Captain Charles Wispington Glover CRAWFORD CBE. "Some Considerations Regarding the Flaw Varieties of the Electrotyped Issues of Queensland." London Philatelist, 1920, pp. 191-4, 213-5, 238-40, 263-4
With the approaching publication of a new work, Queensland, by the Royal Philatelic Society, London, considerable attention is now being devoted to the stamps of that country.
The electrotyped issues have never been in much favour with collectors, largely owing to their poor appearance and the comparative ease in getting a representative collection together. However, taken as a group, they present an unrivalled opening for a specialist when we take into account the various papers, watermarks, and perforations, together with the four type varieties and the possibilities of plating by the aid of the flaw varieties.
Some of these flaws are not of the standard of interest to get "mention" in a catalogue, because they do not cause a misspelt inscription or are not so noticeable as to be easily described, and yet they may be just as interesting to philatelists as a QOE or pence "error."
With the exception of articles by Mr. A. F. Basset Hull in Vindin's Philatelic Monthly, Vol. VII, Mr. J. Bornefeld in the Monthly Journal, Vol. XVIII, and Mr. S. Dalby in the P.J.G.B., Vols. XXIII-XXV, very little has been written about these stamps except for occasional chronicles giving the description of some flaw or other.
Before discussing the relative merits of the different flaws met with I will give a brief account of the process of making an electrotype and then give further details as to the probable method adopted in Queensland.
First of all we have a steel die engraved en epargne, i.e. the portions which appear in colour on the stamp are in relief on the die, the uncoloured parts of the stamp being therefore in recess. A mould is then taken by impressing the die on lead, wax, or other suitable material, the mould is then placed in the electrotyping bath and a coating of copper electrically deposited on it. When the copper shell is thick enough to stand the "handling" the mould is removed from the bath, the copper shell is then stripped from the mould and backed with type metal to printing height in the press. It must be remembered that the printing side of the "electro" (copper shell) is that which was in contact with the mould.
The advantage of the lead mould is that it can be used over and over again, and that the lead is more easily retouched to remedy flaws or damage. Obviously, additions cannot be made to the lead mould, but engraving on the mould (if deep enough) would produce a coloured reproduction in the finished stamp. Similarly, engraving on the die or copper electro would produce an uncoloured replica in the finished stamp.
In a letter to the Treasury reproduced in Vindin's Philatelic Monthly, Vol. VII, p. 140, Mr. William Knight, the Government Engraver, who initiated the electrotype process in Queensland, writes:
"Impressions are taken in lead by means of a drop hammer. A sufficient number of these are soldered together to form a part, or the whole of a sheet numbering 120 stamps; this mould is then placed in the battery to receive a deposit of copper, which, when sufficiently thick (taking two or three days), is separated from the lead, backed type high with metal, and is then ready for the press."
However, the means at Mr. Knight's disposal only allowed the use of a small bath and thus the electros were prepared in blocks of four, the electric current being derived from a Grove's or other primary battery.
Thus the first or matrix mould would be prepared after striking four impressions in lead from the steel die. The mould might possibly require some retouching to repair inequalities in the impressions, or perhaps there might be some slight damage, and such marks would be reproduced in all electros manufactured from that mould, and thus give the different type varieties running regularly in each block of four throughout the sheet.
In the later medallion types it has not yet been possible to give reliable marks for distinguishing the type varieties, but in the case of the One Shilling value Type II can be identified in each block of four in the sheet, and I see no reason to suppose that the other values of the medallion types were not also produced from quadruple moulds. The Halfpenny (one figure) and the first and second redrawn types of the Twopence with four figures were produced by a photographic process and do not show type varieties.
The first finished electro from the matrix mould, i.e. the first electro completed with type-metal backing, would be retained as a "quadruple die" for use similarly as the steel die in the manufacture of new moulds when required. This procedure, however, was not adopted in the case of the One Penny and Twopence of the 1879 issue, all values being found with the characteristics of the second mould known as Die II (vide Mr. Bornefeld in the Monthly Journal, Vol. XVIII, pages II and 32). In the later issues there is every reason to believe that the retention of a quadruple die was a regular practice.
