According to Ince, 86,460 stamps were dispatched between March 31, 1887 and August 4, 1900. Of this number, 26,220 were unsold at the end of 1903. It is not clear whether or not the 86,460 includes the plate 2 printings. My guess is that it does not, but the 26,220 does include remainders from this printing. The relative prevalence of used examples from this last printing relative to mint examples will give us some idea of whether or not this is actually the case. The period covered by these releases is 13 years, or 53 calendar quarters. So, if one assumes that new supplies were sent to Lagos every quarter, the absolute maximum number of printings that could have been sent is 53, and the number of stamps sent each time would be 1,629 on average each time. Of course, there probably were not that many printings, given that this would have been one of the less commonly used stamps at the time. However, it is probable that at least 25 printings were made, and maybe as many as 30 or so.
As with all the other stamps of this series, I will begin by looking at the stamps from the first state of the plate, which shows full detail in the hair at the back of the head, and little to no merging of shading lines in the hair at the top of the head. Then, I will look at the printings from the second state. These printings are very similar to those of the first state, the chief difference being a slight loss of sharpness in the detail of the hair at the back and top of the head. However, there is little to no merging of the hairlines, and all the major detail in the hair, especially at the back of the head is still visible.
One problem that complicates the classification of printings on this stamp is the fact that the green ink is doubly fugitive, and highly susceptible to fading from exposure to moisture. The original colour is usually a dull yellowish green or dull blue green. When this fades, it first turns to blue green, but not dull blue green, then bright yellow green and then to bright greenish yellow. So used stamps in those shades are faded, and are of little to no use in classifying printings. However, as long as the colour retains the overall dull tone, than it should be assumed, unless there is persuasive evidence to the contrary, that the colour is original. some used examples will be found that have gum. These generally have been sweated off the envelopes in order to avoid exposing the ink, and the colour on these should be good unless it has clearly faded.
Group 1: Printings From the First State of the Plate - Printings 1-6 - March 1887-Approximately March 1890
The head plate colour of this printing lacks the greyish undertone of the first two printings above, appearing somewhat yellowish in comparison. It is not however, a yellow green in its own right. It is closest to a deep version of Gibbons's deep dull green. It is quite a bit duller though than the deep dull green shown on the Gibbons colour key.
Here I have one mint example and a used example that has a completely unreadable CDS cancel. This later stamp has full original gum, so it is possible that this is a false cancel that was added posthumously, since the used stamps of this value are generally worth quite a bit more than the mint.
On this printing, the head plate shade is a soft version of the deep dull green. It is brighter and slightly more bluish compared to the shades of the third and fourth printing.
I do not have any mint copies unfortunately which would corroborate the shade. The colour on the second used example on the right appears to be fairly intact, but the stamp is faulty. The first stamp on the left unfortunately is somewhat faded, with the solid areas of colour to the left and right of the medallion being used to match the colour. This first stamp appears to be cancelled with an 8-bar oval, while the second stamp appears to be cancelled with a strike of a 24 mm Lagos CDS.
Group 2: Printings From the Second State of the Plate - Printings 7-15 - Approximately July 1890 to December 1893
The head plate colour of this printing is an almost perfect match to Gibbons deep dull green.
I have two mint examples as shown above, and two used examples. The first of the used stamps matches the shade almost perfectly, and is cancelled with a nice strike of a 21 mm Lagos CDS, with the 4 mm space between W and A, and dated June 27, 1894. This is likely a year or two later than this printing was released, but that is not a particularly late usage, as far as late usages from this issue go, especially given the low volume nature of the use for a stamp of this denomination. The second stamp appears quite yellowish by comparison, but I think the colour is slightly faded, which is why the colour has taken on that yellowish tone. I think that the underlying colour is the same as the other stamps. This one though is a much later use, being cancelled with a 22 mm Lagos CDS dated November 27, 1899.
The head plate colour of this printing is closest to Gibbons's dull blue green, but just a bit deeper.
I have one mint example and one used example as shown above. The used stamp is a very, very late use, being canceled with a 24 mm Lagos CDS dated January 14, 1904. That would suggest the possibility of it being from plate 2, but that possibility is quickly ruled out by the fact that it is from the second state. The plate 2 stamps have a much crisper impression, and so this is simply an extremely late usage of this printing.
The colour of this printing is very close to being a perfect match to Gibbons's dull blue green. It is very, very slightly paler than the Gibbons swatch.
I have three mint examples, as shown above, and one used example. The used example is cancelled with an unreadable CDS, which looks like it might be a 24 mm Lagos CDS. On the other hand, it lacks the side dots and it may simply be another example of a posthumous cancel that was applied to a mint stamp to create a more valuable used stamp. However, it is not a particularly attractive strike, which tends to support the notion that it is a genuine contemporary cancellation.
The head plate colour of this printing is similar to the dull blue green, but is both deeper and brighter. Yet, it is not bright enough to be simple blue green, and it is too bluish to be deep dull green. Therefore, I would call this colour deep dull blue green.
I have one mint example, and a beautiful used example, shown above, which has been cancelled by a superb strike of the 21 mm Lagos CDS, with 3 mm space between W and A, and dated October 7, 1892.
The head plate colour of this printing is closest to Gibbons' deep dull green, perhaps with the slightest hint of blue to the green, but not enough for it to be a blue green shade.
I have three mint examples and two used examples as shown above. The used example on the right has been floated off the envelope of form that it was attached to, and has just begun to be affected by moisture, turning that unmistakable bright blue green colour. The first used example is cancelled with a partial strike of what appears to be a 9-bar oval obliterator.
Before I end, I want to show some examples of what the severely faded used stamps look like. The scan below shows 4 used stamps, all from this state that are severely faded to the point that they cannot be used to identify specific printings: