium grains. Shiny (sul-

fide?) and black opaque (manganese oxide?) patches wereobserved on one sample. Brown stains between grainswere observed in two samples. The largest and mosttransparent crystals were found in a 3.01 ct pear-shapedcabochon (again, see figure 9), which was chosen for fur-ther testing. (This sample also had the highest specificgravity.) EDXRF revealed major Mn, Si, and possibly Ca;minor Sr and Ba; and traces of Zn and possibly Ni. An X-ray diffraction powder pattern was an excellent match toour rhodonite standard pattern (a sample from Australia,pattern measured in 1971) and also to JCPDS pattern13–138, a rhodonite from Franklin, New Jersey. The lat-ter was important because the sample that provided ourrhodonite standard pattern also might have come fromthe Woods mine.

Figure 9. These rough andfashioned pieces of rho-donite from the Woodsmine in Australia illus-trate the range of color inthis material. The largestpiece measures 66.0 ×58.0 × 19.0 mm. Photo byMaha DeMaggio.

Figure 10. With magnification, it is evident that thedark samples in figure 9 are an aggregate of fine-grained deep pink crystals, showing random orientation. Photomicrograph by John I. Koivula;magnified 15×.


Although we are confident that the sample we test-ed is rhodonite (based on the relative intensities in itsdiffraction pattern), there is no guarantee that all of thematerial in this deposit is rhodonite.

Mr. Polk is the exclusive agent for this handsomepink material, which is being mined by John and AlexTaggart. Large quantities are available: About six tons ofvarious qualities is expected to be on the market soon,and Mr. Polk tells us that at least two solid pieces weigh-ing 2 tons each have been found. Mr. Polk and his col-leagues have graded the material into several categoriesbased on transparency and depth of color. “Pencil-sized”unterminated crystals have been found, and some of thematerial may be suitable for mineral specimens. Mr.Polk informs us that the Woods mine has produced sev-eral additional unusual minerals; the mineralogy of thisdeposit warrants further study.

TREATMENTS

“Pink geuda” sapphires from Vietnam and their treat-ment. Ted Themelis (Gemlab Inc., Athens, Greece) hasprovided the following information to the Gem Newseditors. During a March 1997 visit to Vietnam, he notedthe abundance of geuda-type sapphires, locally referred toas “white sapphires,” that had a very pale pink overtone.These reportedly are found in the Hoang Lien Son area,Lao-Cai Province, of northwest Vietnam. Most of thestones were opaque and cloudy, with pronounced milki-ness; they ranged in size to a little over five grams. Rutilesilk was seen in all specimens, either concentrated inpatches or irregularly distributed throughout the stone.Some specimens were reminiscent of Sri Lankan geuda.Mr. Themelis observed that the milkier or more includedthe stone was originally, the more intense was the colorproduced by treatment.

Mr. Themelis performed three heat-treatment experi-ments on rough samples of this Vietnamese “pink geuda.”First, he heated 118 pieces at 1650°C for five minutes inair, followed by rapid cooling (30°C/minute). Of these, 31pieces turned blue, ranging from light to very dark; 54developed patches or zones of blue; 21 turned pink or pur-ple; and 12 pieces turned very light pink. Although thetreatment noticeably improved the diaphaneity of all ofthese samples, they were still semi-translucent to opaque,suitable for cutting only as cabochons.

In a second experiment, Mr. Themelis heated 58pieces of Vietnamese “pink geuda” that had a more no-ticeable pink tinge and obvious milkiness, at 1675°C for10 minutes in pure oxygen, again followed by rapid cool-ing. Of these stones, 28 pieces turned medium-to-intensepink, with improved diaphaneity and luster (althoughstill of cabochon quality); 13 pieces turned medium blue(also cabochon quality); and 17 pieces showed a shift incolor from light purple in fluorescent light to very lightpink under incandescent light (with improved diaphane-ity, but still semi-translucent to opaque).

Last, Mr. Themelis heated 86 pieces of theVietnamese “pink geuda” at 1700°C for 10 minutes in areducing atmosphere, again followed by rapid cooling. Ofthese stones, all still opaque, 51 turned very dark blue,almost black; 23 turned purplish blue to blue; and 12pieces turned mottled blue, with concentrated patches ofblue color all over their surfaces. He believes that thesenewly discovered Vietnamese “pink geuda” sapphiresshow potential for commercial heat treatment, as a goodpercentage of the treated material is suitable for cuttinglow-quality cabochons.

SYNTHETICS AND SIMULANTS

Change-of-color synthetic sapphires represented as yetanother “new find.” The range of colors seen in mostchange-of-color synthetic sapphire is quite distinctive(described as “greenish blue in fluorescent or naturallight and pinkish purple in incandescent light”; see, e.g.,Gem Trade Lab Notes, Summer 1995, p. 127), and usual-ly the material is represented to the unwary as alexan-drite. However, the distinctive set of colors of these syn-thetic sapphires may cause suspicion in the mind of anexperienced gemologist who sees such a stone, regardlessof what it is stated to be.

The three synthetic sapphires in figure 11 were partof a lot of six samples (weighing from 1.32 to 2.28 ct) pur-chased from Afghani “freedom fighters”; these supposed-ly were natural sapphires from a new locality in theregion. The “native cut” faceting they had undergoneadded to their air of authenticity. Examination with amicroscope and a diffused transmitted light sourcequickly confirmed the suspicions of contributing editorShane McClure: Curved striae were present in all three

Figure 11. These three “native cut” synthetic sapphires(weighing 2.03 to 2.28 ct) were represented as change-of-color sapphire from a new locality in or near Afghan-istan. Courtesy of Dr. Horst Krupp, Firegems, La Costa,CA; photo by Maha DeMaggio.

stones. The source of our information, Dr. Horst Krupp,told us that clean pieces of rough as large as 40 gramswere said to be available; no doubt larger pieces wouldhave resembled their parent boules too closely for effec-tive deception.

GGG from Russia. The Spring 1995 Gem News section(p. 70) contained an entry on a number of synthetic andimitation gem materials that were available at the 1995Tucson show. These included materials being offered bythe Morion Co. of Cambridge, Massachusetts; amongthose mentioned in that entry, but not described, wasgadolinium gallium garnet (GGG) in a range of colors.The material was being sold primarily as unpolisheddisks about 80 mm in diameter by 5 mm thick. Whenasked why the material had been pre-formed in this fash-ion, the firm's president, Dr. Leonid Pride, stated that hepurchased them in this shape from a factory in Russiathat originally grew the crystals for use in electronicmemory elements for military applications.

Since 1995, additional colors of GGG have becomeavailable. For instance, in September 1997 rough wasavailable in: “intense pink, pink, light pink, almandine[red?], green-blue, aquamarine blue, sky blue, intenseblue, raspberry, and lilac,” according to Dr. Pride. TheMorion Co. also has had a small number of faceted sam-ples available at the annual Tucson shows; five of these(ranging in weight from 2.41 to 2.75 ct; figure 12) wereobtained for characterization. We determined the follow-ing properties: color––various (see table 1); body colordistribution––even; diaphaneity––transparent; R.I.––overthe limits of the standard refractometer (greater than1.81); S.G. (four of the five samples)––7.11 to 7.14, darkblue stone 6.64 (see below); singly refractive; pleochro-ism––none; magnification––no inclusions noted.Luminescence to ultraviolet radiation, Chelsea filterreaction, and the spectrum seen with the handheld prismspectroscope depended on the color of the stone (again,

see table 1). On the basis of its appearance and theseproperties, GGG is easily separated from any naturalgem materials.

EDXRF analysis showed that the dark blue speci-men contained calcium and zirconium in addition togadolinium and gallium. This detectable difference inchemistry most likely accounts for the significantlylower S.G. All of the specimens showed X-ray diffractionpatterns consistent with GGG.

MISCELLANEOUS

Gems & Gemology author wins European cutting com-petition. Arthur Lee Anderson of Speira Gems, Ashland,Oregon, recently won first prize in the gemstone cuttingcompetition, the “28th German Award for Jewellery andPrecious Stones Idar Oberstein 1997.” Held every threeyears, the competition draws entries from all over theworld and is juried by members from the European gem-stone and jewelery industry. Mr. Anderson's winningentry was a 21.70 ct “Webbed Pear Cut” citrine (figure13), a design that he developed over the past year. To cre-ate this design, Anderson employed several different cut-ting techniques, as well as the use of optics and reflec-tions in the citrine. Mr. Anderson described some of hisfashioning techniques in the article “Curves and Optics


TABLE 1. Color-dependent properties of five samples of GGG produced in Russia.

UV fluorescence Chelsea Absorption

Color Long-wave Short-wave filter spectrum (nm)

Purple Weak Weak No Weak absorptionorangy red orangy red reaction at 500–520

and 530–550,strong bands at 570–580 and 610–620,650 cutoff

Purplish Weak Weak No Sharp lines at pink orangy red orangy red reaction 465, 485, 515,

520, 540; weak 550; doublets at530, 570, 580and 590

Orangy Very weak Weak red Red Vague absorp-red red tion at 400–440

and 500–600;690 cutoff

Green- Inert Inert No Weak 540 band;blue reaction moderate bands

at 570 and 620;650 cutoff

Blue Inert Inert Red Weak 440 and 455 bands; 460–470, 590–620,and 640–690bands

Figure 12. These five samples of gadolinium galliumgarnet (2.41–2.75 ct) were fashioned from material

reportedly produced in a military facility in Russia.Photo by Maha DeMaggio.


in Nontraditional Gemstone Cutting,” which appearedin the Winter 1991 issue of Gems & Gemology (pp.234–239).

