praxim Posted April 2, 2017 Share Posted April 2, 2017 (edited) First my request, then some background. I am asking you please to post, for each vintage Aurora 88 / 88K / 88P that you own,- the model and the serial number- if it is an unserialised 88P, please state that.- if you have documentary evidence of it, the year of manufacture / sale (not a seller's opinion please). BackgroundI have collected quite a few models and serials from FPN, fleabay and my own few pens, and have compared this with the information in the Aurora 88 Dynasty and other threads. Now, I am seeking more data to improve upon it. I have concluded from looking at existing pens that the serial number offers a useable source for dating the pens. It appears that they were applied as would be expected to each successive pen, regardless of model. The lowest serial I have found to date is under 47000, so we know that they did not kick off the process somewhere in the hundred thousands. From the Dynasty thread, one would expect to see 88 and 88K serials intermingled for the period 1951 to 1953, and this happens. One would also expect from that source to see no overlap between the 88K and 88P and so far this too is the case. However, based on total production information in the Dynasty, taking that as cumulative, one would expect a serial in the 2.5 M range to appear only on an 88K, but it appears on an 88 which ceased production in 1953. That implies that either fewer 88P were later produced to stay within 3.8 M total, or else more than 3.8 M were later produced if the inferred 1.3 M 88P were made. An alternative that practically no 88K were produced during 1954-57 seems unlikely! However, no 88P so far has a serial as high as 3.3 M, far from 3.8 M. Casual reading also suggests there are more unserialised 88P around than the small number expected (I have one myself). This may account for the gap, if serialising ceased around 1961. For these reasons I am asking for the above data to try to fill in some gaps. I am already close to having a fairly good translation from serial to year, to counter optimistic sellers and simply for one's own entertainment. To do this I am obliged to make an assumption about non-linear production patterns in which production for each model ramps up quickly in year 0-1, plateaus around 2-4 years and declines in years 5-7. These mild curves are assumed but appear to give better results on known data than assuming flat production rates. edit:minor and formatting Edited April 2, 2017 by praxim X Link to comment Share on other sites More sharing options...
ink-syringe Posted April 3, 2017 Share Posted April 3, 2017 unserialised 88P in the house. Looking for a cap for a Sheaffer Touchdown Sentinel Deluxe Fat version Link to comment Share on other sites More sharing options...
jar Posted April 3, 2017 Share Posted April 3, 2017 (edited) 888 206991388P 3418050 There's an 88 and another 888 around somewhere but right now I don't know where they are hiding. Those serial numbers are secret known only to me and the manufacturer so make sure you don't tell anyone. Edited April 3, 2017 by jar My Website Link to comment Share on other sites More sharing options...
Vintagepens Posted April 3, 2017 Share Posted April 3, 2017 Such a project is long overdue! I asked Giovanni Abrate a while back if anyone in Italy had compiled such a list; he didn't know of anyone, but thought it a worthwhile task. Whether Aurora retains the info in its archives is another question that needs to be pursued. Here are some numbers for 88s:766562289733487204407675245825648336685826720807850728606828833041273437127894313145171450530164400918858021916010 (GL) 88K:203829920519652067505217239825375792556231 (GL)guarantee for 1917686 88P:30084463127365314077531886413288165 David Nishimura www.vintagepens.comvintagepensblog.blogspot.com Link to comment Share on other sites More sharing options...
stephanos Posted April 3, 2017 Share Posted April 3, 2017 Excellent project. Here is what I can contribute: 1.Aurora 88 (gold-plated cap, bleached ebonite section and turning-knob). 88889The piston seal on this looks not so much like a cork as a series of felt rings tightly packed together 2.Aurora 88 with silver-coloured (Nikargenta?) cap, marked: 836914 3.Also, though you didn't ask about Duocart models:88 Duocart with metal cap, marked: AV.0996 No idea about the year of manufacture or sale for any of them, I'm afraid. Link to comment Share on other sites More sharing options...
