To avoid fraudulent use, the stamps were sent to the districts where they were supposed to be overprinted with the district name. Stamps without this overprint were not valid for postage, so the use of un-overprinted stamps constitutes illegal usage.

Early plate, no overprint. Early plate, APAM in black

Just as the postal clerks sometimes forgot to overprint a stamp, they also managed to apply the overprint more than once. Though not a common event, it is possible to find stamps with two or more overprints.

Double overprint, tete-beche. Two different overprints ("Habilitado").
Chalco and Mexico. (Chalco and Mexico cancels).
Triple overprint.

It was up to the individual districts to produce their own overprints, just as they had to use whatever canceling device(s) they found suitable.

It is unknown how these handstamps were made. Most information about them is based on speculation and indirect evidence. It is thought that they were often made from some kind of type metal, perhaps made from newspaper types, held together in some unknown way.

Some of the larger districts used more than one kind of handstamps, necessitated by the large number of stamps they were overprinting. As an example, Mexico used several types, with different fonts, and even seemingly identical overprints may differ on enough points that it is most likely they were actually different handstamps. However, due to the method of application, most likely being done by hand individually on each stamp, some overprints that appear different may have been made with the same device. Slight differences in the angle, pressure, inking and many more factors could produce seemingly different results. Much more research needs to be done in this field to draw more meaningful conclusions.

Various MEXICO overprints.

Since the overprints were often applied by hand they can be found in just about any position possible on the stamps, reading top to bottom, bottom to top, diagonally, upside down and somewhere in between. In rare cases they were even applied after the stamp had been placed on an envelope and cancelled. Pairs showing one stamp with an overprint and the other without are also known.

Overprints in different directions.

There's also some evidence to suggest that overprints may, in some districts, have been applied in larger units, for example by arranging four or more handstamps in some kind of frame. This would have increased efficiency, and was most likely done at a later period in time. So far, all we have is speculation.

It should be noted that in some cases the overprint could be an aid in determining which plate a stamp originated from. For example, for the 2 reales used in Mexico the solid serif font overprints are mainly found on plate I & II. The hollow type overprints are found mainly on plate III stamps. Bill Wagner wrote to the Meximail forum, regarding overprints on stamps used in Mexico:

"It would appear that the switch-over from the solid serif to the common hollow-letters type was synonymous with the delivery of plate III 2 reales (and perhaps the plate III & IV 1 reales) on Jan. 10, 1857. This is hard to prove, since Mexico postmarks with legible dates from 1857 almost don't exist. But a very large percentage of the early, wide plate 1 and 2 reales (60-subject plates) have the solid serif overprint, and a very large percentage of the later plates 1 and 2 reales appear to have the hollow letters overprints, so in the absence of contradiction from stamps that break this pattern, it's a pattern. The first hollow letters handstamps produced impressions 15 mm. in length. On 2 reales, these show up on deep blue greens and light yellow greens; known (to me) dates range across 1857 into May of 1858 (on a 1 real). Then followed the most common hollow letters type, which is 14.5 mm. 1858 being a good year for postmarks in Mexico, there are a good number of dates on these, from July 15 (if not earlier) through at least October of 1859 (probably later). The last hollow letters overprints are the shortest (notice that these start out long, drop to medium, then finish up short): lengths under 14.5 mm (14.25 down to 13.75) show up by February 1861 (probably before this). These short, hollow letters handstamps are also the first ones used on the 1861s - oftentimes they look like they've been dropped several times or otherwise abused. The 13.75 lengths run until at least Dec. 1861 (with an occasional longer one), overlapping the solid serif type (in use by September 29, 1861)."

Normally the overprints are in black, but a few other colors were used occasionally. Campeche, Hermosillo, Pachuca and Saltillo can be found in red, Apam, Chiapas, Guadalajara, Isla del Carmen and Saltillo in blue, and Mexico exists in green.

