Skip to content


Anaglyptography.  A mechanical process of converting a three-dimensional design – an anaglyph – to a line drawing, or an engraved plate. The process, also called medallic engraving, requires a mechanical device similar to a ruling machine. Developed first in the United States in 1817, by Christian Gobrecht, later U.S. Mint engraver, modified somewhat in England in 1829, and independently invented in France in 1830, the process was relatively short lived. It was used, however, for engraving book illustrations and banknotes displaying small reliefs as design elements.

A three-dimensional relief item, as a medal, coin, seal, cameo – or its cast replica, pattern or negative – is fastened in the device. A tracing point is repeatedly played over the object's surface, hundredths of an inch apart, while the design is transferred to either a drawing pen reproducing the design by multiple close-line drawings, or to an etcher inscribing the line on a printing plate.

Anaglyptography – literally embossed writing – was popular 1831-61 but was replaced by the rise of photoengraving following the introduction of that graphic art process in 1880. Modern references to anaglyptography often refer to the technique as medal-engraving or medallion-engraving  (but both these terms wrongly confuse the general public as a machine that engraves the modulated relief in dies).

Numismatic connection.  In 1835 the inventor of the medallic engraver, then a die engraver at the Philadelphia Mint, Christian Gobrecht employed his device to transfer a glyptic design to the surface of a die for engraving a coin design. He did this by making an exact size model of the Liberty Seated device he had chosen to appear on the obverse of this coin.

Gobrecht set up his tracer to reproduce this design on the face of a smooth die blank properly coated with zinc oxide (CHINESE WHITE). The device drew the exact design on the die's surface and he was able to create the modulated relief by engraving this design into the die. This is the only reported incidence of this technique in the numismatic field. Though highly attractive, the resulting design was somewhat stiff and mechanical looking.

History of the anaglyptography. The medallic engraving machine was based on early 19th century models of the ruling pen and a mechanical contrivance, the Collard engraving machine, used for watch case and snuff box decoration (by flat engraving).  In Philadelphia in 1817, Christian Gobrecht (1785-1844), developed the first successful, but primitive, medallic engraving device, even though he had been an engraver for only seven years, a craft he had taken up as a 25-year old watchmaker apprentice in Lancaster County Pennsylvania.

In 1818 Gobrecht had a fellow Philadelphian build for him a somewhat more refined machine. This was accomplished by Joseph Saxton (1790-1873), a machinist, inventor, and later, constructor of weighing equipment for the U.S. Mint. Little was done with the machine for a decade, but Saxton did take it with him to London and improved upon it there in 1829.

Meanwhile, an American artist and banknote engraver, Asa Spencer (ca 1805-?), had built one or more machines, based on Gobrecht's design, and sent engravings to London. A gentleman there purchased one of Spencer's machines, and, upon the gentleman's death, it passed to a technician in London, John Bate. Little is known of Bate, who is described only as "optician to the admiralty," but he is to play an important role in the story of the anaglyptograph.

Bate claims to have improved upon "an existing machine" in 1830; but nowhere did he credit the machine to Spencer (or Gobrecht or Saxton) or the gentleman he acquired it from. Instead, in 1832, John Bate obtained British patent number 6254 for the medallic engraving machine (as if it was his own creation).

Completely independent of these efforts, in the year 1830, a French mechanic, Achille Collas (1795-1859), spent six months developing a medallic engraving machine after he received an order to build a ruling pen. Collas had knowledge of the Collard machine (called tour a guillocher) since it had appeared in French references as early as 1816.

Collas produced his first successful engraving in 1831, but it was the following year Collas was to meet the one man who was to do more for medallic engraving than anyone else, Vincent Nolte (1779-1856). Nolte was a French businessman who recognized the commercial possibilities when shown Collas's device.

Nolte acquired the rights from Achille Collas, hired him to perform the work, and established a company to capitalize on the process. In the period 1832-34 Nolte persuaded Alexander de Lachevardiere (c1790-1855), printer, publisher of encyclopedias and Magasin Pittoresque, somewhat of a "penny magazine" of Paris, to join the venture. Lachevardiere became the manager and gave the company his name; Nolte was the firm's major stockholder.