Standards issued for the jewelry industry in China. TheChinese State Bureau of Technological Supervision(CSBTS) issued three national standards for the jewelryindustry on October 7, 1996; they went into effectthroughout China on May 1, 1997. The three documentsdefining these standards (GB/T 16552-1996 Gems––Nomenclature, GB/T 16553-1996 Gems–– Testing, andGB/T 16554-1996 Diamond Grading) were drafted by theNational Gemstone Testing Center in Beijing, and arewritten in Chinese; research associate Yan Liu of GIAResearch provided us with the following summary.

The first, GB/T 16552-1996 Gems––Nomenclature,specifies the rules for naming gemstones. This standardis usable for identifications, as well as for descriptionsused for imports and exports, insurance, and the trade.The term gems refers to both natural and manufacturedgem materials. Natural gems include single-crystal (ortwinned) gem materials (“natural gemstones”); naturalaggregate and amorphous materials, including jadeite,nephrite, chalcedony, opal, serpentine, and natural glass(all of these are included in the heading “natural jades,”which could confuse Western purchasers); and “naturalorganic substances.” Manufactured materials––includingsingle-crystal synthetics, assemblages, “reconstructedstones,” and imitations––are placed in the category of“artificial products.” Enhancement is also defined, as areoptical phenomena. The Chinese standard was createdwith AGTA, CIBJO, and All Japan Gem Society stan-dards taken into consideration, and the documentincludes a table translating between English and Chinesegem names.

The second document, GB/T 16553-1996 Gems––Testing, provides terminology and methods for gem iden-tification, and can be used as a gem identification manu-

al. It contains identification criteria for 49 natural gem-stones, 31 natural aggregate or amorphous materials,eight natural organic substances (including petrifiedwood, but not ivory), and 17 “artificial stones.”Enhancements described include heating, bleaching,waxing, the use of colorless and colored oils, filling andimpregnation, dyeing, irradiation, laser drilling, coating,and surface diffusion.

The last standard, GB/T 16554-1996 DiamondGrading, is a manual for the grading of the color, clarity,cut, and weight of a faceted diamond. This standard isonly suitable for grading natural colorless to light yellow(or light brown, or light gray) diamonds of 0.20 ct or larger.Twelve color grades are used (D through N, and below N),and master stones are used to judge color and fluorescence.Clarity is evaluated under a 10× loupe, using “European”terminology: LC (loupe clean), two grades of VVS, twogrades of VS, two SI grades and three P (piqué) grades.There are three cutting grades based on diamond propor-tions. The standard also specifies notation for diamondplots and explains how to grade mounted diamonds.

The term natural jades, used in the first two stan-dards to classify aggregate or amorphous materials,requires some additional discussion. The Chinese termyu, although often considered equivalent to the Englishjade, in fact might be better translated throughoutChinese history as “most resistant carvable stone.”Although many of the aggregate materials included inthis classification might only cause puzzlement in theoverseas gem trade if they were described as “naturaljades” (for instance, turquoise or rhodochrosite), thedescription as “natural jade” of some materials discussedin these standards (e.g., serpentine, quartzite, aventurinequartz) could generate considerable, and potentiallyexpensive, confusion. To prevent misuse of this term,the Nomenclature standard puts further restrictions onthe use of “yu” or “jade” to describe materials. No mate-rial––even nephrite or jadeite––can be called simply

Figure 13. This 21.70 ct“Webbed Pear Cut” citrinebrought gem cutter ArthurAnderson top honors in the28th German Award forJewellery and PreciousStones. Photo by ManfredGrebel.


“jade” or “jade rock;” thus, nephrite is “nephritejade”––but serpentine is also “serpentine jade.” Materialscannot be described by their shapes: “round jade” is notan acceptable definition. Except for specific, well-definedmaterials cited in the standard, place names are notacceptable in jade descriptions. Hence, “Yunnan jade” isnot a valid term. (One exception to the “no place names”rule is “Dushan jade,” which is the accepted name inthese Chinese standards for a specific zoisite/plagioclasefeldspar rock.)

ANNOUNCEMENTS

Dates set for February 1998 Tucson shows. The Am-erican Gem Trade Association (AGTA) GemFair will runfrom Wednesday, February 4, through Monday, February9, 1998, at the Tucson Convention Center. Followingthat show at the Convention Center will be the TucsonGem and Mineral Society show from February 12 to 15.At the Holiday Inn City Center (Broadway), fromFebruary 4 to 11, will be the Gem & Lapidary DealersAssociation (GLDA) show. The Gem and JewelryExchange (GJX) show will run from February 5 to 12,across Grenada Street from the Holiday Inn City Center.Other show venues featuring gem materials include the:Pueblo Inn, Rodeway Inn, Tucson Showplace, SonoranDesert Marketplace, Discovery Inn, Holiday Inn Express,Quality Hotel and Suites, Tucson Scottish Rite Temple,

Howard Johnson’s, Tucson East Hilton & Towers, LaQuinta Inn, The Windmill Inn, Southwest Center forMusic, Holiday Inn Palo Verde/Holidome, and Days Inn.Times and dates of shows vary at each location. Consultthe show guide, which will be available at the differentTucson venues, for further information.

Visit Gems & Gemology in Tucson. Gems & GemologyEditor Alice Keller and Senior Editor Brendan Laurs willbe staffing the Gems & Gemology booth in the Galleriasection (middle floor) at the Tucson Convention Centerfor the duration of the AGTA show, February 4–9. Stopby to ask questions, share information, or just say hello.Many back issues will be available.

International Society of Appraisers. The 19th AnnualInternational Conference of the International Society ofAppraisers (ISA) will be held March 22–25, 1998, in SanDiego, California. The ISA is a multidiscipline profes-sional association, with members in the United States,Canada, and other countries; it is dedicated to advancingthe theory, principles, techniques, and ethics of apprais-ing personal property. Founded in 1979, it is the largestassociation of its kind. At the upcoming 19th program,concurrent programs will focus on gems and jewelry,antiques and residential contents, as well as fine art.Further information about this conference is available onthe internet at http://www.isa-appraisers.org.

Proving once again that Gems & Gemology readers are dedicated

to their education and knowledge of the field, almost 400 people particpated in the 1997 Gems & Gemology

Challenge. Entries came from all over the world, as readers tested their knowledge on the questions from the

Spring 1997 issue. Those who earned a score of 75% or better received a GIA Continuing Education Certificate

recognizing their achievement. Those listed below received a perfect 100% score. Congratulations!

USA: Alabama Guntersville: Cindy K. Fortenberry; Mobile: Lori Bryant �Alaska Craig: Deborah Anne Welker �Arizona

Flagstaff: Robert Brent Fogelberg; Fountain Hills: Hank Wodynski; Scottsdale: Kenyon V. Painter, Norma B. Painter;

Tucson: Dave Arens, Luella Dykhuis, Geraldine Alex Towns � California Burlingame: Sandra MacKenzie-Graham;

Carlsbad: Jan L. Arnold, Marla Belbel, Lori J. Burdo, Carl Chilstrom, Diane Flora, Brian I. Genstel, Becka Johnson, Mark

S. Johnson, Douglas Kennedy, Wendi Mayerson, Jana E. Miyahira, Kyaw Soe Moe, David Peters, Diane Saito, Jim Viall,

Melissa Watson-Lafond, Michael Wobby, Philip G. York; Carmel: Douglas R. Mays; Chino Hills: Virgilio M. Garcia, Jr.;

Elk Grove: Michael Pace; Los Angeles: Todd Buedingen, Grace Lee; Los Gatos: G. Donald Eberlein; Mill Valley: Susan

Bickford; Oceanside: LaVerne M. Larson; Palo Alto: Joyce Henderson; Penn Valley: Nancy Marie Spencer; Sacramento: J.

Marlene White; San Diego: Tracy Nuzzo; San Jose: Willard C. Brown; San Rafael: Robert A. Seltzer; Santa Clarita:

Beverly Nardoni-Kurz; Santa Cruz: Tony Averill, Linda M. Bork; Santa Rosa: Keith M. Davie; Santa Ysabel: Glenn

Shaffer; Tustin: Alvin Ip; Ukiah: Charles “Mike” Morgan; Walnut Creek: Michael W. Rinehart; Watsonville: Janet S.

Mayou; West Hills: Bradley A. Partington � Colorado Colorado Springs: Molly K. Knox; Crested Butte: Nancy Y.

Blanton; Denver: Alan J. Winterscheidt � Connecticut Essex: Nancy N. Richardson; New Haven: Matilde Paolini

McAfee; Rockville: Barbara A. Orlowski; Simsbury: Jeffrey A. Adams; Westport: William A. Jeffery, Mary B. Moses �

Florida Cape Coral: Jeffrey R. Hicks; Deland: Sue Angevine Guess; Hollywood: Barry Belenke; Indialantic: Luis Angel

Magana; Miami Beach: Pinchas Schechter; Naples: Michael J. McCormick; Palm Harbor: Timothy D. Schuler;

Plantation: Garrett Walker; Satellite Beach: Consuelo Schnaderbeck; West Palm Beach: Mathew Mooney � Hawaii •

Maui, Makawao: Alison Hutchison-Fahland; • Oahu, Honolulu: Brenda K. Reichel; Mililani: Abe L. Wilson � Illinois

Bloomington: Anne Blumer; Geneseo: Mark O. Arnold; Geneva: Lori M. Mesa; Normal: William A. Lyddon; Springfield:

Pat Pikesh; Troy: Bruce Upperman � Indiana Brownsburg: Charles Robert Scanlan; Indianapolis: Mark Ferreira, Mary

Wright; Jeffersonville: Seth Christian Simpson � Iowa Fairfield: Richard C. Kurka �Kansas Topeka: John C. Goldsmith �

Maine Portland: Arthur E. Spellissy, Jr. � Maryland Davidsonville: Helen Serras-Herman; Temple Hills: William R.

