The 1923 Currency Issue (Business Strike) SA Gold Sovereign - A Numismatic Enigma

[I N Collectables 20]

Good Day Everyone,

Please forgive me if this topic has been previously discussed however I am quite intrigued by what I feel is a somewhat undiscovered/unknown rarity of SA numismatics.

As the title of this thread states, I refer to the 1923 Currency Issue Gold 1 Sovereign.

Whilst the proof version of this year is a well documented and popular issue amongst collectors and dealers alike, the currency issue seems somewhat "swept under the rug".

What the literature says:

Below I have quoted some internet literature that I could find on this coin: 

"There are two key dates to the South African sovereign series, and while the 1924-SA is widely-collected and highly-valued, the 1923-SA currency issue could almost be described as obscure. Overshadowed by the proof version of the date, which is scarce in its own right, Pretoria's currency-issue sovereign of 1923 is a somewhat recent find that was not known to collectors until its discovery and authentication in the 1970s. The exact mintage is not known, but the estimates range from a low of 64 pieces to a high 406. Whatever the mintage, the number of coins trading the market today is small, with perhaps only a dozen sighted in the last couple of years. Despite what is says in the price guides, the 1923-SA currency issue sovereign is rarer than the 1924-SA sovereign and much rarer than the 1923-SA Proof Sovereign."

Description of the only example to have sold on Heritage Auctions:

"George V gold Sovereign 1923-SA, KM-21, Spink-4004, Marsh-287 (mintage 406 pieces), Hern-S338, MS66 Prooflike PCGS; in fact, this coin is not prooflike at all, but satiny in texture with bright luster, the strike not at all that of a proof, sharp but not minutely detailed on the highest points, some small abrasions also attesting to the intended commercial nature of this piece, which glows with brilliance and shows reddish tints to the fine gold. On the rims, the small denticles of the obverse rim are sharp, and the tiny incuse blocks forming the reverse rim are finely detailed. As well, there is a knifelike quality to each rim where it joins the edge, which is sharply reeded. Clearly, this is a special coin, not a proof, carefully made and lovingly saved by somebody. In recent years, the Terner Collection contained another gem, graded MS65 by PCGS. The cataloguer purchased that coin in January of 1994 on behalf of Dr Terner. It and several others  (all seen by the present cataloguer at the end of 1993) showed up in London not long before, coming from the estate of Raoul Robellaz Kahan, whose career took him to the mints at Perth and Bombay during the period of World War One, and a few years later to the Rand Refinery in South Africa, which would supply gold specie to the mint at Pretoria beginning at the end of the year 1921. The mint came into existence officially on December 14, 1922, by proclamation of King George V, but only opened for operation at an unknown date during 1923. Gold came tumbling into the refinery a little more than a year later, but during 1923 it was in scant supply. Citing a report written by the Deputy Master of the Mint in 1924, Marsh pointedly states (see page 89 of his book The Gold Sovereign) that "During the period covered by this report only one deposit of gold was received at the mint. This consisted of old jewellery, which after the Rand Refinery's charge for refining, and the Mint charges, was found to be of the net value of £406.0s.9d. The resulting sovereigns were delivered to the depositor." This evidence suggests, but does not prove, that 406 currency sovereigns were minted at Pretoria during 1923; the mint master mentioned only "resulting sovereigns," which is suggestive but not definitive. Nor did he identify the "depositor." What is definitive is the survival of those coins. Evidently most of the mintage perished. Marsh further reported that his correspondence with the South African Mint in 1979-1980  confirmed the statement from that report of 1924. Information dating from 1980 to 1985 also confirmed the existence of a currency specimen of the 1923-SA sovereign held in the Mint Collection at Pretoria. It is one of just a few pieces known to exist. The issue is rated Extremely Rare. As concerns the presently offered example, only a few individuals throughout history have found themselves in such a position as did Kahan, clever enough to obtain for himself a few samples of the brand-new gold sovereigns bearing the mint's initials "SA" embedded in the ground line just above the date upon which the mounted knight is seen slaying the mythical dragon. This specimen was obtained from the Bentley Collection, Part 2, lot 920, sold last September in London, where its source was cited, dating to exactly the same time during which the present cataloguer obtained Dr Terner's specimen. The provenance is one of the clearest possible and most direct links to any great rarity, and is beyond dispute. This beautiful, gleaming gold coin was surely among the first sovereigns struck at Pretoria in 1923, was saved by Kahan at issue, and is now certified as being at the top of the population of only a handful of known examples."

