Turkey: 1930 Sivas Railway Stamps: Surcharge Varieties

The 1930 Sivas Railway set is a wonderful source of errors and varieties. The major catalogues either list a few particularly noteworthy varieties or simply mention that many varieties exist. It seems to be symptomatic of the attitude that some countries' stamps are more important than others, when it comes to detailed listings, which I suspect is more due to commercial realities than to philatelic elitism. If my series of articles seem light on Europe, USA, Japan, Canada, Australia etc., it is because, for the most part, there is no shortage of information about them and the information is not too difficult to acquire. Not so Benin, not so Turkey, to name but two.

Amongst the varieties you will find listed below are errors of date, an error of value and wrong letters. I have also listed other varieties, from the significant to the very minor. It's your collection, so I leave it to you to decide which varieties are important and which you can 'live without' but I think there is every chance that, if you acquire a set of these stamps, you will, at the very least, find some of the minor varieties in it.

The table has a 'Yes' against any stamps I have seen or which have been reported to exist by reliable sources and a 'No' against stamps which are generally believed, or definitely known, not to exist. Some varieties, particularly some which I am not aware as having been previously listed, I have left blank. In some such cases, I have added some thoughts on the likely existence of other values. In this, I have been aided by being given access to complete sheets of the two 10P surcharges, one the wolf design, the other the train-on-bridge design. The train-on-bridge design, due to its being horizontal in design, had a quite different setting for the surcharge. I suspect it was done after the others, as it has more varieties, such as inverted 930, a likely result of trying to transfer whole sections of typeface across.

The other five designs are all vertical but the top four values (Head of Ataturk) are slightly larger and the surcharge is different again in appearance. I assume that the typeface was simply spaced wider apart to accommodate this, as most varieties are common to all designs and in the same positions in the sheet (except for the re-set train-on-bridge design).

Explanation of Vertical Columns in the Table below:
Column 1 (white): A letter I have assigned to each variety, solely for ease of reference
Column 2 (white): A brief description of the variety (expanded upon in the accompanying text)
Next 22 columns (except green column): The 22 values in the set, each stamp design represented by a different colour of background.
Column 5 (green): The sheet position for this variety in the train-on-bridge value (sheets of 100, 10 x 10)
Penultimate column (green): The sheet position for all the other five designs (all are sheets/100, 10 x 10)
Last Column (white): Repeat of Column 2 (as the page is too wide to see across and I am kind and considerate!)

In the first column of the table at the foot of this article, I have assigned a letter (or letters) to each of the varieties, simply for ease of reference. I have illustrated all the varieties and, under each scan, I have added a fuller description of the variety - plus some notes, where appropriate.

As part of the bold print in each descriptio , I have commented on how many values exist. Where I have specified the number of values recorded, there is very little doubt that these are the only values you will find with the variety in question. I have tended towards caution by not actually placing a 'No' in several boxes, where I might have done so wtih 99% confidence. To save you from constantly referring to the notes when looking at the table, I have added an asterisk (*) after each of the varieties where I am certain or 99% certain that the known varieties have all been recorded. Therefore, if there is an asterisk against that description, you may assume, fairly safely, that any blank box may be read as a 'No'. Conversely, if there is no asterisk, then there may well be other values with this variety of which I am unaware (but the notes on each variety may shed more detailed light on this).

Sivas - varieties on S A:- S-arrow: All 22 values recorded: There is a small tail, roughly like an upward-pointing arrowhead (sometimes fairly open, sometimes a thick triangle of ink),just touching (sometimes just clear) of the centre of the base of the large 'S' in 'Sivas'. This is a slightly tricky identification as there are several examples per sheet of this letter having some sort of 'tail' at this point. The most noteworthy of the other marks is often, but not always, a quite substantial one in position 17 (except the railway bridge, where it is in position 13). This tends to be more like a cedilla and is always attached.

