Telegraph stamps of Great Britain. Until recently, only the Post Office telegraphs were well known. This is changing but the printed information is still out of date. I hope to gather further information and present it here. This site has been expanding. Click here for a site-map.

Back to HOME
General Telegraph	Electric Telegraph	Submarine	British Telegraph	English & Irish	British & Irish	London District	UK Electric	S.E.R.	L.C. & D.R.
Bonelli's	Universal Tel.	National Telephone	Army Telegraphs 1	Army Telegraphs 2	Railway	Post Office	Unusual	Unexpected	Contributions

The London District Telegraph Company (Limited).

Established 1859.

The original page became very large, so I have split it into two parts.
This part contains my evidence for claims made in the other part that are
either not given or disagreed with by earlier writers.

The first set had the values 3d, 4d and 6d, Perf.12½.



Steven Roberts also says on How The Companies Worked (about two thirds of the way down)
"The London District Telegraph Company sold its 6d message stamps bound into small books with ten pages of six stamps in 1861."
According to Steve Hiscocks, these were reported to have been issued in booklets of 10 panes of six stamps, but he didn't give any dates.

So far, I have 115 scans of the following:

* this is apparently a proof. Evidence suggests that these were made some time after production started to test surface coloured rather than coloured paper production.

I have more scans of the 6d than the other two together, so I started with those.

6d  -  sheets of 40, 5 rows of 8 stamps per row.

Consecutive serial numbers are usually vertical pairs numbered downwards:

   
Three digitally reconstructed vertical pairs, the left pair (A1967/A1968 are the lowest numbers I have seen) - courtesy of Steve Lawrie.
The next pair, A2041/A2042 look like they fitted, though with the bottom having additional trimming - courtesy of Steve Lawrie.
The last pair, A2114 of mine and A2115, source unknown, look like they fit.

This pair additionally has A3548 underneath in a horizontal pair with A3543 (see below). - A3546 courtesy of Martien Blank and A3547 courtesy of the late Mike Holt.	This pair seems to fit. In addition they are the only apparently used ones that I have seen. - A4418 courtesy of Steve Lawrie and A4419 courtesy of Andrew Higson.	This pair, A2210 and A2211, clearly does not fit, indicating that the stamps are from different vertical columns - both courtesy of Steve Lawrie.

Horizontal pairs have serial numbers 5 apart, increasing right to left:

These two look like they were once joined. - A2059 and A2064 Courtesy of Steve Lawrie.	These two also, A1979 Courtesy of Steve Lawrie and A1984 courtesy of Derek Harborne.

This 6d pair of stamps indicate that the numbering system is vertical with 5 rows of stamps on the sheet. A3543 and A3548 ourtesy of Andrew Higson.

Until recently, these were the highest controls I had seen for the 6d. Again these indicate 5 rows of stamps on the sheet. The selvedge on the right indicates a sheet edge. A7404 and A7409 courtesy of Steve Lawrie.

Next I need to know how many stamps there were in a sheet.

Sources of flaws.

There is an excellent description of stamp production using lithography at stampsofvictoria.com

Basically, for high-quality work, a steel die was engraved and used to make a transfer stone of secondary images,
The transfer stone was then applied multiple times to create the printing stone.

For low quality work, images could be drawn directly onto a stone with a water-repelling substance.

Flaws could arise at each step of the process :

In addition, the printing stone did not last forever. With use, the impressions degraded, altering existing flaws and perhaps adding new ones.
ultimately the surface of the limestone had to be ground down and new impressions laid down from the transfer stone.

Flaws can comprise either marks where they were not intended, of an absence of ink. Sometimes flaws were repaired by 'retouching' by hand.

Distinguishing between these layers of flaws necessarily entails many images and more would be welcome.

Pairs with matching flaws have serial numbers a multiple of 40 apart:

A1991 is mine and A2111 (C2R1) (120 apart) courtesy of Steve Lawrie.


A2114 is mine, A2194 (C2R4) (80 apart) and A6274 (C2R4) (another 4080 apart) courtesy of Steve Lawrie.


A2108 courtesy of Steve Lawrie and A3548 (C3R3) (1440 apart) courtesy of Andrew Higson.

A2341 and A6261 (C4R1) (160, 3920 (98x40) apart) courtesy of Steve Lawrie, together with A2181 (one of mine).

A2064 and A5304 (C4R4) (1760, 1480 apart) courtesy of Steve Lawrie, together with A1984 courtesy of Derek Harborne, and A3544, postally cancelled at Totton (mine).


A2017, A2057 and A2217 (C5R2) (200, 40 apart) courtesy of Steve Lawrie.


A2338 courtesy of Martien Blank and A4418 (C5R3) (2080 apart) courtesy of Steve Lawrie.


A1979 and A2059 (C5R4) courtesy of Steve Lawrie, A4419 (2360, 80 apart) courtesy of Andrew Higson. A2099 shown below also matches.


A2211 and A2371 (C6R1) (160 apart) courtesy of Steve Lawrie.


A2174 and A2414 (C6R4) (240 apart) courtesy of Steve Lawrie.


A2006 Courtesy of Steve Lawrie and A2166 (C7R1) (160 apart) courtesy of Mark Gibson.


A1968 Courtesy of Steve Lawrie and A6248 (C7R3) (4280 apart) courtesy of Martien Blank.


A2122 courtesy of capital-collectables on eBay, with A2042 and A4722 (C8R2) (2600, 80 apart) courtesy of Steve Lawrie.
Note straight edges on the right of 2 of these. A2122 remains puzzling.



A7404 Courtesy of Steve Lawrie and A6244 (C8R4) (1160 apart) , now mine Ex-Andrew Higson.

Pairs with serial numbers 20 apart do not have matching flaws.

A1993 (C2R3) - courtesy of  Mike Jackson.	A2013 (C6R3) - courtesy of Martien Blank.

Other non-matching pairs are 2115-2135, 2174-2194 and 5304-6244 which is 940 apart (47x20).

A2200 (C1R5) - This is a recent acquisition and was the first example I had seen from Column 1. I show it here as column 1 is important in the following analysis. The selvedge proves it is from the left edge of the sheet. One of the problems with surface coating is that it is susceptible to damage by water or solvents. This has no gum, has some surface abrasion and a closed tear at the top. It has survived though, where apparently most did not.

6d Scenarios.

If we assume that each matching pair is from the same position on different sheets, then the difference between their
control numbers (assuming consistent numbering) must be a multiple of the number of stamps on the sheet.

The next step was to look at these differences:

It would seem that regardless of how they were sold to the public, they were printed and consistently numbered in sheets of 40 stamps.
From the horizontal pairs, we can see this was in 5 rows of 8 stamps per row.

This gave me the characteristic flaws of 14 out of the 40 stamps, or 35% of them.

To me the biggest difference was whether or not the stamps were included in booklets (I stand to be corrected).
If we assume that column-1 stamps are rare because they were not included in booklets, then other stamps not included in booklets would have equal scarcity.
If it was decided to have only one kind of booklet and to secure the stamps by the selvedge, then the most likely ones they chose would be these two.

Of these, the horizontal format is the most likely, otherwise it would be difficult to account for the number of row 1 and 2 stamps but no column 1 stamps.
With the horizontal format, there is still the problem of accounting for all the row-3 stamps we have with no column-1 stamps.
Another possibility is to have two types of booklet, one with a horizontal format, the other vertical:

Together these give 5 horizontal panes and 5 vertical panes from two sheets leaving 20 (including column 1) leftover stamps for sale over the counter.
An alternative would be these (or mirror images):

It still requires an explanation as to what happened to the column-1 stamps ?
However panes of 6 are selected, it leaves some loose stamps that still survived, why not column 1.

I have scans of only two intact pairs. There is no way to know if they were ever in booklets, but if the column-1 stamps are scarce
because they were not in booklets, then that implies that most of the stamps I have scans of, were from booklets.

One other constraint is that the layout has to account for the existing still-joined pairs.. The pairs are 3543/3548 and 7404/7409. These were from R3-C3/4 and R4-C7/8 The layout also has to account for the known stamps with selvedge. Selvedge is known only on the left and right sides of stamps, not top or bottom, however they could have been perforated with it later being torn off. The likely scenarios we were left with included no stamps with selvedge at left or right, basically because column 1 was not included. This makes it clear that the lack of column-1 stamps makes it unlikely that stamps with selvedge on the left or right sides were from booklets. Though most stamps may have been from booklets, stamps with selvedge may have been preferentially collected. (Sorry for the vagueness, but the lack of concrete evidence causes recourse to statistics with associated get-out clauses ;o) The scenario I suggest, implies one pair from a booklet and one from sheet. It will be very interesting to see further pairs.

