I. In the spring of 1713 the War the Spanish Succession formally concluded with treaties signed at Utrecht. Communications and commerce between Spain and Mexico began to return to normal. Mexican viceroy Don Fernando de Alencastre, First Duke of Linares, permitted the first fleet since 1709 to sail in January (still with a French escort). General Don Pedro Ribera in Nuestra Senora de Guadalupe led eight merchantmen out of Vera Cruz on January 19th, arriving safely in Spain on March 30th. Seven weeks before Ribera’s departure, the Flota of 1712 had arrived in Vera Cruz under the command of quarrelsome General Don Juan de Estaban Ubilla. Due in part to a very difficult relationship with Linares, Ubilla began one of the longest enforced stays in Vera Cruz in the history of the Treasure Fleets. For 28 months Ubilla sat in Vera Cruz, until finally released on May 6, 1715. In the holds of Ubilla’s Capitana and Almirante were all the escudos of 1713 that survive today.
We do not yet know the amount of gold Mexico City struck in 1713, although prospects for recovering this information from the Mexican archives are now good. Numbers published in Pradeau (1938, 1950) and Cespedes (1996), proposing a constant Fleet-era mintage of 158,440 pesos, are ludicrous guesstimates based on misreading of a 1858 encyclopedia article. A Vera Cruz manifest for Ubilla’s ships has been located, but, with the exception of a 1000 special production onzas being sent to the Queen’s Treasury (all later recovered), there is no inventory of the gold coinage carried on Ubilla’s fleet.
No one has undertaken a careful census of the surviving population of 1713 Mexican onzas. With no intent to produce a complete census, we have assembled a sample of ninety-six specimens for our study. All of the 1713 onzas in our study were selected because they are published with photographs: 48 in the State of Florida Collection, 42 in Tauler’s on-line census of the marketplace, and 7 from other commercial sources. We believe this sample is a large relative to the surviving population, perhaps approaching 50% of it. All 96 coins in our census appear to be from the 1715 Fleet. Prior to 1963, Mexican onzas of 1713 were unknown to scholars like Lopez Chavez. His 1962 monograph on eight escudos records no known 1713 Mexican onzas. It seems to fair to say that the extant population of the 1713 onzas is a gift of salvages from the Fleet. The State of Florida received about fifty-five 1713 onzas in annual divisions (1964-present) with the salvors.
From our study we conclude that four shield and at least five cross dies were used to strike the 1713 onzas. These are more dies than most numismatists would expect for a coinage that coiners knew might not exceed the smaller mintages of 1711=12 onzas. All shield dies are new with no overdating of dies recycled from 1711-12. An average shield die life of only about 2300 coins seems to suggest either problems with the dies or with planchet preparation or with the manner of striking the coins. Interestingly, none of these problems, and especially no rapid shield die deterioration, is immediately evident in our sample. The overabundance of cross dies in 1713 can be partially explained by holdovers from the cross dies of 1711-12. We know that at least one cross die (R) was used first in 1712 and suspect another (D). The cross fleury design ended in 1713, making dies of that design useless after that year. Coiners could be forgiven for using still serviceable dies from 1711-12 before they became useless. The question remains why they needed to use five cross dies, and why, as we shall see, they made such extensive use of the Royal die (R).
With the exception of shield die 4 and cross die R—the Royal dies for 1713—all the shield and cross dies used in 1713 were poorly engraved. Horizontal and vertical lines within the shield waver and bulge and slant. Lions and castles and fleurs are ridiculously crude. Tressures and brackets and crosslets never manage to follow a straight line or even a smooth curve. The only explanation for this abysmal level of die sinking is that the dies were cut by a very unskilled tallador. For some reason Mexico City apparently had no skilled engraver available in 1713. Fortunately, the problem does not continue into 1714, when a new design and new, well prepared dies appear. Without question, 1713 represents the nadir of Fleet-era die production at Mexico City.