B:- Broken 'S': Quantity Unknown: I have seen this variety only on the values recorded in the table below. It consists of a break (inkless strip completely breaking the letter) near the top of the diagonal stroke of the large 'S' of 'Sivas'. It is not present in the complete sheet of the 10 pa train-on-bridge design that I have examined. There is a small possibility that this was a broken letter which needed to be replaced when the train-on-bridge design setting of the surcharge was prepared (assuming this was the last design printed). There is a possibility that this accounts for the Divas variety, present only on the bridge design, but this is only speculation. I have put a 'No?' against the 10 Pa bridge in the table, the (?) because there is a small possibility that it exists in part of the printing.

C:- SLvas : All 22 values recorded: In fact it is Slvas, with a small 'l' instead of an 'i'. This sometimes appears as an 'l' (a small 'L', not the numeral '1') and sometimes like an 'i' with the upstroke and the dot joined, with a slight tapering effect where the two parts meet.

D:- Sivae: Exists only on 5 values: 'e' for small 's' in 'Sivas'

E:- Sivav: Exists only on 5 values: 'v' for small 's' in 'Sivas'

F:- Si vas: 2 values recorded: Wide space between 'Si' and 'vas'. I have only seen the two values with this variety and they coincide with the only two I have seen listed. Unlikely that there are any other values. Illustrated, a pair, normal on the left, Si vas variety on the right.

G:- Siva6: Quantity Unknown: Not stictly a '6' but the lower hook of the second 's' in 'Sivas' almost closed by an extension of the hook. The same stamps exhibit a second minor variety, the base of the large 'S' in 'Sivas' ending in an arrowhead, due to damage to the letter. Evidence to date suggests it may well be a variety unique to the train-on-bridge design

H:- S ivas: Quantity Unknown: I only know of one example of this, wide spacing between 'S' and ' i ' in 'Sivas', as shown in the table. It is possibly only found in the Head of Ataturk design, created when the plates were slightly altered to accommodate the slightly larger size of this design.

J:- Chipped D: All 22 values recorded: The damage is at the top left corner of the 'D'. In some cases, this amounts to little more than the top left point of the 'D' being a little rounded. This damge perhaps worsened during the printing. Certainly, on the train-on-bridge values, which I believe to have been the last designs surcharged, there is much more extensive damage to this part of the 'D'. Illustrated, a pair of 3k/4k, the right stamp with the damaged 'D', an example with this variety at its least impressive.

K:- D no stop: 17 values recorded: No stop after D (D Y.)

L:- D raised stop: 9 values recorded: The stop after the 'D' is raised and is located very close to the top left corner of the 'Y'

M:- D, Y.: Quantity Unknown: D followed by a comma instead of full stop. In the 10pa/20pa train-on-bridge design complete sheet/100 which I have examined, this variety is not present, despite being present on the 2.5k/3k, hence the 'No?' against the 10pa/20pa in the table, the '?' in case it possibly exists in part of the printing - unlikely.

N:- Y no stop: 6 values recorded: No full stop after Y, so D. Y

P:- Y comma: Quantity Unknown: I have only seen the one value with this variety and it is not present on the full sheet of the 10 pa / 20 pa in the same (train) design. Same comment as D,Y. above - this variety is not present in the full sheet of the 10 pa / 20 pa that I have seen in the same (train) design, hence the 'No?' against this in the table, the '?' in case it possibly exists in part of the printing - unlikely.

Q:- 30 ag.930: 19 values recorded: Curiously, this variety does not appear on just three of the values, all different designs. In this variety, the 'ag.' is right up against the '930'

R:- 30ag. 930: All 22 values recorded: This spacing variety is much less pronounced than in the above case. The space between '30' and 'ag.' is perhaps equal to about 65%-70% of the space between 'ag.' and '930'. In the normal stamp, the three elements of this line of overprint are equally spaced. There are several examples of this variety per sheet, although most of them are not quite as pronounced as in the illustrated example above. In the train values, there are few examples of this variety per sheet, whereas in all the other designs, most of the stamps in the first column display it The sheet position for the variety in the table below is for the best example in each format.

S:- Accent on 'g' of 'ag': All 22 values recorded: This variety takes the form of a thin half-moon accent on the letter 'g'.