Should it be required to produce new values a finished electro would be taken, the value label blocked out as necessary with plaster of Paris or other suitable material, and the resulting mould from this would give electros with value labels blank. New values would be then engraved and the electros used to produce the matrix moulds for their respective values.
The sheets contain one hundred and twenty impressions arranged in twelve horizontal rows of ten, and thus thirty electrotypes (blocks of four) were utilized to make up the forme or printing plate and arranged in six rows of five electros each. About 1897 a larger plant was installed and electrotypes of sixty impressions were utilized, thus effecting a saving of time in the preparation of a plate. Each of these large electros was produced from a quadruple die (electro), fifteen impressions being "struck" to form the large mould. In or about 1899, forty stamp electros were in use for certain values; perhaps the smaller electros were found to be more convenient.
As regards keeping the printing plate in a good printing condition the earlier method with thirty cliches was the best as a worn or defective electrotype could be removed and repaired or replaced by a new one if required. With the large electros, however, it is most probable that faults were allowed to accumulate before it was thought worthwhile to replace the electro, a considerable amount of work being required to replace a single impression. This partly explains why the later printings of the issues on Crown A paper are generally rough and defective.
When a defect was noticed in an electrotype, block of four, the general practice seems to have been not to repair it but to replace it by another, as was the case with the Twopence penge error in the 1879 issue.
When faulty printing was caused by dirt on the plate the forme would be taken apart and the electros taken out, cleaned, and replaced. The electrotype was not always put back in the same position and in the case of the Twopence, 1894 "retouched" mould, I have discovered two such occurrences in the plate.
In 1902 the One Penny plate was damaged and five impressions in the upper row of a sixty stamp electro were cut out and replaced, and in 1906 a single impression was similarly replaced in a large electro of the Twopence, vide P.J.G.B., Vol. XXIII, pages 137, 157. It is most interesting to note that in both these cases the proper order of the type varieties was interfered with and consequently abnormal combinations of the types may be found.
It would seem that retouching of an electro was seldom, if ever, resorted to; I have not seen a specimen of a stamp showing an unmistakable retouch as distinct from a possible flaw.
Two other methods of repairing an electrotype present themselves but, for the reason just given, I very much doubt if either of them was used in Queensland. They both result in adding metal to the damaged portion of the electrotype which can then be engraved or retouched as necessary.
1. The electrotype is coated with a thin layer of paraffin wax except for the part which is damaged. On placing the electro in the electrotyping bath, copper will only be deposited on the part free from wax, and this can be continued until the copper is the requisite thickness.
2. Fresh copper is blown on the electro by means of a blow-pipe [vide footnote on page 106, Vol. I, Postage Stamps in the Making, by F. J. Melville).
We will now consider the preparation of a plate for printing. Having obtained a matrix mould, as already described, electrotypes are manufactured from it, backed to the printing height with type metal, and then fitted in place in the forme, care being taken to preserve the correct alignment and distance apart of the blocks of four impressions.
Obviously the time taken to construct the plate would be halved if two moulds were used so that one mould was in the bath while an electro was being stripped, from the other.
I think this procedure was generally adopted after the first plate of the Twopence, 1879, in which all the electrotypes were from Die I. In the One Penny, 1879, we find the plate built up of electros from two moulds. Dies I and II, which differ by reason of the curved outlines of the network spandrels having been retouched in the electro utilized to make the latter (vide Monthly Journal Vol. XVIII, pages 11 and 12).
Under the best conditions the second mould would be exactly similar to the first, and it is only by damage or retouching that they can be distinguished. We would thus get a certain type variety only occurring in each block of four in the sheet derived from that particular mould.
A very good example of this is the plate of the Halfpenny of 1890, with lined background, where Types II and IV of one of the moulds sustained damage with the result that the defect is repeated in eight electros in the plate, besides which, one electro only shows the flaw in Type II, and two electros show the flaw in Type IV (vide Monthly Journal, Vol. XVIII, page 222). I have found several similar occurrences in the 1894 and 1895 plates of the Twopence value.
