FORVM`s Classical Numismatics Discussion Board
Numismatic and History Discussion Forums => Ancient Coin Forum => Topic started by: Steve Moulding on November 17, 2021, 07:37:10 pm
-
Hi everyone,
This is Steve Moulding. This is my first post, but I've been a quiet board reader and user of the excellent resources here for quite some time.
I want to let folks know about a couple of online projects I've been working on in my spare time for the past few years. This is not my day job, and very much a long-term labor of love. My website is rnumis.com.
The first project is an online database of historical numismatic auctions. I have entries for over 4000 sales covering 167 houses in 16 countries. There are some summary details for each sale and for many - but especially for those with Ancient coins - I try to provide links to any online versions of the auction catalog that I know of. I also list whether the catalogs (real or scans) are in my library...useful mainly for me, but also helpful if someone is looking for a particular lot description or image - always happy to help, if I can.
As an example, the auctions of Ars Classica - Naville & Cie (1921-1938) can be found here: https://www.rnumis.com/house_auctions.php?house=ARS
The second project has seen much more recent progress, but has been a long term goal of mine for many years. For the Ancient Greek coins of Italy and Sicily I'm building a searchable resource using coins from older sales. The goal is to help uncover - or just to see - provenances. I have just over 3000 lots entered so far...only the beginning but it's already fun to browse these older lots and play around. Filtering is currently by Region, Issuer, Weight, and Metal, but I'll probably add more filters later. This resource can be found at https://www.rnumis.com/greek_coins_top.php.
Well, these are very much works in progress but I hope somebody will find them useful as they currently stand.
-
WOW!
What an amazing stealth project!
I can see a lot of old catalog browsing coming up ... :)
It'd be great if it was possible to filter the auctions by keywords, and availability of an online catalog, as well as by country and date (e.g. search for all auctions with online catalog matching keyword(s) "roman").
Ben
-
Thanks Ben! :)
Great suggestions, shouldn't be too hard to implement. There are some amazing old sales that are now online thanks to Heidelberg, Gallica, NNP and others. I'm still finding them and adding the links. Enjoy browsing!
Steve
-
Hi all, over at my rnumis website I've implemented one of Heliodromus/Ben's suggestions. In the Numismatic Auction Database you can now filter to show only those sales that also have links to catalogs available online, eg at archive.org, Gallica/BNF, or Heidelberg.
I'm actively populating links as I find them and there are still more to be done, but there are already 1300 catalog links. This should be a fast way to get at the catalog for many of the older (and some newer) auctions.
You can find the database at https://www.rnumis.com/auctions_top.php (https://www.rnumis.com/auctions_top.php)
Steve
-
Steve,
Your website works very well for quickly identifying catalogs that are online and getting to their download locations. Good suggestion by Ben to have an 'online only' filter.
Today, I was able to find a Merzbacher catalog I didn't have, which is on Spring's most important Greek auctions list. Nice.
-
Hi Kevin, great! Thanks for the feedback and I'm happy to hear it's useful for you.
The ANS has recently wrapped up an amazing job of digitizing an almost complete run of Glendining catalogs and posting them at archive.org (thank you Lara Jacobs). I've added links to around 170 sales, many with plated ancients. 70 are in Spring's book.
Steve
-
Hi, Steve. Thanks for your introduction. I’ve been referring to your “old” site (http://www.rnumis.com) listing of coin catalogs for a couple of years now. The new link you shared shows something wonderfully different.
Like you, in some ways, I've taken a big plunge into literature, but not so much catalogs as collections (yes, there’s some overlap). But when I need a catalog, I often struggle to locate it. I REALLY look forward to perusing your coin pedigree database, because that’s the large part of my desire to consult those old sale catalogs in the first place.
Those links to online catalog collections, like Gallica BnF, will be helpful. So many were largely unknown to me. Not for the first time, l’m learning about sites that I thought I knew….
Kudos!
-
Hi Chris, thank you - really thank you - for that.
For me, the whole catalog digitization, auction database and online link resources...it's useful, but is actually just several years of prelude to what I really want, which is the provenance database. That part is just getting off the ground. It's early stages, but with a few of the older sales entered (3500 lots or so), there's now enough to show people and get some feedback. Is it useful yet? Maybe, maybe not. With so few sales entered, you'd have to be really lucky to find a missing pedigree from a coin in hand today. But as more auctions get added the chances will get better. The direction is good. And it's still fun to play around.
