Valira Torrent - bulletin of the Andorran Philatelic Study Circle. Issue 42, p7-14 (Nov 1995).
It is now thirty-five years ago that the late Derek Tanner wrote his article on this subject; it was entitled "Provisional Cancellations of Andorra" and was published in an issue of "Stamp Collecting", dated 12th August 1960. His article has been the only authoritative study published on these elusive and seldom seen cancellations. The time has now come to update and slightly expand Derek's original study.
In 1959 these comprised of Canillo, Encamp, La Massana, Les Escaldes, Ordino, San Julian de Loria and Santa Coloma (Soldeu had been closed for many years). At this time, all the above Agencies still had their original datestamps which had been supplied to them wayback in 1928 when they were first opened.
However, it was realised in 1959, that these datestamps could not print dates after 1959 - this was because there was not a "6" on the penultimate band of the dater. An examination of the dating mechanism reveals the problem that occurred in 1959, and also what may have happened in 1939 as well. The dater was comprised of wheels or revolving bands as set out in the diagram. When issued in 1928 the penultimate wheel had only four numbers 2 2 3 3 plus some blanks - thus it could only print dates for the 1920s and 1930s (A). It is thought that it contained two each of 2 and 3 so as to allow for possible wear. It would seem that in late 1939 or early l940, these handstamps must have been returned to Spain, in order to have the numbers 4 and 5 added, so that dates of the l940s and 1950s (B) could be used. What happened during this period is not clear, and the only evidence that we have that something was different at that time, is a printed paper rate letter of March l940 to Barcelona. This is cancelled, most unusually, with the Head Office VALORES DECLARADOS handstamp - an item usually reserved for Insured Mail only. The letter is dated -9 MAR 0, there being a blank where the 4 should be. This could indicate that this handstamp had been retained for cancelling mail whilst the others were away being altered.
Similarly, at the end of 1959, it became necessary for the numbers 6 and 7 to be added to the penultimate wheel, so that dates of the 1960s and 1970s (C) could be printed. It is not recorded exactly when all the agency handstamps were withdrawn and sent to Madrid for the alterations, but late September/early October seems the most likely. The earliest "provisional" that I have seen is 12 or 13 of October 1959 (this is a postcard from San Julian de Loria to Barcelona, but the 12/13 is "lost" in the colour of the airmail 1pta stamp - however, the text on the card is clearly dated 12/10/59).
Whilst the alteration of these handstamps should not have been a major task, they were not returned to Andorra until the middle of February 1960. During the period of absence, the Agencies had no official cancelling devices whatsoever; also it would seem that they did not receive any directions from Head Office at Andorra la Vella as to a uniform method of cancelling the mail.
And so the scene was set for the "provisionals" and a very interesting period of Andorran postal history.
a) pen cancellation - Canillo and date in manuscript
b) a most curious two-line combined name/datestamp, comprising a mixture of capital and small letters. It was made up of rubber type held in a wooden slotted holder and very similar to that found in the "JOHN BULL" printing sets. This provisional always seems to be struck in palish purple, and the day of the month is always handwritten. There are some interesting variations - January is shown as ENR - this is incorrect and should be ENE (perhaps there was only one E in the printing set!); and February dates always show the year abbreviated to 60 instead of 1960 as for January.
a) pen cancellation - Encamp and date in manuscript
b) datestamp only; usually in black
c) two line handstamp inscribed ENCAMP / ANDORRA with additional separate datestamp (b) in centre, usually struck in black. This handstamp, made in Barcelona, was originally of three lines ENCAMP / CORREOS / ANDORRA - and was made on the initiative of the old postmaster. However, his son thought that, as it was not a strictly official device, the central CORREOS should be removed, and this was done with a knife.
d) an oval rubberstamp inscribed with the postmaster's name ALBERTO MARTISELLA, usually struck in black and used in conjunction with the datestamp (b).
a) pen cancellation - La Massana and date in manuscript, and sometimes with additional manuscript a (c.f. greek alpha) or similar, usually in blue-black.