In 1834 the company, Lachevardiere & Cie, launched one of the largest undertakings in numismatic publishing: Tresor de Numismatique et de Glyptique.  The company gathered medals, coins, seals, small bas-reliefs from many sources in Europe and with Achille Collas operating his medallic engraver, created and published four folio pages and one page of text each week. An astounding enterprise, this process was to continue for 24 years and result in 20 published volumes!

By 1836 this operation was underway and progressing smoothly in Paris, Nolte visited London to search for more ventures for Collas's talents. Since medals were the primary source for the engravings, Nolte went seeking medals at the British Museum. There he had unexpected fate in being introduced to a curatorial assistant there, Edward Hawkins (1780-1867). Hawkins – who later was to become the museum's Keeper of Antiquities and Coins – prophetically had in hand a manuscript on the medals of Great Britain.

With Hawkins approval, Nolte immediately launched a drive to publish this work and to illustrate the book with medallic engravings to be prepared by his company and Collas's machine in Paris. Nolte was unprepared for what was to transpire.

News of the impending publication of a British medallic history ultimately reached John Bate, who was unhappy that a foreigner would produce the plates by the same process for which he held the British patent! – despite the fact he had done little to create or promote the machine.

However, just as aggressively as Nolte sought publishers, Bate sought to block the project. Nolte finally succeeded in persuading the British Museum to publish – or at least cooperate – with a private London publisher, Charles Tilt (1797-1861) who would oversee and finance the project.

The ruckus increased and led all the way to the British Parliament. Because it was a British Museum publication, it came under the jurisdiction of a committee in the House of Commons. Hearings were held, charges and countercharges; Bate versus Nolte, and each had gathered allies and experts to give testimony.

It was the prominent British engraver and medallist, William Wyon (1795-1851), who was the pivot in the judicial proceedings. Nolte had shown Wyon his work, even engravings of Wyon's own medals, and had Wyon lined up on his side. But when time came to give testimony, Wyon swung over to the Bate side and Nolte lost.

By historical perspective the hearings are amusing. It was like a sub-committee of

the U.S. Congress meeting to determine which engraver would be employed to produce plates for a book to be published by the Smithsonian Institution.

Each party was to walk away and utilize his own machine in his own way. Both illustrated books during the mid-19th century. Nolte got in his last licks, however – he enticed Charles Tilt to publish in 1838 the book Authors of England by H.F. Chorley. This was illustrated, of course, with Collas medallic engravings of British medals of its famed writers. Nolte appended in this book, however, a long description of the House of Commons testimony, his rebuttals to Bate's statements, and letters from everyone concerned, including Joseph Saxton who recounted Gobrecht's early efforts in inventing the first anaglyptographic machine.

Perhaps this was an attempt to prove John Bate was not the originator of the anaglyptograph, but Nolte cared little who was given credit. He wanted work for his company and to capitalize on Collas's talents.

And what of Hawkins's manuscript? It wasn't published until 1885, seventeen years after the author's death. But what is even more incredible, the illustrations weren't published until 1904-11. A delay of 68 years after Vincent Nolte wanted to commence the work! Delayed because the holder of the British patent didn't want a foreigner, particularly a Frenchman, to prepare the engravings.

Meanwhile in the early 1840s, Saxton returned to Philadelphia to work for the U.S. Mint. In addition to building new scales he built an even more improved anaglyptograph based on the tracing machine Gobrecht had used in 1835 for the Liberty Seated dollar. (Gobrecht was unable to use this for another coin design because he died in 1844).

Saxton's machine was steam powered and automatic; an operator did not need to reset it constantly. It was used to prepare engravings for two books, published in 1842 and 1846, by the two assayers at the Philadelphia Mint, Jacob B. Eckfeldt (1803-1872) and William E. Du Bois (1810-1881). These were A Manual of Gold and Silver Coins of All Nations and New Varieties of Gold and Silver Coins, Counterfeit Coins, and Bullion.

