Serial numbers: what do they mean?

Discussion in 'European & Other Pocket Watches' started by jboger, May 24, 2020.

  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  1. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    I have a Louisville, Kentucky, watch sold in the 1830s. It has the serial number 3. This number was of significance to the retailer, not the maker. I have many other Anglo-American fusees with much higher serial numbers, in the tens of thousands. Those sorts of numbers were not of significance to the retailer. In general, who decided what number was engraved on the plates and why?

    Any chance it was the engraver? It seems to me one of the last steps in the manufacture of a watch movement was to have the plates engraved. The engraver may have been only loosely connected to the watch industry as he and/or his shop would have engraved any work that came in, whether a watch, a scientific instrument, or jewelry. If he consecutively engraved whatever entered his shop, then this might (partially?) explain the apparently weak correlation between serial number and assay year that we observe today.
     
  2. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    11,249
    1,556
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi John,

    I think it was generally whoever commissioned the watch, commonly the retailer, or in many of the instances of very high numbers, the workshops where they were made. I believe that the engravers did what they were told, (more or less, there are always examples of mis-reading or mis-hearing instructions, or lapses of concentration!). The larger, high-end 'makers', who kept proper records, did, of course use what appear to be meaningful number sequences; Vulliamy's letter codes have so far resisted complete deciphering however.

    Regards,

    Graham
     
  3. John Matthews

    John Matthews Registered User
    NAWCC Member

    Sep 22, 2015
    2,311
    1,031
    113
    Male
    Retired Systems Architect
    France
    Country Flag:
    Region Flag:
    #3 John Matthews, May 24, 2020
    Last edited: May 24, 2020
    John - I would go a step further than Graham ..

    My answer to your question would be - it is absolutely impossible to generalise across time, makers, finishers, retailers, first owners.

    There are just too many variables. You can find unfinished movements at an early stage of manufacture with serial numbers, you can find finished movements with the same number engraved on the barrel plate as the original movement number stamped next to the movement maker, you find barrel plates engraved with what we assume is the retailer, together with a serial number and beneath the plate a different number, you can discover a movement maker's number beneath the dial, with the first owners name engraved on the barrel plate together with his date of birth, in a case hallmarked in his 21st year.

    At best for a single movement maker/finisher/retailer over a short period of time you may find a set of serial numbers that follow in reasonable chronological sequence. Some authors have provided data from which they have inferred that there may be more than one sequence of serial numbers in use at one time. In my opinion, in order to start to make inferences about a given 'individual's' serial numbers, you either need a surviving work/day book, a few exist, or you need to collect a data set that encompasses a significant portion of the output of the 'individual' of interest e.g.the Johnson database maintained by Oliver Munday. Reference to that database, reveals the additional problem in drawing inferences when dealing with a data set that is dominated by movement only examples, that are so common with movements that were exported to your side of the pool, in Liverpool's heyday.

    John

    Edit - my comments, obviously, apply principally to English watches - they clearly do not apply to American factory produced movements nor to some Swiss manufacturers.
     
  4. Dr. Jon

    Dr. Jon Moderator
    NAWCC Member

    Dec 14, 2001
    5,740
    464
    83
    Aerospace Engineer
    New Hampshire
    Country Flag:
    Region Flag:
    Serial numbering has different origins and meanings.

    Major manufacturers such as Waltham in the US and Longines used them to track assembly and identify replacement parts and were assigned in blocks. They claimed that with the number they could provide an exact replacement part.

    Smaller high end watch making operations operated out of ledgers with printed numbers on each page and assigned this number to the watch recorded on that page which identified the various specialists and what they were paid. This was common in Swiss English and French watchmaking. These numbers may have been printed to order .

    Some retailers such as Breguet had their own system for numbering watches and gave those numbers to teh makers they commissioned to make watches.

    Tiffany prevented makers from putting their names on the watches they bought. They used the maker's numbers to identify them for service.

