Most forgeries start with a photograph. The earliest so far identified are thought to date from the late 19th century They are usually then printed by some variant of the photo-lithographic process. Since the forgery is now two removes from the original, it is not surprising that there has been a loss of image quality. The original sharpness that came from the direct contact between paper, printing ink and copper has been lost. The lines will be blurry and, where close together in the original, will have run together.

However, there are recent indications that some modern forgers are using sophisticated techniques to replicate features of the original. At least two forgeries are known, printed from actual copper plates (with their genuine characteristics) that had been mechanically copied from an original model. [See the discussion of the 1624 Smith title-page map on the MapHist Illustration page ('duplicate Intaglio plate'), and the note on Ortelius America (variant (C)).]

The Italian firm < http://www.cose-belle.com/content-categories/cat-393/maps_and_instruments_of_observat.html > Cose-Belle may be offering examples printed from a modern plate:
"All maps are Made in Italy, outlined using a hand tooled copper press, painted by hand using water color, mounted on an antiquated, linen and cotton canvas in sections which fold easily and are encased in a soft bag, inside of a luxurious burgundy gift box with the map’s title written across the corner cover." [No website for Cose-Belle visible, February 2012.]

EXPLICABLE DIFFERENCES:
All early maps were hand-printed, one at a time. To do that, the copperplate or woodblock was cleaned and re-inked to produce each impression. During that process various things might happen to make one impression very slightly different from another. There might be too much or too little ink in one place, leading to smudging or faintness. A piece of dust might get onto the plate, the paper might be creased in the press, and so on. There could be lots of innocent reasons why the map you are looking at is very slightly different from a known original. And that is before considering what happened afterwards: the applying of colour by hand, folds, damage, staining, etc.

THOSE THAT INDICATE A COUNTERFEIT:
Each genuine impression will be different - even if it might take an expert with a magnifying glass to see that. Each forged map will be exactly the same, since they are machine-produced copies from a single photograph. That photograph will have captured the unique features (the 'copy-specific details') of a single original impression. Put two examples of the forgery together and you see those 'unique' features repeated. It is even possible to find that a forgery reproduces damage on the model (perhaps a tear or a small missing piece) whereas the forgery's paper is undamaged at that point. From my time at Weinreb & Douwma I recall two examples of that, both woodcut originals. The first was the Apian world map of 1520 and the second a map from the Ulm Ptolemy - particularly interesting because it reproduced a blemish found only in the Theatrum Orbis Terrarum facsimile edition of 1963.

Various devices are used to give a modern reproduction at least the superficial appearance of age. Tea or coffee are popular applications. One popular (and very cheap) series of John Speed maps was as brown as chestnut and, being on crinkled paper to give the supposed appearance of vellum, looked like overcooked poppadoms. In a message to Maphist on 13 December 2005, however, Daan Strebe described a more sophisticated attempt at deception (though without identifying the map involved):
"What caught my attention was the sudden revelation of typical and heavy aging. Yellowing. Mottling, particularly toward the edges. Foxing. Everything I would expect to see on a 400-year old map that had not been carefully preserved. Yet turning it away from the light, it reverts to its too-perfect condition. It is as if the map had been resurfaced somehow while somehow preserving the original image."

[Update 20 August 2008:] 'Old Map Tutorial' (Directory Aviva. Photoshop Tutorials. Learn About Using Adobe Photoshop) discussed a way of taking a 'simple black and white drawing and converting it into a mysterious, old map, lost for hundreds of years.' This was intended, presumably, for innocent web-use but the result (which looks as if it has suffered from the worst kind of 1860s brittle paper damage) is similar to some of the forgeries circulating on paper.

Original early maps were sometimes coloured, sometimes not. There is no general rule, but some maps are more usually found in original colouring (e.g. Blaeu), others in black and white (e.g. Speed). The modern colouring of maps dates, largely, from the 1930s onwards, and was directly linked to the sale of maps as individual sheets. Colouring was applied by hand, in much the same way as in earlier centuries (though the pigments were now artificial not natural). This was done because most purchasers wanted it and would pay more for a coloured example. So modern colouring does not make the underlying map suspect. If, however, the colour could be shown to be old, that of course would remove doubt about its authenticity. The most usually cited feature of early colouring is when certain colours (usually green), but not all, bleed through the paper and can be seen from the back. In extreme cases, e.g. roughly the period 1650-1750, the green may have eaten its way through the paper and caused breaks.

For a brief, illustrated article see Dieter Duncker's 'Antique Maps - Recognising the difference between old and modern colouring' on Kunstpedia.

Colour was not applied to maps by machine until the second half of the 19th century, and that can be the most obvious way to tell a modern reproduction. "Most early maps were colored by hand and brush strokes can often be discerned. Reproductions usually use printed halftones and the small dot patterns can be seen with a magnifying glass" (The Old Map Gallery).

Most original maps have no date; most fakes and some 'improved' forgeries are dated. Many starting collectors with little experience of early maps are worried by the absence of a date. They do not like to have to rely on somebody else for what they may feel is the map's most important feature, namely its age. Paradoxically, it leads them to doubt the genuineness of an undated map. A number of forgeries in this listing are really a kind of fake/forgery, in that a date, and sometimes the name of a publisher, have been added to what would otherwise have been a straightforward forgery. [Those instances are indicated by '(F)' after the date in the Index table.] Not only is a date no guarantee a map is genuine, it may itself be a faked element. Several manuscript maps are known to have had their dates altered, for a variety of possible reasons.