A2122 courtesy of capital-collectables on ebay This is from column 8, Row 2, along with A2042 and A4722 having matching flaws. The last two have the expected straight edges on the right, as do the others. Why doesn't this ? The perforations have every appearance of being genuine. Though the sheets were all printed the same, they seem not to have all been perforated the same.

A possibility that has to be considered is that the statement "reported to have been issued in booklets of 10 panes of six stamps"
was in error and should have been "reported to have been issued in booklets of 6 panes of ten stamps".
This would certainly be easier to manage with sheets of 8 x 5 stamps, but there is still the problem of the scarcity of column-1 stamps.

If both left and right sides were horizontally printed through the selvedge, but with no perforations vertically between the last stamp and sheet edge,
it may have been convenient to remove left and right columns and just use the middle 3 blocks of 10 stamps for booklets.
There is still no explanation though as to why column 8 has survived in preference to column-1.
Perhaps Column-1 was given to counter staff and did not generally get bought by the public to take away, whilst column-8 stamps were used as samples by salesmen.

Knowing the pattern makes it possible to predict which should match.

A6272 (C2R2) now one of mine.	A5312 (C2R2) courtesy of Paul & Les Bottomley.	A2032 (C2R2) courtesy of Auktionshaus Christoph Gärtner.

These three are from the same position. Difficult to see if the last matches.




The system of numbering told me that these two formed a pair.
A5317, courtesy of Steve Lawrie, from the rare left side of the sheet, unused, digitally reunited with the used A5312 from above.
Used are almost as scarce as left side of sheet.

These two match well.

A2387 (C3R2) - A new one of mine	A3547 (C3R2) courtesy of the late Mike Holt.

So do these.

A2146 (C3R1) Courtesy of capital-collectables	A2226 (C3R1) Courtesy of Steve Lawrie.	A3546 (C3R1) Courtesy of Martien Blank.

Two more that should match.

A2135 (C6R5) - Anonymous.	A9855 (C6R5) the highest number I have seen. courtesy of Auktionshaus Christoph Gärtner.

Not very convincing, but likely.

A2229 (C3R4) - From Langmead & Huggins book Fig.33 Courtesy of the Great Britain Philatelic Society.	A6269 (C3R4)

This is a good match.

A2435 (C2R5)	A2115 (C2R5) - Anonymous

The right-hand image is rather poor, but some features match.

A2065 (C4R5) capital-collectables on eBay.	A2425 (C4R5) courtesy of Mark Gibson. This should match the last one, A2345.	A2345 (C4R5) courtesy of Edward Coombes.

These have a rather conspicuous flaw to give them a match.

A2092 (C6R2) one of mine.	A8852 (C6R2) courtesy of Steve Lawrie.

These have a rather conspicuous flaw to give them a match.

There are 5 other scans of 6d stamps not yet shown, these are shown below.

A2023 (C4R3) courtesy of Steve Lawrie.	A2102 (C4R2) courtesy of Steve Lawrie.	A4110 (C3R5) courtesy of Steve Lawrie.

A2099 (a fourth C5R4) one of mine.	A2020 (C5R5) courtesy of Andrew Higson.

4d  -  sheets of 42 (6 rows of 7).

To work that out from only 12 scans involved a lot of luck. (Now I have 17 scans)

Consecutive serial numbers are usually vertical pairs:

These seems to fit : A2464 (courtesy of Steve Lawrie) and A2465 (courtesy of Martien Blank) and A2466 (courtesy of Andrew Higson). They are the only vertically connecting stamps so far.

Despite only having 12 scans of different 4d examples, I was lucky enough to find two matches.

A2466(courtesy of Andrew Higson) and A2340(courtesy of Mark Gibson) are 126 apart. this is (42 x 3), (21 x 6), or (7 x 18).
However, 18 stamps per sheet would put A2356 and 2392 in the same position and also A2315 and A2333. But neither seems to match well.
A2333 does match A2375 though.

A2333 courtesy of Steve Lawrie.	A2375

These numbers are 42 apart, suggesting 6, 7 or 42 stamps per sheet.

A2342 courtesy of Steve Lawrie.	A2468 courtesy of Andrew Higson.

A new scan provides another match. These are 126 apart, or 3 x 42.


If there were only 6 or 7 stamps per sheet, I would have a lot more than 3 matches with 17 stamps.
The answer would indeed appear to be 42 stamps per sheet, since it is also compatible with the match above.
The next question is, is this 6 rows of 7 or 7 rows of 6? (assuming it isn't 14 x 3)



These are 7 apart, if there were 7 stamps to a column, they would be adjacent and the same height. They are clearly not the same height.
The stamps with edging on them shown below make it certain that these two are on the same sheet.
That means it must be 6 stamps to a column, i.e. 6 rows, and therefore 7 stamps to a row.

A2315 courtesy of the late Mike Holt   - It is the only one I've seen with a broken line under 'LIMITED'.	A2354 courtesy of Steve Lawrie.	A2356 courtesy of Steve Lawrie.	A2358

Experience with the 6d makes me confident that these are on the edge of a sheet.
Although, considering the last 3 have to be on the edge of the same pane, it looks like some have been trimmed.

A2392 courtesy of Mark Gibson.

This seems to show a straight edges on the left side like the 6d and 3d ones do.
This also though shows an indications of a single perforation on that side.

There is another constraint on the number of stamps per sheet:

4d control *A2466 on dull paper.	4d control *A2468 on translucent paper.
Courtesy of Andrew Higson who remarks that for such close serial numbers, the paper and colour are very different. A2466 and A2468 are clearly from different sheets. Assuming an unbroken numbering sequence, A2466 would be the last in its sheet if there were 6 or 18 stamps per sheet. They would both be on the same sheet if it were 12, 20, 21, 24, 30, 32, 36, 40, 42, 46 or 48 stamps per sheet. This puts a fly in the ointment, as we have now concluded there are 42 stamps per sheet. We have to throw out our initial assumption. Things appear to be consistent up to A2375 that matches A2333, but then A2340 matches A2466, so it is hard to see how A2468 would be on another sheet !

This would give us:

A2364 - courtesy of Martien Blank.	A2358
Were these once joined together ?

There is 3 other scan of a 4d stamps not yet shown, A2367, A2378 and A2383.
These are shown below.

A2367 - courtesy of Steve Lawrie.	A2378 - courtesy of Steve Lawrie.	A2383 - courtesy of Steve Lawrie.

3d  -  probably sheets with 2 panes vertically, each having 42 stamps (6 rows of 7).

I initially thought 28 stamps per sheet which would have been good for producing booklet panes.
The problem was that it would imply that I should have a lot of matching scans that do not in fact match.

Perhaps surprisingly, I have only one scan of consecutive numbers of the 3d. Also (perhaps) one horizontal pair.
I do have several scans with matching flaws however:


A661 and A1221 (560 apart) courtesy of Steve Lawrie. I originally matched these because of the bottom-right corner,
however it causes problems for me (see below). I am now inclined to think this similarity is a coincidence. There are differences.



A1320 courtesy of Steve Lawrie, and A1404 (84 apart) - courtesy of the late Mike Holt.


A1112 courtesy of Steve Lawrie and A1448 (336 apart) courtesy of Martien Blank.


A1144 (anonymous) and A1480 (336 apart) courtesy of Auktionshaus Christoph Gärtner.
I didn't expect to be able to match the image on the right, but the numbers said I should check against A1144. Here it is.

3d Scenarios.

One of them could be mis-numbered, but the 6d stamps seem very consistent.
With just the first pair it looked like the early ones were in sheets of 40, but later ones were in sheets of 12 or 6.

Now though, the ranges overlap.
The first pair (A661/A1221) do not look very convincing, perhaps they are not really a match. What else do we have?

A1267 and A1262 (with selvedge) are just 5 apart.
If they were both from sheets with 5 stamps per column, then they should be a horizontal pair.

A1267 courtesy of Steve Lawrie and A1262 now mine, Ex-Andrew Higson.
Clearly they are not from the same row, they have different heights. If they were numbered like the 6d stamps, they did not have 6 stamps per column.
If there were 5 stamps per row, A1267 should also have selvedge.

The selvedge on A1262 (courtesy of Andrew Higson) has to be accounted for.
If they were all in sheets of 28, this would be the second stamp on its sheet.

A sheet starting with A1261 (following 105 sheets of 12) could be like this:

The problem with this is that it puts 6 other pairs in the same locations that do not actually appear to match.
Indeed, my simulations suggest that to get only two matching pairs out of 22 stamps, there should be about 80 or so stamps per sheet.

A sheet starting with A1261 (following 45 sheets of 28) could be like this:

This would put A1267 on the right edge under A1262. However it doesn't have selvedge !

Control Sheet Column Row A661 24 3 1 A1112 40 3 4 A1144 41 2 4 A1221 44 3 1 A1262 46 7 2 A1267 46 6 3 A1310 47 2 2 A1320 48 7 4 A1324 48 6 4 A1336 48 3 4 A1338 48 2 2 A1357 49 4 1 A1364 49 3 4 A1379 50 6 3 A1386 50 4 2 A1395 50 2 3 A1401 51 7 1 A1404 51 7 4 A1415 51 4 3 A1423 51 2 3 A1424 51 2 4 A1429 52 7 1 A1443 52 4 3 A1448 52 3 4 A1463 53 6 3 A1469 53 4 1 A1480 53 2 4 A2121 76 2 1 A2382 86 7 2 A4853 174 5 1 A4999 179 4 3

However it could also look like this (adding in additional recent scans): R\C 1 2 3 4 5 6 7 1 1285 1281 1277 1273 1269 1265 1261 2 1286 1282 1278 1274 1270 1266 1262 3 1287 1283 1279 1275 1271 1267 1263 4 1288 1284 1280 1276 1272 1268 1264   R\C 1 2 3 4 5 6 7 Totals 1 0 1 2 2 1 0 2 8 2 0 2 0 1 0 0 2 5 3 0 2 0 3 0 3 0 8 4 0 3 4 0 0 1 2 10 Totals 0 8 6 6 1 4 6 31 Giving the 31 stamps with none in the first column. Just like the 6d sheets were at the time. This divides easily into 4 panes of 6 for booklet use plus column 1 leftover for counter use. The only thing is that this gives other matches that I had not noticed. I had matched A1112 and A1448 at Column 3, Row 4, A1364 does also, but according to this A1336 should also match them. In addition there should be 4 others matches: C2/R3-A1395/1423,   C4/R1-A1357/1469,   C6/R3-A1267/1463   and   C7/R1-A1401/1429. As shown below, these do not appear to match.

The perforations between these two are not entirely convincing, but together with the straight-edge on the left of both of them strongly suggest that are a pair from the left of a pane. This scenario though, puts them in column 2. A1423 courtesy Mark Gibson, A1424 courtesy Steve Lawrie.

I like widgits.

They take a little while to make, but make life much easier.
Click on 'popup' to open it in another small window.

This gives positions for the early 6d type on a 5 rows of 8 basis.
for the 3d it does the same for the first 31 sheets (1240 stamps) then switches to 12 stamps per sheet.

Enter a Control number to check            

A1112 courtesy of Steve Lawrie and matching A1448 courtesy of Martien Blank from above, together with A1336 courtesy of Andrew Higson,
and now A1364. With 4 rows of 7 stamps, these would all be Col.3, Row 4.
The first two and last match, but not the third.

A1267 and A1463 courtesy of Steve Lawrie, with A1379 courtesy Geoff Wright of Sandysstamps
With 4 rows of 7 stamps, these would all be Col.6, Row 3. Again, the first two match but not the third.

A1423 courtesy of Mark Gibson. A1395 is illustrated by Langmead and Huggins (colour plate 2) and is Courtesy of the Great Britain Philatelic Society.
With 4 rows of 7 stamps, these would both be Col.2, Row 3. They do not appear to match.

A1357 courtesy of Steve Lawrie and A1469 courtesy of Mark Gibson.
With 4 rows of 7 stamps, these would both be Col.4, Row 1. They do not appear to match either.

A1401 courtesy of Mark Gibson and A1429 courtesy of Steve Lawrie.
With 4 rows of 7 stamps, these would both be Col.7, Row 1. They look about as different as they can get.

The large number of false matches suggests that there are more stamps in the sheet.
But if there are more than 6 rows then A1267 will be in column 7 under A1262, but it does not have selvedge.
If we ignore the A661/A1221 match as being a coincidence, there is the possibility of 84 stamps per sheet.
If we keep 4 rows, then this would need 21 columns (3 panes of 7).
If there are 6 rows, then this would need 14 columns (2 panes of 7).

Modifying the 3d for 21 x 4 gives :

Enter a Control number to check        

Modifying the 3d for 14 x 6 gives :

Enter a Control number to check        

Here are A1463 and A1469. It's close and the straight edge on 1469 suddenly finds an explanation (which it doesn't in the other scenarios).
Looking at the other 3d examples with straight edges, A1221 and A1429 are at 7/3 and 14/1, both on the right sides (of a block of 7 columns) and both
having straight edges that side. A1386 is from 8/6 on the left side with that edge straight.
A bit much for a coincidence, each of the other scenarios get 1 out of 3 correct, but the fact of straight edges on 7/3 and 14/1 may be significant.
Admittedly someone could have snipped off perforations on a whim, but if not perhaps the sheets were actually 12 rows of 7 to make 84.
So does this pair fit together? What do you think ?

That would suggest that 12 rows of 7 stamps arranged in panes as shown (for the straight edges) as the most likely option.

Here is a potential fly in the ointment. The 3d is normally on yellow paper, but of variable transparency.

3d control A1262   (16/14/2 ?)	3d control A1336   (16/2/4 ?)
Reverse of stamps shown above, courtesy of Andrew Higson. Are these from the same sheet? They were on different panes, could the different panes have been on different sheets of paper? A1379 (sheet 17) appears to be on the same kind of paper as A1336. There is a similar problem with the 4d stamps. The bottom-right corner of A1262 though looks a bit more translucent, perhaps some were affected by oil or something.

This A1262 example presents another problem. It is very tall, how far apart were the rows of stamps?
It looks like it is from the top of a sheet but in our current scenario it isn't.

Here is another problem.

The stamp below presents something of an enigma. This is one of 5 scans I have between A1400 and A1430, but the others are all normal.
It is imperf. and surface coated rather than being on yellow paper. It was sold as a proof. The number appears to have been random. There was almost certainly a real A1415.

3d control A1415 sold as a proof.	3d control A1415 sold as a proof.
Courtesy of Andrew Higson.

One more fly to add to the ointment :



This should be on the right side of the sheet. Other stamps in that position have a straight edge on the right.
Is there a problem with this proposed arrangement? This is the highest number that I've seen for the 3d. Perhaps something changed,
perhaps they started perforating around the edges ? On the other hand it may have been torn from a stub on the left-hand side.

There are 8 other scans of 3d stamps not yet shown.
These are A1310, A1324, A1338, A1386, A1443, A2121, A2382 and A4853 and are shown below.

       
A1310 courtesy Derek Harborne, A1338 courtesy of Martien Blank, A1324 and A1386 courtesy of Steve Lawrie.

       
A1443 (mine); A2121 courtesy of Mark Gibson; A2382 and A4853 courtesy of Steve Lawrie.

More scans needed.

1865   Perf.11½

Later the design was changed, initially with manuscript controls.

3d on blue-green paper,  RH4 Number 240	6d on vermilion paper,  RH5 Number 860	6d  RH5 Number 427	6d  RH5 Number 446 ?	6d  RH5 Number 363 changed to 863 ?
From Langmead & Huggins, Colour Plate 1, Courtesy of the Great Britain Philatelic Society.		Courtesy of Steve Lawrie.

If I assume that these are the same sheet format and numbering as the later stamps, 5 rows of 8, horizontally from top-left, I get:

I have now one scan of the 3d, control 240, as illustrated in the Langmead and Huggins book.
This matches row 5 stamp 8 of the later type (see below), the last stamp on the 6th. sheet of 40.
240 is the correct control number for the last stamp on the sixth sheet.

6d Control 860 is illustrated by Langmead and Huggins and matches row 5, stamp 4, though 860 would be expected at row 3 stamp 4.
6d control 427 is a Group 2 stamp matching row 2 stamp 4 of the later type stamps, though 427 would be expected at row 4 stamp 3.
6d control 446 is a Group 2 stamp matching row 1 stamp 6 of the later type stamps.
6d control 863(?) is also a Group 2 stamp but matches row 3 stamp 5 of the later type stamps, though 863 would be expected at row 3 stamp 7.