    Pre Industrial watches and 1830 is of that era are less clear. It seems to me that the serial numbers for many of these are too high to be sequential, I believe they corresponded ot some ledger entries but I suspect that the numbers were whimsical.
     
  5. John Matthews

    John Matthews Registered User
    NAWCC Member

    Sep 22, 2015
    2,311
    1,031
    113
    Male
    Retired Systems Architect
    France
    Country Flag:
    Region Flag:
    Jon - There is no reason at all why high 'whimsical' serial numbers could not have been produced/marked in a sequential order - I think the whimsical aspect is the belief that anyone would ever believe that the named maker, every achieved the output that might be inferred from the sequence of serial number that were used.

    John
     
  6. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    11,249
    1,556
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi Jon,

    This is where the proposition grew from, that these high numbers couldn't have originated with the retailer, but were job numbers from the workshops who supplied them. There are also instances of the movement supplier and the retailer both putting their numbers to a watch, sometimes engraved on the top plate as a fraction.

    Regards,

    Graham
     
  7. John Matthews

    John Matthews Registered User
    NAWCC Member

    Sep 22, 2015
    2,311
    1,031
    113
    Male
    Retired Systems Architect
    France
    Country Flag:
    Region Flag:
    Graham - if, for example, we consider Johnson movements with a distribution of numbers up to 30,000 and then a few further examples from 30,000 right up to 90,000. I think we need need to ask a number of questions relating to who was responsible for assigning the numbers.
    • was it the 'finisher' Jos Johnson?
      • were the serial numbers only used in the Johnson 'workshop'?
    • if not and the job numbers were 'carried down' the supply line ...
      • were all the movements supplied by the same movement maker (very unlikely), who used the same job numbers?
      • more likely multiple movement makers who had multiple customers:
        • they could be dealing with multiple job numbers relating to different customers;
        • they might also have their own serial numbers, for smaller customers.
    I suspect, the former and it was the 'finisher' who assigned the job number - but I do not think that this job number inevitably became the serial number ...

    If the finisher only produced watches that carried his signature it was his 'job number' that became the serial number. Some finishers who sold to the trade may, optionally, have added a serial number, again probably the job number, when they sold watches on. 'Finishing retailers' could then add their signature and, optionally their serial number, often on the barrel bar. Finishers who supplied retailers with a 'ready for sale' product were probably more flexible and would add what was requested, be that the retailer's name and serial number or whatever was specified on the order.

    We do know from surviving examples, that high end small throughput makers, based in London, kept numbered day or work books, containing a detailed record of the work undertaken both by in and outworkers.. The number of each page identified the job. These makers often sold directly to private customers and unless requested otherwise, used the job numbers as their serial number for the finished watch.

    John
     
  8. Dr. Jon

    Dr. Jon Moderator
    NAWCC Member

    Dec 14, 2001
    5,740
    464
    83
    Aerospace Engineer
    New Hampshire
    Country Flag:
    Region Flag:
    I did a quick check on Daniels Book Watches. If its examples are representative numbering did not start until about 1750. Mudge numbered his watches as did Arnold and the others after about that time.

    The essential questions to me are:

    1) When and why did watches get numbers. In factories they are used to keep related pearts together ut what is the need before factories?
    2) What was the purpose of these numbers?
     
