Rob

Clearly I'm too late to make such an observation and in any case I don't have enough digits to count the mistakes. I suspect that his name isn't John Kendall. Random letters put together in a haphazard fashion, will on occasion form sensible words.

Now you've got me COMPLETELY baffled! Database fields only have the validation rules you give them, and the default is usually "Put something in here - but if you don't want to it's ok with me". Like an inflatable d... No no! NOT going there!! I didn't set up the access database for my collection, my offspring did. Unfortunately he's no longer at home, so I live with what I've got. He said it was something to do with sorting the data, but as I don't understand it anyway because my computer programming skills are non-existent, I just do as I was told.

I think your second point (in purple) answers what seems to be the flaw in the logic of the first (in blue), Rob? The survival of both types in high grade being proportional to the ratio of mintage, only applies if both appeared concurrently. But as we know, the survival in quite reasonable grades of rarities such as the 1895 2mm and 1902 LT pennies is because - though both are varyingly scarce - both were also the very first of their kinds and so proportionally more would have been put aside than you might otherwise expect. However, in the case of the 1926, it was the commoner variety that appeared first and would have largely satisfied the demands of those just 'wanting the date'. Maybe a few (lucky) souls were aware of, interested in, and able to distinguish between, the two varieties and knew what to look out for. But, if the total mintage was - say - 100,000 then some of those few would have been looking forever with no luck of ever seeing one. I don't know if the 1st effigy dies were used up before the ME was used or not, as it is equally possible they were used concurrently if the mint was using more than one machine. In fact there would be a very good reason for running them side by side as this would allow the die characteristics to be properly evaluated. No newly introduced design would have been previously subjected to a test to destruction if that involved striking 50-100000 or so coins. That is why I was hoping somebody might have the mint records for this period on the off chance that it might shed light on the dates when 1926 pennies were struck and when the ME dies were introduced. Whatever date this happened, the pennies weren't struck until the second half of the year with either obverse design as per my previous post regarding the articles in the SNC, although modified effigy dies must have been produced in 1925 as the halfpenny is known with this date, but again presumably not until the end of the year as its existence wasn't recorded by Garside until the August 1926 article. Even if the precise dates are not available, it should be possible to make an educated guess of the limits based on the striking rate of the equipment in use at the time and total output. At the moment we are all hypothesising without firm evidence, so could do with a bit of research. Any members of the BNS in the London area could take time out to visit the Library in the Warburg Institute for example as they have probably got a copy of the Mint records for 1925, 1926 & 1927.

Shouldn't that be thong.

Prsumably the starting price is a reflection (sorry) of the labour time required to get it to that state. An hour and a bit at minimum wage rates is about right.

Looks machined to me.

I'm not sure I entirely agree here. Any fool can spot an H or a KN but it takes practice to isolate an ME from the ordinary effigy. Coin collectors are like anybody else and their level of interest and thence skill will vary widely. Only a tiny minority will have had the interest to buy or borrow a publication which shows them what to look out for. My change collecting of pennies as it related to MEs went in three stages; 1) I had no knowledge of there being two varieties of 1926, although I did know about Hs and KNs; 2) I found out that there were two types but initially thought they had a small head like 1928-36. I hadn't picked up on 1927 being a different design; 3) Somebody pointed out to me what I should be looking for and after a while I could recognise one at ten paces. As £400 and 1949 Threepence have pointed out, pre-internet collectors were often lone wolves and it took time to garner information about the subject. Many will have learnt more by borrowing Seaby's (now Spink's) from the library, as I did, but library books have to be returned and you can't learn everything in the two week period before the fines start to kick in. My point is that because of the level of skill involved there were always going to be less people looking out for MEs, although I suspect that like KNs (not Hs) the majority were eventually harvested from circulation but as this took a long time, they were typically in worn condition. On the assumption that nobody recognised the ME and the collector wanted the best example of a 1926 penny he or she could find, then the percentage of mint state pieces of either 1st or 2nd effigies should be directly proportional to the ratio in which they were produced. There is no reason to discriminate in favour of or against either type if both were freely circulated. If you did know of the ME's existence, then it is reasonable to assume that you would look out for both varieties. I can accept that conceivably there were few collectors in proportion to the total collector base who were aware of it, but the limited number of coins collected as opposed to freely circulated should mean that the first sentence is a very good approximation to reality. Once the existence of a rarity, perceived or real, enters general knowledge there will be a tendency to hoover up all examples encountered whatever the grade. KN & H pennies (or even the 20p mule) are good examples of this. The result is that they are relatively common in circulated grades when compared to actual mintage figures for all types because they survived the great melt at decimalisation. This should also apply to the MEs because they would not have been withdrawn from circulation until decimalisation. The key to the mintages should therefore lie in the number of mint state examples extant. In an information free environment the statistics are going to be as unbiased as you can get. Once you have the knowledge that there were two types of 1926 penny and have read about the differences. I don't think it is too much to expect that the collector looks for a date first of all and then the position of the BM. And if they don't care or can't remember these two features, then the first sentence will apply in whatever grade.

