First Contact

Map    Postal History    Watercolors    1877-78    New Rectangulars    Implements
Reading Covers    Non-postals    Forgery    Certificates    Defacement

The stamp collector who is familiar with finely-wrought engravings of the 19th century will likely find himself at a loss when first dealing with the early handstamped impressions of Jammu & Kashmir. Truth to tell, some of us never quite regain composure. These curious stamps were done in watercolor pigments, which may still smear in high humidity.

Well, know at least that the paper is a gloriously resilient local production that may outlive all other stamp papers. We are spared the usual frets about hinge marks (no gum) and short perfs (no perfs). We do still care about good centering, wide margins, and clean impressions in all but the rarities. The image above, rare as rare on cover, is that of the ½-anna yellow watercolor circular, SG23, a detail from a railway cover to Calcutta, 17 July 1876, collection Hellrigl.

But let us not distort things too much; along its own dimensions, much Kashmir material is attractive indeed. The scan above is taken from the cover of the Dawson auction catalog 1967. In addition to the workhorse ½a ultramarine, this cover shows the 4-annas “myrtle green” watercolor, which is attested perhaps only once more on cover. Four annas was the registered rate.

Post-watercolor fare is pedestrian by comparison, and some of it is downright homely. But it is subtle and complex, and like some homely people has a way of growing on one over time.

The price-to-rarity ratio of J&K stamps is still remarkably low compared with that of more popular and better understood collecting areas. The collector of ordinary means and modest patience can acquire items of true rarity. Only a hundred or so copies of certain of the middle-priced catalogue entries are likely existent now, while much of the >£600 fare is attested by fewer than a couple of dozen decent copies. Disconcertingly diverse material in the way of shades and papers is often bundled up under a single entry in the catalogues so that rare subtypes pass from hand to hand at much lower prices than they might. The number of extant covers in the West surely cannot exceed a couple of thousand, including postcards.

Our field lacks an up-to-date specialist handbook. The most detailed work is that of A.J. Séfi and C.H. Mortimer, which was published posthumously in 1937. After more than eight decades since the time of the writing it is still the master reference, though it has naturally become dated in parts, particularly in the accounting of the postal markings. Philatelic auctions report that only 100 copies of the book were printed. (The text is reproduced here on-site.) Séfi also produced works on Grenada [1912] and Malta [1913]. Clearly a gentleman of tastes.

So Where Are We?

And when? The map below (not really meant to be legible) shows the administrative lay of the land, and our stamping ground, in the mid-1890s when the postal arrangements of the entire region, Kashmir, Jammu, and the Punjab, were formalized under British authority. Our story pertains to the preceding three decades (1866-1894) when Jammu-Kashmir issued its own stamps. The British post office worked in a kind of collaboration with these native offices for handling mail crossing the border to more overt British territory. Puńch nestles in there somewhere too, and she issued her own stamps for part of the same period. These were also watercolors, and the designs may have been cut by the same seal cutter, a certain Rahat Ju of Srinagar. Map: D. Appleton Co. NY, 1890s:

Kashmir and Jammu are respectively the northern and southern provinces of an unlikely State that straddles the western-most reaches of the Himalayas. In our postal period, the rough distance across this domain, either east-west or north-south was about 300 miles, Himalayan-crow fashion.

Despite the manifest presence of the Himalayas, J&K philately sits uncomfortably under the purview of “Himalayan philately,” for most of the postal activity looked south to the hot plains of the British Punjab and beyond. Tibet in the east was non-existent so far as the postal record of our period goes, and there are only a few Kashmir-Afghan covers still extant. That is not so surprising considering the physical barriers in all directions but south, which were daunting enough. There is not exactly a mountain of material left to us of from local mailings within the State. Such “internal covers” tend to have a leg-up in philatelic cachet, and naturally the more so the more remote the postal routes.