It can be well imagined that stripping the copper electrotype from the lead mould is an operation of some nicety, otherwise one or the other will be damaged. The copper, as deposited by the electric current, is of a fine crystalline composition, hard and brittle, and thus is liable to be broken, and if cracked, the crack may increase during use and a portion break away finally. This latter is more noticeable in the dotted frame portions of the design.
Thus, if a portion of the mould is torn away, we get a flaw occurring in all future electros from that mould. On the other hand, if the electro is damaged we get a flaw occurring only once in the sheet. In both the Twopence plates previously mentioned I have found specimens showing the growth of a flaw in a mould. In my investigations (not yet completed) of the 1890 plate of the Twopence I find many pronounced examples of Types I and II with the top frame damaged, which probably illustrates the growth of a flaw in a mould as well as in an electro.
After a mould had been in use for some time it would become worn and be replaced, as would, of course, be the case if it was damaged. These additional moulds would not necessarily differ from the original mould, and in the case of the later issues it is most probable that more than two moulds were in general use or each denomination. As an example we may take the Twopence of 1882, for which three moulds were used (vide Monthly Journal, Vol. XVIII, page 132). In this case Mould C was probably derived from Mould B or vice versa, the distinctive mark in the "w" of "two" in Type III being common to both moulds and neither of them having the "hook" as in Mould A.
In addition to the flaws caused by damage to the plate, or dirt when printing, there is sometimes a flaw noticeable in every stamp in the sheet. Obviously this flaw must have been present in the original steel die or its substitute, a very good example being the white scratch across the network in the S.W. spandrel in all values of the 1879 issue.
The flaws in connection with the printing plates and moulds may be conveniently classified as follows :
I. Uncoloured Flaws.
(a) Caused by dents or other damage to an electro. This produces one constant variety in the sheet during the period of use of that particular electro.
(b) Caused by dirt or broken metal lodging in the mould and so affecting all subsequent electros from that mould. This produces several similar constant varieties in the sheet which may be sometimes distinguished from each other by other flaws which may be present.
II. Coloured Flaws.
(a) Caused by dirt, solid ink, etc., accumulating on the plate until it gets to printing height. This flaw may be expected to develop during the use of the plate and produces one variety in the sheet. It is generally noticed at an early stage and corrected by cleaning the electro.
From the above it will be seen that the probable origin of a flaw can generally be determined. Of course the flaws, I (b) and II (b), only show in the electros made from the mould subsequent to the damage, etc., thus the number of varieties in the sheet would depend on the progress made in building up the plate when the damage occurred. Possibly the last electro manufactured might be the first electro to show the flaw, in which case there would be only one variety in the sheet printed from that plate.
In addition to the above flaws there are those to be attributed to bad printing, varying impressions of the same stamp in the sheet being frequently found. These are largely due to the mixing of the ink, etc., and possibly also to varying pressure in the press or to the electrotype having become worn or worked loose in the forme. Mr. Bornefeld, writing in the Monthly Journal, Vol. XVIII, page 167, says : "It must not be forgotten that each of the defects described above occurs, presumably, on one stamp only on the sheet; but as I have listed no varieties of which I have not found at least five copies, these defects were in the plates and not mere blemishes due to bad printing." My experience is that with the exception of the flaws listed under II (a), Mr. Bornefeld is rather on the safe side and that, generally, three or four copies should be sufficient.
Apart from the above considerations flaws may be grouped in varying degrees of interest as follows:
A. Those which occur regularly in each block of four throughout the sheet. These are the most important as they give the marks for distinguishing the four types and determine the use of a quadruple mould in manufacturing the electros.
B. Those flaws which do not occur regularly in the blocks of four but which have two or more regular positions in the sheet, thus determining the number of large electros composing the plate.
C. Those flaws which occur irregularly and consequently have no significance as regards the method of preparation of the plate. A certain amount of interest, however, attaches to the more permanent flaws, some of which were corrected later. The flaws in this section are all important for the purpose of reconstruction of the sheets.
The life of a flaw, as obtained from dated copies and other considerations, is a subject full of interest, as it determines the period of use of a mould or a particular electro, and in a measure may be evidence of the actual life of the plates themselves.