Mann �Massachusetts Attleboro: Jeff Wong; Braintree: Alan R. Howarth; Brookline: Martin Haske; Lynnfield: John A.

Caruso; Milton: Beth I. Fleitman; Norwood: Edward F. McDonough �Michigan Bay City: Bradley J. Payne; Escanaba:

Laurence Bell � Missouri Jefferson City: Margaret A. Gilmore; Kansas City: Greg Nedblake, J. Michael Tracy; Perry:

Bruce L. Elmer �Montana Great Falls: Amy Wolfe-VanCleave �Nevada Las Vegas: Kelley E. Thomas; Reno: Terence E.

Terras � New Jersey Phillipsburg: Robin A. Stokes; Roselle Park: Jo Anne M. Whitteaker � New Mexico Albuquerque:

Susan M. G. Wilson �New York Brooklyn: Istvan Rudas; Gansevoort: Clifford H. Stevens; New York: Lisa H. Thierman,

Brandon Tso; Rye: Gregory J. Cunningham � North Carolina Advance: Blair Tredwell;

Challenge Winners GEMS & GEMOLOGY Fall 1997 231

continued on following page

Manteo: Eileen Alexanian; Shelby: Sue Whitaker; Tryon: Matthew Randolph; Wilmington: Ben H. Smith, Jr. � Ohio

Akron: Lynn L. Myers; Dayton: Stephanie M. Weber, Michael Williams; Toledo: Mary C. Jensen, Jack Schatzley �

Oregon Beaverton: Robert H. Burns; Milwaukie: Cinda V. di Raimondo; Newport: Richard Petrovic � Pennsylvania

Jeffersonville: Aurora A. Stuski-Ryley; Womelsdorf: Lori Perchansky; Yardley: Jon Barry Dinola, Peter R. Stadelmeier �

South Carolina Sumter: James S. Markides � Tennessee Coalmont: Clayton Lee Shirlen � Texas Arlington: A. Thomas

Light; Austin: Corey Shaughnessy; Dallas: Ray Zajicek; El Paso: Jim Ferguson; Flower Mound: Elizabeth M. Roach;

Lamesa: Fontaine Cope; Richardson: Becky Templin Shelton �Virginia Blacksbury: Scott Steward; Fredericksburg: Ted

Kowalski; Hampton: Edward A. Goodman; Harrisonburg: Daniel M. Scanlan; Roanoke: Stephen O. Phillips; Springfield:

Robert G. Davis �Washington Bellingham: D. Hurlbert; Lakebay: Karen Geiger �Wisconsin Beaver Dam: Thomas G.

Wendt; East Troy: William Bailey; Mequon: Kathleen J. Molter; Racine: Kathi Vallner-Villarreal � AFRICA Zimbabwe

Harare: Lesley Faye Marsh �AUSTRALIA • New South Wales Concord West: Steve John Puz; Miranda: C. V. Russell •

Queensland Logan: Ken Hunter • Western Australia Coogee: Helen Judith Haddy � BELGIUM Brussels: Jo Delcoigne,

Léon Rubin; Diegem: Guy Lalous; Hemiksem: Daniel De Maeght; Ruiselede: Lucette Nols � BRAZIL Rio de Janeiro:

Luiz Angelo; São Paulo: Alejandro B. Ferreyra, Maria Amelia Franco �CANADA • Alberta Calgary: Diane Koke, Janusz

J. Meier • British Columbia Aldergrove: Ron Plessis; Chemainus:

Mark S. Curtis; Delta: Barbara Muir; Vancouver: Michael J. P.

Cavanagh; Victoria: Fred G. Billcock, Anthony De Goutiére •

Ontario Bobcaygeon: David R. Lindsay; Cookstown: Michael

Hyszka; Etobicoke: David N. Carabott; St. Catharines: Alice J. Christianson • Québec Laval: Monique Savard; Montréal:

Christiane Beauregard, Lina Massé, Yves Morrier � CHINA Fuzhou: Tang Deping � ENGLAND Kent: Linda Anne

Bateley; London: P. D. Chib � FINLAND Kajaani: Petri Tuovinen � FRANCE Herserange: Frederic Pracucci; Paris:

Marie-France Chateau, Stephen Perera; St. Ismier: Laurent Sikirdji � GERMANY Idar-Oberstein: Josef Bogacz, Erik

Vadaszi �HONG KONG Kowloon: Edward Johnson �GREECE Athens: Anna-Maria Falaga; Thessaloniki: Ioannis Xylas

� INDIA Bangalore: S. R. S. Murthy � INDONESIA Jakarta: Warli Latumena, Sarono; Jember: Adi Wijaya � ITALY

Ferrara: Sonia Franzolin; Florence: Rita Parekh, Paolo Penco; Genova: Mafalda Pasqui; Lucca: Roberto Filippi; Modena:

Apicella Tania; Pordenone: Barbara Pistuddi; Porto Azzurro: Diego Giuseppe Trainini; Roma: Andrea Damiani; Sanremo:

Enrico Cannoletta; Valenza: Rossella Conti � NETHERLANDS Amsterdam: P. Horninge; Rotterdam: E. Van Velzen;

Voorburg: Wilma Van der Giessen � NEW ZEALAND Lower Hutt: Dennis Blacklaws � PHILLIPINES Mandaluyong:

Mark Alexander B. Velayo; Manila: Colleen C. Mangun � POLAND Lublin: Marek Prus � PORTUGAL Algarve:

Johanne C. Jack; Viseu: Rui P. Branco � SCOTLAND Edinburgh: James W. M. Heatlie, Margery E. Watson � SPAIN

Agramunt: Santiago Escolà Villalta; Madrid: María Isabel Cereijo Hierro, José Antonio Gutiérrez Martínez, Ricardo

García de Vinuesa; P. P. Farnals: Monika Bergel-Becker; Puerto de Soller-Balears: J. Maurici Revilla Bonnin; Villauiciosa

Asturias: Claudio Argüelles Moris � SRI LANKA Kandy: Senarath B. Basnayake � SWEDEN Järfallä: Thomas Larsson;

Västerhaninge: Peter Markstedt � SWITZERLAND Rodersdorf: Heinz Kniess; Wolfertswil: Hedley Prynn; Zollikon:

Adrian Meister; Zürich: Eva Mettler � TAIWAN Taichung: Ho-Chi Yang � THAILAND Bangkok: Patricia Goossens �

TURKEY Istanbul: Mehmet Celal Yahyabeyoglu �VIRGIN ISLANDS St. Thomas: Nancy L. Fernandes

Challenge Answers (see pages 71–72 of the Spring 1997 issue for the questions): (1) a, (2) c, (3) c, (4) b, (5) b, (6) b, (7) c, (8) a, (9) b, (10) d, (11) d, (12) c, (13) b, (14)b, (15) a, (16) d, (17) c, (18) d, (19) c, (20) a, (21) b, (22) c, (23) a, (24) c, (25) d

232 Challenge Winners GEMS & GEMOLOGY Fall 1997

Book Reviews GEMS & GEMOLOGY Fall 1997 233

BIRTHDAY BOOK OF DIAMONDSPhotos by Harold and Erica Van Pelt,112 pp., illus., publ. by Harold &Erica Van Pelt, Los Angeles, CA,1997. US$19.95.*

The crisp, perfect color photographsof Harold and Erica Van Pelt, no lessthan 54 in number and all full page,depict an astonishing variety of loosediamonds as well as jewels andwatches set with diamonds and othergems. Some of the diamonds are veryfamous indeed, such as the Hope, theEugénie Blue, the Dresden Green, andthe Star of the East. Other spectacularpieces include the Cullinan blue-and-white diamond necklace, the dia-mond-studded Order of the GoldenFleece, and the Marie Antoinette ear-rings. It is amazing that these ear-rings, set with large and fine dia-monds, descended unharmed fromthe French queen’s jewel case aftershe met her death at the guillotine in1793 (they are now in the collectionof the Smithsonian Institution).

Although diamonds and dia-mond-set jewels dominate the illus-trations, a number of other gems areincluded to illustrate the gemstonefor each month. A nice touch is alsoprovided by heading each monthwith a crystal drawing of one of theforms diamond crystals take innature.

As mentioned in the foreword byRichard T. Liddicoat, Chairman of theBoard at GIA, the Birthday Book ofDiamonds indicates the dates foreach month, but not the specific daysof the week. Consequently, it is “per-fect for noting birthdays, anniver-saries, and other special occasionsthat repeat from year to year.”Although the pages assigned to thispurpose are the functional aspect ofthe book, the illustrations and theircaptions (the latter by Alice Keller)provide valuable facts about thesenotable jewels and gems, many ofwhich appear in print here for the firsttime.