The following districts and sub-offices were sent stamps, but did not overprint them according to regulations, or used already overprinted stamps from the district they were dependent on:

District Normal dependency Temporary dependency Overprint
Colima     None
Guadalupe Hidalgo Mexico   MEXICO with arrows
Perote Jalapa Mexico None
Polotitlan     None
San Felipe del Obraje Maravatio Mexico None
Tacubaya Mexico   None
Tepeji del Rio Mexico   None or MEXICO
Texcoco Mexico most of the time.   None
Tlalpam Mexico   MEXICO
Villa del Valle Toluca Mexico None
Zitacuaro Maravatio Mexico None

Perote, Polotitlan, San Felipe del Obraje, Villa del Valle and Zitacuaro were not independent districts, but were sent stamps directly to meet emergency needs.

Stamps without overprints can only be recognized by the cancellation. However, the cancel is no proof that the stamp was issued in the cancelling district, since the stamp could potentially have been bought in another district.

From the above it can be concluded that Colima and Polotitlan were the only independent districts which did not follow the regulations regarding overprinting.

The following table shows the names of the districts, which overprinted the stamps as they were supposed to do. The numbers indicate a relative scarcity coefficient, adapted from Celis Cano. They can be used to derive the price for a stamp by multiplying the coefficient with the price of the cheapest copy of the same stamp. For example, if a 1 real from Mexico cost $5, then a 1 real from Ciudad Bravos would probably cost around $5 * 100 = $500. This is of course only a guideline. This coefficient is based on the availability of the stamps on the market.


Known overprints ½ real 1 real 2 reales 4 reales 8 reales
Acapulco 6 10 15 5 2
Aguascalientes 60 60 60 U R
Aguascaltes U R R R U
Apam 1 2.5 10 10 1.2
B. California U R R U U
Campeche 4 3 2 1.5 2
Chalco (*) 4 2 50 U U
Chiapas 2.4 10 30 4 1.6
Chihuahua 5 2 1.5 1.5 1.2
C. Bravos R 100 R U R
Cordova 4 2 1.5 2 2
Cuernavaca 2 1 2 2.5 1.2
Durango 2 1 1 1.5 1.2
Guadalajara 1 1 1 1 1
Guanajuato 1 1 1 1 1
Hermosillo 6 7.5 10 2 2
Huejutla 6 6 20 2.5 3
I. del Carmen R R 100 R R
Ixtlahuaca R 75 40 R R
Jalapa 1 1 1 1 1
Lagos 4 4 1 1 1.6
Lerma 90 R R R R
Maravatio 4 10 10 4 2
Mazatlan 1 1 1 1 1
Merida 1 1 1 1 2
Mexico 1 1 1 1 1
Monterrey 4 1.5 1.5 3 3
Morelia 1 1 1 1 1.2
Oaxaca 1 1 1 1 1.2
Orizava 1.2 1 1 1.2 1.4
Pachuca 1.2 1 1 1.5 2
Puebla 1 1 1 1 1
Queretaro 1 1 1 1 1
Saltillo 10 2.5 5 R R
S.L. Potosi 1 1 1 1 1
Sultepec R R 100 R R
Soyaniquilpan 90 50 50 R R
Tabasco 5 4 3 2 1.5
Tampico 1 1 1 1 1
Temascaltepec 20 40 40 R R
Tixtla Guerrero R 30 15 R R
Tlalpujahua 90 25 10 R R
Toluca 1 1 1 2.5 2
Tula 10 20 10 2.5 R
Tulancingo 2.4 5 5 2.5 3
Veracruz 1 1 1 1 1
Victoria 8 15 20 R R
Yguala R R R (Plate III only?) R U
Zacatecas 1 1 1 2.5 5

(*) The coefficient for Chalco is derived from the Follansbee catalog. The numbers found in Celis Cano are far off compared to the more recent prices quoted. U indicates that the stamp is unknown, R means rare. Celis Cano sometimes indicates that a stamp is rare, but Follansbee indicates it is unknown. I have chosen to make the indication rare, unless both of them indicate that it is unknown. Rare should be taken to mean a coefficient of over 100, but often the stamps are so rare that you may not be able to buy them at all. In this case, the coefficient has no meaning.