T:- '0' of '930' solid colour: I have seen this on only two values, the 1k and the 20k. However, the surcharges applied in red (10pa wolf, 1k wolf, 1.5k train, 2k train, 5k wolf, 7.5k hill, 15k tracks, 17.5k hill, 20k tracks, 50k head, 100k head and 250k head) tend to have more varieties, due solely to the nature of the ink. As these are not constant varieties, I have not included them in the table below, in keeping with normal practice. There are also numerous examples of stamps where the surcharge is significantly misplaced, particularly it seems with the 6k tracks; these, too, have been omitted as they are not constant.

U:- '309' for '930': 11 values recorded: A major variety, and one which is listed by Michel

V:- '930' inverted: 4 values recorded: Another major variety, again listed by Michel

W:- '390' for '930': All 22 values recorded: Another major variety and the third of the three varieties listed by Michel. THE TRAIN VALUES HAVE AN ADDITIONAL VARIETY ON THIS STAMP 'Divas' for 'Sivas'

X:- Surcharged 9k instead of 6k: 1 value only: This stamp is illustrated at the top of this article in a block/4. Also in this block are examples (the two left-hand stamps) of the 'ag.' closer to '30' variety. You will note the difference in spacing is relatively small. The 9k error is, in my opinion, the most important variety in this set, an error of value, yet it is all but ignored by most of the major catalogues. You may be sure that, if this were a Western European or North American error, it would be numbered, priced and much valued !

Y:- Italic '4' in '4k': 1 value only: Hard to see and I have therefore provided a negative image of the scan to display it better, resulting in the stamp appearing in entirely wrong colours. Does anyone know the sheet position for this variety, please? Illustrated above, in order, are the normal 4 (used with 30ag variety), the Italic '4' and another copy which may have been an attempt to correct the italic '4', a stamp which also illustrates the fairly common variety, weak or partly- missing base to the 'g' of 'ag'.

Z:- Comma after the surcharged value: I have only seen 1 value with this variety: There are only three values surcharged in Paras and I have not found this variety on the 10 para train sheet. It may well be a variety unique to the 10p/10p. This image is also shown in negative; the stamp is in fact green, with a red surcharge!

(-) D.Y. normal: 70 per sheet : On the train values only, there is a mixture of round and square full stops after D and Y (which stands for 'Demir Yol', I believe, the Turkish for 'Railway'). The first scan shows the 'Normal' format, two square stops.

AA:- D square stop Y round stop: 8 per sheet: The scarcest of the full stop formats after D and Y

BB:- D round stop Y square stop: 22 per sheet: The sheet quantities for each type I have given coincide with the generally accepted view of the distribution of these types. In some cases, stops are a little difficult to define with total accuracy and confidence.

Other Varieties:
Raised final 's' in 'Sivas' (seen on the 30k/40k)
Short top to right arm of 'Y' (seen on 10P/10P and 1k/1.5k)
Numerous other damaged, deformed and badly printed letters !

In the table below, the vast majority of the spaces left blank (i.e. without a 'Yes' or 'No') are probably stamps which do not exist (but see the text above for detailed evaluation):