Internal mailings between Srinagar (the summer capital) and Jammu (the city, the winter capital) would take only a couple of days (the yellow route on the map when that became available) with relays of runners pressing on through the night with torches. Those important local routes were onerous enough, and snow often blocked the high passes, such as the 8,600' Hājī Pīr north of Puńch, the 8,900' Banihāl (which means ‘blizzard’), or the notable Pīr Panjāl at 12,200'. The philatelically significant town of Leh in Ladakh is east, off the map at the upper right. From Srinagar, you have to go via the Zoji La at 11,500'.

Though the northern frontier is formed from the Hindu Kush and the frightful Karakoram range (with a pass at 18,000') there was a curious and sporadic postal connection between Yarkand in Chinese Turkestan to Hoshiarpur in India along an old traders’ route. Since the route passed through Ladakh, Kashmir postage was added at Leh where the item entered the postal system for onward transmission. The passage one way took a couple of months by caravan.

Old Srinagar on the Jhelum. Srinagar was the northern/summer capital of the union. The town itself was often referred to as “Cashmere,” and the early British postmarks used at Srinagar were so inscribed. The population in 1891 was about 120,000. This scan of Srinagar’s Third City Bridge is taken from Winthrop Boggs’ 1941 philatelic article, which is reproduced on-site.

Modern Jammu on the Tawi. Jammu was the southern/winter capital of the union. The river, which was crossed by a suspension bridge in our postal period, flows roughly southwest in a narrow ravine to join the Chenab. The population in 1891 was about 35,000. Our thanks to Paul La Porte of Chicago for making this painterly photograph available to the public.

Panj-āb itself means “five waters,” so-named for some larger five of the many tributaries of the Indus River. By tradition, these rivers are: Jhelum, Chenab, Ravi, Sutlej, and Beas. The Jhelum was known to the Greeks as the Hydaspes, reached by the armies of Alexander. The Punjab came under British control in the spring of 1849, and so-held for about a century.

Winthrop Boggs also provides this photo of a “typical Kashmir post office.” Maybe it is really one of the many chaukis, the kind of postal runners’ depot that eventually dotted the main routes. The Stamp Specialist, Blue Book, p 57 (1941).

Postal History

Postal history is the soul of the subject, and real discovery in this curious philatelic niche is open to all comers; indeed, especially now, internet and all.

Each province supported about three dozen POs, but covers attesting to some of them are hardly known. British Indian postage normally had to accompany native stamps on mail crossing the border. There was also an autonomous “Maharaja’s Post” to a few towns in Punjab and mail thus borne avoided British postage. So-called “javab” covers, so named for a characteristic manuscript notation that they bear, also often crossed the border without British postage. A number of prominent specialists had thus misconstrued a number of these as internal mailings, including the notable example upscreen with the myrtle stamp. That some javabs did carry British postage is one of our (many) unsolved puzzles.

Above: British Indian and Kashmir stamps together. This detail is from a letter that took the most popular route of all: Srinagar to Jammu and onward via Sialkot in India to Amritsar. Apart from being a notable religious site, Amritsar was an important commercial junction for the fabric trades. One will find many a Kashmir cover destined to the station at Katra Ahluwalia in Amritsar’s woolens district (somewhere north of Golden Temple and beside is own nice body of water, says Google Map).

Sialkot and Amritsar are about 70 or so miles apart. Relays of runners would take about half a day to traverse the route between them. The short Sialkot to Jammu rail extension (about 27 miles) did not exist for most of our postal period. From some high vantage point in Sialkot one can see Jammu (unless, that is, one cannot). The main stretch of the line between Lahore and Gujarat on the map (and onward ultimately to Peshawar) had already been built by the end of the 1870s. The Wazirabad-Sialkot branch was completed in 1880, and the extension to Jammu in early 1889. The map detail is from an early 1890s Constable.

There are a couple of desirable railway sorting-office markings that attest to mail on these two sections of line, the “L.7” for the Wazirabad-Sialkot branch, and the “L.51” for the Sialkot to Jammu extension (ref. Bard). The internet has repetitious entries to the effect that the Sialkot to Jammu branch was completed in 1890, though the datestamp shown above is from a May 1889 cover. (Most every claim on these pages thus comes with an invisible tag: Is it really so?)