praxim Posted April 3, 2017 Author Share Posted April 3, 2017 Thank you. I will add these to my ex-ebay list and see whether I can confirm or reject any notions. For the record, I have:88 1039789, 1677535 and 1847540 (one of those is not yet in my hands)88P 3210650 and unserialised. I did not included 888 and Duocart because I was less clear about their start and end dates for production, but happy to add them as well, especially if someone has that production data? Would their production count be within or additional to the Dynasty thread data for 88s? I am assuming it would be additional. Thanks to David's significant list, I should be able give some results later today. X Link to comment Share on other sites More sharing options...
praxim Posted April 3, 2017 Author Share Posted April 3, 2017 Here is a preliminary puzzle. If you look at this ended ebay sale I read in the last two photographs that the pen is labelled 88 and that the serial is a seven digit number starting with 25, i.e. in the 2.5 M range. This is the big anomaly. Based on what I have so far (I welcome more) then, barring this particular pen, all 88 serials are less than 2M, all 88P serials are greater than 3M, and there are no overlaps. The first 88K so far has the serial 1917686 whereas the last 88 is 1916010. We know that the 88 and 88P caps are not interchangeable, but those of the 88 and 88K seem to have been. However, that anomalous pen is marked on the section both front and back, not on different parts of a pen which might have been put together. The pen itself is quite corroded in both cap and ring. Reading the Dynasty thread again, I see that the 88K had a black insert in the cap, whereas the 88 (and early 88K) did not. The ebay pen has an 88K cap and to all appearances an 88K serial number, a late one at that. Any explanations anyone wishes to offer? If this pen is a real 88 then it was produced in 1957, four years after production of the 88 ceased. I may simply ignore it as an outlier unless there is confirmation through another anomalous 88. Did anyone on FPN buy it? X Link to comment Share on other sites More sharing options...
praxim Posted April 4, 2017 Author Share Posted April 4, 2017 (edited) Reflecting on the above anomaly, I conclude from the serial number and cap that the pen is an 88K made in 1956 or 1957. I account for the '88' on the pen (rather than 88K) by assuming worker error. Looking at the engravings on a few pens, they are slightly different in the way 'Aurora' or '88' or other digits are written. The 1950s is a bit early for computerised NC machine tools ( ) so engraving would be using hand tools or at best a hand-guided engraving head. My conjecture is that someone wrote 88 instead of 88K, and that will do for me unless unusual new data comes to light. edit: I notice on my two 88P that one says "Aurora" below 88P where the other simply says 88P. The latter is the serialised one. With that conjecture incorporated, here is what I know so far: It was suggested in the Dynasty post that production of the 88 and 88K may have overlapped. I am convinced this did not happen. diplomat based their view on a presumption that not much over a million 88s were made, yet they knew of 88 serials as high as 1.7M. Well, they hit 1.9M and no 88K has a lower serial than the highest 88. I conclude that there were more 88s produced than was presumed. I have 33 serials for 88s, ranging from 46498 to 1916010 in a fairly uniform distribution. Assuming (as I do) normal sequential numbering, this gives a production run between 1.87M and 1.91M, a bit higher than the 1.5M estimated in Dynasty and about 240,000 annually. If I decide to set production at half that in the first year, 1946, and the last when the 88K was introduced,1953, then I get the following basic table giving for each year the last expected serial for that year. A serial close to that number may be from that or the next year.CAVEAT: The iguanasell history suggests first production in 1947 while this article says it commenced in November 1948. If the last were true then they were producing far more pens each year than any of the articles suggests. Pentrace also suggests overlap of the 88 and 88K, which is not supported by serial numbers. I am writing to Aurora to ask. Meanwhile...1946 1197501947 3993751948 6789381949 9585991950 12380631951 15176251952 17971881953 1917000 My assumption of a flat production rate with half as many in the first and last years is just that, a working assumption (I abandoned my hand drawn curve of production as having no greater merit). However, I doubt that numbers would shift a lot. Only when someone can turn up a few pens with fully verifiable production dates will I be able to improve on this. On the above basis, my own pens are 1950, 1952 and 1953. I have less than half as much data on the 88K and 88P serials but will offer what I have in a further post, later. More data welcome! Edited April 4, 2017 by praxim X Link to comment Share on other sites More sharing options...
