Roman shield patterns are very difficult to find, particularly for the Republican period. The only evidence comes from things like wall paintings and sculpture. The carving from the Basilica Aemeliana in the Roman Forum shows a warrior carrying an oval shield, on which is sculpted what appears to be a winged horse (i.e. a Pegasus). Similarly, there is a sculpture that is actually an Imperial copy of a much older Republican frieze. This shows a fully armed horseman. His round shield has some sort of an animal head in the centre (wolf, lion, bear, etc) - alternatively it may be the head of a gorgon. A number of coins show devices as well. One of these shows a bar across the centre of the shield and above this in the top 'half' the letter 'W' (or, depending on your viewpoint, an 'M'!) A carving from Tarragona in Spain shows a shield with a cat's head in the centre. A terracotta sarcophagus shows oval shields that are decorated with geometrical patterns and another relief shows a round shield decorated with a lion's head.

One can surmise that many of these shield designs may well have had some religious significance, or possibly something to do with an ancestor. It is possible that they may have been tribal totems or possibly the zodiacal birth sign of the owner. This may extend the number of possibilities (e.g. a scorpion symbol - such as those born by the Praetorian Guard).

In the very early period, peoples such as the Etruscans were much influenced by Greek ideas. It is very likely that the Hoplon or Argive style of shield may have had devices that were heavily influenced by the Greeks. This would apply to the very early Republican Roman period (7th-6th centuries BC, say).

We have very few examples of Roman shields, perhaps only one from the Republican period (from Egypt) and a few (15, I think) from the mid 3rd century AD (all found at Dura-Europos). The information from the latter is that the inside of the shield was likely to be painted a dull red, although some of those shields show a fancy pattern of blue hearts outlined in white and arranged in lines, radiating out from the centre (and that's on the back of the shield!)

That individual legions had their own shield designs is probably the case. The most telling piece of evidence is from Tacitus’ description of the 2nd Battle of Bedriacum (Cremona), where he says that two Praetorians took up the shields of dead enemy soldiers and used this to approach a catapult that was doing them serious harm. They got close enough to disable the catapult before they were cut down. From this we can infer that there was something about the shields that fooled the enemy long enough to allow them to get close and the obvious thing is that there was some distinctive shield motif.

Where the ‘ordinary’ legions are concerned, we are on much shakier ground. Trajan’s column does show legionary shield patters. However, we have no ‘proof-positive’ idea which legions they refer to. There is a thought that at least one the standards was that of legio I Minerva, as this legion was known to use the open hand as its symbol.

There are a number of recurring themes. These may be summarized as follows:

1. eagle wings (arranged both vertically and horizontally)
2. wreaths of laurel
3. lightning flashes – from two to eight, but always in pairs
4. thunderbolts
5. stars – both five and six-pointed
6. crescent moons
7. tabula ansata – a rectangle with a triangle at each end. This may have carried the legionary number, but none survive so they were probably originally painted onto the column. This motif is always shown (where it is shown) in the centre of the shield, either side of the boss.

Late Roman shield designs are much better known and identified. Our source here is the Notitia dignitatum, known from two 14th century manuscripts, which differ slightly as to details. The information they contain is dated to the late 4th/early 5th century. The work shows all the command areas of the Empire at this time, together with coloured drawings of the regiments/legions that were based in the various vigariges (as the sub-command areas were known).

We can make a fair guess at the colors, if only from knowledge of the likely pigments that would have been available. Ochre could provide the color for the yellows and possibly the reds as well (iron oxide comes in a huge range of colors and shades, from pale yellow to purple). Black could come from powdered charcoal or soot. White would be provided by powdered chalk/magnesia. Blue would be more of a problem, but could well be from woad or powdered indigo, perhaps mixed with chalk to give a lighter color. Green, of course, would be a mix of blue and yellow. Generally, we rule out very bright yellows and reds. Not only would such colors be expensive but it is unlikely that they would stay attached to the shield for very long on campaign! Modern re-enactors have considerable trouble keeping the shield designs in place. We don’t know what the Romans used as a medium. Possibly things like egg-white or drying oils such as poppy oil or linseed may have been employed.

Republican shields are an even worse problem than those for the Empire! We are restricted to no more than a few sources for these. The frieze of the temple of Domitius Ahenobarbus is one - but this does not show any design on the shields. There is a carving showing what may be marines, which dates from very early Principate times (around the Battle of Actium, ca. 30 BC). This shows a trident being held by a hand, around which is wound a snake.). There is also a carving from Spain (city wall of Tarragonna) and this shows a plan shield with a cat's head in the centre which is dated to the 2nd Punic War (218-201 BC).

These designs were, after all, hand painted by the soldiers themselves. They were not factory produced and there were bound to be some errors. Also be aware that, unless they were freshly painted, there would have been areas at the top and bottom that would have been discolored, the top from being handled and the bottom from contact with the ground. The inside of the shield is generally assumed to have been painted red, but there is no real evidence for this. The outside surface was leather (bull hide or calf leather). If the inside had a felt coating, then it would have been left in its natural state, rather than painting it. The likely 'natural' felt color would have been a yellow-brown or a grey-brown shade.

Also remember is that the pigments used would have been inorganic (i.e. made from earths, rocks, etc.) and would have been rather dull colors. The red background, for example, would have most probably have been a brick red, rather than a bright scarlet. Ochre can come in all sorts of shades, from a dull yellow to a red to a purple - but not scarlet. The pigment 'red lead' is scarlet but I would think that its availability would have been rather limited.

A number of tomb carvings that date from the first quarter of the 1st century BC show shields with very simple geometric designs. These are diagonal saltires, with the central vertical 'barley-corn' rib showing. Whether the rib was also painted, we have no means of knowing. Another monument shows a hexagonal cavalry shield decorated with the same sort of geometric pattern in the centre, but with a 'V' at the top and bottom. Another, round, shield shows a lion's head.

There are also small numbers of carvings showing round cavalry shields. One shows a gorgon's head, while another looks like a letter 'M' (in one half of the shield only), with a bar across the centre of the shield.

There is a sculpture in the Basilica Aemilia in Rome, which shows an oval shield with a winged animal on it - probably a horse (Pegasus). There is some dispute about the date for this sculpture. The building was erected in 179 BC, but was reconstructed several times after that, so the date of the sculpture is difficult to pin down. It is, however, Republican in origin.