Most maps were originally published inside a book. If the map was illustrating a text, it would usually be folded (sometimes more than once) and stuck in, rather than forming an integral part of the paginated text. Such a map would show the lines of the original folds, even after conservation. [It would be right, however, to ask (in the light of recent thefts) how and why the map came to be separated from the book.] If the marks of the folds are visible across the surface of the map, but it is clear that the paper has never been folded, then this will be a forgery - based on a photograph that had picked up that detail.

It is more likely that the map started life in an atlas. Often it will have been folded vertically down the middle and pasted onto a 'guard', a blank strip sewn into the volume. Even after conservation, the mark of the fold, and the traces of paste down the spine, can usually be seen. A forgery will not have a genuine old fold. After being pressed together for centuries, the two sides of a genuine map may show signs of colour or ink offset from one side to the other. Matching staining or worm-holes are other indications of authenticity.

A responsible publisher, when producing a facsimile map, will place a printed statement to that effect. That would normally be outside the map border, at the bottom. However, if there was too big a gap between that text and the map it would be easy to cut that statement off (or hide it under a mount).

Always hold a supposedly pre-1800 map up to the light to see if it has chain-lines and watermark(s), or a date, or some indication of modern origin.

An expert can tell by the look and feel of paper if it is old or not. Others need a helping hand. The essential difference is that, around 1800, machine-made ('wove') paper (made from wood pulp) began to be introduced. When held up to the light, this has an even appearance, though there may be a watermark. It may well be lighter than old paper and almost certainly smoother.

The earlier, hand-made ('laid') paper has a grid of chain lines {the larger ones about 2.5 cm apart and the smaller ones just over 1 mm each}, to which one, or on a folio-sized map, two watermarks would be attached by wire. The chain lines reflect the tray in which the papermaker shook the rag-based material. Some forgeries do not make any effort to reproduce the chain lines of laid paper.

Paradoxically, old, hand-made paper, usually lasts much better than that made after 1800. The strong browning of Janssonius maps is one exception. Otherwise, an overall browned appearance is likely to represent the forger's (incorrect) idea of how old paper should look. However, an early map will usually have some damage or discolouration. The complete absence of these is likely to mean either that it has been heavily restored, or that it is counterfeit.

The study of watermarks is complicated and confusing. The essential guides are large and available only in a research library (apart from the 92,000 images of the "Piccard" Collection of Watermarks). Map publishers tended to continue issuing their maps over long periods, not unnaturally using different batches of paper. Therefore, attempts to tie a particular watermark to a particular atlas are usually not successful. And, of course, like everything else, a watermark can be imitated.

Some old paper carries a genuine date. Confusingly, though, the French continued to use the date 1742 for a long period. From the 1790s, English machine-made Whatman paper can be found with its date of manufacture. That is an invaluable bibliographical aid, rather than a mark of forgery.

Unfortunately, some forgeries are produced on genuinely old paper, supplies of which still exist (particularly for smaller sizes). For example, a book's blank end-paper might be used.

Some facsimile publishers, to ensure their products are not sold as originals, will use paper with an obviously modern watermark.

The platemark is the main identifying feature of a genuine copper engraving. It represents the edge of the copperplate from which the map was printed. However it can be imitated.

GENUINE PLATEMARK:
Since copper was expensive, the printed surface on a genuine map will usually go close to the edge of the plate, i.e. the platemark will be very small, perhaps 1-2mm in the case of an Ortelius. The plate will have been trimmed and then rounded off, leaving a slightly irregular line. The plate will also tend to collect small amounts of ink residue, which will appear as a grey background. That will stop at the edge of the plate.

Confusingly, Manasek (2001) describes how the platemark on a small map may not be visible. He illustrates an example of A. Bell's map of Asia where this is the case.

The area between the map edge and the platemark may have been trimmed off a genuine map, for example if a multi-sheet map was being joined up.

FAKE PLATEMARK:
Because the platemark is visually so important, forgers often add a false one. These tend to be much larger (10mm in the case of the Ortelius Iceland) and very regular. Instead of the paper being smooth and flattened inside the platemark and rougher and raised outside, there will be no change in paper appearance either side of the platemark. The fake 'platemark' will actually be just a simple groove. It is likely that the edge of the genuine platemark on the map that was photographed will still be visible, well inside the false addition. Fake platemarks will usually have been applied separately from the (usually) photolithographed map.

Many early maps appeared originally in atlases, for example those by Ortelius. Those maps had explanatory text on their reverse. Forgers did not normally attempt to reproduce that. However a recently discovered Ortelius map of Iceland (see van den Broecke, 2006) does include the text - the first that the author (the expert on Ortelius) had seen.

Parchment, calf-skin or vellum was widely used in the medieval period and, later, for cases where the map would be subject to hard use, for example, an estate map or a sea chart. Is there evidence of maps being produced on fake vellum?