Recreation of Printing Output

This page is a work in progress, as much a place for me to place notes as much as anything else. My collecting interests center around Alphabet 2 (1852-1855), and having had a look at some of the primary sources of information, I would suggest that the task is infinitely harder than early Alphabet1. The very best that can be hoped for is an educated guess.

Lets have a look at the sources.

Printer's record found in the printing account books.

A primary source but during the period in question there is zero information about what plates were at press, when. The books appear to have been used solely for the recording of the daily printing output of each printer. The week ends on each Friday, which of course must relate to payday. The printers were paid based on a piece work basis.

The good news here is that these records are likely to be absolutely correct, because there is money at stake. Total figures should be possible and we do know how many presses were at wok at any time.

The difficulties include:

• • the fact that no plate information is shown;

  • there is not necessarily any indication where a printer switches from one plate to another;

  • there is no way of knowing that a break is due to the printer doing another job, going off sick or taking leave;

  • we do not know whether the press was left empty and the same plate went back on it after a break in printing;

  • production was high at this point, so a new plate could be worked up to speed without necessarily seeing a drop in production.

See general comment at the foot of this page.

Engraving books

These are likely to be reasonable accurate, despite Statham's words of warning. Statham also comments (p6) that the date recorded in these books includes the hardening of the plates. The hardening books do not start until 1864, and the following table shows that finishing the plate as far as the engraver is concerned does not necessarily mean that the plate has been hardened. 

Plate 82 is the earliest plate recorded in the hardening book. Presumably, plate 84 was hardened before the 24th February. It could be that the policy surrounding the recording had changed between the 1850s and 1864, but there is no evidence that I know of that suggests this.

This information is important as we know for sure that a plate cannot possibly be put to press before it has been completed. The available date will be at the very least a day after that recorded to allow for hardening, unless of course that it was one of the unhardened plates (173-177). Update 4/2023. I have been working through some of the earlier years print figures and need to report that the recording in the engraving book was in some cases somewhat slapdash! As an example, the engraving book shows plates 37 and 38 being finished in August 1843, yet we have notes within the printers print records that they took these new plates in July 1843. Care and discernment are needed.

Wright and Creek

Whether justifiably or not, in my mind the work by these gentlemen has always been suspect (probably caught this attitude from ED Bacon). It was printed in 1899, so the writers could be forgiven an awful lot. One point that caught my eye was the fact that Creek claims that much of the information that the authors recorded came from sources within the Inland Revenue. Many of these sources of information do not now exist, so it would perhaps be foolish to cast aside what they have recorded. Recently, I asked a prominent philatelist for his estimation of numbers printed from each sheet in imperforate Alphabet 2 area, and he sent me his 'wants list'. That list reflected the numbers printed in Wright and Creek very well indeed. In the absence of any conflicting evidence, why would I discard the (probably official) figures quoted. If I can match up these figures with dates and print runs by individual printers, it may well assist in the recreation that I am aiming for. Update 4/2023. I have since found that much of the information presented by W&C matches the information found in IR79/79, which is an Inland Revenue primary source. Bacon had close contacts within the firm of Perkins Bacon. But W&C must have had a close contact within the Government department. I have warmed to W&C.

Here are the relevant details for 1852 - 1855. I have highlighted those plates where they have been used out of order. I did think that there was some significance to this, but am now starting to think that that it is no more significant than a plate being at the top of the pile in the strongbox.

Seymour

The book The Postage Stamps of Great Britain part 1 by Seymour has a breakdown of the number of stamps found from a sample of 20,000. There is a problem with this. There is no way of knowing if the plating of these stamps were correct. In fact it's probably safe to assume that about 25% of these platings were wrong. This is in no way meant to be derogatory remark, as I am a firm believer that the use of computers and graphics programs is almost cheating. Today, we have it easy.

There are some thing to be said for taking these findings and using them as an approximate benchmark for my guestimates.

• • If the error rate was 25%, then 75% were correct.

  • Many of those errors are going to be compensating. For example, a stamp from 150 was plated as151, one from 151 plated as 152 and one from 152 plated as 150. All wrong, but the net effect are correct totals.

  • It was a large sample, one which would be difficult to achieve today. The larger the sample, the more likely that errors will compensated as described above.

  • Having already had a play with the figures found in part 2 in relation with die 1 perforated, I know that they are not that far out from reality.

Figures for 176 and 177 have been left out of the figures, which is perhaps just as well as they would have had the ability to skew the figures. I have listed the figures from the book below next to their plates (column 1 and 2). From this I have worked out a simple percentage figure (column 3). I have applied this percentage figure to what I think is the total print figure from alphabet 2 perforated using the actual print figures from the printers accounts (column 4). Wright and Creek's figures are included in column 5 for comparison. The start of 1852 include stamp details from alphabet 1 and it is a possibility that I have miscalculated when such and such plate was withdrawn from press, but the total is going to be close to being correct. At the end of the imperf era in 1854 the picture is also fuzzy. Plates 176 and 177 both went to press on the 12th of January and perforation started on the 27th, with only a few sheets from both plates were perforated. Taking an arbitrary point between these two dates gives a print figure of about 100.000 sheets that were not perforated. Some of the product of this period was distributed in an imperforated state as there was not enough capacity to perforate the whole product with just one Napier machine. A second machine was installed at the end of February which proved more than adequate to perforate all sheets printed.