The Jammu-Sialkot railroad link became defunct at the time of Partition in 1947, with Jammu remaining in India and Sialkot going to Pakistan. The India and Pakistan border also interrupts the line between Amritsar and Lahore, and thus (by a topological theorem) the old runner line that connected Sialkot and Amritsar. Routes in the west were also broken. Just as Jammu was divorced from Sialkot, so Puńch became separated from its erstwhile postal affiliate, Kahuta. One stretch of the current India-Pakistan border zigzags twice across the old runner’s line north of Puńch. Those of us who prefer to dwell in the 19th century find this new-fangled geometry passing strange.

Kashmir mail destined abroad to locations other than the British Isles are very scarce. Covers to or from Germany, Austria[-Hungary], France, and the USA are commanding high prices these days. And come to think upon’t, what others abroad are there? There just must be something to the Orient somewhere, or South-East Asia?

The Watercolor Decade 1866-76

Three triplets of implements were used for the production of the watercolors: Three circular dies at Jammu, three rectangular dies at Srinagar, and three composite plates (one for Jammu, two for Srinagar). By ‘composite’ is meant that two different denominations are accomodated on a single plate. The dating of covers of this early period is hampered by the fact that the native cancellation was done by pen-marking or with three dateless seals (described next.) The addressors of mail, however, could often be counted on to write the date somewhere on the envelope, usually in the lower-left part, in either the Hindu Samvat or the Mohammedan Hejri system, occasionally both.

The Kashmir Seal: A brick-red circular marking means cancelled at Srinagar, 1866-77. The same implement was used with black in the 1877-79 period.

Two Jammu Seals: The first cancelling implement at Jammu was a circular seal used with purple pigment in the 1866-68 period and with black in the 1868-70 period. That seal was superseded by a square seal with cut-off corners (so technically an octagon). This implement had been fashioned for use at an iron mine some eight years before the advent of the native postal service itself. Yes, an iron mine; such are the charms of J&K philatelly. We have tried to find out about the existence of an iron mine, a name and whereabouts, etc., but without real success, so far, except for a reference to some iron-works at Soap, “the ore for which is obtained from a bed of impure calcareous limonite intercalated in the limestones and rocks to the east of Achibal,” if you want to know. (Tom D. Latouche, May 1890.)

Iron-Mine Seal as Postage Stamp? Item left is what is meant by the catalogue entry SG86. It was always self-obliterated in black. Hellrigl collection. The item at the right is a bit of waste paper that was produced in the modern period from the catalogue’s schematic for the item in question. Beady-eyed specialists make the distinction.


Beginning in late April or early May of 1877 at Jammu, lively spates of paper and pigment experiments, mostly in oilcolors on native or on new European laid paper, were carried out with the original three circular dies and the Jammu plate. Which of the test material found its way into actual postal use (what philatelists call an “issue”) appears to have been a matter of some accident. Some of this production was in clear anticipation of a new regime of plate-printing on European paper that was to ensue in May 1878.

Many of the oilcolor transitionals are delightfully awful, but at least they don’t drip. The stamp shown above, in fact, is a particular delight for those who have been properly reared. The only thing wrong with it is that it is not postally used with some black smudge, and the catalogue even agrees.

Paper and pigment experiments were likely done at Srinagar during this same period. These productions included watercolors on a variety of unusual European papers, both laid varieties and woves. Unlike the Jammu operation, the Srinagar office was evidently more careful in not allowing this test material out as actual postage. Here are candidate examples:

Left: The 2as yellow watercolor on thin pelure wove, collection Hellrigl. Middle: Our own mystery item, a 2as yellow on thin horizontally laid paper. Right: The 1a chestnut [?] watercolor on sturdy wove paper, unrecorded, from the Lunn Collection.

New Rectangulars 1878-94

The Old Period ended in May 1878 with the introduction of new plates and new cancelling implements, which now bear despatch and dating information. Five new plates came into use, with a laggard to follow five years later. Even though this later material is generally much more accessible to collectors, it has been studied with less penetration—in print at least. Even with the help of better triangulation from a broader range of postal markings, much remains murky about the dating of varieties. Some of the material is far (far) rarer than catalogues would intimate, especially on dated cover.