Auction Appearances:

The only record of one having sold at public auction that I could dig up was the Heritage Auctions offering as referred to above. The particular coin was graded as MS66 by the PCGS and realized a whopping $49,937.50 (approx R736,000 at today's spot rate) on 07 January 2013.

The Grading stats:

NGC: 12 graded in total - 1 in 45, 1 in 50, 6 in 55, 1 in 61, 1 in 63, 1 in 64, 1 in 66

PCGS: 9 graded in total - 1 in 50, 2 in 55, 1 in 58, 1 in 62, 1 in 63, 1 in 65, 1 in 66

 

I approached a well-known numismatist about the currency issue gold sovereign and its potential value, his response was "no business strikes minted...Hern gives prices for EF etc but these are for proofs that went into circulation...so I guess look at Hern's book and see what he stipulates for EF and work from there"

I have had the honour of inspecting 1 of these rarities graded AU and the differences from the proof sovereign are clear. There is no question that the currency issues are ex/impaired proofs.

 

Any input would be very welcome...

Regards,

Mohammed.

 

 

 

 

 

[jwither 17]

I don't believe this coin is overlooked.  It just isn't available for sale like many other Union coins, so no one can buy it.

I recall the Heritage sale.  My recollection is Goldbergs sold the same coin  as part of the Millenia collection maybe ten years ago or so.  I believe it sold for $23,000 (or near it) which I suspect is more reflective of it's current value than almost $50K USD.  DNW also offered one in the Bentley Sovereign collection and to my knowledge, it remains ungraded.

I still believe there are noticeably more out there proportionately in other Sovereign collections, but the population count hasn't budged in a long time. If 12 is reflective of the actual supply, this would lead me to believe that the mintage of 406 might be overstated.  The accepted mintage today (in Krause and my prior Hern guides) is 64.  However, DNW also stated 406 in the Bentley sale.

As for the 1923 proof version and the 1924, I don't consider either to be really that scarce, not in the context of KGV coinage in comparable quality.  The proof sovereign is common and the 1924 is also almost certainly a lot more common in higher grades than much harder to find but much cheaper KGV dates in other denominations .  Last time I checked, the population count was over 70 but even with potential duplicates, I believe several hundred exist out of the original accepted mintage of 3100+.  I wouldn't buy it at its current price.

[Pierre_Henri 94]

This is a very interesting subject. 

I did a little bit of research but did not consult any internet sources – only hard copies of old time SA numismatic publications. 

Most old SA Coin Catalogs state a total mintage of 719 of which 655 are proofs and thus 64 are “currency” issues”.  (E.g. Bickels 1970/1, Nicholas 1979, Eastgate 1980/81, Randcoin 2000/1, Hern 2017/18, MTB 2017)  

The one exception is actually the very first SA Coin Catalogue (Alex Kaplan 1950) who states that only 586 Sovereigns were minted in 1923. In later catalogues he corrected this to the well-known figure of 719.  

Until very recently, that first catalogue of 1950 was also the only one that gave a higher catalogue value to the 1923 vs. the 1924 issue.  For many years after that, the 1924 issue was always valued higher than the 1923 date (both proof and non-proof issues). 

Herns’ last catalogues and the MTB 2017 catalogue corrected this and recognize the extreme value of the 1923 non-proof £1.  

Hern actually mentions that a mint official purchased 19 “Uncirculated Pounds“ in 1923 at the time of issue.  