II. THE SHIELD DIES. Our study of ninety-six 1713 Mexican onzas confirms that only one shield design was used at Mexico City in 1713. The new design was hardly a radical redesign of the Bourbon arms. Its most conspicuous change was removing the small triangular shield of Granada. This small triangle, originally placed centrally atop the arms of Flanders and Tyrol, had since 1702 had been positioned just above and to the left of the Bourbon escutcheon. Beyond die simplification, no (political) reason is known for the removal of the triangle of Granada. No shield of 1713 has the triangle of Granada. The lower castle of Castile and Leon is absent as in previous years. A small rectangular space remains where the lower castle had been, now filled with two or three thin vertical lines on most dies. Added to the 1713 dies are four quatrefoil floral stops flanking the shield above and below oXM J and VIII. Six stops of his sort had been present in 1711-12 legends, but never inside the inner border adjacent to the shield. On some lightly struck coins these four-pedaled stops look like crosses. See the 2nd shield die below.
A complete discussion of the 1713 shield dies would have to mention the distinctive crown of 1713, but this rarely appears on business strikes and is best seen on the 1713 Royal onza. We will discuss it presently. Some doubt exists about the number of fleurs in the field of New Burgundy (on the lower right side of the shield). Poor engraving exacerbates this issue. What we can say is that it seems Mexico City tried to engrave three fleurs on all of its 1713 shield dies, but poor positioning of the fleurs meant that on many coins only part of the design is visible and sometimes it seems only two fleurs are present. The shield side (obverse) legend, unchanged from previous years except for the date, reads +PHILIPPUS+V+DEI+G+1713.
In what follows all the 1713 onzas pictured or referenced have a Tauler or Florida State designation. (T1) is the first coin in Tauler’s 43 coin online census of Mexican 1713 onzas available here http://onzasmacuquinas.com/rafa11.php. (T25) is the 25th coin in that census. Florida State coin are all identified by their seven digit State inventory number 11.00XXX. See appendices 1 and 2 for more information on each coin.
The First Shield Die. Pictured is the Caballero coin (T1) sold by Aureo in 2009. Notice immediately the third digit of the date. Some sort of die error at this point left a circular mark over which the 1 had to be cut. The Caballero cataloguer, perhaps in gest, wondered whether it was a 1/0 error. 1713 over 1703 seems a strange error. Other diagnostics of the 1st shield include the o of oXM J positioned very high right at the left corner of the shield, sandwiching the floral stop into a very small space between the o and the crown. The stop is almost unidentifiable. The normally shaped X is nearly centered opposite the castle, and the well-formed M opposite the lion. Engraving within the shield is not terrible (for 1713), but the top of the Bourbon escutcheon slopes noticeably to the left and intersects the left central fleur. The bars of Aragon lean right. Finally, please note that calling this die the “first” shield die is an arbitrary designation and is not meant to imply that this die was the first shield used in 1713. For reasons that will become clear, we believe that there was no orderly sequence of die use in 1713, and that all the dies, including this shield, were used on multiple occasions in 1713. We have no idea whether this shield was in fact the first shield die used in 1713.
The Second Shield Die. Pictured is a fine example (T25) of the 2nd Shield sold by Aureo in 2000. Notice first the full crown. The distinctive crown design of 1713 is clearly shown with its unique double jeweled upper arches. Diagnostics of the 2nd Shield include two four-pedaled floral stops positioned right the corners of the shield. The upper left stop on the example shown here is not fully struck up and looks like a cross leaning left. The X in oXM J is asymmetric and the left down stroke touches the small M below it. The horizontal divider between the castle and lions is not straight. It dips below the castle and makes contact with the lion’s head. The leftmost vertical bar of Aragon is not straight but swerves to the left almost touching the right lion. The Bourbon escutcheon is not symmetric but bulges to the left. Above it, in the space originally for a lower castle, there are now several vertical lines.
