top of page
  • wcnumsoc

The 1931 Silver Coinage of the Union of South Africa

By Pierre H. Nortje (November 2023)

​For readers of this paper who are new to numismatics, it is important to understand the difference between proof coins and coins that are struck for general circulation. Proof coins are specially-made collectable coins that have highly polished fields (backgrounds) that create a mirror effect. They are usually sold by the Mint as boxed sets for each year of minting. Coins that are struck for general circulation are sometimes called “business strikes”, uncirculated, MS (for mint state) or non-proof issues. In this paper, we will use the latter term.

For both collectors of the coins of South Africa and the old British Commonwealth of Nations, especially those of King George V, the year 1931 will always be an extremely difficult year to complete. This is due to the scarcity of the silver coins of that year of the Union of South Africa. Collectors with deep pockets and lots of patience should be able to obtain a proof set of 1931 as these sets do become available from time to time. However, to complete a set in non-proof condition would prove much more difficult and although it is theoretically possible, it is doubtful that any collector has ever succeeded in this goal, especially in certified (graded) condition.

For some unknown reason, there is a possibility that the 1931 proof sets were issued in both blue and white boxes. Source: Bonhams and Bridgeman Images (blue box) and eBay (white box).

In South Africa’s first coin catalogue published in 1950, Dr. Alec Kaplan states that 110 of the 1931 proof sets were issued by the Mint and then adds that additional proof issues were struck for that year being threepence (18) sixpence (4695) shilling (6493) two-shilling (335) and half-crown (742). Dr. Kaplan’s figures are clearly incorrect and he probably meant that the additional mintages were for non-proof coins (business strikes). But that is also incorrect. In later editions of his catalogues, he corrected the quantities, which were 62 proofs for each denomination and then the non-proof quantities being threepence (tickey) (66) sixpence (4743) shilling (6541) two-shilling (383) and half-crown (790).

Considering the fact that the Union coins of 1931, especially the tickey, have gained legendary status over the years, it is surprising that so little has been published about them previously.

In Bickels Coin & Medal News of March 1967, it was announced that a 1931 tickey was sold in Richard Aron’s auctions to a Mr. Ivor Lazerson of Johannesburg for the amount R245, “the highest price yet paid of a 1931 tickey ”.

The coin was considered to be in VF to EF condition. In today’s terms, the sales price was approximately R23 000. (The Krugerrand price in 1967 was R27 so the buyer could alternatively have bought 9 Krugerrands with his money that would be worth approximately R340 000 today).

Source: Sixbid auctions

Earlier this year an NGC graded VF25 1931 tickey that was sold in London for £13 000 (just over R300 000). It is one of only three coins graded by NGC in non-proof condition. It came from the estate of the late Natalie Jaffe of City Coins in Cape Town.

The first mention of their scarcity that the author could find was in the South African Numismatic Magazine of 1947 where Mr. J.T. Becklake, the last Deputy Mint Master of the Royal Mint in Pretoria, reflected on the mintages for the years 1923 to 1942 and states for the year 1931 “In this year the silver series issues were extremely small, and all these are rare”.

The only actual article on the 1931 issues that we could find was written by Sam Lieberman in Journal number 2 of the Association of South African Numismatic Societies (undated but probably 1989 or 1990).

He tells the interesting story of Dr. Froelich of Port Elizabeth. He says that Froelich was one of the few subscribers to the Mint at that time, and in 1931 ordered his usual set for that year. When the Mint realized that they could not find buyers for all their 1931 sets, they asked Dr. Froelich to buy the last eleven unsold sets. The price was 13/6 but he offered 12/6 and the two parties eventually settled on 13/- (which was less than three times the face value of a set).

As a matter of interest, it was Dr. Froelich who originally bought what is considered South Africa’s most valuable coin, the so-called “single 9 Kruger Pond”, in 1954 from the estate of King Farouk of Egypt. When Dr. Froelich died in 1969, the coin was auctioned off in Day’s postal auction number 9. In the same auction, one of Froelich’s twelve 1931 proof sets was sold for R1650 and the other eleven sets were then sold at that same price to various bidders.