The Birthday Book is stronglyand handsomely bound. With a little

care in making the entries neatly, itwill become a lasting treasury of refer-ences and reminiscences.

JOHN SINKANKASPeri-Lithon Books

San Diego, California

RUBY & SAPPHIREBy Richard W. Hughes, 511 pp.,illus., publ. by RWH Publishing,Boulder, CO, 1997. US$98.00.*

This is Mr. Hughes’s second book onruby and sapphire. The first, titledCorundum, was published by Butter-worth-Heinemann in 1990. Ruby andSapphire is self-published by theauthor, which allows him full free-dom in his writing. Tasteful advertise-ments, placed at the end of somechapters, helped fund the publishingcosts. I did not find the ads disruptiveand, in fact, believe they will be use-ful to some readers.

On opening the book, I found abookmark on which Mr. Hugheswarns the reader about his parochialwriting style. Mr. Hughes makes noapologies for this style; rather, hestates his desire to create a complete,factual book in which he interjects hisopinions and observations for enter-taining and stimulating reading. I per-sonally enjoyed this style very much.While all of the factual informationwas at my fingertips, Mr. Hughes’spersonal interjections kept the read-ing lively. I think anyone involved inthe gem trade will find these insightsenlightening, if sometimes irreverent.

The first chapter presents thehistory of rubies and sapphires.Chapters 2 through 9 cover chemistryand crystallography, properties, color,

spectra and luminescence, inclusions,treatments, synthetics, assembledstones, and fashioning. In chapter 2,the text and diagrams that describethe morphology of ruby and sapphirefrom various sources provide informa-tion that is often difficult for thegemologist to find. The followingchapters (3–6) supply the kinds ofdetails that many gemologists willfind useful for identifying corundumand its treatments. Inclusions areexplained in general in chapter 5,whereas inclusions specific to treat-ments, synthetics, and particularsources are found in those respectivechapters (6, 7, and 12). I would haveenjoyed more of the excellent pho-tomicrographs of inclusions catego-rized by source.

Chapter 10 discusses the gradingand valuation of ruby and sapphire, aswell as the world market. AlthoughMr. Hughes does not give a pricingbreakdown that uses a colored stonegrading system, he does explain thefundamentals of modern coloredstone grading, along with how tojudge the quality of these gem vari-eties of corundum. Quality rankingby source is also discussed. A sectionof this chapter is devoted to famousrubies and sapphires, with a summaryof these magnificent stones listedtogether with auction prices whereapplicable.

Chapter 11 covers the geology ofruby and sapphire, and chapter 12 isentirely devoted to world sources.With in-depth information on the loca-tion and history of individual sources,as well as the characteristics of thecorundum produced in each, chapter12 is also a remarkable reference.

The book is lavishly illustratedwith color photos, and each chapterends with a detailed bibliography forfurther research. This book is themost in-depth publication on rubyand sapphire I have seen. I highly rec-

__________________________________*This book is available for purchase throughthe GIA Bookstore, 5345 Armada Drive,Carlsbad, CA 92008. Telephone: (800)421-7250, ext. 4200; outside the U.S. (760)603-4200. Fax: (760) 603-4266.

SUSAN B. JOHNSON ANDJANA E. MIYAHIRA, EDITORS

B O O KB O O K

ReviewsReviews

ommend it to anyone involved in thebuying, selling, grading, identifying,or appraising of rubies and sapphires.

ANDREW LUCASGemological Institute of America

Carlsbad, California

THE NATIONAL GEM COLLECTIONBy Jeffrey Post, 144 pp., illus., publ.by Harry N. Abrams, Inc., New York,1997. US$39.95.*

The heralded redesign of the famousgem and mineral hall of the NationalMuseum of Natural History at theSmithsonian Institution (Washington,DC) has been completed. To celebratethe recent opening of the new hall, Dr.Jeffrey Post, Curator of Gems and Min-erals, describes the fabulous collec-tion––arguably the finest on public dis-play in the world—in this new book.

Dr. Post has provided an attrac-tively presented and entertainingshort course on the world of gems, inconjunction with brief histories ofsome of the major items in the collec-tion. Although the Hope diamond isthe best-known piece, the collectionhas many other unique stones andfamous jewelry pieces that are beauti-fully portrayed in the book. Amongthe remarkable gems are the RosserReeves star ruby; the Hooker,

Mackay, and Chalk emeralds; and theLogan sapphire. Historically signifi-cant jewelry––such as the Napoleondiamond necklace, the MarieAntoinette earrings, and the SpanishInquisition necklace––add breadth tothe appeal. In the nearly four decadessince Mr. Winston donated the Hopediamond to the Smithsonian Institu-tion, many important gifts havehelped build the collection.

Excellent photography by ChipClark and quality color printing addthe necessary element to make JeffreyPost’s effort a superb coffee table bookon gems.

RICHARD T. LIDDICOATGemological Institute of America


CLASSICAL LOOP-IN-LOOPCHAINS AND THEIRDERIVATIVESBy Jean Reist Stark and JosephineReist Smith, 190 pp., illus., publ. byChapman and Hall, New York, 1997.US$44.95.*

Loop-in-loop chains are some of theoldest forms of jewelry known. Infact, they were the principal types ofchains fabricated from the Bronze Agethrough the Middle Ages. This bookis a comprehensive instruction manu-al for making these elegant chains,

written for the person with experi-ence working at the jeweler’s bench.

Developed over many years ofpractical instruction, the text isextraordinarily detailed, well orga-nized, and clear. It is the authors’ stat-ed intention that the book be suffi-cient by itself as a guide to makingthese chains, and in this they havesucceeded. The hands-on portion ofthe book contains 34 projects. Itbegins with simpler chains (such assingle loop-in-loop chains) and thenmoves, chapter by chapter, throughchains of increasing complexity. Itends with multiple woven loop-in-loop chains. Each chapter is accompa-nied by high-quality drawings andphotos of the finished chains, as wellas by numerous excellent workingdrawings. There are also sectionsdealing with history, hand tools andequipment, metals, clasps and termi-nations, and much more. In addition,the book includes an appendix, a glos-sary, and an index.

Although there are other bookson chain-making––and numerousbooks on jewelry-making that containinformation on fabricating chains––Classical Loop-in-Loop Chains is thedefinitive work on the subject.

STEPHEN B. WORKMANGemological Institute of America


234 Book Reviews GEMS & GEMOLOGY Fall 1997

C A L L F O R P O S T E R SThe Gemological Institute of America will host the International Gemological Symposium in SanDiego, California on June 21–24, 1999. More than 2000 people are expected to attend this pivotalevent. The symposium program––with the theme “Meeting the Millennium”––will feature technicalsessions and panel discussions on a variety of topics of vital interest to all members of the gem andjewelry industry. In addition, there will be an open Poster Session featuring original presentations onsuch topics as new gem materials, synthetic gem materials, treatments, gem identification and grad-ing, instrumentation and techniques, gem localities, gem exploration, jewelry manufacturing, andjewelry design.

Contributions are being solicited for this Poster Session. To be considered for this importantevent (space is limited), please submit a preliminary abstract (no more than 250 words) to one of thePoster Session organizers by October 1, 1998. For further information on the Poster Session, contactDr. James Shigley at 760-603-4019 (Fax: 760-603-4021, or e-mail: [email protected]) or Ms. DonaDirlam at 760-603-4154 (Fax: 760-603-4256, or e-mail: [email protected]). For information on theInternational Gemological Symposium, contact Carol Moffatt at 760-603-4406 ([email protected]) orwrite to any of these individuals at GIA, 5345 Armada Drive, Carlsbad, CA 92008.

DIAMONDSArgyle Diamonds’ pink diamond tenders (1985–1996).

Australian Gemmologist, Vol. 19, No. 10, 1997, pp.415–418.

Pink diamonds, and other fancy colors such as purplishred, have been marketed by Argyle Diamonds in a tenderprocess since 1985. This article describes these stonesyear-by-year, through 1996, noting individual sizes,important stones, and other interesting facts. During thisperiod, nearly 650 faceted stones were offered, weighingalmost 550 carats.

The annual offerings have ranged from 33 to 88stones. In general, the stone sizes (carat weight) haveincreased over the years. The first stones over 3 ct wereoffered in 1989; only eight stones over 3 ct (the largest at3.66 ct) have ever been tendered.

Traditionally, the collection is viewed in Geneva, butin recent years Tokyo and Hong Kong have become estab-lished venues. The 1996 tender of 47 stones was pur-chased by 13 individual dealers and jewelers for a per-carat price of over $100,000. However, on at least threeprevious occasions (the last in 1993), the entire collectionwas sold to a single bidder, Geneva-based RobertMouawad. AAL

Ashton finds diamonds in Alberta kimberlites. DiamondIntelligence Briefs, June 6, 1997, p. 1523.

Ashton Mining of Canada Inc. has found diamonds inkimberlites in yet another Canadian province, Alberta.

Kimberlite pipe K12 (also reported as K14), in the BuffaloHills, has yielded 11 diamonds 0.5 mm or larger in maxi-mum dimension, as well as 139 smaller “micro-dia-monds”; two other pipes, K2 and K4, have returned “verylow” diamond counts. Exploration continues. MLJ

Automation v cheap labour. J. Lawrence, Diamond Inter-national, No. 44, November-December 1996, pp.87–88.

