The Control Numbers of the 20 Lepta 1870 Issue of the Large Hermes Heads
by Louis Basel
(This article was originally published in Philotelia, Nos. 536 and 537, in 1989. In 2000, John G. Coundouros published an excellent book The Control Numbers and the Classification of the Stamps of the Large Hermes Heads which covered in extensive detail the control numbers and their errors. Readers should consult this important work for additional information regarding the subject of this article.)
Introduction
Control numbers were printed on the backs of the Large Hermes Head issues printed from 1861 to 1880. These numerals were the same as the denomination printed on the face of the stamps but of a larger size, 6.5 vs. 1.1 mm (0.26 vs. 0.04 inches), and it is believed that their purpose had been to assist the buyers of the stamps in reading the denomination. In the Paris issues, only the 10 lepta had control numbers, but these were of a larger size than those printed on the later issues. In the Athens issues, control numbers were printed only on the 5, 10, 20, 40 and 80 lepta stamps issued before 1881.
The control numbers were printed by means of ordinary printer’s type whose individual characters were arranged on a metal plate and fixed there by metal spacers in such a way as to form 15 rows of 10 numerals each. Spacers of a suitable thickness were used to locate the numerals so that when printed, they would appear in the approximate centers of the backs of the stamps.
It is believed that there was only one plate for the control numbers and that the type had to be replaced whenever a different denomination stamp was to be printed. This frequent removal and replacement of the printer’s type and its use for almost 20 years subjected it to extensive damage and resulted in breaks or indentations in the lines forming the numerals. The control numbers printed from this type when it was new, of course, show very few if any signs of damage, but control numbers printed after about 1864 begin to show an increasing number of distinctive characteristics in both of the digits. This damage increased steadily until 1880 when the use of control numbers was discontinued.
In early studies, Groom showed that:
It became obvious that there was only one set of 150 zeros, possibly with a few extra for use in the event of loss or extensive damage, and that the same zeros were used when printing the 10, 20, 40, and 80 lepta stamps. Because the same zeros were found in the same plate positions on different denomination stamps, it was evident that the zeros were left in the same position in the plate when changing between those denominations requiring a zero. Naturally, however, whenever the printers started to print 5 lepta control numbers, the zeros had to be removed, and, after the printing of the 5 lepta was completed, the zeros were placed in new locations for the printing of the next required denomination. In his pioneering studies, Groom identified many of these characteristic numerals and their plate positions for several of the Large Head issues.
Bellas, in his monumental study of the 20 lepta, displayed six charts of the control number compositions for specific issues between April 1867 and July 1872 and he included images of 80 “twos” and 80 “zeros”. Coundouros and Stassinopoulos in 1986 presented a study of the “zeros” of various denominations and increased the number of “zeros” identified to 86.
Description of the Present Study
In this study, one particular issue of the 20 lepta was chosen as the basis for studying the varieties of the control numbers, namely the Special Printing of 1870, also known as the German Workers Issue. This issue was chosen because it was printed in the middle of the period that the control numbers were used, when the printer’s type had already sustained considerable damage, and, thus, more of the 150 sets of digits could be identified.Another important reason for selection of the 1870 issue is that the stamps themselves are easily identifiable, unlike many of the other issues of the Large Heads whose classification is problematical. The 1870 issue used the so?called “hard” printing method in which a thin “découpage” was placed under the sheet on which the stamps were to be printed. This departed from the method previously used in the Athens printings from 1862 in which a thick felt “blanket” (the so?called “soft” method) was placed under the sheet to be printed. The “hard” method had been used very successfully in the first Paris printings, but it required careful attention to the alignment of the découpage, and in the 1870 issue, the printers were not successful and, generally, very bad impressions of the 1870 stamps resulted.
The characteristic signs of this issue are: whitened corners (the four roughly triangular sections containing the wavy lines and dots between the inner frame and the circle of pearls); whitish shadows on the blue portions of the engraving; and a generally light impression of the entire engraving causing breaks in many of the lines which are normally solid. Because of these many characteristics, there is little difficulty in identifying this issue and we can be certain that all of the identified control numbers belong to the same composition of the control number plate.
The author recently reconstructed the complete plate of the 20 lepta issue of 1870, thanks to the kind assistance of a few friends who supplied some missing positions. The reconstructed sheets of the stamps and their control numbers were displayed at the Collectors Club one?frame exhibit at Interpex in New York City in March 1989. Enlarged images, approximately 1.0 by 1.4 inches, of each set of numerals “20” were also exhibited for each of the 150 positions of the sheet.
