Examining the data on the plate repairs

The information below has been extracted from the text above and relates to the number of impressions/heads re-entered or repaired on six of the earliest plates that printed this stamp. The re-entry details stated by E D Bacon have come from the Perkins Bacon daybooks. However the plate defacement dates which come from Inland Revenue document IR79/79, pages 41-42, 43- 44 and 45-46 which is held at Kew have now been added. This now gives us the following most interesting table 1. The defacement dates below tally, more or less, in the majority of cases, with the Perkins bacon record books.

Table 1

Data from IR 79/79 on the above mentioned repaired plates.

Plate         Put to press                     Repair date               Number repaired             Partially Defaced                       Defaced

72               1st Mar 1864                     Feb 5 1868                  13                                      1868 Mar 23                                 1868 Jun 23
73               1st Mar 1864                     Jan 31 1868                 67                                     1868 May 5                                   1868 Jun 23
80               1st Mar 1864                     Feb 5 1868                    3                                      1868 May 12                                1868 Jun 23
81               1st Mar 1864                     Jan 31 1868                 49                                     1868 May 12                                 1868 Jun 23
85               1st Mar 1864                      Feb 4 1868                  87                                     1868 Mar 23                                1868 Jun 23
90               30th Mar 1864                    Feb 25 1868                 3                                      1869 Jan 18                                 1869 Oct 1

Studying the above data we immediately notice an interesting point, this is that plates 72, 73, 80, 81 and 85, the repaired plates, were all put to press on the same day and were all defaced early in 1868. They would have therefore all had a similar life span. Furthermore it must be borne in mind that by early 1868 these repaired plates would have printed the vast majority, if not all, of the total number of sheets of stamps printed by them and which would have amounted to over half a million sheets or so each, see table 2.

An interesting question to ask now is why was it that seven of the earliest printing plates from the first batch of plates to be put to the press needed to be repaired? Was the early method of making them or the quality of steel used for the plates, for whatever reason, somewhat wanting? It does certainly appear that the later plates were better prepared for the rigours of printing the large number of sheets required from them.

Table 2

Dates the ‘repaired’ plates were put to press and the number of sheets printed from them.

Plate       Put to press            Number of sheets printed

72           1st Mar 1864             522,800
73           1st Mar 1864             529,900
80           1st Mar 1864             495,200
81           1st Mar 1864             520,300
85           1st Mar 1864             510,300
90           30th Mar 1864           471,700

It is perfectly clear from the details in tables 1 and 2, that we now have a fundamental question to ask. Why would the printers carry out such major work on six of the earliest plates (72, 73, 80, 81, 85 and 90), only a few months, in fact just over one month in the case of plate 72, before their defacement, when there were several other working plates available for printing these stamps at that time? See table 3.

Apart from the minor repair work on the plate, it must be noted that re-entering a head from the roller onto the plate in order to re-enter an impression, strengthen an existing worn impression or to correct a major fault in it, is an exacting, time consuming process and one not to be taken lightly. Now the idea that over 219 head re-entries were carried out on five of the earliest, and by then very worn, plates in the space of six days, Jan 31st-Feb 5th 1868, and three months before the plates were defaced and four years after they were first put to press having most probably reached the end of their working life, would take some understanding.

Examining the information on the on the dates the plates were registered and put to press

Let us now examine the details behind Bacon’s statement regarding the reasons for the plate repairs and the plates that were available for printing stamps. E D Bacon cites the following :

“The reason for repairing this batch of plates was due to the stock in hand having got very low at the commencement of 1868, which is not surprising, as no plates of the One Penny value had been registered between April 4th, 1866, and March 23rd, 1868. Plates 108 to 111 were registered on the latter date. Plates 112 to 116 on May 12th and Plates 117, 118, on June 9th, 1868, and every one of these eleven plates was put to press on the day it was registered, shewing the urgent need that then existed for new plates.”