The Third Shield Die. Pictured is an obviously early die state example (T31) of the 3rd shield. Heritage sold it in 2010 as part of the Don Benito Collection. For a later die state example of the 3rd shield, see the coin Stack’s sold in 2011 (T28). Diagnostics of the 3rd Shield include a usually mushy floral stop at the left corner of the shield. The stop overlaps the inner border. On no other shield die does the upper stop overlap the border. The o of oXM J is centered opposite the castle in a position somewhat lower than on the second shield die. Notice the curving die scratch running alongside it to the right. The X is also lower and horizontally asymmetric to the point that on some specimens that it looks more a V on a base. The horizontal divider between castles and lions bows slightly upward between the left sections, the opposite of the dipping horizon on the second shield.
Pictured is coin Ars Classica sold in 2002 (T14). Diagnostics of the 4th Shield include much better engraving and definition throughout the shield. The three fleurs of Burgundy, for example, are much sharper on this die than on any other 1713 die. So too the lions and castles. The horizontal dividers and vertical bars of Aragon are almost straight without the noticeable bows and bulges we see on the other 1713 dies. The o of oXM J is higher, almost touching the floral at the left corner. The X is also higher and horizontally asymmetric with the space between the bottom strokes filled in on all specimens. The M is nearly centered opposite the bottom lion. The position of the stop below J is lowest on any 1713 shield.
The unusually good engraving and definition on this shield die finds an explanation in the fact that it is Royal die for 1713, for some reason released for use for business strike production. We shall discuss this further in section IV below.
Tabulation of the frequency of the four shield dies yields one surprise. The 1st shield is used most often, the 3rd shield least often, but look at the frequency with which the 4th shield die is used. The fourth shield appears 24 times in our sample. The 4th shield die, remember, is the Royal die for 1713. 25% of the business strikes in our sample are struck from the Royal shield dies.
III. CROSS DIES. Our study confirms that only one cross design was used in 1713, the cross with crosslets or cross fleury design begun in 1711. 1713 is the final year of the cross fleury design. Despite what some numismatic guides allege, no Box Cross reverses (used before 1711) or Ornate Cross (used 1711-12) are found associated with shields of 1713. The cross fleury of 1713 is the same as the crosses of 1711 and 1712. Comparison with a limited number of 1711 and 1712 onzas suggest that the 1711-12 cross fleury dies were somewhat better engraved (as the 1711-12 shield dies certainly were), but there is no design difference in the cross fleury dies for 1711-13.
Our study found at least five cross dies in use with the 1713 onza coinage. The poor quality of many, perhaps most of the 1713 reverses, exacerbated by bad photography, causes us to hesitate in saying with complete confidence that exactly five cross dies are used with the surviving 1713 onzas. For a few of the 97 onzas in our study no high confidence identification of the cross die is possible (at least with our skills).
Cross Die A. Pictured to the left is a coin (T19) Aureo sold in 2008, one of the few 1713 onzas showing a complete cross and tressure. Notice that every tressure loop has a different shapes and so too the brackets connecting them. Notice, for example, how the top of the left bracket juts horizontally well beyond the tressure and almost touches the fleur. All the crosslets also have different shapes, often distorted by poor striking. The bottom terminus of the main vertical crossbar is bent to the left, sometimes appearing as angling off at as much as 45 degrees, sometimes more modestly as here. The fleurs all have different positioning with respect to the tressures. All in all, a masterpiece of engraving!
A second example of Cross die A is the Caballero coin (T1) shown to the left. This one of the few 1713 onzas with a good section of legible legends. We can read REX + HISPANIA… Notice the RE angle noticeably upward, a good diagnostic for this die, except it is clearly visible on only three or four coin. The vast majority of 1713 cross dies show no legible trace of legends, making correct orienting of the cross die difficult. Most catalogue photos, including those in Tauler, also must be rotated 90 or 180 degrees to view the cross on its proper axis.
Cross die A is the most common reverse with the first shield die. Tauler 1-3, 6, and 19 are clear examples. Cross die A also pairs with the second and fourth shield dies. No cross die A reverse shows the kind of wear and deterioration we sometimes see with cross dies R and D.
