<P> A third Greek author, Biton (fl . 2nd century BC), whose reliability has been positively reevaluated by recent scholarship, described two advanced forms of the gastraphetes, which he credits to Zopyros, an engineer from southern Italy . Zopyrus has been plausibly equated with a Pythagorean of that name who seems to have flourished in the late 5th century BC . He probably designed his bow - machines on the occasion of the sieges of Cumae and Milet between 421 BC and 401 BC . The bows of these machines already featured a winched pull back system and could apparently throw two missiles at once . </P> <P> From the mid-4th century BC onwards, evidence of the Greek use of crossbows becomes more dense and varied: Arrow - shooting machines (katapeltai) are briefly mentioned by Aeneas Tacticus in his treatise on siegecraft written around 350 BC . An Athenian inventory from 330--329 BC includes catapults bolts with heads and flights . Arrow - shooting machines in action are reported from Philip II's siege of Perinthos in Thrace in 340 BC . At the same time, Greek fortifications began to feature high towers with shuttered windows in the top, presumably to house anti-personnel arrow shooters, as in Aigosthena . </P> <P> The transition to torsion catapults, which are not considered crossbows and came to dominate Greek and Roman artillery design, is first evident in inventories of the Athenian arsenal from between 338 and 326 BC . </P> <P> The ancient world knew a variety of mechanical hand - held weapons similar to the later medieval crossbow . The exact terminology is a subject of continuing scholarly debate . Roman authors like Vegetius (fl . 4th century) note repeatedly the use of arrow shooting weapons such as arcuballista and manuballista respectively cheiroballista . While most scholars agree that one or more of these terms refer to handheld mechanical weapons, there is disagreement whether these were flexion bows or torsion powered like the recent Xanten find . </P>

Who used the crossbow in the middle ages