If I assume that these are 5 rows of 8, vertically numbered from top-left, I get:

It makes no difference to the 3d, it is still the last stamp in the sheet,
but now the 6d stamps numbered 860 and 863 are correct, but the last would need to be 417 to match.

Transitional VERTICAL numbering.

Enter a Control number to check          

From Langmead & Huggins, Colour Plate 1. Courtesy of the Great Britain Philatelic Society.	One of mine.	Courtesy of  Roger de Lacy-Spencer.	Courtesy of Steve Lawrie.
240 (6/5/8)	4000 (100/5/8)	1520 (38/5/8)	6800 (170/5/8)

The 6d stamps, though of the same type, appear to have had a different arrangement on the plate.

From Langmead & Huggins, Colour Plate 1. Courtesy of the Great Britain Philatelic Society.	One of mine.	Courtesy of the late Mike Holt.	Anonymous.
860 (22/5/4)	2836 (71/5/4)	4036 (101/5/4)	4436 (111/5/4)

Courtesy of Steve Lawrie.	One of mine.	Courtesy of Steve Lawrie.	Courtesy of Martien Blank.
417 misnumbered? (11/2/4)	3292 (83/2/4)	3932 (99/2/4)	3092 (78/2/4)

Courtesy of Steve Lawrie.	Courtesy of  Steve Lawrie.	Courtesy of  Andrew Higson.	One of mine.
446 (12/1/6)	3486 - 88/1/6 (Group 3)	2806 - 71/1/6 (Group 3)	3806 - 96/1/6 (Group 3)

Courtesy of Steve Lawrie.	Courtesy of Mark Gibson.	Courtesy of mirak5 on ebay.
863 (22/3/5)	9141 (229/3/5)	3141 (79/3/5)

6d Group 2 stamps have the characteristic mark above 'ES' of 'MESSAGE' and extended bottom line of 'E' in 'CHIEF'.

There are matching flaws between the manuscript and later types, however the 6d later types share common flaws that do not appear to be on the manuscript types.
These presumably were picked up on a transfer from the manuscript type.
Additionally, and more importantly the later type appear to be in different positions.
This may account for the 'scrambled' Group arrangement of the later type.

The 3d later types may have been printed from the same plate as the manuscript 3d types.

More scans needed.

Later type

Despite L&H saying these are Perf. 12½, I make them Perf. 12
The controls were printed in either large or small type. The colours were changed to those below.

Hiscocks H6	Hiscocks H6a courtesy of Mark Gibson.	Hiscocks H7	Hiscocks H7a

Layouts.

In the beginning I had little in the way of multiples, re-joined pairs at best and it was painstaking to make a case for the sheet layouts.
Fortunately larger multiples have now come to light that make a simpler and more convincing case.
The block below is courtesy of Steve Lawrie.



The difference of 8 between the top stamps and the bottom, makes it clear that there were 8 stamps in a row. It also implies that this sheet ended with number 16440, so 16440 should be divisible by the number of stamps per sheet.

For a long time, it bothered me that I never saw a pair crossing the left-right division but, getting ahead a bit, the strip below does, though it is re-joined. I suspect that the complete sheets were routinely quartered before use.

Re-joined strip of 4 courtesy of Andrew Higson.

4850 to 4853 are from row 2, columns 2 to 5

The next step is to determine the number of stamps (and thus rows) in a sheet.
As with the first series, matching flaws are needed.

I now have many choices, but this is a good example.

Courtesy of Mark Gibson.	Anonymous	Anonymous.	Anonymous
1569 (40/2/1)	4529 (114/2/1)	8249 (207/2/1)	1693 (43/2/5)

One thing that quickly became apparent is that there were slight differences between types.
1693 has an extra flaw (left of the 'L' corner) that the others do not have. It is from a different (though related) position.
I started to use the term 'alt' for alternative to indicate the similar ones.
Below is a more complete list of stamps like this:

Splitting the two types gives:

Knowing there are 40 stamps per sheet and 8 stamps per row tells me that there are 5 rows per sheet.
I have broken the control number down into Sheet(1+), Row(1 to 5) and Column(1 to 8) using the 'widget' below.

Tha 'alt's are 4 columns away. The right and left halves of the sheet appear to be close copies of each other.

Notice that this works with large and small controls.
Images of all of these can be found under 'Row-2' lower down.

This calculator made my life easier.

On the evidence that at least some sheets with 5 rows of 8 stamps were used, I made a 'widget' to decode sheet position from control number.
Click on 'popup' to open it in another small window.

Enter a Control number to check          

Characteristics on the 3d.

I originally chose arbitrary groups until I understood the pattern. I will now list by Row Number / Stamp Number.
I have put a thick green line around some major characteristics and a thin one around lesser associated ones I have listed.
The magenta lines represent smaller ones I did not list. Smaller marks may be visible or not depending on under/over inking and age of the stone.
I have used green to indicate primary flaws in which case the blue indicates secondary flaws.

One of mine.	Courtesy of the late Mike Holt.	SECONDARY flaws
		This pair have the following flaws in common that the other pairs do not have: 1) Black mark between 'ST' of 'DISTRICT'. 2) Frame break above first 'S' in 'MESSAGE'. 3) Frame break on top border. 4) White dot in 'D' corner. 5) White dot above 'M' of 'MESSAGE'. 6) White dot below-left of 'L' in 'L' corner. Row-1, Stamps-1 & 5  Controls: 885, 1481, 1485, 1685, 2085, 2845 small: 685, 10325 Using my 'widget' at the bottom gives me: 685 = 18/1/5 885 = 23/1/5 1481 = 38/1/1 1485 = 38/1/5 1685 = 43/1/5 2085 = 53/1/5 2845 = 72/1/5 10325 = 259/1/5
1481 (38/1/1)	885 (23/1/5)

One of mine.	Courtesy of Steve Lawrie.	SECONDARY flaws
		This pair have the following flaws in common that the other pairs do not have: 1) Frame-break above 'T' corner. 2) White dot below 'A' of 'STAMP'. 3) Mark between frames to right of 'STREET'. 4) Broken leg of 'R' in 'DISTRICT'. 5) Black mark above 'SS' of 'MESSAGE'. 6) Black mark between 'Ch" of 'Charles'. 7) Several frame breaks. Row-1, Stamp-2     Controls: 1482, 1562, 4242 Using my 'widget' at the bottom gives me: 1482 = 38/1/2 1562 = 40/1/2 4242 = 107/1/2 .
4242 (107/1/2)	1562 (40/1/2)

One of mine.	Anonymous.	SECONDARY flaws
		This pair have the following flaws in common that the other pairs do not have: 1) Frame-break above 'T' corner. 2) White dot right of 'T' in 'T' corner. 3) Mark in 2nd. 'O' of 'LONDON'. 4) Broken leg of 'R' in 'DISTRICT'. 5) White mark above 'M' of 'MESSAGE'. 6) Numerous marks between frame lines. Row-1, Stamp-6     Controls: 206, 1286, 1646, 3966, 4886 Using my 'widget' at the bottom gives me: 206 = 6/1/6 1286 = 33/1/6 1646 = 42/1/6 3966 = 100/1/6 4886 = 123/1/6 A mark appears between 'LE' of 'TELEGRAPH' on 1646 and after.
206 (6/1/6)	1646 (42/1/6)

One of mine .	Courtesy of ibredguy.co.uk.	SECONDARY flaws
		This pair have the following flaws in common: that the other pairs do not have: 1) Black mark on 'R' of 'DISTRICT'. 2) Black mark between 'ON' of 'LONDON'. 1/3 has an additional broken 'C' in 'District'. Row-1, Stamps-3 & 7     Controls: 843, 1363, 1487, 1767, 2163, 2763, 4207, 4247 small 3887, 10327 Using my 'widget' at the bottom gives me: 843 = 22/1/3 1363 = 35/1/3 1487 = 38/1/7 1767 = 45/1/7 2163 = 55/1/3 2763 = 70/1/3 3887 = 98/1/7 4207 = 106/1/7 4247 = 107/1/7 10327 = 259/1/7
843 (22/1/3)	4247 107/1/7)

One of mine.	One of mine.	SECONDARY flaws
		Characteristics: 1) Dot above 'T' in 'T' corner. 2) Dot below 'ME' in 'MESSAGE'. 3) Dot below 'GE' in 'MESSAGE'. 4) Dot between 'EC' in 'SECy'. On 1/4 only: 5) Dot below 'L' in the 'L' corner. 6) Dot above-left of 'G' in 'MESSAGE'. 7) Mark on 'D' in 'D' corner. On 1/8 only: 8) Dash outside right frame. 9) Dot below 'M' of 'MANAGER'. Row-1, Stamp-4 & 8     Controls: 928, 2124, 2444, 3164             small: 1648, 6164, 10328 Using my 'widget' at the bottom gives me: 928 = 24/1/8 1648 = 42/1/8 2124 = 54/1/4 2164 = 55/1/4 2444 = 62/1/4 6164 = 155/1/4 10328 = 259/1/8
2444 (62/1/4)	10328 (259/1/8)

Examples from row 1.

