The single-cross coins exhibit a diversity of letter forms and other design elements. These differences form the basis of the classificatory system, but some inconsistencies both within individual coins and more generally, may pose difficulties in attribution. These vagaries are both a delight and a stumbling block. The constancy of other features point to purposeful design. The article describes and illustrates some of these issues.
…. and be sure to see the other two articles in this series on Alexander III
BNS Research Blog 
I can confirm that a number of 20 point reverse dies coupled with Class E2 have the three pellets as described here for E1 20 point coins
Thank you for this confirmation.
Hi David, Ron,
I will start by saying this isn’t my area and whilst I knew the various classes existed, the subtleties had eluded me. Thanks for the interesting articles and comments. So here are a few daft questions:
1. Do you have an estimate of the total number of dies used for this coinage?
2. Has anyone published a fully illustrated class and die study? Even if not every die is illustrated, just illustrating many dies for each currently recognised class would give a feeling for the variability and unusual varieties, sub varieties, errors etc.
3. I don’t underestimate the task, but would this be suitable for the blog?
All the best
Gary
Thank you Gary for your interest and for your pertinent questions.
In answer to your first question, I can do no better than quote directly from Stewart and North (p.38) – “If die-output was comparable to that of the Edwardian mints, the production of some forty or fifty million pennies would have required between a thousand and two thousand obverse dies, and many more reverses”. These authors drew on data available for English coins of the period, there being no comparable documentary evidence for Scottish coins. (see: Martin Allan, ‘Mints and Money in Medieval England’, Cambridge University Press, 2012, p. 132) In the absence of mint records, it is possible to estimate die numbers using mathematical formulas, as I did for the silver ryals coinage of Mary, but this is only possible when there are relatively small numbers of extant coins. (see: David J Rampling, The Silver Ryals coinage of Mary, Queen of Scots, JNAA (2016) 27, 90-148) http://www.numismatics.org.au/pdfjournal/Vol27/naa-journal-vol-27.pdf#page=96
There is no fully illustrated class and die study of the singe-cross sterlings. I agree that the publication of a large group of appropriate images would be a boon in illustrating the variety of dies used in the coinage. It would be a Herculean task in view of the number of enlarged high quality images that would be required.
Hi David,
Many thanks for the comments and also the link to the JNAA – an excellent piece of work – especially the die study and estimates. I was aware of the Stewart/North estimate of dies, which seems very broad. Can a better (statistical) estimate be made with existing data? My suggestion for using the blog for illustrations is that the pdfs can be much higher resolution and more numerous than the images in the BNJ and their scans online.
In the same vein I am working on some short notes that will add illustrations to old articles in the Num Chron and BNJ.
Thank you Gary for your comments and for the drill down on the question of die estimates. The statistical methods I am aware of are based on the number of identical dies in a known number of randomly garnered coins of the series, estimates being calculated separately for the obverse and reverse dies. A large assemblage of coins, while in all probability harbouring a number with identical dies, will tax the scrutinising powers of the researcher. It might prove more practicable and meaningful to study a sub-group of modest size, such as the coins with 20 points reverses which account for approximately 5% of extant coins, but even here the task will be considerable. Calculations on much smaller groups, such as coins with 21 points reverses, may produce inflated estimates of original die numbers as, in this case where there are at least four non-identical reverse dies exhibited by a very small number of extant coins. Such a result would suggest a mintage much larger than that evidenced by the few surviving coins, a remarkable, if improbable conclusion. I add the disclaimer that I am not a statistician, and would advise verification of these remarks. I also defer to your knowledge of pdfs and image resolution. I look forward to seeing the illustrated notes you are preparing for the Numismatic Chronicle and BNJ.