I do discover missing pedigrees all the time between the coins already in the database. It's quite easy - visually - once the coins are weight-ordered. For example, many coins in Pozzi will show up again in another Naville sale soon after, but the Pozzi provenance is never mentioned. So I'll record those as well and so build up a record of where the coins have been seen...the Provenance Chains which are shown on the website. That's fun too. I think I have almost 80 new cases so far out of these 3500 records.
The bad news is how long it will take to build this database. For a single catalog with a couple hundred Italy/Sicily lots, it's at least a week of spare time work to do the image extraction from scans plus all the data entry. At that rate it's going to be 2-3 years to get up to the likely 25,000 or so records I think will constitute a reasonable Italy/Sicily database.
Gallica BnF, Heidelberg and archive are incredibly useful. I can't say it enough. I keep looking for a similar resource that covers the Italian houses but so far little luck (gallica has a few Rattos). Perhaps you know of one?
Published collections I do buy and digitize as well (de Nanteuil, Locker Lampson, Lucien Hirsch etc), but mostly auction catalogs. Jameson I don't have, but it's online so I can refer to that for now. I've been recently getting into and buying the SNGs which I'd not paid much attention to, until recently. They are amazing sources of provenance information.
Well, thank you again Chris. Let's keep talking, and if you have feedback - good or bad - please let me know. I want those resources to get better.
Cheers,
Steve
-
Hi Chris/Anaximander - so for fun I actually started checking some of your amazing gallery coins against the database over at rnumis.
I have to say your Lucania Metapontion Nomos looks an awful lot like Ars Classica V, Duplicates of the British Museum (1923), lot 469.
For example, look at the defect right of MA. Weights are very close too: 7.88g (Ars Classica V) vs 7.89g (yours). I saw the CNG 45 lot description doesn't mention any provenance.
https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469 (https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469)
What do you think?
Steve
ps I took the liberty of showing your coin here for comparison, hope that's ok.
-
Looks to me like you just added a great 99-year provenance to this coin. Nice!
-
What a valuable reference! I'm certain I'll be a regular searcher.
Laney
-
Great resource, is it a bridge too far to ultimately index and make the primary coin reference numbers searchable? I'm thinking probably yes.
-
What a valuable reference! I'm certain I'll be a regular searcher.
Laney
Thank you Laney! I hope it's useful and let me know anytime what would make it better.
Cheers,
Steve
-
Great resource, is it a bridge too far to ultimately index and make the primary coin reference numbers searchable? I'm thinking probably yes.
Hi Ron and thank you. I definitely want to get better indexing and I think we're on the same page re: the primary references. I did start on that path at the beginning but found it hugely time consuming and postponed it in favor of just getting basic lot information and images into the database (which still takes 95% of my time).
So, certainly not a bridge too far and it's a great suggestion. It will happen, just not sure yet how or when. One possibility is the project becomes more collaborative and interested people can reference tag by themselves, but we're not there yet.
Steve
-
Looks to me like you just added a great 99-year provenance to this coin. Nice!
Thank you, Kevin. Personally, I don't have any doubts it's the same coin and was happy find the possible connection. Waiting for Chris/Anaximander's feedback. I was actually even happier that it turned up so easily in the search, even with so few of the older sales added so far.
Cheers,
Steve
-
Steve - what was the search process you followed to notice that the Ars Classica 1923 auction coin is the same as the gallery coin? If you have a good method for working this out, I'm sure many of us would try and repeat it for coins in our collections.
-
Hi Ron - I typically follow the same basic procedure. In this case
- Region=LUCANIA
- Issuer=Metapontion
- Weight Lo=7.87
- Weight Hi=7.91
That returns 42 results, so I also set 48 results-per-page (not necessary, but just easier). Now it's just a case of making a visual match.
If something looks promising, click the coin card for the details where you can usually get links to the Heidi/gallica/archive plate page for larger images (description page too). This is important as the resolution on the images I show on the website is nowhere near high enough. I have my own hi-res scans of course (from which the website images are derived) but those images aren't hosted and anyway I find that seeing the whole description and plate pages via Heidi is great for context.