Mohammed writes above  “Citing a report written by the Deputy Master of the Mint in 1924, Marsh pointedly states (see page 89 of his book The Gold Sovereign) that "During the period covered by this report only one deposit of gold was received at the mint. This consisted of old jewellery, which after the Rand Refinery's charge for refining, and the Mint charges, was found to be of the net value of £406.0s.9d. The resulting sovereigns were delivered to the depositor." This evidence suggests, but does not prove, that 406 currency sovereigns were minted at Pretoria during 1923; the mint master mentioned only "resulting sovereigns," which is suggestive but not definitive” 

So, if true the person who sent the unrefined gold to the Mint received 406 minus 19 Sovereigns = 387 coins. 

The information regarding the £406.0s.9d. worth of gold sent to the Mint are confirmed by  J.T. Becklake (From Real to Rand published in the 1960s) who gave the source as the “Fifty-fourth Annual Report of the Deputy Master and Comptroller of the Royal Mint, 1923”.  London, H.M.S.O., 1924. 

If one, for example looks at the NGC pop report regarding the non-proof issues, it says that 8 circulated issues (XF45 to AU 55) were graded and 4 in Mint State. For me that is proof enough that non-proof issues were indeed struck. If one compare this to the ultra-rare 1931 2-Shilling where only one non-proof coin is recorded, it places serious doubt that any non-proofs 1931 2/- coins were actually struck (383 reportedly were). Kaplan in his 1950 catalogue actually said that none were struck but later changed his mind for some reason.  

I am not sure how many non-proof 1923 Pounds still exists in non-graded form, but the number must be very low. Maybe some are still kept in a hand full of old-time family collections, but how many could that be?

[jwither 17]

On ‎11‎/‎30‎/‎2019 at 10:42 AM, Pierre_Henri said:

 

If one, for example looks at the NGC pop report regarding the non-proof issues, it says that 8 circulated issues (XF45 to AU 55) were graded and 4 in Mint State. For me that is proof enough that non-proof issues were indeed struck. If one compare this to the ultra-rare 1931 2-Shilling where only one non-proof coin is recorded, it places serious doubt that any non-proofs 1931 2/- coins were actually struck (383 reportedly were). Kaplan in his 1950 catalogue actually said that none were struck but later changed his mind for some reason.  

I am not sure how many non-proof 1923 Pounds still exists in non-graded form, but the number must be very low. Maybe some are still kept in a hand full of old-time family collections, but how many could that be?

For 1923 sovereign, I believe most proportionately might be outside of South Africa. There are certainly far more sovereign collectors elsewhere than in your country, though most collect UK, Canadian and Australian coins and not those from South Africa and the few from India.

I don't doubt that 1931 2/- were struck for circulation.  In the copy of the 1950 Kaplan you gave me, it co-mingles the mintage for all 1931 silver denominations.  Here are some additional thoughts:

The catalog implies that circulation strike presentation sets were issued.  That's what my South African contact claimed the UK based family who purportedly owned one these sets had; in a presentation case similar to a proof set.

The mintages for all coins except the tickey are higher than every other KGV set aside from the 1923, so I doubt that all of these coins were either proofs or issued in presentation sets.  I certainly don't doubt that 6D or 1/- were issued as circulation strikes.  The mintages are too high and don't believe all of these coins are actually circulated proofs either.  More likely that the few surviving 3D, 2/- and 2/6 could be proofs but there are enough of the 2/6 holdered by NGC (10) to confirm otherwise.

I suspect that the Mitchell collection contains a circulation strike complete set.  Scott Balson mentioned seeing the 3D (an MS) in person in a prior post.

The limited or lack of a recorded prior sales history is definitely a mystery.  But then, this is equally true for other Union coins (that 1934 farthing in my recent topic) or for that matter, any number of coins from the pillar series I collect.  Most of the latter aren't prominent even now but enough are and I have never heard of any prior sales.

Edited December 3, 2019 by jwither

[Pierre_Henri 94]

19 hours ago, jwither said:

I suspect that the Mitchell collection contains a circulation strike complete set.  Scott Balson mentioned seeing the 3D (an MS) in person in a prior post.

I am a friend of Dr Frank Mitchell's son, Rob Mitchell, himself a big numismatic collector. 

https://www.dnw.co.uk/auction-archive/special-collections/foreword.php?specialcollection_id=321

I asked him a few weeks ago if his father ever owned a non-proof 1931 Tickey and he was sure that he never did.

