When l started collecting early Bulova ladies watches from the art deco period between 1922 and 1930, and got to learn about how to date them, I was surprised to find so few 1927 watches in the two main on-line databases of myBulova and Watchophilia. Of the total 380 ladies watches I could find, including the ones in my own collection, just 16 were from 1927. Looking at men’s watches, the statistics are not different: just 12 watches over a sample that must be almost double the size of the ladies watches.
The convention is to date the watches on the basis of the first digit of the case serial number, normally a 7 digit number. So case serial 7528946 is from 1927 while case serial 6653422 is from 1926 etc. Although that was a great discovery attributed to Mark Lawrence, in this forum discussion I want to make a case that there is an exception to that rule. And it is an exception that seems to apply to 1927 only, although further research is needed.
The exception is that there seem to be specific series of case serial numbers starting with an 8 or 9 that seem to be actually watches from 1927. They are identified by their (deviating) case signatures. While Bulova started indicating “New York” in their case signature as from 1928, these deviating 8 and 9 series still use the “Bulova – American Standard” co-signed case signatures as used from 1925 to 1927. This was discovered by me while studying how the case signatures evolved over time between 1922 and 1930. Once I had identified those specific 8 and 9 series with the American Standard signatures, I did a double check on their movement symbols. And indeed, over 90% of those cases had a 1927 movement.
In below link you will find a PDF with more detail. It will explain first the general rule of dating your Bulova, followed by the study I did on case signatures, to conclude on the exceptions I found on the general rule of dating Bulova watches of 1927. I finish with a complete overview of all the case serial numbers that carry the “Bulova-American Standard” signatures with 1927 movements.
Hopefully this forum will start a debate leading to a final conclusion how we are going forward from here on dating 1927 (ladies) watches.
Here's the link to the PDF supplied by Alex. It lays out in detail the points he mentions above.
https://drive.google.com/file/d/0B1mKgMZb4oRQUkt3VS1Yd0IzQUU/view?usp=s…
This will help others follow and participate in the discussion, with graphical examples and tables which would not otherwise display properly on the forum boards.
Edmundo (Alex) I appreciate the time and effort you've put into your research. It is quite remakable.
Is it possible that what we're seeing around 1927 isn't case dating but in fact movement availability? On the other hand could it be attributed to the case manufacturing change or perhaps even something else.
In your research, you find a number of 1928 and 1929 cases (using the 'first digit' dating method) with 1927 movements. Is is possible that Bulova made too many 1927 movements and had to carry them into the following case production years?
We have about twice as many watches in 1928 and 1929 as we do 1926 or 1927. Is it possible that market demand was increasing between these years? I also note that 1/3 of all 1927 watches we have are one model, the Lone Eagle. Is it possible that production or demand of that model caused all three events (excess movements in 1927/case manufacturing change over/increase in sales in 1928-9)?
At this point I'm still not sure I can draw a definite conclusion from the data. There is clearly something happening around case manufacturing AND serial numbering in these years, I'm just unclear what was driving it.
We have previously dissucssed and established the notion that pre 1923 (and in some case up to 1925) the second number in the case serial number coresponded to the year of manufacture.
I don't believe Bulova started at 1000000 and simply continued up. It's my belief that each year they adjusted the second number to correspond with the new year in manufacturing. I still think that there is still some missing information about these early cases and the numbering.
Regarding the 8/9 serial numbers, it's my belief that due to a spike in sales in 1927 (and as such profit), Bulova went all out in manufacturing to keep ahead of the expected demand. This I believe also happenend in 1920/21. The timeframe to order, produce, ship, assemble and test I imagine is not something that could be done in a short timeframe. (weeks, months ???)
The years 1927(later), 1928 (all) and 1929(early), could and may very well be considered as one very busy manufacturing period got Bulova.
The excellent work Alex has done is a great example in showing the complexity and at sometime confussing serial numbering system Bulova used during the 1920s.
I think that due to the shear number of movements and cases being made, and the different plants manufacturing and assembling the parts, lead to a clunky numbering system that we are still struggling to understand.
Whilst I can see some patterns in Alex's work I'm still not 100% convinced that cases starting with an 8 or 9 were manufactured in 1927. They may have been, but I cannot at this stage totally agree with the notion.
What I believe (and this may very well change over time) is that in 1927 Bulova set about manufacturing a great many movement parts in their Swiss plant, shipping them to the US as they became available to be assembled. This process, I believe continued on well into 1928 and as such we see a good number of 1927 movements in 1928 cases.