About 50% of India’s diamond-cutting workforce are con-tract workers who are paid $1 for each small diamondthat they polish. This price is the “base cost” for theIndian cutting center and has been constant for the pasttwo decades, primarily because of the gradual decline inthe rupee’s value relative to the dollar (about 12 to thedollar in 1976 and about 33 to the dollar when the article

Charles E. Ashbaugh IIIWoodland Hills, California

Anne M. BlumerBloomington, Illinios

Jo Ellen ColeGIA, Carlsbad

Maha DeMaggioGIA Gem Trade Lab, Carlsbad

Emmanuel FritschUniversity of Nantes, France

Michael GrayMissoula, Montana

Patricia A. S. GrayMissoula, Montana

Professor R. A. HowieRoyal HollowayUniversity of LondonUnited Kingdom

Mary L. JohnsonGIA Gem Trade Lab, Carlsbad

A. A. LevinsonUniversity of CalgaryCalgary, Alberta, Canada

Loretta B. LoebVisalia, California

Elise B. MisiorowskiLos Angeles, California

Jana E. MiyahiraGIA, Carlsbad

Himiko NakaPacific Palisades, California

Gary A. RoskinLos Angeles, California

James E. ShigleyGIA, Carlsbad

Carol M. StocktonAlexandria, Virginia

Rolf TatjeDuisburg UniversityDuisburg, Germany

This section is designed to provide as complete a record as prac-tical of the recent literature on gems and gemology. Articles areselected for abstracting solely at the discretion of the section edi-tor and his reviewers, and space limitations may require that weinclude only those articles that we feel will be of greatest interest to our readership.

Inquiries for reprints of articles abstracted must be addressed tothe author or publisher of the original material.

The reviewer of each article is identified by his or her initials at theend of each abstract. Guest reviewers are identified by their fullnames. Opinions expressed in an abstract belong to the abstrac-ter and in no way reflect the position of Gems & Gemology or GIA.© 1997 Gemological Institute of America

C.W. FRYER, EDITOR

REVIEW BOARD

Gemological Abstracts GEMS & GEMOLOGY Fall 1997 235

was written). In the highly competitive world of diamondfashioning, the article asks: “Will a centre with a stable$1 base charge prove unassailable by new technology?” Inother words, can new technology negate the competitiveadvantage enjoyed by India because of its low wages, andenable diamond-cutting centers to be established else-where?

When considering automation, additional manufac-turing costs must be factored in, such as for capital equip-ment, amortization, and the salaries of people needed tooperate small, semi-skilled production lines. To balancethese extra costs, one needs added value in terms of bet-ter yields and polished make. Such advances requireequipment that is technically feasible but not presentlyavailable, equipment that probably would be expensive todevelop.

After presenting several scenarios, the authors con-clude that automation would only boost yield by one ortwo percent over that achievable by India’s good artisanworkforce. AAL

Dia Met developments. Mining Journal, London, March7, 1997, p. 184.

Between 1988 and 1995, small-scale operations recoveredmore than 250,000 carats of diamonds from a 3225hectare area at Salvación in Venezuela. Dia Met Minerals(of Canada) signed a letter of intent with Cooperativa LaSalvación (CS), a Venezuelan corporation, to exploit thisdeposit. Diamonds recovered to date have come fromalluvial workings, but these are believed to have originat-ed from kimberlite dikes on the property. Dia Met plansto explore the kimberlites while CS continues to minethe alluvial sources. MLJ

Diadem touts California’s first diamond pipe since 1849.The Diamond Registry, Vol. 29, No. 6, 1997, p. 5.

Macro- and micro-diamonds of unspecified sizes havebeen found in 11 holes drilled in the 1120 acre LeeksSprings, California, property held by Diadem ResourcesInc. The source rock is apparently olivine lamproite brec-cia and agglomerate. MLJ

Diamond—Birthstone for April. H. Bracewell, AustralianGold, Gem & Treasure, Vol. 12, No. 4, April 1997,p. 24.

Hylda Bracewell’s series on birthstones concentrates ontheir lore rather than their gemological properties. Welearn that diamond is a “help to lunaticks,” that a personwearing a red diamond will achieve heroic deeds, but ablack diamond will attract misery. Diamonds with blackspots, “bubbles,” or “lines” in them are eunuchs andshould be avoided. Much of this lore is very old, and all ofthe above apparently came from Camillus Loenardus’sSpeculum Lapidum Venice (1502). Another tidbit fromthat work: “the Diamond is a most precious Stone of thecolour of polished iron. . . .” Some modern lore is includ-ed, such as what personality type should wear which dia-

mond shapes. According to Ms. Bracewell, round bril-liants should be worn by the home loving, pears by extro-verts, marquise cuts by the emotional, ovals by the enter-prising, emerald cuts by the logical, and heart shapes bythe romantic. MLJ

Diamonds from the deep. M. D. Riley, American JewelryManufacturer, Vol. 41, No. 11, November 1996, pp.18, 20, 22, 24.

Diamond exploration has surged worldwide, especiallysince 1990. This article reviews marine exploration, par-ticularly marine diamonds off the shores of Namibia.

At the time this article was written, about 35% ofNamibian diamond production, almost all of which isgem quality, was being recovered from the sea by DeBeers Marine, operating as a contractor to Namdeb (anequal partnership between De Beers and the Namibiangovernment). Several other companies also recover dia-monds from the sea, but these are smaller operations andcloser to the shore. Onshore reserves in Namibia are near-ing depletion, but increased marine production is offset-ting that decrease. Similar marine deposits, although notyet as significant as those off Namibia, are successfullybeing mined off South Africa’s west coast.

With Namibian offshore deposits as a model for theorigin and occurrence of marine diamonds, explorationhas been under way in recent years offshore ofKalimantan (Borneo), Western Australia (Bonaparte Gulfand Cambridge Gulf), Canada (Coronation Gulf in theBeaufort Sea), northern Russia, and Sierra Leone. To date,none of these exploration projects has found an econom-ically viable diamond deposit. AAL

India’s eximports: Gems & jewellery exports recorddowntrend in 1996–97. Diamond World, Vol. 24,No. 3, 1997, pp. 41–42.

Diamonds constitute more than 90% of India’s gem andjewelry exports. For many years, cut and polished dia-monds led exports, both in value and carat weight.However, the financial year 1996–1997 (ending March 31,1997) was the first exception in the long trend of contin-ually increasing exports—diamonds led a decline.Polished diamond exports decreased by 9.16% on a valuebasis (to $4,235 million), and by 1.72% on a carat weightbasis (to 18.88 million carats), compared to 1995–1996.

Other export items also decreased in 1996–1997because of lackluster overseas markets. These includedcut and polished colored gemstones (-5.89%), pearls(-35.89%), costume jewelry (-91.96%), and syntheticgems (-96.06%).

One of the few bright spots in export figures was goldjewelry, which increased 25.24% to $713 million. AAL

Japan holds the key. M. Cockle, Diamond International,No. 45, January-February 1997, pp. 41–44.

Since 1980, the world’s retail diamond jewelry sales haveincreased more than 2.5 times (to about $52.7 billion in

236 Gemological Abstracts GEMS & GEMOLOGY Fall 1997

1995). The U.S. and Japan are the mainstays of the indus-try. Interestingly, De Beers has had to raise its estimate ofannual retail sales for the U.S. by about 6% for the pastdecade. It was discovered that jewelry containing inex-pensive (predominantly Indian-cut) diamonds had notbeen included in previous surveys.

The two major markets have not grown equallysince the late 1980s. Monetary factors, primarily, haveweakened the Japanese market. (The yen depreciated15% against the dollar in 1995 alone.) Because Japanaccounts for about one-third of the world’s retail sales,sluggish performance in that market affects the entireindustry. Conversely, any growth in the world market isdependent on Japan’s recovery.

This article reviews the history of De Beers’sinvolvement in the Japanese diamond jewelry marketfrom 1966 (when that retail market was $255 million) to1996 (when retail sales totaled about $17 billion—a 66-fold increase in 30 years). Marketing campaigns and theappreciation of the yen (until recent years) are the mainfactors behind the spectacular rise in popularity of dia-mond jewelry in Japan. In addition, recent growth of theJapanese market is attributed to: (1) an increase in the sizeof the average engagement-ring stone (from 20 points in1980 to 40 points in 1995); and (2) the development of the“single-woman market,” so that today over 60% of singleJapanese women own a diamond. This latter market hasbeen incredibly successful: Three out of 10 unmarriedJapanese women buy diamond jewelry once a year.

Clearly, such achievements cannot be repeated adinfinitum. In fact, there is evidence of a change in attitudeamong younger Japanese women toward jewelry con-sumption in general, and the diamond engagement ringin particular. Some young people do not want to spendthree months’ salary on a ring. Further, there is a trendtoward less expensive weddings, and Japanese touristsfind that they can buy costly items, such as jewelry, morecheaply overseas.

This paper contains the most up-to-date (through1995) statistical compilations for the retail diamond jew-elry business worldwide. These compilations includeworld retail diamond jewelry sales for 1980–1995, retailsales by country, average price paid for jewelry, and num-ber of pieces sold. AAL

Kensington considers Chinese diamonds. MiningMagazine, Vol. 176, No. 2, February 1997, p. 119.