Table 3

The dates that plates 110-118 were registered and put to press (nine plates)

Plate 110 registered March 23 1868 put to press March 23 1868
Plate 111 registered March 23 1868 put to press March 23 1868
Plate 112 registered May 12 1868 put to press May 12 1868
Plate 113 registered May 12 1868 put to press May 12 1868
Plate 114 registered May 12 1868 put to press May 12 1868
Plate 115 registered May 12 1868 put to press May 12 1868
Plate 116 registered May 12 1868 put to press May 12 1868
Plate 117 registered June 9 1868 put to press June 9 1868
Plate 118 registered June 9 1868 put to press June 9 1868

However what Bacon fails to mention or cite are the dates that plates 104-109 (six plates) were put to press:

Plate 104 registered April 4 1866 put to press Jan 22 1868
Plate 105 registered April 4 1866 put to press Jan 31 1868
Plate 106 registered April 4 1866 put to press Feb 29 1868
Plate 107 registered April 4 1866 put to press Mar 18 1868
Plate 108 registered March 23 1868 put to press Mar 23 1868
Plate 109 registered March 23 1868 put to press Mar 23 1868

It is abundantly clear from this table that plates 104-109, six plates in total, of which four where already registered, were all put to press around the dates the plates were said to have been repaired i.e. between January and February 1868. Why then consider repairing exhausted plates when fresh plates were available for press? Furthermore between March and June of the same year nine more plates were registered and put to press. Not forgetting that the following plates which were not defaced at that particular time were all available for printing: 71, 74, 76, 78, 79, 80, 86-87!, 89, 92, 93-94!, 95, 96, 97, 98, 99, 100, 101, 102, and 103. Twenty one in total and evidently a large number.

Examining the information on the on the reasons why the plates were repaired

If we now look at the sentence below in Bacon’s book in which he tries to justify the repairs he states the following:

“This is also confirmed by Mr Ormond Hill's message to Mr J P Bacon on April 2nd, 1868, in which he asked him "to take steps to prepare plates faster." Old plates could be repaired in much less time than new ones could be constructed and hence it was determined to repair eight of the plates that were at press in January and February, 1868.”

This sentence is very questionable and has a subjective element. It will be investigated.

This is the relevant text, verbatim, from the lengthy letter cited above which was sent from Bacon to Ormond Hill on April 2nd 1868.

“I have just received a message from you through Mr Peacock urging me to take steps to prepare plates faster.”

Now if Ormond Hill was serious about this request in an official capacity then I believe he would have written directly to the printers regarding this matter and would not have left it to a third party to pass the ‘message’ on. No doubt there was a demand for new plates but I feel that this was more of a nudge than an official request. The correspondence between both Perkins Bacon and Ormond Hill was profuse and dealt with all manner of subjects ranging from mundane matters, official requests and technical matters etc. Surely a matter such as this, if fundamental, would have been dealt with in writing.

Secondly the plate repair dates cited by E D Bacon which presumably were taken from the record books were carried out in January and February 1868, i.e. about two months before the date of this letter! This instruction which was made in April cannot therefore be taken an indicator in reply and compliance with this request.

In my view Ormond Hill wanted new plates ‘prepared’ and not old plates ‘repaired’. Furthermore E D Bacon’s assertion that plates could be repaired faster than prepared is correct. However if there was an urgency to repair them, as he says, then why take about one month to repair the six exhausted plates, between 31st January and 25th February when we know it takes 9 days to prepare a brand new plate (letter from Bacon to Ormond Hill dated Feb.17th 1864)?

E D Bacon’s statement ‘it was determined to repair eight of the plates’ is purely subjective and has no foundation whatsoever in view of the fact the we know that six plates, 104-109, were ready and put to press between January and March 1868. It is clear however that the demand for new plates was addressed with some urgency as we can see from table 3 that plates 110- 118 were prepared and put to press in the space of three months, i.e. March-June 1868.

The lifetime of the printing plate

To further endorse the argument against the repair date of ‘1868’ being valid, a rather laborious but a very important and informative piece of work was needed to be carried out in order to see if there is any correlation between the printing lifetime of the plate i.e. the number of years it was available for press and the number of sheets printed from it. For the purpose of this exercise it is the approximate figures we are looking for and not the exact preciseness of the data.

Table 4 below shows ’A’ the plate number, ‘B’ the approximate number of years it was in commission, ’C’ the approximate number of sheets printed from it (1,000s) and ‘D’ the average number of sheets per year printed (1,000’s). The data is within c.5%.