The item on the left was printed at Jammu in a shade similar to that of the transitional circular oilcolor shown with it, both spring 1878. Indeed some of the New Rectangular “inks” seem to be just slightly thinned-out oilcolor. Some of the late circulars were even printed on precisely the same type of European laid paper that characterizes the first of the New Rectangulars.

This item, true to the new-is-always-worse theory in stamps, was printed from the same plate as was the preceding, but on thin wove paper at Srinagar. Notice the impression of a repairing screwhead in the floral margin. The plates may have been rebedded to a new base after their removal to Srinagar, perhaps from April 1881, when Srinagar resumed its annual stint of being capital of the State. The large stock of stamps that had been optimistically produced in this early Srinagar period were partially depleted during the dying years of the native post office, 1890-94. The term “re-issue” is often used to describe the late deployment of this early stock. (The earliest we have seen is March 1890.) Still, oranges are sometimes seen in the intervening years; notable instances are the rare bisects:

A unique Srinagar-to-Leh bisect with a 1a orange (one supposes) serving as ½a postage was mailed in September 1886, both late and early for an orange. Collection Bard. The catalogued bisects go in the other direction, from Leh. A bisected ½a orange is known that did do postal service as a full ½a, not ¼a.

New Colors. After 1883 the different denominations could be distinguished by color, and a new eighth-anna plate used with yellow pigments was introduced in that year. The new colors were slow to appear in postal use, and examples from their advent year are scarce, which is not so strange if their purpose was more philatelic than postal. In this way the New Colors stand as conceptual counterparts to the Special Printings of the old watercolor period.

The ⅛a mustard yellow on thin toned-wove paper. The failure with some of these yellows was due to turmeric mixed in the pigment. As with the yellows of the Special Printings, the pigments proved a particular failure, and like the others too, postally used copies are notably the scarcer. By the way, Gulmarg was a mountain retreat not far from Srinagar. There was enough foreign business (with sufficient clout at least) that a British Srinagar branch post office was set up there. The ⅛a was intended for the visitors’ half-rate privilege on postcards, though it is hardly seen in that usage (certainly we have never encountered one.) More often they are seen in multiples to make up larger amounts of postage on regular mail. That, and for philatelic concoctions for the unwary.

This printing in pink was done in the very late period, just prior to the closing of the native offices in 1894. Since there was a good deal of early (pre-New Color) stock that was still being sopped up for postage, a number of these late items were likely also intended more for the philatelic market; certainly some are scarce or unknown in postally used condition.

Postcards. The New Colors period was also endowed, not to say blessed exactly, with a native postcard of ¼-anna denomination. It was printed from 1883 onward in different sizes, papers, colors, shades, and plate states. Production anomalies, such as a back printing and bounce impressions are attested. Dating is sorely complicated by the fact that older stock, as told by plate states, appeared regularly in the mails alongside later stock.

Above: The four main plate-states (I-IV) of the postcard plate owe themselves to the fact that the block of inscription that included the Dogra sun symbol started shifting downward relative to the block of inscription to the right. Dating uncertain, probably 1891. After the configuration became perilously askew (State III), a repairing rivet was driven into the left section to bring it back into line, thus yielding State IV, a return to State-I positioning but augmented with the pigment dot lying to the lower-left of the sun-emblem.

Spook Tale. Certain New Rectangular ½a blacks appear and disappear from the stocks of collectors on certain anniversaries (typically late autumn when the wind is high) and invariably without an owner’s awareness. The story behind these disappearances and invasions remains mysterious.

