A geophysical survey was carried out across the area of the Lexden Tumulus, Colchester. Initially the positions of the seven Wellingtonia trees that encircle the mound were established on a site grid aligned to the Ordnance Survey, and indirectly enabled the likely position of the 1924 excavation to be re-established. The survey methods, topology, magnetic gradiometry and ground penetrating radar, have produced two and three dimensional images. A feature of the magnetometry that may have archaeological significance is the collection of unsharp moderate-amplitude positive anomalies in the southwest quadrant lying just outside that conjectured position of the 1924 excavations. They correlate with and unsharp lower resistance feature at nearly the same position and also with a group of returns on the ground radar timeslice images first detected at 1.4m below mound surface, or approximately at the pre-mound ground level: at this location the present mound surface is 1.26m above ground level away from the tumulus. The possibility exists that the group represents a separate area of activity outside the main grave pit, and like the burial pit adjacent to the chamber at Folly Lane, Verulamium, which contained among other fire-affected metallic debris, ‘…numerous fragments of iron and nails, burnt clay, amphorae sherds and crumbs of cremated bone’ <2>. This theory is supported by Foster <3> whose Fig. 17 reproduces the original section-sketch of the western side of the 1924 excavation: iron fragments were found 0.6m into the grave pit at a depth 1.5m below the mound surface, together with iron nails nearby. All other significant iron objects at Lexden came from the base of the grave pit. The resistance data are mainly dominated by near-surface disturbance, much of it recent, but do show evidence of the 1924 excavation trench that approached the mound from the northeast. A curved feature is visible to the southwest of the mound, but is likely also to be of recent origin. A pair of linear features marked by dashed lines on the resistance image (fig 6), give some highly speculative evidence of a larger enclosure surrounding the tumulus. The GPR area survey was confined to the mound itself, but several extended radial traverses were also made. A feature of all traverse datasets is the presence of a clear but characteristically irregular boundary between a surface soil layer and the underlying gravel. Independent estimates suggest a soil propagation velocity up to 0.3 m.ns-1, for the near surface at least. The maximum depth of the grave pit is given by Laver (1927) as 13 ft 6 in or 4.11m below the top of the mound. Taking into account a loss of height around 0.3m since that time, then provided propagation velocity was greater than about 0.127 m.ns-1, the GPR will have just reached the base of the pit. The single GPR traverses are represented as distance-time plots (B-scans) while the area surveys are best viewed as timeslices (C-scans), of which examples are given. Short-delay slices match known near-surface features, and closely resemble the resistance plots. Most of the periphery of the mound, especially on the eastern side, shows a scatter of strong returns over a range of delays; the origin of these is unknown, but the shallower ones are likely to be a result or gardening or other activity on the part of the site occupied since the 1930s. Evidence of the 1924 east trench confirms that of the resistance image. At 1.5m below surface, the timeslices show evidence of a discontinuity in positions matching the locations of what was interpreted as a circular ditch in 1924, but they are not obviously curved and do not make a continuous connected feature. The assumption of circularity is in any case flawed since the 1924 excavation trench width of about 1m means that it is unlikely sufficient length was exposed to reveal visible curvature: approximate tangents are the best that could be expected, and these would fit both curved and straight-sides features. A more distinct quadrangular feature appears at slightly greater depth, the dimensions of which are approximately 17m by 15m. An interpretation of this feature is that it might represent an enclosure surrounding the burial. Approaching maximum depth, the signals that we take to represent the boundary from topsoil to hard gravel converge towards but do not clearly extend into a 5.5 by 8.5m rectangular region. At the same stage, strong returns are detected from within the presumed location of the 1924 excavation pit and therefore the grave, and continue to maximum range. The rectangle is on the same orientation but no longer than suggested by Foster where a rectangular grave pit of dimensions 5.5 by 7.4m is proposed, based on reconstruction of the find locations <3>. However, the Lexden Tumulus grave pit was recorded by the Lavers as oval. This suggests two possibilities. Either the standard of the Lavers’ excavation was much poorer than supposed and they completely misread the shape of the chamber, or the Lexden Tumulus chamber was indeed not rectilinear. The latter explanation would be surprising since anything other than a rectilinear shape would make constructing the plank roof for the chamber unnecessarily complicated. The ‘grave pit’ at the Lexden Tumulus has other clear affinities with other chamber pits at Folly Lane <2>, and Stanway, all unknown to Foster in 1986. The fragmented nature of the funerary goods was similar in all chamber pits at Stanway as was the deposition of that material (concentrated on the floors with more material mixed throughout the lower fill). The Lavers show a dark line which not only is at the top of the buried land surface but also extends down the sides of the pit. One explanation is that part of Laver’s black line was different to elsewhere and was the remains of mineralised planks. Dark material of this nature was observed on the sides of the two largest chamber pits at Stanway <4> where it represents thin lenses of mineral-replacement wood from the plank sides of the mortuary chambers which had been constructed inside the ‘grave’ pits. Further excavation of the Lexden Tumulus is unlikely, but additional non invasive activity may be beneficial. Our survey, especially the ground radar, supports the presence of three nested rectangular structures: a grave pit within an ‘enclosure’ of some kind not previously identified, and the Lavers’ ditch, but not of the circular form proposed. There is also the suggestion of further archaeological remains both within the pit and outside it not located in 1924, though the likelihood that some of this results from discarded 1924 excavation debris cannot be ruled out. <1>