praxim Posted April 6, 2017 Author Share Posted April 6, 2017 I am waiting on a reply from Italy. It may not eventuate so I will continue with what I have. Firstly, I am revising the above list to start production in 1947. It certainly does not seem to have started in 1946 despite the claims of some sellers, and I am a little unwilling to go to 1948 without more certain information. Other than the first production year, these changes make surprisingly little difference to dating.1947 137,1871948 466,4371949 795,6861950 1,124,9351951 1,454,1851952 1,783,4341953 1,920,621 To use the list, travel the list until the last number which is greater than your serial number. That is the year. If you are within a few tens of thousands of the given number then it could be either side of the that number. For example, a pen marked 1450530 is nominally 1951 but might easily be 1952 whereas 799238 is nominally 1950 but might be 1949. I will update this if I receive better information from Aurora, else this will do for the original Aurora 88. X Link to comment Share on other sites More sharing options...
praxim Posted April 6, 2017 Author Share Posted April 6, 2017 (edited) 88K and 88P production appears, on the serials available, to have been about a third (each) of that for the original 88. This is quite different from the "Dynasty" information. However, it is pretty consistent with a count of pens available on ebay on sampling over a few days, despite the fact one might expect newer pens to be more likely to survive to today and thus be over-represented. I am unable to glean anything about 888 or duocarts. There is also the confounding factor that we do not know when Aurora ceased serialising the 88P. I have a dozen serials for 88K pens and 14 for 88P, with three more known to exist unserialised. Given the overlap with the 88, I have put half of the first year of 88K production into 1953 where I also have half of a year of 88 production.1953 1,986,5021954 2,134,7341955 2,282,9661956 2,431,1991957 2,579,431 For the 88P, I have no overlap and have gone to 1962. I believe there may have been some overlap with the 98 in 1963 but have eliminated that to account somewhat arbitrarily for unserialised pens.1958 3,084,0001959 3,168,0001960 3,252,0001961 3,336,0001962 3,420,000 I hope these two posts are helpful in dealing with wild ebay claims and in simply being able to say something sensible when asked the age of your Aurora 88 variant. edit: Note that it appears that 88P models started at 3,000,000, not at a number sequential with the last 88K. This will stand until such time as someone can show an 88P with a serial under that. Edited April 6, 2017 by praxim X Link to comment Share on other sites More sharing options...
praxim Posted April 8, 2017 Author Share Posted April 8, 2017 (edited) A point arose that I think deserves some explanation here. diplomat's figures suggest about 1.5M 88, 1M 88K and 1.3M 88P. My figures suggest over 1.9M 88, under 700,000 88K and about 550,000 88P; total < 3.2 M. Do I think my numbers more likely and if so, what are my justifications? The 88 is simple. Unless one presumes that serials were assigned randomly rather than in sequence (and the separation of pen serials between models rebuts that) then it is clear that there were over 1.9M of the original 88, simply because that is the range of serial numbers, and known 88K serials start very shortly after the last available 88 serial. The 88P is relatively simple in that I can not be clear about it, because we do not know when they ceased to serialise them, a cost-saving measure (it was a manual task). We know it was at least 420,000 and probably not a lot more, which the 88K discussion below will help to explain. Either way, there are more serialised than unserialised extant so the odds are that you could top the 88P out at five or at most six hundred thousand, not 1.3 million. I suspect that the 88P ballpoint may be counted in the higher figure from diplomat. I have an 88P set myself. Let us say that you asked, "Why should there not be 88K serials up to or close to 3 million, but in only 12 examples you just have not seen anything above 2.55 million yet?" The answer involves working out this: if there were truly one million rather than 660,000 pens, how likely is it that one could choose 12 pens randomly (random selection being a reasonable assumption here) and see not one above a serial of 2.56 M? The answer to that is about half a percent, 0.5%. Clearly, if there were that many pens we ought to have seen one by now, 995 times out of a thousand. Let us try some other figures. Based on the range of serials, the chances of there being 700,000 rather than 660,000 pens is about fifty-fifty. That there might be 800,000 is about 10%, 900,000 only 2% and a million the aforementioned 0.5%. Note that the only serial I have for an 888 falls within the 88K range of serials, so not all of the 660,000 will be exactly the 88K anyway. Barring new information we can be quite confident that the figures here better represent actual production of those three models than is proposed in the "88 Dynasty" post, where I believe ballpoints may be in the mix in later years. edit: Modified numbers in first paragraph to the even chance level of maximum production, beyond the highest serial seen. For 88P I took 20% unserialised, not a reliable estimate but the best I can do on scant data for it. Edited April 9, 2017 by praxim X Link to comment Share on other sites More sharing options...