Pictured to the left is another exceptional Caballero coin (T26). Notice the asymmetry of the tressures and brackets. No two are the same, nor do they match any of the tressures and brackets in cross die A. The crosslets in this die seem noticeably larger than those in cross Die A (at least in this exceptionally clear striking). The fleurs look larger and have strange distinctive shapes on this well struck coin. If the cross die A and R fleurs look like dolls, these fleurs look like primitive totems or ornaments standing on a three-prong base. Look especially at the fleurs below the horizontal crossbar. The horizontal crossbar also appears to dip down at its right terminus and turn up at the left. HISPANIAR… can be read in the legend, allowing us to orient the coin correctly.
Cross die B pairs with both the 1st, 2nd and 3rd shields, but not the Royal (4th) shield. No Cross die B shield shows evident die deterioration pointing to imminent failure.
Cross Die C. Pictured to the left is an example (T20) that Aureo sold in 2013 as part of Tauler’s own Oro Macuquino Collection. Notice the very straight crossbars. The upper terminus of the vertical crossbar shows a spike extending into the top bracket. The right bracket is very large and nearly touches the inner border at two points. The bottom bracket has a distinct convex curve. The fleurs are in the “small doll” style. The fields show heavy diagonal polishing marks on most specimens. Cross die C pairs with all four shield dies.
Cross Die D. Pictured here is one of two known examples (GC1,T28) of the reclusive die D. Cross die D occurs only with shield die 3 and only in a late die state with conspicuous die failure at the bottom crosslet and elsewhere. The shield side is of this coin is shown in section 1 above. It also shows wear. An initial survey of the 1713 crosses confused the rare die D with die B, but the tressures and brackets are clearly different. No 1712 onza yet confirms this, but we suspect that die D (like die R) is a holdover from 1712, briefly pressed into service late in the 1713 mintage. Why a worn die like D would be used if crosses B, C, and R were still available and in good condition remains a question.
Cross Die R. Cross die R is so named because it is the same die used to strike the two known 1713 Royal onzas. We know from other years that Royal dies could be pressed into service to strike business strikes. It happened in 1714 and perhaps in other years in the Fleet era. The excuse in 1714 seems to have been that the standard dies for 1714 became unusable prematurely, and, whilst new 1714 dies were being prepared, the Royals dies were used briefly (probably just to complete one delivery). By contrast, cross die R pairs with all four of the 1713 shield dies and seems to strike a large percentage of the 1713 onza mintage. In fact, 38% of the 1713 onzas in our study are struck with cross R! Cross die R is not used as a brief, temporary fill-in.
Pictured to the left is an example of a cross die R coin (T15) sold by Cayon in 2002. Another picture of cross die R is in section IV below, where we see that it used to strike the 1713 Royal. This cross die is better engraved than the other 1713 dies, but it is still surprising crude for a Royal. Tressures and brackets and crosslets are more carefully engraved than on the other dies, but notice all the irregularities. Fleurs are in the same “large doll” style as used in 1711-12, but positioned more irregularly than we would expect on a Royal.
Notice the small spike extending from the left end of the horizontal crossbar into the left bracket. This enlarged as the die was used. A second spike developed between the upper terminus of the cross and the top bracket. A curving die break developed between the top of the right bracket and the right crosslet. The Royal cross shows a little of this deterioration. Cross die R pairs with all four 1713 shields.
The story of cross die R is still more complicated. It is in fact a holdover die from 1712. A single (now three) Royal onzas from 1712 survive and its reverse is not struck with die R. Pictured to the left is Florida State coin 11.00022, clearly a business strike. The obverse is from the same die as the Royal for 1712, but the reverse is our cross die R. Consider what this means. The Royal cross die for 1713, our cross die R, was first used in 1712 to produce business strikes. Remarkable!
A tabulation of the 1713 cross dies is below. Note that 38% of the business strike in our sample are struck from the Royal cross die R.