Canadian diamond mining company KensingtonResources has optioned a 50% share in a diamond mine300 km south of Beijing. This [name unspecified] minecurrently produces 40,000–45,000 carats per year, with a“measured and indicated” resource believed to be 2 mil-lion carats (ore grade 1.17 ct/ton) to 300 m deep in themine; an additional one million carats may be found atgreater depth (to 600 m) at a lower ore grade (0.77 ct/ton).Kensington has also signed joint-venture agreements fortwo diamond prospects nearby. MLJ

Lac de Gras revision. Mining Journal, London, February14, 1997, p. 130.

BHP delivered the final feasibility study for the BHP/DiaMet Lac de Gras project in Canada’s NorthwestTerritories; the study covers 17 years’ worth of plannedoperations. The Panda, Misery, Koala, and Fox pipes willbe developed. One change from preliminary plans is thatthe Sable pipe, with a value of US$63/ton, will be devel-oped during the first 17 years, and the Leslie pipe, with avalue of $28/ton, will not. Capital cost estimates remainunchanged, at C$900 million to bring the mine into pro-duction and another C$300 million for expansion in the10th year of operations. Two additional pipes (plus Leslie)may prove economic, and another five await large-diam-eter drill core sampling for evaluation purposes. MLJ

Mwadui diamonds again flow from Williamson mine. In-Sight—The CSO Magazine, Winter 1996–97, p. 11.

The Mwadui mine (originally known as the Williamsonmine) is the world’s largest economic kimberlite pipe(146 hectares in surface area). [Abstracter’s note: From1958 until 1973, it produced over 500,000 carats annual-ly.] After Tanzania became independent, productiondeclined as the mine fell into disrepair. In 1994, produc-tion ceased entirely.

Recently, De Beers and Tanzania agreed to rejuvenatethe mine, with De Beers getting a 75% stake andTanzania, 25%. [Abstracter’s note: De Beers will invest$16 million to refurbish the mine with new infrastruc-ture and equipment. Limited mining resumed inOctober 1996 (16,000 carats were produced that year).]Open-pit mining of the kimberlite pipe will begin inabout two years, and is projected to last for another sevenyears. More than 3,000 tons daily (about 1 million tons ayear) will be processed. [Abstracter’s note: Annualcaratage would be about 200,000 if published reports arecorrect that the kimberlite yields 0.2 ct of diamonds permetric ton.] Clearly, the Mwadui mine is poised to regainmuch of its former glory.

De Beers has a license to explore a large area (23,000km2) around the Mwadui mine. Already, some 60 previ-ously unreported kimberlites have been discovered.

AAL

New company to develop diamond pipes . . . MiningJournal, London, July 25, 1997, p. 65.

Russian diamond producer Almazy Rossii Sakha (ARS orAlrosa) has formed a joint-stock mining company todevelop the Botuobinskaya and Nyurbinskaya kimberlitepipes in the Nakyn region of the Nyurba district in west-ern Sakha (Yakutia). The new company, Alrosa-Nyurba,has a charter capital of 20 billion rubles (about US$3.5million); development of the two pipes is expected to cost$350 million. The mine life for both pipes is estimated at25 years; they could produce about $900 million of dia-monds annually within five years. MLJ


New diamond targets in Mali. Mining Journal, London,May 16, 1997, p. 392.

Several kimberlite “targets,” some containing diamonds,have been discovered at the Kenieba project in Mali. Nineareas have chromites with chemical compositionsmatching those found in diamondiferous kimberlites; 23areas show “anomalous” indicator minerals, and somecontain macro diamonds. Thirty magnetic targets havebeen found. So far, at least two kimberlite pipes areknown, the 19 hectare Cirque pipe and the 4 hectareSekonomala pipe. A 70.62 ct diamond found in a “paleoplacer” in late April 1997 originally may have come fromthe Cirque pipe. MLJ

New facility helps Finnish diamond search. DiamondNews and SA Jeweller, February 1997, p. 13.

A pilot-scale plant that uses flotation technology for con-centrating small diamonds has been developed at theOutukumpu laboratory of VTT Chemical Technology–Mineral Processing in eastern Finland. Plant capacity isabout five tons of ore per hour.

Flotation depends on the separation between waterand hydrophobic materials (substances that repel water,either naturally or through chemical treatment). In thisconcentration technique, small hydrophobic particles—inthis case, diamonds—are gathered in the froth that formsas air bubbles rise through a tank containing water, care-fully ground ore, and (sometimes) additives. Althoughflotation is a common technique for concentrating certainmetal ores, its application to diamond recovery is unex-pected. MLJ

Redaurum finds diamonds in Colorado. DiamondIntelligence Briefs, July 30, 1997, p. 1546.

Redaurum has found two more large diamonds—28.2 ctand 16.3 ct—at the Kelsey Lake mine near Fort Collins,Colorado. The former is the sixth largest stone found todate in North America. A 28.3 ct stone found last yearsold for $90,000. MLJ

Russian diamond pipe tender. Mining Journal, London,May 30, 1997, pp. 425–426.

Russia is seeking foreign participation to develop theLomonosova kimberlite pipe, in the Zolotitsa kimberlitefield on the edge of the Arctic Circle in northern Russia,about 90 km north of Archangelsk. Lomonosova, a largepipe, has an estimated grade of 0.3 ct per ton; Russianpress reports indicate a potential output of 3–6 millioncarats (Mct) for 30–40 years, or possibly even more than atotal of 250 Mct. Their possible worth could be in excessof US$12 billion. The state enterprise Severalmaz holdsthe official license to develop the Lomonosova pipe.

Russia’s diamond production is now dominated byAlmazy Rossii Sakha (ARS), which operates large minesin Sakha (Yakutia). Annual production is currently run-ning about 13 Mct, with stones averaging US$90 per ct.Eighty-five percent of this production presently comes

from the Udachny open pit. Total ARS rough diamondproduction in 1996 was worth about US$1.35 billion.

MLJ

Scandinavian progress for Cambridge. Mining Journal,London, November 15, 1996, pp. 393–394.

Cambridge Resources has been looking for diamonds inthe Alno and Kalix permit regions in Sweden; the area isadjacent to the Baltic-Bothnian megashear, a large tec-tonic feature similar to that in Canada’s NorthwestTerritories. Sixty magnetic anomalies were found in the1996 prospecting season, and five “areas of particularinterest” have been defined—based on the presence ofheavy minerals (e.g., diamond- and kimberlite-indicatorminerals) and topographic depressions. MLJ

Statistical survey Belgian diamond sector 1996.Diamant, Vol. 38, No. 390, February-March 1997,pp. 6–8, 10, 26.

By importing over 210 million carats (Mct) of rough dur-ing 1996, Antwerp further solidified its reputation as theworld’s largest diamond supply center. Of these, 130 Mct(62%) were gem or near-gem and 80 Mct (38%) wereindustrial.

The import figures are truly staggering, consideringthat only about 113 Mct of rough diamonds were minedworldwide in 1996. Two factors explain the discrepancybetween the 210 and 113 Mct figures. First, there weresignificant “double counts” (e.g., sight goods were sentfrom the CSO in London for sorting in Antwerp; thesewere sent back to London, then returned later toAntwerp, and they were counted both times they enteredBelgium). Second, diamonds were sent to Antwerp fromthe Russian stockpile.

Diamonds entering Belgium from the CSO consti-tuted 33% of the rough, while those from the outsidemarket constituted 67%. In addition to Russia, outsidegoods primarily came from Zaire and other west and cen-tral African countries via Liberia and Congo-Brazzaville.Since mid-1996, diamonds from Australia’s Argyle minealso have been part of the outside market.

Most (79%, or 92.5 Mct after deduction of double-counts) of Belgian rough gems and near-gems exported in1996 went to India. This 12% increase over 1995 clearlypoints to more lower-quality goods entering the market.Little is cut locally in Belgium.

Antwerp’s imports of polished diamonds are alsoimpressive: They reached 6.8 Mct in 1996 (an 18% in-crease over 1995), with most coming from India (36%),Russia (10%), the U.S. (10%), and Israel (8%). The re-export of almost 7 Mct of polished goods primarily wentto the U.S. (24%), Hong Kong (14%), Israel (9%),Switzerland (8%), and Japan (7%). Of the major marketsfor polished goods exported from Belgium, Japan sufferedthe most severe year-to-year decline (-19% in U.S. dol-lars), which was attributed to continued recession there.

AAL


Supply set to be outstripped by demand. L. Rombouts,Diamond International, No. 45, January-February1997, pp. 57–59, 61, 62, 64.

Diamond mining, with few exceptions, is very profitable.As a result, particularly since 1990, it has attracted manymining companies that previously showed little interest.Recent statistics and economic studies show a growingdemand for diamonds and a shortfall in better-qualitygoods in a few years—because of the gradual depletion ofreserves at existing mines.

This article reviews worldwide diamond mining andexploration for 1996. New mine supply is estimated at111 million carats [Mct], valued at $6.44 billion. Russianstockpile contributions to world supplies in 1996 were12.3 Mct, so world rough supplies for the year totaled123.3 Mct (a decrease of about 6.7 Mct over 1995).

Mine production in each major source country isreviewed. Key facts include:• By the year 2000, Botswana’s Orapa mine will be the

world’s second largest (after Argyle), with production of12–13 Mct annually.

• Eighty-five percent of Russian production comes fromthe Udachny mine, which is reaching its limit of prof-itability as an open-pit mine.

• Production declined markedly in 1996 at the (nowunderground) Finsch mine in South Africa.

• Production levels remain at about 5.5 Mct at the Mbuji-Mayi mine in Zaire, despite generally chaotic condi-tions in that country.