One of mine.	Courtesy of ibredguy.co.uk.	Courtesy of Steve Lawrie.
1481 (38/1/1)	1482 (38/1/2)	1562 (40/1/2)

One of mine.	Courtesy of Steve Lawrie.	One of mine.
4242 (107/1/2)	843 (22/1/3)	1363 (35/1/3)

Courtesy of Steve Lawrie.	One of mine.	Anonymous.	One of mine.
2163 (55/1/3)	2763 (70/1/3)	2124 (54/1/4)	2164 (55/1/4)

One of mine.	Courtesy of Martien Blank.	Courtesy of Steve Lawrie.	Courtesy of the late Mike Holt.
2444 (62/1/4)	6164 (155/1/4)	685 (18/1/5)	885 (23/1/5)

One of mine.	Courtesy John A. McCulloch, © 2010 ARR.	Courtesy of Steve Lawrie.
1485 (38/1/5)	1685 (43/1/5)	2085 (53/1/5)

Courtesy of ibredguy.co.uk.	Courtesy of Steve Lawrie.
2845 (72/1/5)	10325 (259/1/5)

One of mine.	Courtesy of Steve Lawrie.	Anonymous.
206 (6/1/6)	1286 (33/1/6)	1646 (42/1/6)

One of mine.	Courtesy of Steve Lawrie.	Courtesy of ibredguy.co.uk.	One of mine.
3966 (100/1/6)	4886 (123/1/6)	1487 (38/1/7)	1767 (45/1/7)

Courtesy of Mark Gibson.	Anonymous.	Courtesy of ibredguy.co.uk.	One of mine.
3887 (98/1/7) small.	4207 (106/1/7)	4247 (107/1/7)	10327 (259/1/7)

One of mine.	Courtesy of Steve Lawrie.	One of mine.
928 (24/1/8)	1648 (42/1/8)	10328 (259/1/8)

These have a 'notch' on the left side of the 'T' corner, but it is higher than the other ones.

Anonymous	Anonymous	SECONDARY flaws
		This pair has the following flaws in common that the others do not have: 1) 'T' joined to frame in 'T' corner. 2) White dot under 'AM' of 'STAMP'. 3) White dot between 'SA' of 'MESSAGE'. The '-alt' has an additional mark left of the 'L' corner. Row-2, Stamps-1 & 5                 Controls: large 853, 1413, 1653, 1693, 3973, 4529, 4853, 4893 small 1569, 1649, 8209, 8249 These stamps are quite distinctive. Using my 'widget' at the bottom gives me: 853 = 22/2/5 1413 = 36/2/5 1569 = 40/2/1 1649 = 42/2/1 1653 = 42/2/5 1693 = 43/2/5 3973 = 100/2/5 4529 = 114/2/1 4853 = 122/2/5 4893 = 123/2/5 8209 = 206/2/1 8249 = 207/2/1
4529 (114/2/1)	1693 (43/2/5)

One of mine.	Courtesy of Mark Gibson.	SECONDARY flaws
		Characteristics: 1) Mark on top of 'L' corner. 2) Line through 'M' of 'STAMP'. On 2/6 only: 3) Mark top-right of 'E' in 'MESSAGE'. 4) Mark in 'Co' corner. Row-2, Stamps-2 & 6   Controls: 1450, 1490, 2130, 2134, 2690, 3974, 4254, 4850 small: 1654 These stamps are quite distinctive. Using my 'widget' at the bottom gives me: 1450 = 37/2/2 1490 = 38/2/2 1654 = 42/2/6 2130 = 54/2/2 2134 = 54/2/6 2690 = 68/2/2 3974 = 100/2/6 4254 = 107/2/6 4850 = 122/2/2
1450 (37/2/2)	2134 (54/2/6)

Courtesy of Andrew Higson.	Courtesy of Andrew Higson.	SECONDARY flaws
		Distinctive varieties of note: 1) Marks in 'D' and 'T' corners. 2) Many marks in 'Co' corner. 3) Marks around frames. 4) Diagonal scratch across 'L' corner. The 'Alt' has less marks in 'Co' corner. but marks in 'T' and 'L' corners. Row-2, Stamps-3 & 7   Controls: 931, 1131, 2055, 4851, 4855 small: 10331, 10335 These stamps are quite distinctive. Using my 'widget' at the bottom gives me: 931 = 24/2/3 1131 = 29/2/3 2055 = 52/2/7 4851 = 122/2/3 4855 = 122/2/7 10331 = 259/2/3 10335 = 259/2/7
4851 (122/2/3)	4855 (122/2/7)

Courtesy of ibredguy.co.uk.	Courtesy of the late Mike Holt.	SECONDARY flaws
		These have the following flaws in common: that the others do not have: 1) White dot left of 'D' in 'D' corner. 2) 2 white dots under 'D' of 'D' corner. 3) Black protrusion below 'T' corner. 4) White dot under 'T' of 'STAMP'. 5) Dot before 'C' of 'CHIEF'. 6) Black mark under 'D' corner on '-alt'.) 7) Black marks right of 'D' corner on '-alt'.) Row-2, Stamp-4 Controls: large 972, 1372, 2012, 4216, 4772, 4852 small 5252, 10572, 24932 Using my 'widget' at the bottom gives me: 972 = 25/2/4 1372 = 35/2/4 2012 = 51/2/4 4612 = 116/2/4 4772 = 120/2/4 4852 = 122/2/4 5252 = 132/2/4 10572 = 265/2/4 24932 = 624/2/4
4612 (116/2/4)	5252 (132/2/4)

Courtesy of the late Mike Holt.	Courtesy of ibredguy.co.uk.	SECONDARY flaws
		These have the following flaws in common: that the others do not have: 1) White dot left of 'D' in 'D' corner. 2) 2 white dots under 'D' of 'D' corner. 3) Black protrusion below 'T' corner. 4) White dot under 'T' of 'STAMP'. 5) Dot before 'C' of 'CHIEF'. 6) Broken top serif 'L' corner. (not on '-alt') Row-2, Stamp-8 Controls: large 1576, 1776, 2216, 2256, 4216, 4456, 4536 small 22656 Using my 'widget' at the bottom gives me: 1576 = 40/2/8 1776 = 45/2/8 2216 = 56/2/8 2256 = 57/2/8 4216 = 106/2/8 (2 copies) 4456 = 112/2/8 4536 = 114/2/8 22656 = 567/2/8
4216 (106/2/8)	4456 (112/2/8)

'r-dot' : The signature near the bottom is that of the Secretary and Manager Charles Curtoys.
There is a dot above the signature written as if the 'r' in 'Charles' was an 'i' so I will refer to it as the 'r-dot'
It can be clearly seen on the mint stamp above..

The 'r-dot' can help us. It can be as above or missing or over to the right near the 'l' of 'Charles'.
It depends on the stamp position, but not necessarily on the control number.

Examples from row 2 .