I'd be happy to be paid full Spink book price for one of those too. If it's that collectable, surely the right place is in an auction because more eyeballs will see it. Anyway, if it is serial number 1, then all you have for certain is the first COA printed but not necessarily the first coin. The coins are not going to be struck next to the printing presses with limited runs fully automated including packaging.

The mouse over effect sounds neat, but is of no immediate use to me as I would normally go to half screen images with the unknown image on one side and the pre-existing stored images appearing on the other as each link is clicked in turn. Another advantage of Excel is the effectively unlimited capacity storage facility to the right hand side of the sheet. If I can identify a coin as an ABC123 but don't have an image, I can record the listing for future reference next to those I have imaged in the hope that one day a reference will tie the unknown listing to a specific illustrated coin. However, I still see n lines of provenance in a single Excel box preferable to n lines of Access fields when calling up the complete provenance. I can't see any difference between 6 column wide Excel data and 6 column wide Access data in terms of ease of use. My collection details are stored in an Access database, but I find the requirement to fill in all the boxes before you can move on a bit of a pain as I may not have all the info to hand. I prefer Access when it comes to sorting columns because Excel doesn't always sort things correctly if I remember correctly leaving you with an irreversible layout. Sorry, slightly off topic here.

It isn't rocket science though when all's said and done. You are only going to look for something you are aware of; but once you know, a quick check of the date and if 1926 a check on the position of the BM. I haven't got the slightest interest in pennies, but wouldn't have any difficulty recognising one. Whether a date collector or a full-blown variety collector, both would check the date first because it's the most likely gap in the collection. There seems to be a presumption that previous collectors weren't able to recognise the variety, but the information has always been there if they were so minded. The obvious conclusion is that either collectors took no interest in contemporary currency (similar to the attitude of many collectors today regarding current mint output), or that the coins were never that common in the first place. The former is more likely in my view as the great collections in the first half of the 20th century were focussed on hammered coinage of all ages with a smattering of milled thrown in. Lockett sold his milled currency at the end of the 1920's to concentrate on building up the best collection since Montagu and Murdoch. Would he have had a 1926ME penny? I don't know because I don't have a copy of his milled sale, so possibly, but the attraction would have been as great as putting a low denomination 2010 currency piece in your cabinet today. Most collectors of contemporary currency tend to be younger or poorer collectors as this is the cheapest way into the hobby. These are also people who may need to spend their collection on cars or other consumables as they get older or as fate dictates. Several factors mitigated against the hoarding of 1926 pennies such as economic depression, war and significantly, coin collecting became less popular throughout the 1930s compared to the 1920s, possibly as a result of the economic mess. All in all not a happy mix for collecting high grade pennies or indeed for ensuring a good supply of pieces for the future because fully two decades of potential circulation ensued before coin collecting became more popular again.