Summary of Printing Implements
Our TagPlate or DieDescriptionAdvent
A1st Jammu Die½a circularMar 1866
B2nd Jammu Die1a circularMar 1866
C3rd Jammu Die4as circularApr? 1866
D1st Kashmir Die½a rectangularSep? 1866
E1st Kashmir Plate½a rectangulars (20 in upper sector)Oct? 1866
F1st Kashmir Plate1a rectangulars (5 in bottom strip)Oct? 1866
G2nd Kashmir Plate¼a rectangulars (5 in top strip)Jun? 1867
H2nd Kashmir Plate2as rectangulars (5 in bottom strip)Jun? 1867
IIron-mine Sealas “½a”Sep? 1877
JJammu Plate½a+1a rectangulars (3+1)Aug? 1867
K2nd Kashmir Die4as rectangularOct? 1867
L3rd Kashmir Die8as rectangularOct? 1867
M1st New Plate½a (15 rectangulars)May 1878
N2nd New Plate¼a (15 rectangulars)May 1878
P3rd New Plate1a (20 rectangulars)May 1878
Q4th New Plate2as (20 rectangulars)May 1878
R5th New Plate4as (8 in upper sector)1879?
S5th New Plate8as (8 in lower sector)1879?
T6th New Plate⅛a (15 rectangulars)1883
U“Unissued” Plate¼a (12 rectangulars)1886?

Reading Covers

“Top edge eaten by rat, but still looking very good.”
—anonymous J&K auction dealer

The covers collector will definitely need to decipher a certain amount of Dogri and Indo-Persian script for learning dates and destinations. Having no real knowledge of the languages is no bar to espying matters of philatelic note. Familiar old covers gradually give up more of their secrets, and the puzzles left lurking are footholds for the next stage.

Right to left (the direction of the writing): az Jammūn, az Kashmīr, az Ladākh, Srinagar, and dar Amritsar. The az is Persian for “from,” and dar is “to.” Many destinations or despatch points simply have to be recognized as recurrent logograms, so cursive they may be. There is a glossary of such things on-site.

Sialkot markings in Dogri script. Sialkot operated what was in effect an extraterritorial office for Jammu for some of our period. A postage due seal (above left) and an inseparable ‘duplex’ obliterator and datestamp (right) have come down to us from this operation. Syalkoţ is inscribed at the top of the datestamp, followed by the Hindu month maghar. Strangely, a number of prominent collectors of the past did not care to broach matters Dogri, and for many decades the duplex was thought to have been a Jammu marking proper. Speaking of such:

Both Masson’s and Séfi & Mortimer’s influential works listed this marking as being from an unidentified office of Jammu Province. While the name at the top is usually a blobby affair (left) there are many examples that are legible enough (right). In fact the name is Srinagar. One reason for not suspecting a Srinagar venue was that overt symbols of Dogra power were diplomatically downplayed in largely Mohammedan locales. Still, for such a prominent postmark, it was a curious and long-lasting lapse. Postal use alone should have made the matter clear. Belated realizations of such basic facts is a common experience for us J&Kers. May we all look forward to kicking ourselves about other blindness in the near future.

Hmmmm. Some forms of the Dogri or Takri class of scripts, especially certain cursives casually known as “merchant scripts,” are notoriously difficult to decipher. Most specialists do learn to pick out dating information and the odd postal destination, but seldom more. Some of it may well harbor matter of postal significance that has so far eluded us.

Non-postal Productions

A good deal of material produced with the official printing implements never saw use as actual postage. It ranges from such rarified fare as unique engraver’s proofs to lowly reprints in wrong colors (“fancy colors” is the literature’s term) on a circus of wove papers. Between the extremes, there is interesting and disparate fare of uncertain status, common to rare, for which the tradition likes the term “printing trial,” a term sufficiently safe no matter the detailed story.

Srinagar likely assumed control over the three Jammu circular dies in the spring of 1881 when it reassumed its position as capital of the union. A deluge of oilcolor reprinting on both native and wove papers occurred sporadically over the next decade.

Circular and Kashmir Rectangular reprints on native paper in dull orange, a non-postal color for the old period. Since a rather similar orange is seen on the 4as and 8as New Rectangulars of 1881 on thin wove paper, perhaps the preparation of these reprints (they are quite common) was something of a pigment trial at the same time. The 1a and 4as circulars also come in this way.

While no wove is troublesome, the native-paper reprints in the right colors have always hindered the fortunes of J&K philately. A couple of rules-of-thumb have come down to us for making the distinction, such as the thinner, more-polished character of the “reprint paper.” For items not on such paper, we are instructed to take a dim view of overly sharp impressions, hardly the happy happenstance given the presumed existence of smudgy reprints on typical native paper as well as the more adept productions of the transition year itself, some of which might well have been experimental, i.e., non-postals of decidedly different stripe.