  9. Allan C. Purcell

    Allan C. Purcell Registered User
    NAWCC Member

    Feb 9, 2013
    2,323
    817
    113
    Male
    retired
    Germany
    Country Flag:
    Region Flag:
    Serial numbers lboger, are very much like old paintings, if there is an artist you favour from say the 18th or 19th century, you could spend a lifetime trying to uncover his production, and don´t forget that artists in the17th century did not always put their names to their work. Cataloguing watch serial numbers are very much the same, it takes time, The Roskell file on here as taken five years to reach the standard achieved so far, and probably, will never be finished, that goes of course for other lists and files. They are there though to assist people who want to know more about their watch. Let's go back to the artist, once chosen, and you are determined to find as much of his work as possible, you are in fact concerned with one person. Serial numbers are like that too, stay with the original maker. Most of the problems with serial numbers start to go wrong when other people get involved.(Partners, Family or early death) Take Barraud for example, his serial numbers were well recorded by Cedric Jagger, but in the period he was working in he could not achieve to reach the facilities we have today. When he was writing his book on Barraud they were only two of the Mudge copies known to him, today we know where all of them ended up one way or another. Barraud numbers changed when the firm took a partner Lund. Robert Roskell´s numbers I have found, are mostly reliable, though there are gaps, maybe one day they can be cleared up. There is also the problem of large production been hidden behind work when they sold watches named for other sponsors. This can sometimes create a large gap, much like the 10,000 range in Roskell watches. Serial numbers are a lifetimes work for the firm involved, Roskell´s work dates are from c1798 to 1843 when he retired. Look at the work in that wonderful book on Tompion "300 Hundred Years" remarkable research. So compiling serial numbers can be fun today, though there are those who won't take the time to compile them, each to their own they say. What makes me smile about collecting serial numbers, is when someone asks "why is my watch not on your list" So if you have a watch with a serial number ask first where there is a file on the maker, someone on here will know.

    Allan
     
  10. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    I haven't gone through all the responses--yet. And just so's we're on the same page, my question refers to English practices of the 19th C and earlier, not to the centralized methods practiced by the US makers. Also when I raised the question if the engraver may have been responsible for the serial number, I meant just the serial number; the rest of course was dictated by whomever consigned the watches to the engraver.

    May I toot my own horn? I do rather like my thought that the engraver may have been only loosely connected to the watch trade, and that whether he engraved a movement, a barometer, or a silver object was all the same to him. Whatever crossed his table. That of course is conjecture on my part. Perhaps some engravers had enough work provided by the watch trade. I simply don't know.
     
  11. gmorse

    gmorse Registered User
    NAWCC Member

    Jan 7, 2011
    11,249
    1,556
    113
    Male
    Retired from Xerox
    Breamore, Hampshire, UK
    Country Flag:
    Region Flag:
    Hi Dr. Jon,

    Thomas Tompion and Daniel Quare were numbering their watches rather earlier than this; Tompion started in 1681, Quare around the same time, and so did several other of their contemporaries.

    Although English watches weren't made in anything resembling a factory, as we now understand the term, they were the product of a great many specialised craftspeople working in largely collaborative sequence, mostly not in the same location, and as a frame passed through all the processes it was useful to have an identifying number attached to it and some of its sub-assemblies. If nothing else, it would also be useful in advertising for the return of stolen pieces!

    Regards,

    Graham
     
  12. Jeff Hess

    Jeff Hess Moderator
    Sponsor Gold Business Member

    Sep 3, 2000
    7,067
    356
    83
    Male
    watches
    Florida
    Country Flag:
    Region Flag:
    great thread. can we see the watch?
     
  13. zedric

    zedric Registered User

    Aug 8, 2012
    1,122
    152
    63
    Country Flag:
    Region Flag:
    #13 zedric, May 24, 2020
    Last edited: May 24, 2020
    And its worth noting that some makers used the same sequence numbers for both watches and for clocks, so the sequence relates roughly to their total output, but not to the number of watches. I don't collect watches (yet) but a number of my clocks have three sets of numbers on them, one for the movement (blanc roulant) maker, one for the finisher of the movement, and one for the case maker, and it is not unheard of to have four sets of numbers on one piece.

    Some of the sequence numbers definitely do not start from 1, for example I have a carriage clock by Charles Frodsham that has the Frodsham number 20209 clearly written on the dial and backplate (movement number 32438). This has a presentation engraving for 1897, and is a French clock that was finished by Frodsham.