Lieberman also mentions a few other items of interest. He asserts that it is the only proof 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. The high rim was then filed by hand as will clearly be seen in Fig.1

Lieberman adds that some of the silver coins were varnished, probably by someone at the Mint who must have thought that similar to the copper coins, the silver coins should be varnished as well. In his Fig. II traces of varnish can apparently be seen. However, if one looks at the two figures, neither the file marks nor the traces of varnish are clear. We speculate that Lieberman was referring to the proof issues only.

Almost a decade earlier than Lieberman’s article, Peter & Gales Bowles with Brian Hern, confirmed in the South African Numismatic Trends (1981/82) that “many of the (1931 proof-) pieces have been hand filed with some specimens showing file marks right through the high points on the obverse and reverse”.

On 28 August 2022, Heritage Auctions sold a PF65 Cameo Two-Shillings graded by NGC and says it was “bestowed an appreciable "Cameo" superlative and seemingly deserving of something finer if not for the abundance of die polish to the mirrored expanses”. One wonders if the “die polish” reflected on is not in reality spotty remnants of the “traces of varnish” that Lieberman referred to.

Source: Heritage auctions

The scarcity of the 1931 silver series

A while ago the author worked through 35 old price lists from the pre-internet era (issued between 1967 and 1998) to determine how many 1931 Union silver coins (non-proof) were offered. The lists were issued by the following coin companies in South Africa: Bickels, Collectors Mail Auctions, City Coins, Eddie Absil, Alec Kaplan & Son, Randburg Coin, Good Hope Coins, Chimperie Agencies and the Coin Shoppe.

The coins thought to be unsold re-offerings and also proof coins were deducted from the results. The total number of tickeys offered was zero while the rest were as follows: - sixpence (12), shilling (19), two-shilling (1) and half-crown (3). The highest grades offered were as follows:- sixpence (an example described as VF/EF offered by Chimperie Agencies in May 1996), shilling (an example described as VF-/VF offered by Eddie Absil in July 1984), two-shilling (a VG specimen offered by Bickels in June 1974) and half-crown (VF+ offered by Randburg Coin in September 1997).

The author then checked the BidorBuy (currently named Bobshop) offerings between 2012 and 2013 and the following non-proof graded coins were offered; threepence (zero), sixpence (3), shilling (4), two-shilling (zero) and half-crown (1). Subsequently, if memory serves right, a low grade tickey (NGC graded G or VG) was sold by Randburg Coin and Trust Coins also had one on offer. A few 6d, 1/- and 2/6-pieces were infrequently offered.

Source: Trust Coins

We can recall only one example of a non-proof 2/- piece being sold on BidorBuy, which was a specimen that was sold on 5 October 2009.

Census reports

The combined NGC and PCGS census reports provide the following information (these include “details” issues, being coins with surface problems excluded from receiving a numerical grade, but authenticated nonetheless).



Non- Proof


Threepence (3d)




Sixpence (6d)




Shilling (1/-)




Two-Shilling (2/-)




Half Crown (2/6)








Regarding the proof denominations, it is apparent that there is not that big a difference in the quantities graded. For the non-proofs, the rarity of the threepence and two-shilling coins are obvious. In comparison, the Kruger ZAR shilling of 1893 is regarded as a scarce coin with 323 coins certified, so it is almost unbelievable that there are actually more 1893 shillings graded than all the Union silver coins of 1931 combined.

Some issues and questions

Regarding the NGC & PCGS census reports in comparison to the original mintage figures, the surviving percentage figures (number of coins certified vs. originally minted) are as follows:-




Threepence (3d)



Sixpence (6d)



Shilling (1/-)



Two-Shilling (2/-)



Half Crown (2/6)



Of importance is that collectors will sometimes submit coins for regrading in the hope that better results will be achieved the 2nd time around. In other words, the same coin can be shown on the census reports more than once. However, even if all the proofs were for example submitted twice, more than 25% of all those minted will today be in NGC or PCGS slabs.

Secondly, because 1931 is such an iconic year, one could speculate that proportionally more coins of that year will be sent for certification than most other years. For instance considering modern South African coins that have been minted in their millions, especially lower denominations, very few are sent for certification as the costs involved would not warrant the effort.