The most noticeable examples are those of the Sixpence and One Shilling values of the 1882 type which were issued in November, 1882, and February, 1883, and continued in use until replaced by the four-figure types in 1898 and 1899 respectively. These values were derived from Mould A of the 1882 Twopence, and during the period of sixteen years Type III invariably shows the "hook." It is most unlikely that the same moulds were in use during the whole period; we would expect some to be worn out or damaged, but this proves that all the moulds were derived from the original mould for each value. It also illustrates the persistency with which a small flaw in a mould will repeat itself.
Mr. Bornefeld in the Monthly Journal, Vol. XVIII, page 132, considers that the "hook" with other flaws may have been intentional secret marks, as the "hook" “. . . could easily have been removed in the course of the construction of the electrotypes, when so many other alterations were made. ..." However, in view of the large number of flaws present in most values it seems unlikely that extra flaws would be introduced as secret marks.
Similar examples of flaws in the mould are the letters "la " joined in Type II, which were in the One Penny value from May, 1887, to February, 1895, and in the Twopence from May, 1887, until the issue of the Twopence "retouched mould." I have seen an unused pair of the latter endorsed " Approved 16 Oct 1893," my earliest dated copy being 1st January, 1894.
A most interesting illustration is the case of the long and short "p" found in Type III of the 1887 Twopence, Type II always showing the letters "la" joined. The Australian Philatelist, 4th April, 1916, page 123, gives the following note by the Editor:
"... Mr. H. L. White has in his collection a strip consisting of two rows right across the sheet of the 2d. of the 1887-1889 issue, which shows all the No. 2 with 'LA' joined, and one only. No. 3, with 'p' shortened. This is noted by Stanley Gibbons. He has also a quarter sheet of the same stamp showing all the No. 2 with ' LA ' joined, but with 'p' shortened in all places except two. Here we have the strip contradicting the block."
I have a copy of the short "p" postmarked 21st June, 1887, thus showing it was in the first printing issued on May 5th of that year. It would be interesting to know if the 1887 "Proof" showed a short or a long "p." Thus it would seem that in the course of building up the plate one of the moulds (each of which had the "LA" joined in Type II) received damage, and Type III showed the long "p" in all subsequent electros made from it.
The short "p" variety is not known in a block with comb perforation (1890-4); we get it in the 1894 "retouched" mould, but then the letters "la" in Type II have been corrected and no longer touch each other. The question at once arises; what became of the short "P's" if they were not in the plate from 1890-4?
Possibly the first plate, as well as the mould with the short "p," were damaged beyond repair, and a second plate made up from the mould with long "p." As certain flaws are to be found perf. 12 and also perf. 13 we may further assume that this plate with long "p's" throughout was brought into use before the introduction of the comb machine.
Another and certainly the most interesting illustration is the "red triangle" variety which occurs in Type IV of the One Penny, 1887-95. This flaw shows as a triangular red spot on a white ground in the S.E. spandrel and occurs in several positions in the sheet in the later printings.
It was certainly in the plate before comb perforation was introduced, as it is found on specimens perf. 12; I have an unused pair from the bottom row of the sheet showing the flaw in what might be an initial stage, and I have seen another specimen with the flaw complete.
Thus one of the moulds had developed this flaw before 1890, and all subsequent electrotypes from it utilized to replace damaged and worn electros in the plate would show the red triangle. I am unable to state how many of these flaws there were in the sheets perf. 12, but through the courtesy of Mr. J. H. Chapman I am able to give the positions in the printing on burelé paper in February, 1895, which details are from a complete sheet in his possession. This is of the greatest interest as it gives what is most probably the final state of the One Penny plate, the One Penny "Cameo" being issued on the 28th February, 1895.
In the printing on burelé paper the "red triangle" flaw appears at Nos. 12, 16, 20, 54, 56, 72, 80, 94, and 120. In the sheet in question the flaw is not so pronounced in Nos. 80 and 120.
From the foregoing we see that the "red triangle" variety had a life of at least five years.
As an example from the later issues we may take the One Penny with four figures which was in use for upwards of fifteen years, the constancy of the distinctive marks of the four type varieties showing that all the moulds were derived from the same quadruple die.
