Three and four-bladed arrowheads: Pettigrew 2023 study

The study starts from an archaeological puzzle. In the Iron Age, three-bladed (trilobate) arrowheads, which first appeared on the Eurasian steppe, spread right across the continent, yet an earlier test had found that they penetrate less than two-bladed heads in clay. The authors repeated the clay test and set it against shots through thick leather, chosen to stand in for the leather armour soldiers wore at the time, always using the same modern hunting broadheads fitted with two, three or four blades.

In clay, extra blades reduce penetration: on average 209 mm with two blades, 196 with three and 183 with four, which is 6.2% and 12.5% less. In leather the opposite happens. Through two layers, the force resisting the cut rises with the number of blades (623 N with two, 671 with three, 732 with four), yet penetration rises too: 66 mm with two blades, 75 with three, 82 with four. Through a single layer the difference in cutting force is not significant, but drag on the shaft following the head drops sharply: 48 N with two blades, 27 with three and 21 with four, a reduction of 43% and 56%. With three and four blades the bolt always crossed the 300 mm of air behind the leather and reached the gel block; with two blades it stopped short in five shots out of ten.

The authors' explanation is that more blades open a wider hole in the leather, so the shaft slides through with less friction. In their reading, this makes trilobate heads an adaptation to battlefields where many fighters wore light armour.

No bow was used, but a crossbow: a steel prod tied to an oak stock, with an archery release and a hand-cranked winch, set 950 mm from the target and always drawn to 550 mm. The two bolts were 9.5 mm poplar dowels with aluminium adapters for screw-in points, 618 mm long and weighing 24 g without the head; mean velocity was 35.2 m/s (standard deviation 1.1, over 67 shots), and mean kinetic energy ranged from 18.5 to 18.9 J across the three configurations, with no significant difference. The heads were three 6.5 g Wasp Sharpshooter broadheads with two removable 0.4 g side blades, so the same head, with the same edge, could be shot with two, three or four blades. In the three-blade version the blades were not evenly spaced.

The leather was 3.2-3.6 mm vegetable-tanned cowhide, in strips fixed to a rigid stand, with a block of ballistic gel 300 mm behind it that served only to catch the bolt. The clay was a low-fire pottery clay without grog, in 11 kg blocks. Penetration was measured with tape on the bolt and a ruler; velocity and deceleration through leather came from a camera running at 8,810.57 frames per second and video-analysis software, from which the authors derived the cutting resistance and the drag on the shaft. The comparison rests on 30 shots into clay and 30 through one layer of leather; the two-layer test has only 2 or 3 shots per configuration.

The number of shots is small, and the two-layer leather test, the one that shows the gain in penetration, has 3 shots with two blades and 2 each with three and four: the figures from that part are an indication at best. The heads are modern steel broadheads and the leather is a modern craft tooling leather, not a reconstruction of ancient armour; the authors acknowledge this and argue that the drop in drag does not depend on the metal of the head, but they do not test it with bronze or iron heads. The crossbow shoots at a moderate speed of their choosing, and, as they admit, the experiment does not cover the range of bows, arrows and distances found on ancient battlefields.

Through a single layer, the cutting force with two blades came out high and very variable, which the paper does not fully explain; in the uneven three-blade configuration the bolt flexed and deflected by about 2 mm as it entered the leather. The historical conclusions, about the spread of trilobate heads and the rarity of four-bladed ones, are the authors' interpretation built on these numbers.

For target archers this study changes nothing: it is about hunting and war heads, armour and archaeology. The mechanism it measures does make sense outside that setting, though. In this test, how far the bolt went in depended heavily on the drag of the shaft in the hole the head had cut, not only on sharpness or energy, and two artificial targets can give opposite answers: clay and leather, with the same heads at the same speed, led to contrary conclusions. Anyone reading penetration tests of hunting broadheads will find here a clear example of how much the target material can decide the outcome.