Feanaaro Posted April 9, 2017 Share Posted April 9, 2017 I have a Nikargenta 88 (real Nikargenta as far as I can tell) with the number 928632and a 88P, chrome cap, with no visible serial number, as far as I can see. Link to comment Share on other sites More sharing options...
praxim Posted April 9, 2017 Author Share Posted April 9, 2017 Thank you Feanaaro. The 88P without serial makes me a little more confident that these are less rare than first thought. This may lead me to rescale the 88P production years slightly. by stopping serialisation a bit earlier, perhaps a year. Your 88 looks to be 1950. That does not change the scales but every bit of data consistent with the hypothetical scales is good. X Link to comment Share on other sites More sharing options...
Cryptos Posted April 10, 2017 Share Posted April 10, 2017 I have an 88P without a serial number (could've sworn I'd sold it, but apparently not ) That one has a gold cap. And I have another inbound from Italy that is also an 88P but with a chrome cap - no idea if it has a serial number, let you know when I know. Link to comment Share on other sites More sharing options...
praxim Posted April 10, 2017 Author Share Posted April 10, 2017 Thank E_o_C. I had already noted your unserialised 88P, along with another mentioned by someone on a thread here years ago. I will be interested to see whether your new pen breaks any boundaries in the arcane world of serialisation. X Link to comment Share on other sites More sharing options...
catbert Posted April 10, 2017 Share Posted April 10, 2017 I have an 88, serial number 1317400.It came with the original packaging and papers for a completely different 88, serial number 573307. Link to comment Share on other sites More sharing options...
mallymal1 Posted April 15, 2017 Share Posted April 15, 2017 (edited) 88 wiith Nikargenta cap 585 39588 with gold cap 1 260 52888K 2 137 700 Edited April 15, 2017 by MalcolmH Link to comment Share on other sites More sharing options...
mallymal1 Posted April 15, 2017 Share Posted April 15, 2017 Here is a preliminary puzzle. If you look at this ended ebay sale I read in the last two photographs that the pen is labelled 88 and that the serial is a seven digit number starting with 25, i.e. in the 2.5 M range. This is the big anomaly. Based on what I have so far (I welcome more) then, barring this particular pen, all 88 serials are less than 2M, all 88P serials are greater than 3M, and there are no overlaps. The first 88K so far has the serial 1917686 whereas the last 88 is 1916010. We know that the 88 and 88P caps are not interchangeable, but those of the 88 and 88K seem to have been. However, that anomalous pen is marked on the section both front and back, not on different parts of a pen which might have been put together. The pen itself is quite corroded in both cap and ring. Reading the Dynasty thread again, I see that the 88K had a black insert in the cap, whereas the 88 (and early 88K) did not. The ebay pen has an 88K cap and to all appearances an 88K serial number, a late one at that. Any explanations anyone wishes to offer? If this pen is a real 88 then it was produced in 1957, four years after production of the 88 ceased. I may simply ignore it as an outlier unless there is confirmation through another anomalous 88. Did anyone on FPN buy it? Is this pen back up for sale? Link to comment Share on other sites More sharing options...
praxim Posted April 15, 2017 Author Share Posted April 15, 2017 Yes, that is the one. I am thinking what discreet question I should ask to try to find out more. I also gave passing thought to how much I would be willing to pay simply to get a closer look. Not much; that pen is in poor shape. X Link to comment Share on other sites More sharing options...
praxim Posted April 15, 2017 Author Share Posted April 15, 2017 Now that I have a sample of 38 rather than a dozen, I have recalculated the likelihood that a serial between the highest numbered other 88 and the serial of that mysterious "88" would not be seen in this sample. Accurate to at least four decimal places, it is zero, or < 1 in 20000. Possible of course, but something seems very odd. The number dates to 1957, years after 88 production was superseded by the 88K and nearly at the introduction of the 88P. X Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now