We will now consider the flaws resulting from damage to a mould or electro in the course of preparing the printing plate. A very interesting example is the Sixpence, 1882 type, in which a flaw may be found in Type II in the oval band between the "d" and the upper part of the right scroll ornament. The flaw consists of white and green concentric circles, the central disc being sometimes green and sometimes white. I have specimens showing four well-defined stages of the growth of this flaw in the mould, together with two more or less indefinite or intermediate stages. The material at my disposal does not allow of these varieties being placed in the sheet, but one variety comes from the right margin, another is the second stamp from the left margin, and a third is from the top row of the sheet.
Evidently this is a case of a damaged mould with the damage increasing as each successive electrotype was stripped from it.
Of the flaw in its latest stage I have dated copies, 13th June, 1889, and 9th October, 1893, perf. 12; and 29th June, 1894, perf. 13, by comb machine. The former and latter dates are from copies from the right margin, which tends to show that the same plate had been in use for at least five years. I also have an earlier stage of the flaw, perf. 12, the postmark date 31st January, 1887, apparently adding another two and a half years to the life of the plate.
Another case of a flaw in a mould is the "loop" variety of the One Penny, four figures, issued on Crown Q paper in August, 1897 (vide P.J.G.B., Vol. XXIII, pages 156, 157). This variety, a loop instead of a minute hook at the lower end of the right scroll ornament, occurred in the twenty-fourth impression of a sixty stamp electro, and thus it showed at Nos. 24 and 84 in the sheet. In 1899 the lower two rows of the electros were cut off and a plate made up of three forty stamp electros was in use for a short time, the "loops" appearing at Nos. 24, 64, and 104 in the sheet. New moulds were constructed from the original "quadruple die" for the issue on Crown A paper in December, 1907, when the loop variety disappeared after showing in the sheets for upwards of ten years.
In the One Penny of the 1887 type there is an interesting variety known as the "Pointed Bust," in which the bust is produced through the white oval as far as the "E" of "one." This flaw, Type III, No. 93 in the sheet, is found in the printing on burelé paper in 1895 and, although I have not been able to trace a copy, it was also in the original plate of the 1887 issue, perf. 12 (vide Monthly Journal, Vol. XVI, page iii), and thus giving that particular electro a life of about eight years.
As regards the Twopence of 1894, retouched mould, I have satisfied myself that only one plate was used for the various printings, many of minor varieties being present during the whole period of use, about eighteen months. The '"FWo" error is found on the provisional papers of 1895 as well as the ordinary Crown paper, and was in the plate for at least a year, my earliest dated copy being 28th June, 1894.
As an example of a flaw in a large electro we will take the "Ear Drop" variety of the One Penny, four figures, which developed in Type IV, No. 94 in the sheet, in 1901 and was corrected in 1904, thus having a life of about three years (vide P.J.G.B., Vol, XXIII, pages 156 and 157).
From the foregoing it would seem that the life of a plate was considerably more than would be expected from copper electros under heavy pressure required for surface printing, and this is accentuated when we remember that postal business was continually on the increase, thus necessitating the printing of much larger quantities of stamps to meet the public demand.
The only information I can find on this subject is given by Mr. Basset Hull in Vindin's Philatelic Monthly, Vol. VII, page 141, as follows:
"Between 1882 and 1889 several new plates of the One Penny and Twopenny were prepared. The following particulars are gathered from proof sheets in the Government Engraver's and Post Offices.
Second block prepared in November, 1882 (same date as first, and similar colour and characteristics).
Third block prepared April 7th, 1884, and spoiled in one month through ink containing mercury; colour, pale red. Period after value.
Fourth block prepared June 14th, 1884, new ink used and colour approved June 17th, 1884; bright and pale vermilion. Period after value.
Fifth (?) block, submitted for approval, 5th May, 1887; colour, vermilion. No period after value, and arabesques as in the Twopence.
Sixth (?) block, submitted for approval, 14th August, 1887; colour, salmon-pink. No period after value, and arabesques as in the Twopence.
Second block, commenced October 16th, 1883; colour, full blue.