The nature of the J&K material has invited its share of skulduggery. There has been a deluge of modern facsimiles taken from the old illustrations in the SG catalogue, but these are no threat. The most common type of early forgery is known as the “missing-die” type, the result (as the ► received story has it) of corruption in the native post office. The prospective collector must learn to recognize this and other bogus material from fifty paces by day one, for seldom is the week that eBay and others are not proffering some of it at bad prices. Of course they can also be knowingly collected, and some of them are definitely rare. Not amusing is the prospect of able new forgery. The middle price range (in the logarithmic sense) is the chief worry, for it represents items most dangerous to the ambitious beginner.

Above: A missing-die forgery on the left compared with the authentic design on the right. There are many differences, but notice particularly the curlicue 3-like Dogri element near the top. That is the ḍa- in Dawk Jammu ~ Jammu Post. Its lower tail swoops left in the forgery to touch the neighboring Persian element, while in the original that tail is tucked well back, giving a markedly different slant to the element. Five other of the old-period designs have “missing-die versions”.

It is fortunate that a few basic cues eliminate vast amounts of bogus matter. The field is not nearly so difficult in this respect as is sometimes intimated. Still, what were once widely accepted to be the rare first issues of the State (the ‘Die I’) were accepted eventually as forgeries. It is reported that £50 prices were paid in the 19th-century for examples, clearly a kingly sum for a stamp in those days. Today they are certainly not more plentiful (one hopes) and are of rather indeterminate market value:

The 4as indigo and ½a sap green ‘Die I’ watercolor forgeries. That such were ‘rough sketch’ essays for the first issue is unlikely. For one thing, the Persian betrays features that look slavishly copied from the issued stamps, yet by someone who had no sense about what it was he was copying. It is conceivable that other revelations like the Die I saga lurk today. The nature of the Die I was revealed in 1899 by Sir David Parkes Masson (1846-1915), a figure who still bestrides our subject.

Were it not for Masson’s retrieval, preservation, and analysis of some implausible fraction of the covers that remains to us, J&K stampdom would be impoverished indeed. It is much to his credit, especially given philatelic practice of his day, that he did not soak off or cut stamps away from their covers. Incorrect dating notation in red ink seen on many a cover of the early period are, alas, also his. The image above was nicked from the internet in yet another of its serial thefts. The signature was from a self-mailing sent on 29 May 1890 at Srinagar and which arrived back to him the next day. The full address, apart from his name, was simply Srinagar, such were the days. Jaiswal collection.


A growing danger is the work of a latter-day breed of expertizers (expertisers) who are not qualified for J&K work. Doubtful or debatable certificates are to be expected with some of this material to be sure. The “royal blue” problem is notorious, and starkly different items are armed with contending certificates. But we should be spared the absurd, examples of which accumulate. One cover bears a clear and unproblematic despatch date in Persian, which the self-styled expert simply ignored, construing instead an ambiguous datestamp numeral in a way that would thrust the item into the pre-stamp period. Without comment, the cover was thus certified as bearing an impossible date. Quis custodiet ipsos custodes?

Defacement Proofs and Reprints

In February 1898, all of the printing implements save the Kashmir ½a single die were officially mutilated with deep gashes to the metal. Some ten impressions were made of each, thus making for rarified auction fare today.

Modern Restrikes: The Staal-Sharmas

The wounded implements came to rest at the Sri Pratap Singh Museum in Srinagar. A few impressions of each in purple and black ink were made in August 1981 by Drs Frits Staal and B.P. Sharma in a wonderful summer adventure. The story is recounted in their essay, “Five Fruitful Days in Srinagar” in Staal’s text, The Stamps of Jammu and Kashmir (1983), a must for any J&K collector. The scan is of the ½a+1a Jammu Composite Plate. Five such reprints were produced from this implement. Reference: Staal pp. 148-59.

To top of page