• Offshore production in Namibia (about 550,000 carats)now accounts for about 45% of that country’s total.

Worldwide, diamond exploration activity costs about$350 million per year. Much of it is centered in Canada,Russia (especially in the Arkhangelsk region), Finland,Australia, Angola, Botswana, Zimbabwe, and other partsof Africa (e.g., Sierra Leone, Tanzania). AAL

“A testing time” in diamond jewelry production. MazalU’Bracha, Vol. 13, No. 83, October 1996, pp. 73, 76,77.

The largest diamond-jewelry exporters are Italy, HongKong (including mainland China), Thailand, and India,according to Harry Garnett, director of the De Beers CSOMarketing Liaison Department. However, this has notalways been the case. Labor costs, particularly those asso-ciated with setting loose stones in jewelry, are a majorfactor in jewelry manufacturing, so the relative impor-tance of each center has changed over time.

Hong Kong was the main supplier of low-cost jewel-ry to the U.S. in the 1970s. By the early 1980s, this rolehad shifted to Thailand. More recently, India became themain supplier. That country has an almost inexhaustiblesupply of cheap labor (average labor costs per worker areabout $80–$180 per month).

The 1990s have been a “testing time” for the dia-mond jewelry industry, with lower margins, more con-tract workers, tighter inventory control, and a move

toward last-minute buying of polished goods. This com-petitive market has also resulted in the use of lighter goldsettings, lower-quality diamonds, and more coloredstones to meet lower price points.

The U.S. jewelry manufacturing industry hasbecome more specialized and service oriented; theJapanese have become more insistent on higher designand quality standards; and the Italian industry hasretained its position as the world’s leading manufacturingcountry in terms of size, creativity, and flexibility. Thearticle’s prognosis is that changes will continue in thediamond jewelry manufacturing industry, with contin-ued pressure on profitability and efficiency. This will leadto a more compact pipeline, in which both polished-goodsand jewelry wholesalers will play less important roles.

AAL

GEM LOCALITIESReturning the deposed queen conch to royal status in

Florida. National Geographic, Vol. 192, No. 1, July1997, p. 145.

Queen conchs are desirable for their shells and meat, aswell as for conch pearls. Florida populations had fallendrastically, by 1985, when the state banned conch har-vesting. A hatchery began producing conch larvae onLong Key in 1990, and 5,000 young conch have beenreleased offshore. They are monitored by means of ametal detector, which finds the aluminum tags attachedto their shells; nearly 20% of the young have survivedsince their release at an unspecified date MLJ

Emerald mineralization and metasomatism of amphibo-lite, Khaltaro granitic pegmatite–hydrothermalvein system, Haramosh Mountains, NorthernPakistan. B. M. Laurs, J. H. Dilles, and L. W. Snee,Canadian Mineralogist, Vol. 34, Part 6, 1996, pp.1253–1286.

Emerald formation usually involves two different sourcesof materials: (1) ultramafic (dark) rocks that containchromium, and (2) highly evolved rocks that containberyllium. Most emerald deposits are associated withgranitic pegmatites and hydrothermal quartz veinsintruded into Cr-bearing schist; emeralds are usuallyfound in the schist adjacent to the pegmatites. TheKhaltaro deposit is geologically significant because emer-alds formed within the hydrothermal veins and peg-matites, so Cr (generally an immobile element) musthave migrated from the host rock during their formation.

The Khaltaro deposit is located at an elevation ofnearly 4200 m, in a remote area of northern Pakistan.Small-scale mining by the Gemstone Corporation ofPakistan in 1985–1990 produced about 600 carats of gem-quality emerald from two small prospects. However, fur-ther exploitation has been discouraged by the fact thatthe emeralds are pale and fractured, as well as by the inac-cessibility of the locale.

Emerald at Khaltaro occurs in two types of


hydrothermal veins—quartz and tourmaline-albite––and,less commonly, in the fine-grained outer zone of the peg-matites. Only rarely does it form in the altered amphibo-lite next to the veins. Emerald crystals up to 2.5 cm longwere reportedly found by local miners in the tourmaline-albite veins. Aquamarine and colorless beryl are alsofound locally.

The emerald crystals examined were unmodifiedhexagonal prisms terminated by pinacoids, abundantlycracked and included. A few samples showed potentialfor chatoyancy, because of hollow tubes parallel to the c-axis. Microprobe analyses (some provided in the article)give 0.20–1.27 wt.% Cr2O3 and 0.62–0.89 wt.% Fe2O3.Khaltaro emeralds can be differentiated from otherPakistan emeralds by their cathodoluminescence emis-sion spectra and chemistry (higher Si, Al, and Cs; andlower Mg, Na, Fe, and Sc).

Based on studies of trace elements and oxygen iso-topes in whole-rock samples and individual minerals, theauthors suggest that the Be from the pegmatite-hydrothermal vein system combined with Cr from thehost rock when fluorine-rich hydrothermal fluids alteredcalcic amphibole and zoisite in the amphibolite. Thereleased Cr migrated a short distance into the veins,where it combined with the crystallizing beryl to formemerald. MLJ

Characterization of alexandrite, emerald and phenakitefrom Franqueira (NW Spain). C. Marcos-Pascualand D. B. Moreiras, Journal of Gemmology, Vol. 25,No. 5, 1997, pp. 340–357.

Described are the occurrence and characteristics of theberyllium minerals emerald, alexandrite, and phenakitefrom Franqueira, Spain. The geologic conditions are sim-ilar in many ways to those of the well-known deposits ofTokovaja in the Ural Mountains of Russia; that is, Be andCr were brought together when pegmatites and associat-ed mobile elements (Be, B, P) were emplaced into Cr-bear-ing mafic rocks. The gem-bearing zone is only 3 m thickand crops out for 15 m, although it may extend furtherunderground. The Franqueira minerals are similar totheir counterparts from Tokovaja, as several comparativetables show. The authors present extensive chemical andphysical data on the three minerals, along with discus-sion of their abundant inclusions. If the photos are repre-sentative, then the quality of color or clarity in thesegems, unfortunately, is not high. CMS

Iberian emeralds. Mining Journal, London, December20/27, 1996, p. 496.

The Franqueira deposit, in the Helena permit area ofnorthwestern Spain, contains gem-quality emerald,alexandrite, and phenakite. It was found by geologistsfrom Oveido University. The exploration permit for the30 km2 Helena permit area has been awarded toCambridge Mineral Resources and will be managed local-ly by Stella Exploration Services. The largest emerald to

date, 30 ∞ 10 cm, reportedly had good transparency and adark green color. [Abstracter’s note: Although the articleclaims that this is the first documented deposit of gem-quality emerald and alexandrite in Western Europe,both Austria and Norway have produced gem-qualityemeralds.] MLJ

A new colour-change effect. A. Halvorsen and B. B.Jensen, Journal of Gemmology, Vol. 25, No. 5,1997, pp. 325–330.

Chromium-bearing tourmalines from the Usambararegion of Tanzania exhibit color variations from greenthrough orange to red when viewed in transmitted light,depending on the thickness of the sample. Examinationof pleochroism and spectral features indicates that theapparent color depends on the distance that light travelswithin a sample (path length). Faceted green gems displayflashes of red when internally reflected light travels farenough within a sample to achieve the required selectiveabsorption. The perceived difference in color was attrib-uted to the “psycho-physical effect,” as defined in a 1964paper by C. P. Poole on color change in Cr compounds.

The authors propose the term “Usambara effect” forthis color change that is due to path length. It would beinteresting to try the authors’ experiment (well illustrat-ed in figure 3) on other types of gems, such as some mala-ia garnets, that also exhibit red flashes. The likelihoodthat path length accounts for these flashes has long beenthe subject of speculation, but it has never before beenillustrated so conclusively.

Two issues of this journal later, in a letter to the edi-tor, Dr. Kurt Nassau pointed out that the effect is wellknown in the field of organic dyes, where it has beencalled “dichroism.” Thus, he argued, the new name pro-posed by Halvorsen and Jensen is misleading. Sincedichroism is used differently in gemology, however, Dr.Nassau proposed the terms “concentration dichroism”and “thickness dichroism” as appropriate. CMS

A note on a new occurrence of vanadian grossular garnetfrom Madagascar. A. Mercier, B. Moine, J.Delorme, and M. A. F. Rakotondrazafy, Journal ofGemmology, Vol. 25, No. 6, 1997, pp. 391–393.

A new deposit of green grossular was discovered in late1991 in the Gogogogo area of southwest Madagascar. Itspale yellowish green to dark green color is primarilycaused by vanadium, as is the case with tsavorite fromEast Africa. Overall chemical and physical properties (asrepresented by a single dark green specimen) resemblethose of similar material from East Africa and Pakistan,including an R.I. of 1.742 and S.G. of 3.62. CMS

Pailin’s path to prosperity. S. Kanwanich and R. Saen-grungruang, Bangkok Post, June 1, 1997, p. 1 (Pers-pective section).

Pailin, a Cambodian border town with Thailand, has longbeen known for its gemstones. Although this natural


resource has always been exploited, trade—once bannedby authorities—has recently received official recognitionagain. Most investors are Thai, and gem concessions area major income source in Pailin.

The gem trade largely supports the free health careand education systems. Gem shops abound, and gem con-cessions are sold at rapidly inflating prices. Miners useheavy equipment to sift for gem minerals, but this is usu-ally unproductive because most concessions were previ-ously mined. Although gems are still abundant, they areburied deeper, requiring more-sophisticated machineryand extraction techniques.

