Helvig Conrad Engelhardt worked the moor across campaigns running from the spring of 1858 through 1861, and the volume of material that came up made Thorsberg one of the defining sites of northern European Iron Age archaeology. Its finds sit alongside those from Illerup and Vimose in Denmark as the reference collections for the period. What the objects meant took longer to settle. In 1865 the Danish archaeologist Jens Jacob Asmussen Worsaae was the first to argue that these were not lost goods or a hoard but sacrifices of captured war booty - the equipment of a defeated enemy, given to the water. Engelhardt's finds are now divided between the state archaeological museum at Gottorf Castle in Schleswig and the National Museum of Denmark in Copenhagen, which holds around five hundred pieces, with further material in Flensburg, Hamburg and Berlin.

Much of what went into the bog had been deliberately ruined first. Swords were bent, blades snapped, fittings wrenched off - the objects made useless before they were given up, a practice of ritually killing weapons found widely across Iron Age Europe. The pattern of deposition changes over time. Weapons come to dominate after about 200 AD, and the third and fourth centuries are the peak of the weapon offerings, which suggests conflict on a scale that produced a great deal of captured equipment. Isolated animal bones - a few from horses, along with a bovine horn - lie in the peat among the metal and the cloth. Just outside the moor stands an Iron Age burial mound ringed by a stone circle.

A surprising amount of what the Angles sank here was made inside the Roman Empire, or made in imitation of it. Two phalerae came out of the peat - military decorations in the form of richly worked gold discs 13.2 cm across, produced in the third century in the workshop of a craftsman named Saciro, thought to have been near Cologne. Each shows a seated man holding a spear, possibly Mars. There are Roman helmets and a shirt of chain mail. There is a Germanic face mask of silver, once mounted on a helmet of Roman type, its features composed and entirely blank - a face made to be worn, then given away for good. Some of the fibulae and shield bosses look less like Anglian work than like the possessions of Germanic peoples further south in Greater Germania, who lived in closer contact with Rome.

The textiles are what set Thorsberg apart. Cloaks, tunics and the famous footed trousers survived in the peat, giving a direct view of northern dress at a time when written descriptions of it come almost entirely from Roman observers with an interest in making the north sound barbaric. The trousers were long dated to the fourth century; current opinion puts them no later than 300 AD. Two wooden round shields from the third century came up as well, their boards preserved along with their fittings. Taken together, the clothing and the equipment describe an ordinary fighting man rather than a myth: what he wore, how he carried his sword, how his shield was made.

The single most studied object from the site is small and easy to overlook: a bronze chape from the tip of a scabbard, dated to somewhere around 200 to 260 AD, carrying an inscription in the Elder Futhark on both faces. The runes read owlþuþewaz on one side and niwajemariz on the other - among the oldest runic writing known anywhere. The usual translation runs something like "Wolthuthewaz is well-renowned," with the name understood as servant of Ullr, the god later remembered in Norse sources. That divine name raises a problem with the moor's own. Thorsberg means Thor's hill, and it is tempting to read the whole site as a shrine to Thor - but the name probably comes from a nearby Viking-age hill, and reflects Danish worship centuries after the last offering was made. Engelhardt himself noticed that the hammer symbol appears on several of the finds, and pointed out that the same motif turns up in cultures with no connection to Germanic religion at all.