Regards

Pierre

 

[jwither 17]

On ‎12‎/‎4‎/‎2019 at 10:45 AM, Pierre_Henri said:

I am a friend of Dr Frank Mitchell's son, Rob Mitchell, himself a big numismatic collector. 

https://www.dnw.co.uk/auction-archive/special-collections/foreword.php?specialcollection_id=321

I asked him a few weeks ago if his father ever owned a non-proof 1931 Tickey and he was sure that he never did.

Regards

Pierre

 

Well, so much for that.  Now I have to wonder, if he doesn't then who does?  Maybe the answer is nobody?

I'd still be interested to know what else is in this collection and the presumably few others which I assume own the Union and ZAR coinage essentially never available for sale.  The Kruger and Burgers patterns, the scarcest Union proofs (like the 1926), the 1931 circulation strike silver...even the 1923 red proof bronze mentioned in the Hern catalogue.

If many or most of these coins don't actually exist, collecting South Africa is going to be massively less interesting.

[Pierre_Henri 94]

On 12/5/2019 at 7:25 PM, jwither said:

Well, so much for that.  Now I have to wonder, if he doesn't then who does?  Maybe the answer is nobody?

Regarding the unique 1931 Two-Shilling that has been graded MS63 – the only non-proof 2/- coin in the NGC census (including the details census):-

The question was asked if the coin is really a non-proof coin and not possibly a proof coin that was erroneously graded.

Thirty years ago, Sam Lieberman wrote an article in Journal number 2 of the Association of South African Numismatic Societies entitled ”Coins of South Africa. The 1931 Short Proof Set”.

He made this very interesting observation …

” (a) feature of the 1931 Proof Set is that it is the only set where the coins were finished off by hand. The Master Dies were manufactured in London and then sent to the Pretoria Branch of the Royal Mint. These dies had a very deep “False Rim” giving the coin a higher than usual rim on both obverse and reverse sides. This high rim was then filed by hand as will clearly be seen (here)”

If I understand the pictures correctly – the 1 mm blank strip between the inside reeding (called denticles) and edge reeding (milling) has been filed off so that that there is no blank space between the two.

The big question is if all 1931 proof coins were hand filed or just those that had extra high rims? Because if all were (even only slightly) an expert would be able to pick it up under magnification and THAT would be a tell-tale difference between 1931 proofs and non-proofs – especially for the larger coins like the two-shillings and half-crown.

Here are pictures of the unique MS63 1931 2/- and a proof 2/-. The latter was in my personal collection. I cannot see any edges filed on either coin, but it is obviously very difficult spotting that from pictures than in person.

Edited December 10, 2019 by Pierre_Henri

[jwither 17]

39 minutes ago, Pierre_Henri said:

Regarding the unique 1931 Two-Shilling that has been graded MS63 – the only non-proof 2/- coin in the NGC census (including the details census):-

The question was asked if the coin is really a non-proof coin and not possibly a proof coin that was erroneously graded.

Thirty years ago, Sam Lieberman wrote an article in Journal number 2 of the Association of South African Numismatic Societies entitled ”Coins of South Africa. The 1931 Short Proof Set”.

He made this very interesting observation …

” (a) feature of the 1931 Proof Set is that it is the only set where the coins were finished off by hand. The Master Dies were manufactured in London and then sent to the Pretoria Branch of the Royal Mint. These dies had a very deep “False Rim” giving the coin a higher than usual rim on both obverse and reverse sides. This high rim was then filed by hand as will clearly be seen (here)”

If I understand the pictures correctly – the 1 mm blank strip between the inside reeding (called denticles) and edge reeding (milling) has been filed off so that that there is no blank space between the two.

The big question is if all 1931 proof coins were hand filed or just those that had extra high rims? Because if all were (even only slightly) an expert would be able to pick it up under magnification and THAT would be a tell-tale difference between 1931 proofs and non-proofs – especially for the larger coins like the two-shillings and half-crown.

Here are pictures of the unique MS63 1931 2/- and a proof 2/-. The latter was in my personal collection. I cannot see any edges filed on either coin, but it is obviously very difficult spotting that from pictures than in person.