Why the swap and change of case serial numbers....well that might be due to the different styles of cases being manufactured, each style with a particular batch run of similar numbers....then again there may be another reason.
What we do know is that it was common practice for a watch to have a movement that was dated one year before the case, as it takes more time to manufacture the movements parts, ship, assemble and test than it does to simply manufacture a case locally.
In agreeing with what Geoff said above, the demand in 1927 for Bulova watches spiked and this carried over into 1928 and 1929, before starting to slow in 1930/31.
I don't think that was Alex is saying is incorrect. I think it is certainly noteworthy and has some merit, but for me the first number of the case serial during this period still indicates the year of manufacture of the case.
The quandary still exists however when dating a watch. For example.
Movement manufactured in 1927, cased in 1928, sold to the public in 1929.....
Question: What year is the watch from?
Bottom line, It's the contiuned work like Alex's that will one day hopefully provide us with a solid framework for understanding and dating these early Buylova watches.
I've always been a sucker for the KISS principle.......
My research found two types of case signatures for 1928 and 1929. One that indicates "New York", and one that indicates "American Standard", the latter being the signature as used in 1927, but also 1926.
Focussing on the cases starting with 8 or 9, the ones with the American Standard signature have a specific range of numbers. I give the first 3 digits:
- series starting with 8: 834 to 842 and then 854 to 858. In total this is 140,000 watches of the total 1 million that the full 8 series can respresent given the 7 digits. However, the 8 series only went until 869, so the total quantity of watches starting with an 8 number is 690,000. The difference of 690,000 minus 140,000 gives 550,000 watches that are all signed with "New York".
- series starting with 9: 959 to 963 and then 969 and 970. In total this is 60,000 watches of the total 1 million that the full 9 series can respresent given the 7 digits. However, the 9 series only went until 970, so the total quantity of watches starting with an 9 number is 700,000. The difference of 700,000 minus 60,000 gives 640,000 watches that are all signed with "New York".
So we are talking a very specific and limited number of watches, in total 200,000 of the 1,390,000 watches that are signed with "American Standard". The rest, being 1,190,000 in total have a New York case signature. As said, the American Standard signature was the one used in 1927 (and also 1926). And as it turns out: these cases starting with 8 and 9, but with the American Standard signature have 99% a 1927 movement, while the cases starting with 8 and 9 and the New York case signature have 99% their respective 1928 and 1929 movements.
AS I said in the forum, I still dont understand why Bulova did that, but the evidence is overwhelming: the case signature of "American Standard" is significant in determining the year. Despite the case starting with 8 or 9, they must be from 1927 as confirmed by the movement stamp.
I thought about Geoff's remark if it was possible that Bulova produced too many 1927 movements and put them in 1928 and 1929 cases. Also mybulova_admin is believing that. I did some double checks and my conclusion is: no. The cases within the identified serial number ranges starting with 834 to 842, 854 to 858 and 959 to 970 are genuine 1927 cases, fitted with genuine 1927 movements.
The evidence: let's take two series of watch models I own: the Bernice and Clare. Of the Bernice, I have a 1926 model, signed American Standard and 14kt gold filled case. This is in line with the characteristics of the standard case signature of 1926 and also 1927. I also have Bernices of 1928 and 1929, with serial numbers OUTSIDE the above mentioned case serial number range. Both are signed with New York and made of rolled gold plate. Several models switched to rolled gold plate, not only the Bernice, and this happened in 1928. It is significant to the evidence. I also have a Bernice with case number starting with 963, so WITHIN the range that I identified as being from 1927. If this watch were from 1929, you would expect a case signature indicating "New York", case should be rolled gold plate and it would have a movement of 1929. But is doesn't. It indicates "American Standard",the case is 14kt gold filled and the movement is from 1927. So, all in line with the characteristics of 1927, despite the 963 number. So a true 1927 case.
Same with my Clares. I have Clares from 1928 and 1929. As expected and in line with the characteristics of those years, they are signed with New York, are rolled gold plate, have case serial numbers starting with 800 and 91, so OUTSIDE the range I identified of being from 1927. It all fits. Then I have one with a serial starting with 836, so WITHIN the range I identified as being from 1927. And indeed, it is 14kt gold filled, signed with American Standard and has a 1927 movement, while the "true" 1928 Clare I have, has a (lower) serial number starting with 800, signed New York, is rolled gold plate with a 1928 movement.
As said before, I have no clue why Bulova used this specific range of numbers starting with 8 and 9 in 1927 and look forward to any hypothesis as to why this happened. Also, for what I see, it only happened in 1927.