The weight window I used here is Chris Coin +/- 0.02g. I've seen different sales of the same coin sometimes differ on weights by 0.05g or more, so widening the window will increase the hit chance, but there will be a lot more results to wade through. Here we were lucky as the Ars Classica coin was only 0.01g different so was seen immediately.
Steve
-
Thanks Steve, now you need to expand it to the entire hellenistic period, and then start of Roman Imperial :)
-
Now we're talking, Ron :) I'd love that, but will either need to clone myself (unlikely) or get other people involved. Italy/Sicily is a good and (almost) manageable-size test of concept.
-
I have to say your Lucania Metapontion Nomos looks an awful lot like Ars Classica V, Duplicates of the British Museum (1923), lot 469.
For example, look at the defect right of MA. Weights are very close too: 7.88g (Ars Classica V) vs 7.89g (yours). I saw the CNG 45 lot description doesn't mention any provenance.
https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469 (https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469)
Brilliant! Yes, this is that coin. I now have, as mentioned by Kevin D in this thread, a pedigree for this coin. This is a "data-mining" triumph of the rNumis database. How exciting!
As for the visual comparison of the catalog and my Forum coin, I do believe that there must be some use of plaster casts. There are two 'bubbles' towards the top of the Naville catalog photograph (obverse top left, reverse top right). As you'll know, plaster casts were commonly used. The SNG Copenhagen relied heavily on them. I don't think that takes away anything from the attribution.
-
Hi Ron - I typically follow the same basic procedure. In this case
- Region=LUCANIA
- Issuer=Metapontion
- Weight Lo=7.87
- Weight Hi=7.91
That returns 42 results, so I also set 48 results-per-page (not necessary, but just easier). Now it's just a case of making a visual match.
If something looks promising, click the coin card for the details where you can usually get links to the Heidi/gallica/archive plate page for larger images (description page too). This is important as the resolution on the images I show on the website is nowhere near high enough. I have my own hi-res scans of course (from which the website images are derived) but those images aren't hosted and anyway I find the seeing the whole description and plate pages via Heidi is great for context.
The weight window I used here is Chris Coin +/- 0.02g. I've seen different sales of the same coin sometimes differ on weights by 0.05g or more, so widening the window will increase the hit chance, but there will be a lot more results to wade through. Here we were lucky as the Ars Classica coin was only 0.01g different so was seen immediately.
Steve
I just tested it with a coin known to be in the Naville sale of the Bement Collection in 1924 (Naville VI Lot 98). The coin sold a couple times in the last 15 years, but the provenance had been lost over the decades (I found it by searching catalogs). I started my search on Steve's website with the weight listed in the coin's most recent auction appearance (no range, just the exact weight of 6.58 grams). It didn't come up (not surprising). Then I ran a search with the upside weight increased by 1/100 gram. It didn't come up. On the third search I decreased the downside weight 1/100 gram and it brought up the coin. All of this took about 10 seconds. I am very impressed. Below is an image of the coin I searched. Comparing CNG 76 Lot 3003 with CNG E341 Lot 88 will show the change in cited provenance.
-
I have to say your Lucania Metapontion Nomos looks an awful lot like Ars Classica V, Duplicates of the British Museum (1923), lot 469.
For example, look at the defect right of MA. Weights are very close too: 7.88g (Ars Classica V) vs 7.89g (yours). I saw the CNG 45 lot description doesn't mention any provenance.
https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469 (https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19230618_469)
Brilliant! Yes, this is that coin...I do believe that there must be some use of plaster casts...I don't think that takes away anything from the attribution.
Yes, many early catalog coin images were produced from casts, which sometimes adds to the challenge of identification, but knowing this helps. Congratulations on your good-looking coin, now with an attractive provenance.
-
Interesting. I assume to make them easier to photograph with the lighting available at that time?
-
I don't know what the answer is, but your suggestion of lighting seems it could have been a factor.
I note that many Glendining catalogs do not list the weight of the coins, which means other parameters would be needed for a database search.