Plating of the 20 lepta Large Heads is greatly facilitated by the excellent work of Bellas who showed enlarged pictures of stamps from each plate position with their characteristic ink spots, scratches, dents or breaks in the frame line, broken or bent shading lines, white spots in the blue background, etc. However, even with the help of the Bellas work, plating of the 1870 issue 20 lepta stamps is sometimes quite difficult because of the preceding cleaning of the printing plate in 1868 and 1869, and the resulting lack of many ink spots which are so helpful in determining plate position. Also, the 1870 issue, having been badly printed using the so?called hard method, as described above, exhibits many white spots in the impression which tend to hide the normally visible dents, breaks, scratches, etc. which aid in the plating. Fortunately, the author was able to use a previously described computer?based program for determining plate positions and to successfully plate all of about 1000 of these 1870 issue stamps.
The computer was useful in another aspect of this study: the photographing, enlargement, identification and printing of the control number images. A black/white video camera connected to a computer was used to snap images of the control numbers of each plate position. The computer digitized these images and displayed them on a computer monitor. A macro lens, attached to the camera, provided greatly enlarged images for ease in viewing on the monitor. With instantaneous display of the snapped image, it was possible to adjust the contrast and brightness settings to obtain clear images. This was necessary because some of the control numbers are so lightly printed that it would have been very difficult and time consuming to use conventional photographic techniques to produce suitable prints for visual examination. With the computer system, it was possible to obtain quickly a clear image on the screen. A graphics program was used which provided an “eraser” to clean the image on the monitor when necessary because of postmarks which showed partially through the back of the stamps or other extraneous markings which interfered with the image of the control number. This program also allowed the addition of small arrows to the image to point out the various characteristic signs of the numerals. The image was then printed on a dot matrix printer with a resolution of 180 by 180 dots per inch. The prints produced in this manner display the various identification marks quite clearly.
Results of This Work
Enlarged prints of the controls of each of the 150 positions are presented in this article. The control number images are presented in groups of 10 to fit them on the pages of this work. They are arranged in a format which corresponds to the layout of the sheet of stamps, i.e. 10 vertical columns and 15 horizontal rows. The images are presented with position 1 on the left and position 10 on the right. This was done so that the numerals would be in the same relative location as the stamp when viewing the face. However, it is recognized that a reconstituted sheet of stamps would have position 1 on the right and position 10 on the left when viewing the back of the sheet.
It is estimated that of the 150 numerals, as many as 125 of the “zeros” and 100 of the “twos” have identifiable characteristics, considerably increasing the number previously known. Furthermore, when looking at the two numerals together, only about ten of the 150 pairs lack any characteristic signs on either the “2” or the “0”.
Also important is the fact that by studying all of the numerals of a single issue, it is possible to distinguish between numerals that display very similar but different characteristics; in past studies this was not possible and there has been some incorrect identification of numerals.
In the 1870 issue of the 20 lepta, two well?known errors are found in position 21, the control number “02”, one with a normal zero and the other with the zero inverted. In both of these cases, the zero has characteristic breaks in the middle of three of the four vertical lines which form the zero. In this study, the normal “20” for position 21 was discovered to have the same zero as found in the two examples of controls “02”. Photocopies and enlarged images of all three of these control numbers are presented in Figure 1.
Figure 1. Inverted and normal controls of position 21.
Another previously described error of the 1870 controls is that of the inverted “20” caused by the insertion of an inverted sheet into the printing press. However, it is generally accepted, although not certain, that the control numbers were printed first and then the stamps were printed on the other side of the sheet. If this is indeed the case, then the inversion of the sheet took place during the printing of the stamps and not during the printing of the control numbers. The author has in his collection five examples of this error on positions 32, 41, 74, 110 and 135. Photocopies of the faces and the backs of these stamps are presented in Figure 2. All of these stamps with inverted controls are postmarked with lozenge or circular date stamps of Patras, which indicates that one or more sheets with inverted controls were sold from that post office.
Figure 2. Inverted controls of the 1870 issue.