Courtesy of Mark Gibson.	Anonymous.	Anonymous.	Anonymous.
1569 (40/2/1)	4529 (114/2/1)	8209 (206/2/1)	1649 (42/2/1)

Anonymous.	One of mine.	Courtesy of ibredguy.co.uk.	Courtesy of ibredguy.co.uk.
8249 (207/2/1)	1450 (37/2/2)	1490 (38/2/2)	2130 (54/2/2)

Anonymous.	Courtesy of Andrew Higson.
2690 (68/2/2)	4850 (122/2/2)

One of mine.	One of mine.	Courtesy of  Andrew Higson.	Courtesy of  Steve Lawrie.
931 (24/2/3)	1131 (29/2/3)	4851 (122/2/3)	10331 (259/2/3)

One of mine.	One of mine.	Courtesy of  Steve Lawrie.
972 (25/2/4)	1372 (35/2/4)	2012 (51/2/4)

Courtesy of ibredguy.co.uk.	Courtesy of  Mike Jackson.	Courtesy of  Andrew Higson.
4612 (116/2/4)	4772 (120/2/4)	4852 (122/2/4)

Courtesy of the late Mike Holt.	Courtesy of Steve Lawrie.	One of mine.
5252 (132/2/4)	10572 (265/2/4)	24932 (624/2/4) - highest number

Courtesy of ibredguy.co.uk.	Courtesy of  Steve Lawrie.
853 (22/2/5)	1413 (36/2/5)

Courtesy of  Mike Jackson.	Anonymous.	Courtesy of Mark Talbot.	Courtesy of Andrew Higson.
1653 (42/2/5)	1693 (43/2/5)	3973 (100/2/5)	4853 (122/2/5)

One of mine.	One of mine.	Courtesy of Mark Gibson.	Courtesy of Steve Lawrie.
4893 (123/2/5)	1654 (42/2/6)	2134 (54/2/6)	3974 (100/2/6)

Courtesy of Steve Lawrie.	Reproduced with permission from http://www.stampboards.com	Courtesy of Andrew Higson.	One of mine.
4254 (107/2/6)	2055 (52/2/7)	4855 (122/2/7)	10335 (259/2/7)

One of mine.	Courtesy of ibredguy.co.uk.	One of mine.	One of mine.
1576 (40/2/8)	1776 (45/2/8)	2216 (56/2/8)	2256 (57/2/8)

Courtesy of Martien Blank.	Courtesy of the late Mike Holt.	Courtesy of ibredguy.co.uk.
4216 (106/2/8) used	4216 (106/2/8) unused	4456 (112/2/8)

Courtesy of Mark Gibson.	Courtesy of  Stephen Gardner.
4536 (114/2/8)	22656 (567/2/8)

One of mine.	One of mine.	SECONDARY flaws
		There appear to be some flaws in common with all, but 3/5 has shared flaws not on 3/1 (red). Also 3/1 has a couple of flaws not on 3/5 (magenta). 1) Many 'speckles' in all four corners. 2) Many 'speckles' in 'MESSAGE STAMP' area. 3) Notches in 'N' of 'No.' Row-3, Stamps 1 and 5        Controls: 1537, 2137, 2141, 2781 Using my 'widget' at the bottom gives me: 1537 = 39/3/1 2137 = 54/3/1 2141 = 54/3/5 2781 = 70/3/5
1537 (39/3/1)	2781 (70/3/5)

Courtesy of  Stephen Gardner.	One of mine.	SECONDARY flaws
		These have the following flaws : 1) White dots in 'T' corner. 2) White protrusion from 'L' in 'L' corner. 3) Frame break under 'S' of 'SECy'. Row-3, Stamps-2 & 6     Controls: 1498, 1978, 2142, 2262, 2782, 2822, 3182, 4862 small 1582, 15742 Using my 'widget' at the bottom gives me: 1498 = 38/3/2 1582 = 40/3/6 1978 = 50/3/2 2142 = 54/3/6 2262 = 57/3/6 2782 = 70/3/6 2822 = 71/3/6 3182 = 80/3/6 4862 = 122/3/6 15742 = 394/3/6 1498 and 1978 differ in having a scratch under the 'L' corner and a feint scratch across '90'.
1978 (50/3/2)	2822 (71/3/6)

Courtesy of Mark Gibson.	Anonymous	SECONDARY flaws
		This pair have the following flaws in common that the other pairs do not have: 1) White mark above 'E' of 'MESSAGE'. 2) White mark right of 'E' in 'MESSAGE'. 3) Higher dot right of 'T' in 'T' corner. 4) Dot below 'T' in 'T' corner. 5) Mark on 'N' of 'LONDON' (not on '-alt'). Row-3, Stamps-3 & 7   Controls: 1499, 2139, 2259, 2379, 2739, 2743, 4259, 4699                 small     23499 Using my 'widget' at the bottom gives me: 1499 = 38/3/3 2139 = 54/3/3 2259 = 57/3/3 2379 = 60/3/3 2739 = 69/3/3 2743 = 69/3/7 4259 = 107/3/3 4699 = 118/3/3 23499 = 588/3/3
2739 (69/3/3)	2743 (69/3/7)

One of mine	Courtesy of  Chris Ceremuga.	SECONDARY flaws
		These have the following flaws : 1) White dot right of 'T' in 'T' corner. 2) White marks in 'L' corner. 3) White marks in 'D' corner. Row-3, Stamps-4 & 8    Controls: 340, 1580, 1860, 2140, 2780, 3660 small 1584, 2384 Using my 'widget' at the bottom gives me: 340 = 9/3/4 1580 = 40/3/4 1584 = 40/3/8 1860 = 47/3/4 2140 = 54/3/4 2384 = 60/3/8 2780 = 70/3/4 3660 = 92/3/4
3660 (92/3/4)	small 1584 (40/3/8)

Examples from row 3 .

One of mine.	One of mine.	One of mine.	One of mine.
1537 (39/3/1)	2137 (54/3/1) Note printing flaw.	2141 (54/3/5)	2781 (70/3/5) Note printing flaw.

One of mine.	Courtesy of  Stephen Gardner.
1498 (38/3/2)	1978 (50/3/2)

Courtesy of  Steve Lawrie.	Courtesy of  Mike Jackson.	Courtesy of Steve Lawrie.
1582 (40/3/6)	2142 (54/3/6)	2262 (57/3/6)

Courtesy of Steve Lawrie.	One of mine.	One of mine.
2782 (70/3/6)	2822 (71/3/6)	3182 (80/3/6)

One of mine.	Courtesy of Steve Lawrie.
4862 (122/3/6)	15742 (394/3/6)

One of mine.	One of mine.	Courtesy of Steve Lawrie.
1499 (38/3/3)	2139 (54/3/3)	2259 (57/3/3)

Courtesy of Steve Lawrie.	Courtesy of Mark Gibson.	One of mine.
2379 (60/3/3)	2739 (69/3/3)	4259 (107/3/3)

Courtesy of Steve Lawrie.	Anonymous.	Anonymous.
4699 (118/3/3)	23499 (588/3/3)	2743 (69/3/7)

Courtesy of Mark Gibson.	Courtesy of  Mike Jackson.	Courtesy of  Steve Lawrie.
340 (9/3/4)	1580 (40/3/4)	1860 (47/3/4)

One of mine.	One of mine.	One of mine.
2780 (54/3/4)	2780 (70/3/4)	3660 (92/3/4)

Courtesy of  Chris Ceremuga.	Courtesy of  Steve Lawrie.
1584 (40/3/8)	2384 (60/3/8)

Courtesy of Mark Gibson.	One of mine .	SECONDARY flaws
		This pair have the following flaws in common: 1) Broken 'L' in 'TELEGRAPH'. 2) White dot under 'AM' of 'STAMP'. Row-4, Stamp-1     Controls: 265, 945, 1105, 1345, 1825, 2065, 2265, 3985, 4705 small: 1625, 13825, 16665 Using my 'widget' at the bottom gives me: 265 = 7/4/1 945 = 24/4/1 1105 = 28/4/1 1345 = 34/4/1 1625 = 41/4/1 1825 = 46/4/1 2065 = 52/4/1 2265 = 57/4/1 3985 = 100/4/1 4705 = 118/4/1 13825 = 346/4/1 16665 = 417/4/1
1105 (28/4/1)	1625 (small) (41/4/1)

Anonymous.	One of mine.	SECONDARY flaws
		This pair have the following flaws in common: 1) Bulge on top of 'E' in 'MESSAGE' . 2) Marks above 'AM' in 'STAMP'. 3) Dot to right of 'T' in 'T' corner. 4) H' of 'TELEGRAPH' broken, and mark to right. 5) Mark between 'DI' of 'DISTRICT'. (not 'ALT') 6) Broken frame on left side. (not 'ALT') Row-4, Stamps-2 & 6     Controls: 906, 1106, 1390, 1506, 2466, 4226, 4910 small: 10866 Using my 'widget' at the bottom gives me: 906 = 23/4/2 1106 = 28/4/2 1390 = 35/4/6 1506 = 38/4/2 2466 = 62/4/2 4226 = 106/4/2 4910 = 123/4/6 10866 = 272/4/2
1106 (28/4/2)	4910 (123/4/6)

