Gold Cobs from the Florida shipwrecks of the 1715 Fleet & other New World wrecks. Spanish Colonial gold cobs from Lima, Mexico, Cuzco, Bogota, and Cartagena.






Los Mimbes (Tumbaga)



M4 sold


M36 sold


M56 sold


M101 sold


M152 sold




Cortes from a 16th c.  

engraving by Holl



   The Tarascan Capital













































































       Charles I


























































-,Treasure from Hernando Cortes'

   Conquest of Tarascan Mexico

"Golden silver" (dorada plata) ingots recovered from

the Los Mimbres shipwreck (circa 1527)

Cortes' conquest of Mexico was by no means a planned affair. It was not until Cortes had completed his conquest of the Aztecs in late 1519 that he even learned of another large and wealthy Mexican kingdom to the west, the Tarascan Empire. The capital of the Tarascan Empire was at Tzintzuntzan on the shores of Lake Patzucuaro in modern Michoacan.  The Tarascan capital was about 225 miles due west of Tenochititlan, the former Aztec capital where Cortes now was based. By 1522 Cortes was ready and eager to take on the Tarascans. In July he send the vanguard of his invasion force west under the command of a trusted lieutenant, Captain Cristobal de Olid. Olid did not disappoint Cortes.

The Aztec (gray) and Tarascan (green) Empires in Central Mexico circa 1500 AD


The Spanish conquest of the Tarascan Empire is documented by several contemporary sources including the marvelous La Relacion de Michoacan penned by an anonymous Conquistador.  According to La Relacion, and Cortes own letters agree, the conquest of the Tarascan Empire was a much less bloody affair than the previous conquest of the Aztecs.  Proceeding more by treaties and alliances than battles, Cortes Spaniards were effectively masters of the Tarascan kingdom by the end of 1523. In the process they had acquired a vast amount of treasure, mostly in the form of Tarascan jewelry and ceremonial artifacts.  Cortes sent several of his most trusted lieutenants to Michoacan to act as veedors or supervisors, overseeing the conversion of the Tarascan treasure into a form that could be appraised,  divided, and the very important royal quinto (20%) collected. Foremost among the veedors was one Bernardino Vazquez de Tapia, who much to Cortes' satisfaction had just overseen the conversion of Aztec treasures into bullion. Now Bernardino was tasked with repeating the process in Tarascan Michoacan.  Several primitive field foundries were set up at places where the treasure had been collected. These field operations were not capable of refining precious metals but merely melted it and cast it into crude bars or discs, while at the same time doing a rough assay.  Each veedor was required to mark the bars cast and assayed under his oversight. In effect he was acting with the responsibilities of a colonial assayer but under informal field conditions. Bernardino marked the bars of Tarascan treasure he oversaw in 1524 with the monogram BVo. 



By late 1524 caravans of Tarascan bars (like the one above) were streaming back to Mexico City, but trouble soon arose. A late 1525 inspection of the Tarascan field operations by royal officials found many people were evading the quinto in various ways, including undervaluing their bars with low assays.  Bernardino's bars proved to be accurately assayed but not so bars produced under other veedors. Cortes was criticized for permitting this, and the governing Council in Mexico City decreed on April 6, 1526 a general recall of all unrefined silver bars and discs circulating in Mexico for proper refining and assay. The recall was to begin immediately and be completed within 3 years, after which unrefined metal like the Tarascan bars would be illegal to hold or ship. Late in 1526 a royal inspector named Juan de Ponce de Leon arrived with new quinto seals, punches, and orders to expedite the recall The recall proceeded apace despite Ponce de Leon early demise. By 1529 the Tarascan bars had disappeared into the melting pots, replaced by refined ingots produced the official foundry in Mexico City. No Tarascan bars were thought to have survived the recall and the subsequent 300 years of remelting older ingots under Spanish Colonial rule.

In the summer of 1992 Marex corporation was treasuring hunting off the Little Bahamas bank in the vicinity of the Maravillas wreck (1656).  On the Los Mimbres shoals they stumbled upon what was obviously an early Colonial-era wreck. The wreck yielded about 160 crudely cast silver bars, some similar silver discs, and a dozen pieces of gold . Experts in Spanish Colonial bars could not identify or date the bars, and Marex, under investor pressure, proceeded to auction the still unidentified bars in April 1993. Later that year a private researcher (Armstrong) who had been working with Marex on bars published a monograph. The monograph correctly identified the bars as cast in the reign of the Spanish monarch Charles I (1516-56) and as composed of a metal alloy favored by several pre-Colombian Meso-American cultures.  Unfortunately that researcher settled on the name "tumbaga silver" for bars of this composition. "Tumbaga" is not a word the Tarascans or the Spanish ever used for this metal. The Spanish called this distinctive metal plata dorada (because of its high gold content) or plata baja (because it was not silver of high fineness) or sometimes just plata de Mechuacan.  Tarascans may have called it "chuperi". Tumbaga properly refers to a copper and gold alloy worked by the pre-Colombian people of South America. The Los Mimbres ingots were predominantly silver and not from South America.