It would therefore not be far-fetched to speculate that the larger percentage of the 1931 issues have already been capsulated and that not many remain in old collections or estates which can still be certified.

A question that has previously been voiced: - Are some of the non-proof coins in the census reports actually proof coins that somehow found their way into circulation making them undistinguishable from normal issues? The answer is most probably affirmative. The following is an example: - The author sold a shilling of the Union of South Africa some time ago which, if a similar coin had been submitted for certification, would probably have received a grade of VF30 or VF35. The problem is the date of the shilling was 1949 and as we know, only proofs were struck in that year, so this well circulated coin was actually an ex-proof.

So even if a very small number of the non-proof issues of 1931 are indeed ex-proofs, the actual number surviving of those struck for circulation purposes is even less than indicated.

Which raises another question: Why are the non-proof tickey and two-shilling so exceedingly rare, especially last mentioned, that if 383 pieces were struck, only 4 coins were ever certified by NGC (PCGS has none certified)? A while back the author took pictures of an unslabbed 1931 two-shilling belonging to an 86-year-old collector in Bellville. Pictures of it were posted on the old BidorBuy forum but the forum and pictures are now deleted. The coin is in VF condition but on closer inspection it looked as if the rim was filed – this prompted the author to then ask on the forum if this could be an ex-proof as we know that some proof coins of 1931 had their rims hand-filed by the Mint.

Could the 4 non-proof coins graded by NGC actually be ex-proofs? Two of the coins are VF-details coins and the other two are graded F and MS 63 respectively. It is a pity that the serial number of the MS 63 coin is not known and the author cannot access its pictures. However, we do have pictures of the unique uncirculated half-crown of 1931, which was graded MS 61. On closer inspection, we actually believe this to be a proof coin wrongly certified as a non-proof (MS) coin.

Source: Numismatic Guaranty Company (NGC®)

Some last thoughts

Generally speaking, the striking of low numbers of both South African proof and non-proof coins in certain years is not that a rare occurrence. Proof sets of years like 1926, 1930, and 1932 up till 1936 are scarcer than the 1931 set and for non-proof coins, as an example, fewer shillings of 1948 were struck than the 1931 date while for the year 1950 the mintage of the 2-shilling coins are on par with the 1931 sixpence. In 1948 and 1949 fewer non-proof half-crowns were struck than both the 1931 sixpence and shilling issues. However, in those years when coins were struck for circulation purposes in low numbers, the number of proof sets issued was relatively high. What makes the 1931 silver denominations virtually unique is that BOTH the proof and non-proof issues were produced in low numbers.

Also, in years like the late 1940s when some non-proof silver denominations were struck in low numbers, the actual surviving coins certified in mint state condition are relatively high. Take the instance of the 1948 shilling for example. Although fewer were struck than the 1931 date, 73 coins are numerically graded by NGC with all but 1 coin in mint state condition. Compare this to the 25 NGC graded issues of 1931 which have NONE graded in mint state condition.

In the author’s view, the 1931 silver coins of the Union of South Africa deserve to be elevated to the title of the jewels in the crown of King George V. The tickey has always been put right at the top – but it now seems the two-shilling deserves that spot!



Bowels, P. & G. and Hern, B., South African Numismatic Trends, Standard Catalogue of South Africa 81/82, P&G Coin Company, 1981.

BidorBuy Numismatic Forum (now defunct).

Census reports of NGC and PCGS, published online

Engelbrecht, C.L., Money in South Africa, Tafelberg Publishers, Cape Town, 1987.

Kaplan, A., The Coins of South Africa, published by the author, editions 1 (1950) and 4 (1966).

Lieberman, S., Coins of South Africa – The 1931 Short Proof Set, Journal No. 2 of the Association of South African Numismatic Societies (undated).

Nortje, P.H., The Rarest of the Rare. Unique and Very Rare Gold Coins of the Zuid- Afrikaansche Republiek. New Voices Publishing, 2022.

The South African Numismatic Society Magazine for 1947, Issued for private circulation.

Van Rensburg, C., The South African Coin Catalogue, Published by Randburg Coin Company, 2002.

408 views0 comments


bottom of page