During the Republican and early Imperial Roman years the army was dominated by heavy infantry, with often small amounts of cavalry usually provided by allies. By the middle of the third century however things were changing, as the vast empire became ever more difficult to defend against the many raids and incursions. Smaller, mounted units could respond more quickly to such problems, but of course in battle both infantry and cavalry were still required. The result was a shift away from the dominance of infantry, although this still remained the majority of the army, so late Roman armies had more of a balanced split between the arms than previously.

As usual there was light and heavy cavalry, plus super heavies such as the cataphracts, but this set concentrates on the light, which was provided by a mixture of archers and javelin-armed troops. Four poses of each are provided, as pictured, and naturally the top row shows the men with javelins. Well actually no, because these are about 29mm long, which at a real length of over 2 metres is rather too long for a javelin and more like a spear. Still simply by cutting them down you would get a javelin so no harm done. As can be seen, three of the four poses have ring hands, and for each of these men the set provides three of the pictured 'javelins' plus a sword. It is a bit tricky to insert weapon into ring hand as the fit is very tight and the soft bendy plastic used makes it hard to push the javelin home. However once there the hands seem to hold the weapon at an angle rather than straight ahead, which makes for a more natural and believable pose. Holding a javelin (or sword) with right arm extended out is perhaps not the most likely of poses, but apart from that they are all fine.

Archery was another feature of Roman warfare that developed greatly by the later Empire, naturally influenced by the many steppe peoples and others to the east. Although Roman light cavalry such as this may well not have been manned by Italians, the skills of mounted archery had become much more common, and again we find four poses in this set depicting such men. The three figures in the act of drawing the bow are pretty good, although the fourth pose seems somewhat less likely to us. Still nothing much wrong here either so some useful archers.

Much late Roman cavalry seems to have been armoured, but light troops were often unarmoured and it seems reasonable to suppose this was often the case with the Romans too. All these figures have no visible armour apart from the two men with helmets, both of which are in styles suitable for the period. Two others wear the pannonian hat, which was very common when not in battle, but if possible men would have preferred helmets. However the men were responsible for purchasing their own kit, and sources of supply were many, varied and often irregular, so it is sensible to suppose such clothing was worn in sight of the enemy. This applies equally to the bare-headed men, so while more helmets might have been better we have no problem with these figures as they stand.

All the figures wear the fashionable long-sleeved tunic, which HaT have engraved to reflect the common patterns of discs and bands. On the lower legs all have cloth bindings (something like the later puttees), which is authentic but was only one of many options so some men with laced socks or long trousers would have been a good idea.

All the men carry a sword, correctly held on the left hip. One man has his suspended by a baldric over the right shoulder, but the rest have it hanging from their waist belt. Both are correct but it seems the baldric was the more common, so somewhat misrepresented here. Most of the men carry a small round shield (oval ones would also have been correct), and the set also includes some spares, although neither of the two poses without an integral shield have any means of attaching these. Still such men did not always carry a shield so this is not a problem.

Although eight poses in a cavalry set is quite good, only two horse poses is not so good. The poses are OK rather than great, but both suggest a fairly gentle pace rather than a full-on charge, which is as it should be for this type of cavalry. Their equipment looks fine although the usual simplifications (such as the lack of a bit) are evident. However the saddle is not good as it lacks the characteristic 'horns' in each corner that allowed the rider to maintain a firm grip despite the lack of stirrups, which were still centuries away at this time. During the fifth century a new type of saddle seems to have been introduced, but this does not resemble that one either.

Our first thought on seeing these was they were a bit small, although certainly the light cavalry role would be more likely to be fulfilled by lighter men for obvious reasons. However they are within the historically correct height bounds for the time, and they look OK on the mounts, which would have seemed small by modern standards too. The sculpting is pretty good, and the men fit their horses very well, while there is no flash.

This is a workmanlike set which depicts its subject pretty well, so long as you cut down the spears to make javelins. The eight poses make a big difference when representing two different types of troops, so while a couple more horse poses would not have gone amiss this box does the job nicely.