The link below is to an interesting article on how one company is using their old catalog images to conduct searches.
https://culturalpropertynews.org/rediscovering-old-provenances-for-ancient-coins/
-
How did I miss this thread? Fantastic resource. Thanks for doing this.
Needless to say, I'll be going through my coins just in case.
-
Thanks Jay! Really hope you find something!
Cheers,
Steve
-
I just tested it with a coin known to be in the Naville sale of the Bement Collection in 1924 (Naville VI Lot 98).
Hi Kevin, thank you. That was a great test! Just to show that coin in a chain I added the later sales that you found, so now we see all 4 appearances together as shown below (taken from the website https://www.rnumis.com/greek_coins_detail.php?dbid=CNG_20141217_88 (https://www.rnumis.com/greek_coins_detail.php?dbid=CNG_20141217_88))
Steve
-
How's this for a coin match to the rNumis provenance database:
Sicily, Selinus. 7.74 g., your Ars Classica XVI (https://www.rnumis.com/imgdbp/37eb897f8cf4618df31c35e995753c13.jpg) Lot 642. vs. my coin (https://www.forumancientcoins.com/gallery/displayimage.php?pid=157643) 7.86 gm, purchased from a dealer in Miami (yes, I get some mileage from having had a dealer in Miami ;)).
I'll put down some of the weight difference to my being at sea level, or even below sea level, when taking the measure.
-
Hi Chris. It looks very good but that 0.12g difference is larger than I'd like to see. Not that it can't happen, of course, but I wish they were closer. I'm looking for a different sale that could confirm but as I don't have those in the database yet, it's the same time-consuming manual hunt.
Cheers,
Steve
-
Regarding the reason early catalog coin images were produced from casts, I found this in my notes:
https://www.forumancientcoins.com/board/index.php?action=printpage;topic=102458.0
Title: Re: Old Auction Catalogue Madness!
Post by: Andrew McCabe on May 12, 2016, 05:12:48 am
Quote from: Molinari on May 12, 2016, 04:15:14 am
Why did they make plaster casts back then instead of just photographing the coins?
Good question.
It provides far better photos (the bad exception above is NOT typical), labour costs were low a century ago, and in advance of photoshop and digital cameras, the chances of getting a direct photo even remotely good were minimal - imagine photographing your coins today in the absence of preview screen - whereas plaster cast photography was a known reliable technique. Also, photographic printing costs were then very high indeed (using a lot of silver) so if you were going to spend a lot of money photographing just a few coins - they always selected just the best coins rather than photographing everything - and the cost was anyway going to be high, you might as well spend the extra to make plaster casts and make the photos as good as possible. Furthermore, to make up the plates needed a second photograph - of the photos sitting on the background of the plate with numbers under (again, in advance of photoshop) using a copy stand. Evidently if you start with prints and then have to photograph those, they'll be a whole lot worse than starting with a sharp plaster cast. To see what "just photographing the coins" resulted in, you need only look at the Rheinhold Faelten 1938 Stacks sale - possibly the worst ever plates - and just about any catalogue from the 1970s which was after printing became cheap and labour costs for plaster casts became dear, but before photoshop allowed digital improvements; the 1970s represents the nadir of coin photography.
-
I just tested it with a coin known to be in the Naville sale of the Bement Collection in 1924 (Naville VI Lot 98).
Hi Kevin, thank you. That was a great test! Just to show that coin in a chain I added the later sales that you found, so now we see all 4 appearances together as shown below (taken from the website https://www.rnumis.com/greek_coins_detail.php?dbid=CNG_20141217_88 (https://www.rnumis.com/greek_coins_detail.php?dbid=CNG_20141217_88))
Steve
Steve,
The Provenance Chain is a great feature, with all the info for every auction appearance just a click away.
-
How's this for a coin match to the rNumis provenance database:
Sicily, Selinus. 7.74 g., your Ars Classica XVI (https://www.rnumis.com/imgdbp/37eb897f8cf4618df31c35e995753c13.jpg) Lot 642. vs. my coin (https://www.forumancientcoins.com/gallery/displayimage.php?pid=157643) 7.86 gm, purchased from a dealer in Miami (yes, I get some mileage from having had a dealer in Miami ;)).