One of mine.	Courtesy of Martien Blank.	SECONDARY flaws
		This pair have the following flaws in common: 1) Mark between 'MESSAGE' and 'STAMP'. 2) Mark between 'E' and 'S' of 'MESSAGE'. 3) Dots in 'T' corner. 4) Various flaws as indicated. Row-4, Stamps-4 & 8  Controls: 988, 1712, 1872, 2148, 2152 small 1588 Using my 'widget' at the bottom gives me: 988 = 25/4/4 1588 = 40/4/4 1712 = 43/4/8 1872 = 47/4/8 2148 = 54/4/4 2152 = 54/4/8 4188 = 105/4/4
2148 (54/4/4)	1872 (47/4/8)

Anonymous.	One of mine.	SECONDARY flaws
		This pair have the following flaws in common: 1) Broken 'L' in 'TELEGRAPH'. 2) Dots each side of 'M' of 'STAMP'. 3) Dot to left of 'T' in 'T' corner. 4) Mark hanging from 'D' in 'D' corner. Row-4, Stamp-5             Controls: 269, 2029, 4629 small 1589 Using my 'widget' at the bottom gives me: 269 = 7/4/5 1589 = 40/4/5 2029 = 51/4/5 4629 = 116/4/5
small 1589 (40/4/5)	269 (7/4/5)

Courtesy of Mark Gibson.	Anonymous.	SECONDARY flaws
		This pair have the following flaws in common: 1) Dot to left of 'L' in 'L' corner. 2) Two dots to right of 'L' in 'L' corner. 3) Additional mark in 'L' corner, not on '-alt'. Row-4, Stamps-3 & 7   Controls: 467, 1507, 1551, 2431, 2471, 4911 small: 10867, 22231 Using my 'widget' at the bottom gives me: 467 = 12/4/3 1507 = 38/4/3 1551 = 39/4/7 2431 = 61/4/7 2471 = 62/4/7 4911 = 123/4/7 10867 = 272/4/3 22231 = 556/4/7
467 (12/4/3) this is the one illustrated by Hiscocks.	1551 (39/4/7)

Examples from row 4 .

Courtesy of the late Mike Holt.	Courtesy of ibredguy.co.uk.	Courtesy of Mark Gibson.	One of mine.
265 (7/4/1)	945 (24/4/1)	1105 (28/4/1)	1345 (34/4/1)

One of mine.	Courtesy of Steve Lawrie.	Courtesy of Mark Gibson.	Courtesy of Andrew Higson.
1625 (41/4/1)	1825 (46/4/1)	2065 (52/4/1)	2265 (57/4/1)

Courtesy of Steve Lawrie.	Courtesy of Mark Gibson.	Courtesy of Steve Lawrie.	Courtesy of Steve Lawrie.
3985 (100/4/1)	4705 (118/4/1)	13825 (346/4/1)	16665 (417/4/1)

One of mine.	Courtesy of Mark Gibson.	Anonymous.	Anonymous.
269 (7/4/5)	2029 (51/4/5)	4629 (116/4/5)	1589 (40/4/5)

Courtesy of ibredguy.co.uk.	Anonymous.	One of mine.
906 (23/4/2)	1106 (28/4/2)	1506 (38/4/2)

Courtesy of Steve Lawrie.	One of mine.	Courtesy of Steve Lawrie.
2466 (62/4/2)	4226 (106/4/2)	10866 (272/4/2)

One of mine.	Courtesy of the late Mike Holt.
1390 (35/4/6)	4910 (123/4/6)

Courtesy of Mark Gibson.	One of mine.	One of mine.
467 (12/4/3)	1507 (38/4/3)	10867 (272/4/3)

Anonymous	One of mine.	Courtesy of Mark Gibson.
1551 (39/4/7)	2431 (61/4/7)	2471 (62/4/7)

One of mine.	Courtesy of Steve Lawrie.
4911 (123/4/7)	22231 (556/4/7)

Courtesy of Steve Lawrie.	Courtesy of Martien Blank.	One of mine.	Courtesy of Steve Lawrie.
988 Group (25/4/4)	1588 Group (40/4/4)	2148 Group (54/4/4)	4188 Group (105/4/4)

Anonymous.	Courtesy of Martien Blank.	Courtesy of Mark Gibson.
1712 (43/4/8)	1872 (47/4/8)	2152 (54/4/8)

Courtesy of Mark Talbot.	One of mine.	SECONDARY flaws
		Flaws in common : (the first 3 are also on stamp 5/5) 1) 'SA' in 'MESSAGE' joined. 2) Dot under 'E' of 'MESSAGE'. 3) Weakness under 'L' corner. 4) 'Blob' on last 'N' of 'CANNON'. 5) Dot below 'SS' in 'MESSAGE'. this is the '-alt' of Stamp-5. Row-5, Stamp-1                 Controls: 2313, 4193, 4633 Using my 'widget' at the bottom gives me: 2313 = 58/5/1 4193 = 105/5/1 4633 = 116/5/1
2313 (58/5/1)	4193 (105/5/1)

Anonymous.	One of mine.	SECONDARY flaws
		Flaws in common : 1) Marks in 'Co' corner. 2) Marks in 'T' corner. 3) Broken frame on right. 4) Weakness under 'L' corner. 5) Marks in 'L' corner (faint on 954). On 5/2 only: 6) Black Mark to right of Control No. Row-5, Stamps-2 & 6           Controls: 954, 998, 1238, 1518, 1558, 2594, 4194 Using my 'widget' at the bottom gives me: 954 = 24/5/2 998 = 25/5/6 1238 = 31/5/6 1518 = 38/5/6 1558 = 39/5/6 2594 = 65/5/2 4194 = 105/5/2
954 (23/5/2)	1238 (31/5/6)

Courtesy of Steve Lawrie.	Courtesy of Steve Lawrie.	SECONDARY flaws
		Characteristics : 1) Black spot right of 'L' corner. 2) Black spot on frameline left of 'L' corner. 3) Spot left of 'T' in 'T' corner (weak on 5/7). 4) below is on 5/3 only, and (5) is on 5/7 only. 4) Black mark above 'ST' of 'DISTRICT'. 5) Black mark between 'IE' of 'CHIEF'. Row-5, Stamps-3 & 7   Controls: 915, 1315, 1355, 1439, 1599, 2635, 4199 small: 6799, 10559, 17679, 20755 Using my 'widget' at the bottom gives me: 915 = 23/5/3 1315 = 33/5/3 1355 = 34/5/3 2635 = 66/5/3 1439 = 36/5/7 1599 = 40/5/7 4199 = 105/5/7 6799 = 170/5/7 10559 = 264/5/7 17679 = 442/5/7 20755 = 519/5/3
915 (23/5/3)	10559 (264/5/7)

Courtesy of the late Mike Holt.	One of mine.	SECONDARY flaws
		Flaws in common : 1) Major flaw in 'L' corner. 2) 'SA' in 'MESSAGE' joined. 3) Dot under 'E' of 'MESSAGE'. 4) Line joining frame-lines left of 'N' of 'No'. Row-5, Stamp-5       Controls: 637, 1557, 2557, 4477 small 2357 These two stamps are very distinctive. Using my 'widget' at the bottom gives me: 637 = 16/5/5 1557 = 39/5/5. 2357 = 59/5/5. 2557 = 64/5/5. 4477 = 112/5/5.
637 (16/5/5)	1557 (39/5/5)

Courtesy of ibredguy.co.uk.	One of mine.	SECONDARY flaws
		Flaws in common : 1) Mark between 'MESSAGE' and 'STAMP'. 2) Short middle horizontal to 'E' in 'CHIEF'. 3) White spot above-left of 'T' in 'T' corner. 3) Smeared comma after '90' (not on 'ALT'). Row-5, Stamps-4 & 8 Controls: 1520, 1920, 4000, 4556, 4880 (small) 1596, 6800, 19876 Manuscript 240 (L&H on Plate 1) Using my 'widget' at the bottom gives me: 240 = 6/5/8 1520 = 38/5/8 1596 = 40/5/4 1920 = 48/5/8. 4000 = 100/5/8. 4556 = 114/5/4 4880 = 122/5/8. 6800 = 170/5/8. 19876 = 497/5/4 Surprisingly the manuscript control and large controls were consistent! 6800 appears to have had the number corrected.
small 19876 (497/5/4)	4000 (100/5/8)

Examples from row 5 .