In 2008 another private researcher (Garcia-Barnecke) made the connection to the Tarascan Empire by drawing on metallurgical analysis (Hosler) of the few surviving Tarascan artifacts. Gas chromographs done by Marex on the seven of the Los Mimbres bars showed precious metal ratios and trace metal content definitively matching pre-Colombian Tarascan metal. One bar (M66) even shows as inclusions un-melted Tarascan jewelry discs!  When this same researcher published his 2010 monograph on the Los Mimbres bars, he also did us the favor of including translations of all the key contemporary documents, so we can read in their own words, from their letters and diaries and official accounts, the story of the conquest of the Tarascan Empire.

Current research on the Tarascan bars is now focused on identifying the unfortunate Spanish galleon that came to grief on Los Mimbres. There is a good chance archival work in Spain will put a name to the Los Mimbres galleon. The window of time possible for the sinking is quite small. Several considerations argue strongly for the spring of 1526 as the earliest possible date the Los Mimbres treasure could have left Vera Cruz.  The spring of 1528 (or possibly 1529) is very likely the latest date a treasure of this composition would have left Mexico.  The most probable date that Los Mimbres galleon sailed was the spring of 1527.

The key considerations in dating the wreck are these. First of all, we know that almost all the Los Mimbres bars were cast in Michoacan in 1524. About 10% of the Los Mimbres bars show a second assay. Re-assaying the Michoacan bars commenced in earnest in late 1525/early 1526 after the inspection by royal officials cast doubt on the assays previously done in Michoacan.  10% of the Los Mimbres bars were re-assayed, but, significantly 90% are not, suggesting the full-scale re-evaluation of the Michoacan assays was in its early stages when the bars left Mexico. Nor do any of the bars show evidence of the new official quinto which arrived in late 1526 and supplanted the crude seals Cortes' ironmongers had fashioned. By late 1527, and certainly by 1528, one would expect to find in a group of 160 silver bars at least one stamped with the new quinto.  By that later date one would also expect a substantially higher percentage of bars to show either re-assays or actual conversion to the refine bars ordered by the Council's Resolution of April 6, 1526.

There is in fact only one refined silver bar (M97) in the Los Mimbres treasure, with silver content precisely at a Spanish standard of .936 fineness and with a copper content at 4% and gold under 1/2%. Unrefined Tarascan bars have a typical silver content in the 55-70% with a high copper content and up to 5% gold (dorada plata).  A ship sailing for Spain as late as 1528 or 1529 would likely have many more than one refined silver bar, as the Mexico City recall was rapidly converting all of the Tarascan bars into legally mandated refined ingots. The one refined Los Mimbres ingot (M97) is also clearly an early product of the Mexico City foundry under assayer MS (Martinez) because it too was subject to a re-assay in the wake of the April 1526 resolution. In sum, the Los Mimbres Treasure seems clearly to pre-date the full-scale recall of Tarascan bars which had begun in earnest by mid-1526.  The Los Mimbres galleon is most likely a Spanish ship that sailed from Vera Cruz in early 1527 or shortly thereafter.

One of the largest Los Mimbres bars (M101) for sale here.



Armstrong, Douglas Tumbaga Silver for Emperor Charles V (1993)

Garcia-Barnecke, Agustin The "Tumbaga" Saga (2010)

Hosler, Dorothy The Sounds and Colors of Power (1994)

Hubert, H.B. History of Mexico (1914)

Proctor, Jorge  personal communications

Warren, Benedict The Conquest of Michoacan (1985)


historical images courtesy of Wikipedia




Nuestra Senora de La d the Dominican Republic on July 23, 1641. The Almiranta of the 1641  carried an exceptionally large treasure cargo of approximately 100 s metals. More than 300 passengers and crew perished with the ship, and the Spanish never salvaged the treasure.  Boston sea captain William Phipps became a national hero in 1687 when he salvaged over 30 tons of silver bars and coins for King James II. The n's site was rediscovered 291 years after Phipps by Burt Webber of Seaquest International. Webber worked the site successful for several seasons after 1978. Webber found thousands of Mexican eight and four reales, some Potosi eight reales, and also a few rare early Colombian eight reales, including the present coin.

The Bogota eight reales that Webber found have been invaluable in establishing or confirming the early history of this mint which opened in 1627. Miguel Pinto Camargo was the first full-time assayer at Bogota, working from 1627 to 1632 and marking his coinage with the letter P, which appears on this coin to the left of shield and below the letters NR. NR which for stand "Nuevo Reino" or the New Kingdom. Colombia was called at the time "the New Kingdom of Granada." The bottom stem of the P is off the flan. To the right of the shield is a bold denomination in Roman numerals, VIII. On the reverse in the legend we can read REX 16... We know the missing digits are 28, because of several very distinctive features of the 1628 issue. In the angles of a very  heavy cross are lions and castles that are too small for an 8 reales. Punches for a 4 reales were used on this and another 1628 reverse die, apparently because the normal punches for the 8 reales were not available. Only 1628 8 reales shows this combination of heavy cross and small lions & castles. See the Lasser/Restropo specimen M44-6a. The total surviving population of 1628 Bogota 8 reales may be less than a half dozen coins.


The preservation of this Concepcion 8 reales is exceptional (weight: 26.07 gm). The vast majority of Concepcion coins show significant corrosion, to the point that the design on one side or other is obscured. The few early Bogotas recovered show especially heavy corrosion, with weights usually below 23 grams.  I have seen (and owned) only one other Concepcion Bogota in comparable condition, a coin that sold as lot 200 in Christensen May 1982 Sale. In sum, the present coin is a rare early Bogota 8 reales with a great shipwreck pedigree. With Webber/ Seaquest certificate.


Now available. $2150 or 480-595-1293



S2.  From the Capitana of the 1733 Fleet, EL RUBI,

this full-date 1731 F Mexico 8 reales.

With a very informative 2 page photo-certificate from Bill Wood.




Unfortunately, most of the silver cobs offered in the marketplace as 1733 Fleet are not! Because choice dated 1733 Fleet eight reales are both rare and expensive, sellers routinely offer Vliegenthart and Rooswijk coins of the same dates, with various confabulated certs, as 1733 Fleet. I strongly recommend that you do not trust the "might be" pedigree of any alleged 1733 Fleet silver cob unless it has indisputable documentation from the original salvors. Dealers certs are not enough in this case. When you wish to sell your "1733 Fleet" coin and no one will pay what an indisputable 1733 Fleet cob is worth, you will discover this.


Bill Wood was one of the last salvors to legitimately work the so-called "Coffins Patch" EL RUBI site in 1986 before the State of Florida turned the wrecksite into a State Park. This 1731 Mexican eight reales was one of the best coins he found. Bill has done a great job in preparing a two page photo-certificate for this coin, detailing both the coinage and the history of the 1733 Fleet. A few other dated 1733 Fleet reales have come into my hands from Bill and John McSherry over the last 20 years, and I can second the opinion that this is an exceptional piece. If you would like to see a scan of Bill's certificate, or if you'd like more information on the coinage of the 1733 Fleet, just contact me.



Available for $465 or 480-595-1293



S3. From the 1715 Fleet,

Reign of Charles II (1665-1700)

a very choice Lima 1699 R eight reales

from the Cabin Wreck, wrecksite of the 1715 Fleet Almiranta San Roman.

 Full weight at 26.7 gms, large 35 mm planchet




The Florida State Collection has individually catalogued 22,963 silver cobs from its vast collection. Exactly 257 of these, or just over 1%, are Lima cobs. By comparison, 21,170 coins, or 91.5%, are Mexican cobs. What this tells us is the Lima Fleet cobs are many times scarcer than  Mexican Fleet cobs, 82 times scarcer according to the Florida State sample. Statistics aside, advanced Fleet collectors have long recognized the rarity of choice Fleet Limas. If we are lucky, one or two Fleet Limas come into marketplace in a year.


It is curious that Cabin Wreck, the site where the Ubilla's Mexican Almiranta SAN ROMAN  wrecked, has been the source of almost all Lima silver cobs. (None of the Terra Firma ships have yielded any significant amount of Lima silver or gold.) The distinctive dark gray, slightly tarry deposits that we see on this 8 reales verify its Cabin Wreck provenience. I personally have seen Cabin Wreck Limas with 1692-1700 dates, and some earlier dates are known. The first coin found by Lou Ullian on Cabin Wreck in the summer of 1960 was a 1692 Lima.