It is easy to infer too much and let the calculations run away under their own steam. The problem is that we do not actually know for sure exactly what happened. This is no more than an imprecise guide that makes no claim on science of any sort. If anybody wants to have a play, my spreadsheet for this can be found here https://drive.google.com/file/d/0B_U99fp7XZZXUFRRS1AyVFo4NWM/view?usp=sharing

Enjoy!

Researcher X

Statham describes finding a loose sheet of paper in the Printing A/C books that stated:

Statham draws attention to the fact that the printer for plate 105, James, does not exist. Having looked through the books myself, it could be a simple case of misreading the handwriting. It could be Hennings Jr perhaps, or possibly Turner. This is not a primary source IMHO, it is just the work of a fellow philatelist. However it is encouraging to note that by and large the figures are close to those quoted by Wright and Creek. Those that are way off, I have highlighted in red.

How much should we rely on this anonymous source? Not at all, unless it coincides with something else we discover. The tantalizing question is whether this person was privy to information that we have since lost. A back up source at best.

Additional note: Alan Druce has independently spotted this list in the front of the 'Gumming Records 1850-58' within the Perkins Bacon archives. Researcher X has clearly copied this source which is in fact a primary source. I now need to backtrack and see how the number of sheets printed match other PB records. [September 2016].

Dated used examples

These can be incredibly useful. When comparing the usage dates and the printers records it is sometimes possible to tie in periods when certain plates were at press and being sold over the PO counter. Especially useful are any early dates for any particular plate. Late usage tends to muddy the water as it clouds what is being sold by the Post Office at any one time. Where I have quoted Earliest Known Use (EKU), it will be from Allan Olliver's excellent 'Lifespan' document, or else documented in the individual plate page.  Logically, stamps will not be found unless they are at press at a minimum of two or three weeks prior to the date of usage, usually longer. Dated covers come into their own when dealing with stamps that come off press, either as a reserve or for repair.

I have started a spreadsheet for alphabet 2 detailing date; plate; state; lettering and source. If anyone wants to add to this, please let me know. Also any scans of EKU covers would be a very gratefully received as a worthy addition to this site. Update 4/2023. Allan Olliver's lifespan document is now much updated and records the first three known uses from a plate with scans. This will counteract glitches such as wrong date plug used, and the plating of scans can be double checked. This is an ongoing project being run through the Mulready forum.

Some general comments

In the Victorian era, it was common practice for a workman to take responsibility for something and follow it through. We see steam engines that had only one driver for many years, some drivers chaining the regulator so that only he could move the engine. In a similar way, once a printer had set up a particular plate on a press, he would continue to use that press/plate combination. I am sure that this was not set in stone, only that it would have been preferred by the printers themselves. There was no such thing as a 'hot desk' in 1852.

As I have just hinted, once a plate was set up on a press, it would generally stay there. Most plates would have been relatively flat and square, they certainly would have been before they had been hardened when some movement out of true could have been a possibility. In any case, no two plates would have had exactly the same size or thickness, so the press had to be adjusted to fit. The way that they did this was to add (wood/hardboard?) packing under the bearings of both the top and bottom rollers, above and below the printing bed. Sometimes it took most of the day to get the amount of packing just right, at other times the printer was luckier or more skillful. Thus the packing on a particular press matched a particular plate. If a printer swapped plates, he would swap press as well. There would have been no point in setting up a new press for a plate that was already set up. I would suggest that even if there was a long gap in printing, the packing would stay on the press, only to be removed when the plate was retired and replaced by a new one. Update 4/2023. I think that any minor disortion caused by hardening that led to plate rocking could have been fixed by pasting paper to the back of the plate at the high corners, with some sanding of that paper to make final adjustments.

One question that I do not have an answer to, was how many presses did they have. I don't know. We can know how many presses were working at any one time, but how many spares? I would guess that they would have one press set up for the 2d blue, and left it empty when not printing blue sheets from it. Perhaps it were possible to take the packing off carefully and keep it tied together if a plate had to be removed temporarily, though my gut feeling is that this did not happen.

It has dawned upon me that some of the chaps that came into the Postage Label room for a week or two and then left it did so for a few reasons. Firstly they were spare and needed something to do. Secondly, PB had a rush on to get enough sheets to complete a warrant. Thirdly, They came in to help by setting up a new plate. A week or two in, they would disappear to wherever they came from and one of the regular Postage Label chaps would take over the press, allowing for a worn plate to be withdrawn smoothly. This would have been a good practice, but from my point of view is something of a nightmare. The problem being that the dip in production that one often sees when a new plate is taken up is no longer there. I might know that a new plate is on line, but have no idea which of the usual printers are using it and as a consequence which plate has come off press.

As I stated at the start of this page, the very best that can be hoped for is an educated guess.

AP

August 2015 - 2023