Over the past six weeks I have attempted to sort all of the 4d lilac and black Queen Victoria keyplate stamps into the various printings that were made between March 1887 and August 1901. In all, I identified 64 different groups of stamps, which was 3 more than the absolute maximum number of printings that could have made if one printing was made and dispatched every quarter.
The postage rates were such in the colony throughout the period that the 4d value would have been in high demand the entire time, for one reason or another. Thus it is curious that out of 258,540 stamps 107,880 would remain unsold at the beginning of 1905. The only plausible explanation for this is that sometime after 1900, the procedure that the Crown Agents followed for supplying stamps to the colony was altered, to send a much larger quantity of stamps less frequently to the colony rather than having many, many small printings every quarter. This would appear to be borne out by what actually did happen with the halfpenny and 1d stamps. These two values each had over 45 printings up to 1901 numbering between approximately 830,000 and 730,000 stamps respectively. The 1901 and 1902 printings of these two values, which were the last ones sent each numbered 219,780 and 375,900 stamps respectively - in other words 25-30% of what the total number had been for all printings combined up to that point. Thus it is quite probable that of the 107,880 unsold remainders half or more may have come from the last few printings, while the rest would have been the unsold remainders of earlier printings. Indeed, the official records do show that the total number of stamps sent to the colony up to September 1893 totaled only 108,900 stamps - approximately 42% of the total printing.
It is likely not possible to assign each printing identified to a definitive order within the entire group, but it is possible to assign the printings to groups that can be arranged into a logical order, based on the following characteristics already examined:
The postage rates were such in the colony throughout the period that the 4d value would have been in high demand the entire time, for one reason or another. Thus it is curious that out of 258,540 stamps 107,880 would remain unsold at the beginning of 1905. The only plausible explanation for this is that sometime after 1900, the procedure that the Crown Agents followed for supplying stamps to the colony was altered, to send a much larger quantity of stamps less frequently to the colony rather than having many, many small printings every quarter. This would appear to be borne out by what actually did happen with the halfpenny and 1d stamps. These two values each had over 45 printings up to 1901 numbering between approximately 830,000 and 730,000 stamps respectively. The 1901 and 1902 printings of these two values, which were the last ones sent each numbered 219,780 and 375,900 stamps respectively - in other words 25-30% of what the total number had been for all printings combined up to that point. Thus it is quite probable that of the 107,880 unsold remainders half or more may have come from the last few printings, while the rest would have been the unsold remainders of earlier printings. Indeed, the official records do show that the total number of stamps sent to the colony up to September 1893 totaled only 108,900 stamps - approximately 42% of the total printing.
It is likely not possible to assign each printing identified to a definitive order within the entire group, but it is possible to assign the printings to groups that can be arranged into a logical order, based on the following characteristics already examined:
- The state of the plate.
- The shades.
- Cancellations.
- Whether or not the printing exists surcharged.
Thus, just because I identified a stamp as being say, the 10th printing, does not mean that it was actually the 10th printing sent. Rather, it was the 10th printing that I identified. So how do I go about coming up with an approximate ordering of the printings of this stamp?
Logically, we need to start by considering each of the above characteristics.
The State of The Plate
Logically, the printings followed a progression which mirrored the state of wear on the plate, with the earliest printings showing all the fine detail of the first state, and the very last printings showing the very blurred detail of the fifth state. A useful point of demarcation in terms of dates is to look at the surcharged stamps because those were issued between August 1893 and approximately April 1894. By looking at the states of the plate on these stamps we can get an idea of what the state of the plate was in 1893, as some of these stamps do not appear to have a corresponding un-overprinted example, which suggests that they are from a printing made close to 1893 that was predominately all surcharged.
Of the total number of printings identified, 37 were identified as existing with the surcharge, with the most common ones being the 9th and 18th printings. There were approximately 12 printings from the first and second states of the plate that I did not have a corresponding non-surcharged stamp for. Out of these 12, three were from the third or very early fourth state of the plate: printings 53, 62 and 64.
The printings identified on the surcharged stamps were as follows:
States 1 and 2: 1, 2, 5, 6, 9, 10, 14, 54, 55, 56, 57, 58, 59, 60, 61, 63.
State 3: 17, 18, 23, 26, 27, 29, 31, 33, 35, 36, 39, 53, 62.
State 4: 40, 48, 49, 64
State 5: 52
It is unlikely that the plate would have reached the advanced stages of wear characterized by states 4 and 5 by 1893, and that what appears to be advanced plate wear on these stamps may simply be the result of over-inking on certain stamps in the plate. The vast majority of the surcharged stamps come from the first two states, and then the third state. As was explained, the surcharged stamps would have been gathered up from existing stocks at the time, which would have consisted of a mixture of early and later printings. To understand this, and the predominance of late usages on this issue, one has to consider that the stamps were sold on a last-in-first-out (LIFO) basis, which is to say that the replenishment stocks would simply be piled on top of the old unsold stamps, and the next stamps would be sold from the top of the pile. So unless the supplies were completely exhausted while awaiting the next shipment, this would guarantee that there would always be some early printings remaining on hand, sometimes several years after the fact. This explains why so many of the surcharged stamps come from the early states of the plate.