Dated 1714-1715 Mexican eight reales now bring thousands of dollars, and there are hundreds if not thousands of them! I doubt if more than a few hundred Limas of this quality survive including all dates. 


Available  for $550 or 480-595-1293




S4. From the  galleon NUESTRA SENORA DE LAS MARAVILLAS (lost 1656)

an exceptional full date 1655 Mexican eight reales




Though quite a few salvors have briefly worked the site since 1972, there have been only been two authorized salvage efforts on the wreck of the Maravillas. Bob Marx discovered and worked the site with his Seafarers group in 1972-73, and then in 1989-1992 Humphreys and the Marex group conducted a very systematic salvage of the site. This exceptional well preserved (25.6 gms) Mexican eight comes from the Marx-era salvage. It sat proudly in a Florida collection for over 35 years, in the company of some equally remarkable 1715 Fleet silver, until recently re-appearing in the marketplace.


Not many well preserved and full date 1655 were recovered from the Maravillas. There is nothing of comparable quality in the coins Marx consigned to the December 1974 Schulman sale. Look at lots 210-247. A near twin to this 1655 does occur at lot 78 of the Christie's Marex sale of May 1992, but it is the only comparable coin in that sale. Notice that several coins in both sales show scrapes and scratches. Quite a few of the coins salvaged in 1970's in particular show scrapes and scratches. I asked about this once and was told it was the unavoidable result of having to dig them out of the Bahamian sand with trowels and shovels and other improvised excavation tools. The Seafarers salvage of the 1970's was not the well-financed, high-tech outing that Marex mounted 17 years later.




Available for $585. or 480-595-1293




S5. A second 1655 Mexican eight reales from the MARAVILLAS,

this one pedigreed to Capt Humpreys' MAREX salvage of the wreck

with his 1991 Marex photo-certificate.

Bold date, some (removable) encrustation on the final digit.  wt . 22.75 gm

ex "Treasure Auction #4" (11/08), lot 314, where it sold for $460.






Available for $460. or 480-595-1293




S6. From the wreck of the VOC East Indiaman Vliegenthart,

lost February 3, 1735 off the Zeeland coast of Holland,

this attractive full-date Mexican 1731/0 eight reales comes with the original Rex Cowan color cert (# 901769).


 Assayer F (Felipe Rivas Angulo) had the distinction of being the last assayer of silver cobs at Mexico City. This eight reales has everything you want in a Vliegenthart Mexican cob. Bold assayer and mint mark and full date---in this case, a scarce 1731/0 overdate. Nice cross with well struck lions and castles. Large 37 mm planchet, good weight at 25 gms, and t5he original Cowan color certificate.






Now available for $325 or 480-595-1293




S7. From the  galleon NUESTRA SENORA DE LAS MARAVILLAS (lost 1656)

A rare date 1650 P Mexican eight reales.

Excellent weight, 26.4 gm, on a large 40 mm planchet.

Only the third 1650 Mexican cob known from the Maravillas.




Although the vast majority of Mexican silver cobs recovered from the Maravillas bear the dates 1654 and 1654, the Maravillas has yielded a small quantity of dated Mexican cobs going back to 1650. 1650 is the easiest and rarest of the dates, with only two other 1650's known.  This coin--and for some reason almost all the better-quality Maravillas silver-- is from the Bob Marx early 1970's salvages. A good representative sample of Marx's finds appeared in Schulman's December 1974 auction. Notice lot 210, the only 1650 in the sale, a rare date (as Calbeto notes) that realized a princely $175 while choice 1655's went for only $100 or so!


This 8 reales is certainly among the top 5% of Mexican cobs recovered from the Maravillas. Compare it with the silver cobs featured in the later 1992 Christie's Marex sale. Without question it has better surfaces, a better strike, and better weight than any of Marex's highlight coins. The Mo mint mark and assayer initial P are bold and perfect, as is most of the distinctive mid-century Mexican style of the Hapsburg shield. The legend date shows 650, the 0 being punched higher and smaller-- I think the lazy die cutters just used the letter o from the mint mark. The 6 shows better to the naked eye than in this photo, though the lower loop  is quite clear. The 1 is even there, though faint.




Dated Philip IV Mexican 8 reales in this condition, with or without shipwreck provenance, now easily bring $700-800 at a major auction. Dateless pieces with 2-3 gms of corrosion bring $400+. This 8 reales is as nice as you can expect to find in a mid-century Philip IV Mexican cob.


Available for $650 or 480-595-1293


)2007 All Rights Reserved.