Thanks, very useful information.

Another indication would be where this coin came from.  Is it a circulation strike mixed in a proof set or not?  This occasionally happens, but if it did, the owner will never admit it.  This would make it more likely that NGC graded the coin in error.  This was the purported source to my recollection with at least one of the 1936 farthings currently graded as MS or "specimen" which I consider proof "rejects". 

I also bought a 1947 proof set with an MS shilling from DNW back in September, 2008.  However, DNW identified the coin as an MS, the distinction was obvious and NGC graded it MS-63.  The 2/6 was somewhat "cloudy" and was initially graded MS-63.  I subsequently had it conserved and NGC graded it PR-63.    I also recall that you had a 1947 2/- that NGC initially graded MS-68.  It was discussed on this forum, briefly.  It's no longer in the NGC census.

In the above images, an interesting difference between your coin and the NGC MS-63 is that the strike appears much sharper on the graded circulation strike.  With the limited production run, it's a lot more likely but not something I would necessarily expect.

Lastly, the NGC MS-63 doesn't have the reflectivity of a proof in the image which is more consistent with a circulation strike.  (More like a specimen to me given the strike which is consistent with the low mintage.)  Yours doesn't really either but I infer it's the image only, as it still looks like a proof.

Aside from an independent opinion by someone who knows more about this than I do, the next best option would be to compare the NGC MS-63 directly with numerous confirmed proofs.

[Pierre_Henri 94]

15 hours ago, jwither said:

Lastly, the NGC MS-63 doesn't have the reflectivity of a proof in the image which is more consistent with a circulation strike.  (More like a specimen to me given the strike which is consistent with the low mintage.)  Yours doesn't really either but I infer it's the image only, as it still looks like a proof.

The “problem” with the NGC picture of the MS63 2/- is that it was taken by a scanner and not with a digital camera. When a picture is taken of a coin with a digital camera it is fairly easy to spot the difference between a proof and a non-proof but with a scanner it is difficult.

Now here is a picture of the 1931 MS63 2/- compared to a 1948 2/- both scanned.

Both look like non-proofs to me, the 1931 date actually more proof-like

Now that same 1948 coin is shown below: – one picture scanned and the other taken with a digital camera.

The coin is a proof.

 

[Pierre_Henri 94]

I visited my old friend this morning to take pictures of his 1931 2-Shillings that he bought many years ago from a dealer in Cape Town.

My friend is 82 years old and started collecting coins in 1964.

Here is the picture.

I then had a closer look at the coin to see if I can spot any file marks that MAY indicate that it is possibly an ex-proof coin - see my posts above.

Here is what I found.

Have a closer look - could this be file marks?

[Pierre_Henri 94]

Here is the link to the 1931 2/- that Georg Jacobs sold a couple of years ago on BidorBuy

https://www.bidorbuy.co.za/item/15812717/1931_Two_Shillings_Mintage_445_only_ultra_scarce_Start_R1.html

If anyone knows who the winning bidder is (Bakerman Can) it would be nice to ask him what happened to the coin and if is available for closeup pictures to spot for file marks. 

(I have contacted Georg but he does not have the pictures anymore)

For those interested, here is a picture of a 1931 proof set with its original box.

As this will probably be my last post of the year (we are leaving on our annual summer holiday tomorrow) I wish to report that  Peter Wilson, the President of the National Numismatic Society, sadly passed away this week.  Our deepest and most sincere condolences to his family.

Pierre & Family

[xsiandreas 11]

On 12/13/2019 at 12:00 PM, Pierre_Henri said:

I visited my old friend this morning to take pictures of his 1931 2-Shillings that he bought many years ago from a dealer in Cape Town.

My friend is 82 years old and started collecting coins in 1964.

Here is the picture.

I then had a closer look at the coin to see if I can spot any file marks that MAY indicate that it is possibly an ex-proof coin - see my posts above.

Here is what I found.

Have a closer look - could this be file marks?

In my opinion that looks more like damage from circulation. I would expect the filing on proof coins to be kept to the minimum so that the coin is not damaged. However, one would have to compare it to some proof coins which have been filed to be certain. Very interesting nonetheless.