Third block, November 17th, 1886; colour, pale blue.
Fourth (?) block, May 5th, 1887; colour, bright blue.
Fifth (?) block, August 14th, 1887; colour, bright ultramarine.
Sixth (?) block, November 25th, 1889; colour, deep blue."
Mr. Bornefeld in the Monthly Journal, Vol. XVIII, pages 165 and 166, considers that as regards the One Penny value, the second "block" was Mould B, which was probably used with Mould A in building up the plate of the One Penny issued on 23rd November, 1882, the third and fourth blocks representing two subsequent plates, the fifth black being undoubtedly the 1887 issue without stop after "penny." Of the Twopence (issued 1st August, 1882) Mr. Bornefeld considers the second and third blocks were the second and third plates made up from Moulds A, B, and C, and that the fourth block was the first plate of the Twopence, 1887 issue, with modified design. From these particulars the life of a plate works out at from two to three years, but we notice only three months' interval between the fifth and sixth blocks of the One Penny, and fourth and fifth blocks of the Twopence.
This, however, does not explain what necessity there was for registering a proof sheet (not a colour trial) from a plate on the occasion of Mould B being prepared unless there were two plates, the "first block" being all Mould A, and the "second block" all Mould B. It is exceedingly unlikely that two plates of the One Penny value should be manufactured for use in November, 1882. Mr. Bornefeld does not say definitely that the first plate contained both Moulds A and B; he thinks that such was the case and states definitely that one plate of the One. Penny was built up from both moulds, and that one plate of the Twopence had electros from Moulds A, B, and C.
If the proof sheets referred to by Mr. Basset Hull are still in existence an examination of the "make-up" as regards the moulds, etc., would be of the greatest value. In the absence of any such information I venture to offer the following explanation regarding them:
It would seem that it all hinges on the use of the word "prepare," which I take it does not necessarily mean that a new plate was manufactured, but that a plate was cleaned and prepared or adjusted in position in the printing press. The resulting proof sheet would then be evidence of the proper state of the plate and at the same time it might be a proof for a new plate, design, or colour. Thus the dates given are really the dates of various printings, and we also notice that although the design was modified yet the "blocks" were numbered consecutively for each value.
We therefore get two printings of the One Penny in November, 1882, the Six Pence was issued in the same month, and printings, of the One Penny and Twopence, modified types, in May and August, 1887.
As regards the "third block" or third printing of the One Penny, the plate was, presumably, spoiled about the 7th May, 1884, and by June 14th a new plate had been made and the fourth printing commenced. It is interesting to note that the interval of thirty-eight days is just about the time required to produce thirty new electrotypes using two moulds.
We can, therefore, come to no definite conclusions as regards the life of the printing plates. In the issues with small electros, blocks of four, it would seem doubtful if an entirely new plate was built up to replace another of the same design. It is probable that faulty or worn electros were replaced from time to time, and this would only be known by the fact of a flaw having disappeared from a sheet, or a new variety introduced, or a change of position of the electros in the plate. Thus we see that an added interest attaches to specimens of flaws with dated postmarks or marginal watermarks. The same remarks apply in a modified extent to the later issues with two or three large electros making up the printing plate.
In December, 1907, on the introduction of Crown A paper, which was double size, we find the One Penny being printed from two plates side by side in the printing press, the sheets being cut in half before being issued to the Post Office (vide P.J.G.B., Vol. XXV, page 22). Mr. Dalby mentions that the Twopence, second redrawn type, and probably the Nine Pence, Commonwealth type, were also printed from two plates, and there is a possibility that this method was applied to other denominations.
In most cases it should be possible to find marks to distinguish the twin plates from each other, and such flaws would be of considerable interest to the specialist.
In 1876, when the introduction of surface printing from electrotypes was being considered, Mr. Knight, the Government Engraver, had proposed printing from two plates at the same time so as to compensate for the reduction of the size of the sheets from 240 to 120 impressions. However, in a letter dated the 9th August, 1878, which is reproduced in Vindin's Philatelic Monthly, Vol. VII, page 72, Mr. Knight says:
"... I have the honour to draw your attention to former correspondence relative to the size of the sheet. In my enclosed letter of the 20th July, 1876, I state in the last paragraph that we shall be able to print two sheets at once; but this, I find from more matured experience, cannot be done in our small press, as much greater power would be required for so large a surface, electrotypes needing greater pressure than type printing. . . ."
Thus the printing of double sheets on Crown A paper was in effect the realization of a proposal made thirty years previously.
In concluding this paper I express the hope that it may arouse further interest in this somewhat intricate subject, and that my conjectures may lead to discussion and new discoveries.
As a result of further investigations into the electrotyped issues of Queensland I have thought it might be of interest to make an addition to my previous notes on the subject which were written nearly a year ago.
I have principally directed my attention to the One Penny, type of 1887, which is found both perf. 12 and with Comb perforation 13, and have almost completed the reconstruction of a sheet showing the state of the printing plate in the period 1892 or earlier to 1894.
This plate is -practically the same as was used for the 1895 provisional printings on the thick and burelé papers after making due allowance for wear of the electrotypes and for the fact that three of the electros (blocks of lour) in the N.E. corner of the sheet were interchanged about the end of 1894, probably in the preparation of the plate for the 1895 printings. During the whole of this period I have been unable to trace the introduction of a new electro into the plate, and thus we may conclude with reasonable certainty that the same electros were in use for the One Penny value from 1892 (or perhaps earlier) until the introduction of the Cameo type in February 1895, and that there were two settings of the electros in the plate.
Working back still further I find a large number of stamps with perforation 12 which can be placed in the sheet, and also evidence of an earlier setting with Comb perforation and therefore in use some time after April (?) 1890, when that machine was introduced. Also it is quite possible, though unlikely, that a few sheets of One Penny and Two Pence values were passed through the "12" machine after the Comb machine came into use. In this respect I have recently seen two blocks of four of the One Penny with perforation 12 and postmarked in 1892. Both blocks fit into my sheet, but it is evident they were old "stock," the flaws not being so fully developed as in the 1892 stage.
On the other hand there are flaws in the One Penny, perf. 12, which do not fit into my sheet and are evidently from electros in an earlier arrangement of the plate and subsequently discarded. Of course, there is always the possibility of a plate having been damaged and replaced by a set of new electros, but this can only be determined by the examination of large blocks which in this particular stamp are exceedingly difficult to find.
I have a block of six One Penny, pert. 12, Nos. 91, 92, l01, 102, 111 and 112 in the sheet, which bear a resemblance to the corresponding numbers in the later plate with two important exceptions: —
No. 91 does not show the square white spot under the "E" of "one."
No. 112 does not show the spot on top of the "E" of "one."
Inferentially No. 93 (vide page 238, par.1), if present, would not have shown the "pointed bust" as it is most probable that all three flaws were produced at the same time as the result of some damage or other to the plate. This is rather borne out by the fact that I am still unable to find or trace a copy of these three varieties with perforation 12. A possible explanation of their scarcity is that the damage to the plate occurred just about the time when Comb Perforation was introduced and that only a few sheets with the "pointed bust" variety, etc., were passed through the "12" machine.
Another variety well known to collectors is the "LA" flaw in which the crossbar of the "A" of "LAND" is defective, broken and sometimes entirely missing. It is by no means scarce and occurs in the One Penny, Type II with "LA" joined, and is found with both perforations on Crown Q paper but does not come on the provisional papers of 1895. It was in the attempt to place this variety in the sheet that it was found to be of an intermittent nature, that is, it always occurs in Type II but not necessarily in the same electro and is not constant. Thus, although interesting, it is of little importance as it is useless for plating purposes. This flaw has been definitely found in some copies of Nos. 82 and 106 in the sheet the following being a general explanation of it: —
Obviously this is a coloured flaw which might be produced by dirt on the plate and removed when the plate was cleaned and so would come under the heading II (a) as given on page 194. Evidently in this case we have a flaw in the mould, probably part of the same damage which produced the "la" joined, and which causes the crossbar of the "A" in all Types II to be represented by an unduly shallow recess in the electro. Thus, a very little dirt or over inking will fill the dent and cause that part of the "A" to be partly or wholly obliterated with colour.