    The number 20209 may be a continuation of a number sequence for watches (carriage clocks were sometimes considered as large "watches", I'm pretty certain this is stated in the jury report from the 1851 great exhibition). Or the number may be something completely different, but is certainly not related to the number of clocks, or carriage clocks, made by the firm to that point in time.
     
  14. John Matthews

    John Matthews Registered User
    NAWCC Member

    Sep 22, 2015
    2,311
    1,031
    113
    Male
    Retired Systems Architect
    France
    Country Flag:
    Region Flag:
    As an example, in a London trade directory for 1880 - the first that came to hand - there is a list of ~20 'Watch Engravers' concentrated in the watchmaking areas. No doubt they did other work e.g. heraldic work, simple inscriptions and monograms etc., but, given their location, I think it is reasonable to infer that for some their principle source of work was from the watch trade.

    John
     
    gmorse likes this.
  15. Allan C. Purcell

    Allan C. Purcell Registered User
    NAWCC Member

    Feb 9, 2013
    2,323
    817
    113
    Male
    retired
    Germany
    Country Flag:
    Region Flag:
    Letting the trade to put your serial number on your watches does not, in my opinion, make sense. It would involve taking the risk of bad workmanship, plus laziness, problems with alcohol, all well known in the trade. I would have thought the numbers would have been put there inhouse. Has Graham says these watches from the early periods went through many hands, this alone would be enough to cause backlogs. The serial number is there for security and a record that could be used as a guarantee. Tompion is credited with some five thousand watches, but it is also thought he made another five thousand-that were not recorded by him-though they had a serial number. The serial number was a control tool, if you have a safe, you don´t give the key to the labour force. There were also firms like Frodsham who put coded numbers and letters on their work, that too would not be put out. Though the serial number is not the be-all in the watch trade it is still one of the most reliable tools in research. The files I have produced, like Barraud, Barwise, Roskell and others prove this. Give it a try some time, the Swiss make money out of it.

    Allan
     
  16. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    I've now read through the posts in this thread. It seems there were multiple people for multiple reasons who had serial numbers engraved on watch movements--and I mean English watch movements of the 19th C and earlier. Under the US factory system, I think it's clear why individual movements were numbered: inventory control. Various parts of the watch--the plates, the barrel, the cock, and the balance wheel--were all given the same number as inventory control and a way to track parts during production, not a number assigned at the end of production. Given that most English fusees have only one number, and that number seems to have been applied near the end of production, it would seem that the serial number is not of significance to the actual process of production. There may be other numbers or letters stamped into parts of the watch that may relate to inventory control whilst the watch is being made, but not so for the serial number. (Note to self: need to think about this some more.)

    May I back up in time? Is it safe to say that the first generation of English watchmakers--Tompion and Quare, for example--made their watches on demand and not for a market or a retailer other than themselves? That the names engraved on the plates are the names of the makers who themselves sold the watches? If so, I could see the serial number having significance to the maker as part of his signature, a sort of declaration: "This was the 368th watch I made." And it would be a way of keeping track to whom the watch was sold in case it ever came back for repair. So in the early days, perhaps, the serial number originated with the sole maker = seller.

    Still, I'm trying to wrap my mind around who else was motivated to have a serial number engraved on a movement and why. Clearly in the case of my Louisville, KY, watch with a serial number of 3, it was the retailer (although I personally believe the watch, case and all including all engraving was done in England). I have a Mott's NYC verge from the 1830s with a three-digit serial number. Rich Newman also has a very fine lever fusee from Mott's with a three-digit serial number. I suspect those numbers were of significance to that retailer and indicate the number of watches or time pieces sold. These are Anglo-American fusees, the three of them. But I bet there were watches that never left the country with low serial numbers put thereon by the retailer.

    If that's the case, then watchmaking and selling had changed radically from the days of Tompion to when my Louisville watch was put in the store window. My Louisville watch was made for a faceless market.