Spink Numismatic Circular. 1926 p.441 and 1927 p.413 refer. Unfortunately the pages are too big to reduce the file size to less than 150kB if you want image clarity. Another possible reason for their lack of inclusion in contemporary collections is that anybody who collected copper or bronze was considered a bit of an oddball. Even coins such as Soho patterns would be lumped together in bulk lots of dozens whereas today they would be given individual lot status.

Their presence was certainly known within the collecting fraternity from the start. Henry Garside issued a monograph on the subject of British Imperial bronze coins in the Circular. He added to this in the August 1927 circular listing both the old effigy and the new one described as quote " O. - Similar to the obverse of 1, but the King's effigy has been remodelled, His Majesty's hair is treated differently, the eye has a different appearance, the top of the ear is more rounded, the initials BM (Sir Bertram Mackennal, K.C.V.O.), on the truncation of the neck, are more to the right, with less space betwen them, and no stop between and after them. R. - Similar to the reverse of 1, but dated 1926" Garside's list was updated every August. In 1926 this update included only the 1925 halfpenny with the revised reverse. The 1927 update included both types of 1926 penny and the 1927 penny; 1925, 1926 & 1927 ME halfpennies; 1926 & 1927 ME farthings. Therefore we can assume that the ME pennies appeared in circulation after the August 1926 Circular went to press. This would be in keeping with the low mintage for the year irrespective of type and is not unreasonable given there were no pennies dated 1923-5. Rather more surprising is the inclusion of the 1925 ME halfpenny in the 1927 update as this should have been reported in the previous year assuming the update was timely. Whether this means that the 1925 coins were actually issued in 1926 I can not say as I don't know the date the ME dies were first used, but whatever, there is an inconsistency here which needs to be resolved. The Mint Records would be helpful here if anybody has them.

I'm still not convinced about the wisdom of having the provenance information (which is the raison d'etre for the entire exercise) stored in a separate layout away from the main page (or the concordance in yet another). It also begs the question as to what I would put in the main database. It seems more sensible to have everything under one roof rather than in several databases. The only things I store away from this are the images which are systematically named as the standard commonly used detailed classification (eg. Besly or Morrieson die pairings) together with the image sale reference, and stored in clearly identifiable folders giving the mint and denomination. This also allows me to run down the image folders and files in quick time without entering the database if I want to compare an image with one already stored. I don't know if the attached will show up adequately, but it illustrates the 6 columns as used with the individual coin provenance forming the link. As the full sheet displays all the examples I have recorded to date, a quick scan down the list (16 in the case of the Exeter C11 crown) will tell me what to look for. For example, Cumberland Clark 95 I know to be the C11 crown with the 5 legged horse. Click on the box with the provenance and the image pops up. It's simple, but quite effective.

Years is not uncommon on higher end stuff. You need a lot of capital and nerves of steel! Oh and an understanding spouse, bank manager etc Drug dealing would show a quicker return although the results for failure are much worse That is why I try to buy decent coins for selling on that wouldn't look out of place in the collection. If I don't sell them within a reasonable time frame, the temptation to incorporate them in the collection sometimes becomes too great to resist.

Given that nobody knows the exact number produced and all Freeman's (as are every other's) estimates of rarity are numbers based on conjecture to some extent, then maybe the R5 is not a true reflection of reality. Comparing various writers' attributions of rarity values for issues over the past few hundred years, I would say that in general there is a tendency to underestimate the numbers of older coins in existence, but possibly also to overestimate some of the modern rarities. It is quite possible that R5 (50-100K) is in fact an optimistic estimate of the quantity in existence. Given that nobody has a crystal ball or the ability to oversee the entire market, all rarity numbers must be treated with a degree of scepticism. If the market place doesn't bear out the quoted numbers, then it is probably wrong. Freeman based these assessments on his pile mostly accumulated from circulation, but it is equally possible that he was not averse to picking up small batches from markets stalls etc just as we all are today which could bias the numbers somewhat. What I am saying is that there is no guarantee that the sample of approx. 60000 pennies was entirely representative of the numbers in circulation. A parallel conundrum is that of the F465A 1953 proof halfpenny with the 2+B die combination. Mintage of the 1953 proof sets was 40,000. Freeman gives the F465A as R14 (251-500), but apart from my own example I only know of a handful of people with one. I would expect to see a number appearing on ebay, but they don't. You don't see any in auctions, yet they are rare enough to warrant a separate lot. I think you would have to do your own survey and generate your own data to arrive at a rarity with which you feel comfortable. Above all, question anything that is written elsewhere. Whilst there is not likely to be any intention to deceive, the market wants to be given an indication of rarity and the author will attempt to fill the demand. That's human nature. Do your own work and draw your own conclusions.