Courtesy of Mark Talbot.	One of mine.	One of mine.
2313 (58/5/1)	4193 (105/5/1)	4633 (116/5/1)

Courtesy of the late Mike Holt.	One of mine.	One of mine.	Courtesy of Steve Lawrie.
637 (16/5/5)	1557 (39/5/5)	2357 (59/5/5)	2557 (64/5/5)

Courtesy of Steve Lawrie.	Anonymous.	One of mine.	Courtesy of Steve Lawrie.
4477 (112/5/5)	954 ( 24/5/2)	2594 ( 65/5/2)	4194 ( 105/5/2)

One of mine.	One of mine.	Courtesy of ibredguy.co.uk.	Courtesy of Steve Lawrie.
998 (25/5/6)	1238 (31/5/6)	1518 (38/5/6)	1558 (39/5/6)

Courtesy of Steve Lawrie.	One of mine.	Courtesy of Mark Gibson.
915 (23/5/3)	1315 (33/5/3) (damaged)	1355 (34/5/3)

Anonymous.	Courtesy of Steve Lawrie.	Courtesy of Steve Lawrie.
2635 (66/5/3)	20755 (519/5/3)	1439 (36/5/7)

One of mine.	Courtesy of Steve Lawrie.	Courtesy of Steve Lawrie.	Courtesy of Steve Lawrie.
1599 (40/5/7)	4199 (105/5/7)	6799 (170/5/7)	10559 (264/5/7) - with 1886 postal cancel

Courtesy of Andrew Higson.	Courtesy of Mark Gibson.	One of mine.	Courtesy of ibredguy.co.uk.
17679 (442/5/7)	1596 (40/5/4)	4556 (114/5/4)	19876 (497/5/4)

One of mine.	Anonymous.	One of mine.	Courtesy of Steve Lawrie.
1520 (38/5/8)	1920 (48/5/8)	4000 (100/5/8)	6800 (170/5/8)

4879 is 122/5/7 and 4880 is 122/5/8 Unfortunately I do not know the source of this scan, anyone recognise it? I would like to give credit and perhaps get a better scan. These are also nearly the highest control numbers I have seen for these.

A similar analysis with the 6d.

Re-joined strip of 3 courtesy of Andrew Higson.
2806 is Row-1, stamp 6	2807 is Row-1, stamp 7	2808 is Row-1, stamp 8

Re-joined strip of 4 (Ex Iain Stevenson, now mine)
4501 is Row-3, stamp 5	4502 is Row-3, stamp 6	4503 is Row-3, stamp 7	4504 is Row-3, stamp 8

This is a reconstructed strip, held together at the back. These, like many of the others still have full gum despite being cancelled. It appears that they were generally pre-cancelled.

Until recently, the lack of any blocks and rarity of even pairs made this much more difficult.
This scan kindly provided by Martien Blank has confirmed much of the 'detective work'.
It certainly makes it clear that there were 8 stamps per row.


I will refer to this in the text as the 'MB-block'.

The arrangement of the 3d plate was fairly simple with the whole of each row derived from the same prototype (Group).
I expected the same of the 6d, but it was clearly not the case. some features on stamps also appear to have 'faded' over time.
See Row-1, stamp 1 , Row-2, stamp 1 , Row-4, stamp 2 and Row-5, stamp 7

I also had an expectation that some may have been printed as booklet panes.
Despite looking for evidence of this, I have to conclude that all the examples I have seen were printed in sheets of 40 stamps as with the 3d stamps.
The large and small controls fit the same pattern, but some slight differences suggest some plate-cleaning at some stage.
I have only scans of two examples of the earlier manuscript controls which are very similar to the 'Group 2' stamps below.

It is possible that the layout was intended to be the same as with the 3d stamps. My suspicion is as follows:
Two intermediate stones of 4 stamps derived from the same prototype (a manuscript type ?) were prepared,
but instead of the pairs staying together in the same finished row, they got mixed up.
It is possible that the manuscript type were in booklets of six and that five of these were used as prototypes for the later plates of 40.
With the manuscript types being so rare though, this has to remain conjecture.

Anonymous.	One of mine.	SECONDARY flaws
		Common features: This group has the following flaws in common: 1) Damage to top of 'L' corner. 2) Mark between frames left of 'D' corner. 3) Dot above top of frame line. Large controls only: 4) Extra mark on top of 'C' in 'Charles'. Row-1, Stamp 1               Controls: 1681, 3961 small: 2121, 11961, 15081, 15841 Using my 'widget' at the bottom gives me: 1681 = 43/1/1 2121 = 54/1/1 3961 = 100/1/1 11961 = 300/1/1 15081 = 378/1/1 15841 = 397/1/1 On 1681, you can see the white marks in the 'L' corner of the stamp below. That is 1689.
1681	2121 (small)

One of mine.	One of mine.	SECONDARY flaws
		These have the following flaws in common: 1) Quite a few small inner-frame breaks . 2) White mark under 'ST' of 'STAMP'. 3) White mark in 'T' corner. 4) White marks in 'Co' corner. 5) Black mark on 'N' of 'No'. 6) Black mark on 'M' of 'MANAGER'.   Row-1, Stamp 2           Controls: 4002 small: 2562, 37882 Using my 'widget' at the bottom gives me: 2562 = 65/1/2 4002 = 101/1/2 37882 = 948/1/2 37882 was cancelled by favour in Totton, 1956. (despite the fact that it was already used)
4002	37882

Courtesy of Mark Gibson.	Anonymous.	SECONDARY flaws
		These are similar to Row-1, stamp 1 above. 1) Damage to top of 'L' corner. 2) Mark between frames left of 'D' corner. 3) Dot above top of frame line. 4) Dot above 'ET' of 'STREET'. Row-1, Stamp 3      Controls: 2843, 3363, 4843         small     8283, 34683 Using my 'widget' at the bottom gives me: 2843 = 72/1/3 3363 = 85/1/3 4843 = 122/1/3 8283 = 208/1/3 34683 = 868/1/3
3363	8283 (small)

One of mine.	Courtesy of Mark Gibson.	SECONDARY flaws
		These have the following flaws in common: 1) Two small frame-breaks right of 'STREET'. 2) Large white marks in 'D' corner. 3) Dot above 'ET' of 'STREET'. 4) Numerous frame defects as indicated. Row-1, Stamp 4           Controls: 524, 3684         small     21444 Using my 'widget' at the bottom gives me: 524 = 14/1/4 3684 = 93/1/4 21444 = 537/1/4
524	21444 (small)

One of mine.	Courtesy of Mark Gibson.	SECONDARY flaws
		This is similar to Row-3 stamp 1 but without the join between 'Ch' of 'Charles'. 1) Frame-break right of signature. 2) White mark top-right from 'P' in 'STAMP'. 3) Mark on inner frame right of control Num. 4) Dot above 'ET' in 'STREET'. 5) Many other minor flaws as indicated. Row-1, Stamp 5             Controls: 3125, 3805, 4005 Small: 37885 Using my 'widget' at the bottom gives me: 3125 = 79/1/5 3805 = 96/1/5 4005 = 101/1/5 37885 = 948/1/5 3805 adjoins 3806 below. 37885 adjoins 37886 below.
37885	3805

Courtesy of Steve Lawrie.	One of mine.	SECONDARY flaws
		These have the following flaws in common. 1) Comma after 'Office' very weak. 2) White spot by frame right of 'T' in 'T' corner. 3) The frame-breaks on left side. Row-1, Stamp 6             Controls: 2806, 3486, 3806 Small: 37886 Using my 'widget' at the bottom gives me: 2806 = 71/1/6 3486 = 88/1/6 3806 = 96/1/6 37886 = 948/1/6 2806 forms a strip with 2807 and 2808 3806 adjoins 3805 above.
3486	3806

Courtesy of Steve Lawrie.	Courtesy of Andrew Higson.	SECONDARY flaws
		This shares the green flaws with 1/6 above. It shares the light blue flaws with 3/3 below, and has its own flaws shown in magenta. 1) Two white spots in 'T' corner. 2) The frame-breaks on left side. 3) Frame-break around 'L' corner. 4) Frame-break below 'Co' corner. 5) Black mark below right dot in 'T' corner. Row-1, Stamp 7             Controls: 1687, 2807, 4847 small: 31847 Using my 'widget' at the bottom gives me: 1687 = 43/1/7 2807 = 71/1/7 4847 = 122/1/7 31847 = 797/1/7 2807 forms a strip with 2806 and 2808 1687 shows part of the 2/7 stamp below with a frame-break in the right corner.
1687	2807