A very similar effect is noticeable in the companion Two Pence, with both perforations, in which the "P" of "pence" in Type II (with "la" joined) is defective and varies considerably as also the "s" in "Queensland." The crossbar of the "A" is mostly normal in this type but sometimes it is thinned much as in the One Penny. These defects were corrected in the Two Pence retouched mould, 1894, in which the lettering is practically normal in all four types.
It is definitely certain that the electros containing Nos. 82 and 106 in the sheet, which occasionally showed the "LA" variety, were present in the plate used for the provisional printings of 1895, neither of which show the flaw. Possibly greater care was taken in these printings, besides which we know the plate received considerable attention, the forme being taken apart towards the end of 1894. I can detect no sign of a retouch of the electros.
We will now consider a subject to which only a brief reference has been made in these notes. I refer to the question of chemical action taking place between mercury salts in the composition of the vermilion ink and the copper electrotype. The result is that the electro suffers damage, the colour of the stamps changing to the yellow and orange shades so frequently noticed in the 1890-95 issues and especially in the provisional prints of the latter year.
This trouble with the ink was first mentioned by the Government Engraver in a letter dated August 9th, 1878, which is reproduced in Vindin's Philatelic Monthly, Vol. VII, page 72, in which Mr. Knight refers to the "... proof sheet of 1d. Postage Stamps just completed and submitted for your approval ..." the 1879 issue, and later on says: "It has been found necessary to alter the colour of the id. stamp in consequence of the material used containing mercury which acts injuriously on copper. ..." Next we find that a plate of the One Penny was actually spoiled about May 1884, vide page 238. Mr. Bornefeld tells us in the Monthly Journal, Vol. XXVIIT, page 186, that the One Penny was appearing in a canary yellow shade about May, 1894.
The late Mr. Samuel Dalby in the P.J.G.B., Vol. XXIII, page 51, states: "... Probably the earlier printings, when the face of the electrotype was fairly clean, would be affected to a greater extent than when the copper had become covered with ink . . ." thus accounting for the various yellow shades met with and explaining why those shades are more pronounced in the earlier printings on thick paper than in the printing on burelé paper. A little farther on Mr. Dalby says: " . . . with the 1896 and subsequent issues the electrotypes were given a thin coating of silver which prevented the copper coming into contact with and decomposing the mercury in the vermilion ink. ..."
Although I have seen no record of it there is a possibility that there was similar trouble with the blue inks for the Two Pence which comes in so many different shades, vide Postage Stamps in the Making, Vol. I, pages 162 and 163 (Melville), where the subject of protective coatings for stamp plates is discussed.
Now the coating of the plate of the 1896 One Penny (figures in lower corners) with silver, though efficient, would not give continued satisfaction owing to the deposit of silver being soft and wearing away quickly, and it is to this fact that the numerous flaws and uneven printing may possibly be attributed.
About this time, 1896, however, we know that the plant for producing postage stamps was considerably improved and brought up-to-date, for the Two Pence, with figures in all four corners and printed from large electros containing sixty impressions, was issued in April, 1897. It seems quite likely that at the same time a nickel plating bath may have been established or "acierage" (coating with steel) resorted to. In each case the plate would have the appearance of having been silvered, which might account for the information given to Mr. Dalby.
The "acierage" and nickelling processes, both described by Mr. Melville, increase the life of the electros very considerably and their use would account for such small flaws as the "loop "variety (vide page 215, last paragraph) showing in the sheets for nearly ten years during which period very large printings of the One Penny value must have been made.
It is quite possible that the use of protective coatings was extended to other denominations besides the One Penny, and if such was the case, and the coatings were allowed to get worn and patchy, we have another reason why the later printings of the issues on Crown A paper are generally rough and defective (vide page 192, last paragraph).
Quite recently I have seen a lower half sheet of the Halfpenny of 1890, with lined background, and find that the information given on pages 193 and 194 is not wholly correct. The defects referred to occur in ten electros in the plate and possibly also in an eleventh. I have also seen two blocks in the later printing in pale green which show that there was a second setting of the electros in the printing plate.