Quote from Steve on July 25, 2023, 7:58 am

This is a very interesting and useful article, well-presented and illustrated. It is offers practical advice based on hands-on experience and knowledge from working with hundreds, if not 1000s, of ZAR / Transvaal stamps. When I asked Bas to jot down his thoughts on the subject, this is exactly what I hoped to receive. It clarifies what I thought I knew and points me in the right direction to invest in a measuring  solution that is right for me. I am not sure how many of us postal historians will want to go so far as to measure postmarks - the response to this article will give us a clue - but for those who do this is an excellent starting point. Thank you for sharing this with us.

Quote from Steve on July 25, 2023, 10:20 am

AT THE RISK OF MAKING A FOOL OF MYSELF, AGAIN....

Among many projects, the current one I am interested in is determining the size of text / type in the hooded of the CoGH Hooded Circular Datestamp. Simon Kelly has identified three text sizes - 2.5, 3 and 3.5 mm. Previously the only person to attempt to describe these was Franco Frescura who limited his description, almost certainly wisely, to 'shorter' and 'taller' typefaces. This saved him a heck of a lot of trouble.

To begin at the very beginning..... Most Cape and Natal datestamps would have been manufactured in Great Britain while some OFS and ZAR ones were made there and or in Europe. At the time Britain used imperial measurements, miles, yards, feet, inches, fractions, etc. Europe did not. Today we measure everything in European metric, including our British-made datestamps that were not made in metric units. Is that an issue?

The countries that used Imperial measurements, Britain, its Empire and the USA, worked to tolerances of a 'thou', a thousandth of an inch. I wonder what the results would be if we attempted to measure our datestamps in imperial units. Would we see a sensible pattern of original design measurements emerge? If we took our Metric measurements and converted them to Imperial would we see that they approximated a likely unit of Imperial measurement used at the time? Shoot me down in flames. I am beginning to worry about my sanity here.

The observant among you will note that the difference between Kelly's measurements from 2.5 - 3.5 mm is just 1 mm across three type sizes. This is minute and, given the vagaries of datestamp-struck ink on old paper, it is something that is very difficult (some would say impossible) to measure with any real accuracy unless what you are measuring is an early example of a postmark that is both clean and sharp. This is all confirmed to Bas' article which, it must be reiterated, is based on years of experience.

Kelly's metric measurements converted to thou are:

2.5 mm - 98.425196 thou, say 98

3.0 mm - 118.110236 thou, say 118

3.5 mm - 137.795275 thou, say 138

The interesting thing about this is the roughly 20 thou increments in type size. (20 thou is .508 mm). Is this the start of the pattern I hoped to see? The thing is that despite their roughly 20 thou increments, 98, 118 and 138 are all such arbitrary numbers. How did the datestamp designer come by them? Was it possibly the design and aesthetic considerations of fitting them into the Hood?

I question the value of any of this. Have I not just proved that Kelly's metric measurement are sufficient as a guide? I imagine that we would be placing additional obstacles in our path if we were to measure in Imperial units. If we did what difference would it make? Can we draw anything positive from this post?

I look forward to being corrected. Engineering is not my strongpoint.

Quote from yannisl on July 25, 2023, 11:41 am

Steve

Some comments as an Engineer and a postmark collector.

1.0 All books on postmarks not only for the Cape provide guideline measurements to an accuracy of 0.5 mm. 

2.0 When the eye compares it can measure down to 1 micron (1/1000 of a mm). If you make a slit on a piece of paper and you put light behind it, you can measure possibly 1/100000 of a mm. We used this in metrology labs with slip gauges. 

3.0 In Engineering measurements are usually specified with a tolerance as well. One would say 2.5 mm +-0.01 or similar. As you might remember from school when rounding up there is never a real number that is perfect, so saying 2.5 can be 2.51, 2.52...2.59 rounded to a 4.0 For technical drawing we used to use Rotring pens that could draw lines from 0.1 mm to 1.0 mm. We also had stencils for letters to write letters also for similar ranges. 

5.0 So even my ageing body and eye can see and feel very low lengths. A thin plastic shopping bag is about 12 microns, a normal heavier type 30 microns. A human hair is about 120-150 microns. 

POSTMARKS

1.0 For postmarks if it was punched, I guess it used points as a measurement. Like in typography or Word. You also need some tracing paper. Find the center of the circle, with a set of squares. Use a compass to measure to say bottom of letter H, at middle. Do the same for the top. Repeat through a few and average. 

Sadly I don't have these instruments any longer. This is what the architects that drew postmarks for Postal History books, used at the time. I know it can be done with software. I could draw a a gear in about 5 minutes, I would guess a postmark in 1 minute. With software now is a hopeless exercise at least for me. 

Hope you find this helpful. You can still use this measurement and say tall approx 2,8 mm and short 2.5 mm. 

At this point I wouldn't worry about the original dimension was in imperial, point sizes or metric units. 

SUMMARY

Specify what you measured. Use a 0.5 tolerance or less if the difference is small, say 0.25. Use a compass to measure. Trust your eye. 

 

 

 

 

Quote from yannisl on July 25, 2023, 11:49 am

You can also use vernier calipers. The plastic one are a few euros in a hardware store. Electronic ones are a bit more expensive. 

 

Quote from Steve on July 25, 2023, 2:55 pm

Yes, I agree, one shouldn't 'worry about the original dimension was in imperial, point sizes or metric units'. I had more-or-less stumbled to that conclusion myself. Bas has emiled me on another matter and took the opportunity to comment on this. I think it is worth repeating here.

"As regards imperial and metric, I haven't discussed this mainly because if you want to find what units people were using (the opposite way of thinking about the question), what's important isn't the unit you measure in, but whether you find recurring patterns in the measurements you make, regardless of unit.  That's how, for instance, archaeologists think they have identified the "megalithic yard".

 

In fact,  the available evidence suggests that most pre-1894 ZAR cancellers were made in the UK!  The Ditsong Museum has a collection of about 30 pre-1894 ZAR cancellers, several of which are marked with the names of their manufacturers, all of which are London firms ...  My guess is that later ZAR and, probably, Transvaal cancellers were made in Europe, mainly because they are wheel-change cancellers, while Britain at the time was still firmly wedded to slug-change cancellers. I haven't yet come across any corroborative evidence."

 

 

Quote from Bas PAYNE on July 26, 2023, 4:13 pm

Many thanks for comments and discussion.  I attach a revised and, I hope, improved note.

Two main points:

The fuzziness of most postmarks makes very precise measurement unrealistic and potentially misleading.  However precise the instrument, in general +/- 1 mm is usually realistic; +/- 0.5 mm is often possible; +/- 0.1 mm is rarely realistic.  

Vernier callipers aren't very useful for measuring postmarks, for reasons set out in the note.