There have been no recorded examples of pen cancellations or provisional handstamps from this Agency. Letters that were posted at Las Escaldes were sent on to the Head Office, Andorra la Vella, where they were cancelled with the current Andorra la Vella handstamp (type 2), or the machine canceller (type 1). This is the only Agency that did not utilise some sort of provisional cancellation.
a) pen cancellation - Ordino and date in manuscript. It is found both in blue-black and red.
a) pen cancellation - San Julian and date in manuscript in blue-black
b) pen cancellation - date only
c) pen cancellation - postmasters "squiggle" without date
d) date only - applied by rubberstamp in grey-black
e) similar to Les Escaldes, some mail was forwarded directly to the capital and received the Andorra la Vella cancellation
a) pen cancellation - Sta Coloma and date in manuscript in blue or blue-black.
Very few letters or cards of this period seem to have survived. It must also be remembered that internal mail from the Agencies did not always receive a cancellation as a matter of course; therefore it is quite possible that a fair proportion of internal mail, during the relevant period, did not receive any cancellations whatsoever - unless specifically requested!
Registered mail of the period is rare. One philatelic item has recently appeared at auction in France, being a letter from Canillo to Switzerland, cancelled Canillo (b) 3 DIC 1959. It is interesting to note that the stamps were all "over-cancelled" with the Andorra la Vella certificado of the same date.
The up-dated handstamps are believed to have been returned to the Agencies on 15th February 1960. However, some provisionals have been seen dated later in this month, particularly from Encamp. It would seem that these have been applied to a philatelic request. I have never seen any provisionals dated later than February 1960.
Many external letters of the period are franked with Spanish stamps, or a mixture of Spanish and Spanish Andorran stamps. This is normal and quite correct, as there were very few "denominationally different" stamps of Spanish Andorra available at this time. (see "MODERN STAMPS OF SPAIN OFFICIALLY SOLD AND USED IN ANDORRA" by Carlos Romo - Valira Torrent, Bulletin No 19).
Forgeries exist, and collectors should be very careful when purchasing "provisional" material because, by their very nature, the pen cancellations can easily be faked. In addition, several faked Encamp letters and cards are known, featuring the Alberto Martisella oval namestamp (d) and a datestamp inscribed FEVR i.e. in French! The same faker is also known to have used Spanish 1 and 5pta Francos with the imprint F.N.M.T.- B , but these two stamps were not issued until May 1960.
As mentioned earlier, the exact date of the withdrawal of the original handstamps for alterations is not known; therefore, if any members have any material with original type 1 handstamps during late September/early October 1959, then I would be very pleased to hear from them.
Head Office was in a different situation when compared to the Agencies. A new handstamp (2) had been introduced in 1954, and thus was able to print dates of the 1960s and 1970s. The original handstamp (type 1) was therefore withdrawn on 31/12/59 as there was no need to bother to alter it. In addition, Andorra la Vella also had a Machine Canceller in use, being brought into service in December 1958.
The problem at Andorra la Vella concerned the registered (CERTIFICADO) handstamp - this was the original (type 1) issued in 1928 and could not print dates after 1959. It is believed that this was sent to Madrid for alteration much later than the Agency handstamps, probably at the end of 1959. In the event, it would seem that it was not altered but withdrawn, and a new Certificado was made (type 2). This new Certificado was first seen used in September 1960, but was quickly withdrawn a few months later because of the LERIDA inscription - but that's another story!
Therefore, Andorra la Vella was without a Certificado for about eight months, and registered mail during this period was cancelled by a handstamp inscribed "C.P.DE A." (CAJA POSTAL DE AHORROS).
This was not actually a postmark, but a handstamp used on forms and documents related to Savings Bank transactions. Although this "provisional" was in use for some months in 1960, not many examples are recorded; like the agency cancellations, it is much sought after.
The C. P. DE A. cancellation has been used instead of the Certificado on other occasions. I have a registered letter from Canillo (dated 19 OCT 51) to England, and this has the Savings Bank handstamp as a transit marking on the reverse. However, it is not known whether its use in this instance was deliberate or accidental.
Provisional Cancellations of Andorra, D. W. Tanner (1960)
Fausses Oblitérations, Philandorre (1979)
My sincere thanks to Messrs David Lamb, Ted Jewell and Bostjan Petauer for the loan of photocopies. In addition, I am pleased to acknowledge considerable help from Ted Jewell on several aspects of the article.
Articles index | Home Page