That's a 'how long is a piece of string?' question. No two coins are the same, some dates are more desirable than others and some denominations are more desirable than others. Coupled with the fact that not all desirable dates for one denomination are the same as those for a different denomination and you can easily see that there is no straightforward answer. The first thing you need to do is acquaint yourself with grades and grading. Get some experience in grading coins and see what they sell for when compared with references such as CCGB or Spink. ebay is not the best indicator of pricing because too many things sell for far more than they should and conversely so. In the case of pennies and farthings, 1849 is a good starting point, though you might take a while to get out of the blocks. Obviously the price to buy for needs to be lower than a selling price, but things starting at an elevated price on ebay frequently don't sell because most buyers are greedy. They want everything for 99p and consequently most will disregard the £100 coin priced at £75 BIN because they live in hope that the seller will give up trying to realise a decent price and offer them the chance of a real steal rather than a bargain.

You are quite right that I don't know all the ins and outs of databases (or spreadsheets for that matter), in fact I find computers incredibly depressing things to use as they frequently don't give me the answer I want - usually because I don't have enough in depth knowledge of a program. All I need is something that I can understand and use easily to provide me with the information I am seeking to collate together with an image of the coin which I can compare with an illustration in another catalogue and so record the new coin within an existing provenance, or I can generate another known example of that type. In the case of the example above I know that its provenance is ex E W Wigan (collection bt by Rollin & Feuardent 1872), H Webb 560, J G Murdoch 194, G Hamilton-Smith (1913) 126, K Vaughan-Morgan 336, V J E Ryan 1307, J R Vincent, J G Brooker 1153, 3 x SNC references, A Morris (from Roddy Richardson) and finally me (from Lloyd Bennett). It is no help to have the details neatly tabulated out of view from my perspective as the list of names immediately tells me which coin it is, just as the auction catalogue will have a list of past owners underneath the description. This is why putting all the names into one box works so well for me. Any system that requires a single field entry for each name dismembers the provenance. Having a link to the image allows me to compare a new catalogue reference with an existing provenance.

Bidders on a limitetd budget wouldn't realistically expect to buy an 1827 1d for a quid or even to be in the running to buy one at all. It's horses for courses. Presumably you wouldn't complain that you had bought a coin too cheaply and if a coin got lost in the post that wasn't insured you would also have to take the loss on the chin. There are two parties to every transaction and both have interests that need to be considered. I send everything out at least signed for. If I didn't then buyers could make a claim on their credit card against me for non shipment of a paid for item and I would end up footing the bill. In the case of Paypal it is even worse because the seller is automatically the guilty party and so there is an even greater incentive to cover your backside.

Why? If you are going to purchase items that are going to cost more than the statutory compensation limit of 100x first class postage, then it is reasonable to insure against loss. OK, in this case it looks like a case of a balls up in the listing because you can't expect the £1 BIN to be intended, but if the price had been listed at a reasonable level, then to cover both seller and buyer against loss or claimed loss would not be outrageous given special delivery is £5.05 up to £500 compensation. It's only excessive if you charge SD postal rates for crap items of minimal value.

I'd go for lower relief which would be easier to block up such that clarity is reduced. Fill a 1mm deep hole to a depth of 0.1mm and you probably won't notice; fill a 0.15mm or 0.1mm deep hole to the same depth and any design will almost or completely disappear.

I've no idea at present as I haven't checked out 1/2 sovs, but will keep an eye open for a reference. They aren't a commonly collected denomination to any depth, so any occurence is likely to be random which is unhelpful when searching catalogues. It's likely to be a Captain Oates moment.