So what this suggests is that nearly all examples of the fourth and fifth states can safely be reckoned as coming from after April 1894, i.e. printings 36 through 52.
The Shades
All other factors being equal, we would expect the shades to follow a gradual progression, so that stamps of a similar, but different shade, should be from printings that would have been made in close proximity to one another. There has been quite a range of shades ranging from deep dull purples, reddish lilacs, bluish lilacs and grey lilacs.
It is worth noting that the deepest purples and palest lilacs do not seem to occur very frequently among the surcharged stamps, which does suggest that where these extreme shades come from the first, or second states of the plate, that they are made well before those printings that are found surcharged. Likewise any extreme shades found from the third state are likely to have come from printings made after 1893.
Cancellations
The cancellations on these stamps are of somewhat limited use for assigning printings with certainty. The main reason has to do with the amount of overlap that exists with respect to dates, where several types of postmark were in use concurrently, and also due to the fact that late usages are common due to the (LIFO) flow of the stamps in the postal system. Nonetheless, general overall patterns in the types of postmarks that predominate certain groups of printings can give some clues about the ordering of printings.
The types of postmarks generally seen on these stamps are:
- An 8-bar barred oval grid measuring 24 x 19 mm that was in use from July 1880 to April 1899.
- A 9-bar barred oval grid measuring either 26.5 x 20 mm or 24 x 19 mm, that was in use from May 1887 until July 1897.
- A Lagos CDS that is 21 mm wide where the "L" of "Lagos" and "W"of "W. Africa" are 4 mm apart, which was in use from July 1887 until November 1895.
- A similar Lagos CDS in which the gap between the "L" and "W" is 3 mm, that was in use from August 1891 to July 1896.
- A larger 24 mm wide Lagos CDS with a dot at each side of the circle, that was in use from March 1897 until December 1904.
If we look at the surcharged stamps, the most common cancellations by far are CDS's, followed by the 9-bar oval. All of the CDS's that I looked at were the first type with the 4 mm gap. I only had one example with the 3 mm gap. I did not see any 8-bar oval obliterators used to cancel the surcharged stamps. Thus, it would appear that even though the 8-bar oval can be found used up to April 1899, its use should be very infrequent around 1893.
If we look at the printings identified as the first state, most used examples are cancelled with an 8-bar oval, and there are a few of the 9-bar types of the larger size used at Ibadan. There is one single example with the 21 mm CDS cancel dated June 2, 1890.
In the second state, we find a mixture of 8-bar and 9-bar ovals with the 9-bar ovals being more prevalent. There were three examples of the large 24 mm Lagos CDS's, which were likely late usages.
In the third state we find the largest range of cancels. The most common still appears to be the 9-bar oval, through there are a few 8-bar ovals. There is one example of the 21 mm CDS with the 3 mm gap dated October 7, 1892 and an example of the 4 mm gap, dated March 16, 1894.
In the fourth state, all the cancels except for one example of the 24 mm Lagos CDS cancel are the 9-bar ovals. There are no 8-bar ovals, which is consistent with what we would expect, given their absence on the surcharged issue. But there are also no examples of the 21 mm CDS.
In the fifth state, I only have one used example, and it is an Abeokuta CDS that is similar to the 24 mm Lagos CDS, and dated August 21, 1899.
All of this suggests that:
- The 8-bar oval cancels were the earliest, and were all but gone by 1893.
- The stamps of the fourth state are likely to have been printed after July 1896. Those with the 9-bar type are likely printed before July 1897.
- Within the third state, the stamps with the 8-bar ovals will be earlier than the 9-bar ovals, and the earliest printings will be those with the 21 mm Lagos CDS.
Based on all of these considerations, the printings can be re-ordered as follows:
- Non-surcharged stamps of the first state, printed in the deeper reddish lilac shades and the deep, dull purple shades, starting with those cancelled with 8-bar oval obliterators and ending with the CDS cancelled examples.
- Surcharged and non-surcharged stamps of the first state printed in the reddish lilac shades.
- Surcharged and non-surcharged stamps of the second state printed in the dull purple shades, starting with the 8-bar cancels and ending with the 9-bar cancels.
- Surcharged and non-surcharged stamps of the second state printed in the paler reddish lilac shades.
- Surcharged and non-surcharged stamps of the third state, printed in the deeper shades first, followed by the paler shades, and then, by cancellation type.
- The stamps of the fourth state for which I have used examples first, and then all the mint, arranged with the deeper shades first and paler shades last.
- The stamps of the fifth state, with the paler and greyer shades of lilac being issued first, and the redder shades being issued later.
So this concludes my discussion of this value. So far, I have covered all values of this set up to the 5d, the 7.5d, and 10d. The next value logically is the 6d, which I will begin examining next week. After this the 1/- and then the three high values: the 2/6d, 5/- and 10/-. After this, I will be ready to circle back and begin tacking the most difficult values: the 1/2 and 1d.