It is interesting to consider the relative location of each of these inverted numerals on the backs of their respective stamps as a possible clue regarding the number of sheets that bore inverted numerals. As discussed above, the printer’s type for the numerals had to be spaced in the control number plate in such a manner so that the numerals would appear in the approximate centers of the backs of the stamps. If a numeral was misplaced in one direction on a stamp from one position, then it would be misplaced in the same direction on all of the other positions of the same sheet. Observing the photocopies in Figure 2, it will be seen that the numerals of positions 32 and 41 are both located in the lower left corners of the backs of the stamps, with a portion of the numerals cut off. These two stamps also have the same sky blue color which indicates that they probably came from the same sheet. Looking at the numerals of positions 74 and 135, it is observed that the numerals, although shifted in the same direction as the previous two stamps, their displacement is not as extreme and they appear to be very similarly located; both of these stamps have a bright Sevres blue color. In the author's opinion, these two stamps came from the same sheet but one that was different than that of the first pair. In the last example, position 110, the numerals are misplaced slightly above and to the left of the center. This example is definitely from a different sheet than the other four stamps with inverted controls. We can conclude that there were at least 2 sheets with inverted controls and probably 3 such sheets which were circulated from the Patras Post Office. Coundouros, in his extensive and excellent study of the control numbers, shows several examples of inverted controls which appear to be in different locations on the backs of the stamps. Also, he states that stamps with inverted controls exist with postmarks of Kerkyra, Aigion, Pyrgos and Patras, with the latter having the most examples. A careful study of all of the known examples of this error to determine their colors, plate positions, and relative locations on the back of the stamps would shed more light on this subject.
As explained by Coundouros, if the inverted control number were caused by an entire sheet being inverted in the press, then the characteristic “20” of the inverted numeral must belong to the position number calculated by subtracting the position number of the face of the stamp from 151. Thus, for example, the inverted number of position 32 should have the characteristics of the numerals of position 119 (151 ? 32 = 119), which is in fact the case. The inverted number of position 41 is badly misplaced to the upper right corner of the stamp and the top and right sides of the numeral are cut off. For this reason it is not possible to determine with certainty that the inverted numeral of position 41 is that of position 110, but there are no characteristic signs that negate this possibility. For the other inverted numerals of positions 74, 110 and 135, it is found that their numerals are indeed the normal numerals of positions 77, 41, and 16 respectively, as expected. Thus, we can say with a certain degree of certainty that the cause of these inverted numerals was the inverted insertion of the sheet into the press and not to individual instances of printers type of one or more positions being placed inverted into the printing plate.
The controls of the 1870 issue 20 lepta display many of the usual errors found on other Large Head issues such as misplaced numerals, in some cases so extreme that a portion of the numeral is cut off and sometimes a small portion of the next numeral from the stamp above or below appears on the opposite edge. These as well as other errors, such as: double impressions, the “2” and the “0” on different levels or spaced further apart than normal, and combinations of these errors with the inverted “20” are all described in the study of Coundouros.
Significance of the Control Numbers
The usefulness of the control numbers in the classification of the Large Hermes Head issues has been adequately discussed by many authors.,,, The determination of the specific issue to which an unknown stamp belongs is facilitated by the heaviness of the impression of the numerals, their color, the relative thickness of the left and right sides of each numeral, the number of ink spots, etc. However, even with the aid of the well?described general characteristics of the engraving and the controls, there are often uncertainties as to the exact classification of the Large Head issues. This situation can be greatly improved by the plating of a stamp whose classification is in doubt and by then comparing its control number characteristics with previously determined settings of the controls for the same position in each of the Large Head issues. Once a match is found, then the issue to which the stamp belongs is immediately known.
Another interesting aspect of this method is that the “zeros” of stamps of one denomination can be compared with the “zeros” of other denominations, since it is known that the “zeros” were not removed from the plate when changing denominations between the 10, 20, 40 and 80 lepta. Thus, for example, if a match is found between an unknown 10 lepta “zero” and the “zero” of a known 20 lepta in the same plate position, it can be concluded that the 10 lepta stamp belongs to an issue immediately before or immediately after that of the 20 lepta. This method requires: 1) plating characteristics for each of the denominations with controls (similar to the work of Bellas), and 2) control number settings for each of the issues of each denomination. Obviously, the method is not applicable to those issues without control numbers, but the majority of the Large Head stamps can be checked in this manner. It is hoped that other workers in this field will prepare similar control number settings for other issues and other denominations of the Large Head issues. The author wishes to especially thank John Coundouros and George Trefonas for furnishing stamps of some of the missing positions and thus allowing the reconstruction of the full sheet of the 1870 issue.