This well-written article also looks at the history ofthe area and provides a fascinating, first-hand glimpseinto the excitement and complex interaction of merchantand military in what is today a gem boomtown.

Alison Mader

Physicochemical structural characteristics of ambersfrom deposits in Poland. F. Czechowski, B. R. T.Simoneit, M. Sachanbinski, J. Chojcan, and S.Wolowiec, Applied Geochemistry, Vol. 11, No. 6,1996, pp. 811–834.

Amber has been mined from, or known to occur at, over600 localities in Poland. At present, the richest occur-rence is being mined near Gdansk (on the Baltic Coast),particularly between the mouths of the Stogi and Vistularivers. This paper compares physical and chemical prop-erties of ambers from the Gdansk region with those fromnewly discovered occurrences in central Poland nearBelchatów (associated with brown coal) and in southwestPoland near Jaroszów (associated with refractory clay).Very sophisticated analytical techniques (e.g., gas chro-matography, nuclear magnetic resonance, positron anni-hilation spectroscopy, and nitrogen sorption) were used tomake comparisons.

The results show that these ambers from distinctlydifferent geological environments are similar in somerespects, such as in their chemical character: They con-sist of mixtures of the same compounds, namely succi-nates and n-alkanes. However, relative proportions ofthese compounds differ in ambers from the three locali-ties and even within the amber samples themselves (exte-rior versus interior zones). Similar variations are reportedfor other properties, such as porosity, reflectance, and sol-vent-extraction yields.

On the basis of the chemical similarities (particular-ly the succinates), the authors suggest that amber from allthree areas is of similar origin and can be called “Balticamber.” The authors assume that all three ambers camefrom the same hypothetical conifer of the Pinus suc-cinifera family. AAL

A study of New Zealand Kauri copal. S. J. A. Currie,Journal of Gemmology, Vol. 25, No. 6, 1997, pp.408–416.

Kauri copal, a resin produced by the long-lived Kauri pine,

has been “mined” for more than 150 years from NewZealand coal beds. Its interest to gemologists lies largelyin its resemblance to amber, but this role pales by com-parison to the almost half million tons once exported foruse in varnishes and linoleum. Some fine varnishes stilluse it today. The Otamatea Kauri Museum in northernNew Zealand preserves the material’s history.

Twenty-five specimens of widely varying color (color-less to dark brown), diaphaneity (transparent to opaque),and age (“recent” to 40 million years) were studied by theauthor. The results are compiled in a comprehensive tablethat includes source, age, specific gravity (1.03–1.095),refractive index (1.540), solubility in ether and alcohol, flu-orescence, and smell. Tests and observations, other thanR.I. and S.G., showed a broad range of reactions, but fossilsamples of copal could be distinguished from recent spec-imens in that the former are insoluble in alcohol. As such,the fossil samples are indistinguishable from most amber.Moreover, because the alcohol test is destructive, it hasonly very limited application in gemology. This article isa must-have reference on a poorly documented (yet abun-dant) gem material. CMS

INSTRUMENTS AND TECHNIQUESA Raman microscope in the gemmological laboratory:

First experiences of application. H. A. Hänni, L.Kiefert, and J. P. Chalain, Journal of Gemmology,Vol. 25, No. 6, 1997, pp. 394–406.

The Swiss Gemmological Institute (SSEF) in Basel hasfound its recent acquisition of a Raman microscope wellworth the investment, to judge by this excellent accountof its initial applications at this laboratory. Much moreaccurate, easier to use, and considerably less expensivethan its predecessors, the Renishaw Raman microscopeat SSEF easily fits on a desktop. It has enabled SSEF staffmembers to identify CO2 inclusions in sapphire that arediagnostic of heat treatment, detect and identify fracturefillings in emerald, easily identify mounted gemstones,and characterize specimens with irregular surfaces. It canalso distinguish component minerals in many aggregates.

The method is nondestructive in all but a few well-documented conditions and works on all types of materi-als, except those having simple compositions and highsymmetry, such as metals and alloys. Raman spectroscopydoes require that the user be aware of its limitations andparameters; also, the lack of a large, published databasemeans that each laboratory must obtain many of its ownreference spectra from known samples. However, Ramanspectroscopy clearly has earned a place of respect amongthe few high-tech laboratory techniques that have provedindispensable to gemological testing today. CMS

JEWELRY RETAILINGDue to shortage, prices increase for Tahitian cultured

black pearls. National Jeweler, July 1, 1997, p. 26.

‘‘

‘


Because of a pearl shortage, prices of Tahitian culturedblack pearls jumped 7.9% during the first quarter of 1997over the same period the previous year. The amount ofpearls exported during that time decreased by 109 kg,according to customs figures provided by the Tahitiangovernment.

The biggest export market for loose Tahitian pearlsduring the first quarter was Japan, although the volume(440.7 kg worth $7.8 million) was down this year fromthe first three months of 1996 (547 kg worth $9.7 mil-lion). The U.S. moved from fourth to second place, im-porting a total of 61.7 kg of Tahitian pearls, comparedwith 39.8 kg imported during the same period last year.

MD

Top of the heap. Retail Jeweller & British Jeweller, July10, 1997, pp. 10–11.

Christie’s London jewelry sales in June 1997 brought in£7.03 million, a massive 61% increase over last year’sJune sales. The sale included the Hatot collection and animportant privately owned European collection. About30% of sales were to individuals from Asia and theMiddle East.

The highlight of the sale was a spectacular fancy-col-ored diamond pendant/brooch that brought the top priceof £441,500, almost double its pre-sale estimate. Thispendant was purchased by an American dealer. In theBijoux Français sale, a diamond pendant by Van Cleef &Arpels sold above estimate for £199,500 to a member ofthe British trade, while a private Hong Kong buyer paid£155,500 for a 1925 ruby-and-diamond necklace byMauboussin, originally estimated at £60,000–£80,000.

At Sotheby’s June 19 sale, diamonds took all the topprices. An arrow and quiver brooch—set with diamonds,emeralds, and rubies—sold for £25,300, much higher thana pre-sale estimate of £4,000–£6,000. A ruby-and-diamondbracelet, circa 1930, sold for £40,000; while a sapphire-and-diamond tiara went for £26,450, double its estimate.

At Bonham’s June 20 sale, an attractive diamond-setswan brooch by Van Cleef & Arpels sold for £11,000. ABurmese ruby-and-diamond cluster ring went for£22,000.

At Sotheby’s June 11 New York sale, private Ameri-can buyers purchased eight of the 10 top lots. A fine cul-tured pearl necklace sold just below estimate at £68,680;the same price paid for a necklace of natural black pearlswas substantially above estimate.

The article also reports on other auctions, includingfurther examples of pieces sold. MD

PRECIOUS METALSPlatinum demand rises in U.S., China. M. Kletter,

National Jeweler, June 16, 1997, p. 3.According to Johnson Matthey’s Platinum 1997, demand

for platinum in jewelry applications rose in both Chinaand the United States in 1996. Worldwide, total demandfor jewelry reached 1.85 million ounces, up from 1.81 mil-lion ounces in 1995. Demand for platinum jewelry inNorth America rose around 38%, to 90,000 ounces from65,000 ounces in 1995. Platinum jewelry in the U.S. is stilllargely concentrated in the bridal market. China became asignificant market for platinum jewelry in the last threeyears, according to the report, because of demands by anew middle class in that country. Demand was flat in1996 in Japan, the largest market for platinum jewelry.

Platinum supplies fell by 90,000 ounces to 4.9 mil-lion ounces in 1996, largely because of declining ship-ments from Russia and South Africa, the largest platinumsuppliers. MD

SYNTHETICS AND SIMULANTSThe identity of reddish-brown inclusions in a new type of

Russian hydrothermal synthetic emerald. K.Schmetzer and H. J. Bernhardt, Journal of Gem-mology, Vol. 25, No. 6, 1997, pp. 389–390.

Reddish brown transparent platelets in recent Russianhydrothermal synthetic emeralds are conclusively identi-fied as hematite. They differ from previously observedopaque hematite platelets only in that they are thinenough to achieve transparency.

CMS

TREATMENTSIdentification of B jade by diffuse reflectance infrared

Fourier transform (DRIFT) spectroscopy. P. L. Quekand T. L. Tan, Journal of Gemmology, Vol. 25, No.6, 1997, pp. 417–427.

The advantages of DRIFT spectroscopy to detect wax andpolymer impregnation of jadeite are spelled out in thisarticle. DRIFT can obtain results even from thick and/ormounted specimens; its sensitivity exceeds that of tradi-tional FTIR spectroscopy.

Three samples of “wax-buffed” (otherwise untreated)jadeite, three of bleached and wax-impregnated jadeite,and four of bleached and polymer-impregnated jadeite (allgreen) were examined. Ultraviolet (long- and short-wave)fluorescence could not distinguish between the first twosample types, and it only sometimes indicated the third.FTIR results proved similarly ambiguous. However, finepeaks in the 700–6000 cm-1 region of the DRIFT spectraenabled the authors to conclusively separate the samplesinto their three groups. Although 10 samples represent asmall population from which to draw sweeping conclu-sions, this method (and its usefulness in detecting certaintreatments) clearly deserves serious consideration andfurther study. CMS


fall 1997 gems & gemology - gia · louderback’s docu-mentation of the ditrigonal-dipyramidal...

Documents