I'll put down some of the weight difference to my being at sea level, or even below sea level, when taking the measure.
As Steve has said, usually the weights are closer. This could be just a typo error for the middle weight digit, in one listing or the other. Or, one of the scales could have been a tenth gram off.
-
I weighed the Selinus tetradrachm anew and got the same reading: 7.86 gm (or 7.87 gm, depending on how well centered the coin is on the scale).
Given that coin weights on my scale align with weights provided by coin dealers, my scale is reasonably accurate (picture attached).
If my coin is the same as the one shown in the Ars Classical catalog of 1933, then the catalogers must have had an inaccurate scale (unlikely), some human error (most likely), or the difference in altitude (Switzerland vs. sea level) had a measurable impact.
Interesting to hear about the use of plaster casts in early catalogs. It's all about the costs and difficulties of coin photography. I was betting that getting both sides of a coin into one photoshoot was the cincher, back in the day.
-
I note that many Glendining catalogs do not list the weight of the coins, which means other parameters would be needed for a database search.
I ran another test, looking for the same coin that I ran a test on previously. This time I left the weight unspecified, just entering Calabria and Tarentum. This brought up 10 pages of coins, 48 results per page. I found the coin in 3 minutes of looking, on page 5.
Even if the database were to be expanded many times over, known coin weight or not, this would still be a GREAT improvement on what I did to find this provenance originally.
Congratulations Dr. Moulding, on creating this successful and wonderful provenance research program.
-
If my coin is the same as the one shown in the Ars Classical catalog of 1933, then the catalogers must have had an inaccurate scale (unlikely), some human error (most likely), or the difference in altitude (Switzerland vs. sea level) had a measurable impact.
Hi all, I'd agree with Chris that human error is the most likely of those three explanations. I do see occasional mistakes Ars Classica catalogs. I think the weight drop due to the altitude of Lucerne (not very high) would be smaller than what we see. Also, I believe their scales were probably ok. Looking at other lots, we see for example
Ars Classica XVI, Jul 1933 Lot 153 7.24g = Sotheby, May 1916 (Headlam) Lot 230 7.23g = Jacob Hirsch XXVI, May 1910 Lot 227 7.23g.
All within 0.01g of each other.
(https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19330703_153 (https://www.rnumis.com/greek_coins_detail.php?dbid=ARS_19330703_153))
Steve
p.s. Very nice to see that coin as it is today on your scale!
-
I note that many Glendining catalogs do not list the weight of the coins, which means other parameters would be needed for a database search.
Thanks, Kevin! I'm glad you're finding the website useful. Yes, you're absolutely right...some Glendining catalogs didn't record the lot weights. I don't know why. Thankfully, many do though. So, all that makes things a little more fun (=challenging), but not impossible.
Along the same lines, I sometimes see sales from other houses where they didn't record the weights of some the bronze coins, even when they've been plated at some expense. Curious.
Steve
-
For anyone interested in the Provenance Explorer over at rnumis I've made a significant upgrade to the code, adding the ability to do Multi-Level filtering beyond the initial Region/Issuer/Weight/Metal.
This makes it much easier to spot lost provenances (I found 3 already today).
Also, searches can now be bookmarked.
So far, I've only tagged Messana lots at Level 1 (183 of them). It's going to be a very lengthy process to go through the 8000 or so current lots and tag everything, but at least you can now see how it works:
https://www.rnumis.com/greek_coins_top.php?gk_region=SICILY&gk_issuer=Messana&metal=All%20Metals&mgkdb_minwt=0&mgkdb_maxwt=9999&FL=1&FN=1 (https://www.rnumis.com/greek_coins_top.php?gk_region=SICILY&gk_issuer=Messana&metal=All%20Metals&mgkdb_minwt=0&mgkdb_maxwt=9999&FL=1&FN=1)
Thanks for the continued feedback and especially for the help that some of you have very graciously provided.
Cheers,
Steve
edit: The 303 lots from Velia (Lucania) are now also tagged
https://www.rnumis.com/greek_coins_top.php?gk_region=LUCANIA&gk_issuer=Velia%20%28Hyele%29&metal=All%20Metals&mgkdb_minwt=0&mgkdb_maxwt=9999&FL=1&FN=1