IV. KNOWN DIE COMBINATION. Below we tabulate the fifteen die combinations we encountered in our study. Twenty die combinations are mathematically possible from 5 cross and 4 shield dies. Two die combinations we did not see but may still exist are A3 and B4. It seems unlikely that worn die D made more than one brief appearance with shield 3. Fifteen varieties is an unusually large number to encounter with a small coinage.
V. THE ROYAL ONZA FOR 1713. Until recently, a single 8 escudos Royal was known for 1713. It appeared in a January 1985 FUN auction described as a “beach find” from the 1715 Fleet [Krueger]. No further details on its provenience were offered. Collectors and salvors know that if a coin is found in the water in Florida, it must be surrendered to the lease-holders for division with the state (which would claim a unique coin like the 1713 Royal). But if it is found on the beach, it is the sole property of the finder. Royals and other expensive Fleet onzas seem to be especially buoyant objects as many of late are discovered “on the beach”.
The first question we must address is whether this 8 escudos should in fact be considered a Royal. Clearly it is not a business strike. From the 1714 gold coinage we are familiar with a few coins properly called “failed Royals”. Royals were multiply struck on specially prepared planchets. A press was used but without coin collars, which meant the uniform expansion of the planchet to a desired round 35.5 mm was not guaranteed, especially if there were any irregularity in the thickness of the planchet. Some multiply struck coins went out of round or failed to expand enough. They could not be used as Royals. Apparently the ever frugal Mexican coiners just consigned their failures to the circulating coinage and tried again. See the discussion of failed Royals in our essay on the 1714 coinage [Flemming].
The question is whether this 1713 onza should be considered a failed Royal rather than a Royal. The coin obviously did not expand to its full desired diameter, and it is noticeably out of round. Look at the edge where the top of PHILIPPUS is truncated in a wavy fashion by an out of round planchet. Look also the misshapen Bourbon escutcheon. But this was not a fault in the striking. The die is poorly engraved at this point, certainly not up to the standards of previous years (1702, 1711, 1712).
There can be no certainty in this matter, but we believe this onza passed muster with the Mexican coiners in 1713 and was presented as a Royal. Mexico City struggled mightily in 1713 to produce decent gold dies. We demonstrated above the poor quality of the business dies they produced, but the Royal shield die for the 1713 shows at least an effort was made to do fine engraving. That the 1713 Royal was probably the poorest quality Royal produced by Mexico City speaks again to the special problems in 1713.
The reason why the Royal dies for 1713 were used, singly or together, to strike much of the 1713 onza coinage remains a mystery. In the case of the use of 1714 Royal dies to produce business strikes, a likely scenario suggests itself. After a fairly modest coinage both sets of standard 1714 dies expectantly failed, and Mexico City had no choice but to press the Royal dies for 1714 briefly into service until new dies for 1714 could be prepared. That explanation does not fit 1713 at all, where the Royal shield (4th shield) pairs with all cross dies and shows deterioration consistent with significant use. The Royals dies for 1713 were not brief fill-ins but workhouse dies for 1713. Why was this necessary or allowed? None of the regular shield or cross dies (except D) shows deterioration consistent with imminent failure. The argument that the Royal dies needed to be used in 1713 is not sustained by our 97 coin study. After a certain point it almost seems that the Mexican coiners were encouraged and preferred to use Royal dies in place of the poorer quality business dies.
Conclusions. We end our study with more questions rather than answers. We have discovered that the 1713 onza coinage was accomplished with 4 shield and at least 5 cross dies. Except for a pair of Royals dies (one recycled from 1712), these dies were all poorly engraved. The 1713 onzas cannot tell us why their dies were so poorly engraved. The 1713 onzas also cannot tell us why 15 die combinations were needed to strike probably a small number of coins. If we reject the unlikely scenario that there were 15 separate deliveries in 1713, then the coiners were using multiple dies in a single production run. If the coiners needed to produce 2000 onzas, why would they switch between three sets of dies during production? But that is what 15 varieties imply.