. Surcharged........: 10pa 10pa .. 20pa 1 k 1.5k 2k 2.5k 3k 4k 5k 6k 7.5k 12.5k 15k 17.5k 20k 25k 30k 40k 50k 100k 250k .. ...........................
. * = Listing very likely complete Wolf Loco .. Loco Wolf Loco Loco Loco Wolf Hill Wolf Rail Hill Hill Rail Hill Rail Hill Rail Head Head Head Head .. * = Listing very likely complete
. SG number 1099 1100 .. 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 .. Variety
A S-arrow* Yes Yes 49 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 35 S-arrow*
B Broken 'S' Yes No? .. . Yes . . . . . Yes . . Yes . . . . Yes . . . . 69 Broken 'S'
C SLvas* Yes Yes 79 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 22 SLvas*
D Sivae* No Yes 26 Yes No Yes Yes Yes No No No No No No No No No No No No No No No .. Sivae*
E Sivav* No Yes 35 Yes No Yes Yes Yes No No No No No No No No No No No No. No No No .. Sivav*
F Siva6 . Yes 20 . . . Yes Yes . . . . . . . . . . . . . . . .. Siva6
G Si vas . . .. . . . . . . . . . . . Yes . Yes . . . . . . 47 Si vas
H S ivas . . .. . . . . . . . . . . . . . . . . . Yes . . ?? S ivas
J Chipped D* Yes Yes 37 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 90 Chipped D*
K D-no stop* Yes No .. No Yes No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 44 D-no stop*
L D-raised stop* Yes . .. . Yes . . . . . Yes . Yes Yes Yes Yes Yes Yes . . . . . 40 D-raised stop*
M D, Y. Yes No? ?? . Yes . . Yes . Yes Yes . Yes Yes . Yes . Yes Yes . . . . 24 D, Y.
N Y no stop* Yes . .. . Yes . . . . . Yes . . . . . . . . . Yes Yes Yes .. Y no stop*
P Y comma . No? ?? . . . . Yes . . . . . . . . . . . . . . . .. Y comma
Q 30 ag.930* Yes Yes 73 Yes Yes Yes Yes Yes No No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 57 30 ag.930*
R 30ag. 930* Yes Yes 89 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 91 30ag. 930*
S accent on g* Yes Yes 11 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 9 accent on g*
T 0 of 930 solid . . . . Yes . . . . . . . . . . . Yes . . . . . . ? O of 930 solid
U 309 for 930* No Yes 56 Yes Yes Yes Yes Yes No No Yes No No No No No No No No Yes Yes Yes Yes 14 309 for 930*
V 930 inverted* No Yes 98 Yes No Yes Yes No No No No No No No No No No No No No No No No .. 930 inverted*
W 390 for 930* Yes Yes 85 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 45 390 for 930*
X 9k for 6k* No No .. No No No No No No No No YES No No No No No No No No No No No ?? 9k for 6k*
Y Italic 4(k)* No No .. No No No No No No YES No No No No No No No No No No No No No 62 Italic 4(k)*
Z Comma after val. Yes . .. . . . . . . . . . . . . . . . . . . . . 64 Comma after val.
AA Sq+Rnd stops* No Yes .. Yes No Yes Yes Yes No No No No No No No No No No No No No No No .. Sq+Rnd stops*
AB Rnd+Sq stops* No Yes .. Yes No Yes Yes Yes No No No No No No No No No No No No No No No .. Rnd+Sq stops*
. Michel Nr. 913 914 .. 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934
. Surcharged: 10pa 10pa .. 20pa 1 k 1.5k 2k 2.5k 3k 4k 5k 6k 7.5k 12.5k 15k 17.5k 20k 25k 30k 40k 50k 100k 250k
. . Wolf Loco .. Loco Wolf Loco Loco Loco Wolf Hill Wolf Rail Hill Hill Rail Hill Rail Hill Rail Head Head Head Head

I would be surprised if inspection of sets by various railway thematic collectors worldwide did not shed a little more light on this fascinating issue. In common with normal practice, my listing has concentrated on what appear to be CONSTANT varieties, rather than freak errors which may have only affected one stamp or one sheet.
Please see our retail list of railways, accessed via our home page, for a list of varieties currently in stock. If the whole thing seems a little too large and overwhelming, here's a thought. It is possible to put together a complete set of 22 stamps, one of each value, each stamp showing a different variety!

I suspect it would be possible to print out the table above on a single sheet in 'landscape' format, rather than the standard 'portrait' page set-up. Those of you who are more gifted than I with computers might know how to do it, simply and quickly! Neither Netscape nor Internet Explorer seem to provide this facility. In case it helps, I have reproduced the same table, but on its own, on a separate web page, where you can perhaps manipulate it more easily. If you know either how to send it to a printer in landscape format, or, even better, how to produce it sideways on the web page, please let me know.
Click here for table only, opens in new tab

  • Return to Listing of all Railway Articles (with links). Note: You may already have an open tab