Scott cataloguing these farthings will be an absolute nightmare, I have several that do not fit perfectly into the "varieties listed" but there are that many subtled differences that the page would be about 3 miles long!! I only record those farthings where there is no mistaking the variety, and usually would look for a higher grade example to act as confirmation before adding it to the site. I get numerous submissions daily, and it is a constant battle trying to determine what warrants inclusion and what does not. Spotting the different Obverse on that coin would be impossible. It is determined by the brooch design, and therefore any other subtle differences could not be used as a determining factor without careful comparison against numerous other coins. "so what to do, is this now the offical 9+B? or is this 11+B" I only wish everyone saw my numbering as the official system I understand what you are saying in that it appears to my 9b prior to it having the 3 recut, but because the exact combination could not be determined with certainty from that coin, I would only record it as a potential variety to keep an eye out for, it would not become 11B just yet Herein lies the real difficulty of producing definitive lists of varieties. In the era before reducing machines, every die was punched in by hand and every die will therefore be different. Add in the number of different punches used for what on the face of it is a similar die and you could easily write a 50 page article on just one issue such as the 1771-5 farthings. All of these common early issues would give the same results if accorded the same research. There were a minimum of 50 ship punches used and probably more for the 1799 or 1806-7 Soho halfpennies. All are different varieties as a result. Where do you stop? Probably when you think it is all getting a little silly, but don't worry, there is always someone waiting in the wings to carry the baton to a new level. God - we're a wierd bunch.

I have a bookmarks folder which expands into a column high enough to fill the screen but no more. I therefore tend to add a useful one at the expense of a less used (or useless) one. You remember those that are of use in any case without needing to bookmark.

I can see that if you are only recording a fixed number of data points per item, then Access has to be easier due to its versatility for searching, but in my instance I found it easier with Excel to add in new data due to a variable number of data points for any single coin. Take the example of the unique D23 Worcester 2/6d (Brooker 1153 for those in the dark). I have 14 references to this coin in sales or lists over the past 140 years for which said number may well be increased if I can find the lot bought by the owner who sold it in 1872. Those 14 (or potentially more) individual sales would each require a field if I am to be able to search using the sale date. Many individual coins will only have a single sale reference, but I will still have an image for it in just the same way as I do the coin with the long provenance. Add in the concordance for the various (numismatic) references through the ages which may well number 8 or 10, a brief description of the defining points of the variety, any notes of interest for the coin and you are looking at a Access database that is say 30 fields wide. This is unwieldy compared to what I currently have in place which has only 6 fields - Date, Reference, Variety Identifier, Metal/Mintmark/Moneyer (depends on coin type), Provenance and finally notes of interest such as unique/ so many known/ no of die varieties known for the generic design etc. This isn't searchable, but you would never have to look far as any systematic entry will automatically be in the correct denomination file and listed under the correct monarch or any other chronological attribute even if not dated. So for example, an Exeter C6 will always be cunningly interposed somewhere between the last entry for a C5 and the first C7. It is the guaranteed inconsistency in the number of fields required that made me use Excel and condense them into a single box for multiple data points of the same feature. The only way I can see that this would be improved is if I had some means of sorting the group of coins of an individual die combination into the best known down to the worst, but that isn't a problem until you have a few dozen or more examples imaged.

My son's school library was recently sent a copy of "the Phoenix" which is Coincraft's sales paper. I can't believe the prices or for that matter the content. Front page news is the announcement that they have purchased about 380,000 predecimal coins and are selling pouches of 65 coins for just £28.50 including a Churchill crown, silver 3d, and 1 each of a Victoria, Edward VII and George V penny. Inside the paper you can get a scratched grotty John penny for £145, or a WWII set of brass threepences in average circulated condition for £7.95, or a very good 1797 2d for £50. Anyone for 1968 date sets at £9.95 when you can't give them away? Who buys these things? On the only occasion I've been into the shop, I saw a 1942 Ag 3d for £125 and walked out again. Tourists must get stung, badly.