Table 4

Data showing the average number of sheets per year printed from plates 71-118 (E&OE)

A         B        C          D
71      4.2      557       129
72      4         523       130
73      3.8      530       139
74      4.1      531       130
76      5         555       111
78      4.8      615       128
79      5.2      639       123
80      4.2      495       118
81      4.2      520       123
82      2.1      263       125
83      2.1      200         95
84      3.8      369         97
85      4         510       127
86      4         460       115
87      4         462       115
88      1.8      199       110
89      4.7       503      107
90      4.10     472      115
91      3.10     384      124
92      5.1       568      111
93      4          455      114
94      4          478      119
95      4.5       533      118
96      4.3       488      113
97      4.10     537      131
98      3.2       351      110
99      3.4       355      104
100    2.4       256      107
101    3.9       372        95
102    4          495      124
103    3.3       400      121
104    1.8       176        97
105    1.11     203       102
106    3          391       130
107    2.7       321       119
108   1.8        213       118
109    2          237       118
110    3.1       317       102
111    3.5       453       141
112    2.7       299       111
113    3          366       122
114    2          233       116
115    2          214       107
116    2.9        351      121
117   4.11      479       116
118   4.11      441       107

We can clearly see beyond any doubt and without any exception that each printing plate 71-118 averaged to print between circa. 100,000 and 130,000 sheets per year. If we refer to table 4 above we will note that all the six repaired plates (72, 73, 80, 81, 85 and 90) by the time of their defacement have printed this average, some, such as plates 72 and 73, printing the higher end of this average. So the question to ask again is why would the printers wish to carry out extensive repairs to plates that have been exhausted a short time before the date they were defaced?

The repaired plates have evidently printed their full share of stamps.

Where other plates also similarly re-entered and not recorded in the 'Record Book'?

In my view this possibility certainly does exist and I would like to cite plate 78 as one example.

A final thought! - ‘1865’ and not ‘1868’

What if the stated year of repair of 1868 was in fact 1865, whereby the figure ‘8’ was either written as an ‘8’ in error or was perhaps misread as an ‘8’ when it was in fact a figure ‘5’ instead i.e. 1865. The two numbers could quite look similar written by hand!

Would this possibility not be one to keep in mind in view of the arguments made above? The year 1865 for these repairs would then make a lot of sense in that the plates would have been in use for almost one year and in view of the fact that they were the mainstay for printing this issue then perhaps they would have been somewhat worn due to the printing pressure exerted on them. It also appears that the early plates were the only ones that had been re-entered which perhaps could mean that they may have been manufactured from a more inferior metal or perhaps the impression may not have been embedded as deep, as this problem does not seem to arise with the later plates.

It is worthwhile repeating here that by March 1868 a good number of plates were fully available for printing and this would, once again, beg the question of why would the printers wish to repair plates that have reached the end of their lifetime when brand new plates, plates 104-109 available for printing?

The re-engraving of some heads on plate 73 with the number 77

As far as plate 73 is concerned, the plate on which some heads were re-engraved to show a plate number 77, it is recorded that on January 31st 1868, 67 heads were re-entered according to Perkins Bacon day books and the plate was defaced on May 5th 1868. The plate by January 1868 would have reached the end of its working life having printed in excess of 529,000 sheets, please refer to the correlation study, table 4. To blindly accept that this major repair work was carried out three months before the plate was partially defaced would in my view be wrong. I would therefore very strongly suggest that the re-entries made on plate 73 were made quite early during its working lifetime and not at the published dates given above by E D Bacon. Perhaps in 1865 as this is the year date of the cancellation on the stamps on cover.

In my view the re-engraving of the number 77 and the head re-entry to correct it was a very early event in the life of plate 73 and one which was quickly corrected leaving the plate to print thousands of sheets thereafter. Ackermann and Osborne confirm that positions SK and RL were amongst the 67 re-entered impressions. We now also know that stamp SL was re-entered in view
of the example on the cover showing a number 77.

As to why engrave a number 77 on some of the heads, one can only guess at an answer. Was it engraved in error, perhaps a misunderstanding between the printers and the plate makers?

There is no doubt that some impressions were worn and the plate number was imperceptible on some. Is it then beyond the bounds of reason that the number 77 was used as a marker for all those heads that needed re-entry? Altering the number by engraving is a quick method for correcting a number and keeping the plate out of commission for a short time only before the roller can be re-applied to correct the heads on the plate at a later date. After all the plate number 77 was not used and is an easy number to engrave quickly on a plate.

There is no doubt a coincidental re-entry to correct this problem is very possible and would give a new impression that would be bold and clear and not show any signs of doubling or otherwise. It is certainly not beyond the bounds of reason that this could have happened. The
siderographers and engravers of that time were more than competent of such an act. In fact anything could have happened to a plate that needed major repairs during its working life and one that has printed over half a million sheets of stamps and was in commission for over four years. We know that 66/67 heads were re-entered and so why is it beyond reason that some heads, for whatever reason, were engraved with the number 77 and then corrected back?

One thing is for certain- We may never know everything.