Also two other Americans had taken an interest in anaglyptographic engravings, both steel banknote engravers, James W. Steel (1799-1879) and Waterman Lilly Ormsby (1809-1883). Both of these gentlemen had built or acquired medallic engraving machines.

Orsmby was a prominent banknote engraver, foe of forgers, founder of the Continental Bank Note Company, and author of a number of books on these subjects. He maintained that the greatest use of anaglyptographic engravings was – not for book illustrations – but for banknote vignettes. He reproduced anaglyptographic samples, described the process, but we are mostly indebted to him for illustrating his anaglyptograph machine in his 1852 book, A Description of the Present System of Bank Note Engraving.

A few books and journals were illustrated with anaglyptographic engravings in the U.S. after the Civil War, but the far larger use by then was banknote vignettes. Nevertheless, the process rapidly fell into disuse in the 1880s with the rise of photoengraving – making printers' plates from photographic prints – with a continuous tone (as in colotype or heliotype), or a screened process (as in halftone).

Mechanical process.  Gobrecht's first engravings were made from paper impressions of embossed blocks. The original relief object is not used directly because the tracer must constantly ride its surface and possibly damage it. Instead, plaster casts were first employed, even shellacking the plaster to give additional surface strength.

Final solution was to make an electrotype of the relief and use this as the actual object to come in contact with the tracer. The electrotype was also advantageous in that it provided the negative from which a positive drawing was obtained.

After the electrotype is fixed in position in the machine, the tracer is setup to cross over its surface. At the end of each path the machine is set to advance only a small distance for the next path. Some engravings exhibit as many as 200 lines per inch, so this setting would be 200th of an inch.

If the surface is completely flat, the etcher makes a straight line. The mechanism is such that when the tracer reaches a rise in relief, it will cause the etcher or drawing pen to deviate from that straight course in proportion to the height of the relief. The adjacent line will make a similar deviation in proportion to its relief. The accumulated effect of all the lines gives a somewhat topographic drawing of the object.

There are some technical aspects of anaglyptographic engravings. Most evident: the accumulated effect of ascending relief will be a closeness of many lines with a darkness to the drawing. The opposite holds true: the effect of descending relief will be widespread lines and a lightness in shading.

Also there is a point of entrance where the first line is drawn; and a point of exit where the last line appears. In most instances the point of entrance would be at 6 o'clock or 9 o'clock on a round object; this would provide horizontal or vertical lines respectively.

For artistic reasons if the operator had a choice he would choose a point of entrance somewhere between 4 and 8 o'clock on the side facing the portrait. This would provide a dark concentration of lines around the facial profile and a light shade areas at the back or top of the head, as if sunlight were falling from above and the face was outlined with a strong shadow.

Retouching.  Some retouching was done by adding additional hand-engraved or hand-drawn lines. Most prevalent of these was at the point of exit where an arc was drawn to outline the object, or to add additional shading. A variety of additional shading was effected by adding crosslines, infrequently by solid shading.

Artistic effect.  Anaglyptographic engravings had an advantage over freehand drawings in that they were somewhat more accurate to the original contours of the bas-relief object. But in comparison to the photographic process there was no contest and it is certainly understandable the process died with the rise of photoengraving. However, for the brief period it reigned in the mid-19th century it served well its purpose.

At best anaglyptography rendered a relief illustration suitable for printing, but the optical effect on the human eye was one of monotonous gray with a loss of the sharp angular detail so often desirable in small objects like coins and medals. Finally, the requirement for a pattern the tracer wouldn't damage was a generation away from the original and a further loss of some detail.

Even so it served numismatics in a way no other technique has done: In copying medal designs for printed illustrations, in one instance of aiding a coin die to be engraved, and the engraving process for banknote vignettes for over half a century!


A32 {1972} Gentleman.

X1  {1838} Nolte.

X2  {1967} Stannard.

X3  {1975} Frazier.

excerpted with permission from

An Encyclopedia of Coin and Medal Technology

For Artists, Makers, Collectors and Curators


Roger W. Burdette, Editor

NNP is 100% non-profit and independent // Your feedback is essential and welcome. // Your feedback is essential and welcome.