    John B
     
  17. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    Jeff: when you wrote, Can we see the watch? was that directed to me? If so, here is the Louisville watch:

    DSCF1906.jpg
     
  18. Allan C. Purcell

    Allan C. Purcell Registered User
    NAWCC Member

    Feb 9, 2013
    2,323
    817
    113
    Male
    retired
    Germany
    Country Flag:
    Region Flag:
    #18 Allan C. Purcell, May 25, 2020
    Last edited: May 25, 2020
    Hi John, I like your post and the way you thought it through, but yet again I have to bring in "The Book" the earliest known watch by Robert Roskell 110 is in there for service in New York in the 1820s. That also goes for Jos. Johnson, and very low numbers for Litherland, not to mention the London makers, the Swiss and the French watches. I think Roskell started with No.100, but it would be nice to know he started with No.1 You wouldént happen to own it by any chance?:cool:

    Best wishes,

    Allan

    Edit: Page 4 1825-for the early 110 Roskell pluss 130 is in there too.
     
  19. zacandy

    zacandy Registered User
    NAWCC Member

    Nov 6, 2019
    15
    21
    3
    Male
    Country Flag:
    An interesting post thank you

    May I add a small comment from business nothing to do with horology really.

    Nobody in business starts with a serial number of one. One does not start with zero.

    It is 1840. I am a lawyer buying a very expensive pocket watch - an item it is vital to me to enable me to go about my business but I do not really understand. I rely on the honesty of a unknown young looking watchmaker near Royal Exchange recommended to me by a friend to sell me a reliable item. I would not buy from him if this was only the first or even tenth item he has made and sold!! I want to know he has done this before at least say 100 times.

    It is the same in any business of course.

    Hence the high looking numbers usually imv.
     
  20. Allan C. Purcell

    Allan C. Purcell Registered User
    NAWCC Member

    Feb 9, 2013
    2,323
    817
    113
    Male
    retired
    Germany
    Country Flag:
    Region Flag:
    John Arnold is credited with 30 number one´s, and the Mudge copies started with 1 to 27, but you are right about the attitude of modern business. Numbers under one hundred by the great names in Horology bring huge prices. One of those Mudge copies is now known as "The Antique Road Show Mudge" its on youtube. Please you like this thread.

    Allan
     
    zacandy likes this.
  21. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    Allan, were you asking me if I owned Roskell No. 1? That's a question I can answer with one word. Nope.
     
  22. jboger

    jboger Registered User

    Jan 7, 2019
    347
    114
    43
    Male
    Country Flag:
    I've read John M's post #7 several times, the one that discusses Jos Johnson's serial numbers. I think I got a handle on low serial numbers, those put on by the first generation of makers = sellers, and those put on by retailers like my Louisville watch. This leaves the great bulk of watches with serial numbers that range typically from thousands to tens of thousands. Just to be clear, I mean only the engraved serial numbers we see on the plates when we remove the cap. Who generated these numbers and when were they generated in the production process? It seems to me these numbers are late arrivals, applied somewhere near the end of production but before the movement went off to the case maker. If the casemaker stamped the serial number into the case, then he had this number when given the movement. Ditto for the engraver. Whatever the situation, these numbers date from the time when watches were made for a mass market and not by high-end makers who made individual pieces for the carriage trade.

    John B
     
  23. Allan C. Purcell

    Allan C. Purcell Registered User
    NAWCC Member

    Feb 9, 2013
    2,323
    817
    113
    Male
    retired
    Germany
    Country Flag:
    Region Flag:
    Hi John B, Serial numbers started with clocks, but it depends how far you can go back to the first set on watches, but first, you had to have a series to put the numbers on. Thank you for letting me know you do not have Roskell number 1, it was just a dream. :(. Have you had a look at the Roskell file, he started his serial in 1798?

    Allan
     

Share This Page