Four shield and five cross dies would not have been prepared in anticipation of a small coinage. Surplus cross dies from 1711-12 help explain the availability of five crosses in 1713. No 1711-12 shield dies were overdated for use in 1713, but the extensive use of 1713 Royal dies in regular production added to the total of production dies available. The availability of 4 shields and 5 crosses dies, however, does not explain why all these dies were actually used. Why in particular did the Mexican coiners choose to use Royal dies for about a third of the business coinage? There is no evidence that shield dies 1, 2, and 3 or cross dies A, B, and C were failing. In circumstances in which serviceable business dies were still available, Mexican coiners were apparently permitted and preferred to use Royal dies for regular production.
One piece of evidence we have not tried to extract from the coins is a die sequence. This is not completely due to sloth. The fact that cross dies A, C, and R were used with all four shield dies (and cross dies B with three shield dies) suggests that there was not a simple, orderly sequence of use. Fifteen die combinations suggest promiscuous and almost random die pairing. Many of the old photos in our study barely allow die identification. We happily resign to others, if they should be interested, an attempt to judge the die states of all the coins. Some coins are clearly struck from new dies, others from deteriorating dies, but in most cases it seems to us that it is not possible to make reliable die state calls. We are content to have begun a detailed study of the 1713 onza coinage.
Appendix 1: Tauler’s online census of 1713 Mexican onzas classified by varieties. All shield identifications are confident, a half dozen cross die identifications are not and are so marked*.
Appendix 2: Some 1713 Mexican onzas in the Florida State Collection classified by varieties. All Florida State gold cobs are identified by a seven digit number 11.0XXXX. We omit the 11.0 in the tabulation that follows. Florida acquired about fifty-five 1713 Mexican onzas in Divisions since 1964, of which 27 are dated and listed as 1713’s in the State’s inventory. By trades and thefts in the 1970’s that number has been reduced. No traded or stolen coins are included in our sample. All shield identifications are confident, but all cross identifications are not and are so marked*
Barriga Villalba, A.M. HISTORIA DE LA CASA DE MONEDA. Bogota, 1969
Bowers & Ruddy Galleries. The Harold A Blauvelt,…& 1715 Spanish Treasure Fleet Auction. Los Angeles, February 17-19, 1977.
Cespedes del Castillo, Guillermo LAS CASAS DE MONEDA EN LOS REINOS DE INDIAS. Vol 1. LAS CECAS INDIANAS EN 1536-1825. Madrid, 1997.
Craig, Alan SPANISH COLONIAL GOLD COINS IN THE FLORIDA COLLECTION Gainsville, 2000.
Flemming, Philip THE GOLD COBS OF MEXICO 1697-1732. Privately printed. Scottsdale, 2013.
“Varieties of the 1714 Mexico City Eight Escudos” Journal of the US Mexican Numismatic Assoc. Carefree, AZ., June 2014
Krueger, Kurt The 1985 FUN Official Auction. Orlando, Jan 3-5, 1985.
Lazo Garcia, Carlos ECONOMIA COLONIAL Y REGIMEN MONETARIO PERU: SIGLOS XVI-XIX. Lima, 1992
Pradeau, Alberto NUMISMAIC HISTORY OF MEXICO. Los Angeles, 1938.
Proctor, Jorge “Caudales y Flotas 1700-1715”, unpublished research paper, 2014.
Schulman, Hans SPANISH GALLEON TREASURE. Schulman Coin & Mint. New York, November 27-29, 1972
PART I. Portrait courtesy of Wikipedia Commons, coin photo courtesy of the Gold Cobs Co.
Part II. All coin images courtesy of Rafael Tauler’s online catalogo Escudos Macuquinos.
Part III. All coin images courtesy of Rafael Tauler’s online catalogo Escudos Macuquinos, except for Florida State coin 11.00022, courtesy of the Florida State Collection, BAR.
Part V. Coin image courtesy of Kurt R Krueger auctions.